From phil at bolthole.com Mon Jun 1 00:36:17 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 31 May 2009 15:36:17 -0700 Subject: [csw-maintainers] Bug in our uwatch stuffs Message-ID: <20090531223617.GB32936@bolthole.com> BTW: there appears to be a bug in our "uwatch" definition and/or code. I an maintainer for the obsolete "glib"(version 1) NOT "glib2". the uwatch stuff is incorrectly reporting glib2 updates, to the glib maintainer -- me. ----- Forwarded message from Upstream Package Watch ----- Date: Sat, 30 May 2009 04:46:49 +0200 (CEST) From: Upstream Package Watch To: phil at opencsw.org Subject: [svn] glib upstream update notification Hello dear glib maintainer, The upstream notification job has detected the availability of new files for glib. The following upstream file(s): glib-2.20.3.tar.bz2 is/are available at the following url(s): http://ftp.gnome.org/pub/GNOME/sources/glib/2.20/ Please consider updating your package. -- Kindest regards upstream notification job ----- End forwarded message ----- From skayser at opencsw.org Mon Jun 1 12:21:22 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 01 Jun 2009 12:21:22 +0200 Subject: [csw-maintainers] Anyone successfully downloaded gnulib? Download problems here. Message-ID: <4A23ABA2.5080001@opencsw.org> Hi, has anyone successfully cloned the gnulib git repository? When i try to follow the download instructions from their homepage [1] the clone operation times out. # git clone git://git.savannah.gnu.org/gnulib.git Initialized empty Git repository in /root/tmp/gnulib/.git/ cdgit.savannah.gnu.org[0: 199.232.41.69]: errno=Connection timed out fatal: unable to connect a socket (Connection timed out) I am doing this from my home box, i.e. there is no proxy required and no restrictive firewall in between. Sebastian [1] http://www.gnu.org/software/gnulib/ From skayser at opencsw.org Mon Jun 1 12:29:48 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 01 Jun 2009 12:29:48 +0200 Subject: [csw-maintainers] Anyone successfully downloaded gnulib? Download problems here. In-Reply-To: <4A23ABA2.5080001@opencsw.org> References: <4A23ABA2.5080001@opencsw.org> Message-ID: <4A23AD9C.4030101@opencsw.org> Sebastian Kayser wrote: > has anyone successfully cloned the gnulib git repository? When i try to > follow the download instructions from their homepage [1] the clone > operation times out. Ok, never mind, i just noticed the news on their front page [1]. A couple of services is down, git is one of them. Has anyone ever thought of packaging gnulib so that we have it centrally available on the buildfarm? Sebastian [1] http://savannah.gnu.org/forum/forum.php?forum_id=5826 From dam at opencsw.org Mon Jun 1 12:43:51 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 1 Jun 2009 12:43:51 +0200 Subject: [csw-maintainers] Anyone successfully downloaded gnulib? Download problems here. In-Reply-To: <4A23AD9C.4030101@opencsw.org> References: <4A23ABA2.5080001@opencsw.org> <4A23AD9C.4030101@opencsw.org> Message-ID: <8221E473-F04F-4A59-858D-BDA1EC489722@opencsw.org> Hi Sebastian, Am 01.06.2009 um 12:29 schrieb Sebastian Kayser: > Sebastian Kayser wrote: >> has anyone successfully cloned the gnulib git repository? When i >> try to >> follow the download instructions from their homepage [1] the clone >> operation times out. > > Ok, never mind, i just noticed the news on their front page [1]. A > couple of services is down, git is one of them. > > Has anyone ever thought of packaging gnulib so that we have it > centrally > available on the buildfarm? gnulib is not meant to be packaged. The projects needing them should include the functionality they actually need: > Gnulib takes a different approach. Its components are intended to be > shared at the source level, rather than being a library that gets > built, > installed, and linked against. Thus, there is no distribution tarball; > the idea is to copy files from Gnulib into your own source tree. Best regards -- Dago From dam at opencsw.org Mon Jun 1 12:46:14 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 1 Jun 2009 12:46:14 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> Message-ID: Hi, Am 31.05.2009 um 23:00 schrieb Dagobert Michelsen: > 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- > packages/xcbgen/xtypes.py I guess this is the reason why the py-files aren't gettint compiled: /opt/csw/X11/ Should I put the py-files into /opt/csw/lib/python* instead regardless of the libdir setting? Best regards -- Dago From skayser at opencsw.org Mon Jun 1 12:55:00 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 01 Jun 2009 12:55:00 +0200 Subject: [csw-maintainers] Anyone successfully downloaded gnulib? Download problems here. In-Reply-To: <8221E473-F04F-4A59-858D-BDA1EC489722@opencsw.org> References: <4A23ABA2.5080001@opencsw.org> <4A23AD9C.4030101@opencsw.org> <8221E473-F04F-4A59-858D-BDA1EC489722@opencsw.org> Message-ID: <4A23B384.3090302@opencsw.org> Dagobert Michelsen wrote: > Am 01.06.2009 um 12:29 schrieb Sebastian Kayser: >> Sebastian Kayser wrote: >>> has anyone successfully cloned the gnulib git repository? When i >>> try to >>> follow the download instructions from their homepage [1] the clone >>> operation times out. >> Ok, never mind, i just noticed the news on their front page [1]. A >> couple of services is down, git is one of them. >> >> Has anyone ever thought of packaging gnulib so that we have it >> centrally >> available on the buildfarm? > > gnulib is not meant to be packaged. The projects needing them should > include the functionality they actually need: > >> Gnulib takes a different approach. Its components are intended to be >> shared at the source level, rather than being a library that gets >> built, >> installed, and linked against. Thus, there is no distribution tarball; >> the idea is to copy files from Gnulib into your own source tree. That's exactly what i was aiming at. Having all the source files handy in one location (package) gives our maintainers instant access to them. No hassle cloning non-available git repositories. Debian has a gnulib package also. I wonder why, but heck, when even they as a GNU platform have one we should have one too ;) Sebastian From pfelecan at opencsw.org Mon Jun 1 17:43:31 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 01 Jun 2009 17:43:31 +0200 Subject: [csw-maintainers] assembler in shared libraries Message-ID: Trying to package/port projects using optimizations implemented in assembler I encounter the same issue for which I didn't found a solution. For the moment, the issue manifests itself only on i386 architecture as the porting has as source, projects developed on Linux and concerns mainly video applications, e.g., ffmpeg, xvid, &c. In the linking phase of the share library I get the following message: Text relocation remains referenced against symbol offset in file 0x7 bitstream/x86_asm/cbp_mmx.o . . . 0xc6 utils/x86_asm/cpuid.o ld: fatal: relocations remain against allocatable but non-writable sections collect2: ld returned 1 exit status The result of my research can be summarized as follows: don't use hand written assembly in shared libraries if you're not using the GNU linker. The consequences of this are: - don't activate assembler optimizations: poor performance - where the previous is not possible, don't generate shared libraries containing the assembler code: build only statically linked binaries --- at least the corresponding library is statically linked Is there a fellow maintainer having a positive experience on building shared libraries for Solaris containing assembler code and willing to share his solution? -- Peter From bwalton at opencsw.org Mon Jun 1 17:50:42 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 01 Jun 2009 11:50:42 -0400 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: References: Message-ID: <1243871326-sup-7314@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Mon Jun 01 11:43:31 -0400 2009: > Is there a fellow maintainer having a positive experience on building > shared libraries for Solaris containing assembler code and willing to > share his solution? This isn't something I've bumped into personally, but would this be a candidate for using the gnu linker instead of the solaris one? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Mon Jun 1 17:51:55 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 10:51:55 -0500 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: References: Message-ID: <4A23F91B.1020803@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter FELECAN wrote: > Trying to package/port projects using optimizations implemented in > assembler I encounter the same issue for which I didn't found a solution. > > For the moment, the issue manifests itself only on i386 architecture as > the porting has as source, projects developed on Linux and concerns > mainly video applications, e.g., ffmpeg, xvid, &c. > > In the linking phase of the share library I get the following message: > > Text relocation remains referenced > against symbol offset in file > 0x7 bitstream/x86_asm/cbp_mmx.o > . > . > . > 0xc6 utils/x86_asm/cpuid.o > ld: fatal: relocations remain against allocatable but non-writable sections > collect2: ld returned 1 exit status > > The result of my research can be summarized as follows: don't > use hand written assembly in shared libraries if you're not using the > GNU linker. > > The consequences of this are: > > - don't activate assembler optimizations: poor performance > - where the previous is not possible, don't generate shared > libraries containing the assembler code: build only statically linked > binaries --- at least the corresponding library is statically linked > > Is there a fellow maintainer having a positive experience on building > shared libraries for Solaris containing assembler code and willing to > share his solution? > The solution I used was to use gnu linker. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoj+RsACgkQLrhmsXMSLxcL8wCggoidR+PkaZpTmwwLADpfWA59 OqkAoJRJiNVJjHkYG9PPBZ2bksBhRihv =etrg -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 1 18:02:17 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 11:02:17 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> Message-ID: <4A23FB89.50006@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Dago, Dagobert Michelsen wrote: >> root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/xtypes.py > > I guess this is the reason why the py-files aren't gettint compiled: > /opt/csw/X11/ > Should I put the py-files into /opt/csw/lib/python* instead regardless > of the > libdir setting? if the library files are to be used within python programs yes. that is to say, if a python script does something like "from xtypes import foo" or "import xtypes" then it needs to go under /opt/csw/lib/python/site-packages - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoj+4kACgkQLrhmsXMSLxf6HwCfUVAjnDAkrfyB3te0iX/Cy/hc W1oAnA3fZy5Rz644LID5BOaeDrczbVQX =SkxE -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 1 18:19:13 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 11:19:13 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> Message-ID: <4A23FF81.9070903@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi, > > Am 31.05.2009 um 23:00 schrieb Dagobert Michelsen: >> 0 17552 cpio 3 >> root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/xtypes.py > > I guess this is the reason why the py-files aren't gettint compiled: > /opt/csw/X11/ > Should I put the py-files into /opt/csw/lib/python* instead regardless > of the > libdir setting? I did some testing on the class script this morning, and it works as designed. as long as your .py library files are under /opt/csw/lib/python the class script will attempt to compile them. **** NOTE **** The default location for the library files is currently NOT /opt/csw/lib/python The version in testing has fixed this problem. Please hold your python packages until the new one is on the build farm. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoj/4EACgkQLrhmsXMSLxfnMwCcDoXFiOBp9rlahNMshfqetRPj vRkAoKjnUl5AbjEzgMf3kb4EGkyzM1O8 =IcHd -----END PGP SIGNATURE----- From skayser at opencsw.org Mon Jun 1 18:37:36 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 01 Jun 2009 18:37:36 +0200 Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") Message-ID: <4A2403D0.4090109@opencsw.org> Hi, a couple of RFEs for the "GAR package status" [1] page. Can we get: * an "last updated" field at the top of the page * the numbers / stats right at the top of the page (instead of further down) * a contact address for the maintainer of the page on the page (for queries like this one and to get packages that are in GAR but not listed as such fixed) Ah, and some sort of monthly historical data would be cool also. So that we can see some progress. Sebastian [1] http://www.opencsw.org/package-gar-status.html From dam at opencsw.org Mon Jun 1 18:51:05 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 1 Jun 2009 18:51:05 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A23FF81.9070903@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> <4A23FF81.9070903@opencsw.org> Message-ID: Hi Mike, Am 01.06.2009 um 18:19 schrieb Mike Watters: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Dagobert Michelsen wrote: >> Hi, >> >> Am 31.05.2009 um 23:00 schrieb Dagobert Michelsen: >>> 0 17552 cpio 3 >>> root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/xtypes.py >> >> I guess this is the reason why the py-files aren't gettint compiled: >> /opt/csw/X11/ >> Should I put the py-files into /opt/csw/lib/python* instead >> regardless >> of the >> libdir setting? > > I did some testing on the class script this morning, and it works > as designed. > as long as your .py library files are under /opt/csw/lib/python the > class > script will attempt to compile them. > > **** NOTE **** > The default location for the library files is currently NOT /opt/ > csw/lib/python > The version in testing has fixed this problem. Please hold your > python > packages until the new one is on the build farm. Just for reference, why aren't you compiling just the pathes you get and then installf them? I remember you said performance, but if you only installf -f once in postinstall this shouldn't be an issue. Best regards -- Dago From mwatters at opencsw.org Mon Jun 1 20:24:18 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 13:24:18 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> <4A23FF81.9070903@opencsw.org> Message-ID: <4A241CD2.7050803@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Dago, Dagobert Michelsen wrote: > > Just for reference, why aren't you compiling just the pathes you get and > then installf them? I remember you said performance, but if you only > installf -f once in postinstall this shouldn't be an issue. The compile of all the files was done very quickly ~10 - 15 seconds. Even with the single installf -f it took forever between 1 and 3 seconds per file. that does not sound like much, but when you start talking about 2000 or 3000 files it adds up quickly. I am not sure if that is a problem with my machine? large contents file, or some other issue? or if it is normal for installf to be that slow. I would rather do it this way, which does not show that kind of response time. I think this makes for a Better user experience. installed on a fairly quiet v490 running solaris 10 for my benchmark. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkokHNEACgkQLrhmsXMSLxfnogCdEpHayfPHrzWeHdiYse7HUk1o a/cAoIhjUk9UzpB6obefhfmFVqaTCurq =io4B -----END PGP SIGNATURE----- From phil at bolthole.com Mon Jun 1 20:27:46 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 1 Jun 2009 11:27:46 -0700 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A241CD2.7050803@opencsw.org> References: <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> <4A23FF81.9070903@opencsw.org> <4A241CD2.7050803@opencsw.org> Message-ID: <20090601182746.GI35922@bolthole.com> On Mon, Jun 01, 2009 at 01:24:18PM -0500, Mike Watters wrote: > Even with the single installf -f it took forever > between 1 and 3 seconds per file. that does not sound like much, > but when you start talking about 2000 or 3000 files it adds up quickly. > I am not sure if that is a problem with my machine? large contents file, or > some other issue? or if it is normal for installf to be that slow. I would > rather do it this way, which does not show that kind of response time. > I think this makes for a Better user experience. > > installed on a fairly quiet v490 running solaris 10 for my benchmark. I am not sure what "this way" is. I looked at http://wiki.opencsw.org/cswclassutils-package and it SEEMS to have a very brief gar directive. Would you please update the wiki page to state that explicitly... and ALSO to give a more "manual" description of what is neccessary, if one is putting together a package by hand rather than using gar? From mwatters at opencsw.org Mon Jun 1 20:34:49 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 13:34:49 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: <20090601182746.GI35922@bolthole.com> References: <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> <4A23FF81.9070903@opencsw.org> <4A241CD2.7050803@opencsw.org> <20090601182746.GI35922@bolthole.com> Message-ID: <4A241F49.9060401@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > I am not sure what "this way" is. > I looked at http://wiki.opencsw.org/cswclassutils-package and it SEEMS > to have a very brief gar directive. > > Would you please update the wiki page to state that explicitly... and ALSO > to give a more "manual" description of what is neccessary, if one is > putting together a package by hand rather than using gar? "this way" == classutils Yes, I am aware of the minimalistic approach on the wiki. I gave Peter just a tiny description for the README.CSW file and he was kind enough to put that on the wiki. I have not got around to updating it with a more detailed description/usage. it is on my todo list ;-) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkokH0kACgkQLrhmsXMSLxdpPgCfT6FXv+VuGu+QJ/nEPm5A6Ri8 sl0An1pj3O7db1gWU0jwRkyBch1CvTf3 =7YFf -----END PGP SIGNATURE----- From pfelecan at opencsw.org Mon Jun 1 20:42:39 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 01 Jun 2009 20:42:39 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <4A23F91B.1020803@opencsw.org> (Mike Watters's message of "Mon\, 01 Jun 2009 10\:51\:55 -0500") References: <4A23F91B.1020803@opencsw.org> Message-ID: Mike Watters writes: > Peter FELECAN wrote: >> Trying to package/port projects using optimizations implemented in >> assembler I encounter the same issue for which I didn't found a solution. >> >> For the moment, the issue manifests itself only on i386 architecture as >> the porting has as source, projects developed on Linux and concerns >> mainly video applications, e.g., ffmpeg, xvid, &c. >> >> In the linking phase of the share library I get the following message: >> >> Text relocation remains referenced >> against symbol offset in file >> 0x7 bitstream/x86_asm/cbp_mmx.o >> . >> . >> . >> 0xc6 utils/x86_asm/cpuid.o >> ld: fatal: relocations remain against allocatable but non-writable sections >> collect2: ld returned 1 exit status >> >> The result of my research can be summarized as follows: don't >> use hand written assembly in shared libraries if you're not using the >> GNU linker. >> >> The consequences of this are: >> >> - don't activate assembler optimizations: poor performance >> - where the previous is not possible, don't generate shared >> libraries containing the assembler code: build only statically linked >> binaries --- at least the corresponding library is statically linked >> >> Is there a fellow maintainer having a positive experience on building >> shared libraries for Solaris containing assembler code and willing to >> share his solution? >> > > The solution I used was to use gnu linker. Sure. I'm not comfortable with that but why not? However, I'm flabbergasted: I thought that I knew how to do that... -B/opt/csw/bin or GCC_EXEC_PREFIX... none works! Still uses /usr/ccs/bin/ld -- Peter From bwalton at opencsw.org Mon Jun 1 20:52:50 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 01 Jun 2009 14:52:50 -0400 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: References: <4A23F91B.1020803@opencsw.org> Message-ID: <1243882320-sup-282@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Mon Jun 01 14:42:39 -0400 2009: > Sure. I'm not comfortable with that but why not? However, I'm > flabbergasted: I thought that I knew how to do that... -B/opt/csw/bin > or GCC_EXEC_PREFIX... none works! Still uses /usr/ccs/bin/ld I've seen some autoconf/configure scripts that accept the path to ld as an option and don't work against some of the other things you've tried. HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Mon Jun 1 21:01:37 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 1 Jun 2009 12:01:37 -0700 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: References: <4A23F91B.1020803@opencsw.org> Message-ID: <20090601190137.GK35922@bolthole.com> On Mon, Jun 01, 2009 at 08:42:39PM +0200, Peter FELECAN wrote: > Sure. I'm not comfortable with that but why not? However, I'm > flabbergasted: I thought that I knew how to do that... -B/opt/csw/bin > or GCC_EXEC_PREFIX... none works! Still uses /usr/ccs/bin/ld > -- a WELL-BEHAVED set of Makefiles, will allow you to override linker by simply specifying LD=/path/to/ld similar to setting LDFLAGS and so on. in one or both of the Makefile or your environment vars. From william at wbonnet.net Mon Jun 1 21:33:07 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 01 Jun 2009 21:33:07 +0200 Subject: [csw-maintainers] Bug in our uwatch stuffs In-Reply-To: <20090531223617.GB32936@bolthole.com> References: <20090531223617.GB32936@bolthole.com> Message-ID: <4A242CF3.70809@wbonnet.net> Hi Philip > BTW: there appears to be a bug in our "uwatch" definition and/or code. > It is not a bug it is a functionality :) > I an maintainer for the obsolete "glib"(version 1) NOT > "glib2". > > the uwatch stuff is incorrectly reporting glib2 updates, to the glib > maintainer -- me. > The problem comes from the fact that both package have the same GARNAME in the Makefile. glib2 is using glib as it's name, thus uwatch considers it is your package. Maybe we should consider to move GARNAME to glib2 for glib2 :) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From phil at bolthole.com Mon Jun 1 21:43:35 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 1 Jun 2009 12:43:35 -0700 Subject: [csw-maintainers] Bug in our uwatch stuffs In-Reply-To: <4A242CF3.70809@wbonnet.net> References: <20090531223617.GB32936@bolthole.com> <4A242CF3.70809@wbonnet.net> Message-ID: <20090601194335.GN35922@bolthole.com> > The problem comes from the fact that both package have the same GARNAME > in the Makefile. glib2 is using glib as it's name, thus uwatch considers > it is your package. Maybe we should consider to move GARNAME to glib2 > for glib2 :) sounds sensible. would the glib2 maintainer please fix their gar configs? :) From william at wbonnet.net Mon Jun 1 22:09:10 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 01 Jun 2009 22:09:10 +0200 Subject: [csw-maintainers] Packages age statistics : May '09 In-Reply-To: <49DA5BED.9010908@wbonnet.net> References: <49DA5BED.9010908@wbonnet.net> Message-ID: <4A243566.90903@wbonnet.net> Hi Here is the June update of package age statistics for the current source. The third column show the delta between this month and previous month, the last one the cumulated delta over the year. As you can notice packages total from years 2003 to 2008 decreased, which is good (they have been updated). Delta of packages in may is +105, which is very good. This following statistics show that 76 packages have been update over the last month, and 29 new packages added. Thanks to you all. Year Total Delta Year 1997 1 0 0 1998 1 0 0 2001 3 0 0 2002 4 0 0 2003 23 0 -3 2004 98 0 -9 2005 162 -11 -17 2006 177 -4 -27 2007 147 -26 -52 2008 304 -35 -173 2009 508 +105 +530 cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Mon Jun 1 22:50:25 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 01 Jun 2009 22:50:25 +0200 Subject: [csw-maintainers] Packages age statistics : May '09 In-Reply-To: <4A243566.90903@wbonnet.net> References: <49DA5BED.9010908@wbonnet.net> <4A243566.90903@wbonnet.net> Message-ID: <4A243F11.7080809@opencsw.org> Hi William, William Bonnet wrote: > Here is the June update of package age statistics for the current source. > > The third column show the delta between this month and previous month, > the last one the cumulated delta over the year. As you can notice > packages total from years 2003 to 2008 decreased, which is good (they > have been updated). Delta of packages in may is +105, which is very > good. This following statistics show that 76 packages have been update > over the last month, and 29 new packages added. > > Thanks to you all. > > Year Total Delta Year > 1997 1 0 0 > 1998 1 0 0 > 2001 3 0 0 > 2002 4 0 0 > 2003 23 0 -3 > 2004 98 0 -9 > 2005 162 -11 -17 > 2006 177 -4 -27 > 2007 147 -26 -52 > 2008 304 -35 -173 > 2009 508 +105 +530 would you mind elaborating on these statistics a bit more? Right now i don't know whether i am putting them together correctly. - What exactly are the figures in the Total column? Currently available packages released in that year? So we would still have one package each from back in 1997 and 1998? So taking for example the "Delta of packages in may is +105". Does it mean, we released 105 packages in May (either updated or new) that haven't been available in a 2009 revision before (thus contributing to the Total 2009 figure)? This would be quite neat. And +530 would be 530 updated/released packages in 2009? Why would it differ from the Total number of 508 then? Getting back to another year, taking 2008 for example. Do we have 304 packages in the catalog which are from 2008? Have we updated/dropped 35 of them last month and 173 over the course of this year? Sebastian From Joerg.Schilling at fokus.fraunhofer.de Mon Jun 1 23:57:39 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Mon, 01 Jun 2009 23:57:39 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <20090601190137.GK35922@bolthole.com> References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> Message-ID: <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> Philip Brown wrote: > a WELL-BEHAVED set of Makefiles, will allow you to override linker by > simply specifying > > LD=/path/to/ld Well, if the "new" LD behaves significantly similar to the expected LD, you are correct. For simple makefiles you may be correct too...... For more complex makefiles you are not correct as these makefiles may have more internal knowledge than you may expect. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From mwatters at opencsw.org Tue Jun 2 00:23:39 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 17:23:39 -0500 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <4A2454EB.7000308@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Joerg Schilling wrote: > Philip Brown wrote: > >> a WELL-BEHAVED set of Makefiles, will allow you to override linker by >> simply specifying >> >> LD=/path/to/ld > > Well, if the "new" LD behaves significantly similar to the expected LD, > you are correct. > > For simple makefiles you may be correct too...... > > For more complex makefiles you are not correct as these makefiles may > have more internal knowledge than you may expect. > > J?rg > It is all up to the writer of the Makefile, Makefiles can force whatever the creator wants it to force by default. we always have the option of changing it. either with a patch to Makefile.in or autoconf. however, if the package uses libtool, the LD environment variable should be honored. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkokVOsACgkQLrhmsXMSLxf49wCeLk27ZKfkiKfB7Eo0WMOHdY/0 KrUAni52pJ7UXIyJgbs+R1Ge35U7On91 =32kt -----END PGP SIGNATURE----- From car at opencsw.org Tue Jun 2 08:54:46 2009 From: car at opencsw.org (Chris Reece) Date: Tue, 2 Jun 2009 18:54:46 +1200 (NZST) Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> References: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> Message-ID: <36490.131.203.105.13.1243925686.squirrel@webmail.jessies.org> On Thu, May 28, 2009 9:12 am, Dagobert Michelsen wrote: > Chris: As you already did findutils with glocate, it would be > really cool if you could also package up slocate > as more secure alternative. I looked over the GAR documentation at the weekend and an hour or so later I was none the wiser. I checked out a local copy on my one Sun box at home, found what looks like an attempt at findutils in there, not mine. Didn't get anywhere with it. In the interim, there's a traditionally-rolled, up to date package in testing: /home/testing/findutils-4.4.1,REV=2009.06.01-SunOS5.8-i386-CSW.pkg.gz /home/testing/findutils-4.4.1,REV=2009.06.01-SunOS5.8-sparc-CSW.pkg.gz There's a postinstall script in there to add a gupdatedb job in the small hours. This is the first time I've packaged in a long time, so I'd be happy to receive feedback. I have no way to test the i386 package, in particular. As for slocate, I'm guessing that it's going to require chunks of GNUlib. We're missing fts, for instance. Have you used a contemporary version of this, or seen it running on Solaris? Cheers, Chris. From pfelecan at opencsw.org Tue Jun 2 11:01:13 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 02 Jun 2009 11:01:13 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <1243882320-sup-282@ntdws12.chass.utoronto.ca> (Ben Walton's message of "Mon\, 01 Jun 2009 14\:52\:50 -0400") References: <4A23F91B.1020803@opencsw.org> <1243882320-sup-282@ntdws12.chass.utoronto.ca> Message-ID: Ben Walton writes: > Excerpts from Peter FELECAN's message of Mon Jun 01 14:42:39 -0400 2009: > >> Sure. I'm not comfortable with that but why not? However, I'm >> flabbergasted: I thought that I knew how to do that... -B/opt/csw/bin >> or GCC_EXEC_PREFIX... none works! Still uses /usr/ccs/bin/ld > > I've seen some autoconf/configure scripts that accept the path to ld > as an option and don't work against some of the other things you've > tried. The autotools for the projects that I'm porting don't offer this kind of option. This is why I tried the 2 possibilities, the environment variable being the most adequate in my case. -- Peter From skayser at opencsw.org Tue Jun 2 11:05:22 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 02 Jun 2009 11:05:22 +0200 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <36490.131.203.105.13.1243925686.squirrel@webmail.jessies.org> References: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> <36490.131.203.105.13.1243925686.squirrel@webmail.jessies.org> Message-ID: <4A24EB52.3000206@opencsw.org> Hi Chris, Chris Reece wrote: > On Thu, May 28, 2009 9:12 am, Dagobert Michelsen wrote: > >> Chris: As you already did findutils with glocate, it would be >> really cool if you could also package up slocate >> as more secure alternative. > > I looked over the GAR documentation at the weekend and an hour or so later I > was none the wiser. thanks for your effort. Could you tell us, what exactly you were missing about the documentation so that we can improve on it? Any feedback appreciated (off-list is also fine if you prefer to). Dago has just recently written some updated information on how to set up the basic compilation / GAR environment [1,2] and how to set up build descriptions for simple packages [3]. Just this weekend, I moved away most of the outdated stuff from the main page, put Dago's documents first and tried to structure the main page a bit more. > I checked out a local copy on my one Sun box at home, > found what looks like an attempt at findutils in there, not mine. Didn't get > anywhere with it. > > In the interim, there's a traditionally-rolled, up to date package in testing: > > /home/testing/findutils-4.4.1,REV=2009.06.01-SunOS5.8-i386-CSW.pkg.gz > /home/testing/findutils-4.4.1,REV=2009.06.01-SunOS5.8-sparc-CSW.pkg.gz > > There's a postinstall script in there to add a gupdatedb job in the small > hours. This is the first time I've packaged in a long time, so I'd be happy > to receive feedback. I have no way to test the i386 package, in particular. Just installed it on an i386 box, looks mostly good to me, just two minor suggestions: * Could you please use /var/opt/csw instead of /opt/csw/var This has been changed a while ago http://opencsw.org/standards/layout * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to have it in there, just for the sake of standardization. /opt/csw/bin# dump -Lv glocate | grep R*PATH [5] RUNPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib [6] RPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib > As for slocate, I'm guessing that it's going to require chunks of GNUlib. > We're missing fts, for instance. I have submitted a rudimentary build description to GAR this weekend [1], but as the savannah.gnu.org git repository was down i couldn't get on with integrating the fts module (has quite a few dependencies to other GNUlib modules from what i saw). Do we have someone fluent in autotools and/or integrating GNUlib things? Are you, Chris? So far i had only integrated minor modules w/o dependencies, maybe this one needs some more fiddling with autotools. Sebastian [1] http://apps.sourceforge.net/trac/gar/wiki/GarSetup [2] http://apps.sourceforge.net/trac/gar/wiki/GarRepository [3] http://apps.sourceforge.net/trac/gar/wiki/GarPackagingTutorial From skayser at opencsw.org Tue Jun 2 11:13:56 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 02 Jun 2009 11:13:56 +0200 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <4A24EB52.3000206@opencsw.org> References: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> <36490.131.203.105.13.1243925686.squirrel@webmail.jessies.org> <4A24EB52.3000206@opencsw.org> Message-ID: <4A24ED54.4080705@opencsw.org> Sebastian Kayser wrote: > Chris Reece wrote: >> On Thu, May 28, 2009 9:12 am, Dagobert Michelsen wrote: >> >>> Chris: As you already did findutils with glocate, it would be >>> really cool if you could also package up slocate >>> as more secure alternative. >> I looked over the GAR documentation at the weekend and an hour or so later I >> was none the wiser. > > thanks for your effort. > > Could you tell us, what exactly you were missing about the documentation > so that we can improve on it? Any feedback appreciated (off-list is also > fine if you prefer to). On a related note. If you could share your build descriptions with us (mainly configure flags or any special handling required) i can try to submit them as a GARized version to the repository. This way, or we help you get started using GAR so that you can do it yourself :) Just let us know. Sebastian From pfelecan at opencsw.org Tue Jun 2 11:38:45 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 02 Jun 2009 11:38:45 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <4A2454EB.7000308@opencsw.org> (Mike Watters's message of "Mon\, 01 Jun 2009 17\:23\:39 -0500") References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> <4A2454EB.7000308@opencsw.org> Message-ID: Mike Watters writes: > Joerg Schilling wrote: >> Philip Brown wrote: >> >>> a WELL-BEHAVED set of Makefiles, will allow you to override linker by >>> simply specifying >>> >>> LD=/path/to/ld >> >> Well, if the "new" LD behaves significantly similar to the expected LD, >> you are correct. >> >> For simple makefiles you may be correct too...... >> >> For more complex makefiles you are not correct as these makefiles may >> have more internal knowledge than you may expect. >> >> J?rg >> > > It is all up to the writer of the Makefile, Makefiles can force whatever the > creator wants it to force by default. we always have the option of changing > it. either with a patch to Makefile.in or autoconf. however, if the package > uses libtool, the LD environment variable should be honored. The projects that I try to port/package use some autotools but not libtool --- which I can understand. Using the an environment variable LD implies the knowledge by the subsequent tools of this mechanism; it's true for libtool but not for the gcc driver. Here is some additional information: - the linking is done through the gcc driver, with classic argument passing by -Wl - originally, the linker arguments are from the super-set of the GNU -linker, e.g., soname, --version-script, &c - when asking to show the invoked sub-processes: /opt/csw/gcc4/bin/gcc -v -B /opt/csw/gnu t.c -o t we obtain, for the link phase: COLLECT_GCC_OPTIONS='-v' '-B' '/opt/csw/gnu' '-o' 't' '-mtune=generic' /usr/ccs/bin/ld -V ... ld: Software Generation Utilities - Solaris Link Editors: 5.10-1.490 which is not what we expect when using -B - note that when using gcc3 there is a subtle difference: /opt/csw/gcc3/bin/gcc -v -B /opt/csw/gnu t.c -o t ... /opt/csw/gcc3/libexec/gcc/i386-pc-solaris2.8/3.4.5/collect2 -V ... ld: Software Generation Utilities - Solaris Link Editors: 5.10-1.490 the linker is called by collect2; here I'm a little bit confused as this is the first time that I see a gcc shipped without the collect2 binary. But this is another issue which is probably not related. - exploring the compiler driver search paths: gcc -print-search-dirs | grep '^programs:' programs: =/opt/csw/gcc3/libexec/gcc/i386-pc-solaris2.8/3.4.5/:...:/usr/ccs/bin/ but gcc -B /opt/csw/gnu -print-search-dirs | grep '^programs:' programs: =/opt/csw/gnu/i386-pc-solaris2.8/3.4.5/:/opt/csw/gnu/:...:/usr/ccs/bin/ shows that this has no effect even though the path is modified as expected. - trussing the driver: truss -e -f -o gcc.truss gcc -v -B /opt/csw/gnu t.c -o t shows that there is no tentative to search for a linker binary in /opt/csw/gnu There is a remark in the GCC FAQ: http://gcc.gnu.org/faq.html#gas about using --with-ld when configuring the compiler: "One drawback of this option is that it won't allow you to override the search path for assembler and linker with command-line options -B/path/ if the specified file names exist." My experiments show that this is also true when trying to use the environment variable GCC_EXEC_PREFIX -- Peter From william at wbonnet.net Tue Jun 2 13:15:11 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 02 Jun 2009 13:15:11 +0200 Subject: [csw-maintainers] pixman 0.15.8 is in testing In-Reply-To: <4A1C518A.6090503@wbonnet.net> References: <4A1C518A.6090503@wbonnet.net> Message-ID: <4A2509BF.5020706@wbonnet.net> Hi Pixman package has been updated to latest version (0.15.8) and is available from testing Any feedback is welcome kind regards, W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From pfelecan at opencsw.org Tue Jun 2 20:36:45 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 02 Jun 2009 20:36:45 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: (Peter FELECAN's message of "Tue\, 02 Jun 2009 11\:38\:45 +0200") References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> <4A2454EB.7000308@opencsw.org> Message-ID: Just for the thread completeness and reference: I'm patching the Makefile --- brutal kludge --- to use explicitly gld on i386, thus having the best performance for the platform; for sparc, the project doesn't have assembler crafted code. -- Peter From dam at opencsw.org Tue Jun 2 23:19:19 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 2 Jun 2009 23:19:19 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> <4A2454EB.7000308@opencsw.org> Message-ID: <38AF7A48-097B-4F6E-8C7F-AFB40AAF7401@opencsw.org> Hi Peter, Am 02.06.2009 um 20:36 schrieb Peter FELECAN: > Just for the thread completeness and reference: I'm patching the > Makefile > --- brutal kludge --- to use explicitly gld on i386, thus having the > best performance for the platform; for sparc, the project doesn't have > assembler crafted code. Thanks for your effort, I have a similar problem with libgmp, which also has assembler optimizations for x86. Best regards -- Dago From james at opencsw.org Wed Jun 3 11:14:19 2009 From: james at opencsw.org (James Lee) Date: Wed, 03 Jun 2009 09:14:19 GMT Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <38AF7A48-097B-4F6E-8C7F-AFB40AAF7401@opencsw.org> References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> <4A2454EB.7000308@opencsw.org> <38AF7A48-097B-4F6E-8C7F-AFB40AAF7401@opencsw.org> Message-ID: <20090603.9141900.1268094198@gyor.oxdrove.co.uk> On 02/06/09, 22:19:19, Dagobert Michelsen wrote regarding Re: [csw-maintainers] assembler in shared libraries: > > --- brutal kludge --- to use explicitly gld on i386, thus having the > > best performance for the platform; for sparc, the project doesn't have > > assembler crafted code. > Thanks for your effort, I have a similar problem with libgmp, which > also has assembler optimizations for x86. Are you doing this because you believe it's better or because you proved it's better? I had a similar situation (with mesa3d) and, after fighting to get it to compile, the assembler was slower. James. From dam at opencsw.org Wed Jun 3 12:48:13 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 3 Jun 2009 12:48:13 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> Message-ID: Hi, Am 01.06.2009 um 12:46 schrieb Dagobert Michelsen: > Am 31.05.2009 um 23:00 schrieb Dagobert Michelsen: >> 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- >> packages/xcbgen/xtypes.py > > I guess this is the reason why the py-files aren't gettint compiled: > /opt/csw/X11/ > Should I put the py-files into /opt/csw/lib/python* instead > regardless of the > libdir setting? I relocated the files to /opt/csw/lib/python/site-packages/xcbgen Works like charm now. New packages in testing/: xcbproto-1.5,REV=2009.06.03-SunOS5.8-sparc-CSW.pkg.gz xcbproto-1.5,REV=2009.06.03-SunOS5.8-i386-CSW.pkg.gz Best regards -- Dago From pfelecan at opencsw.org Wed Jun 3 19:25:34 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 03 Jun 2009 19:25:34 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <20090603.9141900.1268094198@gyor.oxdrove.co.uk> (James Lee's message of "Wed\, 03 Jun 2009 09\:14\:19 GMT") References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> <4A2454EB.7000308@opencsw.org> <38AF7A48-097B-4F6E-8C7F-AFB40AAF7401@opencsw.org> <20090603.9141900.1268094198@gyor.oxdrove.co.uk> Message-ID: James Lee writes: > On 02/06/09, 22:19:19, Dagobert Michelsen wrote regarding > Re: [csw-maintainers] assembler in shared libraries: > >> > --- brutal kludge --- to use explicitly gld on i386, thus having the >> > best performance for the platform; for sparc, the project doesn't have >> > assembler crafted code. > >> Thanks for your effort, I have a similar problem with libgmp, which >> also has assembler optimizations for x86. > > Are you doing this because you believe it's better or because you > proved it's better? > > I had a similar situation (with mesa3d) and, after fighting to get > it to compile, the assembler was slower. As usual, James ask the good question. Trying to answer it by running a benchmark I discovered that the "brutal kludge" didn't work in a not so subtle way: - when linking, without error, with the static library, the execution halts with an "illegal instruction" in _rt_boot (real time library: using sched_yield, &c). - when linking with the shared library I get unresolved symbols; after exploring the library I conclude that there is a mistake of local instead of global for a bunch of symbols. After that, I repackage a version which didn't use assembler code, consequently being able to use the system's linker. None of the above phenomena appears and the benchmark programs ran with success. Lessons learned: 1. Test thoroughly your package 2. The GNU linker didn't work in some cases --- at least in this one 3. Sacrifice hypothetical gains for usability --- well, mine is an extreme case I'll explore the behaviour of the GNU linker as I have some more candidates with the issue that I tried to solve initially. -- Peter From bwalton at opencsw.org Thu Jun 4 02:16:19 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 03 Jun 2009 20:16:19 -0400 Subject: [csw-maintainers] cswpycompile Message-ID: <1244074532-sup-287@ntdws12.chass.utoronto.ca> Hi Mike and Peter, Any idea when the updated cswclassutils will hit the streets? I've got a package I'd like to leverage cswpycompile from. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Thu Jun 4 05:02:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 03 Jun 2009 22:02:51 -0500 Subject: [csw-maintainers] cswpycompile In-Reply-To: <1244074532-sup-287@ntdws12.chass.utoronto.ca> References: <1244074532-sup-287@ntdws12.chass.utoronto.ca> Message-ID: <4A27395B.3060504@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Hi Mike and Peter, > > Any idea when the updated cswclassutils will hit the streets? I've > got a package I'd like to leverage cswpycompile from. > > Thanks > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I have tested the fix for bugs 3633, 3634, Sebastian is testing out the fix for 3635, we need a tester for 3636. http://www.opencsw.org/mantis/view.php?id=3636 if you can test out the fix for that bug, we can get this rolled out faster. if there is no tester for 3636, I believe Peter is going to write a test case to test himself. it just may take longer then if another maintainer tests it. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkonOVsACgkQLrhmsXMSLxfsKwCgjKnDX415ybn56kadQ8jgaowl dO0AnjXj8mNGI1l5IfNjiuDuVAueIcgh =4qfW -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu Jun 4 05:17:30 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 03 Jun 2009 23:17:30 -0400 Subject: [csw-maintainers] cswpycompile In-Reply-To: <4A27395B.3060504@opencsw.org> References: <1244074532-sup-287@ntdws12.chass.utoronto.ca> <4A27395B.3060504@opencsw.org> Message-ID: <1244085411-sup-6497@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Wed Jun 03 23:02:51 -0400 2009: > I have tested the fix for bugs 3633, 3634, Sebastian is testing out > the fix for 3635, we need a tester for > 3636. http://www.opencsw.org/mantis/view.php?id=3636 if you can test > out the fix for that bug, we can get this rolled out faster. Ok, let me see what I can do over the next few days. > if there is no tester for 3636, I believe Peter is going to write a > test case to test himself. it just may take longer then if another > maintainer tests it. Of course. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Thu Jun 4 05:20:28 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 03 Jun 2009 23:20:28 -0400 Subject: [csw-maintainers] gitosis in testing Message-ID: <1244085481-sup-5308@ntdws12.chass.utoronto.ca> Hi All, I've rolled a package for gitosis[1], which is a cool way to host multiple git repositories[2]. I'm waiting for the cswpycompile addition to cswclassutils before I release it, so I can have the .py files handled automatically, but the base package seems pretty good at this point. It's available from testing/. Any feedback is welcome. Thanks -Ben [1] http://swik.net/gitosis [2] http://scie.nti.st/2007/11/14/hosting-git-repositories-the-easy-and-secure-way -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From car at opencsw.org Thu Jun 4 08:49:09 2009 From: car at opencsw.org (Chris Reece) Date: Thu, 4 Jun 2009 18:49:09 +1200 (NZST) Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm Message-ID: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> On Tue, June 2, 2009 9:05 pm, Sebastian Kayser wrote: > Could you tell us, what exactly you were missing about the documentation > so that we can improve on it? Any feedback appreciated (off-list is also fine > if you prefer to). Specifically, no. I recall being daunted by finding three versions of GAR, none of which appeared to have a full set of GAR newbie documentation. I found a "Getting started with GAR" document that, depending on which link I followed to it, from where: was not found; contained a page blank except for "redirect"; or was there, but was listed under some manner of "this is for an older GAR -- here be dragons" warning. I was in a hurry, may have missed something and, believe me, I know what it's like to not have enough hours in the day. Documentation's no damned fun. Your suggestion of hand-holding to en-GAR findutils sounds like a fine idea, if you've the time. I'd rather put these things in a format that they'll be useful, for what that's worth. But if I am just being dense, feel free to point me at the docs that you know tell me everything I need and tell me to RTFM. > Just installed it on an i386 box, looks mostly good to me, just two > minor suggestions: > > * Could you please use /var/opt/csw instead of /opt/csw/var > This has been changed a while ago http://opencsw.org/standards/layout I considered this, and thought better of it because I was worried about people upgrading from the current (!) and previous versions, that put locatedb in /opt/csw/var. But, thinking about it now, it's as simple as testing for the presence of the old file at install time and moving it. I'll do that. > * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to have > it in there, just for the sake of standardization. > /opt/csw/bin# dump -Lv glocate | grep R*PATH > [5] RUNPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib > [6] RPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib I set LD_OPTIONS as per OpenCSW documentation, like so: LD_OPTIONS='-R/opt/csw/lib/$ISALIST -R/opt/csw/lib -L/opt/csw/lib' I wondered if I'd done something silly like forgotten that I'd set LDFLAGS too, but a clean build just now repeats the error, even before it's staged: -bash-4.0$ dump -Lv locate/locate | grep R*PATH [5] RUNPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib [6] RPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib Environment looks like this: CPPFLAGS=-I/opt/csw/include LD_OPTIONS=-R/opt/csw/lib/$ISALIST -R/opt/csw/lib -L/opt/csw/lib PATH=/opt/studio/SOS11/SUNWspro/bin:/usr/bin:/bin:/usr/sbin:/sbin:/opt/csw/bin:/usr/ccs/bin:/usr/ucb MAKE=gmake CFLAGS=-fast -xarch=v8 -xnolibmopt CC=cc If I set: LD_OPTIONS='-R/opt/csw/lib -L/opt/csw/lib' as I used to do, then I get: [5] RUNPATH /opt/csw/lib [6] RPATH /opt/csw/lib If I set: LD_OPTIONS='-R/opt/csw/lib/$ISALIST -L/opt/csw/lib' I get: -bash-4.0$ dump -Lv locate/locate | grep R*PATH [5] RUNPATH /opt/csw/lib/$ISALIST:/opt/csw/lib [6] RPATH /opt/csw/lib/$ISALIST:/opt/csw/lib which I assume is what we're aiming for. > Are you, Chris? Not at all. Sorry. Cheers, Chris. From bonivart at opencsw.org Thu Jun 4 09:24:59 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 4 Jun 2009 09:24:59 +0200 Subject: [csw-maintainers] Bug in createpkg Message-ID: <625385e30906040024k635899c7wddb4c90f1457e697@mail.gmail.com> I just manually put together a test package and got this: "ERROR: fix csw/man entry in prototype file" I assume this is supposed to catch using csw/man paths instead of csw/share/doc but I have this in the prototype: "f none /etc/opt/csw/manifest 0644 root bin" This is the line in createpkg that triggers: "grep 'csw/man' $PROTO && die fix csw/man entry in $PROTO file" There's a bunch of similar lines, they should use word boundaries with RE's so not to be so easily fooled. -- /peter From bonivart at opencsw.org Thu Jun 4 10:06:59 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 4 Jun 2009 10:06:59 +0200 Subject: [csw-maintainers] cswpycompile In-Reply-To: <4A27395B.3060504@opencsw.org> References: <1244074532-sup-287@ntdws12.chass.utoronto.ca> <4A27395B.3060504@opencsw.org> Message-ID: <625385e30906040106w2a078e72gc14ace3b1b82fb1b@mail.gmail.com> On Thu, Jun 4, 2009 at 5:02 AM, Mike Watters wrote: > I have tested the fix for bugs 3633, 3634, Sebastian is testing out the fix for > 3635, we need a tester for 3636. http://www.opencsw.org/mantis/view.php?id=3636 > if you can test out the fix for that bug, we can get this rolled out faster. > > if there is no tester for 3636, I believe Peter is going to write a test case > to test himself. ?it just may take longer then if another maintainer tests it. I have now tested 3636 (custom manifest) and it works. Sebastian has tested everything about 3635 until the last part and I just tested that myself and it works as well. I'm releasing 1.16. -- /peter From bwalton at opencsw.org Thu Jun 4 14:59:01 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 04 Jun 2009 08:59:01 -0400 Subject: [csw-maintainers] cswpycompile In-Reply-To: <625385e30906040106w2a078e72gc14ace3b1b82fb1b@mail.gmail.com> References: <1244074532-sup-287@ntdws12.chass.utoronto.ca> <4A27395B.3060504@opencsw.org> <625385e30906040106w2a078e72gc14ace3b1b82fb1b@mail.gmail.com> Message-ID: <1244120299-sup-8752@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Thu Jun 04 04:06:59 -0400 2009: Hi Peter, > I have now tested 3636 (custom manifest) and it works. Sebastian has > tested everything about 3635 until the last part and I just tested > that myself and it works as well. I'm releasing 1.16. That's great! Thanks for letting me know. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Thu Jun 4 15:27:49 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 04 Jun 2009 08:27:49 -0500 Subject: [csw-maintainers] cswpycompile In-Reply-To: <1244120299-sup-8752@ntdws12.chass.utoronto.ca> References: <1244074532-sup-287@ntdws12.chass.utoronto.ca> <4A27395B.3060504@opencsw.org> <625385e30906040106w2a078e72gc14ace3b1b82fb1b@mail.gmail.com> <1244120299-sup-8752@ntdws12.chass.utoronto.ca> Message-ID: <4A27CBD5.7000902@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Peter Bonivart's message of Thu Jun 04 04:06:59 -0400 2009: > > Hi Peter, > >> I have now tested 3636 (custom manifest) and it works. Sebastian has >> tested everything about 3635 until the last part and I just tested >> that myself and it works as well. I'm releasing 1.16. > > That's great! Thanks for letting me know. > Excellent, I am releasing python - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkony9UACgkQLrhmsXMSLxecwACg3cknLgPFCJoIwGp5afmb3yjJ TscAoNWqYfaTRitqyQ+3WJ1SO+z7TcQh =x0hO -----END PGP SIGNATURE----- From phil at bolthole.com Thu Jun 4 18:04:53 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 4 Jun 2009 09:04:53 -0700 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> References: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> Message-ID: <20090604160453.GB2284@bolthole.com> On Thu, Jun 04, 2009 at 06:49:09PM +1200, Chris Reece wrote: > > * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to have > > it in there, just for the sake of standardization. > > > /opt/csw/bin# dump -Lv glocate | grep R*PATH > > [5] RUNPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib > > [6] RPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib having both RUNPATH and RPATH in there is normal, for some reason. dont worry about it. From phil at bolthole.com Thu Jun 4 18:20:19 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 4 Jun 2009 09:20:19 -0700 Subject: [csw-maintainers] Bug in createpkg In-Reply-To: <625385e30906040024k635899c7wddb4c90f1457e697@mail.gmail.com> References: <625385e30906040024k635899c7wddb4c90f1457e697@mail.gmail.com> Message-ID: <20090604162019.GH2284@bolthole.com> On Thu, Jun 04, 2009 at 09:24:59AM +0200, Peter Bonivart wrote: > I just manually put together a test package and got this: > > "ERROR: fix csw/man entry in prototype file" > > I assume this is supposed to catch using csw/man paths instead of > csw/share/doc but I have this in the prototype: > > "f none /etc/opt/csw/manifest 0644 root bin" ahaha. ok, thanks for debugging that :-) From skayser at opencsw.org Thu Jun 4 18:39:34 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 4 Jun 2009 18:39:34 +0200 (CEST) Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <20090604160453.GB2284@bolthole.com> References: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> <20090604160453.GB2284@bolthole.com> Message-ID: <54487.194.246.122.22.1244133574.squirrel@ssl.skayser.de> Philip Brown wrote: > On Thu, Jun 04, 2009 at 06:49:09PM +1200, Chris Reece wrote: >> > * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to >> have >> > it in there, just for the sake of standardization. >> >> > /opt/csw/bin# dump -Lv glocate | grep R*PATH >> > [5] RUNPATH >> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib >> > [6] RPATH >> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib > > having both RUNPATH and RPATH in there is normal, for some reason. dont > worry about it. IIRC RPATH had been used in the past, but was considered evil as it had priority over LD_LIBRARY_PATH and you could not override it. That's why RUNPATH had been introduced. - RPATH (deprecated, not used) - LD_LIBRARY_PATH - RUNPATH - crle settings I would suppose RPATH is still in there for compatibility purposes. Sebastian From phil at bolthole.com Fri Jun 5 00:30:51 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 4 Jun 2009 15:30:51 -0700 Subject: [csw-maintainers] /opt/csw/X11 notes Message-ID: <20090604223051.GB97249@bolthole.com> FYI: with William and Dagobert now releasing packages for it, I have added a note about /opt/csw/X11, to http://www.opencsw.org/standards/layout , so everyone will have a clear idea of their purpose. The note is short enough that I will just replicate it here, to save people some effort: [/opt/csw/] X11 This is for our own packaged X11 libs, loosely equivalent to /usr/X11R6 on other systems. Up until 2009, we have relied upon the system level X11 libraries shipped with Solaris. In most cases, we wish to continue to do so. However, there are some recent applications that require modern versions of X11 libraries. We make them available in this separate subtree, so that applications will not accidentally pull them in at compile time, but default to the system ones. That way, our builds will only pull in our own X11 libs as dependancies, if the maintainer decides it is absolutely required. From mwatters at opencsw.org Fri Jun 5 16:23:00 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 09:23:00 -0500 Subject: [csw-maintainers] pysetuptools now in testing Message-ID: <4A292A44.8000305@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: package now compiles libraries on install. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopKkQACgkQLrhmsXMSLxcuyQCgn/ntf9ps/pH9B7IFq6RN34Qc DD8AniRWnQrttctHMqeB/iigEHK6W4fW =RXm7 -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri Jun 5 16:26:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 09:26:22 -0500 Subject: [csw-maintainers] pydocutils now in testing Message-ID: <4A292B0E.6030808@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: now compiles bytecode upon install. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopKw4ACgkQLrhmsXMSLxeH3wCcDcgl/72RjLWqrLRoqvnUAjBo D/wAoLfThqmtMdSdZghKbSvr+hCtmtZ1 =jc5W -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri Jun 5 16:50:09 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 09:50:09 -0500 Subject: [csw-maintainers] ap2_modpython now in testing Message-ID: <4A2930A1.2040708@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: now compiles bytecode upon install. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopMKEACgkQLrhmsXMSLxeYSgCgiO7bGsfKKQuLAjZ4S2WY7/5j WS4AoMIXA1LS00cbvjQPNSM0/pjxZKxG =xsu2 -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri Jun 5 16:50:53 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 09:50:53 -0500 Subject: [csw-maintainers] mysql-python now in testing Message-ID: <4A2930CD.4000109@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: compiles bytecode on install - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopMM0ACgkQLrhmsXMSLxeENQCbBprvZkRZG+VQsClCRDMWteS0 az4AoNyg+QENACSZgq18rn86eF/I3PRU =jmo4 -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri Jun 5 16:59:43 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 09:59:43 -0500 Subject: [csw-maintainers] pysqlite now in testing Message-ID: <4A2932DF.906@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: location and bytecode - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopMt8ACgkQLrhmsXMSLxeBDQCfaRy2tPEWHg0h7qGXQlNv4Um0 jkEAoOmNZPvWwJh3OI+Hwa5I8pM5qsEc =7Y2h -----END PGP SIGNATURE----- From skayser at opencsw.org Fri Jun 5 18:27:40 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 5 Jun 2009 18:27:40 +0200 (CEST) Subject: [csw-maintainers] Jumpstart post-install recipe for OpenCSW Message-ID: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> Hi, does someone have a working "install me some OpenCSW packages" Jumpstart post-install snippet that he could pass to me? I am in the process of setting up my first Jumpstart server and i would like to have it install a list of CSW pkgs after setting up the machine. Is this known to work with the various postinstall scripts (or cswinitsmf) that we have in the packages? And out of curiosity: What are the specifics to watch out for so that a postinstall script will work in a Jumpstart environment, just $PKG_INSTALL_ROOT [1] or are there any other caveats that pop up frequently? Sebastian [1] http://www.opencsw.org/standards/pkgcreation From mwatters at opencsw.org Fri Jun 5 18:34:00 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 11:34:00 -0500 Subject: [csw-maintainers] SilverCity now in testing Message-ID: <4A2948F8.6050900@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: location and no more bytecode in package. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopSPgACgkQLrhmsXMSLxc7vQCfdytvq+opa5cf1yeHtO3vsniF PcAAoJpEZhcvhe2PmTwBFESNvnGHH0qk =5G7x -----END PGP SIGNATURE----- From phil at bolthole.com Fri Jun 5 18:36:52 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 5 Jun 2009 09:36:52 -0700 Subject: [csw-maintainers] Jumpstart post-install recipe for OpenCSW In-Reply-To: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> References: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> Message-ID: <20090605163652.GC73123@bolthole.com> On Fri, Jun 05, 2009 at 06:27:40PM +0200, Sebastian Kayser wrote: > Hi, > > does someone have a working "install me some OpenCSW packages" Jumpstart > post-install snippet that he could pass to me? I am in the process of > setting up my first Jumpstart server and i would like to have it install a > list of CSW pkgs after setting up the machine. > > Is this known to work with the various postinstall scripts (or cswinitsmf) > that we have in the packages? If it doesnt, its a bug in the postinstall script. > And out of curiosity: What are the specifics to watch out for so that a > postinstall script will work in a Jumpstart environment, just > $PKG_INSTALL_ROOT [1] or are there any other caveats that pop up > frequently? That, and also the issue that you need to be aware that you may not be running on a "live" system. EG: If PKG_INSTALL_ROOT != /, then you cant run smf related commands, since there is no svc.configd running. which kinda sucks. and no you cant run svc.config chrooted, btw. not that i've found anyway. So that sort of stuff needs to be added post-reboot at the moment. * * * * This is why it's important that people use the class utility scripts to add in smf support. So that when and if we figure out improvements in this area, everyone gets to use it without having to rebuild their packages. ADDITIONALLY... it might behoove us to formalize some kind of "post reboot helper script" hook for our packages. That way, we could have those packages/classes that need it, have a specific integrated way to register "run this after reboot", like IPS allows. From mwatters at opencsw.org Fri Jun 5 18:46:16 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 11:46:16 -0500 Subject: [csw-maintainers] Genshi now in testing Message-ID: <4A294BD8.7090706@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 comments welcome: - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopS9cACgkQLrhmsXMSLxcJ+gCg3wwAfQL/hb7WK4nq/al3/gTJ ODwAoNz9oxZY5GAdrvTlkSZRArXT/NbH =DN1A -----END PGP SIGNATURE----- From Darin.Perusich at cognigencorp.com Fri Jun 5 19:01:15 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Fri, 05 Jun 2009 13:01:15 -0400 Subject: [csw-maintainers] Jumpstart post-install recipe for OpenCSW In-Reply-To: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> References: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> Message-ID: <4A294F5B.1070905@cognigencorp.com> Sebastian Kayser wrote: > Hi, > > does someone have a working "install me some OpenCSW packages" Jumpstart > post-install snippet that he could pass to me? I am in the process of > setting up my first Jumpstart server and i would like to have it install a > list of CSW pkgs after setting up the machine. I attached my CSW package installation script. It's nothing fancy but get the job done for me. > Is this known to work with the various postinstall scripts (or cswinitsmf) > that we have in the packages? > > And out of curiosity: What are the specifics to watch out for so that a > postinstall script will work in a Jumpstart environment, just > $PKG_INSTALL_ROOT [1] or are there any other caveats that pop up > frequently? If you chroot the package installation you don't need to worry about $PKG_INSTALL_ROOT being set. I use chroot to install everything except pkg-get and I've never had a problem. Granted I only install a small number of package, sudo, amanda, etc, and build the servers out afterwards. If it's a machine I need to make a ton of clones of I set it up and create flash images of anyways. The biggest issues I've come across was pre/post scripts which ask questions and wait for a response. Kind of defeats the purpose of hands off installs right. I haven't run across this in years but I wanted to mention it as something to watch out for. -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: cswpkginst URL: From mwatters at opencsw.org Fri Jun 5 19:01:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 12:01:26 -0500 Subject: [csw-maintainers] pygobject now in testing Message-ID: <4A294F66.9030506@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: compiled against glib2-2.20.0,REV=2009.04.08 and python 2.6.2 feedback welcome. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopT2YACgkQLrhmsXMSLxf1RgCg4ysLMCvkV4xLZ+sev8bdwyoR jx0AoKyQuGtm+aTuUzU/0T7FSw5MN34M =VFBF -----END PGP SIGNATURE----- From phil at bolthole.com Fri Jun 5 19:13:00 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 5 Jun 2009 10:13:00 -0700 Subject: [csw-maintainers] Jumpstart post-install recipe for OpenCSW In-Reply-To: <4A294F5B.1070905@cognigencorp.com> References: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> <4A294F5B.1070905@cognigencorp.com> Message-ID: <20090605171300.GA54426@bolthole.com> Side comment: On Fri, Jun 05, 2009 at 01:01:15PM -0400, Darin Perusich wrote: > > for ii in $PKGLIST > do > chroot /a /opt/csw/bin/pkg-get install $ii > done > This is mildly inefficient. You could just have a single line, chroot /a /opt/csw/bin/pkg-get install $PKGLIST :-) From mwatters at opencsw.org Fri Jun 5 20:42:47 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 13:42:47 -0500 Subject: [csw-maintainers] dbus-python now in testing Message-ID: <4A296727.707@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 feedback welcome! - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopZycACgkQLrhmsXMSLxeU7ACfa6L5nLeEuSN+I+XF/jq72CBa HacAn2U+2UIv3DXk1l1hs+0p+L3QWl/o =DVxp -----END PGP SIGNATURE----- From bwalton at opencsw.org Fri Jun 5 23:21:13 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 05 Jun 2009 17:21:13 -0400 Subject: [csw-maintainers] pysetuptools Message-ID: <1244236697-sup-875@ntdws12.chass.utoronto.ca> Hi Mike, I'm confirming functionality of the updated pysetuptools for you. I can build and install the gitosis package with it. Thanks for the nice work! :) -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From jeff at cjsa.com Fri Jun 5 23:32:44 2009 From: jeff at cjsa.com (Jeffery Small) Date: Fri, 5 Jun 2009 21:32:44 GMT Subject: [csw-maintainers] dbus-python now in testing References: <4A296727.707@opencsw.org> Message-ID: Speaking of dbus, was there any consensus on the status of the dbus service? Can anyone successfully disable the svc:/system/cswdbus:default service on their system with the command: svcadm disable system/cswdbus:default I'm still looking for a resolution to the problem of why my system will not halt successfully. Regards, -- Jeff C. Jeffery Small CJSA LLC 206-232-3338 jeff at cjsa.com 7000 E Mercer Way, Mercer Island, WA 98040 From william at wbonnet.net Sat Jun 6 23:32:18 2009 From: william at wbonnet.net (William Bonnet) Date: Sat, 06 Jun 2009 23:32:18 +0200 Subject: [csw-maintainers] libXdcmp now in testing Message-ID: <4A2AE062.3070707@wbonnet.net> Hi libXdmcp and libXdmcpdevel are now available from testing (version 1.0.2) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From bonivart at opencsw.org Mon Jun 8 11:02:10 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 8 Jun 2009 11:02:10 +0200 Subject: [csw-maintainers] Fwd: [csw-users] I don't know what pkg or pkgs are broken.... In-Reply-To: <200906080849.n588nXJn004314@dfki.uni-kl.de> References: <4A243C2F.5070307@ernieball.com> <200906080849.n588nXJn004314@dfki.uni-kl.de> Message-ID: <625385e30906080202q5c9fda76p89c28dd62256dd04@mail.gmail.com> This seems like a simple init script bug..? William? ---------- Forwarded message ---------- From: Nicolai Schwindt Date: Mon, Jun 8, 2009 at 10:49 AM Subject: Re: [csw-users] I don't know what pkg or pkgs are broken.... To: questions and discussions > I think I recall the same error on Friday: > /var/opt/csw/svc/method/svc-cswdbus: smf_is_nonglobalzone: not found I filed that as bug report in mantis aboput a month ago. If I had not been on vacation I could have saved you some time .) Nicolai _______________________________________________ users mailing list users at lists.opencsw.org https://lists.opencsw.org/mailman/listinfo/users -- /peter From dam at opencsw.org Mon Jun 8 11:16:01 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 8 Jun 2009 11:16:01 +0200 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> Message-ID: <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> Hi, this bug just has been submitted: Anfang der weitergeleiteten E-Mail: > The following issue has been SUBMITTED. ... > ====================================================================== > Summary: libintl > Description: > After upgrading a bunch of CSW-packages binaries linked to libintl > starts > to crash due to a nonexistant symbol > > fatal: relocation error: file /opt/csw/lib/libintl.so.8: symbol > libiconv_set_relocation_prefix: referenced symbol not found > ====================================================================== This looks pretty bad. However, most of the apps still seems to function... Best regards -- Dago From maciej at opencsw.org Mon Jun 8 16:40:25 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Mon, 8 Jun 2009 15:40:25 +0100 Subject: [csw-maintainers] A backed up file in $FILEDIR triggers the 'uncommitted' warning Message-ID: mGAR currently adds 'UNCOMMITTED' when a package is built from sources which are not committed to the repository. When building packages, I often issue 'gmake clean' to make sure that the package is being built from scratch. This deletes everything from downloads/ - all source files need to be downloaded again. To avoid pulling sources multiple times, I copy files from downloads/ to files/ - 'gmake clean' doesn't remove them. In this way, I can rebuild packages from scratch without re-downloading everything. However, an uncommitted file in $FILEDIR causes mGAR to think that these files (tarballs, usually) are some uncommitted material. Perhaps the rule for stuff in files/ could be: 1. If there are changes in the checksums file, it's uncommitted. 2. If there are no changes in the checksums file, and the actual checksums in files/ don't match, it's uncommitted 3. If there are no changes in the checksums file, and the actual checksums in files/ do match, it is not uncommitted, even though some files aren't added to the repository. ...or perhaps there is a way of cleaning the build directory without deleting downloads? Maciej From bwalton at opencsw.org Mon Jun 8 16:49:54 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 08 Jun 2009 10:49:54 -0400 Subject: [csw-maintainers] A backed up file in $FILEDIR triggers the 'uncommitted' warning In-Reply-To: References: Message-ID: <1244472436-sup-6646@ntdws12.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizinski's message of Mon Jun 08 10:40:25 -0400 2009: > mGAR currently adds 'UNCOMMITTED' when a package is built from sources > which are not committed to the repository. When building packages, I > often issue 'gmake clean' to make sure that the package is being built > from scratch. This deletes everything from downloads/ - all source gmake garchive That will stash downloads in your configured source directory. The convention on the buildfarm is that /home/src is used. > files need to be downloaded again. To avoid pulling sources multiple > times, I copy files from downloads/ to files/ - 'gmake clean' doesn't > remove them. In this way, I can rebuild packages from scratch without > re-downloading everything. However, an uncommitted file in $FILEDIR > causes mGAR to think that these files (tarballs, usually) are some > uncommitted material. If you rely on archive, this mostly goes away...everything else in the gar directory should be of the type that you don't mind committing. Does that help? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Mon Jun 8 17:07:15 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 8 Jun 2009 17:07:15 +0200 Subject: [csw-maintainers] New in testing: BerkeleyDB 4.7.25 Message-ID: Hi, I just copied bdb 4.7.25 in testing/: berkeleydb-4.7.25,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz berkeleydb_devel-4.7.25,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz berkeleydb-4.7.25,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz berkeleydb_devel-4.7.25,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz berkeleydb_doc-4.7.25,REV=2009.06.08-SunOS5.8-all-CSW.pkg.gz It has a different layout than the existing bdb4* packages by putting the libs directly in /opt/csw/lib. The last package that did that was bdb 3.3.11 (CSWbdb3), so the current package is incompatible with that (not flagged yet in depend). There are still some packages left which rely on bdb3: - CSWapache - CSWcfengine - CSWwebalizer - CSWdsniff - CSWruby - CSWkdesdk - CSWkdevelop - CSWpostfix I guess the new bdb package should include the old libs for compatibility, CSWbdb3 should then be empty and depend on the new package. Best regards -- Dago From dam at opencsw.org Mon Jun 8 17:11:38 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 8 Jun 2009 17:11:38 +0200 Subject: [csw-maintainers] A backed up file in $FILEDIR triggers the 'uncommitted' warning In-Reply-To: References: Message-ID: <1DE0AB09-5A42-45FC-8858-9F42EC653F7F@opencsw.org> Hi Maciej, Am 08.06.2009 um 16:40 schrieb Maciej (Matchek) Blizinski: > Perhaps the rule for stuff in files/ could be: > > 1. If there are changes in the checksums file, it's uncommitted. > > 2. If there are no changes in the checksums file, and the actual > checksums in files/ don't match, it's uncommitted > > 3. If there are no changes in the checksums file, and the actual > checksums in files/ do match, it is not uncommitted, even though some > files aren't added to the repository. Well, the current rule is "everything which is neither committed not ignored through svn:ignore triggers UNCOMMITTED". Apart from that Ben describes the solution for you problem :-) Best regards -- Dago From ihsan at opencsw.org Mon Jun 8 22:27:12 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 08 Jun 2009 22:27:12 +0200 Subject: [csw-maintainers] test Message-ID: <4A2D7420.8050804@opencsw.org> please ignore -- ihsan at dogan.ch http://blog.dogan.ch/ From hson at opencsw.org Mon Jun 8 22:37:38 2009 From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=) Date: Mon, 08 Jun 2009 22:37:38 +0200 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl In-Reply-To: <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> Message-ID: <4A2D7692.3060902@opencsw.org> Dagobert Michelsen wrote: > Hi, > > this bug just has been submitted: > > > > Anfang der weitergeleiteten E-Mail: >> The following issue has been SUBMITTED. > ... >> ====================================================================== >> Summary: libintl >> Description: >> After upgrading a bunch of CSW-packages binaries linked to libintl starts >> to crash due to a nonexistant symbol >> >> fatal: relocation error: file /opt/csw/lib/libintl.so.8: symbol >> libiconv_set_relocation_prefix: referenced symbol not found >> ====================================================================== > > This looks pretty bad. However, most of the apps still seems to function... > I'm the one who filed the bug report and the problem isn't that the apps doesn't start, but that they crashes after a while in some specific places (within a function call to libexif which is linked to libintl and libiconv). However, a recompile of the app seem to fix the problem. But the question is if there is some option when building libintl whether to include libiconv_set_relocation_prefix or not... (Btw, I'm updating the bug report with this info) From maciej at opencsw.org Mon Jun 8 22:41:14 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Mon, 8 Jun 2009 21:41:14 +0100 Subject: [csw-maintainers] A backed up file in $FILEDIR triggers the 'uncommitted' warning In-Reply-To: <1244472436-sup-6646@ntdws12.chass.utoronto.ca> References: <1244472436-sup-6646@ntdws12.chass.utoronto.ca> Message-ID: On Mon, Jun 8, 2009 at 3:49 PM, Ben Walton wrote: > gmake garchive > > That will stash downloads in your configured source directory. The > convention on the buildfarm is that /home/src is used. I see, cool. Will use that. I was unaware of the gar archive concept. Thanks! From phil at bolthole.com Mon Jun 8 22:48:48 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 8 Jun 2009 13:48:48 -0700 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl In-Reply-To: <4A2D7692.3060902@opencsw.org> References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> <4A2D7692.3060902@opencsw.org> Message-ID: <20090608204848.GA35058@bolthole.com> On Mon, Jun 08, 2009 at 10:37:38PM +0200, Roger H?kansson wrote: > But the question is if there is some option when building libintl whether > to include libiconv_set_relocation_prefix or not... > yes there is. buried in lib/Makefile appears -Dset_relocation_prefix=libiconv_set_relocation_prefix \ From phil at bolthole.com Mon Jun 8 22:51:20 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 8 Jun 2009 13:51:20 -0700 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl In-Reply-To: <4A2D7692.3060902@opencsw.org> References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> <4A2D7692.3060902@opencsw.org> Message-ID: <20090608205120.GB35058@bolthole.com> PS: i should mention that while i am currently the last person to provide the libiconv package, it is essentially orphaned from me. multiple people have "offered" to take it over, and even done some amount of work (in gar) about it. However, no-one has actually DELIVERED a new package. For many weeks now. If you care strongly about this issue, you have your choice of waiting for other people to finish it, or submitting a new package yourself. The first person who submits a fully working package, will become the new maintainer. From harpchad at opencsw.org Mon Jun 8 22:56:18 2009 From: harpchad at opencsw.org (Chad Harp) Date: Mon, 08 Jun 2009 15:56:18 -0500 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl In-Reply-To: <4A2D7692.3060902@opencsw.org> References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> <4A2D7692.3060902@opencsw.org> Message-ID: <4A2D7AF2.3050105@opencsw.org> Which version of libintl are the failing applications linked against? There are currently three active shared libraries, only libintl.so.8 provides libiconv_set_relocation_prefix: $ nm libintl.so.2 | grep libiconv_set_relocation_prefix $ nm libintl.so.3 | grep libiconv_set_relocation_prefix $ nm libintl.so.8 | grep libiconv_set_relocation_prefix [186] | 0| 0|FUNC |GLOB |0 |UNDEF |libiconv_set_relocation_prefix Can you provide a specific test case so I can replicate the problem? (Application, use case, etc) Thanks, Chad From dam at opencsw.org Mon Jun 8 23:07:57 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 8 Jun 2009 23:07:57 +0200 Subject: [csw-maintainers] New in testing: libidl 0.8.13 Message-ID: Hi, new in testing: libidl-0.8.13,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz libidl-0.8.13,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz The following packages rely on it: CSWmozilla CSWorbit2 CSWpyorbit CSWthunderbird CSWfirefox CSWseamonkey CSWsunbird I need it as dependency to orbit2. Please see if you can try it with one of the apps so I can release it and go ahead with the gnome stuff. Best regards -- Dago From dam at opencsw.org Mon Jun 8 23:30:21 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 8 Jun 2009 23:30:21 +0200 Subject: [csw-maintainers] New in testing: libxrender Message-ID: <39724A18-D4F3-46D0-842D-E398B9621EC3@opencsw.org> Hi, a new libxrender is ready in testing/: libxrender-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz libxrender-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz renderdev-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz renderdev-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz Please give it a try, it is a prerequisite on the long way to gnome. Best regards -- Dago From phil at bolthole.com Mon Jun 8 23:43:01 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 8 Jun 2009 14:43:01 -0700 Subject: [csw-maintainers] New in testing: libxrender In-Reply-To: <39724A18-D4F3-46D0-842D-E398B9621EC3@opencsw.org> References: <39724A18-D4F3-46D0-842D-E398B9621EC3@opencsw.org> Message-ID: <20090608214301.GB94732@bolthole.com> On Mon, Jun 08, 2009 at 11:30:21PM +0200, Dagobert Michelsen wrote: > Hi, > > a new libxrender is ready in testing/: > > libxrender-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz > libxrender-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz > renderdev-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz > renderdev-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz > > Please give it a try, it is a prerequisite on the long way to gnome. > i think we should rename renderdev to libxrender_devel. since it hsould have no dependancies, the rename is not as much an issue as in other cases. From car at opencsw.org Tue Jun 9 07:55:27 2009 From: car at opencsw.org (Chris Reece) Date: Tue, 9 Jun 2009 17:55:27 +1200 (NZST) Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <54487.194.246.122.22.1244133574.squirrel@ssl.skayser.de> References: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> <20090604160453.GB2284@bolthole.com> <54487.194.246.122.22.1244133574.squirrel@ssl.skayser.de> Message-ID: <41354.131.203.105.13.1244526927.squirrel@webmail.jessies.org> On Fri, June 5, 2009 4:39 am, Sebastian Kayser wrote: > Philip Brown wrote: >> On Thu, Jun 04, 2009 at 06:49:09PM +1200, Chris Reece wrote: >> >>>> * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to >>>> >>> have >>>> it in there, just for the sake of standardization. >>> >>>> /opt/csw/bin# dump -Lv glocate | grep R*PATH >>>> [5] RUNPATH >>>> >>> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib >>> >>>> [6] RPATH >>>> >>> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib >>> >> >> having both RUNPATH and RPATH in there is normal, for some reason. dont >> worry about it. > > IIRC RPATH had been used in the past, but was considered evil as it had > priority over LD_LIBRARY_PATH and you could not override it. That's why RUNPATH > had been introduced. > > - RPATH (deprecated, not used) > - LD_LIBRARY_PATH > - RUNPATH > - crle settings > > > I would suppose RPATH is still in there for compatibility purposes. I think we may be talking at crossed purposes. I was under the impression that you were commenting on the duplication of "/opt/csw/lib". I was saying that this appears to be down to my following the documented OpenCSW standards. I may be wrong. No matter, there are new pacakges: /home/testing/findutils-4.4.2,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz /home/testing/findutils-4.4.2,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz Which are both more up to date (4.4.2 was released in the interim) and should address the points raised. Cheers, Chris. From hson at opencsw.org Tue Jun 9 10:57:53 2009 From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=) Date: Tue, 09 Jun 2009 10:57:53 +0200 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl In-Reply-To: <4A2D7AF2.3050105@opencsw.org> References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> <4A2D7692.3060902@opencsw.org> <4A2D7AF2.3050105@opencsw.org> Message-ID: <4A2E2411.5080102@opencsw.org> Chad Harp wrote: > Which version of libintl are the failing applications linked against? My applications are linked to libintl.so.8 > There are currently three active shared libraries, only libintl.so.8 > provides libiconv_set_relocation_prefix: > > $ nm libintl.so.2 | grep libiconv_set_relocation_prefix > $ nm libintl.so.3 | grep libiconv_set_relocation_prefix > $ nm libintl.so.8 | grep libiconv_set_relocation_prefix > [186] | 0| 0|FUNC |GLOB |0 |UNDEF > |libiconv_set_relocation_prefix > Well, libintl.so.8 doesn't provide libiconv_set_relocation_prefix but have it as a undefined symbol, i.e to be provided by some other library. > Can you provide a specific test case so I can replicate the problem? > (Application, use case, etc) Unfortunately my applications needs a Oracle database and a bunch of data in it to get to the part of the code where there is a problem. From skayser at opencsw.org Tue Jun 9 14:31:15 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 09 Jun 2009 14:31:15 +0200 Subject: [csw-maintainers] mantis config change? In-Reply-To: <49B5ACD5.6010502@opencsw.org> References: <1236642753-sup-662@ntdws12.chass.utoronto.ca> <49B5ACD5.6010502@opencsw.org> Message-ID: <4A2E5613.8050408@opencsw.org> Mike Watters wrote: > Ben Walton wrote: >> Is there an option in Mantis to have only the most recent note and/or >> status change be displayed in the email? I find the output of >> concatenated notes rather nasty (and most times not required). >> >> Would others be in favour of this change if the option exists? > > +1 > I have to agree with you Ben, most of the time you only need the last > comment. and since the email contains a direct link to the mantis bug, > the entire chain IMO is not needed. It has been a while since this was requested and i am happy to say that it is now implemented. It is actually a user-specific setting (My account -> Preferences -> E-mail Notes Limit) which wasn't used by anyone until now. I have now set this to 1 for all existing users and adjusted the configuration so that new users will also have this set to 1. In case someone wants to go back to the old "all bugnotes in a notification" behavior, just set the value to 0. I have also dropped the bug history from the bottom of the notifications. This way there is no distracting text below the bug note; the bug note (the essential part of the notification) can thus be easily identified. For an example bug note, see http://lists.opencsw.org/pipermail/bug-notifications/2009-June/001324.html Thoughts, ideas? Just let me know. Sebastian From skayser at opencsw.org Tue Jun 9 14:40:14 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 09 Jun 2009 14:40:14 +0200 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <41354.131.203.105.13.1244526927.squirrel@webmail.jessies.org> References: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> <20090604160453.GB2284@bolthole.com> <54487.194.246.122.22.1244133574.squirrel@ssl.skayser.de> <41354.131.203.105.13.1244526927.squirrel@webmail.jessies.org> Message-ID: <4A2E582E.1010608@opencsw.org> Chris Reece wrote: > On Fri, June 5, 2009 4:39 am, Sebastian Kayser wrote: >> Philip Brown wrote: >>> On Thu, Jun 04, 2009 at 06:49:09PM +1200, Chris Reece wrote: >>> >>>>> * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to >>>>> >>>> have >>>>> it in there, just for the sake of standardization. >>>>> /opt/csw/bin# dump -Lv glocate | grep R*PATH >>>>> [5] RUNPATH >>>>> >>>> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib >>>> >>>>> [6] RPATH >>>>> >>>> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib >>>> >>> having both RUNPATH and RPATH in there is normal, for some reason. dont >>> worry about it. >> IIRC RPATH had been used in the past, but was considered evil as it had >> priority over LD_LIBRARY_PATH and you could not override it. That's why RUNPATH >> had been introduced. >> >> - RPATH (deprecated, not used) >> - LD_LIBRARY_PATH >> - RUNPATH >> - crle settings >> >> >> I would suppose RPATH is still in there for compatibility purposes. > > I think we may be talking at crossed purposes. I was under the impression > that you were commenting on the duplication of "/opt/csw/lib". I was saying > that this appears to be down to my following the documented OpenCSW standards. > I may be wrong. Sorry Chris for getting you confused here. I was indeed only after the duplicate /opt/csw/lib when commenting on your package (and that is what needs to be taken care of). The RPATH/RUNPATH thingie was just in response to Phil's mail. Sebastian From bwalton at opencsw.org Tue Jun 9 15:05:07 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 09 Jun 2009 09:05:07 -0400 Subject: [csw-maintainers] mantis config change? In-Reply-To: <4A2E5613.8050408@opencsw.org> References: <1236642753-sup-662@ntdws12.chass.utoronto.ca> <49B5ACD5.6010502@opencsw.org> <4A2E5613.8050408@opencsw.org> Message-ID: <1244552681-sup-9540@ntdws12.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Tue Jun 09 08:31:15 -0400 2009: > http://lists.opencsw.org/pipermail/bug-notifications/2009-June/001324.html > > Thoughts, ideas? Just let me know. This is _much_ better! Thanks. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From skayser at opencsw.org Tue Jun 9 16:45:36 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 09 Jun 2009 16:45:36 +0200 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> References: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> Message-ID: <4A2E7590.5040000@opencsw.org> Chris Reece wrote: > On Tue, June 2, 2009 9:05 pm, Sebastian Kayser wrote: >> Could you tell us, what exactly you were missing about the documentation >> so that we can improve on it? Any feedback appreciated (off-list is also fine >> if you prefer to). > > Specifically, no. I recall being daunted by finding three versions of GAR, > none of which appeared to have a full set of GAR newbie documentation. I > found a "Getting started with GAR" document that, depending on which link I > followed to it, from where: was not found; contained a page blank except for > "redirect"; or was there, but was listed under some manner of "this is for an > older GAR -- here be dragons" warning. > > I was in a hurry, may have missed something and, believe me, I know what it's > like to not have enough hours in the day. Documentation's no damned fun. > > Your suggestion of hand-holding to en-GAR findutils sounds like a fine idea, > if you've the time. I'd rather put these things in a format that they'll be > useful, for what that's worth. > > But if I am just being dense, feel free to point me at the docs that you know > tell me everything I need and tell me to RTFM. A couple of options. Option 1 -------- Send me your current build script (or instructions on how to follow the build process you are currently using). I will then GAR-ize and submit it to the repository so you can have a look at it and maybe tweak it a bit. Doing this for others i have seen that this seems to give people a pretty good idea of how GAR works. You can then GAR-ize your remaining packages. Option 2 -------- If you want to learn GAR by yourself, follow the steps below. You might take shortcuts if some points seem obvious to you. 1) Read [1] for setting up your local build enviroment. 2) Check out gperf from the repository which is a known to work, basic build-description. Then build it along with reading the build explanation. [2,3] 3) Do the same for dtracetoolkit to see how different packaging requirements are handled. [4] 4) Peek into the GAR variable reference along with its referenced build descriptions to see how other packages (with varying complexity) are built. [5] Btw. this is roughly the way that the GAR wiki main page [6] is currently structured. Option 3 -------- I will hang out on our IRC channel this evening (#opencsw on freenode.net, starting ~8pm CEST). Come by if i you prefer some Q&A. :) Sebastian [1] http://apps.sf.net/trac/gar/wiki/GarSetup [2] http://apps.sf.net/trac/gar/wiki/GarRepository#Checkingoutonepackage [3] http://apps.sf.net/trac/gar/wiki/ExplainedGperf [4] http://apps.sf.net/trac/gar/wiki/ExplainedDtracetoolkit [5] http://apps.sf.net/trac/gar/wiki/GAR%20Variable%20Reference [6] http://gar.sf.net From dam at opencsw.org Tue Jun 9 17:12:13 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 9 Jun 2009 17:12:13 +0200 Subject: [csw-maintainers] New in testing: libxrender In-Reply-To: <20090608214301.GB94732@bolthole.com> References: <39724A18-D4F3-46D0-842D-E398B9621EC3@opencsw.org> <20090608214301.GB94732@bolthole.com> Message-ID: <4299CDC0-CB4F-4C57-99FC-356BF5C42012@opencsw.org> Hi Phil, Am 08.06.2009 um 23:43 schrieb Philip Brown: > On Mon, Jun 08, 2009 at 11:30:21PM +0200, Dagobert Michelsen wrote: >> a new libxrender is ready in testing/: >> >> libxrender-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz >> libxrender-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz >> renderdev-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz >> renderdev-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz >> >> Please give it a try, it is a prerequisite on the long way to gnome. >> > > i think we should rename renderdev to libxrender_devel. > > since it hsould have no dependancies, the rename is not as much an > issue > as in other cases. Sure. New packages in testing/: libxrender-0.9.4,REV=2009.06.09-SunOS5.8-sparc-CSW.pkg.gz libxrender_devel-0.9.4,REV=2009.06.09-SunOS5.8-sparc-CSW.pkg.gz libxrender-0.9.4,REV=2009.06.09-SunOS5.8-i386-CSW.pkg.gz libxrender_devel-0.9.4,REV=2009.06.09-SunOS5.8-i386-CSW.pkg.gz Best regards -- Dago From phil at bolthole.com Tue Jun 9 17:41:39 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 9 Jun 2009 08:41:39 -0700 Subject: [csw-maintainers] Poll on our future X11 library usage Message-ID: <20090609154139.GK35058@bolthole.com> hi folks, a few maintainers have been discussing what to do about X11 libs, going forward. I just shared the following poll to our user list. Please also weigh in. http://doodle.com/73mdviypknk63y7e Here's the full body of the email i sent to the users list, if you are curious, and not on the list: ------------------------------------------------------------ Hello CSW users, This is a very important poll, for the future of X11 support in CSW packages going forward. We are faced with a tough technical decision to make. While the implementation details are technical, the major impact is more user based. So, we are asking you, the users, to tell us what you would prefer! For 6 years now, we have based virtually all our X11 dependant programs, on the X11 libraries supplied by Solaris itself. Unfortunately, the latest versions of gtk, which are required by the latest versions of things like firefox, etc. require newer versions of basic things like libX11 and so on, than older versions of Solaris provide. We are obviously going to provide firefox. So we are also going to provide the required libraries in one way or another. The question then becomes; from this point on, do we have ALL X11 programs use the CSW X11 libraries? Or do we have only the programs that "need" the new stuff, use the new stuff in /opt/csw/X11, whereas others that are less fancy (ie: not gtk based) use the sun libs in /usr/openwin/lib So far, there is some inclination to no longer use the Sun libs. This would mean that, if you still use some sun graphical programs, you would potentially have a bit more redundancy in the shared library department. The "good news", is that it should only amount to around 4 megabytes of extra ram used up this way, vs using sun X11 libraries. That's total extra; it's a one-time hit, not a per-process cost. Please let us know what you, our users, would prefer. We've set up a poll for this issue, at http://doodle.com/73mdviypknk63y7e From bonivart at opencsw.org Tue Jun 9 17:53:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 9 Jun 2009 17:53:14 +0200 Subject: [csw-maintainers] Poll on our future X11 library usage In-Reply-To: <20090609154139.GK35058@bolthole.com> References: <20090609154139.GK35058@bolthole.com> Message-ID: <625385e30906090853s4f5227c2s9d0e7f656e9eecbb@mail.gmail.com> On Tue, Jun 9, 2009 at 5:41 PM, Philip Brown wrote: > Unfortunately, the latest versions of gtk, which are required by the latest > versions of things like firefox, etc. require newer versions of basic > things like libX11 and so on, than older versions of Solaris provide. Is Solaris 9 too old to build, e.g., Firefox on? -- /peter From william at wbonnet.net Tue Jun 9 21:49:26 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 09 Jun 2009 21:49:26 +0200 Subject: [csw-maintainers] Poll on our future X11 library usage In-Reply-To: <625385e30906090853s4f5227c2s9d0e7f656e9eecbb@mail.gmail.com> References: <20090609154139.GK35058@bolthole.com> <625385e30906090853s4f5227c2s9d0e7f656e9eecbb@mail.gmail.com> Message-ID: <4A2EBCC6.8090303@wbonnet.net> Hi Peter > On Tue, Jun 9, 2009 at 5:41 PM, Philip Brown wrote: > >> Unfortunately, the latest versions of gtk, which are required by the latest >> versions of things like firefox, etc. require newer versions of basic >> things like libX11 and so on, than older versions of Solaris provide. >> > > Is Solaris 9 too old to build, e.g., Firefox on? > I'm not really sure, i haven't tried since i focussed on Solaris 8. I think it is too old, but i have to double check if it you need an answer to this. cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From dam at opencsw.org Wed Jun 10 18:48:58 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 10 Jun 2009 18:48:58 +0200 Subject: [csw-maintainers] Alignment-problem on libmhash References: <20090610162618.5AEFA5C4D@mail.opencsw.org> Message-ID: Hi, I have an alignment problem with libmhash on sparcv9. It looks like this: Checking fragmentation capabilities of HAVAL256: FAIL: frag_test build8s% cd work/build-isa-sparcv9/mhash-0.9.9.9/src/ build8s% mdb .libs/ driver frag_test hmac_test keygen_test rest_test build8s% mdb .libs/frag_test core mdb: warning: core file is from SunOS 5.10 Generic_138888-07; shared text mappings may not match installed libraries Loading modules: [ ] > ::stack libmhash.so.2`mutils_word32nswap+0xb8(100104990, 20, 1f, 6c24cc00, 1, 20754a1616aa27a) libmhash.so.2`havalTransform3+0xc(1001047d4, ffffffff7fffefcd, 100104880, 143367, a19b3906, 1f3213955f32139) libmhash.so.2`havalUpdate+0x130(1001047d0, ffffffff7fffefcc, 81, ffffffffffffffff, 1001047f4, 408) libmhash.so.2`mhash+0x14(100101fd0, ffffffff7fffefcc, 81, fffffffffffffffd, 0, 7ffeff021eb8) frag_test+0x98(100, 81, 1, 958, ffffffff7fffefcc, 80) main+0x8c(1c, 7ffeff1329ce, 1, 100101620, 100001038, 100101588) _start+0x17c(0, ffffffff7ffff238, 0, 1, 7ffeffe0d6cc, 7ffefff10000) > ::dis mutils_word32nswap libmhash.so.2`mutils_word32nswap: save %sp, -0xb0, %sp libmhash.so.2`mutils_word32nswap+4: btst 0xff, %i2 libmhash.so.2`mutils_word32nswap+8: bne,pn %icc,+0x24 libmhash.so.2`mutils_word32nswap+0xc: mov %i0, %l4 libmhash.so.2`mutils_word32nswap+0x10: sll %i1, 2, %l6 libmhash.so.2`mutils_word32nswap+0x14: call +0x12d494 <0x7ffeff131aa0> libmhash.so.2`mutils_word32nswap+0x18: srl %l6, 0, %o0 libmhash.so.2`mutils_word32nswap+0x1c: brnz,pn %o0, +0x14 libmhash.so.2`mutils_word32nswap+0x20: mov %o0, %i0 libmhash.so.2`mutils_word32nswap+0x24: ret libmhash.so.2`mutils_word32nswap+0x28: restore %g0, 0, %o0 libmhash.so.2`mutils_word32nswap+0x2c: mov %l4, %i0 libmhash.so.2`mutils_word32nswap+0x30: cmp %i1, 0 libmhash.so.2`mutils_word32nswap+0x34: bleu,pn %icc,+0x264 libmhash.so.2`mutils_word32nswap+0x38: cmp %i1, 4 libmhash.so.2`mutils_word32nswap+0x3c: bl,pn %icc,+0x200 libmhash.so.2`mutils_word32nswap+0x40: mov %i0, %l6 libmhash.so.2`mutils_word32nswap+0x44: sra %i1, 0, %o0 libmhash.so.2`mutils_word32nswap+0x48: sub %i0, %l4, %l7 libmhash.so.2`mutils_word32nswap+0x4c: srax %l7, 0x3f, %g4 libmhash.so.2`mutils_word32nswap+0x50: sethi %hi(0x7ffffc00), %l1 libmhash.so.2`mutils_word32nswap+0x54: srl %i1, 0, %l2 libmhash.so.2`mutils_word32nswap+0x58: xor %l7, %g4, %g5 libmhash.so.2`mutils_word32nswap+0x5c: sllx %o0, 2, %g3 libmhash.so.2`mutils_word32nswap+0x60: add %l1, 0x3fe, %l0 libmhash.so.2`mutils_word32nswap+0x64: sub %g5, %g4, %g2 libmhash.so.2`mutils_word32nswap+0x68: sub %l2, %l0, %o2 libmhash.so.2`mutils_word32nswap+0x6c: srlx %o2, 0x3f, %o1 libmhash.so.2`mutils_word32nswap+0x70: cmp %g2, %g3 libmhash.so.2`mutils_word32nswap+0x74: clr %g1 libmhash.so.2`mutils_word32nswap+0x78: and %o1, 1, %i4 libmhash.so.2`mutils_word32nswap+0x7c: movge %xcc,1, %g1 libmhash.so.2`mutils_word32nswap+0x80: btst %g1, %i4 libmhash.so.2`mutils_word32nswap+0x84: be,a,pn %icc,+0x1bc libmhash.so.2`mutils_word32nswap+0x88: clr %l2 libmhash.so.2`mutils_word32nswap+0x8c: sethi %hi(0xfc00), %l5 libmhash.so.2`mutils_word32nswap+0x90: add %i1, 0, %g4 libmhash.so.2`mutils_word32nswap+0x94: add %l5, 0x300, %l3 libmhash.so.2`mutils_word32nswap+0x98: add %i1, -1, %i2 libmhash.so.2`mutils_word32nswap+0x9c: clr %l1 libmhash.so.2`mutils_word32nswap+0xa0: sethi %hi(0xff000000), %l2 libmhash.so.2`mutils_word32nswap+0xa4: sethi %hi(0xff0000), %l5 libmhash.so.2`mutils_word32nswap+0xa8: cmp %g4, 5 libmhash.so.2`mutils_word32nswap+0xac: bl,pn %icc,+0x13c libmhash.so.2`mutils_word32nswap+0xb0: sub %i2, 3, %o7 libmhash.so.2`mutils_word32nswap+0xb4: add %l1, 1, %l1 libmhash.so.2`mutils_word32nswap+0xb8: ld [%l4], %g5 libmhash.so.2`mutils_word32nswap+0xbc: add %l4, 4, %l4 libmhash.so.2`mutils_word32nswap+0xc0: and %g5, %l5, %o1 > ::regs %g0 = 0x0000000000000000 %l0 = 0x000000007ffffffe %g1 = 0x0000000000000001 %l1 = 0x0000000000000001 %g2 = 0x00000001801059c3 %l2 = 0x00000000ff000000 %g3 = 0x0000000000000080 %l3 = 0x000000000000ff00 %g4 = 0x0000000000000020 %l4 = 0xffffffff7fffefcd %g5 = 0x00000001801059c3 %l5 = 0x0000000000ff0000 %g6 = 0x0000000000000000 %l6 = 0x0000000100104990 %g7 = 0x0000000000000000 %l7 = 0x00000001801059c3 %o0 = 0x0000000000000020 %i0 = 0x0000000100104990 %o1 = 0x0000000000000001 %i1 = 0x0000000000000020 %o2 = 0xffffffff80000022 %i2 = 0x000000000000001f %o3 = 0x000000007f7e7e7e %i3 = 0x000000006c24cc00 %o4 = 0x00000000007e0000 %i4 = 0x0000000000000001 %o5 = 0x0000000100104990 %i5 = 0x020754a1616aa27a %o6 = 0xffffffff7fffe151 %i6 = 0xffffffff7fffe201 %o7 = 0x000000000000001c %i7 = 0x00007ffeff0190ec libmhash.so.2`havalTransform3+0xc %ccr = 0x00 xcc=nzvc icc=nzvc %y = 0x0000000000000000 %pc = 0x00007ffeff0046b0 libmhash.so.2`mutils_word32nswap+0xb8 %npc = 0x00007ffeff0046b4 libmhash.so.2`mutils_word32nswap+0xbc %sp = 0xffffffff7fffe151 %fp = 0xffffffff7fffe201 %asi = 0x82 %fprs = 0x07 So, the problem is that 0xffffffff7fffefcd is not 32 bit aligned, but I can't see it in the source why this should be the case. Maybe I'm just too blind today and someone other can have a look? Everything has been committed to pkg/libmhash. Best regards -- Dago From rmacduff at opencsw.org Wed Jun 10 19:03:43 2009 From: rmacduff at opencsw.org (Ross Macduff) Date: Wed, 10 Jun 2009 13:03:43 -0400 Subject: [csw-maintainers] gsed 4.2 in testing Message-ID: <1244653354-sup-1474@frog.chass.utoronto.ca> Hello, gsed 4.2 is in testing: gsed-4.2,REV=2009.06.10-SunOS5.8-i386-CSW.pkg.gz gsed-4.2,REV=2009.06.10-SunOS5.8-sparc-CSW.pkg.gz Best, Ross From dam at opencsw.org Wed Jun 10 22:29:03 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 10 Jun 2009 22:29:03 +0200 Subject: [csw-maintainers] gsed 4.2 in testing In-Reply-To: <1244653354-sup-1474@frog.chass.utoronto.ca> References: <1244653354-sup-1474@frog.chass.utoronto.ca> Message-ID: <87834F2F-1E74-437F-BEC0-2186A0EC2653@opencsw.org> Hi Ross, Am 10.06.2009 um 19:03 schrieb Ross Macduff: > gsed 4.2 is in testing: > > gsed-4.2,REV=2009.06.10-SunOS5.8-i386-CSW.pkg.gz > gsed-4.2,REV=2009.06.10-SunOS5.8-sparc-CSW.pkg.gz Looks good. Thanks for the quick response! Best regards -- Dago From mwatters at opencsw.org Thu Jun 11 06:59:42 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 10 Jun 2009 23:59:42 -0500 Subject: [csw-maintainers] another bug in upstream. Message-ID: <4A308F3E.9020306@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I received an upgrade notice for GD ( pasted below ) I am the GD maintainer, however all the links to the download location are for the pm_gd to which I am not the maintainer. ( I believe it was Alex who is retired.) Further investigation showed me that I do in fact have the latest "stable" release of GD packaged, but the perl module is what is in need of upgrade. GD-2.43.tar.gz is the filename of pm_gd gd-2.0.35.tar.gz is the filename of gd note the regex for upstream watch should be the same. I have no solution off hand to this bug, but I will take a look when I get a few spare cycles. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Hello dear GD maintainer, The upstream notification job has detected the availability of new files for GD. The following upstream file(s): GD-2.43.tar.gz is/are available at the following url(s): http://search.cpan.org/CPAN/authors/id/L/LD/LDS/ ftp://ftp.nrc.ca/pub/CPAN/authors/id/L/LD/LDS/ ftp://ftp.nas.nasa.gov/pub/perl/CPAN/authors/id/L/LD/LDS/ http://mirrors.ibiblio.org/pub/mirrors/CPAN/authors/id/L/LD/LDS/ ftp://cpan.pair.com/pub/CPAN/authors/id/L/LD/LDS/ http://mirrors.kernel.org/cpan/authors/id/L/LD/LDS/ Please consider updating your package. - -- Kindest regards upstream notification job - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkowjz4ACgkQLrhmsXMSLxcZzQCgjY0dRLLSaGvrVnFFd13FG7kH 458AoJkUUd5vdjv9sis7uhxniiNbNZRY =fC8f -----END PGP SIGNATURE----- From bonivart at opencsw.org Thu Jun 11 15:13:43 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 11 Jun 2009 15:13:43 +0200 Subject: [csw-maintainers] /testing clamav 0.95.2 Message-ID: <625385e30906110613v59a347f1xd24f33ae77e2c406@mail.gmail.com> A new version of Clam AntiVirus is in testing. http://mirror.opencsw.org/testing/clamav-0.95.2,REV=2009.06.11-SunOS5.8-sparc-CSW.pkg.gz http://mirror.opencsw.org/testing/clamav-0.95.2,REV=2009.06.11-SunOS5.8-i386-CSW.pkg.gz http://mirror.opencsw.org/testing/libclamav-0.95.2,REV=2009.06.11-SunOS5.8-sparc-CSW.pkg.gz http://mirror.opencsw.org/testing/libclamav-0.95.2,REV=2009.06.11-SunOS5.8-i386-CSW.pkg.gz -- /peter From skayser at opencsw.org Fri Jun 12 01:55:49 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 12 Jun 2009 01:55:49 +0200 Subject: [csw-maintainers] Bugtracker (Mantis) patched for performance, feedback welcome Message-ID: <4A319985.8040904@opencsw.org> Hi everyone, i have patched a Mantis test installation today to improve the long page load times we have been seeing. A couple of features that we don't use as of yet (sub-projects, private projects, product versions) have been de-facto deactivated as a result of this. The changes are targeted at data display only and should be non-disruptive. The patched version is available at [1], if some of you guys could have a look and check whether something obvious feels wrong to you i would appreciate it. Heads up!: it works with our live bug data and sends out bug update notifications, so don't just close or delete random bugs. I will be on vacation until Sunday. Pending some feedback and finishing touches i aim to patch the productive installation early next week. If you spot pages that still feel overly sluggish to you, let me know. Note that this is a first attempt at improving performance. In case we might need one of the now missing features, we can always revisit the changes. Sebastian [1] http://opencsw.org/mantis-test/ From bwalton at opencsw.org Fri Jun 12 02:04:48 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 11 Jun 2009 20:04:48 -0400 Subject: [csw-maintainers] Bugtracker (Mantis) patched for performance, feedback welcome In-Reply-To: <4A319985.8040904@opencsw.org> References: <4A319985.8040904@opencsw.org> Message-ID: <1244765051-sup-8822@ntdws12.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Thu Jun 11 19:55:49 -0400 2009: > i have patched a Mantis test installation today to improve the long page > load times we have been seeing. A couple of features that we don't use > as of yet (sub-projects, private projects, product versions) have been > de-facto deactivated as a result of this. The changes are targeted at > data display only and should be non-disruptive. +1. It looks good to me and it's a _lot_ faster. Nice work! :) Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From ihsan at dogan.ch Fri Jun 12 16:38:16 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Fri, 12 Jun 2009 16:38:16 +0200 Subject: [csw-maintainers] Bugtracker (Mantis) patched for performance, feedback welcome In-Reply-To: <1244765051-sup-8822@ntdws12.chass.utoronto.ca> References: <4A319985.8040904@opencsw.org> <1244765051-sup-8822@ntdws12.chass.utoronto.ca> Message-ID: <4A326858.4010405@dogan.ch> Ben Walton wrote: >> i have patched a Mantis test installation today to improve the long page >> load times we have been seeing. A couple of features that we don't use >> as of yet (sub-projects, private projects, product versions) have been >> de-facto deactivated as a result of this. The changes are targeted at >> data display only and should be non-disruptive. > > +1. It looks good to me and it's a _lot_ faster. Nice work! :) Wow, it's a rocket now. Thanks Sebastian. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at dogan.ch Fri Jun 12 16:44:11 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Fri, 12 Jun 2009 16:44:11 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp 2009 In-Reply-To: <4A11B3AF.7090709@opencsw.org> References: <4A11B3AF.7090709@opencsw.org> Message-ID: <4A3269BB.4090604@dogan.ch> Ihsan Dogan wrote: > The OpenCSW Summer Camp 2009 will take place in Oslo/Norway. To figure > out when it should take place, I've create a Doodle poll. Please fill > out the link bellow: > > http://www.doodle.com/6zr9kmdr8p9vv69p The summer camp will take place on the weekend of 22/23 August 2009 in Olso/Norway. Thanks to all, who participated with the poll. @Trygve: Could you please arrange the location and provide a list of recommended hotels? Next task: - Define the topics... Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Fri Jun 12 17:41:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 12 Jun 2009 11:41:47 -0400 Subject: [csw-maintainers] exim maintainer Message-ID: <1244821262-sup-9225@ntdws12.chass.utoronto.ca> Hi Markus, Are you still out there? If so, we've got a bug to file against the exim package. If not, we may consider taking it over. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From rmacduff at opencsw.org Fri Jun 12 18:10:08 2009 From: rmacduff at opencsw.org (Ross Macduff) Date: Fri, 12 Jun 2009 12:10:08 -0400 Subject: [csw-maintainers] Exim and SMF Message-ID: <1244821559-sup-5458@frog.chass.utoronto.ca> Hello, During a recent install of Exim on a Solaris 10 box it was noticed that the postinstall script doesn't register it with SMF. It looks like this is because we have autoenable_deamons set to 0 in our csw.conf. It would be nice to have it register the service, yet leave it disabled. Is this something that could be done? Would making use of cswclassutils solve this problem? Is this worthy of a bug report? Thanks, Ross From phil at bolthole.com Fri Jun 12 18:20:44 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 12 Jun 2009 09:20:44 -0700 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <1244821559-sup-5458@frog.chass.utoronto.ca> References: <1244821559-sup-5458@frog.chass.utoronto.ca> Message-ID: <20090612162044.GB93732@bolthole.com> On Fri, Jun 12, 2009 at 12:10:08PM -0400, Ross Macduff wrote: > Hello, > > During a recent install of Exim on a Solaris 10 box it was noticed that > the postinstall script doesn't register it with SMF. It looks like this > is because we have autoenable_deamons set to 0 in our csw.conf. It > would be nice to have it register the service, yet leave it disabled. > Ummmm.. out of curiosity.. WHY would that be nice? From rmacduff at opencsw.org Fri Jun 12 18:32:08 2009 From: rmacduff at opencsw.org (Ross Macduff) Date: Fri, 12 Jun 2009 12:32:08 -0400 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <20090612162044.GB93732@bolthole.com> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> Message-ID: <1244823762-sup-4624@frog.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 12 12:20:44 -0400 2009: > On Fri, Jun 12, 2009 at 12:10:08PM -0400, Ross Macduff wrote: > > Hello, > > > > During a recent install of Exim on a Solaris 10 box it was noticed that > > the postinstall script doesn't register it with SMF. It looks like this > > is because we have autoenable_deamons set to 0 in our csw.conf. It > > would be nice to have it register the service, yet leave it disabled. > > > > Ummmm.. out of curiosity.. WHY would that be nice? > Perhaps it isn't desirable to have it start up right after installing it, yet you don't want to go through the steps necessary to register it with SMF (I know it's only one step...). That way, it is only a matter of enabling it when it's ready to go. It seems that cswclassutils has this type of behaviour in mind based on what I read here: http://wiki.opencsw.org/cswclassutils-package#toc3 So if use_smf isn't set to 'no' and autoenable_daemons is set to 'no', then it won't start but is ready to be enabled. Best, Ross From trygvis at opencsw.org Fri Jun 12 18:42:08 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Fri, 12 Jun 2009 18:42:08 +0200 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <20090612162044.GB93732@bolthole.com> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> Message-ID: <4A328560.4070805@opencsw.org> Philip Brown wrote: > On Fri, Jun 12, 2009 at 12:10:08PM -0400, Ross Macduff wrote: >> Hello, >> >> During a recent install of Exim on a Solaris 10 box it was noticed that >> the postinstall script doesn't register it with SMF. It looks like this >> is because we have autoenable_deamons set to 0 in our csw.conf. It >> would be nice to have it register the service, yet leave it disabled. >> > > Ummmm.. out of curiosity.. WHY would that be nice? It would be nice to have it registered so that all the user would have to do was to enable it, just like the regular Solaris services. -- Trygve From trygvis at opencsw.org Fri Jun 12 18:43:39 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Fri, 12 Jun 2009 18:43:39 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp 2009 In-Reply-To: <4A3269BB.4090604@dogan.ch> References: <4A11B3AF.7090709@opencsw.org> <4A3269BB.4090604@dogan.ch> Message-ID: <4A3285BB.4000705@opencsw.org> Ihsan Dogan wrote: > Ihsan Dogan wrote: > >> The OpenCSW Summer Camp 2009 will take place in Oslo/Norway. To figure >> out when it should take place, I've create a Doodle poll. Please fill >> out the link bellow: >> >> http://www.doodle.com/6zr9kmdr8p9vv69p > > The summer camp will take place on the weekend of 22/23 August 2009 in > Olso/Norway. Thanks to all, who participated with the poll. > > @Trygve: > Could you please arrange the location and provide a list of recommended > hotels? I'll try to ask around next week. I think I've found a nice place for us to work during the day too. The place is the same that the JUG that I lead use so it should be good enough for us :) -- Trygve From bwalton at opencsw.org Fri Jun 12 18:47:10 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 12 Jun 2009 12:47:10 -0400 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <4A328560.4070805@opencsw.org> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <4A328560.4070805@opencsw.org> Message-ID: <1244825151-sup-6189@ntdws12.chass.utoronto.ca> Excerpts from Trygve Laugst?l's message of Fri Jun 12 12:42:08 -0400 2009: > > Ummmm.. out of curiosity.. WHY would that be nice? > > It would be nice to have it registered so that all the user would have > to do was to enable it, just like the regular Solaris services. Yes, this is our position. If the package is updated, it should be moved to use class utils, which would get this behaviour 'for free.' This was the reason for my ping to Markus earlier today. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From phil at bolthole.com Fri Jun 12 19:11:26 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 12 Jun 2009 10:11:26 -0700 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <1244823762-sup-4624@frog.chass.utoronto.ca> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> Message-ID: <20090612171126.GE95242@bolthole.com> On Fri, Jun 12, 2009 at 12:32:08PM -0400, Ross Macduff wrote: > Excerpts from Philip Brown's message of Fri Jun 12 12:20:44 -0400 2009: > > Ummmm.. out of curiosity.. WHY would that be nice? > > > > Perhaps it isn't desirable to have it start up right after installing > it, yet you don't want to go through the steps necessary to register it > with SMF (I know it's only one step...). That way, it is only a matter > of enabling it when it's ready to go. Ah, got it. So to phrase it more in the general case: When something requires hand-chosen configuration, it can be nice to be able to install a buncha things in batch mode, but actually configure it to start properly, later, when batch install is all done. However, for simple services, its not so much of an issue. Either you want it started, or you dont. From bwalton at opencsw.org Fri Jun 12 19:19:12 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 12 Jun 2009 13:19:12 -0400 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <20090612171126.GE95242@bolthole.com> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> <20090612171126.GE95242@bolthole.com> Message-ID: <1244827079-sup-1294@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 12 13:11:26 -0400 2009: > However, for simple services, its not so much of an issue. > Either you want it started, or you dont. And in either case, it should be one command away...svcadm or /etc/init.d/. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 12 19:22:16 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 12 Jun 2009 10:22:16 -0700 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <1244827079-sup-1294@ntdws12.chass.utoronto.ca> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> <20090612171126.GE95242@bolthole.com> <1244827079-sup-1294@ntdws12.chass.utoronto.ca> Message-ID: <20090612172216.GH95242@bolthole.com> On Fri, Jun 12, 2009 at 01:19:12PM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Fri Jun 12 13:11:26 -0400 2009: > > However, for simple services, its not so much of an issue. > > Either you want it started, or you dont. > > And in either case, it should be one command away...svcadm or > /etc/init.d/. yeah.... what I think would benefit us, would be if the cswinitsmf class, came with an additional util script, to enable things after the fact. From bwalton at opencsw.org Fri Jun 12 21:19:08 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 12 Jun 2009 15:19:08 -0400 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <20090612172216.GH95242@bolthole.com> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> <20090612171126.GE95242@bolthole.com> <1244827079-sup-1294@ntdws12.chass.utoronto.ca> <20090612172216.GH95242@bolthole.com> Message-ID: <1244834220-sup-7183@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 12 13:22:16 -0400 2009: > yeah.... what I think would benefit us, would be if the cswinitsmf class, > came with an additional util script, to enable things after the fact. It doesn't really need to, unless I'm misunderstanding you. After cswinitsmf is done its job, the service can be started with a simple command after placing appropriate config files in place. As far as I can see, the manifest gets imported and any init scripts installed. The state of the service after that depends on the csw.conf settings. Am I missing your point? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 12 22:24:27 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 12 Jun 2009 13:24:27 -0700 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <1244834220-sup-7183@ntdws12.chass.utoronto.ca> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> <20090612171126.GE95242@bolthole.com> <1244827079-sup-1294@ntdws12.chass.utoronto.ca> <20090612172216.GH95242@bolthole.com> <1244834220-sup-7183@ntdws12.chass.utoronto.ca> Message-ID: <20090612202427.GI95242@bolthole.com> On Fri, Jun 12, 2009 at 03:19:08PM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Fri Jun 12 13:22:16 -0400 2009: > > yeah.... what I think would benefit us, would be if the cswinitsmf class, > > came with an additional util script, to enable things after the fact. > > It doesn't really need to, unless I'm misunderstanding you. After > cswinitsmf is done its job, the service can be started with a simple > command after placing appropriate config files in place. As far as I > can see, the manifest gets imported and any init scripts installed. > The state of the service after that depends on the csw.conf settings. > > Am I missing your point? yes, i think so :-) such a script has multiple uses. One,is like the exim case mentioned. Otheres, are when you nfs-share out /opt/csw. Or, replicate it. Most people on this list wouldnt think of doing that sort of thing. But a lot of large places, DO do that sort of thing. there are a few other reasons too i think, but I'm writing in a hurry :) From bwalton at opencsw.org Fri Jun 12 22:31:20 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 12 Jun 2009 16:31:20 -0400 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <20090612202427.GI95242@bolthole.com> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> <20090612171126.GE95242@bolthole.com> <1244827079-sup-1294@ntdws12.chass.utoronto.ca> <20090612172216.GH95242@bolthole.com> <1244834220-sup-7183@ntdws12.chass.utoronto.ca> <20090612202427.GI95242@bolthole.com> Message-ID: <1244838650-sup-8959@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 12 16:24:27 -0400 2009: > yes, i think so :-) It happens! :) > such a script has multiple uses. > One,is like the exim case mentioned. Otheres, are when you nfs-share out > /opt/csw. > Or, replicate it. Ok, that makes sense now. Thanks for clarifying. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From maciej at opencsw.org Sat Jun 13 13:04:40 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Sat, 13 Jun 2009 12:04:40 +0100 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt Message-ID: In order to support sparse zones, paths such as /opt/csw/var/... must become /var/opt/csw/... - otherwise applications in non-global zones won't be able to write there. Paths are being set in the $(DIRPATHS), as a space-separated list. What's the proper way of tweaking it? (say, define sysconfdir as /etc/opt/csw) I guess I could add sysconfdir to NODIRPATHS and then append my own sysconfdir, but I'm not sure if it's the way Dago intended. Maciej From maciej at opencsw.org Sat Jun 13 15:06:59 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Sat, 13 Jun 2009 14:06:59 +0100 Subject: [csw-maintainers] File from pkgroot not present on disk after installation Message-ID: Today's quandary: The file mpd.conf.CSW is present in work/pkgroot: http://dpaste.com/54904/ It's also present in the prototype and in resulting pkgmap after installation: http://dpaste.com/54911/ ...but the file is not there in the disk! (after installing the package) Do you have any ideas why might that be? More info: It's mGAR v2, rev. 5237, the Makefile is here: http://dpaste.com/54926/ Maciej From maciej at opencsw.org Sat Jun 13 15:32:43 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Sat, 13 Jun 2009 14:32:43 +0100 Subject: [csw-maintainers] File from pkgroot not present on disk after installation In-Reply-To: References: Message-ID: Found the problem: cswcpsampleconf != cswsampleconf in SPKG_CLASSES From dam at opencsw.org Sat Jun 13 21:29:02 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sat, 13 Jun 2009 21:29:02 +0200 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: References: Message-ID: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> Hi Maciej, Am 13.06.2009 um 13:04 schrieb Maciej (Matchek) Blizinski: > In order to support sparse zones, paths such as /opt/csw/var/... must > become /var/opt/csw/... - otherwise applications in non-global zones > won't be able to write there. Paths are being set in the $(DIRPATHS), > as a space-separated list. What's the proper way of tweaking it? (say, > define sysconfdir as /etc/opt/csw) > > I guess I could add sysconfdir to NODIRPATHS and then append my own > sysconfdir, but I'm not sure if it's the way Dago intended. Well, in gar.conf.mk the definitions for the directories are conditional, that means you can safely set sysconfdir = /etc/opt/csw localstatedir = /var/opt/csw and everything should be fine. We could make it the default, but I am afraid that this may break existing build descriptions. Best regards -- Dago From phil at bolthole.com Sat Jun 13 22:04:34 2009 From: phil at bolthole.com (Philip Brown) Date: Sat, 13 Jun 2009 13:04:34 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> Message-ID: <20090613200434.GA19773@bolthole.com> On Sat, Jun 13, 2009 at 09:29:02PM +0200, Dagobert Michelsen wrote: > Well, in gar.conf.mk the definitions for the directories are > conditional, > that means you can safely set > sysconfdir = /etc/opt/csw > localstatedir = /var/opt/csw > and everything should be fine. We could make it the default, but I am > afraid that this may break existing build descriptions. > > i think we should make it the default. From bwalton at opencsw.org Sat Jun 13 22:31:51 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sat, 13 Jun 2009 16:31:51 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090613200434.GA19773@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> Message-ID: <1244925073-sup-2633@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Sat Jun 13 16:04:34 -0400 2009: > > sysconfdir = /etc/opt/csw > > localstatedir = /var/opt/csw > > i think we should make it the default. I second this. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From maciej at opencsw.org Sun Jun 14 15:53:54 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Sun, 14 Jun 2009 14:53:54 +0100 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> Message-ID: On Sat, Jun 13, 2009 at 8:29 PM, Dagobert Michelsen wrote: > Well, in gar.conf.mk the definitions for the directories are conditional, > that means you can safely set > sysconfdir = /etc/opt/csw > localstatedir = /var/opt/csw > and everything should be fine. We could make it the default, but I am > afraid that this may break existing build descriptions. You could grep for /opt/csw/var instances in all the build files and replace them accordingly. I see two main issues: - potential problems will be seen at runtime, not at the time of packaging - already installed programs will have their data in /opt/csw/var, some form of data migration might be necessary Perhaps focus on just few important packages would be enough. Postfix and cups come to mind. Maciej From ihsan at dogan.ch Sun Jun 14 16:01:12 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Sun, 14 Jun 2009 16:01:12 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp 2009 In-Reply-To: <4A3285BB.4000705@opencsw.org> References: <4A11B3AF.7090709@opencsw.org> <4A3269BB.4090604@dogan.ch> <4A3285BB.4000705@opencsw.org> Message-ID: <4A3502A8.5020903@dogan.ch> Am 12.6.2009 18:43 Uhr, Trygve Laugst?l schrieb: >>> The OpenCSW Summer Camp 2009 will take place in Oslo/Norway. To figure >>> out when it should take place, I've create a Doodle poll. Please fill >>> out the link bellow: >>> >>> http://www.doodle.com/6zr9kmdr8p9vv69p >> >> The summer camp will take place on the weekend of 22/23 August 2009 in >> Olso/Norway. Thanks to all, who participated with the poll. >> >> @Trygve: >> Could you please arrange the location and provide a list of recommended >> hotels? > > I'll try to ask around next week. > > I think I've found a nice place for us to work during the day too. The > place is the same that the JUG that I lead use so it should be good > enough for us :) Thanks a lot Trygve. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Sun Jun 14 16:02:20 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 14 Jun 2009 10:02:20 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> Message-ID: <1244988012-sup-5133@ntdws12.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizinski's message of Sun Jun 14 09:53:54 -0400 2009: > You could grep for /opt/csw/var instances in all the build files and > replace them accordingly. I see two main issues: > > - potential problems will be seen at runtime, not at the time of packaging > - already installed programs will have their data in /opt/csw/var, > some form of data migration might be necessary If this is announced loudly enough, all maintainers should take precautions at the next update of affected packages. Could a class script mitigate some of the pain here? The script would either install the new file or iff the file exists in the old location, copy that in instead? There are some corner cases that would need to be handled, so it's not as simple as described, but I think it could be done. Thoughts? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From ihsan at opencsw.org Sun Jun 14 16:11:18 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 14 Jun 2009 16:11:18 +0200 Subject: [csw-maintainers] Summer Camp 2009 (Please book your flights) Message-ID: <4A350506.1000804@opencsw.org> Hello, As already mentioned, the date and the location is fixed. I've updated the wiki as well, but the information is not complete. --> http://wiki.opencsw.org/summercamp-2009 The information for accommodation will be available soon. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From Murray.Jensen at csiro.au Sun Jun 14 18:34:16 2009 From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au) Date: Mon, 15 Jun 2009 02:34:16 +1000 Subject: [csw-maintainers] /testing: nasm 2.05.01 Message-ID: <21894.1244997256@gerd> The following files have been put into /testing ... 9aea1d1df09e7f51525d8a9fee95b0ee nasm-2.05.01,REV=2009.06.14-SunOS5.8-i386-CSW.pkg.gz 0fa0d0eebd39d15f6c7ae7336764b4fa nasm-2.05.01,REV=2009.06.14-SunOS5.8-sparc-CSW.pkg.gz I haven't even run them yet - let me know if they work. Cheers! Murray... From ihsan at opencsw.org Sun Jun 14 20:02:26 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 14 Jun 2009 20:02:26 +0200 Subject: [csw-maintainers] exim maintainer In-Reply-To: <1244821262-sup-9225@ntdws12.chass.utoronto.ca> References: <1244821262-sup-9225@ntdws12.chass.utoronto.ca> Message-ID: <4A353B32.2000607@opencsw.org> Am 12.6.2009 17:41 Uhr, Ben Walton schrieb: > Are you still out there? If so, we've got a bug to file against the > exim package. If not, we may consider taking it over. He should be still available. At least he logs in from time to time on the mail server. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From william at wbonnet.net Mon Jun 15 00:27:43 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 15 Jun 2009 00:27:43 +0200 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config Message-ID: <4A35795F.8000005@wbonnet.net> Hi I am building some Xfce packages and i have the following error : checking for pkg-config... /opt/csw/bin/pkg-config checking for pkg-config >= 0.9.0... 0.23 checking for gtk+-2.0 >= 2.10.0... not found *** The required package gtk+-2.0 was not found on your system. *** Please install gtk+-2.0 (atleast version 2.10.0) or adjust *** the PKG_CONFIG_PATH environment variable if you *** installed the package in a nonstandard prefix so that *** pkg-config is able to find it. I checked for gtk+ libs, and they are up to date, version is superior to 2.10.0 (2.12.0). It seems the arguments passed to pkg-config are not correct. Any one already faced this ? cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From mwatters at opencsw.org Mon Jun 15 00:38:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 14 Jun 2009 17:38:26 -0500 Subject: [csw-maintainers] OpenCSW is now listed on subversion's binary download page. Message-ID: <4A357BE2.3080001@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 http://subversion.tigris.org/getting.html - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAko1e+EACgkQLrhmsXMSLxdQXwCgwVHuKXRGCBQqiB9JG7a8d+Sy s7UAniekUhnHMhwAtkMhGiTQHRPPcFcI =5OMg -----END PGP SIGNATURE----- From phil at bolthole.com Mon Jun 15 06:02:21 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 14 Jun 2009 21:02:21 -0700 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <4A35795F.8000005@wbonnet.net> References: <4A35795F.8000005@wbonnet.net> Message-ID: <20090615040221.GD12886@bolthole.com> On Mon, Jun 15, 2009 at 12:27:43AM +0200, William Bonnet wrote: > Hi > > I am building some Xfce packages and i have the following error : > > checking for pkg-config... /opt/csw/bin/pkg-config > checking for pkg-config >= 0.9.0... 0.23 > checking for gtk+-2.0 >= 2.10.0... not found well it's worse than that. pkg-config --modversion gtk+-2.0 Package xrender was not found in the pkg-config search path. Package 'xrender', required by 'cairo', not found I'm guiessing with the recent rebilds of the xrender package, someone removed the CSWxrender package from the build machines, but forgot to put it back. From dam at opencsw.org Mon Jun 15 11:32:51 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 15 Jun 2009 11:32:51 +0200 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <20090615040221.GD12886@bolthole.com> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> Message-ID: <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> Hi Phil, Am 15.06.2009 um 06:02 schrieb Philip Brown: > On Mon, Jun 15, 2009 at 12:27:43AM +0200, William Bonnet wrote: >> Hi >> >> I am building some Xfce packages and i have the following error : >> >> checking for pkg-config... /opt/csw/bin/pkg-config >> checking for pkg-config >= 0.9.0... 0.23 >> checking for gtk+-2.0 >= 2.10.0... not found > > well it's worse than that. > > pkg-config --modversion gtk+-2.0 > Package xrender was not found in the pkg-config search path. > Package 'xrender', required by 'cairo', not found > > I'm guiessing with the recent rebilds of the xrender package, someone > removed the CSWxrender package from the build machines, but forgot > to put > it back. No. In fact, gtk+-2.0 requires cairo, which requires xrender >= 0.6. I removed the .pc-files for the legacy .so as there should be no compiles against th legacy library. This breaks cairo-xlib-xrender, cairo and xft, all of which I am going to release soon. Best regards -- Dago From pfelecan at opencsw.org Mon Jun 15 16:02:49 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 15 Jun 2009 16:02:49 +0200 Subject: [csw-maintainers] contact maintainer page Message-ID: I'm trying to use the site's mailer: http://www.opencsw.org/eml.php?devname=pfelecan After entering my e-mail address it shows me a page with the text "Debug: part 1". Nice but not useful if somebody tries to write me something through this... -- Peter From dam at opencsw.org Mon Jun 15 16:40:18 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 15 Jun 2009 16:40:18 +0200 Subject: [csw-maintainers] New in testing: libxft2 and libcairo Message-ID: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> Hi, new packages for libxft and libcairo are in testing: libcairo-1.8.6,REV=2009.06.15-SunOS5.8-i386-CSW.pkg.gz libcairo-1.8.6,REV=2009.06.15-SunOS5.8-sparc-CSW.pkg.gz libcairo_devel-1.8.6,REV=2009.06.15-SunOS5.8-i386-CSW.pkg.gz libcairo_devel-1.8.6,REV=2009.06.15-SunOS5.8-sparc-CSW.pkg.gz libcairo_doc-1.8.6,REV=2009.06.15-SunOS5.8-all-CSW.pkg.gz libxft2-2.1.13,REV=2009.06.15-SunOS5.8-i386-CSW.pkg.gz libxft2-2.1.13,REV=2009.06.15-SunOS5.8-sparc-CSW.pkg.gz libxft2_devel-2.1.13,REV=2009.06.15-SunOS5.8-i386-CSW.pkg.gz libxft2_devel-2.1.13,REV=2009.06.15-SunOS5.8-sparc-CSW.pkg.gz I am installing them now on build8st. William: Do you mind giving them a try with your compilation problem? Best regards -- Dago From phil at bolthole.com Mon Jun 15 16:51:08 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 07:51:08 -0700 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> Message-ID: <20090615145107.GD34818@bolthole.com> On Mon, Jun 15, 2009 at 11:32:51AM +0200, Dagobert Michelsen wrote: >> I'm guiessing with the recent rebilds of the xrender package, someone >> removed the CSWxrender package from the build machines, but forgot to >> put >> it back. > > No. In fact, gtk+-2.0 requires cairo, which requires xrender >= 0.6. I > removed > the .pc-files for the legacy .so as there should be no compiles against > the legacy library. But.... it was working before. How did it break now?? we dont have a "new" cairo. so its a mystery why cairo is complaning now, other than "we dont have xrender installed any more". Seems like you could just install xrender again and it would be happy From phil at bolthole.com Mon Jun 15 16:59:06 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 07:59:06 -0700 Subject: [csw-maintainers] contact maintainer page In-Reply-To: References: Message-ID: <20090615145906.GE34818@bolthole.com> On Mon, Jun 15, 2009 at 04:02:49PM +0200, Peter FELECAN wrote: > I'm trying to use the site's mailer: > http://www.opencsw.org/eml.php?devname=pfelecan > > After entering my e-mail address it shows me a page with the text > "Debug: part 1". Nice but not useful if somebody tries to write me > something through this... posix_getpwnam() [a php call] seems to be broken somehow on the website now. I have asked Ihsan to look into it. Until that is resolved, the mail form doesnt work :-( From william at wbonnet.net Mon Jun 15 16:59:17 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 15 Jun 2009 16:59:17 +0200 Subject: [csw-maintainers] New in testing: libxft2 and libcairo In-Reply-To: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> References: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> Message-ID: <4A3661C5.5030602@wbonnet.net> Hi Dago > William: Do you mind giving them a try with your compilation problem? Thanks a lot I'm testing it. Cheers W. From skayser at opencsw.org Mon Jun 15 17:04:12 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 15 Jun 2009 17:04:12 +0200 (CEST) Subject: [csw-maintainers] contact maintainer page In-Reply-To: <20090615145906.GE34818@bolthole.com> References: <20090615145906.GE34818@bolthole.com> Message-ID: <55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de> Philip Brown wrote: > On Mon, Jun 15, 2009 at 04:02:49PM +0200, Peter FELECAN wrote: >> I'm trying to use the site's mailer: >> http://www.opencsw.org/eml.php?devname=pfelecan >> >> After entering my e-mail address it shows me a page with the text >> "Debug: part 1". Nice but not useful if somebody tries to write me >> something through this... > > posix_getpwnam() [a php call] > seems to be broken somehow on the website now. Might be related to the recent php5 modules split-off? [1] Sebastian [1] http://lists.opencsw.org/pipermail/announce/2009-May/000059.html From mwatters at opencsw.org Mon Jun 15 17:22:11 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 15 Jun 2009 10:22:11 -0500 Subject: [csw-maintainers] contact maintainer page In-Reply-To: <55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de> References: <20090615145906.GE34818@bolthole.com> <55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de> Message-ID: <4A366723.3050806@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Sebastian Kayser wrote: > Philip Brown wrote: >> On Mon, Jun 15, 2009 at 04:02:49PM +0200, Peter FELECAN wrote: >>> I'm trying to use the site's mailer: >>> http://www.opencsw.org/eml.php?devname=pfelecan >>> >>> After entering my e-mail address it shows me a page with the text >>> "Debug: part 1". Nice but not useful if somebody tries to write me >>> something through this... >> posix_getpwnam() [a php call] >> seems to be broken somehow on the website now. > > Might be related to the recent php5 modules split-off? [1] Installing the following should fix php5_posix 5.2.9,REV=2009.04.29 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko2ZyMACgkQLrhmsXMSLxcNhACbBXdVFDyrBIcG+33PFeBX6aCf 6vYAoKu0QSLYN+M8f2+Wa1lrllPLTx3D =U4et -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 15 17:39:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 15 Jun 2009 10:39:26 -0500 Subject: [csw-maintainers] New in testing: libxft2 and libcairo In-Reply-To: <4A3661C5.5030602@wbonnet.net> References: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> <4A3661C5.5030602@wbonnet.net> Message-ID: <4A366B2E.9010503@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 William Bonnet wrote: > Hi Dago >> William: Do you mind giving them a try with your compilation problem? > Thanks a lot > > I'm testing it. > > Cheers > W. > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I am trying to build pycairo against it, can you please install python_devel on build8st. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko2ay4ACgkQLrhmsXMSLxegOgCdGcDZllvG7q1yQY0Q/t5uR8yO j2EAnRcHC7+Sgb3T4mDbRQ0k+htKSMA9 =KNfh -----END PGP SIGNATURE----- From dam at opencsw.org Mon Jun 15 17:42:11 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 15 Jun 2009 17:42:11 +0200 Subject: [csw-maintainers] New in testing: libxft2 and libcairo In-Reply-To: <4A366B2E.9010503@opencsw.org> References: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> <4A3661C5.5030602@wbonnet.net> <4A366B2E.9010503@opencsw.org> Message-ID: <0A545E97-991F-47D8-9DD0-45A58F349165@opencsw.org> Hi Mike, Am 15.06.2009 um 17:39 schrieb Mike Watters: > I am trying to build pycairo against it, can you please install > python_devel on > build8st. Done. Best regards -- Dago From william at wbonnet.net Mon Jun 15 18:33:51 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 15 Jun 2009 18:33:51 +0200 Subject: [csw-maintainers] New in testing: libxft2 and libcairo In-Reply-To: <4A3661C5.5030602@wbonnet.net> References: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> <4A3661C5.5030602@wbonnet.net> Message-ID: <4A3677EF.8050103@wbonnet.net> Hi > I'm testing it. It seems to fix my problems :) Thanks a lot. Now i have to install all dependencies on the build8*t machines (to recompile Xfce 4.6.1 on these boxes). Dago can you check that i can sudo pkgrm and pkgadd on the tests machines please ? (i never tried ...). cheers W. From william at wbonnet.net Mon Jun 15 18:35:38 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 15 Jun 2009 18:35:38 +0200 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <20090615145107.GD34818@bolthole.com> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> <20090615145107.GD34818@bolthole.com> Message-ID: <4A36785A.7040801@wbonnet.net> Philip Brown a ?crit : > On Mon, Jun 15, 2009 at 11:32:51AM +0200, Dagobert Michelsen wrote: > >>> I'm guiessing with the recent rebilds of the xrender package, someone >>> removed the CSWxrender package from the build machines, but forgot to >>> put >>> it back. >>> >> No. In fact, gtk+-2.0 requires cairo, which requires xrender >= 0.6. I >> removed >> the .pc-files for the legacy .so as there should be no compiles against >> the legacy library. >> > > But.... it was working before. How did it break now?? > > we dont have a "new" cairo. so its a mystery why cairo is complaning now, > other than "we dont have xrender installed any more". > > Seems like you could just install xrender again and it would be happy > It now works on the test machines :) From mwatters at opencsw.org Mon Jun 15 18:42:42 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 15 Jun 2009 11:42:42 -0500 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <4A36785A.7040801@wbonnet.net> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> <20090615145107.GD34818@bolthole.com> <4A36785A.7040801@wbonnet.net> Message-ID: <4A367A02.4090801@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 William Bonnet wrote: > Philip Brown a ?crit : >> On Mon, Jun 15, 2009 at 11:32:51AM +0200, Dagobert Michelsen wrote: >> >>>> I'm guiessing with the recent rebilds of the xrender package, someone >>>> removed the CSWxrender package from the build machines, but forgot >>>> to put >>>> it back. >>>> >>> No. In fact, gtk+-2.0 requires cairo, which requires xrender >= 0.6. >>> I removed >>> the .pc-files for the legacy .so as there should be no compiles >>> against the legacy library. >>> >> >> But.... it was working before. How did it break now?? >> >> we dont have a "new" cairo. so its a mystery why cairo is complaning now, >> other than "we dont have xrender installed any more". >> >> Seems like you could just install xrender again and it would be happy >> > > It now works on the test machines :) > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers DOH, What happened to libcairo on build8st? I finally get everything ready to build pycairo, type make build and get CAIRO_PS_LEVEL_2 undefined... I check the version of libcairo again and it has reverted to 1.4.10 ;-( guess I need to be faster next time... ;-) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko2egIACgkQLrhmsXMSLxez7ACfc1XO33DqSPRcBR9dRb2bXYCc lr4AoOGWIMnoXBYVr46dBoxfEb1Y+rts =Is1H -----END PGP SIGNATURE----- From phil at bolthole.com Mon Jun 15 19:02:41 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 10:02:41 -0700 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <4A36785A.7040801@wbonnet.net> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> <20090615145107.GD34818@bolthole.com> <4A36785A.7040801@wbonnet.net> Message-ID: <20090615170241.GN95242@bolthole.com> On Mon, Jun 15, 2009 at 06:35:38PM +0200, William Bonnet wrote: > Philip Brown a ?crit : >> Seems like you could just install xrender again and it would be happy >> > > It now works on the test machines :) well, that's nice and alll.. but we need our standard build machines functional as well!! From bwalton at opencsw.org Mon Jun 15 19:45:54 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 15 Jun 2009 13:45:54 -0400 Subject: [csw-maintainers] versioning /etc Message-ID: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> Hi Phil and Peter, I wondered what both you think about supporting a tool like etckeeper[1] from pkg-get and pkgutil. With the recent discussion about defaulting to /etc/opt/csw for our stuff, I'm considering packaging it up. The tools would need to grow the idea of pre- and post- install hooks, so that etckeeper could be driven to snapshot /etc before and after packages are manipulated with the high-level tools. These hooks may be useful for other things too...it seems that apt, yum and co support a generic hook mechanism for things like this to plug in to. I've been dabbling etckeeper a bit on a few systems and I like aspects of it, although I'm still not ready to do a wholesale rollout. Thoughts? Thanks -Ben [1] http://joey.kitenet.net/code/etckeeper/ -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Mon Jun 15 19:55:11 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 10:55:11 -0700 Subject: [csw-maintainers] versioning /etc In-Reply-To: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> Message-ID: <20090615175511.GV95242@bolthole.com> On Mon, Jun 15, 2009 at 01:45:54PM -0400, Ben Walton wrote: > > Hi Phil and Peter, > > I wondered what both you think about supporting a tool like > etckeeper[1] from pkg-get and pkgutil. With the recent discussion > about defaulting to /etc/opt/csw for our stuff, I'm considering > packaging it up. > > The tools would need to grow the idea of pre- and post- install hooks, > so that etckeeper could be driven to snapshot /etc before and after > packages are manipulated with the high-level tools. These hooks may > be useful for other things too...it seems that apt, yum and co support > a generic hook mechanism for things like this to plug in to. I think that sounds rather overly broad. Not to mention, it sorta assumes that "you run an install tool, and you DONT KNOW WHAT IT DOES, so you use this tool to make a copy of everything". If one of our packages falls in the category of "we dont know what it does in etc", it's a bad package, and needs to be fixed!! If you want to package up etckeeper... great! That being said... As far as integrating it in our install tools, however, I dont think it's a good idea. It's extra complication, and it encourages sloppy package creation. From bwalton at opencsw.org Mon Jun 15 20:03:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 15 Jun 2009 14:03:47 -0400 Subject: [csw-maintainers] versioning /etc In-Reply-To: <20090615175511.GV95242@bolthole.com> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <20090615175511.GV95242@bolthole.com> Message-ID: <1245088612-sup-5510@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Mon Jun 15 13:55:11 -0400 2009: > I think that sounds rather overly broad. > > Not to mention, it sorta assumes that > > "you run an install tool, and you DONT KNOW WHAT IT DOES, so you > use this tool to make a copy of everything". No, it's not to mitigate this at all... > If one of our packages falls in the category of > "we dont know what it does in etc", > it's a bad package, and needs to be fixed!! Agreed, but again, that's not the purpose of this tool. > As far as integrating it in our install tools, however, I dont think > it's a good idea. You wouldn't be integrating it _specifically_, but rather providing a framework for other things to be transparently integrated, thus allowing a site to leverage this tool or others that may find running code pre and post package actions to be useful. etckeeper would simply leverage these hooks. Nobody would be forced to use it, as absence of the package (and thus the hooks it would install) would see pkg-get/pkgutil operate as it had historically. > It's extra complication, and it encourages sloppy package creation. Nope. The point of etckeeper is to version /etc...it's not to mitigate sloppy packaging. I think it's fairly common to version parts of /etc these days, whether you use some vcs or cp -p. This wraps it up into a nice package and allows for a single repo for all of /etc instead of one for apache, one for foo, one for...The beauty of this is that it can help in cases where a new version of a package breaks your existing install by trampling a config file (incompatible changes, etc)...or if you want to see exactly what it changed. It goes beyond CSW packages, to everything that lives in /etc. Again, as I mentioned, I'm not totally sold on this particular package myself, but the idea of pkg-get/pkgutil providing hooks for other things can be completely independent of this... -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Mon Jun 15 20:52:34 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 11:52:34 -0700 Subject: [csw-maintainers] versioning /etc In-Reply-To: <1245088612-sup-5510@ntdws12.chass.utoronto.ca> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <20090615175511.GV95242@bolthole.com> <1245088612-sup-5510@ntdws12.chass.utoronto.ca> Message-ID: <20090615185234.GX95242@bolthole.com> On Mon, Jun 15, 2009 at 02:03:47PM -0400, Ben Walton wrote: > Nope. The point of etckeeper is to version /etc...it's not to > mitigate sloppy packaging. [...] > The beauty > of this is that it can help in cases where a new version of a package > breaks your existing install by trampling a config file (incompatible > changes, etc)...or if you want to see exactly what it changed. Hmm. that sounds potentially useful. > It > goes beyond CSW packages, to everything that lives in /etc. And that sounds... like a lot of work. At a very high level, do you have architectural suggestions as to how this sort of integration would function? From rupert at opencsw.org Mon Jun 15 21:08:21 2009 From: rupert at opencsw.org (rupert THURNER) Date: Mon, 15 Jun 2009 21:08:21 +0200 Subject: [csw-maintainers] versioning /etc In-Reply-To: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> Message-ID: <6af4270906151208q757f503bs1fb2022bb0f068a4@mail.gmail.com> On Mon, Jun 15, 2009 at 19:45, Ben Walton wrote: > > Hi Phil and Peter, > > I wondered what both you think about supporting a tool like > etckeeper[1] from pkg-get and pkgutil. ?With the recent discussion > about defaulting to /etc/opt/csw for our stuff, I'm considering > packaging it up. > > The tools would need to grow the idea of pre- and post- install hooks, > so that etckeeper could be driven to snapshot /etc before and after > packages are manipulated with the high-level tools. ?These hooks may > be useful for other things too...it seems that apt, yum and co support > a generic hook mechanism for things like this to plug in to. > > I've been dabbling etckeeper a bit on a few systems and I like aspects > of it, although I'm still not ready to do a wholesale rollout. > > Thoughts? this does not mean that the default /opt/csw/etc goes away, isn't it? because we enjoyed a lot that one mountpoint is sufficient to run csw. rupert. From bwalton at opencsw.org Mon Jun 15 21:31:35 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 15 Jun 2009 15:31:35 -0400 Subject: [csw-maintainers] versioning /etc In-Reply-To: <20090615185234.GX95242@bolthole.com> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <20090615175511.GV95242@bolthole.com> <1245088612-sup-5510@ntdws12.chass.utoronto.ca> <20090615185234.GX95242@bolthole.com> Message-ID: <1245092338-sup-6372@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Mon Jun 15 14:52:34 -0400 2009: > > It goes beyond CSW packages, to everything that lives in /etc. > > And that sounds... like a lot of work. > At a very high level, do you have architectural suggestions as to how this > sort of integration would function? It actually shouldn't be that bad. My suggestion would be the tried and true .d directory. I'd suggest something like: /etc/opt/csw/pkg-hooks/pre.d /etc/opt/csw/pkg-hooks/post.d (no difference between -i and -u is required, I don't think). pkg-get/pkgutil would then do: ---> before doing work cd /etc/opt/csw/pkg-hooks/pre.d/ for f in *; do [ -x $f ] && ./$f done ---> after doing work cd /etc/opt/csw/pkg-hooks/post.d/ for f in *; do [ -x $f ] && ./$f done An alternate sort order could be applied if required...this is just a simplistic approach. There could (possibly) be a user settable option in csw.conf as to how non-zero exits would be handled, thus allowing for various hooks to prevent a run if some trigger is detected, etc. Then, etckeeper (just as an example) would, as part of it's package, drop a file named 01-etckeeper in pkg-hooks/pre.d/ with a script like: #!/bin/sh /opt/csw/bin/etckeeper preinstall A similar post install hook would be placed in post.d. How does that sound? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Mon Jun 15 21:39:22 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 12:39:22 -0700 Subject: [csw-maintainers] versioning /etc In-Reply-To: <1245092338-sup-6372@ntdws12.chass.utoronto.ca> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <20090615175511.GV95242@bolthole.com> <1245088612-sup-5510@ntdws12.chass.utoronto.ca> <20090615185234.GX95242@bolthole.com> <1245092338-sup-6372@ntdws12.chass.utoronto.ca> Message-ID: <20090615193922.GA95242@bolthole.com> On Mon, Jun 15, 2009 at 03:31:35PM -0400, Ben Walton wrote: > It actually shouldn't be that bad. My suggestion would be the tried > and true .d directory. I'd suggest something like: > > /etc/opt/csw/pkg-hooks/pre.d > /etc/opt/csw/pkg-hooks/post.d > Hmmm. interesting idea. still pondering it. but i'll make the comment now, that I think we shouldnt abbreviate. so, "preinstall.d", or simply "preinstall", vs "pre.d" > (no difference between -i and -u is required, I don't think). not so sure about that one either. From bwalton at opencsw.org Mon Jun 15 21:48:16 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 15 Jun 2009 15:48:16 -0400 Subject: [csw-maintainers] versioning /etc In-Reply-To: <20090615193922.GA95242@bolthole.com> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <20090615175511.GV95242@bolthole.com> <1245088612-sup-5510@ntdws12.chass.utoronto.ca> <20090615185234.GX95242@bolthole.com> <1245092338-sup-6372@ntdws12.chass.utoronto.ca> <20090615193922.GA95242@bolthole.com> Message-ID: <1245095107-sup-9601@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Mon Jun 15 15:39:22 -0400 2009: > interesting idea. > still pondering it. but i'll make the comment now, that I think we shouldnt > abbreviate. > > so, "preinstall.d", or simply "preinstall", vs "pre.d" Ok, as long as it's clear what the directory is for, the name isn't important. Since it's meant for computer consumption, longer names are fine with me too. > > (no difference between -i and -u is required, I don't think). > > not so sure about that one either. My argument for the simplification is that there is no underlying concept in pkg{add,rm} of -i being different from -u. There wouldn't be anything inherently wrong with providing more options up front though. There is also the wrinkle that a -u can imply -i for other packages, but maybe that's "just one of those things"... -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Mon Jun 15 22:31:32 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 15 Jun 2009 22:31:32 +0200 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <4A367A02.4090801@opencsw.org> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> <20090615145107.GD34818@bolthole.com> <4A36785A.7040801@wbonnet.net> <4A367A02.4090801@opencsw.org> Message-ID: <8BE61299-F4BB-4992-960D-D30953C4BA8F@opencsw.org> Hi Mike, Am 15.06.2009 um 18:42 schrieb Mike Watters: > William Bonnet wrote: >> It now works on the test machines :) > > DOH, What happened to libcairo on build8st? > I finally get everything ready to build pycairo, type make build > and get > CAIRO_PS_LEVEL_2 undefined... I check the version of libcairo > again and it has > reverted to 1.4.10 ;-( > > guess I need to be faster next time... ;-) :-) Hopefully cairo will be released soon, so you should be ok on all machines then. Best regards -- Dago From mwatters at opencsw.org Mon Jun 15 22:35:03 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 15 Jun 2009 15:35:03 -0500 Subject: [csw-maintainers] lcms now in testing Message-ID: <4A36B077.6030108@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 "Little cms intends to be a small-footprint, speed optimized color management engine in open source form." --> http://www.littlecms.com/ pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u lcms Feedback is always welcome. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko2sHcACgkQLrhmsXMSLxeoKwCdHmuCLc/zkwVn8Q1nzJuOgqbf 7W8AoMkrYOfRt0tXGbItP42VMyNpH1W/ =7JRL -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 15 22:47:23 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 15 Jun 2009 15:47:23 -0500 Subject: [csw-maintainers] lcms now in testing In-Reply-To: <4A36B077.6030108@opencsw.org> References: <4A36B077.6030108@opencsw.org> Message-ID: <4A36B35B.5080405@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Rodger, Sorry for popping this in testing, I missed you owning it. (I looked for liblcms and not lcms ) I will remove from testing. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko2s1sACgkQLrhmsXMSLxfI2gCgnkqiSScb3Rm1E4H+zSuaVy5Y 5qwAnjMaXHPvGyWab7Ep70SCJ5h5/vhb =+9dH -----END PGP SIGNATURE----- From hson at opencsw.org Tue Jun 16 00:34:40 2009 From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=) Date: Tue, 16 Jun 2009 00:34:40 +0200 Subject: [csw-maintainers] lcms now in testing In-Reply-To: <4A36B35B.5080405@opencsw.org> References: <4A36B077.6030108@opencsw.org> <4A36B35B.5080405@opencsw.org> Message-ID: <4A36CC80.4060103@opencsw.org> Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Rodger, > Sorry for popping this in testing, I missed you owning it. > (I looked for liblcms and not lcms ) > > I will remove from testing. No problem, I just took over it from the previous "owner" due to the fact that imagemagick needs it. But I'll push out a new package ASAP. From phil at bolthole.com Tue Jun 16 00:58:17 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 15:58:17 -0700 Subject: [csw-maintainers] versioning /etc In-Reply-To: <6af4270906151208q757f503bs1fb2022bb0f068a4@mail.gmail.com> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <6af4270906151208q757f503bs1fb2022bb0f068a4@mail.gmail.com> Message-ID: <20090615225817.GI95242@bolthole.com> On Mon, Jun 15, 2009 at 09:08:21PM +0200, rupert THURNER wrote: > this does not mean that the default /opt/csw/etc goes away, isn't it? nope. From william at wbonnet.net Tue Jun 16 09:42:22 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 16 Jun 2009 09:42:22 +0200 Subject: [csw-maintainers] Is maintainer still active Message-ID: <4A374CDE.6020006@wbonnet.net> Hi I was wondering ... is Nathan Coraor still active ? I don't remember to have heard from him over the last month, and i opened two upgrade request against two packages he maintains. I am asking this just in case it is known Nathan is retired :) cheers W. From nate at blastwave.org Tue Jun 16 16:05:36 2009 From: nate at blastwave.org (Nate Coraor) Date: Tue, 16 Jun 2009 10:05:36 -0400 Subject: [csw-maintainers] Is maintainer still active In-Reply-To: <4A374CDE.6020006@wbonnet.net> References: <4A374CDE.6020006@wbonnet.net> Message-ID: <4A37A6B0.4070506@blastwave.org> William Bonnet wrote: > Hi > > I was wondering ... is Nathan Coraor still active ? I don't remember to > have heard from him over the last month, and i opened two upgrade > request against two packages he maintains. > > I am asking this just in case it is known Nathan is retired :) I've tried to get back into it a few times, but yes, I'm retired. --nate From phil at bolthole.com Tue Jun 16 18:22:28 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 16 Jun 2009 09:22:28 -0700 Subject: [csw-maintainers] Is maintainer still active In-Reply-To: <4A37A6B0.4070506@blastwave.org> References: <4A374CDE.6020006@wbonnet.net> <4A37A6B0.4070506@blastwave.org> Message-ID: <20090616162227.GA49403@bolthole.com> On Tue, Jun 16, 2009 at 10:05:36AM -0400, Nate Coraor wrote: > > I've tried to get back into it a few times, but yes, I'm retired. > noted From william at wbonnet.net Tue Jun 16 18:28:27 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 16 Jun 2009 18:28:27 +0200 Subject: [csw-maintainers] Is maintainer still active In-Reply-To: <4A37A6B0.4070506@blastwave.org> References: <4A374CDE.6020006@wbonnet.net> <4A37A6B0.4070506@blastwave.org> Message-ID: <4A37C82B.9090606@wbonnet.net> Hi Nate >> >> I am asking this just in case it is known Nathan is retired :) > > I've tried to get back into it a few times, but yes, I'm retired. Thanks for answering Nate. So, at short term, i propose to update three of your packages (perl module Glib and its depends). I need some fresh version to build Xfce 4.6.1. I hope you will find time to get back in the project, in the next weeks or months cheers W. From skayser at opencsw.org Tue Jun 16 21:54:39 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 16 Jun 2009 21:54:39 +0200 Subject: [csw-maintainers] Bugtracker (Mantis) patched for performance, feedback welcome In-Reply-To: <4A326858.4010405@dogan.ch> References: <4A319985.8040904@opencsw.org> <1244765051-sup-8822@ntdws12.chass.utoronto.ca> <4A326858.4010405@dogan.ch> Message-ID: <4A37F87F.1020900@opencsw.org> Ihsan Dogan wrote: > Ben Walton wrote: > >>> i have patched a Mantis test installation today to improve the long page >>> load times we have been seeing. A couple of features that we don't use >>> as of yet (sub-projects, private projects, product versions) have been >>> de-facto deactivated as a result of this. The changes are targeted at >>> data display only and should be non-disruptive. >> +1. It looks good to me and it's a _lot_ faster. Nice work! :) > > Wow, it's a rocket now. Thanks Sebastian. Thanks for the feedback. Our productive installation is now also tweaked. Happy bug squashing ;) Sebastian From william at wbonnet.net Tue Jun 16 21:59:31 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 16 Jun 2009 21:59:31 +0200 Subject: [csw-maintainers] Bugtracker (Mantis) patched for performance, feedback welcome In-Reply-To: <4A319985.8040904@opencsw.org> References: <4A319985.8040904@opencsw.org> Message-ID: <4A37F9A3.1050209@wbonnet.net> Hi Sebastian > i have patched a Mantis test installation today to improve the long page > load times we have been seeing. A couple of features that we don't use > as of yet (sub-projects, private projects, product versions) have been > de-facto deactivated as a result of this. The changes are targeted at > data display only and should be non-disruptive. > Thanks ! cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Tue Jun 16 22:31:59 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 16 Jun 2009 22:31:59 +0200 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1244925073-sup-2633@ntdws12.chass.utoronto.ca> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> Message-ID: <4A38013F.60500@opencsw.org> Ben Walton wrote: > Excerpts from Philip Brown's message of Sat Jun 13 16:04:34 -0400 2009: > >>> sysconfdir = /etc/opt/csw >>> localstatedir = /var/opt/csw >> i think we should make it the default. > > I second this. Can we ponder about the /etc stuff a bit? From working with other people using OpenCSW i know that there is always some kind of confusion as to where to find configuration files. Most are in /opt/csw/etc, but some are in /etc/opt/csw. Our file layout standards page [1] says that machine-specific files should go to /etc/opt/csw, but even i as a maintainer sometimes find it difficult to decide whether to place them there or not. Furthermore IIRC there have been discussions on the list about issues with sparse zone installations and configuration files beneath /opt. >From that perspective i would second this change also (only one place to look for configs, smoother integration with zones), but then there has been Rupert who raised some concerns about not being able to just share out /opt/csw any more [2]. Is this common? Is it a requirement that we can break in order to eventually just have /etc/opt/csw or how would it work? Sebastian [1] http://opencsw.org/standards/layout [2] http://lists.opencsw.org/pipermail/maintainers/2009-June/002914.html From ihsan at opencsw.org Tue Jun 16 23:23:47 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Tue, 16 Jun 2009 23:23:47 +0200 Subject: [csw-maintainers] Summer Camp 2009 Message-ID: <4A380D63.6020706@opencsw.org> Hello, I've updated the wiki page with the attendees and the hotel. I've booked my flight and hotel already and it would be great if we could stay all together in one hotel. --> http://wiki.opencsw.org/summercamp-2009 PS: Topics and schedule will follow... Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From phil at bolthole.com Tue Jun 16 23:24:24 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 16 Jun 2009 14:24:24 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <4A38013F.60500@opencsw.org> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> Message-ID: <20090616212424.GD51834@bolthole.com> On Tue, Jun 16, 2009 at 10:31:59PM +0200, Sebastian Kayser wrote: > Our file layout standards page [1] says that machine-specific files > should go to /etc/opt/csw, but even i as a maintainer sometimes find it > difficult to decide whether to place them there or not. when in doubt, discuss it on the maintainers list. > Furthermore IIRC > there have been discussions on the list about issues with sparse zone > installations and configuration files beneath /opt. they should be treated pretty much in the same way as if /opt was NFS-shared between separate hosts. Sometimes, it is DESIRABLE to have the configs globally in /opt/csw/etc. Ideally, a program could be configured to first chec /opt/csw/etc, and then override with any local configs in /etc/opt/csw But if not, then it's up to the maintainer to decide. From mwatters at opencsw.org Tue Jun 16 23:58:18 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 16 Jun 2009 16:58:18 -0500 Subject: [csw-maintainers] squid now in testing Message-ID: <4A38157A.2080002@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: Changed the default Cache location to /var/opt/csw/squid from /opt/csw/var - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko4FXoACgkQLrhmsXMSLxdkGACgvkknL3YgYzKDxVxJYmYqLQSm hYUAoNfPacS5YNoZqGmYUjA5Q7VvO/Hj =mDxZ -----END PGP SIGNATURE----- From mwatters at opencsw.org Wed Jun 17 00:04:28 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 16 Jun 2009 17:04:28 -0500 Subject: [csw-maintainers] maintainer for gmp? Message-ID: <4A3816EC.3060103@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Is anyone working on gmp? I am updating libmpfr, and would like to compile it 64bit. if not, I will rebuild gmp - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko4FusACgkQLrhmsXMSLxdE6ACgsekxESdj/NQqQgMMErVa+Oth 3ykAoL6tj6rnkN9df3jeCrX918WEAJO6 =E9Jx -----END PGP SIGNATURE----- From william at wbonnet.net Wed Jun 17 00:07:56 2009 From: william at wbonnet.net (William Bonnet) Date: Wed, 17 Jun 2009 00:07:56 +0200 Subject: [csw-maintainers] Anyone for Eclipse Message-ID: <4A3817BC.6030101@wbonnet.net> Hi, I am current maintainer of Eclipse, and i am a bit overloaded ;) Eclipse is a very important package, which should be kept up to date. So if anyone is interested in having this package, i'll be happy to step down. Otherwise i'll update it ... not before September (for a real target date). cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Wed Jun 17 00:08:56 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 17 Jun 2009 00:08:56 +0200 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090616212424.GD51834@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> Message-ID: <4A3817F8.50202@opencsw.org> Philip Brown wrote: > On Tue, Jun 16, 2009 at 10:31:59PM +0200, Sebastian Kayser wrote: >> Our file layout standards page [1] says that machine-specific files >> should go to /etc/opt/csw, but even i as a maintainer sometimes find it >> difficult to decide whether to place them there or not. > > when in doubt, discuss it on the maintainers list. > >> Furthermore IIRC >> there have been discussions on the list about issues with sparse zone >> installations and configuration files beneath /opt. > > they should be treated pretty much in the same way as if /opt was NFS-shared > between separate hosts. > > Sometimes, it is DESIRABLE to have the configs globally in /opt/csw/etc. Does "desirable" go hand in hand with the core notion of this thread to change the default $(sysconfdir) to /etc/opt/csw? Have all configuration files in /etc/opt/csw eventually unless a maintainer decides otherwise for his packages? > Ideally, a program could be configured to first chec /opt/csw/etc, and then > override with any local configs in /etc/opt/csw Ok, this is easy for daemons via their init scripts. Not so easy for most other programs. They read one config file location and usually a dot file. > But if not, then it's up to the maintainer to decide. And this leaves the user at the point that i mentioned initially. "Mhh, there are configuration files in two places. Where is the one for package foo located?". Not one of my top concerns right now, just something that i have observed and wanted to share while we are talking about configuration file locations. Sure, they can look it up, but it lends itself to confusion. Sebastian From skayser at opencsw.org Wed Jun 17 00:32:41 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 17 Jun 2009 00:32:41 +0200 Subject: [csw-maintainers] Summer Camp 2009 In-Reply-To: <4A380D63.6020706@opencsw.org> References: <4A380D63.6020706@opencsw.org> Message-ID: <4A381D89.3080506@opencsw.org> Ihsan Dogan wrote: > I've updated the wiki page with the attendees and the hotel. I've booked > my flight and hotel already and it would be great if we could stay all > together in one hotel. > > --> http://wiki.opencsw.org/summercamp-2009 > > > PS: Topics and schedule will follow... For the topics, how about each topic "owner" fills in some information on the respective topic (similar to how it was done for the meeting Zurich [1]) so that the topic itself is prepared and others can give it a thought (and come up with possibly helpful ideas) in advance? Especially if there is something that you might need others to think about a bit first in order to have a good discussion. Sebastian [1] http://wiki.opencsw.org/suggestions From dam at opencsw.org Wed Jun 17 09:14:22 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 17 Jun 2009 09:14:22 +0200 Subject: [csw-maintainers] libtool on build8st downgraded References: <4A388CC6.4090905@opencsw.org> Message-ID: <5AB3EAB5-DBEC-4CA0-A6E0-45D155E821F8@opencsw.org> Hi, Anfang der weitergeleiteten E-Mail: > K?nntest du nicht auf build8st das alte Libtool installieren? FYI: libtool on build8st has been downgraded temporarily to 1.5.24. Best regards -- Dago From maciej at opencsw.org Wed Jun 17 09:37:24 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Wed, 17 Jun 2009 08:37:24 +0100 Subject: [csw-maintainers] Broken links on Trac timeline page Message-ID: I've noticed that the links on the Trac timeline page[1] use IP addresses instead of DNS names. The IP currently present in the links[2] times out. Pages can still be reached by replacing the IP address with the domain name: s/172\.29\.29\.21/sourceforge.net/, perhaps it is a retired IP address? Maciej [1] https://sourceforge.net/apps/trac/gar/timeline [2] https://172.29.29.21/apps/trac/gar/wiki/GarRepository?version=7 From ihsan at opencsw.org Wed Jun 17 10:09:25 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Wed, 17 Jun 2009 10:09:25 +0200 Subject: [csw-maintainers] libtool on build8st downgraded In-Reply-To: <5AB3EAB5-DBEC-4CA0-A6E0-45D155E821F8@opencsw.org> References: <4A388CC6.4090905@opencsw.org> <5AB3EAB5-DBEC-4CA0-A6E0-45D155E821F8@opencsw.org> Message-ID: <4A38A4B5.80106@opencsw.org> Dagobert Michelsen wrote: >> K?nntest du nicht auf build8st das alte Libtool installieren? > FYI: libtool on build8st has been downgraded temporarily to 1.5.24. > Thanks Dago. Ihsa -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Wed Jun 17 11:57:04 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 17 Jun 2009 11:57:04 +0200 Subject: [csw-maintainers] maintainer for gmp? In-Reply-To: <4A3816EC.3060103@opencsw.org> References: <4A3816EC.3060103@opencsw.org> Message-ID: <459B7BA3-8F64-4300-BE69-06C9F7D96E65@opencsw.org> Hi Mike, Am 17.06.2009 um 00:04 schrieb Mike Watters: > Is anyone working on gmp? > I am updating libmpfr, and would like to compile it 64bit. > if not, I will rebuild gmp I did some work on it in the past, but failed x86 assembly optimization. Apart from that the package should be usable. Please take it over and rebuild it :-) Best regards -- Dago From skayser at opencsw.org Wed Jun 17 12:08:19 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 17 Jun 2009 12:08:19 +0200 (CEST) Subject: [csw-maintainers] Broken links on Trac timeline page In-Reply-To: References: Message-ID: <57452.78.140.68.9.1245233299.squirrel@ssl.skayser.de> Maciej (Matchek) Blizinski wrote: > I've noticed that the links on the Trac timeline page[1] use IP > addresses instead of DNS names. The IP currently present in the > links[2] times out. Pages can still be reached by replacing the IP > address with the domain name: s/172\.29\.29\.21/sourceforge.net/, > perhaps it is a retired IP address? After briefly checking with the guys on #sourceforge i just created a ticket for the issue. https://sourceforge.net/apps/trac/sourceforge/ticket/1374 Sebastian From phil at bolthole.com Wed Jun 17 15:47:45 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 17 Jun 2009 06:47:45 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <4A3817F8.50202@opencsw.org> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> Message-ID: <20090617134745.GA46139@bolthole.com> On Wed, Jun 17, 2009 at 12:08:56AM +0200, Sebastian Kayser wrote: > Philip Brown wrote: > > Sometimes, it is DESIRABLE to have the configs globally in /opt/csw/etc. > > Does "desirable" go hand in hand with the core notion of this thread to > change the default $(sysconfdir) to /etc/opt/csw? Have all configuration > files in /etc/opt/csw eventually unless a maintainer decides otherwise > for his packages? "defaults" are usually set to be whatever is used most commonly. my non-quantified belief is that most of our packages that have config files, are/should have them in /etc/opt/csw If so, then the default should point to there. > And this leaves the user at the point that i mentioned initially. "Mhh, > there are configuration files in two places. Where is the one for > package foo located?". a good point. I'll update our User guide to mention to check both locations. From skayser at opencsw.org Wed Jun 17 20:12:58 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 17 Jun 2009 20:12:58 +0200 Subject: [csw-maintainers] Broken links on Trac timeline page In-Reply-To: <57452.78.140.68.9.1245233299.squirrel@ssl.skayser.de> References: <57452.78.140.68.9.1245233299.squirrel@ssl.skayser.de> Message-ID: <4A39322A.9020303@opencsw.org> Sebastian Kayser wrote: > Maciej (Matchek) Blizinski wrote: >> I've noticed that the links on the Trac timeline page[1] use IP >> addresses instead of DNS names. The IP currently present in the >> links[2] times out. Pages can still be reached by replacing the IP >> address with the domain name: s/172\.29\.29\.21/sourceforge.net/, >> perhaps it is a retired IP address? > > After briefly checking with the guys on #sourceforge i just created a > ticket for the issue. > > https://sourceforge.net/apps/trac/sourceforge/ticket/1374 Update: I haven't had any SF feedback on the ticket, but for me the links on the timeline page now contain a domain name again and are working fine. Thanks for the heads up Maciej. Sebastian From skayser at opencsw.org Thu Jun 18 01:45:28 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 18 Jun 2009 01:45:28 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? Message-ID: <4A398018.7050003@opencsw.org> Hi, our perl 5.8.8 contains Storable.pm 2.15 as a core module. Now i would like to package POE [1] which requires Storable.pm >= 2.16. What would be the options to deliver a Storable.pm >= 2.16 that POE can use? >From looking at @INC our /opt/csw/lib/perl/csw comes after the core module locations (which makes sense somehow), so i suppose it won't be as easy as just to package the current Storable.pm (2.20) from CPAN. Sebastian [1] http://search.cpan.org/dist/POE/ From bwalton at opencsw.org Thu Jun 18 01:59:19 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 17 Jun 2009 19:59:19 -0400 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <4A398018.7050003@opencsw.org> References: <4A398018.7050003@opencsw.org> Message-ID: <1245282863-sup-6225@ntdws12.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Wed Jun 17 19:45:28 -0400 2009: > our perl 5.8.8 contains Storable.pm 2.15 as a core module. Now i would > like to package POE [1] which requires Storable.pm >= 2.16. What would > be the options to deliver a Storable.pm >= 2.16 that POE can use? Woudl updating to 5.10.0 (the current stable) get you what you need? > >From looking at @INC our /opt/csw/lib/perl/csw comes after the core > module locations (which makes sense somehow), so i suppose it won't be > as easy as just to package the current Storable.pm (2.20) from CPAN. If you bundle the module with the package and force the package to use the file from your package (by munging @INC), you could likely stay with the current perl...this gets really ugly, really fast though. It's more work, but updating perl is likely the best option...Peter? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From skayser at opencsw.org Thu Jun 18 02:11:22 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 18 Jun 2009 02:11:22 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <1245282863-sup-6225@ntdws12.chass.utoronto.ca> References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> Message-ID: <4A39862A.5030309@opencsw.org> Ben Walton wrote: > Excerpts from Sebastian Kayser's message of Wed Jun 17 19:45:28 -0400 2009: >> our perl 5.8.8 contains Storable.pm 2.15 as a core module. Now i would >> like to package POE [1] which requires Storable.pm >= 2.16. What would >> be the options to deliver a Storable.pm >= 2.16 that POE can use? > > Woudl updating to 5.10.0 (the current stable) get you what you need? According to the POE dependency tree, yes. See http://deps.cpantesters.org/?module=POE;perl=latest Sebastian From mwatters at opencsw.org Thu Jun 18 03:59:29 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 17 Jun 2009 20:59:29 -0500 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <1245282863-sup-6225@ntdws12.chass.utoronto.ca> References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> Message-ID: <4A399F81.8010100@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Sebastian Kayser's message of Wed Jun 17 19:45:28 -0400 2009: >> our perl 5.8.8 contains Storable.pm 2.15 as a core module. Now i would >> like to package POE [1] which requires Storable.pm >= 2.16. What would >> be the options to deliver a Storable.pm >= 2.16 that POE can use? if moving to 5.10.0 is not possible. It is technically be possible to replace the Storable.pm in core with a "new" version. I don't know what the policy would be on this, but you would need a "pre-install" that does a backup of the existing files. and a post-remove that puts them all back -- only if the core perl package has "not" been removed. this may be more hassle then hacking @INC - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAko5n4EACgkQLrhmsXMSLxcJ9wCdGU3cCXfjx8YIGnFOP5Wq4Fvp ilIAoJnAfk5JMziF1RTJRg/ayUd1S2eV =sfVx -----END PGP SIGNATURE----- From mwatters at opencsw.org Thu Jun 18 04:48:41 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 17 Jun 2009 21:48:41 -0500 Subject: [csw-maintainers] maintainer for gmp? In-Reply-To: <459B7BA3-8F64-4300-BE69-06C9F7D96E65@opencsw.org> References: <4A3816EC.3060103@opencsw.org> <459B7BA3-8F64-4300-BE69-06C9F7D96E65@opencsw.org> Message-ID: <4A39AB09.9030604@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > I did some work on it in the past, but failed x86 assembly optimization. > Apart from that the package should be usable. > Please take it over and rebuild it :-) using Studio 11 I can't get it to compile on any platform ;-( it compiles fine using gcc4 currently looking through the forums for patches that may assist. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAko5qwkACgkQLrhmsXMSLxe/BQCeNnid01ukmhSVJst2JJYef9V1 mHYAn3AnE65dE4uF8Qgw/Ywax3p4rs+3 =3MAf -----END PGP SIGNATURE----- From bonivart at opencsw.org Thu Jun 18 09:22:27 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 18 Jun 2009 09:22:27 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <4A399F81.8010100@opencsw.org> References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> <4A399F81.8010100@opencsw.org> Message-ID: <625385e30906180022x60b998fcl8bd0660c02d4ac8d@mail.gmail.com> On Thu, Jun 18, 2009 at 3:59 AM, Mike Watters wrote: > It is technically be possible to replace the Storable.pm in core with a "new" > version. This is what I usually do. > I don't know what the policy would be on this, but you would need a > "pre-install" that does a backup of the existing files. and a post-remove that > puts them all back -- only if the core perl package has "not" been removed. Please explain this further, is this specifically for this module or in general? -- /peter From pfelecan at opencsw.org Thu Jun 18 10:26:49 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 18 Jun 2009 10:26:49 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <4A399F81.8010100@opencsw.org> (Mike Watters's message of "Wed\, 17 Jun 2009 20\:59\:29 -0500") References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> <4A399F81.8010100@opencsw.org> Message-ID: Mike Watters writes: > Ben Walton wrote: >> Excerpts from Sebastian Kayser's message of Wed Jun 17 19:45:28 -0400 2009: >>> our perl 5.8.8 contains Storable.pm 2.15 as a core module. Now i would >>> like to package POE [1] which requires Storable.pm >= 2.16. What would >>> be the options to deliver a Storable.pm >= 2.16 that POE can use? > > if moving to 5.10.0 is not possible. > > It is technically be possible to replace the Storable.pm in core with a "new" > version. > I don't know what the policy would be on this, but you would need a > "pre-install" that does a backup of the existing files. and a post-remove that > puts them all back -- only if the core perl package has "not" been removed. > > this may be more hassle then hacking @INC Maybe, but it's an ugly kludge doomed to fail if somebody out of our control needs the module in the current form. Subsidiary question, how comes that we are still with Perl 5.8.8? -- Peter From bonivart at opencsw.org Thu Jun 18 10:33:47 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 18 Jun 2009 10:33:47 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> <4A399F81.8010100@opencsw.org> Message-ID: <625385e30906180133w7cf717e2s98760085eae14d85@mail.gmail.com> On Thu, Jun 18, 2009 at 10:26 AM, Peter FELECAN wrote: > Subsidiary question, how comes that we are still with Perl 5.8.8? 1. I haven't had time to work on it. 2. Demand has been zero. -- /peter From dam at opencsw.org Thu Jun 18 15:04:42 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 18 Jun 2009 15:04:42 +0200 Subject: [csw-maintainers] maintainer for gmp? In-Reply-To: <4A39AB09.9030604@opencsw.org> References: <4A3816EC.3060103@opencsw.org> <459B7BA3-8F64-4300-BE69-06C9F7D96E65@opencsw.org> <4A39AB09.9030604@opencsw.org> Message-ID: Hi Mike, Am 18.06.2009 um 04:48 schrieb Mike Watters: >> I did some work on it in the past, but failed x86 assembly >> optimization. >> Apart from that the package should be usable. >> Please take it over and rebuild it :-) > > using Studio 11 I can't get it to compile on any platform ;-( > it compiles fine using gcc4 > > currently looking through the forums for patches that may assist. Ok, last time I tried it compiled nicely without assemblers optimization. Please note that Solaris 8 w/Studio 11 is best effort, but Solaris 9 w/Studio 12 is baseline. Best regards -- Dago From mwatters at opencsw.org Thu Jun 18 16:22:59 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 18 Jun 2009 09:22:59 -0500 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <625385e30906180022x60b998fcl8bd0660c02d4ac8d@mail.gmail.com> References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> <4A399F81.8010100@opencsw.org> <625385e30906180022x60b998fcl8bd0660c02d4ac8d@mail.gmail.com> Message-ID: <4A3A4DC3.4010402@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter Bonivart wrote: > On Thu, Jun 18, 2009 at 3:59 AM, Mike Watters wrote: >> It is technically be possible to replace the Storable.pm in core with a "new" >> version. > > This is what I usually do. > >> I don't know what the policy would be on this, but you would need a >> "pre-install" that does a backup of the existing files. and a post-remove that >> puts them all back -- only if the core perl package has "not" been removed. > > Please explain this further, is this specifically for this module or in general? > Just in general. when I replace a "core piece" of software with a newer version, I like to backup and restore the original if the new version is removed. as long as you know the install package names for all software involved it is fairly easily accomplished with installf/removef. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko6TcMACgkQLrhmsXMSLxdREgCfUP6M1JKfIaHPBokM9maRSVd0 CUIAniHVL8ClDh4LNTf+3NfslkDgoOqL =Dg9Y -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu Jun 18 16:32:23 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 18 Jun 2009 10:32:23 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090617134745.GA46139@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> Message-ID: <1245335357-sup-8252@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed Jun 17 09:47:45 -0400 2009: > On Wed, Jun 17, 2009 at 12:08:56AM +0200, Sebastian Kayser wrote: > > Philip Brown wrote: > > > Sometimes, it is DESIRABLE to have the configs globally in /opt/csw/etc. > > > > Does "desirable" go hand in hand with the core notion of this thread to > > change the default $(sysconfdir) to /etc/opt/csw? Have all configuration > > files in /etc/opt/csw eventually unless a maintainer decides otherwise > > for his packages? > > "defaults" are usually set to be whatever is used most commonly. > my non-quantified belief is that most of our packages that have config > files, are/should have them in /etc/opt/csw > If so, then the default should point to there. I'm combining threads here, since they are of overlapping subject matter. > On Mon, Jun 15, 2009 at 09:08:21PM +0200, rupert THURNER wrote: > > this does not mean that the default /opt/csw/etc goes away, isn't it? > > nope. ...but, if the GAR default is changed, it will gradually become less prominent. Lets have a real discussion about this, weighing pros and cons of the move from a default of /opt/csw/etc to a default of /etc/opt/csw. Pros: 1. A single /etc tree (eventually), and not including some of the sfw stuff, etc. 2. Still maintains a partition between system packages and csw packages that might overlap. Cons: 1. For people sharing out a single install (possibly per $arch) with NFS, /etc/opt/csw may make things more difficult, although I'd argue that symlinks could work around this quite easily. [For people in this situation, would creating symlinks for you?] 2. Potentially painful for maintainers during the transition. 3. User confusion during the transition, although I'd argue it's already a confusing situation. Personally, I really dislike having multiple etc directories...it feels wrong to me, but that is likely due to my Linux roots where the FHS has been in place for a long time and it isn't carrying some of the legacy aspects that make solaris attractive in other ways. I also dislike having /opt/csw/apache2/etc and would prefer /etc/opt/csw/apache2, for example. Should this discussion be taken to users@ to 'poll the audience'? I'm curious what others think. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Thu Jun 18 18:04:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 18 Jun 2009 11:04:51 -0500 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245335357-sup-8252@ntdws12.chass.utoronto.ca> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> Message-ID: <4A3A65A3.7000003@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Philip Brown's message of Wed Jun 17 09:47:45 -0400 2009: >> On Wed, Jun 17, 2009 at 12:08:56AM +0200, Sebastian Kayser wrote: >>> Philip Brown wrote: >>>> Sometimes, it is DESIRABLE to have the configs globally in /opt/csw/etc. >>> Does "desirable" go hand in hand with the core notion of this thread to >>> change the default $(sysconfdir) to /etc/opt/csw? Have all configuration >>> files in /etc/opt/csw eventually unless a maintainer decides otherwise >>> for his packages? >> "defaults" are usually set to be whatever is used most commonly. >> my non-quantified belief is that most of our packages that have config >> files, are/should have them in /etc/opt/csw >> If so, then the default should point to there. > > I'm combining threads here, since they are of overlapping subject matter. > > >> On Mon, Jun 15, 2009 at 09:08:21PM +0200, rupert THURNER wrote: >>> this does not mean that the default /opt/csw/etc goes away, isn't it? >> nope. > > ...but, if the GAR default is changed, it will gradually become less > prominent. Lets have a real discussion about this, weighing pros and > cons of the move from a default of /opt/csw/etc to a default of > /etc/opt/csw. > > Pros: > > 1. A single /etc tree (eventually), and not including some of the sfw > stuff, etc. a single etc tree can be accomplished with either directory > > 2. Still maintains a partition between system packages and csw > packages that might overlap. > > Cons: > > 1. For people sharing out a single install (possibly per $arch) with > NFS, /etc/opt/csw may make things more difficult, although I'd > argue that symlinks could work around this quite easily. [For > people in this situation, would creating symlinks for you?] symlinks may or may not work if you are trying to run on zones from a single installation on a global zone. > > 2. Potentially painful for maintainers during the transition. yes, but I wouldn't consider that a con per se > > 3. User confusion during the transition, although I'd argue it's > already a confusing situation. not just confusing for the users, confusing for the maintainers as well, at least for me. -- but that is not saying much ;-) > > Personally, I really dislike having multiple etc directories...it > feels wrong to me, but that is likely due to my Linux roots where the > FHS has been in place for a long time and it isn't carrying some of > the legacy aspects that make solaris attractive in other ways. I also > dislike having /opt/csw/apache2/etc and would prefer > /etc/opt/csw/apache2, for example. I agree with that, anything that has it's own etc tree should be consolidated to whatever "new standard" location is chosen. > > Should this discussion be taken to users@ to 'poll the audience'? I think it is a good Idea to take it to the users. ultimately they are the ones driving this project, without them our work is moot. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko6ZaIACgkQLrhmsXMSLxcV/gCffjgNpSNKCvBYOJZfpYXFHz61 CYsAoMDYLB/xMounpQG61d+D52sndjBu =ByWA -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu Jun 18 18:18:34 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 18 Jun 2009 12:18:34 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <4A3A65A3.7000003@opencsw.org> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <4A3A65A3.7000003@opencsw.org> Message-ID: <1245341535-sup-5196@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Thu Jun 18 12:04:51 -0400 2009: > > 1. A single /etc tree (eventually), and not including some of the sfw > > stuff, etc. > a single etc tree can be accomplished with either directory Not if you take single to mean 'all etc content' Ultimately, I'd like the CSW etc to be nested in the system /etc, which is my 'definition' of single. If you take single to mean single within CSW, then yes, you're right. > > 1. For people sharing out a single install (possibly per $arch) with > > NFS, /etc/opt/csw may make things more difficult, although I'd > > argue that symlinks could work around this quite easily. [For > > people in this situation, would creating symlinks for you?] > symlinks may or may not work if you are trying to run on zones from a single > installation on a global zone. Yes, zones certainly complicate things. As we are zone-free here, I'd appreciate others chiming in here. What is the best way to handle this wrt zones. Does the best option for zone support help or hinder the idea of nfs sharing as a goal? > > 2. Potentially painful for maintainers during the transition. > yes, but I wouldn't consider that a con per se Fair enough. Any change will see some short-term (or long, depending on package update cycle) issues, I guess. It does get tricky though in some cases. Hypothetically take the libxml and docbook packages. If docbook moved etc's first, it would still have to reference the old locations, since libxml is rooted in the old location. When libxml is updated, docbook would need to be updated too, to handle the new location. Shall we start a new thread on users@ then? Any objections? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 18 18:46:51 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 18 Jun 2009 09:46:51 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245335357-sup-8252@ntdws12.chass.utoronto.ca> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> Message-ID: <20090618164651.GG63121@bolthole.com> On Thu, Jun 18, 2009 at 10:32:23AM -0400, Ben Walton wrote: > I'm combining threads here, since they are of overlapping subject matter. Unfortunately, you quoted too much, and overly long emails tend to get zero replies. I'm going to try to shorten it a lot in the interests of more focused discussion:-) > 1. For people sharing out a single install (possibly per $arch) with > NFS, /etc/opt/csw may make things more difficult, although I'd > argue that symlinks could work around this quite easily. [For > people in this situation, would creating symlinks for you?] For people with a shared /opt (whether that be NFS or zones), the same issues still come into play. the issue of, "should configuration for software X, be unique to each box, or globally shared?" You dont see to address that type of issue at all. > Personally, I really dislike having multiple etc directories...it > feels wrong to me, but that is likely due to my Linux roots Or bsd. yes, i understand your background :-) Personally, I think /etc/opt/VENDOR makes a lot of sense. I also think that, in the rare cases global configs are desired (and it really is quite rare) /opt/csw/etc makes sense for those also. > I also > dislike having /opt/csw/apache2/etc and would prefer > /etc/opt/csw/apache2, for example. I think most everyone dislikes that. it is legacy configuration, and should really be changed to use /etc/opt/csw/apache2, in my opinion. > Should this discussion be taken to users@ to 'poll the audience'? I think we should have the discussion on the maintainers list first, to pre-air out any controversial issues. A fully public, large sized mailing list, is never a good place to have initial "exploratory" type technical discussions. Better to distill the core issues first. From mwatters at opencsw.org Thu Jun 18 19:18:30 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 18 Jun 2009 12:18:30 -0500 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090618164651.GG63121@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> Message-ID: <4A3A76E6.6060406@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > For people with a shared /opt (whether that be NFS or zones), the same > issues still come into play. the issue of, "should configuration for > software X, be unique to each box, or globally shared?" > > You dont see to address that type of issue at all. Actually, and correct me if I am wrong, I think "global vs unique" is a moot point if you have a shared /opt (NFS or zones.) I believe you will get a global config regardless of /opt/csw/etc or /etc/opt/csw. Example: shared /opt via NFS and linking /etc/opt/csw to /opt/csw/etc you get the same result as if you just put it in /opt/csw/etc. if you change it to suit one box, everything that has the NFS mount will get changed. the alternative is to instruct the user to copy the contents of /etc/opt/csw from the shared server to each NFS mount, but that IMHO defeats the purpose of a "shared" installation. if we want a unique etc for each box, I say we: * make the default /etc/opt/csw. * create a package "CSWcswshared" that has a script with a config, that will allow the user to pick and choose what packages he/she wants local configs for and copy those to /etc/opt/csw/{catalogname} -- skipping files that already exist. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko6duYACgkQLrhmsXMSLxc4DQCfeb4SVqCYoqwhD60IJ7GUyyf7 EVMAoJRhpv+b2JTMW6b737er59DVaEe6 =EiYA -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu Jun 18 19:21:41 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 18 Jun 2009 13:21:41 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090618164651.GG63121@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> Message-ID: <1245344654-sup-7349@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 18 12:46:51 -0400 2009: > For people with a shared /opt (whether that be NFS or zones), the same > issues still come into play. the issue of, "should configuration for > software X, be unique to each box, or globally shared?" > > You dont see to address that type of issue at all. I think that's driven by the software, not the packaging. While some things (eg: sudo) allow for a single, globally shared configuration and other things (eg: apache) can use includes to share a global template exist, many packages don't offer this flexibility outside of environment variables or other hacks. For the people doing nfs/zone sharing of /opt/csw, how are local configurations being handled presently? Something like apache is simple to handle, but other things aren't as easy. It seems like reconciling some of these things would only work if the software allowed for config1 -> global, config2 -> local (as Phil mentioned previously)...I haven't seen much of that though. Even if this were to be handled by some packages, I'd vote for /etc/opt/csw/$foo and /etc/opt/csw/global/$foo. If we have /etc/opt/csw at all, that means machine local or a second mount of some sort anyway, so why not go all the way? > Personally, I think /etc/opt/VENDOR makes a lot of sense. > I also think that, in the rare cases global configs are desired > (and it really is quite rare) /opt/csw/etc makes sense for those > also. Can you point out one of these rare cases so that we have a good example for discussion? I think sudo is (possibly) a good choice, but only because it _can_ be global, not because it typically is (at least in my experience). > I think we should have the discussion on the maintainers list first, > to pre-air out any controversial issues. A fully public, large > sized mailing list, is never a good place to have initial > "exploratory" type technical discussions. Better to distill the core > issues first. Ok. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 18 19:35:16 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 18 Jun 2009 10:35:16 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245344654-sup-7349@ntdws12.chass.utoronto.ca> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> Message-ID: <20090618173516.GA59068@bolthole.com> On Thu, Jun 18, 2009 at 01:21:41PM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Thu Jun 18 12:46:51 -0400 2009: > > For people with a shared /opt (whether that be NFS or zones), the same > > issues still come into play. the issue of, "should configuration for > > software X, be unique to each box, or globally shared?" > > ... > > I think that's driven by the software, not the packaging. i agree. > Even if this were to be handled by some packages, I'd vote for > /etc/opt/csw/$foo and /etc/opt/csw/global/$foo. anything in /etc, is by definition, NOT global :-} "global/shared" stuff, is under /opt/csw. Whether that be because it is NFS-shared, or lofs-shared across zones. > Can you point out one of these rare cases so that we have a good > example for discussion? I think sudo is (possibly) a good choice, but > only because it _can_ be global, not because it typically is (at least > in my experience). Funnily enough, at my place of work, the sudo config is indeed a global one across all machines. however, it is replicated out :-) I'm not sure of many "global-only config" examples. I can think of one or two good "global AND local" examples. For example... pkg-get :-) I think generally, examples of things that benefit from a global configuration, are things that are commonly used by users, and just dont work right without a detailed etc file. For example, netpbm? or xpdf? It is not likely that there will be any machine-local configuration tweaks needed. It is quite likely that a site-wide configuration is the best style of config, for those type of things. Or to put it another way, things that get some kind of conf file installed at pkgadd time, and the user will probably never bother to change, because the config is just so standard. Another example, might be mailcap files. From william at wbonnet.net Fri Jun 19 00:19:44 2009 From: william at wbonnet.net (William Bonnet) Date: Fri, 19 Jun 2009 00:19:44 +0200 Subject: [csw-maintainers] pixman 0.15.12 is in testing Message-ID: <4A3ABD80.1030700@wbonnet.net> Hi I have update pixman to the latest version. It is available from testing Feedback will be welcomed cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Fri Jun 19 01:02:39 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 19 Jun 2009 01:02:39 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <625385e30906180133w7cf717e2s98760085eae14d85@mail.gmail.com> References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> <4A399F81.8010100@opencsw.org> <625385e30906180133w7cf717e2s98760085eae14d85@mail.gmail.com> Message-ID: <4A3AC78F.9010502@opencsw.org> Peter Bonivart wrote: > On Thu, Jun 18, 2009 at 10:26 AM, Peter FELECAN wrote: >> Subsidiary question, how comes that we are still with Perl 5.8.8? > > 1. I haven't had time to work on it. > 2. Demand has been zero. Demand++ ;) Would a perl update require us to re-package modules that contain compiled .so s or would you expect any other major breakage? Sebastian From bwalton at opencsw.org Fri Jun 19 01:30:55 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 18 Jun 2009 19:30:55 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090618173516.GA59068@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> Message-ID: <1245367441-sup-4715@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 18 13:35:16 -0400 2009: > > Even if this were to be handled by some packages, I'd vote for > > /etc/opt/csw/$foo and /etc/opt/csw/global/$foo. > > anything in /etc, is by definition, NOT global :-} "global/shared" > stuff, is under /opt/csw. Whether that be because it is NFS-shared, > or lofs-shared across zones. Ok, but if we're already in the position of straddling shared vs non-shared, which presumably sees the nfs/zones people need to make arrangements for sharing the things living in /etc already, why not go all the way? > Funnily enough, at my place of work, the sudo config is indeed a > global one across all machines. however, it is replicated out :-) We'd use cfengine for this task too. > I'm not sure of many "global-only config" examples. I can think of > one or two good "global AND local" examples. For example... pkg-get > :-) I think it's safe to say that not many software authors make this consideration...nor should they be expected to, imo. Maybe the emergence of zones changes the game some, but nfs sharing of software volumes must be on the decline by now? > Or to put it another way, things that get some kind of conf file > installed at pkgadd time, and the user will probably never bother to > change, because the config is just so standard. Again though, if we already have /etc/opt/csw in use, why not go all the way? I understand that things that never need to be fiddled with _could_ live in /opt/csw/etc, but since the local version exists and will cause problems (presumably) for the nfs/zones people, solving it for all packages should be the same work as solving it for half the packages. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 19 01:49:07 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 18 Jun 2009 16:49:07 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245367441-sup-4715@ntdws12.chass.utoronto.ca> References: <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> Message-ID: <20090618234907.GA52737@bolthole.com> On Thu, Jun 18, 2009 at 07:30:55PM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Thu Jun 18 13:35:16 -0400 2009: > > anything in /etc, is by definition, NOT global :-} "global/shared" > > stuff, is under /opt/csw. Whether that be because it is NFS-shared, > > or lofs-shared across zones. > > Ok, but if we're already in the position of straddling shared vs > non-shared, which presumably sees the nfs/zones people need to make > arrangements for sharing the things living in /etc already, why not go > all the way? you make this "go all the way" statement multiple times, but I dont understand what you mean by it. So I will condense and rephrase what i was saying previously: MOST configs, probably belong in /etc/opt/csw. However, there are a few configs, that rarely change, and are presumed to be standardized site-wide. For those things, there is no harm(and sometimes, an actual benefit) in having the configs live in /opt/csw/etc. Things like xpdf configs, and so on, that almost never change, and are presumed normally identical across all machines at a site. .......... Furthermore, in the even that a site administrator decides it is important to hack a local override for the config file, our packages can smoothly and easily handle it, if deployed in a clean manner. if /opt/csw/etc/someglobal.conf is usually copied in via our cswcpsampleconf class action, it will only be copied in or updated, if the site has not created something of their own there. So, if the site admin replaced it, with /opt/csw/etc/someglobal.conf -> /etc/opt/csw/someglobal.conf then that symlink would be left along, and what is normally a globally shared config file, becomes a machine-local config file, for that site only. From mwatters at opencsw.org Fri Jun 19 02:08:28 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 18 Jun 2009 19:08:28 -0500 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090618234907.GA52737@bolthole.com> References: <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> Message-ID: <4A3AD6FC.8030705@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > you make this "go all the way" statement multiple times, but I dont > understand what you mean by it. I have to agree with Ben here, and we should "go all the way" with using /etc/opt/csw. what "I" mean by "go all the way" is no longer use /opt/csw/etc or /opt/csw/$foo/etc and everything gets put in /etc/opt/csw or /etc/opt/csw/$foo > So, if the site admin replaced it, with > /opt/csw/etc/someglobal.conf -> /etc/opt/csw/someglobal.conf I am not sure that is the case. the way you have it linked above... if the admin links the version from /opt/csw/etc to /etc/opt/csw that link would be true for every host. if the /etc/opt/csw file does not exist on a particular host, the config would be non-existent. > then that symlink would be left along, and what is normally a globally > shared config file, becomes a machine-local config file, for that site > only. the link would be left, as well as a someglobal.conf.CSW file which the apps are not currently configured to use. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAko61vsACgkQLrhmsXMSLxcz4wCg0ZmdH0r239M7303r1GuaoM1J LDAAoN9sIpiM5zpnKr9u3+3fteCw4DoM =Hy1F -----END PGP SIGNATURE----- From bwalton at opencsw.org Fri Jun 19 02:40:15 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 18 Jun 2009 20:40:15 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090618234907.GA52737@bolthole.com> References: <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> Message-ID: <1245370024-sup-1205@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 18 19:49:07 -0400 2009: Ok, let me try to rephrase too. I agree with your points, but I still think a single etc is the way to go. People sharing /opt/csw are going to need to handle configuration differently than others with everything local to the machine (where local means not nfs shared and if in a zone, a non-sparse zone). If the config is in /opt/csw/etc and they need it customized per nfs client (or zone), they'll be symlinking it to somewhere on the box or mounting another directory onto /opt/csw/etc. This means that every box using packages from /opt/csw will be required to have this config maintained locally somehow. Also, because some packages installed on the host for /opt/csw have placed config files in /etc/opt/csw, they'll be symlinking the client /etc/opt/csw to some location in the share, rsyncing it from the host, or otherwise taking additional care to make sure all clients have the required files. IOW, they're making accommodations in _both_ directions, ignoring the sites that only shove a few gnu tools or something onto a share. Not to ignore your point about site standard, glacially changing files...For sites not sharing their /opt/csw, this is moot. It doesn't matter where they live, so it might as well be /etc/opt/csw. For sites that are sharing, I think (hope?) that I've demonstrated that they're already doing extra work anyway, so xpdf.conf and the like wouldn't add burden. Does that make more sense? I think we mostly agree, we're just not seeing eye-to-eye on small semantic points. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 19 02:47:22 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 18 Jun 2009 17:47:22 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <4A3AD6FC.8030705@opencsw.org> References: <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> Message-ID: <20090619004722.GA64424@bolthole.com> On Thu, Jun 18, 2009 at 07:08:28PM -0500, Mike Watters wrote: s> > So, if the site admin replaced it, with > > /opt/csw/etc/someglobal.conf -> /etc/opt/csw/someglobal.conf > > I am not sure that is the case. the way you have it linked above... > if the admin links the version from /opt/csw/etc to /etc/opt/csw > that link would be true for every host. yes, i was saying if the site admin decides, "no i want this app configured locally instead of globally", he still technically has that option, relatively easily. Lest anyone misunderstand me, i will reiterate that I am still completely in favour of the DEFAULT for demons and things, being "configs go in /etc/opt/csw". I am just trying to leave open the door, for certain applications to be allowed to have "global" configs in /opt/csw/etc, if it makese sense to do so. Something like xpdf, I think, it makes sense to do so. You can just NFS-share out /opt/csw everywhere, and it will work correctly. If on the other hand, it looks for configs in /etc/opt..... you share out /opt/csw... and then it breaks, until you go and have to copy in something to /etc/opt/csw on to **Every Single Machine*** you want to run it on. Not very large-scale-install friendly. From phil at bolthole.com Fri Jun 19 02:52:02 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 18 Jun 2009 17:52:02 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245370024-sup-1205@ntdws12.chass.utoronto.ca> References: <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <1245370024-sup-1205@ntdws12.chass.utoronto.ca> Message-ID: <20090619005202.GB64424@bolthole.com> On Thu, Jun 18, 2009 at 08:40:15PM -0400, Ben Walton wrote: > ... > People sharing /opt/csw are going to need to handle configuration > differently than others with everything local to the machine (where > local means not nfs shared and if in a zone, a non-sparse zone). > > If the config is in /opt/csw/etc and they need it customized per nfs > client (or zone), they'll be symlinking it to somewhere on the box or > mounting another directory onto /opt/csw/etc. There is a missing piece here. Site that share out /opt/csw, tend to do it for things that are easily sharable. things like xpdf **firefox** You want firefox to have all its default configs in /etc? Individually? For every machine? I dooont think sooo.... You want global configs for a beast like that. If global configs are then "ok" for something like that, then they should be allowed in certain other targetted situations also. > IOW, they're making accommodations in _both_ directions, ignoring the > sites that only shove a few gnu tools or something onto a share. btw, I dont think we should ignore "sites that only shove a few gnu tools onto a share" either. > Does that make more sense? I think we mostly agree, we're just not > seeing eye-to-eye on small semantic points. I think we're actually disagreeing on a rare-case, but conceptually large issue. Hopefully, the firefox example, shows the issue better ;-) From mwatters at opencsw.org Fri Jun 19 19:11:10 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 19 Jun 2009 12:11:10 -0500 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090619005202.GB64424@bolthole.com> References: <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <1245370024-sup-1205@ntdws12.chass.utoronto.ca> <20090619005202.GB64424@bolthole.com> Message-ID: <4A3BC6AE.9070102@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > There is a missing piece here. > Site that share out /opt/csw, tend to do it for things that are easily > sharable. > things like > xpdf > **firefox** > > You want firefox to have all its default configs in /etc? Individually? > For every machine? > > I dooont think sooo.... > You want global configs for a beast like that. Firefox is a different beast, and by default does not store anything in *etc* the global configs for firefox are stored in .*/share and the "local configs" are stored in the user's home directory under .mozilla I would say that firefox is not an issue in sharing /opt. Software like xpdf, I don't think it matters where the configs are located since they normally don't change. ~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~ the whole problem here, I believe, is how do we make CSW packages respect both local configs and global configs without hacking up every piece of software we package. or how do we tell the users to change their sharing schema to allow the same locations across the board. The easiest but least optimal solution I believe is to put everything in /opt/csw/ and link anything back to /etc that needs to be there... i.e. init scripts on solaris 8/9. for smf, /opt/csw/var/manifest and /opt/csw/var/methods This would solve the issue, although I reiterate this is not the best solution, but I am not sure there is a "best solution" with this problem. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko7xq0ACgkQLrhmsXMSLxeZLwCgxd9a6cNvkYnLM468JFhKwl5C xjQAn3DPH7f7bHjzxSxFn9MY+zQ4kWGP =Ksfv -----END PGP SIGNATURE----- From bwalton at opencsw.org Fri Jun 19 19:20:44 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 19 Jun 2009 13:20:44 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090619004722.GA64424@bolthole.com> References: <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> Message-ID: <1245431722-sup-3548@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 18 20:47:22 -0400 2009: > Lest anyone misunderstand me, i will reiterate that I am still > completely in favour of the DEFAULT for demons and things, being > "configs go in /etc/opt/csw". Understood. > You can just NFS-share out /opt/csw everywhere, and it will work > correctly. If on the other hand, it looks for configs in > /etc/opt..... you share out /opt/csw... and then it breaks, until > you go and have to copy in something to /etc/opt/csw on to **Every > Single Machine*** you want to run it on. > Not very large-scale-install friendly. Large-scale sites are exactly the same types of installations that would be using cfengine/puppet/etc to manage things like this. Do you agree that having two etc directories for an installation without nfs/zone sharing is pointless? Do you agree that if everything were in /etc/opt/csw and a site wanted to share their csw install, a move on the server of /etc/opt/csw to /opt/csw/etc and a symlink of /etc/opt/csw -> /opt/csw/etc on the all machines consuming this particular /opt/csw would handle things just fine? (Packages wanting/needing local customization could still symlink individual files to somewhere else as you noted.) I realize that Rupert raised concerns about this, due to restrictions in his environment, so it might be nice to get his thoughts (or the thoughts of others in a shared-csw setup) on this. I'm not in the position of having to deal with something like this. It would also be nice, but difficult to determine, how many sites actually use csw like this. 1%? 5%? More? Less? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 19 19:34:19 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 19 Jun 2009 10:34:19 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245431722-sup-3548@ntdws12.chass.utoronto.ca> References: <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> Message-ID: <20090619173419.GA77530@bolthole.com> On Fri, Jun 19, 2009 at 01:20:44PM -0400, Ben Walton wrote: >.... > Do you agree that having two etc directories for an installation > without nfs/zone sharing is pointless? Why even ask that question? So you can say, "well then, lets forget about sharing and optimize for the non-shared case"? > > Not very large-scale-install friendly. > > Large-scale sites are exactly the same types of installations that > would be using cfengine/puppet/etc to manage things like this. Some do, some dont. Also, using cfengine or puppet to handle this, would merely be a hack around a core problem that the package provider (ie: us) is not providing proper large-install aware packages in the first place! > Do you agree that if everything were in /etc/opt/csw and a site wanted > to share their csw install, a move on the server of /etc/opt/csw to > /opt/csw/etc and a symlink of /etc/opt/csw -> /opt/csw/etc on the all > machines consuming this particular /opt/csw would handle things just > fine? (Packages wanting/needing local customization could still > symlink individual files to somewhere else as you noted.) That is ugly. Besides, you may just as equally say, "everything should use /opt/csw/etc, and people who want local configs, should symlink /opt/csw/etc ->/etc/opt/csw" > It would also be nice, but difficult to determine, how many sites > actually use csw like this. 1%? 5%? More? Less? > I dont think this is a good question to base decisions on. Again, it has behind it, the mindset and assumption of, "well, most of our 'customers' are not large scale, so lets ignore what suits large scale". Forgive me if I'm wrong, but your questions seem to come from a basis of, [I dont care about large installs, i just want to do whats easiest/tidiest for me] is that accurate? From phil at bolthole.com Fri Jun 19 19:40:24 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 19 Jun 2009 10:40:24 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <4A3BC6AE.9070102@opencsw.org> References: <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <1245370024-sup-1205@ntdws12.chass.utoronto.ca> <20090619005202.GB64424@bolthole.com> <4A3BC6AE.9070102@opencsw.org> Message-ID: <20090619174024.GB77530@bolthole.com> On Fri, Jun 19, 2009 at 12:11:10PM -0500, Mike Watters wrote: > > You want firefox to have all its default configs in /etc? Individually? > > For every machine? > > > > I dooont think sooo.... > > You want global configs for a beast like that. > > Firefox is a different beast, and by default does not store anything in *etc* > the global configs for firefox are stored in .*/share and the "local configs" > are stored in the user's home directory under .mozilla but it could. And in some ways, it "should". It just doesnt at the moment :-) > > I would say that firefox is not an issue in sharing /opt. > > Software like xpdf, I don't think it matters where the configs are located > since they normally don't change. So, you seem to be agreeing with my premise, that SOME programs are fine in having their configs on the global /opt/csw, when they dont change. > the whole problem here, I believe, is how do we make CSW packages respect both > local configs and global configs without hacking up every piece of software we > package. We simply cant do that for *every* package. more below. > .... > This would solve the issue, although I reiterate this is not the best solution, > but I am not sure there is a "best solution" with this problem. I believe that is the core truth here. There is no "one size fits all 'best' solution". What we CAN do, is try to make the "best" choice, on a package-by-package basis. with certain guidelines. The most relevant ones being, "treat /opt/csw/* as read-only after install. If something is likely to need significant changing after initial install, then it belongs in /etc/opt/csw or /var/opt/csw" From bwalton at opencsw.org Fri Jun 19 20:03:38 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 19 Jun 2009 14:03:38 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090619173419.GA77530@bolthole.com> References: <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> <20090619173419.GA77530@bolthole.com> Message-ID: <1245433062-sup-2821@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 19 13:34:19 -0400 2009: > > Do you agree that having two etc directories for an installation > > without nfs/zone sharing is pointless? > > Why even ask that question? > So you can say, "well then, lets forget about sharing and optimize for the > non-shared case"? No, just a staring point. > > > Not very large-scale-install friendly. > > > > Large-scale sites are exactly the same types of installations that > > would be using cfengine/puppet/etc to manage things like this. > > Some do, some dont. > Also, using cfengine or puppet to handle this, would merely be a hack > around a core problem that the package provider (ie: us) is not providing > proper large-install aware packages in the first place! I disagree here. The problem isn't that things aren't packaged nicely for this situation, it's that software isn't written to support this situation. If most packages provided support for $localsysconf and $globalsysconf, sourcing global and then overlaying local, I'd say that maintaining 2 etc directories made perfect sense. Given that this certainly isn't the case, I don't think 2 etc's is good. > > Do you agree that if everything were in /etc/opt/csw and a site wanted > > to share their csw install, a move on the server of /etc/opt/csw to > > /opt/csw/etc and a symlink of /etc/opt/csw -> /opt/csw/etc on the all > > machines consuming this particular /opt/csw would handle things just > > fine? (Packages wanting/needing local customization could still > > symlink individual files to somewhere else as you noted.) > > That is ugly. Uglier than having etc directories all over the place? At best, I think we can say that there isn't an optimal solution for consumers of CSW in a setup like this. > Besides, you may just as equally say, "everything should use /opt/csw/etc, > and people who want local configs, should symlink /opt/csw/etc > ->/etc/opt/csw" But then we're still scattering etc directories around the system. To be honest, I'd prefer that all CSW etc be consolidated into a single location, even if the chosen location doesn't see it become a subset of the system etc tree. > > It would also be nice, but difficult to determine, how many sites > > actually use csw like this. 1%? 5%? More? Less? > > > > I dont think this is a good question to base decisions on. > Again, it has behind it, the mindset and assumption of, > "well, most of our 'customers' are not large scale, so lets ignore what > suits large scale". Here, you're making the assumption that all large scale installations would share /opt/csw. I suspect this isn't the case, although please correct me if you think otherwise. The point of this question is not to marginalize sites using CSW like this, but to simply keep in perspective this size of the audience that may or may not benefit from keeping 2 etc directories. > Forgive me if I'm wrong, but your questions seem to come from a basis of, > [I dont care about large installs, i just want to do whats easiest/tidiest > for me] No, I do care. While I think homing things under /etc/ is the better choice, I don't want to discriminate against sites that share /opt/csw. I'm trying to argue that if we've already arrived at the decision that /etc/opt/csw is the better _default_, then we're not going to be discriminating any further by moving everything there. > is that accurate? No. I think we'll end up agreeing to disagree here. I'm in favour of changing the default location, as are you. Lets just carry forward from this decision. We can always look at things when more packages are living in /etc/opt/csw anyway. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 19 22:55:51 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 19 Jun 2009 13:55:51 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245433062-sup-2821@ntdws12.chass.utoronto.ca> References: <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> <20090619173419.GA77530@bolthole.com> <1245433062-sup-2821@ntdws12.chass.utoronto.ca> Message-ID: <20090619205551.GC12157@bolthole.com> On Fri, Jun 19, 2009 at 02:03:38PM -0400, Ben Walton wrote: > I think we'll end up agreeing to disagree here. I'm in favour of > changing the default location, as are you. Lets just carry forward > from this decision. We can always look at things when more packages > are living in /etc/opt/csw anyway. ok, sounds like a working agreement. I think you are proposing, "lets focus on moving what is easily movable, from /opt/csw/etc, to /etc/opt/csw, and then review what is left after that point". I am very happy to move forward on that premise :-) From skayser at opencsw.org Sat Jun 20 02:00:46 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Sat, 20 Jun 2009 02:00:46 +0200 Subject: [csw-maintainers] Heads up: freenode has disabled mibbit access Message-ID: <4A3C26AE.1070601@opencsw.org> Hi, freenode has blocked access via mibbit starting yesterday [1], so don't be surprised when your mibbit sessions ends with "Terminated" soon after you hit "Connect". There is an alternative, a freenode-operated webchat based on qwebirc that can be reached at [2]. Phil, could you please adjust our main page. Sebastian [1]http://blog.freenode.net/2009/06/new-freenode-webchat-and-why-to-use-it/ [2]http://webchat.freenode.net/?channels=opencsw From dam at opencsw.org Sat Jun 20 10:37:37 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sat, 20 Jun 2009 10:37:37 +0200 Subject: [csw-maintainers] We'll always have Paris Message-ID: <42FADF04-9E67-43EC-B4F5-FAA5D9F3CAD0@opencsw.org> Hi, I'll be to a training in Paris next week. Maybe someone wants to have a beer in the evening? William? Best regards -- Dago From william at wbonnet.net Sat Jun 20 11:03:52 2009 From: william at wbonnet.net (William Bonnet) Date: Sat, 20 Jun 2009 11:03:52 +0200 Subject: [csw-maintainers] We'll always have Paris In-Reply-To: <42FADF04-9E67-43EC-B4F5-FAA5D9F3CAD0@opencsw.org> References: <42FADF04-9E67-43EC-B4F5-FAA5D9F3CAD0@opencsw.org> Message-ID: <4A3CA5F8.2040808@wbonnet.net> Hi > I'll be to a training in Paris next week. Maybe someone > wants to have a beer in the evening? William? Yes for sure :) Which days are you staying in Paris ? cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From ihsan at opencsw.org Sat Jun 20 12:56:47 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 20 Jun 2009 12:56:47 +0200 Subject: [csw-maintainers] Heads up: freenode has disabled mibbit access In-Reply-To: <4A3C26AE.1070601@opencsw.org> References: <4A3C26AE.1070601@opencsw.org> Message-ID: <4A3CC06F.10503@opencsw.org> Am 20.6.2009 2:00 Uhr, Sebastian Kayser schrieb: > freenode has blocked access via mibbit starting yesterday [1], so don't > be surprised when your mibbit sessions ends with "Terminated" soon after > you hit "Connect". There is an alternative, a freenode-operated webchat > based on qwebirc that can be reached at [2]. Phil, could you please > adjust our main page. I just updated the website. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sat Jun 20 12:56:48 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 20 Jun 2009 12:56:48 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp: Schedule & Topics Message-ID: <4A3CC070.306@opencsw.org> Hello, I've updated the Wiki with the schedule: http://wiki.opencsw.org/summercamp-2009 This is actually only a draft, we still can change it. @Trygve: Do you think the Schedule is realistic? I didn't know how much time we need from the hotel to the place where it takes place, so I've just wrote something. There are also a few topics that we couldn't discuss in Zurich. Even if it's mainly a technical meeting, it would be really nice if we could talk also about the topics that have been suggested for Zurich. --> http://wiki.opencsw.org/suggestions The topics for now are: - Stable collection - Release process - Tiered packages - Download stats - New website - License discussion and choice I would suggest, that we make a list with the topics that we have to discuss and set priorities. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sat Jun 20 20:48:42 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 20 Jun 2009 20:48:42 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp: Schedule & Topics In-Reply-To: <4A3CC070.306@opencsw.org> References: <4A3CC070.306@opencsw.org> Message-ID: <4A3D2F0A.2040008@opencsw.org> Am 20.6.2009 12:56 Uhr, Ihsan Dogan schrieb: > - Release process I was just pointed out that we have discussed the release process already and that it's already mentioned on the wiki: http://wiki.opencsw.org/release-process I would say, we remove this topic then. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sun Jun 21 15:06:52 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 21 Jun 2009 15:06:52 +0200 Subject: [csw-maintainers] /testing clamav 0.95.2 In-Reply-To: <625385e30906110613v59a347f1xd24f33ae77e2c406@mail.gmail.com> References: <625385e30906110613v59a347f1xd24f33ae77e2c406@mail.gmail.com> Message-ID: <4A3E306C.2050806@opencsw.org> Hello Peter, Am 11.6.2009 15:13 Uhr, Peter Bonivart schrieb: > A new version of Clam AntiVirus is in testing. > > http://mirror.opencsw.org/testing/clamav-0.95.2,REV=2009.06.11-SunOS5.8-sparc-CSW.pkg.gz > http://mirror.opencsw.org/testing/clamav-0.95.2,REV=2009.06.11-SunOS5.8-i386-CSW.pkg.gz > http://mirror.opencsw.org/testing/libclamav-0.95.2,REV=2009.06.11-SunOS5.8-sparc-CSW.pkg.gz > http://mirror.opencsw.org/testing/libclamav-0.95.2,REV=2009.06.11-SunOS5.8-i386-CSW.pkg.gz Thanks Peter, works fine. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From trygvis at opencsw.org Sun Jun 21 15:25:58 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 21 Jun 2009 15:25:58 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp: Schedule & Topics In-Reply-To: <4A3CC070.306@opencsw.org> References: <4A3CC070.306@opencsw.org> Message-ID: <4A3E34E6.2060400@opencsw.org> Ihsan Dogan wrote: > Hello, > > I've updated the Wiki with the schedule: > http://wiki.opencsw.org/summercamp-2009 > > This is actually only a draft, we still can change it. > > @Trygve: Do you think the Schedule is realistic? I didn't know how much > time we need from the hotel to the place where it takes place, so I've > just wrote something. I think that should be ok. The traffic planner say we use 32 minutes from the hotel to the (currently planned) location. -- Trygve From william at wbonnet.net Sun Jun 21 16:33:52 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 21 Jun 2009 16:33:52 +0200 Subject: [csw-maintainers] Using X11 from Sun against using our own version of X11 Message-ID: <4A3E44D0.1060406@wbonnet.net> Hi During the last weeks there was some discussion about reasons and advantages of using our own X11 packages against continuing to use Sun's X11 libs. The work of packaging our own version was started a few months ago when work on FF3 started. It pulled a lot of dependencies because of compilation issues with current libs (either outdated packages, or incompatible Sun's packages including X11). Here is some detailed information about the problem we are facing with using X11 from Sun. This looks like a good reason to start to use now our own X11 packages. I have started FF3 compilation on a freshly installed Solaris 9 box (thanks Dago ;) ), build9st. the conclusion i reached is that i cannot compile latest xrender on this box using Sun's X11. I encountered several problems. Here is the list of what i have done. Start with FF3 compilation -> it requires requires a recent cairo Recompile Cairo -> it requires renderdev Recompile our recent version of xrender -> configure fails, x11.pc is missing This is a know Solaris 9 know bug http://forums.sun.com/thread.jspa?threadID=5204312&tstart=0 http://bugs.opensolaris.org/view_bug.do?bug_id=6571762 The fix is to create by hand /usr/lib/pkgconfig/x11.pc. I did it and resumed compilation -> several *proto.pc are missing. Installation of xproto, kbproto and inputproto from our repository (also missing in Solaris 9) Recompile our recent version of xrender -> This time configure success -> Compilation fails "Picture.c", line 29: cannot find include file: I checked for this file and it is not present in Sun's packages (either Sol 8, 9 or 10 but it seems it is in opensolaris). This missing include file looks to me like a showstopper. Many recent software will use this include, and i have no guarantee it is the only missing file. Thus i can imagine porting effort will be heavy, and it is surely a good reason to use our own, recent version of (Xorg) X11 libs. cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From rupert at opencsw.org Sun Jun 21 20:04:22 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 21 Jun 2009 20:04:22 +0200 Subject: [csw-maintainers] recompiled mod_wsgi now in testing Message-ID: <6af4270906211104q5ffc4648tb2b9a6a3cda67d7@mail.gmail.com> recompiled mod_wsgi against python-2.6.2, is now in testing. rupert. From skayser at opencsw.org Sun Jun 21 22:06:24 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 21 Jun 2009 22:06:24 +0200 Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online Message-ID: <4A3E92C0.8050500@opencsw.org> Hi, to bring some more life to our IRC channel, i have extended an IRC bot framework with a couple of CSW-specific plugins today and hooked a bot named "ceeswie" up to #opencsw for testing purposes. You can either query her with commands or trigger her in the channel by prefixing @ to your commands. Give the following commands a try: @pkginfo trac @buglist trac @fsearch ssl.h There are also some straight-forward URL builders available (no validity checking) @url pkginfo trac @url gar trac @url bugs trac @url bug 3659 The help system (@help) is off-the-shelf and not yet adjusted. The bot is just a glimpse of what can be done. I had the idea recently, when someone came on the channel to request help with his trac installation that went bad on pkg upgrades. Instead of going off to the bug tracker to check for possible bugs (or referring the user there), the bot enables us to query for bugs right away in the channel. It's convenient and fun after all ;) Give it a try, let me know what you think. Expect the one or the other hickup, it's rather basic and doesn't yet handle all exceptions. If someone has prior experience in coding such bots and wants to roll an alternative one, feel free. Feature requests very welcome! Sebastian From maciej at opencsw.org Mon Jun 22 08:08:25 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Mon, 22 Jun 2009 07:08:25 +0100 Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online In-Reply-To: <4A3E92C0.8050500@opencsw.org> References: <4A3E92C0.8050500@opencsw.org> Message-ID: On Sun, Jun 21, 2009 at 9:06 PM, Sebastian Kayser wrote: > to bring some more life to our IRC channel, i have extended an IRC bot > framework Which one? > You can either query her with commands or trigger her in the channel by > prefixing @ to your commands. Give the following commands a try: > > ?@pkginfo trac > ?@buglist trac > ?@fsearch ssl.h I tried them - way cool! > Give it a try, let me know what you think. Expect the one or the other > hickup, it's rather basic and doesn't yet handle all exceptions. If > someone has prior experience in coding such bots and wants to roll an > alternative one, feel free. Feature requests very welcome! A suggestion: would you mind putting your code into the code repository? [1] A feature that comes to mind would be a wiki search: @wiki summer camp http://wiki.opencsw.org/summercamp-2009 - Summer Camp 2009 Maciej [1] http://opencsw.svn.sourceforge.net/viewvc/opencsw/ From dam at opencsw.org Mon Jun 22 08:54:14 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 22 Jun 2009 08:54:14 +0200 Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online In-Reply-To: References: <4A3E92C0.8050500@opencsw.org> Message-ID: <3E5D1D33-B325-47C9-9121-4030BC896ADB@opencsw.org> Hi Maciej, Am 22.06.2009 um 08:08 schrieb Maciej (Matchek) Blizinski: >> Give it a try, let me know what you think. Expect the one or the >> other >> hickup, it's rather basic and doesn't yet handle all exceptions. If >> someone has prior experience in coding such bots and wants to roll an >> alternative one, feel free. Feature requests very welcome! > > A suggestion: would you mind putting your code into the code > repository? [1] YOu can check it in here: http://opencsw.svn.sourceforge.net/viewvc/opencsw/ > A feature that comes to mind would be a wiki search: > > @wiki summer camp > http://wiki.opencsw.org/summercamp-2009 - Summer Camp 2009 and @bugs trac lists bugs, severity etc. Bets regards -- Dago From dam at opencsw.org Mon Jun 22 08:59:16 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 22 Jun 2009 08:59:16 +0200 Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online In-Reply-To: <3E5D1D33-B325-47C9-9121-4030BC896ADB@opencsw.org> References: <4A3E92C0.8050500@opencsw.org> <3E5D1D33-B325-47C9-9121-4030BC896ADB@opencsw.org> Message-ID: Hi, Am 22.06.2009 um 08:54 schrieb Dagobert Michelsen: > and > @bugs trac > lists bugs, severity etc. Ups, never mind. -- Dago From maciej at opencsw.org Mon Jun 22 12:42:08 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Mon, 22 Jun 2009 11:42:08 +0100 Subject: [csw-maintainers] Summer Camp 2009 (Please book your flights) In-Reply-To: <4A350506.1000804@opencsw.org> References: <4A350506.1000804@opencsw.org> Message-ID: On Sun, Jun 14, 2009 at 3:11 PM, Ihsan Dogan wrote: > Hello, > > As already mentioned, the date and the location is fixed. I've updated > the wiki as well, but the information is not complete. > > --> http://wiki.opencsw.org/summercamp-2009 > > The information for accommodation will be available soon. Due to a small mixup, I ended up with an extra, paid reservation for 1 person for 2 nights (Fri/Sat and Sat/Sun). I can't change or cancel it since it was a 'WebDeal' kind of reservation. The price was ?186, I'm willing to let it go for ?120. (I've called Thon and verified with them that it's okay to check a different person in.) Maciej From Darin.Perusich at cognigencorp.com Mon Jun 22 15:17:45 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Mon, 22 Jun 2009 09:17:45 -0400 Subject: [csw-maintainers] Solaris 9 system for test installs Message-ID: <4A3F8479.1000606@cognigencorp.com> Is there a Solaris 9 system in the build farm I can use to install a package? I received a a bug about installation on Sol9 and I don't have a Sol9 host anymore to test the installation. -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From skayser at opencsw.org Mon Jun 22 21:08:10 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 22 Jun 2009 21:08:10 +0200 (CEST) Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online In-Reply-To: References: <4A3E92C0.8050500@opencsw.org> Message-ID: <59133.194.246.122.22.1245697690.squirrel@ssl.skayser.de> Maciej (Matchek) Blizinski schrieb: > On Sun, Jun 21, 2009 at 9:06 PM, Sebastian Kayser > wrote: >> to bring some more life to our IRC channel, i have extended an IRC bot >> framework > > Which one? supybot, based on python. http://sourceforge.net/projects/supybot/ >> You can either query her with commands or trigger her in the channel by >> prefixing @ to your commands. Give the following commands a try: >> >> ??@pkginfo trac >> ??@buglist trac >> ??@fsearch ssl.h > > I tried them - way cool! :) >> Give it a try, let me know what you think. Expect the one or the other >> hickup, it's rather basic and doesn't yet handle all exceptions. If >> someone has prior experience in coding such bots and wants to roll an >> alternative one, feel free. Feature requests very welcome! > > A suggestion: would you mind putting your code into the code repository? > [1] Sure, will do so tomorrow. > A feature that comes to mind would be a wiki search: > > @wiki summer camp > http://wiki.opencsw.org/summercamp-2009 - Summer Camp 2009 Noted. Could also do that one for the GAR reference or maybe even a complete GAR wiki search. Thanks. Sebastian From maciej at opencsw.org Tue Jun 23 00:29:27 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Mon, 22 Jun 2009 23:29:27 +0100 Subject: [csw-maintainers] Summer Camp 2009 (Please book your flights) In-Reply-To: References: <4A350506.1000804@opencsw.org> Message-ID: On Mon, Jun 22, 2009 at 11:42 AM, Maciej (Matchek) Blizinski wrote: > Due to a small mixup, I ended up with an extra, paid reservation for 1 > person for 2 nights (Fri/Sat and Sat/Sun). This has been sorted out. Thanks! Maciej From bwalton at opencsw.org Tue Jun 23 01:56:19 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 22 Jun 2009 19:56:19 -0400 Subject: [csw-maintainers] Fwd: [bug-notifications] [mysql5 0003722]: Request for update to MySQL 5.1 Message-ID: <1245714826-sup-5508@ntdws12.chass.utoronto.ca> Hi All, I haven't had the time to work on mysql5 as I thought I might. The current GAR recipe builds and passes tests for i386 and sparc v8. It fails tests on both v9 and x86_64. I forget which is which now, but on one of the 64-bit platforms, the failed test is a simple 'example plugin' test. All others pass. The failures on the other platform are more serious. If anyone does have the time to take this, I'd appreciate it. If not, I'll see what I can do, but it likely won't be fast. Thanks -Ben --- Begin forwarded message from Mantis Bug Tracker --- From: Mantis Bug Tracker To: bug-notifications Date: Mon, 22 Jun 2009 16:46:07 -0400 Subject: [bug-notifications] [mysql5 0003722]: Request for update to MySQL 5.1 The following issue has been SUBMITTED. ====================================================================== http://www.opencsw.org/mantis/view.php?id=3722 ====================================================================== Reported By: bryanwgraham Assigned To: ====================================================================== Project: mysql5 Issue ID: 3722 Category: upgrade Reproducibility: always Severity: minor Priority: normal Status: new ====================================================================== Date Submitted: 2009-06-22 22:46 CEST Last Modified: 2009-06-22 22:46 CEST ====================================================================== Summary: Request for update to MySQL 5.1 Description: It's out. I'm told this is how you request a new version. ====================================================================== --- End forwarded message --- -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Tue Jun 23 04:29:43 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 22 Jun 2009 22:29:43 -0400 Subject: [csw-maintainers] Solaris 9 system for test installs In-Reply-To: <4A3F8479.1000606@cognigencorp.com> References: <4A3F8479.1000606@cognigencorp.com> Message-ID: <1245724130-sup-6322@ntdws12.chass.utoronto.ca> Excerpts from Darin Perusich's message of Mon Jun 22 09:17:45 -0400 2009: Hi Darin, > Is there a Solaris 9 system in the build farm I can use to install a > package? I received a a bug about installation on Sol9 and I don't have > a Sol9 host anymore to test the installation. I believe there is now a build9st box. You can either mail requests to buildfarm at opencsw or ask Dago for sudo access on that box. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Tue Jun 23 19:32:49 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 23 Jun 2009 12:32:49 -0500 Subject: [csw-maintainers] updating berkeleydb to 4.7.25 on build8st and build8xt Message-ID: <4A4111C1.3080602@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I am going to update to the latest berkeley DB package that Dago put in testing. on the solaris 8 test boxes. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpBEcEACgkQLrhmsXMSLxeLoACgx+nYguH0TpgKS1zCDOzVNywo IAgAnR57650z9joRHRwjJbWdnHc18b+I =Vs9c -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue Jun 23 19:34:28 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 23 Jun 2009 12:34:28 -0500 Subject: [csw-maintainers] updating berkeleydb to 4.7.25 on build8st and build8xt In-Reply-To: <4A4111C1.3080602@opencsw.org> References: <4A4111C1.3080602@opencsw.org> Message-ID: <4A411224.7060405@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > I am going to update to the latest berkeley DB package that Dago put in > testing. on the solaris 8 test boxes. > > Nevermind, someone beat me to it ... ;-) _______________________________________________ maintainers mailing list maintainers at lists.opencsw.org https://lists.opencsw.org/mailman/listinfo/maintainers - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpBEiQACgkQLrhmsXMSLxeS/gCfb4/3sXReuFe5mbtq25C7V7oQ FqEAoJ8omHt5XJuAfveMnxc0Wdr+sSft =eBia -----END PGP SIGNATURE----- From phil at bolthole.com Tue Jun 23 20:12:57 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 23 Jun 2009 11:12:57 -0700 Subject: [csw-maintainers] updating berkeleydb to 4.7.25 on build8st and build8xt In-Reply-To: <4A4111C1.3080602@opencsw.org> References: <4A4111C1.3080602@opencsw.org> Message-ID: <20090623181257.GA56726@bolthole.com> On Tue, Jun 23, 2009 at 12:32:49PM -0500, Mike Watters wrote: > > I am going to update to the latest berkeley DB package that Dago put in > testing. on the solaris 8 test boxes. becuase someone may not be aware that we even HAVE "test boxes", a clarification: we have test boxes, that are separate from our build boxes. the test boxes are build8st build8xt (this is a bit silly... I personally think they should more clearly be named, such as "test8s" and "test8x", but the above, is what they are named at the moment) From skayser at opencsw.org Tue Jun 23 22:17:44 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 23 Jun 2009 22:17:44 +0200 (CEST) Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online In-Reply-To: References: <4A3E92C0.8050500@opencsw.org> Message-ID: <56108.194.246.122.22.1245788264.squirrel@ssl.skayser.de> Maciej (Matchek) Blizinski wrote: > On Sun, Jun 21, 2009 at 9:06 PM, Sebastian Kayser > wrote: >> to bring some more life to our IRC channel, i have extended an IRC bot >> framework > > A suggestion: would you mind putting your code into the code repository? > [1] Done. Please do note: i am not a programmer and touched python for the first time last weekend. That being said, feel free to polish things up. Sebastian > [1] http://opencsw.svn.sourceforge.net/viewvc/opencsw/ From william at wbonnet.net Wed Jun 24 07:53:25 2009 From: william at wbonnet.net (William Bonnet) Date: Wed, 24 Jun 2009 07:53:25 +0200 Subject: [csw-maintainers] Thunderbird 2.0.0.22 is in testing Message-ID: <4A41BF55.9060808@wbonnet.net> Hi all I have updated Thunderbird package to the latest release. It is available from testing. Feedbacks are welcome cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From bonivart at opencsw.org Wed Jun 24 14:32:15 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 24 Jun 2009 14:32:15 +0200 Subject: [csw-maintainers] Minor web site change? Message-ID: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> On the web site we have this written: "CSW (Community SoftWare) is a free, volunteer effort, to provide quality downloadable binary packages for the currently fully supported versions of Sun Solaris (Currently, 8, 9, and 10)." Maybe we should rephrase that somehow now that Solaris 8 is not fully supported any more. -- /peter From pfelecan at opencsw.org Wed Jun 24 15:27:19 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 24 Jun 2009 15:27:19 +0200 Subject: [csw-maintainers] updating berkeleydb to 4.7.25 on build8st and build8xt In-Reply-To: <20090623181257.GA56726@bolthole.com> (Philip Brown's message of "Tue\, 23 Jun 2009 11\:12\:57 -0700") References: <4A4111C1.3080602@opencsw.org> <20090623181257.GA56726@bolthole.com> Message-ID: Philip Brown writes: > build8st > build8xt > > (this is a bit silly... I personally think they should more clearly be > named, such as "test8s" and "test8x", but the above, is what they are > named at the moment) +1 although they are really test build systems but I'm not proposing to name them testbuild8s, but wait, it's not so long... -- Peter From pfelecan at opencsw.org Wed Jun 24 15:28:50 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 24 Jun 2009 15:28:50 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> (Peter Bonivart's message of "Wed\, 24 Jun 2009 14\:32\:15 +0200") References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> Message-ID: Peter Bonivart writes: > On the web site we have this written: > > "CSW (Community SoftWare) is a free, volunteer effort, to provide > quality downloadable binary packages for the currently fully supported > versions of Sun Solaris (Currently, 8, 9, and 10)." > > Maybe we should rephrase that somehow now that Solaris 8 is not fully > supported any more. What about: ..."versions of Sun Solaris (when possible 8, but always 9, and 10)." -- Peter From bwalton at opencsw.org Wed Jun 24 15:49:05 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 24 Jun 2009 09:49:05 -0400 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> Message-ID: <1245851300-sup-4398@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Wed Jun 24 09:28:50 -0400 2009: > > "CSW (Community SoftWare) is a free, volunteer effort, to provide > > quality downloadable binary packages for the currently fully supported > > versions of Sun Solaris (Currently, 8, 9, and 10)." > > > ..."versions of Sun Solaris (when possible 8, but always 9, and 10)." Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From bonivart at opencsw.org Wed Jun 24 15:55:30 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 24 Jun 2009 15:55:30 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <1245851300-sup-4398@ntdws12.chass.utoronto.ca> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> Message-ID: <625385e30906240655t7278640br542b6ff0d0757afc@mail.gmail.com> On Wed, Jun 24, 2009 at 3:49 PM, Ben Walton wrote: > Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." "versions 9 and 10 of Sun Solaris (and 8, when possible) for both Sparc and x86." ??? -- /peter From bwalton at opencsw.org Wed Jun 24 16:05:10 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 24 Jun 2009 10:05:10 -0400 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <625385e30906240655t7278640br542b6ff0d0757afc@mail.gmail.com> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <625385e30906240655t7278640br542b6ff0d0757afc@mail.gmail.com> Message-ID: <1245852281-sup-2721@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Wed Jun 24 09:55:30 -0400 2009: > On Wed, Jun 24, 2009 at 3:49 PM, Ben Walton wrote: > > Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." > > "versions 9 and 10 of Sun Solaris (and 8, when possible) for both > Sparc and x86." +1 -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From pfelecan at opencsw.org Wed Jun 24 16:11:50 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 24 Jun 2009 16:11:50 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <625385e30906240655t7278640br542b6ff0d0757afc@mail.gmail.com> (Peter Bonivart's message of "Wed\, 24 Jun 2009 15\:55\:30 +0200") References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <625385e30906240655t7278640br542b6ff0d0757afc@mail.gmail.com> Message-ID: Peter Bonivart writes: > On Wed, Jun 24, 2009 at 3:49 PM, Ben Walton wrote: >> Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." > > "versions 9 and 10 of Sun Solaris (and 8, when possible) for both > Sparc and x86." +1 -- Peter From james at opencsw.org Wed Jun 24 16:30:22 2009 From: james at opencsw.org (James Lee) Date: Wed, 24 Jun 2009 14:30:22 GMT Subject: [csw-maintainers] Minor web site change? In-Reply-To: <1245851300-sup-4398@ntdws12.chass.utoronto.ca> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> Message-ID: <20090624.14302200.1757446011@gyor.oxdrove.co.uk> On 24/06/09, 14:49:05, Ben Walton wrote regarding Re: [csw-maintainers] Minor web site change?: > Excerpts from Peter FELECAN's message of Wed Jun 24 09:28:50 -0400 2009: > > > "CSW (Community SoftWare) is a free, volunteer effort, to provide > > > quality downloadable binary packages for the currently fully supported > > > versions of Sun Solaris (Currently, 8, 9, and 10)." > > > > > ..."versions of Sun Solaris (when possible 8, but always 9, and 10)." > Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." Except this isn't true, or at best it's misleading. Currently there are no 5.9 packages so 5.9 support is by 5.8 packages, thus 5.8 is as fully supported as 5.9 (a statement of truth about the present). The statement needs to distinguish the aims (CSW's commitment to support Sun supported systems now and in future) and the offerings (what is currently happening). James. From bonivart at opencsw.org Wed Jun 24 16:42:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 24 Jun 2009 16:42:14 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <20090624.14302200.1757446011@gyor.oxdrove.co.uk> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <20090624.14302200.1757446011@gyor.oxdrove.co.uk> Message-ID: <625385e30906240742u3e44cb91u6ee86fee4eebf6f@mail.gmail.com> On Wed, Jun 24, 2009 at 4:30 PM, James Lee wrote: > The statement needs to distinguish the aims (CSW's commitment to support > Sun supported systems now and in future) and the offerings (what is > currently happening). Any suggestions? :-) -- /peter From ihsan at dogan.ch Wed Jun 24 16:46:15 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Wed, 24 Jun 2009 16:46:15 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <1245851300-sup-4398@ntdws12.chass.utoronto.ca> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> Message-ID: <4A423C37.2010101@dogan.ch> Ben Walton wrote: >>> "CSW (Community SoftWare) is a free, volunteer effort, to provide >>> quality downloadable binary packages for the currently fully supported >>> versions of Sun Solaris (Currently, 8, 9, and 10)." >> ..."versions of Sun Solaris (when possible 8, but always 9, and 10)." > > Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." That might be confusing. I would not mention Solaris 8 together with 9 and 10. It should be clearly mentioned, that the Software might run on Solaris 8, but Solaris 9 is the lowest supported version. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Wed Jun 24 17:06:59 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 24 Jun 2009 11:06:59 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090619205551.GC12157@bolthole.com> References: <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> <20090619173419.GA77530@bolthole.com> <1245433062-sup-2821@ntdws12.chass.utoronto.ca> <20090619205551.GC12157@bolthole.com> Message-ID: <1245855797-sup-5350@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 19 16:55:51 -0400 2009: > ok, sounds like a working agreement. I think you are proposing, > "lets focus on moving what is easily movable, from /opt/csw/etc, to > /etc/opt/csw, and then review what is left after that point". > > I am very happy to move forward on that premise :-) Yes, lets do this. I think we should notify users@ of this upcoming change, though. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Wed Jun 24 18:38:55 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 24 Jun 2009 09:38:55 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245855797-sup-5350@ntdws12.chass.utoronto.ca> References: <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> <20090619173419.GA77530@bolthole.com> <1245433062-sup-2821@ntdws12.chass.utoronto.ca> <20090619205551.GC12157@bolthole.com> <1245855797-sup-5350@ntdws12.chass.utoronto.ca> Message-ID: <20090624163855.GC54688@bolthole.com> On Wed, Jun 24, 2009 at 11:06:59AM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Fri Jun 19 16:55:51 -0400 2009: > > ok, sounds like a working agreement. I think you are proposing, > > "lets focus on moving what is easily movable, from /opt/csw/etc, to > > /etc/opt/csw, and then review what is left after that point". > > > > I am very happy to move forward on that premise :-) > > Yes, lets do this. I think we should notify users@ of this upcoming > change, though. it's not a policy change. it would be bringing packages in line with existing guidelines, on a package-by-package basis. the maintainers of each affected package, should make the relevant announcement for their package, when the time comes From bwalton at opencsw.org Wed Jun 24 18:44:11 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 24 Jun 2009 12:44:11 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090624163855.GC54688@bolthole.com> References: <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> <20090619173419.GA77530@bolthole.com> <1245433062-sup-2821@ntdws12.chass.utoronto.ca> <20090619205551.GC12157@bolthole.com> <1245855797-sup-5350@ntdws12.chass.utoronto.ca> <20090624163855.GC54688@bolthole.com> Message-ID: <1245861741-sup-1660@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed Jun 24 12:38:55 -0400 2009: > the maintainers of each affected package, should make the relevant > announcement for their package, when the time comes Ok, this works for me. I think it will actually have a higher impact, since it will be at 'time of delivery' instead of something that might be read and forgotten. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From maciej at opencsw.org Wed Jun 24 19:07:17 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Wed, 24 Jun 2009 18:07:17 +0100 Subject: [csw-maintainers] Debugging with Sun Studio and mGAR Message-ID: I'm currently working on debugging an issue with cups-polld. At the present stage, I'm about to learn how to build all the binaries with debugging symbols and run them under the control of Sun Studio. I'm guessing I'll be using a 'project with existing sources', pointing at the work/${package}-${version} directory and not calling 'gmake clean' not to nuke the compiled objects or binaries. Has anyone here done this kind of work before? If so, do you have any tips or know any gotchas about the process? Maciej From mwatters at opencsw.org Wed Jun 24 19:08:07 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 24 Jun 2009 12:08:07 -0500 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <4A423C37.2010101@dogan.ch> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> Message-ID: <4A425D77.3020103@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ihsan Dogan wrote: > Ben Walton wrote: > >>>> "CSW (Community SoftWare) is a free, volunteer effort, to provide >>>> quality downloadable binary packages for the currently fully supported >>>> versions of Sun Solaris (Currently, 8, 9, and 10)." >>> ..."versions of Sun Solaris (when possible 8, but always 9, and 10)." >> Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." > > That might be confusing. I would not mention Solaris 8 together with 9 > and 10. It should be clearly mentioned, that the Software might run on > Solaris 8, but Solaris 9 is the lowest supported version. > > > > > Ihsan > I know baseline is solaris 9 but solaris 8 is "best effort", but when will we drop the "best attempt"? I think we should leave everything as-is unless we are no longer going to create packages specifically for solaris 8. I suggest the following. as we roll out new versions of the packages, no longer attempt solaris 8 compile. compile for solaris 9+ only. Freeze the existing solaris 8 software catalog and keep for those users that want to run solaris 8 unsupported by sun. if the users want an upgrade to the software in the solaris 8 catalog, they can fill a bug report or request on the mailing list and we will apply the "best effort" to create that package and update the catalog. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpCXXcACgkQLrhmsXMSLxdf1gCfQ8C+1ZDHaj1H75HWlcmILsDc dC0An2CPwYnbzUQ0ZMpzhzzhMP+H86JY =On6N -----END PGP SIGNATURE----- From william at wbonnet.net Wed Jun 24 20:35:07 2009 From: william at wbonnet.net (William Bonnet) Date: Wed, 24 Jun 2009 20:35:07 +0200 Subject: [csw-maintainers] updating berkeleydb to 4.7.25 on build8st and build8xt In-Reply-To: References: <4A4111C1.3080602@opencsw.org> <20090623181257.GA56726@bolthole.com> Message-ID: <4A4271DB.1080409@wbonnet.net> Hi >> build8st >> build8xt >> >> (this is a bit silly... I personally think they should more clearly be >> named, such as "test8s" and "test8x", but the above, is what they are >> named at the moment) >> > > +1 although they are really test build systems but I'm not proposing to > name them testbuild8s, but wait, it's not so long... > I like test8s and test8x. Names and not ambiguous and shorter to type ;) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Wed Jun 24 21:36:43 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 24 Jun 2009 21:36:43 +0200 (CEST) Subject: [csw-maintainers] Debugging with Sun Studio and mGAR In-Reply-To: References: Message-ID: <54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de> Maciej (Matchek) Blizinski wrote: > I'm currently working on debugging an issue with cups-polld. At the > present stage, I'm about to learn how to build all the binaries with > debugging symbols and run them under the control of Sun Studio. I guess it doesn't fully cover your answer, but regarding GAR, you can set GARFLAVOR to DBG if you want your binaries built with debugging symbols. This will also skip the step where GAR strips all the binaries. See [1]. Of course this requires your program to honour CFLAGS correctly, because that is what GARFLAVOR changes. I used it to build packages that i wanted to debug with dbx later on (for example because i had seen crashes). Sebastian [1] http://sourceforge.net/apps/trac/gar/wiki/GAR%20Variable%20Reference#CompilationSettings From bwalton at opencsw.org Thu Jun 25 04:46:51 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 24 Jun 2009 22:46:51 -0400 Subject: [csw-maintainers] package hooks Message-ID: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> Hi All, I thought I'd get the discussion of hooks for the package tools going by posting a bit of code. This is completely untested, but is encapsulates the vision I have of how hooks could be implemented. [Please excuse my rusty perl.] The pkgutil patch implements pre/post install/remove hook capabilities that allow the admin (or other packages) to have code run at various points during the management of CSW packages. It does not differentiate between install and upgrade and it implements 'pre' and 'post' in the batch sense, not the per package sense. More granularity could be had by running hooks at each invocation of pkgadd/pkgrm, but I think that might be overkill. Thoughts? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: pkgutil_cswhooks.patch Type: application/octet-stream Size: 1823 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Thu Jun 25 09:01:17 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 09:01:17 +0200 Subject: [csw-maintainers] Debugging with Sun Studio and mGAR In-Reply-To: <54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de> References: <54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de> Message-ID: Hi Sebastian, Am 24.06.2009 um 21:36 schrieb Sebastian Kayser: > Maciej (Matchek) Blizinski wrote: >> I'm currently working on debugging an issue with cups-polld. At the >> present stage, I'm about to learn how to build all the binaries with >> debugging symbols and run them under the control of Sun Studio. > > I guess it doesn't fully cover your answer, but regarding GAR, you > can set > GARFLAVOR to DBG if you want your binaries built with debugging > symbols. > This will also skip the step where GAR strips all the binaries. See > [1]. Yes, and you can set it transienstly with gmake clean && GARFLAVOR=DBG gmake package Best regards -- Dago From dam at opencsw.org Thu Jun 25 09:03:39 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 09:03:39 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <4A425D77.3020103@opencsw.org> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> Message-ID: <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Hi Mike, Am 24.06.2009 um 19:08 schrieb Mike Watters: > I suggest the following. > as we roll out new versions of the packages, no longer attempt > solaris 8 > compile. compile for solaris 9+ only. Why? Usually it is no different packaging for Solaris 9 than for Solaris 8. Why drop Solaris 8 without the need? If there is *any* problem on Solaris 8 it is of course ok to just release for Solaris 9. Best regards -- Dago From bonivart at opencsw.org Thu Jun 25 09:54:32 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 25 Jun 2009 09:54:32 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <625385e30906250054x3cc31a8fqdd2c4278009d195b@mail.gmail.com> On Thu, Jun 25, 2009 at 9:03 AM, Dagobert Michelsen wrote: > Why? Usually it is no different packaging for Solaris 9 than for > Solaris 8. Why drop Solaris 8 without the need? If there is *any* > problem on Solaris 8 it is of course ok to just release for Solaris 9. We still need to rephrase the claim that Solaris 8 is fully supported by Sun. Maybe just remove the part about Sun..? "...provide quality downloadable binary packages for Sun Solaris 8, 9 and 10, both Sparc and x86." -- /peter From bonivart at opencsw.org Thu Jun 25 10:31:10 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 25 Jun 2009 10:31:10 +0200 Subject: [csw-maintainers] package hooks In-Reply-To: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> Message-ID: <625385e30906250131n4daa51car2a44bbaaf0f80e4c@mail.gmail.com> On Thu, Jun 25, 2009 at 4:46 AM, Ben Walton wrote: > I thought I'd get the discussion of hooks for the package tools going > by posting a bit of code. ?This is completely untested, but is > encapsulates the vision I have of how hooks could be implemented. > [Please excuse my rusty perl.] Looks nice, thanks for the patch! I'll add it to 1.7. :-) > The pkgutil patch implements pre/post install/remove hook capabilities > that allow the admin (or other packages) to have code run at various > points during the management of CSW packages. And these hooks will be provided by packages like etckeeper? -- /peter From james at opencsw.org Thu Jun 25 11:16:59 2009 From: james at opencsw.org (James Lee) Date: Thu, 25 Jun 2009 09:16:59 GMT Subject: [csw-maintainers] Minor web site change? In-Reply-To: <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <20090625.9165900.3503357047@gyor.oxdrove.co.uk> On 25/06/09, 08:03:39, Dagobert Michelsen wrote regarding Re: [csw-maintainers] Minor web site change?: > > I suggest the following. > > as we roll out new versions of the packages, no longer attempt > > solaris 8 > > compile. compile for solaris 9+ only. > Why? Usually it is no different packaging for Solaris 9 than for > Solaris 8. Why drop Solaris 8 without the need? Studio 12. > If there is *any* > problem on Solaris 8 it is of course ok to just release for Solaris 9. Except it's not. If a single useful package is missing it scuppers your whole installation. Due to the strong interdependence of the CSW system we can't realistically pick and choose and offer partial support. Further, we should remain egalitarian and not decide what is important to others, i.e. if just one package is missing we no longer support Solaris 8. James. From james at opencsw.org Thu Jun 25 11:17:00 2009 From: james at opencsw.org (James Lee) Date: Thu, 25 Jun 2009 09:17:00 GMT Subject: [csw-maintainers] Minor web site change? In-Reply-To: <625385e30906240742u3e44cb91u6ee86fee4eebf6f@mail.gmail.com> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <20090624.14302200.1757446011@gyor.oxdrove.co.uk> <625385e30906240742u3e44cb91u6ee86fee4eebf6f@mail.gmail.com> Message-ID: <20090625.9170000.106433457@gyor.oxdrove.co.uk> On 24/06/09, 15:42:14, Peter Bonivart wrote regarding Re: [csw-maintainers] Minor web site change?: > On Wed, Jun 24, 2009 at 4:30 PM, James Lee wrote: > > The statement needs to distinguish the aims (CSW's commitment to support > > Sun supported systems now and in future) and the offerings (what is > > currently happening). > Any suggestions? :-) CSW (Community SoftWare) is a free, volunteer effort, to provide quality downloadable Sparc and x86 binary packages for Sun Solaris. CSW is committed to supporting version of Solaris fully supported by Sun (at present 9 and 10) [1]. CSW's packages currently also work on Solaris 8 however users are urged to migrate to Solaris 9 or 10 as the Solaris 8 packages may be dropped soon. [2] [1] http://www.sun.com/software/solaris/releases.jsp (or similar) [2] Reference/link to time table and methodologies for announcements. James. From glaw at opencsw.org Thu Jun 25 11:39:45 2009 From: glaw at opencsw.org (Gary Law) Date: Thu, 25 Jun 2009 10:39:45 +0100 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: 2009/6/25 Dagobert Michelsen : > Why? Usually it is no different packaging for Solaris 9 than for > Solaris 8. Why drop Solaris 8 without the need? If there is *any* > problem on Solaris 8 it is of course ok to just release for Solaris 9. The problem for me is testing. Rolling out something that compiles is all very well and good, but if it's not tested on Sol8 then it might be buggy, and might in fact be worse than the old Sol8 version we have in the archive. Finding the hardware/virtualisation kit for testing for Sol8 is hard enough now, and it's only going to get worse. Gary -- Gary Law glaw at opencsw.org From glaw at opencsw.org Thu Jun 25 11:40:35 2009 From: glaw at opencsw.org (Gary Law) Date: Thu, 25 Jun 2009 10:40:35 +0100 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <20090625.9170000.106433457@gyor.oxdrove.co.uk> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <20090624.14302200.1757446011@gyor.oxdrove.co.uk> <625385e30906240742u3e44cb91u6ee86fee4eebf6f@mail.gmail.com> <20090625.9170000.106433457@gyor.oxdrove.co.uk> Message-ID: 2009/6/25 James Lee : > CSW (Community SoftWare) is a free, volunteer effort, to provide > quality downloadable Sparc and x86 binary packages for Sun Solaris. > > CSW is committed to supporting version of Solaris fully supported > by Sun (at present 9 and 10) [1]. ?CSW's packages currently also work > on Solaris 8 however users are urged to migrate to Solaris 9 or 10 > as the Solaris 8 packages may be dropped soon. [2] > > [1] http://www.sun.com/software/solaris/releases.jsp > (or similar) > > [2] Reference/link to time table and methodologies for announcements. +1 -- Gary Law glaw at opencsw.org From dam at opencsw.org Thu Jun 25 11:47:05 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 11:47:05 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: Hi Gary, Am 25.06.2009 um 11:39 schrieb Gary Law: > 2009/6/25 Dagobert Michelsen : >> Why? Usually it is no different packaging for Solaris 9 than for >> Solaris 8. Why drop Solaris 8 without the need? If there is *any* >> problem on Solaris 8 it is of course ok to just release for Solaris >> 9. > > The problem for me is testing. Rolling out something that compiles is > all very well and good, but if it's not tested on Sol8 then it might > be buggy, and might in fact be worse than the old Sol8 version we have > in the archive. Finding the hardware/virtualisation kit for testing > for Sol8 is hard enough now, and it's only going to get worse. This will be mitigated when the ESX farm has been upgraded to vSphere 4. It should then be possible to assign dedicated package install testing machines for all operating systems. The other reasons however stay very valid. Best regards -- Dago From pfelecan at opencsw.org Thu Jun 25 14:12:24 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 25 Jun 2009 14:12:24 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <625385e30906250054x3cc31a8fqdd2c4278009d195b@mail.gmail.com> (Peter Bonivart's message of "Thu\, 25 Jun 2009 09\:54\:32 +0200") References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> <625385e30906250054x3cc31a8fqdd2c4278009d195b@mail.gmail.com> Message-ID: Peter Bonivart writes: > On Thu, Jun 25, 2009 at 9:03 AM, Dagobert Michelsen wrote: >> Why? Usually it is no different packaging for Solaris 9 than for >> Solaris 8. Why drop Solaris 8 without the need? If there is *any* >> problem on Solaris 8 it is of course ok to just release for Solaris 9. > > We still need to rephrase the claim that Solaris 8 is fully supported > by Sun. Maybe just remove the part about Sun..? > > "...provide quality downloadable binary packages for Sun Solaris 8, 9 > and 10, both Sparc and x86." We just agreed on a rephrasing which seems alright: "versions 9 and 10 of Sun Solaris (and 8, when possible) for both Sparc and x86." What's the issue with that? -- Peter From bwalton at opencsw.org Thu Jun 25 15:18:35 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 09:18:35 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <625385e30906250131n4daa51car2a44bbaaf0f80e4c@mail.gmail.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <625385e30906250131n4daa51car2a44bbaaf0f80e4c@mail.gmail.com> Message-ID: <1245934781-sup-9640@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Thu Jun 25 04:31:10 -0400 2009: > On Thu, Jun 25, 2009 at 4:46 AM, Ben Walton wrote: > > I thought I'd get the discussion of hooks for the package tools going > > by posting a bit of code. ?This is completely untested, but is > > encapsulates the vision I have of how hooks could be implemented. > > [Please excuse my rusty perl.] > > Looks nice, thanks for the patch! I'll add it to 1.7. :-) Cool! :) Are there points of discussion before it gets 'adopted' though? Should install be differentiated from upgrade? Should the hooks run per-package? Are there any other useful points that should be hooked? Is the path I picked (/etc/opt/csw/pkg-hooks) acceptable? Since this system would (presumably) be implemented by any csw utility that is going to add/remove packages, we should all be happy with the initial choices up front...not to say we couldn't change things later, but smart choices are definitely better. > > The pkgutil patch implements pre/post install/remove hook capabilities > > that allow the admin (or other packages) to have code run at various > > points during the management of CSW packages. > > And these hooks will be provided by packages like etckeeper? Etckeeper will likely be the first to leverage the hooks. On my Ubuntu and RHEL systems, there are various other things that hook into the provided plugin systems. With RHEL5, Redhat moved away from their up2date system and simply provides a yum plugin (rhnplugin) that hooks into the facilities provided by yum. There is also a downloadonly yum plugin that I've installed. Things like the Ubuntu update-notifier use the hooks to maintain a timestamp so that if you've just done a manual package update, it won't bother checking again, etc. As long as the system is flexible enough, there are lots of creative things that can be done. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pfelecan at opencsw.org Thu Jun 25 15:26:45 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 25 Jun 2009 15:26:45 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> (Dagobert Michelsen's message of "Thu\, 25 Jun 2009 09\:03\:39 +0200") References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: Dagobert Michelsen writes: > Hi Mike, > > Am 24.06.2009 um 19:08 schrieb Mike Watters: >> I suggest the following. >> as we roll out new versions of the packages, no longer attempt >> solaris 8 >> compile. compile for solaris 9+ only. > > Why? Usually it is no different packaging for Solaris 9 than for > Solaris 8. Why drop Solaris 8 without the need? If there is *any* > problem on Solaris 8 it is of course ok to just release for Solaris 9. This can be true for pre Solaris 8 packages also. In my opinion it's a question of policy. More than 1 year ago we had this discussion, in a more extreme conditions, and the promise was made that when Solaris 8 is no more supported by SUN we drop also the support. Let it be known that I'm really reluctant to use the SUN compiler until we cannot use the last version of it, i.e. SUN Studio 12. May I suggest to have a ballot on this issue? -- Peter From pfelecan at opencsw.org Thu Jun 25 15:31:54 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 25 Jun 2009 15:31:54 +0200 Subject: [csw-maintainers] issue with mail sent by mantis Message-ID: The mail delivered by mantis has a systematic issue, detected by Amavis and others: X-Virus-Scanned: amavisd-new at opencsw.org X-Amavis-Alert: BAD HEADER SECTION, Duplicate header field: "Message-ID" Can this be corrected? -- Peter From dam at opencsw.org Thu Jun 25 15:38:13 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 15:38:13 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <5954B1F4-D33A-4B36-A5B4-3CBFA484F488@opencsw.org> Hi Peter, Am 25.06.2009 um 15:26 schrieb Peter FELECAN: > This can be true for pre Solaris 8 packages also. In my opinion it's a > question of policy. More than 1 year ago we had this discussion, in a > more extreme conditions, and the promise was made that when Solaris > 8 is > no more supported by SUN we drop also the support. Let it be known > that > I'm really reluctant to use the SUN compiler until we cannot use the > last version of it, i.e. SUN Studio 12. May I suggest to have a ballot > on this issue? No need to. If you don't want to package for Solaris 8 it is perfectly ok to start with Solaris 9 and Sun Studio 12. I was just saying that it is still ok to package for Solaris 8 and the catalog will still receive updates (but not all). Best regards -- Dago PS: Are you in for the beer tonight with William? From mwatters at opencsw.org Thu Jun 25 15:39:24 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 25 Jun 2009 08:39:24 -0500 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <4A437E0C.4070101@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter FELECAN wrote: > Dagobert Michelsen writes: > >> Hi Mike, >> >> Am 24.06.2009 um 19:08 schrieb Mike Watters: >>> I suggest the following. >>> as we roll out new versions of the packages, no longer attempt >>> solaris 8 >>> compile. compile for solaris 9+ only. >> Why? Usually it is no different packaging for Solaris 9 than for >> Solaris 8. Why drop Solaris 8 without the need? If there is *any* >> problem on Solaris 8 it is of course ok to just release for Solaris 9. > > This can be true for pre Solaris 8 packages also. In my opinion it's a > question of policy. More than 1 year ago we had this discussion, in a > more extreme conditions, and the promise was made that when Solaris 8 is > no more supported by SUN we drop also the support. Let it be known that > I'm really reluctant to use the SUN compiler until we cannot use the > last version of it, i.e. SUN Studio 12. May I suggest to have a ballot > on this issue? I prefer Studio 12, Sun made enhancements to that version to make it more gnu friendly. gnu apps that would not compile using studio 11 now have a great chance of compiling and running with studio 12. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpDfgsACgkQLrhmsXMSLxfs3gCgqEHWt+U/4KM2GLCSjHnhWgDH AjMAn1CvfkOwALgTLIliQJM6+99EDsEu =YLZE -----END PGP SIGNATURE----- From dam at opencsw.org Thu Jun 25 15:39:49 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 15:39:49 +0200 Subject: [csw-maintainers] package hooks In-Reply-To: <1245934781-sup-9640@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <625385e30906250131n4daa51car2a44bbaaf0f80e4c@mail.gmail.com> <1245934781-sup-9640@ntdws12.chass.utoronto.ca> Message-ID: <617118FB-4D91-4691-B4CE-0F19B5C97A2C@opencsw.org> Hi Ben, Am 25.06.2009 um 15:18 schrieb Ben Walton: > Are there points of discussion before it gets 'adopted' though? > Should install be differentiated from upgrade? Should the hooks run > per-package? Are there any other useful points that should be hooked? I guess we should talk about arguments passed, like the package name of the package currently modified. Additionally we should have a package "update hook" (useful for nginx, which allows uninterrupted update) and a "package purge hook" which brings the system to a pristine state before package installation. Best regards -- Dago From dam at opencsw.org Thu Jun 25 15:41:55 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 15:41:55 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <4A437E0C.4070101@opencsw.org> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> <4A437E0C.4070101@opencsw.org> Message-ID: Hi Mike, Am 25.06.2009 um 15:39 schrieb Mike Watters: > I prefer Studio 12, Sun made enhancements to that version to make it > more gnu > friendly. gnu apps that would not compile using studio 11 now have > a great > chance of compiling and running with studio 12. Again, this is perfectly ok. All I'm asking for is to try Solaris 8 and Studio 11. If there is any issue, just change the one line in GAR for the compiler change, login to build9 and retry. BTW, should I change the default compiler for Solaris 9 and 10 to Studio 12? Best regards -- Dago From bwalton at opencsw.org Thu Jun 25 15:43:08 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 09:43:08 -0400 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <1245936905-sup-7739@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Thu Jun 25 09:26:45 -0400 2009: > last version of it, i.e. SUN Studio 12. May I suggest to have a ballot > on this issue? I think a vote is in order too. Personally, I'll hate to lose sol8 support as I'm still running a few 8 boxes that depend on CSW. That being said, I think I'd still vote to drop the support, since keeping it allows people to continue hobbling along on the same crutch forever. Losing official support from CSW does not preclude individual sites building packages against the last released packages for 8 either, so just like running an unsupported OS, people can continue to run an unsupported CSW. Maybe somebody will spawn a CSW Legacy project ala the now defunct Fedora Legacy project. [I'm hoping to be completely moved away from 8 by the end of the summer, but we'll see.] Just my $0.02 CDN. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 25 16:17:11 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 25 Jun 2009 07:17:11 -0700 Subject: [csw-maintainers] package hooks In-Reply-To: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> Message-ID: <20090625141711.GA25001@bolthole.com> On Wed, Jun 24, 2009 at 10:46:51PM -0400, Ben Walton wrote: > > Hi All, > > I thought I'd get the discussion of hooks for the package tools going > by posting a bit of code. This is completely untested, but is > encapsulates the vision I have of how hooks could be implemented. > [Please excuse my rusty perl.] its always nice to have someone willing to code. However, this is an API we're talking about. It's usually best for an API to be driven by a specification, not just "well look at the code". So.. how about writing up some docs for the API now? :-) (I sugest putting it on the wiki) From bwalton at opencsw.org Thu Jun 25 16:31:39 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 10:31:39 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <617118FB-4D91-4691-B4CE-0F19B5C97A2C@opencsw.org> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <625385e30906250131n4daa51car2a44bbaaf0f80e4c@mail.gmail.com> <1245934781-sup-9640@ntdws12.chass.utoronto.ca> <617118FB-4D91-4691-B4CE-0F19B5C97A2C@opencsw.org> Message-ID: <1245939157-sup-8479@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Thu Jun 25 09:39:49 -0400 2009: Hi Dago, > I guess we should talk about arguments passed, like the package name > of the package currently modified. Additionally we should have a Ok, so you think the hooks should be run per-package instead of per-batch then? This is heavier, but more granular. On a more macro level, the package info could be provided to hooks through a defined 'protocol' whereby the tool would create a set of files (/var/opt/csw/lib/pkg-hooks/{installs,removes,upgrades}) that would be available to the hooks. As long as these files were created prior to the batch call, the hooks could use this as required. If all hooks are made per-package instead of per-batch, the protocol could simply be a single argument passed to the hook. That's much easier from an implementation perspective simply because there are no filesystem objects to look after. This brings up the issue of hook exit status though. The patch I posted will abort the whole (batch) operation if any hook exits non-zero. Should per-package hooks have the ability to do the same? Should the batch hooks have this ability? My thinking is yes to both, but I'm interested in alternate views on this. > package "update hook" (useful for nginx, which allows uninterrupted > update) and a "package purge hook" which brings the system to a > pristine state before package installation. Ok, so I'm taking this as: If a package is being upgraded (pkgrm followed immediately by pkgadd) that a preupdate and postupdate hook should be called? If a package is being removed (without a new version being added) that there should be a hook called after pkgrm named purge? This would allow for batch hooks as well as per-package hooks. In this case then, should there be a pre/post upgrade batch hook too? This is something I wasn't sure of...since the underlying tools don't have any notion of an upgrade, our distinction would be mostly artificial...not to say useless though. This is exactly the type of discussion I was hoping for. It's good to define the needs up front. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Thu Jun 25 16:36:05 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 10:36:05 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <20090625141711.GA25001@bolthole.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> Message-ID: <1245940343-sup-1963@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 25 10:17:11 -0400 2009: > its always nice to have someone willing to code. > However, this is an API we're talking about. It's usually best for an API > to be driven by a specification, not just "well look at the code". I agree. It was just a good way to grab eyeballs and start the discussion. I wasn't expecting Peter to grab it so quickly, which is why I prefaced it with 'completely untested...' > So.. how about writing up some docs for the API now? :-) I'm willing to do this, but I wanted to make sure it was a community discussion, since it will be a fundamental portion of the package tools. I'll start the wiki document and modify as the discussion goes. When we're happy, I'll update the patch. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 25 17:36:52 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 25 Jun 2009 08:36:52 -0700 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <20090625153652.GC13603@bolthole.com> On Thu, Jun 25, 2009 at 10:39:45AM +0100, Gary Law wrote: > 2009/6/25 Dagobert Michelsen : > > Why? Usually it is no different packaging for Solaris 9 than for > > Solaris 8. Why drop Solaris 8 without the need? If there is *any* > > problem on Solaris 8 it is of course ok to just release for Solaris 9. > > The problem for me is testing. Rolling out something that compiles is > all very well and good, but if it's not tested on Sol8 then it might > be buggy, and might in fact be worse than the old Sol8 version we have > in the archive. Finding the hardware/virtualisation kit for testing > for Sol8 is hard enough now, and it's only going to get worse. for people who run 8, even having "untested packages" is better than "having no packages". From maciej at opencsw.org Thu Jun 25 17:43:41 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Thu, 25 Jun 2009 16:43:41 +0100 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <20090625153652.GC13603@bolthole.com> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> <20090625153652.GC13603@bolthole.com> Message-ID: On Thu, Jun 25, 2009 at 4:36 PM, Philip Brown wrote: > for people who run 8, even having "untested packages" is better than > "having no packages". Perhaps having separate levels of 'testedness' for each Solaris major version would bring more clarity? "This package has been tested on $(TESTED_VERSIONS)." Maciej From maciej at opencsw.org Thu Jun 25 17:58:08 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Thu, 25 Jun 2009 16:58:08 +0100 Subject: [csw-maintainers] gmake hangs on the buildfarm Message-ID: maciej at build10s [build10s]:~/src/cups/trunk > gmake -d package GNU Make 3.81 Copyright (C) 2006 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. This program built for sparc-sun-solaris2.8 Reading makefiles... Reading makefile `Makefile'... Reading makefile `gar/category.mk' (search path) (no ~ expansion)... Reading makefile `gar/categories/net/category.mk' (search path) (no ~ expansion)... Reading makefile `gar/gar.mk' (search path) (no ~ expansion)... Reading makefile `gar/gar.conf.mk' (search path) (no ~ expansion)... Reading makefile `/home/maciej/.garrc' (search path) (don't care) (no ~ expansion)... Reading makefile `gar/gar.lib.mk' (search path) (no ~ expansion)... ...and it hangs there. Is anyone else observing this? Maciej From ihsan at dogan.ch Thu Jun 25 18:08:33 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Thu, 25 Jun 2009 18:08:33 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> <4A437E0C.4070101@opencsw.org> Message-ID: <4A43A101.7000509@dogan.ch> Dagobert Michelsen wrote: > BTW, should I change the default compiler for Solaris 9 and 10 to Studio > 12? Yes, that would make sense. I don't there is any reason to keep the default compiler on Studio 11. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at dogan.ch Thu Jun 25 18:10:44 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Thu, 25 Jun 2009 18:10:44 +0200 Subject: [csw-maintainers] issue with mail sent by mantis In-Reply-To: References: Message-ID: <4A43A184.8040202@dogan.ch> Peter FELECAN wrote: > The mail delivered by mantis has a systematic issue, detected by Amavis > and others: > > X-Virus-Scanned: amavisd-new at opencsw.org > X-Amavis-Alert: BAD HEADER SECTION, Duplicate header field: "Message-ID" Looks like a Mantis issue. Sebastian? Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From pfelecan at opencsw.org Thu Jun 25 18:32:05 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 25 Jun 2009 18:32:05 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <5954B1F4-D33A-4B36-A5B4-3CBFA484F488@opencsw.org> (Dagobert Michelsen's message of "Thu\, 25 Jun 2009 15\:38\:13 +0200") References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> <5954B1F4-D33A-4B36-A5B4-3CBFA484F488@opencsw.org> Message-ID: Dagobert Michelsen writes: > PS: Are you in for the beer tonight with William? I wished to but William forgot to tell me... although I'm thirsty. -- Peter From bwalton at opencsw.org Thu Jun 25 18:54:46 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 12:54:46 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <20090625141711.GA25001@bolthole.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> Message-ID: <1245948672-sup-1293@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 25 10:17:11 -0400 2009: > So.. how about writing up some docs for the API now? :-) Done: http://wiki.opencsw.org/package-hooks This is rough and definitely needs comment, thought and most likely improvement. I've already made some changes from what I posted in the patch last night. Notably, I've incorporated the comments (as I interpreted them) from Dago and I also changed /etc/opt/csw/pkg-hooks to /var/opt/csw/libexec/pkg-hooks, as I felt that was a better location. Lets talk it through. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 25 19:54:07 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 25 Jun 2009 10:54:07 -0700 Subject: [csw-maintainers] package hooks In-Reply-To: <1245948672-sup-1293@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245948672-sup-1293@ntdws12.chass.utoronto.ca> Message-ID: <20090625175407.GH13603@bolthole.com> On Thu, Jun 25, 2009 at 12:54:46PM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Thu Jun 25 10:17:11 -0400 2009: > > So.. how about writing up some docs for the API now? :-) > > Done: http://wiki.opencsw.org/package-hooks it's a good start. I think it would benefit, first and foremost, by an explaination of why this exists. ie: what is it trying to achieve, that cannot be done via the existing mechanisms of preinstall/postinstall scripts. Additionally, an explicit handling of how they are delivered, and more explicit naming, would probably be a good idea. for example, you dont explicitly say that these "hooks" are always delivered as part of a CSW package. (are they?) Also, while you specify a directory of /var/opt/csw/libexec/pkg-hooks, you dont specify or propose any naming standard for the actual files that live in there. (ie: possible a PKG based prefix or postfix to the name?) Additionally.. i actually do NOT think that /var/opt is appropriate. /var/opt is for data that will be changed /etc/opt is for scripts and configs these are scripts, therefore, they belong under /etc/opt. Now, the scripts themselves, might use /var/opt/XXX for scratch space or state info for themselves. But it doesnt make sense to me, that the scripts themselves would be under /var/opt From bwalton at opencsw.org Thu Jun 25 21:25:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 15:25:47 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <20090625175407.GH13603@bolthole.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245948672-sup-1293@ntdws12.chass.utoronto.ca> <20090625175407.GH13603@bolthole.com> Message-ID: <1245957871-sup-1485@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 25 13:54:07 -0400 2009: > I think it would benefit, first and foremost, by an explaination of why > this exists. > ie: what is it trying to achieve, that cannot be done via the existing > mechanisms of preinstall/postinstall scripts. Updated. > Additionally, an explicit handling of how they are delivered, and more > explicit naming, would probably be a good idea. > for example, you dont explicitly say that these "hooks" are always > delivered as part of a CSW package. (are they?) Updated. > Also, while you specify a directory of /var/opt/csw/libexec/pkg-hooks, > you dont specify or propose any naming standard for the actual files that > live in there. (ie: possible a PKG based prefix or postfix to the > name?) Good point. Updated. > Additionally.. i actually do NOT think that /var/opt is appropriate. > > /var/opt is for data that will be changed > /etc/opt is for scripts and configs Ok. I was thinking they were a 'libexec-ish' type thing, but I think you're correct. I've reverted to the old path. Thanks for the feedback. More welcome, as always. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 25 23:13:19 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 25 Jun 2009 14:13:19 -0700 Subject: [csw-maintainers] package hooks In-Reply-To: <1245957871-sup-1485@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245948672-sup-1293@ntdws12.chass.utoronto.ca> <20090625175407.GH13603@bolthole.com> <1245957871-sup-1485@ntdws12.chass.utoronto.ca> Message-ID: <20090625211319.GD71096@bolthole.com> On Thu, Jun 25, 2009 at 03:25:47PM -0400, Ben Walton wrote: > Thanks for the feedback. More welcome, as always. ah, that helps a lot. http://wiki.opencsw.org/package-hooks Some more thoughts: "The package tools shall pass, at most, a single argument to each hook that is run. " I think putting arbitrary limits on that, is unneccessary, and possibly harmful. It's usually better to define a MINIMUM set of arguments that is passed, but allow room for future expansion. Hmm... was going to comment on something else, but will think about it a bit more first.... From skayser at opencsw.org Fri Jun 26 00:01:22 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 26 Jun 2009 00:01:22 +0200 (CEST) Subject: [csw-maintainers] issue with mail sent by mantis In-Reply-To: <4A43A184.8040202@dogan.ch> References: <4A43A184.8040202@dogan.ch> Message-ID: <60284.194.246.122.22.1245967282.squirrel@ssl.skayser.de> Ihsan Dogan wrote: > Peter FELECAN wrote: > >> The mail delivered by mantis has a systematic issue, detected by Amavis >> and others: >> >> X-Virus-Scanned: amavisd-new at opencsw.org >> X-Amavis-Alert: BAD HEADER SECTION, Duplicate header field: >> "Message-ID" > > Looks like a Mantis issue. Sebastian? I have just issued a bug notification to myself and can't see such a duplicate header. Might be that there is something on my side normalizing things. Questions: Are others seeing this issue also? Peter, is every bug notification flagged with this alert on your side (the "systematic" makes believe yes, just want to make sure)? Sebastian From skayser at opencsw.org Fri Jun 26 00:24:32 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 26 Jun 2009 00:24:32 +0200 (CEST) Subject: [csw-maintainers] issue with mail sent by mantis In-Reply-To: <60284.194.246.122.22.1245967282.squirrel@ssl.skayser.de> References: <4A43A184.8040202@dogan.ch> <60284.194.246.122.22.1245967282.squirrel@ssl.skayser.de> Message-ID: <50970.194.246.122.22.1245968672.squirrel@ssl.skayser.de> Sebastian Kayser wrote: > Ihsan Dogan wrote: >> Peter FELECAN wrote: >> >>> The mail delivered by mantis has a systematic issue, detected by Amavis >>> and others: >>> >>> X-Virus-Scanned: amavisd-new at opencsw.org >>> X-Amavis-Alert: BAD HEADER SECTION, Duplicate header field: >>> "Message-ID" >> >> Looks like a Mantis issue. Sebastian? > > I have just issued a bug notification to myself and can't see such a > duplicate header. Might be that there is something on my side normalizing > things. > > Questions: Are others seeing this issue also? Peter, is every bug > notification flagged with this alert on your side (the "systematic" makes > believe yes, just want to make sure)? Ok, now i can see it on my side also. Forget about the questions, i will try to take care of it. Sebastian From bonivart at opencsw.org Fri Jun 26 00:51:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Fri, 26 Jun 2009 00:51:14 +0200 Subject: [csw-maintainers] package hooks In-Reply-To: <1245940343-sup-1963@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245940343-sup-1963@ntdws12.chass.utoronto.ca> Message-ID: <625385e30906251551n7c67da54qca80d28ef42bb1bd@mail.gmail.com> On Thu, Jun 25, 2009 at 4:36 PM, Ben Walton wrote: > ?When we're happy, I'll update the patch. If you want to, I can add your Sourceforge account to the pkgutil project, that way you can check in your changes directly into the repository. http://sourceforge.net/project/memberlist.php?group_id=250920 -- /peter From bwalton at opencsw.org Fri Jun 26 01:58:01 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 19:58:01 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <20090625211319.GD71096@bolthole.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245948672-sup-1293@ntdws12.chass.utoronto.ca> <20090625175407.GH13603@bolthole.com> <1245957871-sup-1485@ntdws12.chass.utoronto.ca> <20090625211319.GD71096@bolthole.com> Message-ID: <1245974162-sup-2859@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 25 17:13:19 -0400 2009: > http://wiki.opencsw.org/package-hooks [Leaves URL included...] > "The package tools shall pass, at most, a single argument to each hook that > is run. " > I think putting arbitrary limits on that, is unneccessary, and possibly > harmful. It's usually better to define a MINIMUM set of arguments that is > passed, but allow room for future expansion. The point with that was to make things a little more rigid, but I've altered the language as you suggest. I also added some thoughts (in the same section) about stdout/stderr from hook scripts. > Hmm... was going to comment on something else, but will think about it a > bit more first.... Fire when ready. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Fri Jun 26 01:59:24 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 19:59:24 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <625385e30906251551n7c67da54qca80d28ef42bb1bd@mail.gmail.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245940343-sup-1963@ntdws12.chass.utoronto.ca> <625385e30906251551n7c67da54qca80d28ef42bb1bd@mail.gmail.com> Message-ID: <1245974288-sup-2717@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Thu Jun 25 18:51:14 -0400 2009: > If you want to, I can add your Sourceforge account to the pkgutil > project, that way you can check in your changes directly into the > repository. > > http://sourceforge.net/project/memberlist.php?group_id=250920 Sure, as long as you trust my perl. It's been several years now since I've written any volume of it. When we've hashed out the standard formally, I'll add support for it to pkgutil as the reference implementation. Sound good? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bonivart at opencsw.org Fri Jun 26 09:53:31 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Fri, 26 Jun 2009 09:53:31 +0200 Subject: [csw-maintainers] package hooks In-Reply-To: <1245974288-sup-2717@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245940343-sup-1963@ntdws12.chass.utoronto.ca> <625385e30906251551n7c67da54qca80d28ef42bb1bd@mail.gmail.com> <1245974288-sup-2717@ntdws12.chass.utoronto.ca> Message-ID: <625385e30906260053s5e6d90a4yc89596143ebfda5e@mail.gmail.com> On Fri, Jun 26, 2009 at 1:59 AM, Ben Walton wrote: > Sure, as long as you trust my perl. ?It's been several years now since > I've written any volume of it. ?When we've hashed out the standard > formally, I'll add support for it to pkgutil as the reference > implementation. ?Sound good? Sounds very good. :-) I have added you to the project now. -- /peter From glaw at opencsw.org Fri Jun 26 14:23:02 2009 From: glaw at opencsw.org (Gary Law) Date: Fri, 26 Jun 2009 13:23:02 +0100 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) Message-ID: 2009/6/25 Philip Brown : > > for people who run 8, even having "untested packages" is better than > "having no packages". I'd disagree if they're displacing tested, known good, older Sol 8 versions. I'd suggest we archive what's available for Sol8, publish a health warning to users about no security or other patches, and move on. Given we support the 'supported versions of Solaris', why no OpenSolaris support? This is now a supported Solaris version, according to Sun, and has been since Nov 2008 AFAIK. Gary -- Gary Law glaw at opencsw.org From dam at opencsw.org Fri Jun 26 14:46:33 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 26 Jun 2009 14:46:33 +0200 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) In-Reply-To: References: Message-ID: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> Hi Gary, Am 26.06.2009 um 14:23 schrieb Gary Law: > Given we support the 'supported versions of Solaris', why no > OpenSolaris support? This is now a supported Solaris version, > according to Sun, and has been since Nov 2008 AFAIK. Yes, we should support OpenSolaris as there is support from Sun available. I am currently planning to set up LDoms on the T5220 for OpenSolaris sparc and get some extra capacity on VMware for OpenSolaris x86, but this all will take some time. Best regards -- Dago From phil at bolthole.com Fri Jun 26 15:46:22 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 26 Jun 2009 06:46:22 -0700 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) In-Reply-To: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> References: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> Message-ID: <20090626134622.GA34937@bolthole.com> On Fri, Jun 26, 2009 at 02:46:33PM +0200, Dagobert Michelsen wrote: > Yes, we should support OpenSolaris as there is support from > Sun available. I am currently planning to set up LDoms on the > T5220 for OpenSolaris sparc and get some extra capacity on > VMware for OpenSolaris x86, but this all will take some time. Do we have anyone willing to step up and make an IPS repository out of our packages? From mwatters at opencsw.org Fri Jun 26 15:54:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 26 Jun 2009 08:54:22 -0500 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) In-Reply-To: <20090626134622.GA34937@bolthole.com> References: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> <20090626134622.GA34937@bolthole.com> Message-ID: <4A44D30E.4040701@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > Do we have anyone willing to step up and make an IPS repository out of our > packages? > I am willing to "help" setup the IPS repo, but I currently don't have the time to champion this. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpE0w4ACgkQLrhmsXMSLxdbkQCfWTNQ9ihOuJrUDZEJUG/yunmD NGAAoI1bwuin2xeh4IELgj7tewX+9+ED =uG2+ -----END PGP SIGNATURE----- From schwindt at dfki.uni-kl.de Fri Jun 26 18:50:54 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Fri, 26 Jun 2009 18:50:54 +0200 Subject: [csw-maintainers] Zope in the package repository Message-ID: <200906261650.n5QGotw2008790@dfki.uni-kl.de> As I have the pleasure to transfer an old zope instance I gave Zope 3.3.1 a shot pkg-get -a | grep zope zope 3.3.1 I vaguely remembered problems with python 2.5 and was proven right. This one goes "boom". Even Zope 3.4 fails with python 2.6, it even fails with /usr/bin/python. Maybe the package should be pulled off the catalog ? I am thinking of building python 2.4.6 together with zope 3.4. Would it be a good idea to package both in /opt/csw/zope3 ? Nicolai From bwalton at opencsw.org Fri Jun 26 19:00:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 26 Jun 2009 13:00:47 -0400 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <200906261650.n5QGotw2008790@dfki.uni-kl.de> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> Message-ID: <1246035332-sup-1101@ntdws12.chass.utoronto.ca> Excerpts from Nicolai Schwindt's message of Fri Jun 26 12:50:54 -0400 2009: > Even Zope 3.4 fails with python 2.6, it even fails with > /usr/bin/python. Yes, Zope is rather stinky in this regard. > Maybe the package should be pulled off the catalog ? > > I am thinking of building python 2.4.6 together with zope 3.4. > Would it be a good idea to package both in /opt/csw/zope3 ? There is a 'Unified Installer' for Plone that bundles the required (and tested combinations of) python, python egg libraries, zope and plone install a single installation point. Is there something similar for just Zope? If so, that might be the way to go in this case. It's similar to what you're proposing, but if the work and testing is already done, it should save you some time. We're using the Plone bundled option here and it works well for our needs... HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 26 19:25:18 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 26 Jun 2009 10:25:18 -0700 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <1246035332-sup-1101@ntdws12.chass.utoronto.ca> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> <1246035332-sup-1101@ntdws12.chass.utoronto.ca> Message-ID: <20090626172518.GB26196@bolthole.com> On Fri, Jun 26, 2009 at 01:00:47PM -0400, Ben Walton wrote: > > There is a 'Unified Installer' for Plone that bundles the required > (and tested combinations of) python, python egg libraries, zope and > plone install a single installation point. Is there something similar > for just Zope? If so, that might be the way to go in this case. It's > similar to what you're proposing, but if the work and testing is > already done, it should save you some time. a special "local python", just for zope?! thats... disgusting :-( What's wrong with python, that it cant support slightlyi older versions? and what's wrong with Zope, that it doesnt work with recent perl versions? From bwalton at opencsw.org Fri Jun 26 19:31:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 26 Jun 2009 13:31:47 -0400 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <20090626172518.GB26196@bolthole.com> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> <1246035332-sup-1101@ntdws12.chass.utoronto.ca> <20090626172518.GB26196@bolthole.com> Message-ID: <1246037257-sup-2162@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 26 13:25:18 -0400 2009: > a special "local python", just for zope?! > > thats... disgusting :-( I completely agree. > What's wrong with python, that it cant support slightlyi older > versions? > and what's wrong with Zope, that it doesnt work with recent perl versions? Zope 'monkey patched' into some python functionality that has changed across the versions of python. I don't know why it hasn't been resolved yet, given how long python 2.5 has been around for. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 26 19:37:10 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 26 Jun 2009 10:37:10 -0700 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <1246037257-sup-2162@ntdws12.chass.utoronto.ca> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> <1246035332-sup-1101@ntdws12.chass.utoronto.ca> <20090626172518.GB26196@bolthole.com> <1246037257-sup-2162@ntdws12.chass.utoronto.ca> Message-ID: <20090626173710.GD26196@bolthole.com> On Fri, Jun 26, 2009 at 01:31:47PM -0400, Ben Walton wrote: > Zope 'monkey patched' into some python functionality that has changed > across the versions of python. I don't know why it hasn't been > resolved yet, given how long python 2.5 has been around for. http://wiki.zope.org/zope3/Zope3UsingPython25 The project is currently completed and the final code uploaded can be obtained from the following link. http://google-summer-of-code-2007-zope.googlecode.com/files/Nikhil_N.tar.gz From phil at bolthole.com Fri Jun 26 19:42:30 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 26 Jun 2009 10:42:30 -0700 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <20090626173710.GD26196@bolthole.com> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> <1246035332-sup-1101@ntdws12.chass.utoronto.ca> <20090626172518.GB26196@bolthole.com> <1246037257-sup-2162@ntdws12.chass.utoronto.ca> <20090626173710.GD26196@bolthole.com> Message-ID: <20090626174230.GE26196@bolthole.com> On Fri, Jun 26, 2009 at 10:37:10AM -0700, Philip Brown wrote: > On Fri, Jun 26, 2009 at 01:31:47PM -0400, Ben Walton wrote: > > Zope 'monkey patched' into some python functionality that has changed > > across the versions of python. I don't know why it hasn't been > > resolved yet, given how long python 2.5 has been around for. > > > http://wiki.zope.org/zope3/Zope3UsingPython25 > > The project is currently completed and the final code uploaded can be > obtained from the following link. > http://google-summer-of-code-2007-zope.googlecode.com/files/Nikhil_N.tar.gz also... http://mail.zope.org/pipermail/zope-tests/2009-June/011951.html So there seems to have been some work done, even in the Zope main "trunk"code. just not released i guess? From bwalton at opencsw.org Fri Jun 26 19:42:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 26 Jun 2009 13:42:47 -0400 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <20090626173710.GD26196@bolthole.com> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> <1246035332-sup-1101@ntdws12.chass.utoronto.ca> <20090626172518.GB26196@bolthole.com> <1246037257-sup-2162@ntdws12.chass.utoronto.ca> <20090626173710.GD26196@bolthole.com> Message-ID: <1246038075-sup-4078@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 26 13:37:10 -0400 2009: > http://wiki.zope.org/zope3/Zope3UsingPython25 > > The project is currently completed and the final code uploaded can be > obtained from the following link. > http://google-summer-of-code-2007-zope.googlecode.com/files/Nikhil_N.tar.gz Ok. This has changed since I last checked. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From schwindt at dfki.uni-kl.de Fri Jun 26 20:04:01 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Fri, 26 Jun 2009 20:04:01 +0200 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: Your message of "Fri, 26 Jun 2009 10:42:30 PDT." <20090626174230.GE26196@bolthole.com> Message-ID: <200906261804.n5QI42qH009352@dfki.uni-kl.de> [...] > > http://wiki.zope.org/zope3/Zope3UsingPython25 I found this one too, an I am tryin to build right now .) > > The project is currently completed and the final code uploaded can be > > obtained from the following link. > > http://google-summer-of-code-2007-zope.googlecode.com/files/Nikhil_N.tar.gz > > also... > > http://mail.zope.org/pipermail/zope-tests/2009-June/011951.html This is the svn trunk - I'd rather wait on a stable. I personally dislike python somewhat and hate zope .) Had enough trouble with them already. What zope calls final would be alpha with me - so imagine what I'd think of ther svn trunk. At our site I will have to even use an zope 2.11 because I found projects still using old Zope Products. But I will also look into the possibilities of upgrade the existing zope in csw to some reasonable version Nicolai From bwalton at opencsw.org Fri Jun 26 22:12:57 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 26 Jun 2009 16:12:57 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <625385e30906260053s5e6d90a4yc89596143ebfda5e@mail.gmail.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245940343-sup-1963@ntdws12.chass.utoronto.ca> <625385e30906251551n7c67da54qca80d28ef42bb1bd@mail.gmail.com> <1245974288-sup-2717@ntdws12.chass.utoronto.ca> <625385e30906260053s5e6d90a4yc89596143ebfda5e@mail.gmail.com> Message-ID: <1246047141-sup-6274@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Fri Jun 26 03:53:31 -0400 2009: > I have added you to the project now. Great. This will also give me a chance to try out the git-svn bridge, since there are no svn:externals involved! :) Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From rupert at opencsw.org Sat Jun 27 13:40:50 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 27 Jun 2009 13:40:50 +0200 Subject: [csw-maintainers] libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive (there is only .so ...) In-Reply-To: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> References: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> Message-ID: <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> i tried to compile serf, and get: "test/test_context.c", line 635: warning: statement not reached "test/test_context.c", line 809: warning: statement not reached "buckets/ssl_buckets.c", line 1267: warning: argument #2 is incompatible with prototype: ? ? ? ?prototype: pointer to pointer to unsigned char : "/opt/csw/include/openssl/x509.h", line 939 ? ? ? ?argument : pointer to pointer to char /opt/csw/apache2/share/build/libtool --silent --mode=link /opt/studio/SOS11/SUNWspro/bin/cc -xarch=v8 -L/opt/csw/lib -L/opt/csw/bdb44/lib -R/opt/csw/apache2/lib -L/opt/csw/apache2/lib -R/opt/csw/apache2/lib -L/opt/csw/apache2/lib -rpath /opt/csw/lib -o libserf-0.la buckets/aggregate_buckets.lo buckets/request_buckets.lo context.lo buckets/buckets.lo buckets/simple_buckets.lo buckets/file_buckets.lo buckets/mmap_buckets.lo buckets/socket_buckets.lo buckets/response_buckets.lo buckets/headers_buckets.lo buckets/allocator.lo buckets/dechunk_buckets.lo buckets/deflate_buckets.lo buckets/limit_buckets.lo buckets/ssl_buckets.lo buckets/barrier_buckets.lo buckets/chunk_buckets.lo -laprutil-1 -lldap -llber -ldb-4.4 -lexpat -liconv ?-lapr-1 -luuid -lsendfile -lrt -lsocket ?-lpthread -ldl -lm ?-lz -lssl -lcrypto grep: can't open /opt/csw/bdb44/lib/libdb-4.4.la /opt/csw/bin/gsed: can't read /opt/csw/bdb44/lib/libdb-4.4.la: No such file or directory libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive gmake[2]: *** [libserf-0.la] Error 1 gmake[2]: Leaving directory `/home/rupert/mgar/pkg/libserf/trunk/work/build-isa-sparcv8/serf-0.3.0' gmake[1]: *** [build-work/build-isa-sparcv8/serf-0.3.0/Makefile] Error 2 gmake[1]: Leaving directory `/home/rupert/mgar/pkg/libserf/trunk' gmake: *** [merge-isa-sparcv8] Error 2 how could one avoid that? rupert. From rupert at opencsw.org Sat Jun 27 14:00:56 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 27 Jun 2009 14:00:56 +0200 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <4A180558.70601@opencsw.org> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> <6af4270905221809u232fdfe7l7f55df4705a0434@mail.gmail.com> <4A180558.70601@opencsw.org> Message-ID: <6af4270906270500nbc25e89rcc96612455fa6c50@mail.gmail.com> hi mike, do you have an example how you put such plugins best into a plugin pack, especially: * multiple downloads * installs then i could give it a try. rupert. On Sat, May 23, 2009 at 16:16, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi rupert, > rupert THURNER wrote: >> >> that is great to hear! i even would appreciate if the most important >> plugins make it into trac as well (not enabled by default), e.g. >> >> easy_install http://trac-hacks.org/svn/tracwysiwygplugin/0.11/ >> easy_install http://trac-hacks.org/svn/ticketdeleteplugin/0.11/ >> easy_install http://trac-hacks.org/svn/ticketchangeplugin/0.11/ >> easy_install http://trac-hacks.org/svn/customfieldadminplugin/0.11/ >> easy_install http://trac-hacks.org/svn/tagsplugin/tags/0.6/ >> easy_install http://svn.ipd.uka.de/repos/javaparty/JP/trac/plugins/tracnav-0.11/ >> easy_install http://trac-hacks.org/svn/sectioneditplugin/0.11/ >> easy_install http://trac-hacks.org/svn/extracturlplugin/0.11/ >> easy_install http://trac-hacks.org/svn/serversideredirectplugin/0.11/ >> easy_install http://trac-hacks.org/svn/masterticketsplugin/0.11/ >> easy_install http://trac-hacks.org/svn/autoupgradeplugin/0.11/ >> > > I am working on updating Trac in the repository. ( after I get python > re-released as separate packages.) ?I will create a second package for trac, > trac_plugin_pack that contains the eggs you listed above. > > also, > your last post on the google group forum link in the ticket. > the '-G' is gone because you export LDSHARED=/opt/SUNWspro/bin/cc > ''' > compiler.linker_so > ['/opt/studio/SOS11/SUNWspro/bin/cc', '-G'] > ''' > means that then setting for compiler.linker_so is > /opt/studio/SOS11/SUNWspro/bin/cc -G > > os.environ settings override what python finds in its config. > export LDSHARED='/opt/SUNWspro/bin/cc -G' > would have saved the -G > > you can easily play with this using the following python script > =============================================================== > #! /usr/bin/env python > > from distutils.ccompiler import new_compiler > from distutils.sysconfig import customize_compiler > > compiler = new_compiler(None,None,None,None) > customize_compiler(compiler) > > print """ > *** Compiler Settings for Python *** > > ? ?Type: %s > ?Linker: %s > Compiler: %s > """ % (compiler.compiler_type, compiler.linker_so, compiler.compiler) > ====================================================================== > > $ ./pycompile.py > > *** Compiler Settings for Python *** > > ? ?Type: unix > ?Linker: ['/opt/studio/SOS11/SUNWspro/bin/cc', '-G'] > Compiler: ['/opt/studio/SOS11/SUNWspro/bin/cc', '-DNDEBUG', '-O'] > > you can set various environment variables and see how that affects those settings. > > - -- > Thanks, > Mike > > "Any intelligent fool can make things bigger, more complex, > and more violent. ?It takes a touch of genius -- and a lot of courage -- > to move in the opposite direction." > > * Albert Einstein 1879 - 1955 > ? ?US German-born Theoretical Physicist > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.9 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iEYEARECAAYFAkoYBVgACgkQLrhmsXMSLxdLswCg4qlEUAdZIpxfr6B/0UpQhQHj > UW4AoLk0dKQ3J5rsmkEwylnShAaWUtrh > =GqGq > -----END PGP SIGNATURE----- > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers > From bwalton at opencsw.org Sat Jun 27 15:13:56 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sat, 27 Jun 2009 09:13:56 -0400 Subject: [csw-maintainers] libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive (there is only .so ...) In-Reply-To: <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> References: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> Message-ID: <1246108151-sup-8443@ntdws12.chass.utoronto.ca> Excerpts from rupert THURNER's message of Sat Jun 27 07:40:50 -0400 2009: > buckets/barrier_buckets.lo buckets/chunk_buckets.lo -laprutil-1 -lldap > -llber -ldb-4.4 -lexpat -liconv ?-lapr-1 -luuid -lsendfile -lrt > -lsocket ?-lpthread -ldl -lm ?-lz -lssl -lcrypto > grep: can't open /opt/csw/bdb44/lib/libdb-4.4.la > /opt/csw/bin/gsed: can't read /opt/csw/bdb44/lib/libdb-4.4.la: No such > file or directory You can try the libtool fixup tools that Mike added a few months back. You'll want to set STRIP_LIBTOOL to 1 (or any value, really) in your GAR Makefile. I hit a situation where this didn't help, but that was for a package that was doing things 'their own way' instead of the more standard way you'll typically encounter. If this variable (and the gar/bin/fixlibtool script) doesn't help in your case, let me know and I'll show you how I fixed it for my oddball. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Sat Jun 27 17:11:35 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 27 Jun 2009 10:11:35 -0500 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270906270500nbc25e89rcc96612455fa6c50@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> <6af4270905221809u232fdfe7l7f55df4705a0434@mail.gmail.com> <4A180558.70601@opencsw.org> <6af4270906270500nbc25e89rcc96612455fa6c50@mail.gmail.com> Message-ID: <4A4636A7.8040602@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > hi mike, > > do you have an example how you put such plugins best into a plugin > pack, especially: > * multiple downloads > * installs > > then i could give it a try. > > rupert. > I don't have an example unfortunately. basically you would have to write a custom Makefile. downloading all the software you want to bundle, building and installing each one in $(DESTDIR) the merge and package should be relatively normal once everything is installed. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkpGNqYACgkQLrhmsXMSLxfuawCeLYqj5aKBIFY+/JlJoduqqTf5 RzgAmwUtancgZXxy0DJepU68IULol0ue =g96R -----END PGP SIGNATURE----- From dam at opencsw.org Sat Jun 27 21:27:47 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sat, 27 Jun 2009 21:27:47 +0200 Subject: [csw-maintainers] /testing: nasm 2.05.01 In-Reply-To: <21894.1244997256@gerd> References: <21894.1244997256@gerd> Message-ID: Hi Murray, Am 14.06.2009 um 18:34 schrieb : > The following files have been put into /testing ... > > 9aea1d1df09e7f51525d8a9fee95b0ee nasm-2.05.01,REV=2009.06.14- > SunOS5.8-i386-CSW.pkg.gz > 0fa0d0eebd39d15f6c7ae7336764b4fa nasm-2.05.01,REV=2009.06.14- > SunOS5.8-sparc-CSW.pkg.gz > > I haven't even run them yet - let me know if they work. Cheers! Looks good. Please update to 2.06 and relase it :-) Best regards -- Dago From rupert at opencsw.org Sun Jun 28 21:14:26 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 28 Jun 2009 21:14:26 +0200 Subject: [csw-maintainers] libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive (there is only .so ...) In-Reply-To: <1246108151-sup-8443@ntdws12.chass.utoronto.ca> References: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> <1246108151-sup-8443@ntdws12.chass.utoronto.ca> Message-ID: <6af4270906281214rd80c747hc04f69b442ea41a2@mail.gmail.com> On Sat, Jun 27, 2009 at 15:13, Ben Walton wrote: > Excerpts from rupert THURNER's message of Sat Jun 27 07:40:50 -0400 2009: >> buckets/barrier_buckets.lo buckets/chunk_buckets.lo -laprutil-1 -lldap >> -llber -ldb-4.4 -lexpat -liconv ?-lapr-1 -luuid -lsendfile -lrt >> -lsocket ?-lpthread -ldl -lm ?-lz -lssl -lcrypto >> grep: can't open /opt/csw/bdb44/lib/libdb-4.4.la >> /opt/csw/bin/gsed: can't read /opt/csw/bdb44/lib/libdb-4.4.la: No such >> file or directory > > You can try the libtool fixup tools that Mike added a few months > back. ?You'll want to set STRIP_LIBTOOL to 1 (or any value, really) in > your GAR Makefile. ?I hit a situation where this didn't help, but that > was for a package that was doing things 'their own way' instead of the > more standard way you'll typically encounter. > > If this variable (and the gar/bin/fixlibtool script) doesn't help in > your case, let me know and I'll show you how I fixed it for my > oddball. i did set it, but got the error anyway. i also found http://trac.macports.org/ticket/19751 which says: serf links to lib_apr, which depends on bdb. apr-1.3.7 fixes this, and dynamically links to bdb. but this would mean we'd need libapr, libaprutil as separate package i guess? rupert. From william at wbonnet.net Sun Jun 28 21:55:52 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 28 Jun 2009 21:55:52 +0200 Subject: [csw-maintainers] another bug in upstream. In-Reply-To: <4A308F3E.9020306@opencsw.org> References: <4A308F3E.9020306@opencsw.org> Message-ID: <4A47CAC8.5080603@wbonnet.net> Hi Mike > I received an upgrade notice for GD ( pasted below ) > Thanks for the bug report. I'll try to fix it in the next days. I am currently changing several things to uwatch to handle this kind of problem, and a adding a few missing features. > I am the GD maintainer, however all the links to the download location are for > the pm_gd to which I am not the maintainer. ( I believe it was Alex who is > retired.) > > Further investigation showed me that I do in fact have the latest "stable" > release of GD packaged, but the perl module is what is in need of upgrade. > GD-2.43.tar.gz is the filename of pm_gd > gd-2.0.35.tar.gz is the filename of gd > > note the regex for upstream watch should be the same. > > I have no solution off hand to this bug, but I will take a look when I get a > few spare cycles. the problem is not the regexp used by uwatch, but the same Phil already reported before. Both package have the same GARNAME. The perl module should be renamed to have a different GARNAME (actually pm_gd garname is GD, and gd garname is gd). Since pm_gd is orphaned, i'll fix this package in the next days. cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From william at wbonnet.net Sun Jun 28 22:16:35 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 28 Jun 2009 22:16:35 +0200 Subject: [csw-maintainers] Adding a new category to mantis ? Message-ID: <4A47CFA3.6040608@wbonnet.net> Hi I would to propose a small evolution to mantis. To add a category for GAR build description issues. Actually the following categories are available : Other Packaging Upgrade Regular Use IMHO "Packaging" category is different of a possible "Build description" category. I do the following distinction between these cat?gories. Packaging can keep track of problem with "package content", and build description of problem about prackage creation process. For exemple, a missing config file or an invalid permission on a file is a packaging issue. A missing uwatch regexp or a duplicated garname is a build description issue. "Build description" for category name is just a proposal. Other names are welcomed :) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From rupert at opencsw.org Sun Jun 28 23:06:28 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 28 Jun 2009 23:06:28 +0200 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) In-Reply-To: <20090626134622.GA34937@bolthole.com> References: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> <20090626134622.GA34937@bolthole.com> Message-ID: <6af4270906281406o60495920lbaa4c8c5789bdb75@mail.gmail.com> On Fri, Jun 26, 2009 at 15:46, Philip Brown wrote: > On Fri, Jun 26, 2009 at 02:46:33PM +0200, Dagobert Michelsen wrote: >> Yes, we should support OpenSolaris as there is support from >> Sun available. I am currently planning to set up LDoms on the >> T5220 for OpenSolaris sparc and get some extra capacity on >> VMware for OpenSolaris x86, but this all will take some time. > > Do we have anyone willing to step up and make an IPS repository out of our > packages? you mean: 1.setting up a repository: http://blogs.sun.com/halton/entry/setup_a_ips_repository 2. http://blogs.sun.com/migi/resource/creating_ips_repo_part_5.html use solaris.py ? (out of http://blogs.sun.com/migi/entry/create_your_own_opensolaris_ips2) or just do something like: for cswpkg in `ls -1 /home/wherever-opencsw-packages-are` do pkgsend -s http://pkg.opencsw.org:10000 send $cswpkg done From phil at bolthole.com Sun Jun 28 23:51:26 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 28 Jun 2009 14:51:26 -0700 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) In-Reply-To: <6af4270906281406o60495920lbaa4c8c5789bdb75@mail.gmail.com> References: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> <20090626134622.GA34937@bolthole.com> <6af4270906281406o60495920lbaa4c8c5789bdb75@mail.gmail.com> Message-ID: <20090628215126.GB83796@bolthole.com> On Sun, Jun 28, 2009 at 11:06:28PM +0200, rupert THURNER wrote: > On Fri, Jun 26, 2009 at 15:46, Philip Brown wrote: > > Do we have anyone willing to step up and make an IPS repository out of our > > packages? > > you mean: > > 1.setting up a repository: > http://blogs.sun.com/halton/entry/setup_a_ips_repository > 2. http://blogs.sun.com/migi/resource/creating_ips_repo_part_5.html > use solaris.py ? > (out of http://blogs.sun.com/migi/entry/create_your_own_opensolaris_ips2) > > or just do something like: > > for cswpkg in `ls -1 /home/wherever-opencsw-packages-are` > do > pkgsend -s http://pkg.opencsw.org:10000 send $cswpkg > done I meant primarily taking on the responsability of converting packages, on an ongoing basis. This is not a trivial task: Quite a few of our packages will need massaging, for them to work under IPS. Additionally, this person would then have de-facto responsability, of giving feedback to the "mainline group" (ie: the maintainers at large :) on things that do not translate well to IPS mechanisms If anyone is interested, please email me privately, and i will discuss more details there. From dam at opencsw.org Mon Jun 29 14:09:00 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 29 Jun 2009 14:09:00 +0200 Subject: [csw-maintainers] Firmware update on BO Buildfarm 2009/06/30 Message-ID: Hi, the T5220 at the BO farm carrying login, build*s, web and mirror will be down tomorrow, on 2009/06/30 from 10:00 CEST to 12:00 CEST (CEST = UTC+2). I am going to update the firmware of the T5220 to prepare it for Logical Domains as prerequisite for an OpenSolaris build host. There will be an additional announcement when the update is finished. You can still use build8s.go.opencsw.org during the downtime. The SVN repository is not affected as it is hosted at SourceForge. Best regards -- Dago From bonivart at opencsw.org Mon Jun 29 15:12:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 29 Jun 2009 15:12:14 +0200 Subject: [csw-maintainers] Duplicate CSWlibgnet2rt in catalog Message-ID: <625385e30906290612l3c07f602k7b9c9856ba67d20e@mail.gmail.com> It seems that CSWlibgnet2rt has changed its catalogname from libgnet2rt to libgnet2_rt but the old entry is still there: libgnet2_rt 2.0.8,REV=2009.05.27 CSWlibgnet2rt libgnet2_rt-2.0.8,REV=2009.05.27-SunOS5.8-sparc-CSW.pkg.gz 61d14c836f09148fd79c1ae86b53e434 78126 CSWcommon|CSWgcc3corert|CSWggettextrt|CSWglib2|CSWiconv none libgnet2rt 2.0.7 CSWlibgnet2rt libgnet2rt-2.0.7-SunOS5.8-sparc-CSW.pkg.gz ee590915e6ee972869c32ee9a372dc47 99258 CSWcommon|CSWggettext|CSWglib2|CSWiconv none -- /peter From rmacduff at opencsw.org Mon Jun 29 17:16:58 2009 From: rmacduff at opencsw.org (Ross Macduff) Date: Mon, 29 Jun 2009 11:16:58 -0400 Subject: [csw-maintainers] gsed 4.2.1 in testing Message-ID: <1246288551-sup-4211@frog.chass.utoronto.ca> Hello, gsed 4.2.1 is now in testing. Best, Ross From mwatters at opencsw.org Mon Jun 29 17:23:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 29 Jun 2009 10:23:51 -0500 Subject: [csw-maintainers] pycairo 1.8.0 now in testing Message-ID: <4A48DC87.7070006@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 you can get it here: pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u pycairo Feedback always welcome. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpI3IYACgkQLrhmsXMSLxdP7QCfbeVU7llZsXacKp7Ejvs3ym1y JDsAoKKwteXvihKNE3MzBXiUe3aJzvjx =iQq+ -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 29 17:48:44 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 29 Jun 2009 10:48:44 -0500 Subject: [csw-maintainers] anyone updating libglade2? Message-ID: <4A48E25C.20309@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 if not, I will do it, once pycairo (in testing) is released it will be the last dependency I need to update pygtk. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpI4lwACgkQLrhmsXMSLxdHugCfT5/WM93dw6YWQbTz9tOmGmk8 cqwAmwf95q59EqQfqs8AYJRy4sGCLzcf =6PhQ -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 29 18:10:30 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 29 Jun 2009 11:10:30 -0500 Subject: [csw-maintainers] anyone updating libglade2? In-Reply-To: <4A48E25C.20309@opencsw.org> References: <4A48E25C.20309@opencsw.org> Message-ID: <4A48E776.4040200@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > if not, I will do it, once pycairo (in testing) is released it will be the last > dependency I need to update pygtk. > well, last dependency besides gtk ;-) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpI53YACgkQLrhmsXMSLxdVxwCbByC5pS7czeK/adnplppPjhma ttkAoMU7Q8ZaUcY3DGQ0vVTTbzzMI4ka =lNYv -----END PGP SIGNATURE----- From benny at opencsw.org Mon Jun 29 18:48:59 2009 From: benny at opencsw.org (Benjamin von Mossner) Date: Mon, 29 Jun 2009 18:48:59 +0200 Subject: [csw-maintainers] pen: initial commit Message-ID: <20090629164859.GA73690@vonmossner.de> Hi maintainers, i just commited my first package (pen) :-) a quite tiny one, but nevertheless. If anyone feels about checking my Makefile and/or the package itself, please see svn and/or /home/testing. Any feedback is welcome. Cheers, benny -- /"\ ASCII RIBBON CAMPAIGN | Benjamin von Mossner \ / AGAINST HTML MAIL | benny at vonmossner.de X / \ multiple exclamation marks are a sure sign of a diseased mind From benny at vonmossner.de Mon Jun 29 18:48:59 2009 From: benny at vonmossner.de (Benjamin von Mossner) Date: Mon, 29 Jun 2009 18:48:59 +0200 Subject: [csw-maintainers] pen: initial commit Message-ID: <20090629164859.GA73690@vonmossner.de> Hi maintainers, i just commited my first package (pen) :-) a quite tiny one, but nevertheless. If anyone feels about checking my Makefile and/or the package itself, please see svn and/or /home/testing. Any feedback is welcome. Cheers, benny -- /"\ ASCII RIBBON CAMPAIGN | Benjamin von Mossner \ / AGAINST HTML MAIL | benny at vonmossner.de X / \ multiple exclamation marks are a sure sign of a diseased mind From phil at bolthole.com Mon Jun 29 19:11:25 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 29 Jun 2009 10:11:25 -0700 Subject: [csw-maintainers] Duplicate CSWlibgnet2rt in catalog In-Reply-To: <625385e30906290612l3c07f602k7b9c9856ba67d20e@mail.gmail.com> References: <625385e30906290612l3c07f602k7b9c9856ba67d20e@mail.gmail.com> Message-ID: <20090629171125.GC98636@bolthole.com> On Mon, Jun 29, 2009 at 03:12:14PM +0200, Peter Bonivart wrote: > It seems that CSWlibgnet2rt has changed its catalogname from > libgnet2rt to libgnet2_rt but the old entry is still there: > Urk. thanks for catching that. From bwalton at opencsw.org Mon Jun 29 19:19:42 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 29 Jun 2009 13:19:42 -0400 Subject: [csw-maintainers] libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive (there is only .so ...) In-Reply-To: <6af4270906281214rd80c747hc04f69b442ea41a2@mail.gmail.com> References: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> <1246108151-sup-8443@ntdws12.chass.utoronto.ca> <6af4270906281214rd80c747hc04f69b442ea41a2@mail.gmail.com> Message-ID: <1246295816-sup-8970@ntdws12.chass.utoronto.ca> Excerpts from rupert THURNER's message of Sun Jun 28 15:14:26 -0400 2009: > but this would mean we'd need libapr, libaprutil as separate package > i guess? I thought that someone was working on this already...? As for your issues, see the Makefile for openjade. Specifically, the post-configure-modulated step. That's how I 'broke' libtool there...not sure if it's helpful for your case or not, since your situation is different. HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From mwatters at opencsw.org Mon Jun 29 20:03:02 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 29 Jun 2009 13:03:02 -0500 Subject: [csw-maintainers] libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive (there is only .so ...) In-Reply-To: <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> References: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> Message-ID: <4A4901D6.50402@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > > how could one avoid that? > the main issue is the Makefile generated by configre is using "libtool" from /opt/csw/apache2/share/build/libtool which does "NOT" get modified by my strip. Here is a patch to the Makefile currently in svn that will allow you to build and test successfully libserf if you cut and paste, please take care to watch for wrapping I have Thunderbird configured to wrap at 100 characters. $ svn diff Makefile Index: Makefile =================================================================== - --- Makefile (revision 5414) +++ Makefile (working copy) @@ -23,13 +23,19 @@ # UPSTREAM_MASTER_SITES = DEPENDS = server/apache2rt - -STRIP_LIBTOOL = 1 CONFIGURE_ARGS = $(DIRPATHS) - -CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 - -CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2 - -#CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config - -#CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config +CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config +CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config TEST_TARGET = check include gar/category.mk + +EXTRA_LDFLAGS = -R/opt/csw/bdb44/lib +APR_LIBTOOL = $(shell /opt/csw/apache2/bin/apr-config --apr-libtool) +post-configure-modulated: + gcp $(APR_LIBTOOL) $(WORKSRC) + perl -i -pne 's|$(APR_LIBTOOL)|$(abspath $(WORKSRC)/libtool)|' \ + $(WORKSRC)/Makefile + $(GARBIN)/fixlibtool $(WORKSRC) + @$(MAKECOOKIE) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpJAdUACgkQLrhmsXMSLxeBXQCgmTxH6hTrxyY2TrT6eSMj+S4t c+4Anj0nU/L6fyoBYYX4VqFjMCGd4iRo =qeGL -----END PGP SIGNATURE----- From skayser at opencsw.org Mon Jun 29 23:20:19 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 29 Jun 2009 23:20:19 +0200 (CEST) Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") In-Reply-To: <4A2403D0.4090109@opencsw.org> References: <4A2403D0.4090109@opencsw.org> Message-ID: <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> Sebastian Kayser wrote: > a couple of RFEs for the "GAR package status" [1] page. Can we get: > > * an "last updated" field at the top of the page > * the numbers / stats right at the top of the page (instead of further > down) > * a contact address for the maintainer of the page on the page (for > queries like this one and to get packages that are in GAR but not > listed as such fixed) > > Ah, and some sort of monthly historical data would be cool also. So that > we can see some progress. > > Sebastian > > [1] http://www.opencsw.org/package-gar-status.html Anyone taking care of the GAR status page? Sebastian From william at wbonnet.net Mon Jun 29 23:41:26 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 29 Jun 2009 23:41:26 +0200 Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") In-Reply-To: <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> References: <4A2403D0.4090109@opencsw.org> <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> Message-ID: <4A493506.9050204@wbonnet.net> Hi Sebastian > Sebastian Kayser wrote: > >> a couple of RFEs for the "GAR package status" [1] page. Can we get: >> >> * an "last updated" field at the top of the page >> * the numbers / stats right at the top of the page (instead of further >> down) >> * a contact address for the maintainer of the page on the page (for >> queries like this one and to get packages that are in GAR but not >> listed as such fixed) >> >> Ah, and some sort of monthly historical data would be cool also. So that >> we can see some progress. >> >> Sebastian >> >> [1] http://www.opencsw.org/package-gar-status.html >> > > Anyone taking care of the GAR status page? > I'm currently working on this page and uwatch. Rewriting it in Perl and improving things. I will modify the page by the end of the week, and uwatch next week. uwatch have several bugs to fix (problem with garname used to retrieve package instead of catalog name). I would like to merge some information from uwatch to this page. cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From dam at opencsw.org Mon Jun 29 23:44:39 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 29 Jun 2009 23:44:39 +0200 Subject: [csw-maintainers] New in testing: gtk2 2.16.1 Message-ID: Hi, there is now a current gtk2 in testing: gtk2-2.16.1,REV=2009.06.29-SunOS5.8-i386-CSW.pkg.gz gtk2-2.16.1,REV=2009.06.29-SunOS5.8-sparc-CSW.pkg.gz gtk2_devel-2.16.1,REV=2009.06.29-SunOS5.8-i386-CSW.pkg.gz gtk2_devel-2.16.1,REV=2009.06.29-SunOS5.8-sparc-CSW.pkg.gz gtk2_doc-2.16.1,REV=2009.06.29-SunOS5.8-i386-CSW.pkg.gz gtk2_doc-2.16.1,REV=2009.06.29-SunOS5.8-sparc-CSW.pkg.gz Feedback as always welcome! (I know, doc should be archall) Best regards -- Dago From william at wbonnet.net Mon Jun 29 23:47:01 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 29 Jun 2009 23:47:01 +0200 Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") In-Reply-To: <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> References: <4A2403D0.4090109@opencsw.org> <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> Message-ID: <4A493655.1090206@wbonnet.net> Hi opps i hit the send key too fast :) Is there any other RFE to come ? I have added a wiki page for this RFE http://wiki.opencsw.org/gar-package-status-rfe cheers W. From skayser at opencsw.org Tue Jun 30 00:31:59 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 30 Jun 2009 00:31:59 +0200 (CEST) Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") In-Reply-To: <4A493655.1090206@wbonnet.net> References: <4A2403D0.4090109@opencsw.org> <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> <4A493655.1090206@wbonnet.net> Message-ID: <51059.217.227.50.24.1246314719.squirrel@ssl.skayser.de> William Bonnet wrote: > opps i hit the send key too fast :) > > Is there any other RFE to come ? > > I have added a wiki page for this RFE > http://wiki.opencsw.org/gar-package-status-rfe Thanks for the status update and thanks for taking care of it. No additional RFEs from my side - at least for now. ;) Sebastian From william at wbonnet.net Tue Jun 30 00:49:54 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 30 Jun 2009 00:49:54 +0200 Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") In-Reply-To: <51059.217.227.50.24.1246314719.squirrel@ssl.skayser.de> References: <4A2403D0.4090109@opencsw.org> <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> <4A493655.1090206@wbonnet.net> <51059.217.227.50.24.1246314719.squirrel@ssl.skayser.de> Message-ID: <4A494512.9040104@wbonnet.net> Hi Sebastian > Thanks for the status update and thanks for taking care of it. No > additional RFEs from my side - at least for now. ;) > Page content is (supposed to be) updated everyday. I will add modification date tomorrow. The rest of the evolutions will come with the perl version. I'm working on this since yesterday, i am also pluging the database backend. Actaully this page is generated by an ugly shell script (generating html...). cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From car at opencsw.org Tue Jun 30 11:08:52 2009 From: car at opencsw.org (Chris Reece) Date: Tue, 30 Jun 2009 21:08:52 +1200 Subject: [csw-maintainers] tcsh 6.16.00 in testing Message-ID: As per subject, current version of tcsh in: /home/testing/tcsh-6.16.00,REV=2009.06.30-SunOS5.8-i386-CSW.pkg.gz /home/testing/tcsh-6.16.00,REV=2009.06.30-SunOS5.8-sparc-CSW.pkg.gz Works on my only test host, would appreciate any additional testing. Cheers, Chris. From bonivart at opencsw.org Tue Jun 30 13:36:47 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 30 Jun 2009 13:36:47 +0200 Subject: [csw-maintainers] tcsh 6.16.00 in testing In-Reply-To: References: Message-ID: <625385e30906300436v1eed482eq4c9f5354ed6415e7@mail.gmail.com> On Tue, Jun 30, 2009 at 11:08 AM, Chris Reece wrote: > As per subject, current version of tcsh in: > > /home/testing/tcsh-6.16.00,REV=2009.06.30-SunOS5.8-i386-CSW.pkg.gz > /home/testing/tcsh-6.16.00,REV=2009.06.30-SunOS5.8-sparc-CSW.pkg.gz > > Works on my only test host, would appreciate any additional testing. It was years ago I used tcsh but it installs/removes cleanly and seems to work properly during my short test on a Solaris 9 Sparc and a Solaris 10 x86. Man page also works. -- /peter From dam at opencsw.org Tue Jun 30 13:45:37 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 30 Jun 2009 13:45:37 +0200 Subject: [csw-maintainers] Firmware update on BO Buildfarm 2009/06/30 In-Reply-To: References: Message-ID: Hi, Am 29.06.2009 um 14:09 schrieb Dagobert Michelsen: > the T5220 at the BO farm carrying login, build*s, web and mirror > will be down tomorrow, on 2009/06/30 from 10:00 CEST to 12:00 CEST > (CEST = UTC+2). I am going to update the firmware of the T5220 > to prepare it for Logical Domains as prerequisite for an > OpenSolaris build host. There will be an additional announcement > when the update is finished. Unfortunately I had an unplanned customer visit and the update wasn't done. I will schedule it for a later time. Best regards -- Dago From ja at opencsw.org Tue Jun 30 15:10:28 2009 From: ja at opencsw.org (Juergen Arndt) Date: Tue, 30 Jun 2009 15:10:28 +0200 Subject: [csw-maintainers] Fwd: [csw-pkgrequests] package request In-Reply-To: References: <20090630115854.1C68B8754@mail.opencsw.org> Message-ID: Hi, I'll try to build packages for the munin server and node. Could take some days, please be patient :) Juergen >> Von: mcb at uib.no >> Datum: 30. Juni 2009 13:58:54 MESZ >> An: pkgrequests at lists.opencsw.org >> Betreff: [csw-pkgrequests] package request >> >> Munin: http://munin.projects.linpro.no/ >> >> I would appreciate packages for both munin and munin-node. >> >> What is Munin? >> Munin the monitoring tool surveys all your computers and remembers what >> it saw. It presents all the information in graphs through a web >> interface. Its emphasis is on plug and play capabilities. After >> completing a installation a high number of monitoring plugins will be >> playing with no more effort. >> >> Using Munin you can easily monitor the performance of your computers, >> networks, SANs, applications, weather measurements and whatever comes >> to mind. It makes it easy to determine "what's different today" when a >> performance problem crops up. It makes it easy to see how you're doing >> capacity-wise on any resources. >> >> Munin uses the excellent RRDTool (written by Tobi Oetiker) and the >> framework is written in Perl, while plugins may be written in any >> language. Munin has a master/node architecture in which the master >> connects to all the nodes at regular intervals and asks them for data. >> It then stores the data in RRD files, and (if needed) updates the >> graphs. One of the main goals has been ease of creating new plugins >> (graphs). >> _______________________________________________ >> pkgrequests mailing list >> pkgrequests at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/pkgrequests > -- Juergen Arndt From mwatters at opencsw.org Tue Jun 30 18:19:16 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 30 Jun 2009 11:19:16 -0500 Subject: [csw-maintainers] New in testing: gtk2 2.16.1 In-Reply-To: References: Message-ID: <4A4A3B04.2040701@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Feedback as always welcome! I have successfully compiled and tested libglade2 v2.6.4, pidgin v2.5.7 against the version in testing. +1 on the release. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpKOwQACgkQLrhmsXMSLxcCCQCfSJ20DzwMjfzWUGNxjzbWT3Yz CvQAoNl7l3cHrmXEXStl7L4WqufJOxWP =DE5/ -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue Jun 30 19:57:24 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 30 Jun 2009 12:57:24 -0500 Subject: [csw-maintainers] New in testing: gtk2 2.16.1 In-Reply-To: <4A4A3B04.2040701@opencsw.org> References: <4A4A3B04.2040701@opencsw.org> Message-ID: <4A4A5204.7070705@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > Dagobert Michelsen wrote: > >> Feedback as always welcome! > Successfully compiled and tested pygtk against 2.16.1 in testing. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpKUgQACgkQLrhmsXMSLxff+QCgvA9nuS4NtF2O+lE6k01+X8rn Dx8AoNJRTWgoLDCdVaRY/3MYrJ34MMiZ =dK3f -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue Jun 30 20:28:42 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 30 Jun 2009 13:28:42 -0500 Subject: [csw-maintainers] New in testing: gtk2 2.16.1 In-Reply-To: <4A4A5204.7070705@opencsw.org> References: <4A4A3B04.2040701@opencsw.org> <4A4A5204.7070705@opencsw.org> Message-ID: <4A4A595A.3010304@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 While playing with pygtk, I found a "nicety" for the package install. have a post install that runs: /opt/csw/bin/gdk-pixbuf-query-loaders >/opt/csw/etc/gtk-2.0/gdk-pixbuf.loaders I am going to file a feature bug against the pango package to request the same for /opt/csw/bin/pango-querymodules >/opt/csw/etc/pango/pango.modules - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpKWVkACgkQLrhmsXMSLxf+MACg5924oC6sIGEB3ZxnLq3CEXvB 930An0teOX8d1T3iI/33diKsJxfM+jNe =/U7x -----END PGP SIGNATURE----- From car at opencsw.org Tue Jun 30 23:57:29 2009 From: car at opencsw.org (Chris Reece) Date: Wed, 1 Jul 2009 09:57:29 +1200 (NZST) Subject: [csw-maintainers] tcsh 6.16.00 in testing Message-ID: <15954.131.203.105.1.1246399049.squirrel@webmail.jessies.org> On Tue, June 30, 2009 11:36 pm, Peter Bonivart wrote: > It was years ago I used tcsh but it installs/removes cleanly and seems > to work properly during my short test on a Solaris 9 Sparc and a Solaris 10 > x86. Man page also works. Thanks, Peter, much obliged. My only remaining problem is that I can't log in to www.opencsw.org, which the documentation suggests is the location of /home/newpkgs. None of my keys are accepted. Or, rather, there are "Too many authentication failures" after all of my keys have been tried. I can log in to login.opencsw.org, and from there to build8s and build8x. From phil at bolthole.com Mon Jun 1 00:36:17 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 31 May 2009 15:36:17 -0700 Subject: [csw-maintainers] Bug in our uwatch stuffs Message-ID: <20090531223617.GB32936@bolthole.com> BTW: there appears to be a bug in our "uwatch" definition and/or code. I an maintainer for the obsolete "glib"(version 1) NOT "glib2". the uwatch stuff is incorrectly reporting glib2 updates, to the glib maintainer -- me. ----- Forwarded message from Upstream Package Watch ----- Date: Sat, 30 May 2009 04:46:49 +0200 (CEST) From: Upstream Package Watch To: phil at opencsw.org Subject: [svn] glib upstream update notification Hello dear glib maintainer, The upstream notification job has detected the availability of new files for glib. The following upstream file(s): glib-2.20.3.tar.bz2 is/are available at the following url(s): http://ftp.gnome.org/pub/GNOME/sources/glib/2.20/ Please consider updating your package. -- Kindest regards upstream notification job ----- End forwarded message ----- From skayser at opencsw.org Mon Jun 1 12:21:22 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 01 Jun 2009 12:21:22 +0200 Subject: [csw-maintainers] Anyone successfully downloaded gnulib? Download problems here. Message-ID: <4A23ABA2.5080001@opencsw.org> Hi, has anyone successfully cloned the gnulib git repository? When i try to follow the download instructions from their homepage [1] the clone operation times out. # git clone git://git.savannah.gnu.org/gnulib.git Initialized empty Git repository in /root/tmp/gnulib/.git/ cdgit.savannah.gnu.org[0: 199.232.41.69]: errno=Connection timed out fatal: unable to connect a socket (Connection timed out) I am doing this from my home box, i.e. there is no proxy required and no restrictive firewall in between. Sebastian [1] http://www.gnu.org/software/gnulib/ From skayser at opencsw.org Mon Jun 1 12:29:48 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 01 Jun 2009 12:29:48 +0200 Subject: [csw-maintainers] Anyone successfully downloaded gnulib? Download problems here. In-Reply-To: <4A23ABA2.5080001@opencsw.org> References: <4A23ABA2.5080001@opencsw.org> Message-ID: <4A23AD9C.4030101@opencsw.org> Sebastian Kayser wrote: > has anyone successfully cloned the gnulib git repository? When i try to > follow the download instructions from their homepage [1] the clone > operation times out. Ok, never mind, i just noticed the news on their front page [1]. A couple of services is down, git is one of them. Has anyone ever thought of packaging gnulib so that we have it centrally available on the buildfarm? Sebastian [1] http://savannah.gnu.org/forum/forum.php?forum_id=5826 From dam at opencsw.org Mon Jun 1 12:43:51 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 1 Jun 2009 12:43:51 +0200 Subject: [csw-maintainers] Anyone successfully downloaded gnulib? Download problems here. In-Reply-To: <4A23AD9C.4030101@opencsw.org> References: <4A23ABA2.5080001@opencsw.org> <4A23AD9C.4030101@opencsw.org> Message-ID: <8221E473-F04F-4A59-858D-BDA1EC489722@opencsw.org> Hi Sebastian, Am 01.06.2009 um 12:29 schrieb Sebastian Kayser: > Sebastian Kayser wrote: >> has anyone successfully cloned the gnulib git repository? When i >> try to >> follow the download instructions from their homepage [1] the clone >> operation times out. > > Ok, never mind, i just noticed the news on their front page [1]. A > couple of services is down, git is one of them. > > Has anyone ever thought of packaging gnulib so that we have it > centrally > available on the buildfarm? gnulib is not meant to be packaged. The projects needing them should include the functionality they actually need: > Gnulib takes a different approach. Its components are intended to be > shared at the source level, rather than being a library that gets > built, > installed, and linked against. Thus, there is no distribution tarball; > the idea is to copy files from Gnulib into your own source tree. Best regards -- Dago From dam at opencsw.org Mon Jun 1 12:46:14 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 1 Jun 2009 12:46:14 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> Message-ID: Hi, Am 31.05.2009 um 23:00 schrieb Dagobert Michelsen: > 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- > packages/xcbgen/xtypes.py I guess this is the reason why the py-files aren't gettint compiled: /opt/csw/X11/ Should I put the py-files into /opt/csw/lib/python* instead regardless of the libdir setting? Best regards -- Dago From skayser at opencsw.org Mon Jun 1 12:55:00 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 01 Jun 2009 12:55:00 +0200 Subject: [csw-maintainers] Anyone successfully downloaded gnulib? Download problems here. In-Reply-To: <8221E473-F04F-4A59-858D-BDA1EC489722@opencsw.org> References: <4A23ABA2.5080001@opencsw.org> <4A23AD9C.4030101@opencsw.org> <8221E473-F04F-4A59-858D-BDA1EC489722@opencsw.org> Message-ID: <4A23B384.3090302@opencsw.org> Dagobert Michelsen wrote: > Am 01.06.2009 um 12:29 schrieb Sebastian Kayser: >> Sebastian Kayser wrote: >>> has anyone successfully cloned the gnulib git repository? When i >>> try to >>> follow the download instructions from their homepage [1] the clone >>> operation times out. >> Ok, never mind, i just noticed the news on their front page [1]. A >> couple of services is down, git is one of them. >> >> Has anyone ever thought of packaging gnulib so that we have it >> centrally >> available on the buildfarm? > > gnulib is not meant to be packaged. The projects needing them should > include the functionality they actually need: > >> Gnulib takes a different approach. Its components are intended to be >> shared at the source level, rather than being a library that gets >> built, >> installed, and linked against. Thus, there is no distribution tarball; >> the idea is to copy files from Gnulib into your own source tree. That's exactly what i was aiming at. Having all the source files handy in one location (package) gives our maintainers instant access to them. No hassle cloning non-available git repositories. Debian has a gnulib package also. I wonder why, but heck, when even they as a GNU platform have one we should have one too ;) Sebastian From pfelecan at opencsw.org Mon Jun 1 17:43:31 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 01 Jun 2009 17:43:31 +0200 Subject: [csw-maintainers] assembler in shared libraries Message-ID: Trying to package/port projects using optimizations implemented in assembler I encounter the same issue for which I didn't found a solution. For the moment, the issue manifests itself only on i386 architecture as the porting has as source, projects developed on Linux and concerns mainly video applications, e.g., ffmpeg, xvid, &c. In the linking phase of the share library I get the following message: Text relocation remains referenced against symbol offset in file 0x7 bitstream/x86_asm/cbp_mmx.o . . . 0xc6 utils/x86_asm/cpuid.o ld: fatal: relocations remain against allocatable but non-writable sections collect2: ld returned 1 exit status The result of my research can be summarized as follows: don't use hand written assembly in shared libraries if you're not using the GNU linker. The consequences of this are: - don't activate assembler optimizations: poor performance - where the previous is not possible, don't generate shared libraries containing the assembler code: build only statically linked binaries --- at least the corresponding library is statically linked Is there a fellow maintainer having a positive experience on building shared libraries for Solaris containing assembler code and willing to share his solution? -- Peter From bwalton at opencsw.org Mon Jun 1 17:50:42 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 01 Jun 2009 11:50:42 -0400 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: References: Message-ID: <1243871326-sup-7314@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Mon Jun 01 11:43:31 -0400 2009: > Is there a fellow maintainer having a positive experience on building > shared libraries for Solaris containing assembler code and willing to > share his solution? This isn't something I've bumped into personally, but would this be a candidate for using the gnu linker instead of the solaris one? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Mon Jun 1 17:51:55 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 10:51:55 -0500 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: References: Message-ID: <4A23F91B.1020803@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter FELECAN wrote: > Trying to package/port projects using optimizations implemented in > assembler I encounter the same issue for which I didn't found a solution. > > For the moment, the issue manifests itself only on i386 architecture as > the porting has as source, projects developed on Linux and concerns > mainly video applications, e.g., ffmpeg, xvid, &c. > > In the linking phase of the share library I get the following message: > > Text relocation remains referenced > against symbol offset in file > 0x7 bitstream/x86_asm/cbp_mmx.o > . > . > . > 0xc6 utils/x86_asm/cpuid.o > ld: fatal: relocations remain against allocatable but non-writable sections > collect2: ld returned 1 exit status > > The result of my research can be summarized as follows: don't > use hand written assembly in shared libraries if you're not using the > GNU linker. > > The consequences of this are: > > - don't activate assembler optimizations: poor performance > - where the previous is not possible, don't generate shared > libraries containing the assembler code: build only statically linked > binaries --- at least the corresponding library is statically linked > > Is there a fellow maintainer having a positive experience on building > shared libraries for Solaris containing assembler code and willing to > share his solution? > The solution I used was to use gnu linker. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoj+RsACgkQLrhmsXMSLxcL8wCggoidR+PkaZpTmwwLADpfWA59 OqkAoJRJiNVJjHkYG9PPBZ2bksBhRihv =etrg -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 1 18:02:17 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 11:02:17 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> Message-ID: <4A23FB89.50006@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Dago, Dagobert Michelsen wrote: >> root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/xtypes.py > > I guess this is the reason why the py-files aren't gettint compiled: > /opt/csw/X11/ > Should I put the py-files into /opt/csw/lib/python* instead regardless > of the > libdir setting? if the library files are to be used within python programs yes. that is to say, if a python script does something like "from xtypes import foo" or "import xtypes" then it needs to go under /opt/csw/lib/python/site-packages - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoj+4kACgkQLrhmsXMSLxf6HwCfUVAjnDAkrfyB3te0iX/Cy/hc W1oAnA3fZy5Rz644LID5BOaeDrczbVQX =SkxE -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 1 18:19:13 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 11:19:13 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> Message-ID: <4A23FF81.9070903@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi, > > Am 31.05.2009 um 23:00 schrieb Dagobert Michelsen: >> 0 17552 cpio 3 >> root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/xtypes.py > > I guess this is the reason why the py-files aren't gettint compiled: > /opt/csw/X11/ > Should I put the py-files into /opt/csw/lib/python* instead regardless > of the > libdir setting? I did some testing on the class script this morning, and it works as designed. as long as your .py library files are under /opt/csw/lib/python the class script will attempt to compile them. **** NOTE **** The default location for the library files is currently NOT /opt/csw/lib/python The version in testing has fixed this problem. Please hold your python packages until the new one is on the build farm. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoj/4EACgkQLrhmsXMSLxfnMwCcDoXFiOBp9rlahNMshfqetRPj vRkAoKjnUl5AbjEzgMf3kb4EGkyzM1O8 =IcHd -----END PGP SIGNATURE----- From skayser at opencsw.org Mon Jun 1 18:37:36 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 01 Jun 2009 18:37:36 +0200 Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") Message-ID: <4A2403D0.4090109@opencsw.org> Hi, a couple of RFEs for the "GAR package status" [1] page. Can we get: * an "last updated" field at the top of the page * the numbers / stats right at the top of the page (instead of further down) * a contact address for the maintainer of the page on the page (for queries like this one and to get packages that are in GAR but not listed as such fixed) Ah, and some sort of monthly historical data would be cool also. So that we can see some progress. Sebastian [1] http://www.opencsw.org/package-gar-status.html From dam at opencsw.org Mon Jun 1 18:51:05 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 1 Jun 2009 18:51:05 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A23FF81.9070903@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> <4A23FF81.9070903@opencsw.org> Message-ID: Hi Mike, Am 01.06.2009 um 18:19 schrieb Mike Watters: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Dagobert Michelsen wrote: >> Hi, >> >> Am 31.05.2009 um 23:00 schrieb Dagobert Michelsen: >>> 0 17552 cpio 3 >>> root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/xtypes.py >> >> I guess this is the reason why the py-files aren't gettint compiled: >> /opt/csw/X11/ >> Should I put the py-files into /opt/csw/lib/python* instead >> regardless >> of the >> libdir setting? > > I did some testing on the class script this morning, and it works > as designed. > as long as your .py library files are under /opt/csw/lib/python the > class > script will attempt to compile them. > > **** NOTE **** > The default location for the library files is currently NOT /opt/ > csw/lib/python > The version in testing has fixed this problem. Please hold your > python > packages until the new one is on the build farm. Just for reference, why aren't you compiling just the pathes you get and then installf them? I remember you said performance, but if you only installf -f once in postinstall this shouldn't be an issue. Best regards -- Dago From mwatters at opencsw.org Mon Jun 1 20:24:18 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 13:24:18 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> <4A23FF81.9070903@opencsw.org> Message-ID: <4A241CD2.7050803@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Dago, Dagobert Michelsen wrote: > > Just for reference, why aren't you compiling just the pathes you get and > then installf them? I remember you said performance, but if you only > installf -f once in postinstall this shouldn't be an issue. The compile of all the files was done very quickly ~10 - 15 seconds. Even with the single installf -f it took forever between 1 and 3 seconds per file. that does not sound like much, but when you start talking about 2000 or 3000 files it adds up quickly. I am not sure if that is a problem with my machine? large contents file, or some other issue? or if it is normal for installf to be that slow. I would rather do it this way, which does not show that kind of response time. I think this makes for a Better user experience. installed on a fairly quiet v490 running solaris 10 for my benchmark. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkokHNEACgkQLrhmsXMSLxfnogCdEpHayfPHrzWeHdiYse7HUk1o a/cAoIhjUk9UzpB6obefhfmFVqaTCurq =io4B -----END PGP SIGNATURE----- From phil at bolthole.com Mon Jun 1 20:27:46 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 1 Jun 2009 11:27:46 -0700 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A241CD2.7050803@opencsw.org> References: <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> <4A23FF81.9070903@opencsw.org> <4A241CD2.7050803@opencsw.org> Message-ID: <20090601182746.GI35922@bolthole.com> On Mon, Jun 01, 2009 at 01:24:18PM -0500, Mike Watters wrote: > Even with the single installf -f it took forever > between 1 and 3 seconds per file. that does not sound like much, > but when you start talking about 2000 or 3000 files it adds up quickly. > I am not sure if that is a problem with my machine? large contents file, or > some other issue? or if it is normal for installf to be that slow. I would > rather do it this way, which does not show that kind of response time. > I think this makes for a Better user experience. > > installed on a fairly quiet v490 running solaris 10 for my benchmark. I am not sure what "this way" is. I looked at http://wiki.opencsw.org/cswclassutils-package and it SEEMS to have a very brief gar directive. Would you please update the wiki page to state that explicitly... and ALSO to give a more "manual" description of what is neccessary, if one is putting together a package by hand rather than using gar? From mwatters at opencsw.org Mon Jun 1 20:34:49 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 13:34:49 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: <20090601182746.GI35922@bolthole.com> References: <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> <4A23FF81.9070903@opencsw.org> <4A241CD2.7050803@opencsw.org> <20090601182746.GI35922@bolthole.com> Message-ID: <4A241F49.9060401@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > I am not sure what "this way" is. > I looked at http://wiki.opencsw.org/cswclassutils-package and it SEEMS > to have a very brief gar directive. > > Would you please update the wiki page to state that explicitly... and ALSO > to give a more "manual" description of what is neccessary, if one is > putting together a package by hand rather than using gar? "this way" == classutils Yes, I am aware of the minimalistic approach on the wiki. I gave Peter just a tiny description for the README.CSW file and he was kind enough to put that on the wiki. I have not got around to updating it with a more detailed description/usage. it is on my todo list ;-) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkokH0kACgkQLrhmsXMSLxdpPgCfT6FXv+VuGu+QJ/nEPm5A6Ri8 sl0An1pj3O7db1gWU0jwRkyBch1CvTf3 =7YFf -----END PGP SIGNATURE----- From pfelecan at opencsw.org Mon Jun 1 20:42:39 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 01 Jun 2009 20:42:39 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <4A23F91B.1020803@opencsw.org> (Mike Watters's message of "Mon\, 01 Jun 2009 10\:51\:55 -0500") References: <4A23F91B.1020803@opencsw.org> Message-ID: Mike Watters writes: > Peter FELECAN wrote: >> Trying to package/port projects using optimizations implemented in >> assembler I encounter the same issue for which I didn't found a solution. >> >> For the moment, the issue manifests itself only on i386 architecture as >> the porting has as source, projects developed on Linux and concerns >> mainly video applications, e.g., ffmpeg, xvid, &c. >> >> In the linking phase of the share library I get the following message: >> >> Text relocation remains referenced >> against symbol offset in file >> 0x7 bitstream/x86_asm/cbp_mmx.o >> . >> . >> . >> 0xc6 utils/x86_asm/cpuid.o >> ld: fatal: relocations remain against allocatable but non-writable sections >> collect2: ld returned 1 exit status >> >> The result of my research can be summarized as follows: don't >> use hand written assembly in shared libraries if you're not using the >> GNU linker. >> >> The consequences of this are: >> >> - don't activate assembler optimizations: poor performance >> - where the previous is not possible, don't generate shared >> libraries containing the assembler code: build only statically linked >> binaries --- at least the corresponding library is statically linked >> >> Is there a fellow maintainer having a positive experience on building >> shared libraries for Solaris containing assembler code and willing to >> share his solution? >> > > The solution I used was to use gnu linker. Sure. I'm not comfortable with that but why not? However, I'm flabbergasted: I thought that I knew how to do that... -B/opt/csw/bin or GCC_EXEC_PREFIX... none works! Still uses /usr/ccs/bin/ld -- Peter From bwalton at opencsw.org Mon Jun 1 20:52:50 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 01 Jun 2009 14:52:50 -0400 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: References: <4A23F91B.1020803@opencsw.org> Message-ID: <1243882320-sup-282@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Mon Jun 01 14:42:39 -0400 2009: > Sure. I'm not comfortable with that but why not? However, I'm > flabbergasted: I thought that I knew how to do that... -B/opt/csw/bin > or GCC_EXEC_PREFIX... none works! Still uses /usr/ccs/bin/ld I've seen some autoconf/configure scripts that accept the path to ld as an option and don't work against some of the other things you've tried. HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Mon Jun 1 21:01:37 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 1 Jun 2009 12:01:37 -0700 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: References: <4A23F91B.1020803@opencsw.org> Message-ID: <20090601190137.GK35922@bolthole.com> On Mon, Jun 01, 2009 at 08:42:39PM +0200, Peter FELECAN wrote: > Sure. I'm not comfortable with that but why not? However, I'm > flabbergasted: I thought that I knew how to do that... -B/opt/csw/bin > or GCC_EXEC_PREFIX... none works! Still uses /usr/ccs/bin/ld > -- a WELL-BEHAVED set of Makefiles, will allow you to override linker by simply specifying LD=/path/to/ld similar to setting LDFLAGS and so on. in one or both of the Makefile or your environment vars. From william at wbonnet.net Mon Jun 1 21:33:07 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 01 Jun 2009 21:33:07 +0200 Subject: [csw-maintainers] Bug in our uwatch stuffs In-Reply-To: <20090531223617.GB32936@bolthole.com> References: <20090531223617.GB32936@bolthole.com> Message-ID: <4A242CF3.70809@wbonnet.net> Hi Philip > BTW: there appears to be a bug in our "uwatch" definition and/or code. > It is not a bug it is a functionality :) > I an maintainer for the obsolete "glib"(version 1) NOT > "glib2". > > the uwatch stuff is incorrectly reporting glib2 updates, to the glib > maintainer -- me. > The problem comes from the fact that both package have the same GARNAME in the Makefile. glib2 is using glib as it's name, thus uwatch considers it is your package. Maybe we should consider to move GARNAME to glib2 for glib2 :) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From phil at bolthole.com Mon Jun 1 21:43:35 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 1 Jun 2009 12:43:35 -0700 Subject: [csw-maintainers] Bug in our uwatch stuffs In-Reply-To: <4A242CF3.70809@wbonnet.net> References: <20090531223617.GB32936@bolthole.com> <4A242CF3.70809@wbonnet.net> Message-ID: <20090601194335.GN35922@bolthole.com> > The problem comes from the fact that both package have the same GARNAME > in the Makefile. glib2 is using glib as it's name, thus uwatch considers > it is your package. Maybe we should consider to move GARNAME to glib2 > for glib2 :) sounds sensible. would the glib2 maintainer please fix their gar configs? :) From william at wbonnet.net Mon Jun 1 22:09:10 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 01 Jun 2009 22:09:10 +0200 Subject: [csw-maintainers] Packages age statistics : May '09 In-Reply-To: <49DA5BED.9010908@wbonnet.net> References: <49DA5BED.9010908@wbonnet.net> Message-ID: <4A243566.90903@wbonnet.net> Hi Here is the June update of package age statistics for the current source. The third column show the delta between this month and previous month, the last one the cumulated delta over the year. As you can notice packages total from years 2003 to 2008 decreased, which is good (they have been updated). Delta of packages in may is +105, which is very good. This following statistics show that 76 packages have been update over the last month, and 29 new packages added. Thanks to you all. Year Total Delta Year 1997 1 0 0 1998 1 0 0 2001 3 0 0 2002 4 0 0 2003 23 0 -3 2004 98 0 -9 2005 162 -11 -17 2006 177 -4 -27 2007 147 -26 -52 2008 304 -35 -173 2009 508 +105 +530 cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Mon Jun 1 22:50:25 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 01 Jun 2009 22:50:25 +0200 Subject: [csw-maintainers] Packages age statistics : May '09 In-Reply-To: <4A243566.90903@wbonnet.net> References: <49DA5BED.9010908@wbonnet.net> <4A243566.90903@wbonnet.net> Message-ID: <4A243F11.7080809@opencsw.org> Hi William, William Bonnet wrote: > Here is the June update of package age statistics for the current source. > > The third column show the delta between this month and previous month, > the last one the cumulated delta over the year. As you can notice > packages total from years 2003 to 2008 decreased, which is good (they > have been updated). Delta of packages in may is +105, which is very > good. This following statistics show that 76 packages have been update > over the last month, and 29 new packages added. > > Thanks to you all. > > Year Total Delta Year > 1997 1 0 0 > 1998 1 0 0 > 2001 3 0 0 > 2002 4 0 0 > 2003 23 0 -3 > 2004 98 0 -9 > 2005 162 -11 -17 > 2006 177 -4 -27 > 2007 147 -26 -52 > 2008 304 -35 -173 > 2009 508 +105 +530 would you mind elaborating on these statistics a bit more? Right now i don't know whether i am putting them together correctly. - What exactly are the figures in the Total column? Currently available packages released in that year? So we would still have one package each from back in 1997 and 1998? So taking for example the "Delta of packages in may is +105". Does it mean, we released 105 packages in May (either updated or new) that haven't been available in a 2009 revision before (thus contributing to the Total 2009 figure)? This would be quite neat. And +530 would be 530 updated/released packages in 2009? Why would it differ from the Total number of 508 then? Getting back to another year, taking 2008 for example. Do we have 304 packages in the catalog which are from 2008? Have we updated/dropped 35 of them last month and 173 over the course of this year? Sebastian From Joerg.Schilling at fokus.fraunhofer.de Mon Jun 1 23:57:39 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Mon, 01 Jun 2009 23:57:39 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <20090601190137.GK35922@bolthole.com> References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> Message-ID: <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> Philip Brown wrote: > a WELL-BEHAVED set of Makefiles, will allow you to override linker by > simply specifying > > LD=/path/to/ld Well, if the "new" LD behaves significantly similar to the expected LD, you are correct. For simple makefiles you may be correct too...... For more complex makefiles you are not correct as these makefiles may have more internal knowledge than you may expect. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From mwatters at opencsw.org Tue Jun 2 00:23:39 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 01 Jun 2009 17:23:39 -0500 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <4A2454EB.7000308@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Joerg Schilling wrote: > Philip Brown wrote: > >> a WELL-BEHAVED set of Makefiles, will allow you to override linker by >> simply specifying >> >> LD=/path/to/ld > > Well, if the "new" LD behaves significantly similar to the expected LD, > you are correct. > > For simple makefiles you may be correct too...... > > For more complex makefiles you are not correct as these makefiles may > have more internal knowledge than you may expect. > > J?rg > It is all up to the writer of the Makefile, Makefiles can force whatever the creator wants it to force by default. we always have the option of changing it. either with a patch to Makefile.in or autoconf. however, if the package uses libtool, the LD environment variable should be honored. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkokVOsACgkQLrhmsXMSLxf49wCeLk27ZKfkiKfB7Eo0WMOHdY/0 KrUAni52pJ7UXIyJgbs+R1Ge35U7On91 =32kt -----END PGP SIGNATURE----- From car at opencsw.org Tue Jun 2 08:54:46 2009 From: car at opencsw.org (Chris Reece) Date: Tue, 2 Jun 2009 18:54:46 +1200 (NZST) Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> References: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> Message-ID: <36490.131.203.105.13.1243925686.squirrel@webmail.jessies.org> On Thu, May 28, 2009 9:12 am, Dagobert Michelsen wrote: > Chris: As you already did findutils with glocate, it would be > really cool if you could also package up slocate > as more secure alternative. I looked over the GAR documentation at the weekend and an hour or so later I was none the wiser. I checked out a local copy on my one Sun box at home, found what looks like an attempt at findutils in there, not mine. Didn't get anywhere with it. In the interim, there's a traditionally-rolled, up to date package in testing: /home/testing/findutils-4.4.1,REV=2009.06.01-SunOS5.8-i386-CSW.pkg.gz /home/testing/findutils-4.4.1,REV=2009.06.01-SunOS5.8-sparc-CSW.pkg.gz There's a postinstall script in there to add a gupdatedb job in the small hours. This is the first time I've packaged in a long time, so I'd be happy to receive feedback. I have no way to test the i386 package, in particular. As for slocate, I'm guessing that it's going to require chunks of GNUlib. We're missing fts, for instance. Have you used a contemporary version of this, or seen it running on Solaris? Cheers, Chris. From pfelecan at opencsw.org Tue Jun 2 11:01:13 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 02 Jun 2009 11:01:13 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <1243882320-sup-282@ntdws12.chass.utoronto.ca> (Ben Walton's message of "Mon\, 01 Jun 2009 14\:52\:50 -0400") References: <4A23F91B.1020803@opencsw.org> <1243882320-sup-282@ntdws12.chass.utoronto.ca> Message-ID: Ben Walton writes: > Excerpts from Peter FELECAN's message of Mon Jun 01 14:42:39 -0400 2009: > >> Sure. I'm not comfortable with that but why not? However, I'm >> flabbergasted: I thought that I knew how to do that... -B/opt/csw/bin >> or GCC_EXEC_PREFIX... none works! Still uses /usr/ccs/bin/ld > > I've seen some autoconf/configure scripts that accept the path to ld > as an option and don't work against some of the other things you've > tried. The autotools for the projects that I'm porting don't offer this kind of option. This is why I tried the 2 possibilities, the environment variable being the most adequate in my case. -- Peter From skayser at opencsw.org Tue Jun 2 11:05:22 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 02 Jun 2009 11:05:22 +0200 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <36490.131.203.105.13.1243925686.squirrel@webmail.jessies.org> References: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> <36490.131.203.105.13.1243925686.squirrel@webmail.jessies.org> Message-ID: <4A24EB52.3000206@opencsw.org> Hi Chris, Chris Reece wrote: > On Thu, May 28, 2009 9:12 am, Dagobert Michelsen wrote: > >> Chris: As you already did findutils with glocate, it would be >> really cool if you could also package up slocate >> as more secure alternative. > > I looked over the GAR documentation at the weekend and an hour or so later I > was none the wiser. thanks for your effort. Could you tell us, what exactly you were missing about the documentation so that we can improve on it? Any feedback appreciated (off-list is also fine if you prefer to). Dago has just recently written some updated information on how to set up the basic compilation / GAR environment [1,2] and how to set up build descriptions for simple packages [3]. Just this weekend, I moved away most of the outdated stuff from the main page, put Dago's documents first and tried to structure the main page a bit more. > I checked out a local copy on my one Sun box at home, > found what looks like an attempt at findutils in there, not mine. Didn't get > anywhere with it. > > In the interim, there's a traditionally-rolled, up to date package in testing: > > /home/testing/findutils-4.4.1,REV=2009.06.01-SunOS5.8-i386-CSW.pkg.gz > /home/testing/findutils-4.4.1,REV=2009.06.01-SunOS5.8-sparc-CSW.pkg.gz > > There's a postinstall script in there to add a gupdatedb job in the small > hours. This is the first time I've packaged in a long time, so I'd be happy > to receive feedback. I have no way to test the i386 package, in particular. Just installed it on an i386 box, looks mostly good to me, just two minor suggestions: * Could you please use /var/opt/csw instead of /opt/csw/var This has been changed a while ago http://opencsw.org/standards/layout * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to have it in there, just for the sake of standardization. /opt/csw/bin# dump -Lv glocate | grep R*PATH [5] RUNPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib [6] RPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib > As for slocate, I'm guessing that it's going to require chunks of GNUlib. > We're missing fts, for instance. I have submitted a rudimentary build description to GAR this weekend [1], but as the savannah.gnu.org git repository was down i couldn't get on with integrating the fts module (has quite a few dependencies to other GNUlib modules from what i saw). Do we have someone fluent in autotools and/or integrating GNUlib things? Are you, Chris? So far i had only integrated minor modules w/o dependencies, maybe this one needs some more fiddling with autotools. Sebastian [1] http://apps.sourceforge.net/trac/gar/wiki/GarSetup [2] http://apps.sourceforge.net/trac/gar/wiki/GarRepository [3] http://apps.sourceforge.net/trac/gar/wiki/GarPackagingTutorial From skayser at opencsw.org Tue Jun 2 11:13:56 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 02 Jun 2009 11:13:56 +0200 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <4A24EB52.3000206@opencsw.org> References: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> <36490.131.203.105.13.1243925686.squirrel@webmail.jessies.org> <4A24EB52.3000206@opencsw.org> Message-ID: <4A24ED54.4080705@opencsw.org> Sebastian Kayser wrote: > Chris Reece wrote: >> On Thu, May 28, 2009 9:12 am, Dagobert Michelsen wrote: >> >>> Chris: As you already did findutils with glocate, it would be >>> really cool if you could also package up slocate >>> as more secure alternative. >> I looked over the GAR documentation at the weekend and an hour or so later I >> was none the wiser. > > thanks for your effort. > > Could you tell us, what exactly you were missing about the documentation > so that we can improve on it? Any feedback appreciated (off-list is also > fine if you prefer to). On a related note. If you could share your build descriptions with us (mainly configure flags or any special handling required) i can try to submit them as a GARized version to the repository. This way, or we help you get started using GAR so that you can do it yourself :) Just let us know. Sebastian From pfelecan at opencsw.org Tue Jun 2 11:38:45 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 02 Jun 2009 11:38:45 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <4A2454EB.7000308@opencsw.org> (Mike Watters's message of "Mon\, 01 Jun 2009 17\:23\:39 -0500") References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> <4A2454EB.7000308@opencsw.org> Message-ID: Mike Watters writes: > Joerg Schilling wrote: >> Philip Brown wrote: >> >>> a WELL-BEHAVED set of Makefiles, will allow you to override linker by >>> simply specifying >>> >>> LD=/path/to/ld >> >> Well, if the "new" LD behaves significantly similar to the expected LD, >> you are correct. >> >> For simple makefiles you may be correct too...... >> >> For more complex makefiles you are not correct as these makefiles may >> have more internal knowledge than you may expect. >> >> J?rg >> > > It is all up to the writer of the Makefile, Makefiles can force whatever the > creator wants it to force by default. we always have the option of changing > it. either with a patch to Makefile.in or autoconf. however, if the package > uses libtool, the LD environment variable should be honored. The projects that I try to port/package use some autotools but not libtool --- which I can understand. Using the an environment variable LD implies the knowledge by the subsequent tools of this mechanism; it's true for libtool but not for the gcc driver. Here is some additional information: - the linking is done through the gcc driver, with classic argument passing by -Wl - originally, the linker arguments are from the super-set of the GNU -linker, e.g., soname, --version-script, &c - when asking to show the invoked sub-processes: /opt/csw/gcc4/bin/gcc -v -B /opt/csw/gnu t.c -o t we obtain, for the link phase: COLLECT_GCC_OPTIONS='-v' '-B' '/opt/csw/gnu' '-o' 't' '-mtune=generic' /usr/ccs/bin/ld -V ... ld: Software Generation Utilities - Solaris Link Editors: 5.10-1.490 which is not what we expect when using -B - note that when using gcc3 there is a subtle difference: /opt/csw/gcc3/bin/gcc -v -B /opt/csw/gnu t.c -o t ... /opt/csw/gcc3/libexec/gcc/i386-pc-solaris2.8/3.4.5/collect2 -V ... ld: Software Generation Utilities - Solaris Link Editors: 5.10-1.490 the linker is called by collect2; here I'm a little bit confused as this is the first time that I see a gcc shipped without the collect2 binary. But this is another issue which is probably not related. - exploring the compiler driver search paths: gcc -print-search-dirs | grep '^programs:' programs: =/opt/csw/gcc3/libexec/gcc/i386-pc-solaris2.8/3.4.5/:...:/usr/ccs/bin/ but gcc -B /opt/csw/gnu -print-search-dirs | grep '^programs:' programs: =/opt/csw/gnu/i386-pc-solaris2.8/3.4.5/:/opt/csw/gnu/:...:/usr/ccs/bin/ shows that this has no effect even though the path is modified as expected. - trussing the driver: truss -e -f -o gcc.truss gcc -v -B /opt/csw/gnu t.c -o t shows that there is no tentative to search for a linker binary in /opt/csw/gnu There is a remark in the GCC FAQ: http://gcc.gnu.org/faq.html#gas about using --with-ld when configuring the compiler: "One drawback of this option is that it won't allow you to override the search path for assembler and linker with command-line options -B/path/ if the specified file names exist." My experiments show that this is also true when trying to use the environment variable GCC_EXEC_PREFIX -- Peter From william at wbonnet.net Tue Jun 2 13:15:11 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 02 Jun 2009 13:15:11 +0200 Subject: [csw-maintainers] pixman 0.15.8 is in testing In-Reply-To: <4A1C518A.6090503@wbonnet.net> References: <4A1C518A.6090503@wbonnet.net> Message-ID: <4A2509BF.5020706@wbonnet.net> Hi Pixman package has been updated to latest version (0.15.8) and is available from testing Any feedback is welcome kind regards, W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From pfelecan at opencsw.org Tue Jun 2 20:36:45 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 02 Jun 2009 20:36:45 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: (Peter FELECAN's message of "Tue\, 02 Jun 2009 11\:38\:45 +0200") References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> <4A2454EB.7000308@opencsw.org> Message-ID: Just for the thread completeness and reference: I'm patching the Makefile --- brutal kludge --- to use explicitly gld on i386, thus having the best performance for the platform; for sparc, the project doesn't have assembler crafted code. -- Peter From dam at opencsw.org Tue Jun 2 23:19:19 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 2 Jun 2009 23:19:19 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> <4A2454EB.7000308@opencsw.org> Message-ID: <38AF7A48-097B-4F6E-8C7F-AFB40AAF7401@opencsw.org> Hi Peter, Am 02.06.2009 um 20:36 schrieb Peter FELECAN: > Just for the thread completeness and reference: I'm patching the > Makefile > --- brutal kludge --- to use explicitly gld on i386, thus having the > best performance for the platform; for sparc, the project doesn't have > assembler crafted code. Thanks for your effort, I have a similar problem with libgmp, which also has assembler optimizations for x86. Best regards -- Dago From james at opencsw.org Wed Jun 3 11:14:19 2009 From: james at opencsw.org (James Lee) Date: Wed, 03 Jun 2009 09:14:19 GMT Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <38AF7A48-097B-4F6E-8C7F-AFB40AAF7401@opencsw.org> References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> <4A2454EB.7000308@opencsw.org> <38AF7A48-097B-4F6E-8C7F-AFB40AAF7401@opencsw.org> Message-ID: <20090603.9141900.1268094198@gyor.oxdrove.co.uk> On 02/06/09, 22:19:19, Dagobert Michelsen wrote regarding Re: [csw-maintainers] assembler in shared libraries: > > --- brutal kludge --- to use explicitly gld on i386, thus having the > > best performance for the platform; for sparc, the project doesn't have > > assembler crafted code. > Thanks for your effort, I have a similar problem with libgmp, which > also has assembler optimizations for x86. Are you doing this because you believe it's better or because you proved it's better? I had a similar situation (with mesa3d) and, after fighting to get it to compile, the assembler was slower. James. From dam at opencsw.org Wed Jun 3 12:48:13 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 3 Jun 2009 12:48:13 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> Message-ID: Hi, Am 01.06.2009 um 12:46 schrieb Dagobert Michelsen: > Am 31.05.2009 um 23:00 schrieb Dagobert Michelsen: >> 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- >> packages/xcbgen/xtypes.py > > I guess this is the reason why the py-files aren't gettint compiled: > /opt/csw/X11/ > Should I put the py-files into /opt/csw/lib/python* instead > regardless of the > libdir setting? I relocated the files to /opt/csw/lib/python/site-packages/xcbgen Works like charm now. New packages in testing/: xcbproto-1.5,REV=2009.06.03-SunOS5.8-sparc-CSW.pkg.gz xcbproto-1.5,REV=2009.06.03-SunOS5.8-i386-CSW.pkg.gz Best regards -- Dago From pfelecan at opencsw.org Wed Jun 3 19:25:34 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 03 Jun 2009 19:25:34 +0200 Subject: [csw-maintainers] assembler in shared libraries In-Reply-To: <20090603.9141900.1268094198@gyor.oxdrove.co.uk> (James Lee's message of "Wed\, 03 Jun 2009 09\:14\:19 GMT") References: <4A23F91B.1020803@opencsw.org> <20090601190137.GK35922@bolthole.com> <4a244ed3.hCB1jkwkOgR8syNY%Joerg.Schilling@fokus.fraunhofer.de> <4A2454EB.7000308@opencsw.org> <38AF7A48-097B-4F6E-8C7F-AFB40AAF7401@opencsw.org> <20090603.9141900.1268094198@gyor.oxdrove.co.uk> Message-ID: James Lee writes: > On 02/06/09, 22:19:19, Dagobert Michelsen wrote regarding > Re: [csw-maintainers] assembler in shared libraries: > >> > --- brutal kludge --- to use explicitly gld on i386, thus having the >> > best performance for the platform; for sparc, the project doesn't have >> > assembler crafted code. > >> Thanks for your effort, I have a similar problem with libgmp, which >> also has assembler optimizations for x86. > > Are you doing this because you believe it's better or because you > proved it's better? > > I had a similar situation (with mesa3d) and, after fighting to get > it to compile, the assembler was slower. As usual, James ask the good question. Trying to answer it by running a benchmark I discovered that the "brutal kludge" didn't work in a not so subtle way: - when linking, without error, with the static library, the execution halts with an "illegal instruction" in _rt_boot (real time library: using sched_yield, &c). - when linking with the shared library I get unresolved symbols; after exploring the library I conclude that there is a mistake of local instead of global for a bunch of symbols. After that, I repackage a version which didn't use assembler code, consequently being able to use the system's linker. None of the above phenomena appears and the benchmark programs ran with success. Lessons learned: 1. Test thoroughly your package 2. The GNU linker didn't work in some cases --- at least in this one 3. Sacrifice hypothetical gains for usability --- well, mine is an extreme case I'll explore the behaviour of the GNU linker as I have some more candidates with the issue that I tried to solve initially. -- Peter From bwalton at opencsw.org Thu Jun 4 02:16:19 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 03 Jun 2009 20:16:19 -0400 Subject: [csw-maintainers] cswpycompile Message-ID: <1244074532-sup-287@ntdws12.chass.utoronto.ca> Hi Mike and Peter, Any idea when the updated cswclassutils will hit the streets? I've got a package I'd like to leverage cswpycompile from. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Thu Jun 4 05:02:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 03 Jun 2009 22:02:51 -0500 Subject: [csw-maintainers] cswpycompile In-Reply-To: <1244074532-sup-287@ntdws12.chass.utoronto.ca> References: <1244074532-sup-287@ntdws12.chass.utoronto.ca> Message-ID: <4A27395B.3060504@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Hi Mike and Peter, > > Any idea when the updated cswclassutils will hit the streets? I've > got a package I'd like to leverage cswpycompile from. > > Thanks > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I have tested the fix for bugs 3633, 3634, Sebastian is testing out the fix for 3635, we need a tester for 3636. http://www.opencsw.org/mantis/view.php?id=3636 if you can test out the fix for that bug, we can get this rolled out faster. if there is no tester for 3636, I believe Peter is going to write a test case to test himself. it just may take longer then if another maintainer tests it. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkonOVsACgkQLrhmsXMSLxfsKwCgjKnDX415ybn56kadQ8jgaowl dO0AnjXj8mNGI1l5IfNjiuDuVAueIcgh =4qfW -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu Jun 4 05:17:30 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 03 Jun 2009 23:17:30 -0400 Subject: [csw-maintainers] cswpycompile In-Reply-To: <4A27395B.3060504@opencsw.org> References: <1244074532-sup-287@ntdws12.chass.utoronto.ca> <4A27395B.3060504@opencsw.org> Message-ID: <1244085411-sup-6497@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Wed Jun 03 23:02:51 -0400 2009: > I have tested the fix for bugs 3633, 3634, Sebastian is testing out > the fix for 3635, we need a tester for > 3636. http://www.opencsw.org/mantis/view.php?id=3636 if you can test > out the fix for that bug, we can get this rolled out faster. Ok, let me see what I can do over the next few days. > if there is no tester for 3636, I believe Peter is going to write a > test case to test himself. it just may take longer then if another > maintainer tests it. Of course. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Thu Jun 4 05:20:28 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 03 Jun 2009 23:20:28 -0400 Subject: [csw-maintainers] gitosis in testing Message-ID: <1244085481-sup-5308@ntdws12.chass.utoronto.ca> Hi All, I've rolled a package for gitosis[1], which is a cool way to host multiple git repositories[2]. I'm waiting for the cswpycompile addition to cswclassutils before I release it, so I can have the .py files handled automatically, but the base package seems pretty good at this point. It's available from testing/. Any feedback is welcome. Thanks -Ben [1] http://swik.net/gitosis [2] http://scie.nti.st/2007/11/14/hosting-git-repositories-the-easy-and-secure-way -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From car at opencsw.org Thu Jun 4 08:49:09 2009 From: car at opencsw.org (Chris Reece) Date: Thu, 4 Jun 2009 18:49:09 +1200 (NZST) Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm Message-ID: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> On Tue, June 2, 2009 9:05 pm, Sebastian Kayser wrote: > Could you tell us, what exactly you were missing about the documentation > so that we can improve on it? Any feedback appreciated (off-list is also fine > if you prefer to). Specifically, no. I recall being daunted by finding three versions of GAR, none of which appeared to have a full set of GAR newbie documentation. I found a "Getting started with GAR" document that, depending on which link I followed to it, from where: was not found; contained a page blank except for "redirect"; or was there, but was listed under some manner of "this is for an older GAR -- here be dragons" warning. I was in a hurry, may have missed something and, believe me, I know what it's like to not have enough hours in the day. Documentation's no damned fun. Your suggestion of hand-holding to en-GAR findutils sounds like a fine idea, if you've the time. I'd rather put these things in a format that they'll be useful, for what that's worth. But if I am just being dense, feel free to point me at the docs that you know tell me everything I need and tell me to RTFM. > Just installed it on an i386 box, looks mostly good to me, just two > minor suggestions: > > * Could you please use /var/opt/csw instead of /opt/csw/var > This has been changed a while ago http://opencsw.org/standards/layout I considered this, and thought better of it because I was worried about people upgrading from the current (!) and previous versions, that put locatedb in /opt/csw/var. But, thinking about it now, it's as simple as testing for the presence of the old file at install time and moving it. I'll do that. > * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to have > it in there, just for the sake of standardization. > /opt/csw/bin# dump -Lv glocate | grep R*PATH > [5] RUNPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib > [6] RPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib I set LD_OPTIONS as per OpenCSW documentation, like so: LD_OPTIONS='-R/opt/csw/lib/$ISALIST -R/opt/csw/lib -L/opt/csw/lib' I wondered if I'd done something silly like forgotten that I'd set LDFLAGS too, but a clean build just now repeats the error, even before it's staged: -bash-4.0$ dump -Lv locate/locate | grep R*PATH [5] RUNPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib [6] RPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib Environment looks like this: CPPFLAGS=-I/opt/csw/include LD_OPTIONS=-R/opt/csw/lib/$ISALIST -R/opt/csw/lib -L/opt/csw/lib PATH=/opt/studio/SOS11/SUNWspro/bin:/usr/bin:/bin:/usr/sbin:/sbin:/opt/csw/bin:/usr/ccs/bin:/usr/ucb MAKE=gmake CFLAGS=-fast -xarch=v8 -xnolibmopt CC=cc If I set: LD_OPTIONS='-R/opt/csw/lib -L/opt/csw/lib' as I used to do, then I get: [5] RUNPATH /opt/csw/lib [6] RPATH /opt/csw/lib If I set: LD_OPTIONS='-R/opt/csw/lib/$ISALIST -L/opt/csw/lib' I get: -bash-4.0$ dump -Lv locate/locate | grep R*PATH [5] RUNPATH /opt/csw/lib/$ISALIST:/opt/csw/lib [6] RPATH /opt/csw/lib/$ISALIST:/opt/csw/lib which I assume is what we're aiming for. > Are you, Chris? Not at all. Sorry. Cheers, Chris. From bonivart at opencsw.org Thu Jun 4 09:24:59 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 4 Jun 2009 09:24:59 +0200 Subject: [csw-maintainers] Bug in createpkg Message-ID: <625385e30906040024k635899c7wddb4c90f1457e697@mail.gmail.com> I just manually put together a test package and got this: "ERROR: fix csw/man entry in prototype file" I assume this is supposed to catch using csw/man paths instead of csw/share/doc but I have this in the prototype: "f none /etc/opt/csw/manifest 0644 root bin" This is the line in createpkg that triggers: "grep 'csw/man' $PROTO && die fix csw/man entry in $PROTO file" There's a bunch of similar lines, they should use word boundaries with RE's so not to be so easily fooled. -- /peter From bonivart at opencsw.org Thu Jun 4 10:06:59 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 4 Jun 2009 10:06:59 +0200 Subject: [csw-maintainers] cswpycompile In-Reply-To: <4A27395B.3060504@opencsw.org> References: <1244074532-sup-287@ntdws12.chass.utoronto.ca> <4A27395B.3060504@opencsw.org> Message-ID: <625385e30906040106w2a078e72gc14ace3b1b82fb1b@mail.gmail.com> On Thu, Jun 4, 2009 at 5:02 AM, Mike Watters wrote: > I have tested the fix for bugs 3633, 3634, Sebastian is testing out the fix for > 3635, we need a tester for 3636. http://www.opencsw.org/mantis/view.php?id=3636 > if you can test out the fix for that bug, we can get this rolled out faster. > > if there is no tester for 3636, I believe Peter is going to write a test case > to test himself. ?it just may take longer then if another maintainer tests it. I have now tested 3636 (custom manifest) and it works. Sebastian has tested everything about 3635 until the last part and I just tested that myself and it works as well. I'm releasing 1.16. -- /peter From bwalton at opencsw.org Thu Jun 4 14:59:01 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 04 Jun 2009 08:59:01 -0400 Subject: [csw-maintainers] cswpycompile In-Reply-To: <625385e30906040106w2a078e72gc14ace3b1b82fb1b@mail.gmail.com> References: <1244074532-sup-287@ntdws12.chass.utoronto.ca> <4A27395B.3060504@opencsw.org> <625385e30906040106w2a078e72gc14ace3b1b82fb1b@mail.gmail.com> Message-ID: <1244120299-sup-8752@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Thu Jun 04 04:06:59 -0400 2009: Hi Peter, > I have now tested 3636 (custom manifest) and it works. Sebastian has > tested everything about 3635 until the last part and I just tested > that myself and it works as well. I'm releasing 1.16. That's great! Thanks for letting me know. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Thu Jun 4 15:27:49 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 04 Jun 2009 08:27:49 -0500 Subject: [csw-maintainers] cswpycompile In-Reply-To: <1244120299-sup-8752@ntdws12.chass.utoronto.ca> References: <1244074532-sup-287@ntdws12.chass.utoronto.ca> <4A27395B.3060504@opencsw.org> <625385e30906040106w2a078e72gc14ace3b1b82fb1b@mail.gmail.com> <1244120299-sup-8752@ntdws12.chass.utoronto.ca> Message-ID: <4A27CBD5.7000902@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Peter Bonivart's message of Thu Jun 04 04:06:59 -0400 2009: > > Hi Peter, > >> I have now tested 3636 (custom manifest) and it works. Sebastian has >> tested everything about 3635 until the last part and I just tested >> that myself and it works as well. I'm releasing 1.16. > > That's great! Thanks for letting me know. > Excellent, I am releasing python - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkony9UACgkQLrhmsXMSLxecwACg3cknLgPFCJoIwGp5afmb3yjJ TscAoNWqYfaTRitqyQ+3WJ1SO+z7TcQh =x0hO -----END PGP SIGNATURE----- From phil at bolthole.com Thu Jun 4 18:04:53 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 4 Jun 2009 09:04:53 -0700 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> References: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> Message-ID: <20090604160453.GB2284@bolthole.com> On Thu, Jun 04, 2009 at 06:49:09PM +1200, Chris Reece wrote: > > * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to have > > it in there, just for the sake of standardization. > > > /opt/csw/bin# dump -Lv glocate | grep R*PATH > > [5] RUNPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib > > [6] RPATH /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib having both RUNPATH and RPATH in there is normal, for some reason. dont worry about it. From phil at bolthole.com Thu Jun 4 18:20:19 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 4 Jun 2009 09:20:19 -0700 Subject: [csw-maintainers] Bug in createpkg In-Reply-To: <625385e30906040024k635899c7wddb4c90f1457e697@mail.gmail.com> References: <625385e30906040024k635899c7wddb4c90f1457e697@mail.gmail.com> Message-ID: <20090604162019.GH2284@bolthole.com> On Thu, Jun 04, 2009 at 09:24:59AM +0200, Peter Bonivart wrote: > I just manually put together a test package and got this: > > "ERROR: fix csw/man entry in prototype file" > > I assume this is supposed to catch using csw/man paths instead of > csw/share/doc but I have this in the prototype: > > "f none /etc/opt/csw/manifest 0644 root bin" ahaha. ok, thanks for debugging that :-) From skayser at opencsw.org Thu Jun 4 18:39:34 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 4 Jun 2009 18:39:34 +0200 (CEST) Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <20090604160453.GB2284@bolthole.com> References: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> <20090604160453.GB2284@bolthole.com> Message-ID: <54487.194.246.122.22.1244133574.squirrel@ssl.skayser.de> Philip Brown wrote: > On Thu, Jun 04, 2009 at 06:49:09PM +1200, Chris Reece wrote: >> > * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to >> have >> > it in there, just for the sake of standardization. >> >> > /opt/csw/bin# dump -Lv glocate | grep R*PATH >> > [5] RUNPATH >> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib >> > [6] RPATH >> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib > > having both RUNPATH and RPATH in there is normal, for some reason. dont > worry about it. IIRC RPATH had been used in the past, but was considered evil as it had priority over LD_LIBRARY_PATH and you could not override it. That's why RUNPATH had been introduced. - RPATH (deprecated, not used) - LD_LIBRARY_PATH - RUNPATH - crle settings I would suppose RPATH is still in there for compatibility purposes. Sebastian From phil at bolthole.com Fri Jun 5 00:30:51 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 4 Jun 2009 15:30:51 -0700 Subject: [csw-maintainers] /opt/csw/X11 notes Message-ID: <20090604223051.GB97249@bolthole.com> FYI: with William and Dagobert now releasing packages for it, I have added a note about /opt/csw/X11, to http://www.opencsw.org/standards/layout , so everyone will have a clear idea of their purpose. The note is short enough that I will just replicate it here, to save people some effort: [/opt/csw/] X11 This is for our own packaged X11 libs, loosely equivalent to /usr/X11R6 on other systems. Up until 2009, we have relied upon the system level X11 libraries shipped with Solaris. In most cases, we wish to continue to do so. However, there are some recent applications that require modern versions of X11 libraries. We make them available in this separate subtree, so that applications will not accidentally pull them in at compile time, but default to the system ones. That way, our builds will only pull in our own X11 libs as dependancies, if the maintainer decides it is absolutely required. From mwatters at opencsw.org Fri Jun 5 16:23:00 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 09:23:00 -0500 Subject: [csw-maintainers] pysetuptools now in testing Message-ID: <4A292A44.8000305@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: package now compiles libraries on install. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopKkQACgkQLrhmsXMSLxcuyQCgn/ntf9ps/pH9B7IFq6RN34Qc DD8AniRWnQrttctHMqeB/iigEHK6W4fW =RXm7 -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri Jun 5 16:26:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 09:26:22 -0500 Subject: [csw-maintainers] pydocutils now in testing Message-ID: <4A292B0E.6030808@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: now compiles bytecode upon install. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopKw4ACgkQLrhmsXMSLxeH3wCcDcgl/72RjLWqrLRoqvnUAjBo D/wAoLfThqmtMdSdZghKbSvr+hCtmtZ1 =jc5W -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri Jun 5 16:50:09 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 09:50:09 -0500 Subject: [csw-maintainers] ap2_modpython now in testing Message-ID: <4A2930A1.2040708@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: now compiles bytecode upon install. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopMKEACgkQLrhmsXMSLxeYSgCgiO7bGsfKKQuLAjZ4S2WY7/5j WS4AoMIXA1LS00cbvjQPNSM0/pjxZKxG =xsu2 -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri Jun 5 16:50:53 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 09:50:53 -0500 Subject: [csw-maintainers] mysql-python now in testing Message-ID: <4A2930CD.4000109@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: compiles bytecode on install - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopMM0ACgkQLrhmsXMSLxeENQCbBprvZkRZG+VQsClCRDMWteS0 az4AoNyg+QENACSZgq18rn86eF/I3PRU =jmo4 -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri Jun 5 16:59:43 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 09:59:43 -0500 Subject: [csw-maintainers] pysqlite now in testing Message-ID: <4A2932DF.906@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: location and bytecode - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopMt8ACgkQLrhmsXMSLxeBDQCfaRy2tPEWHg0h7qGXQlNv4Um0 jkEAoOmNZPvWwJh3OI+Hwa5I8pM5qsEc =7Y2h -----END PGP SIGNATURE----- From skayser at opencsw.org Fri Jun 5 18:27:40 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 5 Jun 2009 18:27:40 +0200 (CEST) Subject: [csw-maintainers] Jumpstart post-install recipe for OpenCSW Message-ID: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> Hi, does someone have a working "install me some OpenCSW packages" Jumpstart post-install snippet that he could pass to me? I am in the process of setting up my first Jumpstart server and i would like to have it install a list of CSW pkgs after setting up the machine. Is this known to work with the various postinstall scripts (or cswinitsmf) that we have in the packages? And out of curiosity: What are the specifics to watch out for so that a postinstall script will work in a Jumpstart environment, just $PKG_INSTALL_ROOT [1] or are there any other caveats that pop up frequently? Sebastian [1] http://www.opencsw.org/standards/pkgcreation From mwatters at opencsw.org Fri Jun 5 18:34:00 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 11:34:00 -0500 Subject: [csw-maintainers] SilverCity now in testing Message-ID: <4A2948F8.6050900@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: location and no more bytecode in package. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopSPgACgkQLrhmsXMSLxc7vQCfdytvq+opa5cf1yeHtO3vsniF PcAAoJpEZhcvhe2PmTwBFESNvnGHH0qk =5G7x -----END PGP SIGNATURE----- From phil at bolthole.com Fri Jun 5 18:36:52 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 5 Jun 2009 09:36:52 -0700 Subject: [csw-maintainers] Jumpstart post-install recipe for OpenCSW In-Reply-To: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> References: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> Message-ID: <20090605163652.GC73123@bolthole.com> On Fri, Jun 05, 2009 at 06:27:40PM +0200, Sebastian Kayser wrote: > Hi, > > does someone have a working "install me some OpenCSW packages" Jumpstart > post-install snippet that he could pass to me? I am in the process of > setting up my first Jumpstart server and i would like to have it install a > list of CSW pkgs after setting up the machine. > > Is this known to work with the various postinstall scripts (or cswinitsmf) > that we have in the packages? If it doesnt, its a bug in the postinstall script. > And out of curiosity: What are the specifics to watch out for so that a > postinstall script will work in a Jumpstart environment, just > $PKG_INSTALL_ROOT [1] or are there any other caveats that pop up > frequently? That, and also the issue that you need to be aware that you may not be running on a "live" system. EG: If PKG_INSTALL_ROOT != /, then you cant run smf related commands, since there is no svc.configd running. which kinda sucks. and no you cant run svc.config chrooted, btw. not that i've found anyway. So that sort of stuff needs to be added post-reboot at the moment. * * * * This is why it's important that people use the class utility scripts to add in smf support. So that when and if we figure out improvements in this area, everyone gets to use it without having to rebuild their packages. ADDITIONALLY... it might behoove us to formalize some kind of "post reboot helper script" hook for our packages. That way, we could have those packages/classes that need it, have a specific integrated way to register "run this after reboot", like IPS allows. From mwatters at opencsw.org Fri Jun 5 18:46:16 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 11:46:16 -0500 Subject: [csw-maintainers] Genshi now in testing Message-ID: <4A294BD8.7090706@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 comments welcome: - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopS9cACgkQLrhmsXMSLxcJ+gCg3wwAfQL/hb7WK4nq/al3/gTJ ODwAoNz9oxZY5GAdrvTlkSZRArXT/NbH =DN1A -----END PGP SIGNATURE----- From Darin.Perusich at cognigencorp.com Fri Jun 5 19:01:15 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Fri, 05 Jun 2009 13:01:15 -0400 Subject: [csw-maintainers] Jumpstart post-install recipe for OpenCSW In-Reply-To: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> References: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> Message-ID: <4A294F5B.1070905@cognigencorp.com> Sebastian Kayser wrote: > Hi, > > does someone have a working "install me some OpenCSW packages" Jumpstart > post-install snippet that he could pass to me? I am in the process of > setting up my first Jumpstart server and i would like to have it install a > list of CSW pkgs after setting up the machine. I attached my CSW package installation script. It's nothing fancy but get the job done for me. > Is this known to work with the various postinstall scripts (or cswinitsmf) > that we have in the packages? > > And out of curiosity: What are the specifics to watch out for so that a > postinstall script will work in a Jumpstart environment, just > $PKG_INSTALL_ROOT [1] or are there any other caveats that pop up > frequently? If you chroot the package installation you don't need to worry about $PKG_INSTALL_ROOT being set. I use chroot to install everything except pkg-get and I've never had a problem. Granted I only install a small number of package, sudo, amanda, etc, and build the servers out afterwards. If it's a machine I need to make a ton of clones of I set it up and create flash images of anyways. The biggest issues I've come across was pre/post scripts which ask questions and wait for a response. Kind of defeats the purpose of hands off installs right. I haven't run across this in years but I wanted to mention it as something to watch out for. -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: cswpkginst URL: From mwatters at opencsw.org Fri Jun 5 19:01:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 12:01:26 -0500 Subject: [csw-maintainers] pygobject now in testing Message-ID: <4A294F66.9030506@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: compiled against glib2-2.20.0,REV=2009.04.08 and python 2.6.2 feedback welcome. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopT2YACgkQLrhmsXMSLxf1RgCg4ysLMCvkV4xLZ+sev8bdwyoR jx0AoKyQuGtm+aTuUzU/0T7FSw5MN34M =VFBF -----END PGP SIGNATURE----- From phil at bolthole.com Fri Jun 5 19:13:00 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 5 Jun 2009 10:13:00 -0700 Subject: [csw-maintainers] Jumpstart post-install recipe for OpenCSW In-Reply-To: <4A294F5B.1070905@cognigencorp.com> References: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de> <4A294F5B.1070905@cognigencorp.com> Message-ID: <20090605171300.GA54426@bolthole.com> Side comment: On Fri, Jun 05, 2009 at 01:01:15PM -0400, Darin Perusich wrote: > > for ii in $PKGLIST > do > chroot /a /opt/csw/bin/pkg-get install $ii > done > This is mildly inefficient. You could just have a single line, chroot /a /opt/csw/bin/pkg-get install $PKGLIST :-) From mwatters at opencsw.org Fri Jun 5 20:42:47 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 05 Jun 2009 13:42:47 -0500 Subject: [csw-maintainers] dbus-python now in testing Message-ID: <4A296727.707@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 feedback welcome! - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkopZycACgkQLrhmsXMSLxeU7ACfa6L5nLeEuSN+I+XF/jq72CBa HacAn2U+2UIv3DXk1l1hs+0p+L3QWl/o =DVxp -----END PGP SIGNATURE----- From bwalton at opencsw.org Fri Jun 5 23:21:13 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 05 Jun 2009 17:21:13 -0400 Subject: [csw-maintainers] pysetuptools Message-ID: <1244236697-sup-875@ntdws12.chass.utoronto.ca> Hi Mike, I'm confirming functionality of the updated pysetuptools for you. I can build and install the gitosis package with it. Thanks for the nice work! :) -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From jeff at cjsa.com Fri Jun 5 23:32:44 2009 From: jeff at cjsa.com (Jeffery Small) Date: Fri, 5 Jun 2009 21:32:44 GMT Subject: [csw-maintainers] dbus-python now in testing References: <4A296727.707@opencsw.org> Message-ID: Speaking of dbus, was there any consensus on the status of the dbus service? Can anyone successfully disable the svc:/system/cswdbus:default service on their system with the command: svcadm disable system/cswdbus:default I'm still looking for a resolution to the problem of why my system will not halt successfully. Regards, -- Jeff C. Jeffery Small CJSA LLC 206-232-3338 jeff at cjsa.com 7000 E Mercer Way, Mercer Island, WA 98040 From william at wbonnet.net Sat Jun 6 23:32:18 2009 From: william at wbonnet.net (William Bonnet) Date: Sat, 06 Jun 2009 23:32:18 +0200 Subject: [csw-maintainers] libXdcmp now in testing Message-ID: <4A2AE062.3070707@wbonnet.net> Hi libXdmcp and libXdmcpdevel are now available from testing (version 1.0.2) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From bonivart at opencsw.org Mon Jun 8 11:02:10 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 8 Jun 2009 11:02:10 +0200 Subject: [csw-maintainers] Fwd: [csw-users] I don't know what pkg or pkgs are broken.... In-Reply-To: <200906080849.n588nXJn004314@dfki.uni-kl.de> References: <4A243C2F.5070307@ernieball.com> <200906080849.n588nXJn004314@dfki.uni-kl.de> Message-ID: <625385e30906080202q5c9fda76p89c28dd62256dd04@mail.gmail.com> This seems like a simple init script bug..? William? ---------- Forwarded message ---------- From: Nicolai Schwindt Date: Mon, Jun 8, 2009 at 10:49 AM Subject: Re: [csw-users] I don't know what pkg or pkgs are broken.... To: questions and discussions > I think I recall the same error on Friday: > /var/opt/csw/svc/method/svc-cswdbus: smf_is_nonglobalzone: not found I filed that as bug report in mantis aboput a month ago. If I had not been on vacation I could have saved you some time .) Nicolai _______________________________________________ users mailing list users at lists.opencsw.org https://lists.opencsw.org/mailman/listinfo/users -- /peter From dam at opencsw.org Mon Jun 8 11:16:01 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 8 Jun 2009 11:16:01 +0200 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> Message-ID: <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> Hi, this bug just has been submitted: Anfang der weitergeleiteten E-Mail: > The following issue has been SUBMITTED. ... > ====================================================================== > Summary: libintl > Description: > After upgrading a bunch of CSW-packages binaries linked to libintl > starts > to crash due to a nonexistant symbol > > fatal: relocation error: file /opt/csw/lib/libintl.so.8: symbol > libiconv_set_relocation_prefix: referenced symbol not found > ====================================================================== This looks pretty bad. However, most of the apps still seems to function... Best regards -- Dago From maciej at opencsw.org Mon Jun 8 16:40:25 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Mon, 8 Jun 2009 15:40:25 +0100 Subject: [csw-maintainers] A backed up file in $FILEDIR triggers the 'uncommitted' warning Message-ID: mGAR currently adds 'UNCOMMITTED' when a package is built from sources which are not committed to the repository. When building packages, I often issue 'gmake clean' to make sure that the package is being built from scratch. This deletes everything from downloads/ - all source files need to be downloaded again. To avoid pulling sources multiple times, I copy files from downloads/ to files/ - 'gmake clean' doesn't remove them. In this way, I can rebuild packages from scratch without re-downloading everything. However, an uncommitted file in $FILEDIR causes mGAR to think that these files (tarballs, usually) are some uncommitted material. Perhaps the rule for stuff in files/ could be: 1. If there are changes in the checksums file, it's uncommitted. 2. If there are no changes in the checksums file, and the actual checksums in files/ don't match, it's uncommitted 3. If there are no changes in the checksums file, and the actual checksums in files/ do match, it is not uncommitted, even though some files aren't added to the repository. ...or perhaps there is a way of cleaning the build directory without deleting downloads? Maciej From bwalton at opencsw.org Mon Jun 8 16:49:54 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 08 Jun 2009 10:49:54 -0400 Subject: [csw-maintainers] A backed up file in $FILEDIR triggers the 'uncommitted' warning In-Reply-To: References: Message-ID: <1244472436-sup-6646@ntdws12.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizinski's message of Mon Jun 08 10:40:25 -0400 2009: > mGAR currently adds 'UNCOMMITTED' when a package is built from sources > which are not committed to the repository. When building packages, I > often issue 'gmake clean' to make sure that the package is being built > from scratch. This deletes everything from downloads/ - all source gmake garchive That will stash downloads in your configured source directory. The convention on the buildfarm is that /home/src is used. > files need to be downloaded again. To avoid pulling sources multiple > times, I copy files from downloads/ to files/ - 'gmake clean' doesn't > remove them. In this way, I can rebuild packages from scratch without > re-downloading everything. However, an uncommitted file in $FILEDIR > causes mGAR to think that these files (tarballs, usually) are some > uncommitted material. If you rely on archive, this mostly goes away...everything else in the gar directory should be of the type that you don't mind committing. Does that help? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Mon Jun 8 17:07:15 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 8 Jun 2009 17:07:15 +0200 Subject: [csw-maintainers] New in testing: BerkeleyDB 4.7.25 Message-ID: Hi, I just copied bdb 4.7.25 in testing/: berkeleydb-4.7.25,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz berkeleydb_devel-4.7.25,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz berkeleydb-4.7.25,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz berkeleydb_devel-4.7.25,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz berkeleydb_doc-4.7.25,REV=2009.06.08-SunOS5.8-all-CSW.pkg.gz It has a different layout than the existing bdb4* packages by putting the libs directly in /opt/csw/lib. The last package that did that was bdb 3.3.11 (CSWbdb3), so the current package is incompatible with that (not flagged yet in depend). There are still some packages left which rely on bdb3: - CSWapache - CSWcfengine - CSWwebalizer - CSWdsniff - CSWruby - CSWkdesdk - CSWkdevelop - CSWpostfix I guess the new bdb package should include the old libs for compatibility, CSWbdb3 should then be empty and depend on the new package. Best regards -- Dago From dam at opencsw.org Mon Jun 8 17:11:38 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 8 Jun 2009 17:11:38 +0200 Subject: [csw-maintainers] A backed up file in $FILEDIR triggers the 'uncommitted' warning In-Reply-To: References: Message-ID: <1DE0AB09-5A42-45FC-8858-9F42EC653F7F@opencsw.org> Hi Maciej, Am 08.06.2009 um 16:40 schrieb Maciej (Matchek) Blizinski: > Perhaps the rule for stuff in files/ could be: > > 1. If there are changes in the checksums file, it's uncommitted. > > 2. If there are no changes in the checksums file, and the actual > checksums in files/ don't match, it's uncommitted > > 3. If there are no changes in the checksums file, and the actual > checksums in files/ do match, it is not uncommitted, even though some > files aren't added to the repository. Well, the current rule is "everything which is neither committed not ignored through svn:ignore triggers UNCOMMITTED". Apart from that Ben describes the solution for you problem :-) Best regards -- Dago From ihsan at opencsw.org Mon Jun 8 22:27:12 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 08 Jun 2009 22:27:12 +0200 Subject: [csw-maintainers] test Message-ID: <4A2D7420.8050804@opencsw.org> please ignore -- ihsan at dogan.ch http://blog.dogan.ch/ From hson at opencsw.org Mon Jun 8 22:37:38 2009 From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=) Date: Mon, 08 Jun 2009 22:37:38 +0200 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl In-Reply-To: <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> Message-ID: <4A2D7692.3060902@opencsw.org> Dagobert Michelsen wrote: > Hi, > > this bug just has been submitted: > > > > Anfang der weitergeleiteten E-Mail: >> The following issue has been SUBMITTED. > ... >> ====================================================================== >> Summary: libintl >> Description: >> After upgrading a bunch of CSW-packages binaries linked to libintl starts >> to crash due to a nonexistant symbol >> >> fatal: relocation error: file /opt/csw/lib/libintl.so.8: symbol >> libiconv_set_relocation_prefix: referenced symbol not found >> ====================================================================== > > This looks pretty bad. However, most of the apps still seems to function... > I'm the one who filed the bug report and the problem isn't that the apps doesn't start, but that they crashes after a while in some specific places (within a function call to libexif which is linked to libintl and libiconv). However, a recompile of the app seem to fix the problem. But the question is if there is some option when building libintl whether to include libiconv_set_relocation_prefix or not... (Btw, I'm updating the bug report with this info) From maciej at opencsw.org Mon Jun 8 22:41:14 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Mon, 8 Jun 2009 21:41:14 +0100 Subject: [csw-maintainers] A backed up file in $FILEDIR triggers the 'uncommitted' warning In-Reply-To: <1244472436-sup-6646@ntdws12.chass.utoronto.ca> References: <1244472436-sup-6646@ntdws12.chass.utoronto.ca> Message-ID: On Mon, Jun 8, 2009 at 3:49 PM, Ben Walton wrote: > gmake garchive > > That will stash downloads in your configured source directory. The > convention on the buildfarm is that /home/src is used. I see, cool. Will use that. I was unaware of the gar archive concept. Thanks! From phil at bolthole.com Mon Jun 8 22:48:48 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 8 Jun 2009 13:48:48 -0700 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl In-Reply-To: <4A2D7692.3060902@opencsw.org> References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> <4A2D7692.3060902@opencsw.org> Message-ID: <20090608204848.GA35058@bolthole.com> On Mon, Jun 08, 2009 at 10:37:38PM +0200, Roger H?kansson wrote: > But the question is if there is some option when building libintl whether > to include libiconv_set_relocation_prefix or not... > yes there is. buried in lib/Makefile appears -Dset_relocation_prefix=libiconv_set_relocation_prefix \ From phil at bolthole.com Mon Jun 8 22:51:20 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 8 Jun 2009 13:51:20 -0700 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl In-Reply-To: <4A2D7692.3060902@opencsw.org> References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> <4A2D7692.3060902@opencsw.org> Message-ID: <20090608205120.GB35058@bolthole.com> PS: i should mention that while i am currently the last person to provide the libiconv package, it is essentially orphaned from me. multiple people have "offered" to take it over, and even done some amount of work (in gar) about it. However, no-one has actually DELIVERED a new package. For many weeks now. If you care strongly about this issue, you have your choice of waiting for other people to finish it, or submitting a new package yourself. The first person who submits a fully working package, will become the new maintainer. From harpchad at opencsw.org Mon Jun 8 22:56:18 2009 From: harpchad at opencsw.org (Chad Harp) Date: Mon, 08 Jun 2009 15:56:18 -0500 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl In-Reply-To: <4A2D7692.3060902@opencsw.org> References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> <4A2D7692.3060902@opencsw.org> Message-ID: <4A2D7AF2.3050105@opencsw.org> Which version of libintl are the failing applications linked against? There are currently three active shared libraries, only libintl.so.8 provides libiconv_set_relocation_prefix: $ nm libintl.so.2 | grep libiconv_set_relocation_prefix $ nm libintl.so.3 | grep libiconv_set_relocation_prefix $ nm libintl.so.8 | grep libiconv_set_relocation_prefix [186] | 0| 0|FUNC |GLOB |0 |UNDEF |libiconv_set_relocation_prefix Can you provide a specific test case so I can replicate the problem? (Application, use case, etc) Thanks, Chad From dam at opencsw.org Mon Jun 8 23:07:57 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 8 Jun 2009 23:07:57 +0200 Subject: [csw-maintainers] New in testing: libidl 0.8.13 Message-ID: Hi, new in testing: libidl-0.8.13,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz libidl-0.8.13,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz The following packages rely on it: CSWmozilla CSWorbit2 CSWpyorbit CSWthunderbird CSWfirefox CSWseamonkey CSWsunbird I need it as dependency to orbit2. Please see if you can try it with one of the apps so I can release it and go ahead with the gnome stuff. Best regards -- Dago From dam at opencsw.org Mon Jun 8 23:30:21 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 8 Jun 2009 23:30:21 +0200 Subject: [csw-maintainers] New in testing: libxrender Message-ID: <39724A18-D4F3-46D0-842D-E398B9621EC3@opencsw.org> Hi, a new libxrender is ready in testing/: libxrender-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz libxrender-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz renderdev-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz renderdev-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz Please give it a try, it is a prerequisite on the long way to gnome. Best regards -- Dago From phil at bolthole.com Mon Jun 8 23:43:01 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 8 Jun 2009 14:43:01 -0700 Subject: [csw-maintainers] New in testing: libxrender In-Reply-To: <39724A18-D4F3-46D0-842D-E398B9621EC3@opencsw.org> References: <39724A18-D4F3-46D0-842D-E398B9621EC3@opencsw.org> Message-ID: <20090608214301.GB94732@bolthole.com> On Mon, Jun 08, 2009 at 11:30:21PM +0200, Dagobert Michelsen wrote: > Hi, > > a new libxrender is ready in testing/: > > libxrender-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz > libxrender-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz > renderdev-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz > renderdev-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz > > Please give it a try, it is a prerequisite on the long way to gnome. > i think we should rename renderdev to libxrender_devel. since it hsould have no dependancies, the rename is not as much an issue as in other cases. From car at opencsw.org Tue Jun 9 07:55:27 2009 From: car at opencsw.org (Chris Reece) Date: Tue, 9 Jun 2009 17:55:27 +1200 (NZST) Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <54487.194.246.122.22.1244133574.squirrel@ssl.skayser.de> References: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> <20090604160453.GB2284@bolthole.com> <54487.194.246.122.22.1244133574.squirrel@ssl.skayser.de> Message-ID: <41354.131.203.105.13.1244526927.squirrel@webmail.jessies.org> On Fri, June 5, 2009 4:39 am, Sebastian Kayser wrote: > Philip Brown wrote: >> On Thu, Jun 04, 2009 at 06:49:09PM +1200, Chris Reece wrote: >> >>>> * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to >>>> >>> have >>>> it in there, just for the sake of standardization. >>> >>>> /opt/csw/bin# dump -Lv glocate | grep R*PATH >>>> [5] RUNPATH >>>> >>> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib >>> >>>> [6] RPATH >>>> >>> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib >>> >> >> having both RUNPATH and RPATH in there is normal, for some reason. dont >> worry about it. > > IIRC RPATH had been used in the past, but was considered evil as it had > priority over LD_LIBRARY_PATH and you could not override it. That's why RUNPATH > had been introduced. > > - RPATH (deprecated, not used) > - LD_LIBRARY_PATH > - RUNPATH > - crle settings > > > I would suppose RPATH is still in there for compatibility purposes. I think we may be talking at crossed purposes. I was under the impression that you were commenting on the duplication of "/opt/csw/lib". I was saying that this appears to be down to my following the documented OpenCSW standards. I may be wrong. No matter, there are new pacakges: /home/testing/findutils-4.4.2,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz /home/testing/findutils-4.4.2,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz Which are both more up to date (4.4.2 was released in the interim) and should address the points raised. Cheers, Chris. From hson at opencsw.org Tue Jun 9 10:57:53 2009 From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=) Date: Tue, 09 Jun 2009 10:57:53 +0200 Subject: [csw-maintainers] Fwd: [bug-notifications] [ggettextrt 0003706]: libintl In-Reply-To: <4A2D7AF2.3050105@opencsw.org> References: <272137b3416fc9ba157b82e1885e1bb2@opencsw.org> <034EEED8-6EB4-4C80-A2E3-7602BAF0CF70@opencsw.org> <4A2D7692.3060902@opencsw.org> <4A2D7AF2.3050105@opencsw.org> Message-ID: <4A2E2411.5080102@opencsw.org> Chad Harp wrote: > Which version of libintl are the failing applications linked against? My applications are linked to libintl.so.8 > There are currently three active shared libraries, only libintl.so.8 > provides libiconv_set_relocation_prefix: > > $ nm libintl.so.2 | grep libiconv_set_relocation_prefix > $ nm libintl.so.3 | grep libiconv_set_relocation_prefix > $ nm libintl.so.8 | grep libiconv_set_relocation_prefix > [186] | 0| 0|FUNC |GLOB |0 |UNDEF > |libiconv_set_relocation_prefix > Well, libintl.so.8 doesn't provide libiconv_set_relocation_prefix but have it as a undefined symbol, i.e to be provided by some other library. > Can you provide a specific test case so I can replicate the problem? > (Application, use case, etc) Unfortunately my applications needs a Oracle database and a bunch of data in it to get to the part of the code where there is a problem. From skayser at opencsw.org Tue Jun 9 14:31:15 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 09 Jun 2009 14:31:15 +0200 Subject: [csw-maintainers] mantis config change? In-Reply-To: <49B5ACD5.6010502@opencsw.org> References: <1236642753-sup-662@ntdws12.chass.utoronto.ca> <49B5ACD5.6010502@opencsw.org> Message-ID: <4A2E5613.8050408@opencsw.org> Mike Watters wrote: > Ben Walton wrote: >> Is there an option in Mantis to have only the most recent note and/or >> status change be displayed in the email? I find the output of >> concatenated notes rather nasty (and most times not required). >> >> Would others be in favour of this change if the option exists? > > +1 > I have to agree with you Ben, most of the time you only need the last > comment. and since the email contains a direct link to the mantis bug, > the entire chain IMO is not needed. It has been a while since this was requested and i am happy to say that it is now implemented. It is actually a user-specific setting (My account -> Preferences -> E-mail Notes Limit) which wasn't used by anyone until now. I have now set this to 1 for all existing users and adjusted the configuration so that new users will also have this set to 1. In case someone wants to go back to the old "all bugnotes in a notification" behavior, just set the value to 0. I have also dropped the bug history from the bottom of the notifications. This way there is no distracting text below the bug note; the bug note (the essential part of the notification) can thus be easily identified. For an example bug note, see http://lists.opencsw.org/pipermail/bug-notifications/2009-June/001324.html Thoughts, ideas? Just let me know. Sebastian From skayser at opencsw.org Tue Jun 9 14:40:14 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 09 Jun 2009 14:40:14 +0200 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <41354.131.203.105.13.1244526927.squirrel@webmail.jessies.org> References: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> <20090604160453.GB2284@bolthole.com> <54487.194.246.122.22.1244133574.squirrel@ssl.skayser.de> <41354.131.203.105.13.1244526927.squirrel@webmail.jessies.org> Message-ID: <4A2E582E.1010608@opencsw.org> Chris Reece wrote: > On Fri, June 5, 2009 4:39 am, Sebastian Kayser wrote: >> Philip Brown wrote: >>> On Thu, Jun 04, 2009 at 06:49:09PM +1200, Chris Reece wrote: >>> >>>>> * The R*PATH contains /opt/csw/lib twice. It doesn't really hurt to >>>>> >>>> have >>>>> it in there, just for the sake of standardization. >>>>> /opt/csw/bin# dump -Lv glocate | grep R*PATH >>>>> [5] RUNPATH >>>>> >>>> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib >>>> >>>>> [6] RPATH >>>>> >>>> /opt/csw/lib/$ISALIST:/opt/csw/lib:/opt/csw/lib >>>> >>> having both RUNPATH and RPATH in there is normal, for some reason. dont >>> worry about it. >> IIRC RPATH had been used in the past, but was considered evil as it had >> priority over LD_LIBRARY_PATH and you could not override it. That's why RUNPATH >> had been introduced. >> >> - RPATH (deprecated, not used) >> - LD_LIBRARY_PATH >> - RUNPATH >> - crle settings >> >> >> I would suppose RPATH is still in there for compatibility purposes. > > I think we may be talking at crossed purposes. I was under the impression > that you were commenting on the duplication of "/opt/csw/lib". I was saying > that this appears to be down to my following the documented OpenCSW standards. > I may be wrong. Sorry Chris for getting you confused here. I was indeed only after the duplicate /opt/csw/lib when commenting on your package (and that is what needs to be taken care of). The RPATH/RUNPATH thingie was just in response to Phil's mail. Sebastian From bwalton at opencsw.org Tue Jun 9 15:05:07 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 09 Jun 2009 09:05:07 -0400 Subject: [csw-maintainers] mantis config change? In-Reply-To: <4A2E5613.8050408@opencsw.org> References: <1236642753-sup-662@ntdws12.chass.utoronto.ca> <49B5ACD5.6010502@opencsw.org> <4A2E5613.8050408@opencsw.org> Message-ID: <1244552681-sup-9540@ntdws12.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Tue Jun 09 08:31:15 -0400 2009: > http://lists.opencsw.org/pipermail/bug-notifications/2009-June/001324.html > > Thoughts, ideas? Just let me know. This is _much_ better! Thanks. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From skayser at opencsw.org Tue Jun 9 16:45:36 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 09 Jun 2009 16:45:36 +0200 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> References: <50719.131.203.105.13.1244098149.squirrel@webmail.jessies.org> Message-ID: <4A2E7590.5040000@opencsw.org> Chris Reece wrote: > On Tue, June 2, 2009 9:05 pm, Sebastian Kayser wrote: >> Could you tell us, what exactly you were missing about the documentation >> so that we can improve on it? Any feedback appreciated (off-list is also fine >> if you prefer to). > > Specifically, no. I recall being daunted by finding three versions of GAR, > none of which appeared to have a full set of GAR newbie documentation. I > found a "Getting started with GAR" document that, depending on which link I > followed to it, from where: was not found; contained a page blank except for > "redirect"; or was there, but was listed under some manner of "this is for an > older GAR -- here be dragons" warning. > > I was in a hurry, may have missed something and, believe me, I know what it's > like to not have enough hours in the day. Documentation's no damned fun. > > Your suggestion of hand-holding to en-GAR findutils sounds like a fine idea, > if you've the time. I'd rather put these things in a format that they'll be > useful, for what that's worth. > > But if I am just being dense, feel free to point me at the docs that you know > tell me everything I need and tell me to RTFM. A couple of options. Option 1 -------- Send me your current build script (or instructions on how to follow the build process you are currently using). I will then GAR-ize and submit it to the repository so you can have a look at it and maybe tweak it a bit. Doing this for others i have seen that this seems to give people a pretty good idea of how GAR works. You can then GAR-ize your remaining packages. Option 2 -------- If you want to learn GAR by yourself, follow the steps below. You might take shortcuts if some points seem obvious to you. 1) Read [1] for setting up your local build enviroment. 2) Check out gperf from the repository which is a known to work, basic build-description. Then build it along with reading the build explanation. [2,3] 3) Do the same for dtracetoolkit to see how different packaging requirements are handled. [4] 4) Peek into the GAR variable reference along with its referenced build descriptions to see how other packages (with varying complexity) are built. [5] Btw. this is roughly the way that the GAR wiki main page [6] is currently structured. Option 3 -------- I will hang out on our IRC channel this evening (#opencsw on freenode.net, starting ~8pm CEST). Come by if i you prefer some Q&A. :) Sebastian [1] http://apps.sf.net/trac/gar/wiki/GarSetup [2] http://apps.sf.net/trac/gar/wiki/GarRepository#Checkingoutonepackage [3] http://apps.sf.net/trac/gar/wiki/ExplainedGperf [4] http://apps.sf.net/trac/gar/wiki/ExplainedDtracetoolkit [5] http://apps.sf.net/trac/gar/wiki/GAR%20Variable%20Reference [6] http://gar.sf.net From dam at opencsw.org Tue Jun 9 17:12:13 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 9 Jun 2009 17:12:13 +0200 Subject: [csw-maintainers] New in testing: libxrender In-Reply-To: <20090608214301.GB94732@bolthole.com> References: <39724A18-D4F3-46D0-842D-E398B9621EC3@opencsw.org> <20090608214301.GB94732@bolthole.com> Message-ID: <4299CDC0-CB4F-4C57-99FC-356BF5C42012@opencsw.org> Hi Phil, Am 08.06.2009 um 23:43 schrieb Philip Brown: > On Mon, Jun 08, 2009 at 11:30:21PM +0200, Dagobert Michelsen wrote: >> a new libxrender is ready in testing/: >> >> libxrender-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz >> libxrender-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz >> renderdev-0.9.4,REV=2009.06.08-SunOS5.8-i386-CSW.pkg.gz >> renderdev-0.9.4,REV=2009.06.08-SunOS5.8-sparc-CSW.pkg.gz >> >> Please give it a try, it is a prerequisite on the long way to gnome. >> > > i think we should rename renderdev to libxrender_devel. > > since it hsould have no dependancies, the rename is not as much an > issue > as in other cases. Sure. New packages in testing/: libxrender-0.9.4,REV=2009.06.09-SunOS5.8-sparc-CSW.pkg.gz libxrender_devel-0.9.4,REV=2009.06.09-SunOS5.8-sparc-CSW.pkg.gz libxrender-0.9.4,REV=2009.06.09-SunOS5.8-i386-CSW.pkg.gz libxrender_devel-0.9.4,REV=2009.06.09-SunOS5.8-i386-CSW.pkg.gz Best regards -- Dago From phil at bolthole.com Tue Jun 9 17:41:39 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 9 Jun 2009 08:41:39 -0700 Subject: [csw-maintainers] Poll on our future X11 library usage Message-ID: <20090609154139.GK35058@bolthole.com> hi folks, a few maintainers have been discussing what to do about X11 libs, going forward. I just shared the following poll to our user list. Please also weigh in. http://doodle.com/73mdviypknk63y7e Here's the full body of the email i sent to the users list, if you are curious, and not on the list: ------------------------------------------------------------ Hello CSW users, This is a very important poll, for the future of X11 support in CSW packages going forward. We are faced with a tough technical decision to make. While the implementation details are technical, the major impact is more user based. So, we are asking you, the users, to tell us what you would prefer! For 6 years now, we have based virtually all our X11 dependant programs, on the X11 libraries supplied by Solaris itself. Unfortunately, the latest versions of gtk, which are required by the latest versions of things like firefox, etc. require newer versions of basic things like libX11 and so on, than older versions of Solaris provide. We are obviously going to provide firefox. So we are also going to provide the required libraries in one way or another. The question then becomes; from this point on, do we have ALL X11 programs use the CSW X11 libraries? Or do we have only the programs that "need" the new stuff, use the new stuff in /opt/csw/X11, whereas others that are less fancy (ie: not gtk based) use the sun libs in /usr/openwin/lib So far, there is some inclination to no longer use the Sun libs. This would mean that, if you still use some sun graphical programs, you would potentially have a bit more redundancy in the shared library department. The "good news", is that it should only amount to around 4 megabytes of extra ram used up this way, vs using sun X11 libraries. That's total extra; it's a one-time hit, not a per-process cost. Please let us know what you, our users, would prefer. We've set up a poll for this issue, at http://doodle.com/73mdviypknk63y7e From bonivart at opencsw.org Tue Jun 9 17:53:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 9 Jun 2009 17:53:14 +0200 Subject: [csw-maintainers] Poll on our future X11 library usage In-Reply-To: <20090609154139.GK35058@bolthole.com> References: <20090609154139.GK35058@bolthole.com> Message-ID: <625385e30906090853s4f5227c2s9d0e7f656e9eecbb@mail.gmail.com> On Tue, Jun 9, 2009 at 5:41 PM, Philip Brown wrote: > Unfortunately, the latest versions of gtk, which are required by the latest > versions of things like firefox, etc. require newer versions of basic > things like libX11 and so on, than older versions of Solaris provide. Is Solaris 9 too old to build, e.g., Firefox on? -- /peter From william at wbonnet.net Tue Jun 9 21:49:26 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 09 Jun 2009 21:49:26 +0200 Subject: [csw-maintainers] Poll on our future X11 library usage In-Reply-To: <625385e30906090853s4f5227c2s9d0e7f656e9eecbb@mail.gmail.com> References: <20090609154139.GK35058@bolthole.com> <625385e30906090853s4f5227c2s9d0e7f656e9eecbb@mail.gmail.com> Message-ID: <4A2EBCC6.8090303@wbonnet.net> Hi Peter > On Tue, Jun 9, 2009 at 5:41 PM, Philip Brown wrote: > >> Unfortunately, the latest versions of gtk, which are required by the latest >> versions of things like firefox, etc. require newer versions of basic >> things like libX11 and so on, than older versions of Solaris provide. >> > > Is Solaris 9 too old to build, e.g., Firefox on? > I'm not really sure, i haven't tried since i focussed on Solaris 8. I think it is too old, but i have to double check if it you need an answer to this. cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From dam at opencsw.org Wed Jun 10 18:48:58 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 10 Jun 2009 18:48:58 +0200 Subject: [csw-maintainers] Alignment-problem on libmhash References: <20090610162618.5AEFA5C4D@mail.opencsw.org> Message-ID: Hi, I have an alignment problem with libmhash on sparcv9. It looks like this: Checking fragmentation capabilities of HAVAL256: FAIL: frag_test build8s% cd work/build-isa-sparcv9/mhash-0.9.9.9/src/ build8s% mdb .libs/ driver frag_test hmac_test keygen_test rest_test build8s% mdb .libs/frag_test core mdb: warning: core file is from SunOS 5.10 Generic_138888-07; shared text mappings may not match installed libraries Loading modules: [ ] > ::stack libmhash.so.2`mutils_word32nswap+0xb8(100104990, 20, 1f, 6c24cc00, 1, 20754a1616aa27a) libmhash.so.2`havalTransform3+0xc(1001047d4, ffffffff7fffefcd, 100104880, 143367, a19b3906, 1f3213955f32139) libmhash.so.2`havalUpdate+0x130(1001047d0, ffffffff7fffefcc, 81, ffffffffffffffff, 1001047f4, 408) libmhash.so.2`mhash+0x14(100101fd0, ffffffff7fffefcc, 81, fffffffffffffffd, 0, 7ffeff021eb8) frag_test+0x98(100, 81, 1, 958, ffffffff7fffefcc, 80) main+0x8c(1c, 7ffeff1329ce, 1, 100101620, 100001038, 100101588) _start+0x17c(0, ffffffff7ffff238, 0, 1, 7ffeffe0d6cc, 7ffefff10000) > ::dis mutils_word32nswap libmhash.so.2`mutils_word32nswap: save %sp, -0xb0, %sp libmhash.so.2`mutils_word32nswap+4: btst 0xff, %i2 libmhash.so.2`mutils_word32nswap+8: bne,pn %icc,+0x24 libmhash.so.2`mutils_word32nswap+0xc: mov %i0, %l4 libmhash.so.2`mutils_word32nswap+0x10: sll %i1, 2, %l6 libmhash.so.2`mutils_word32nswap+0x14: call +0x12d494 <0x7ffeff131aa0> libmhash.so.2`mutils_word32nswap+0x18: srl %l6, 0, %o0 libmhash.so.2`mutils_word32nswap+0x1c: brnz,pn %o0, +0x14 libmhash.so.2`mutils_word32nswap+0x20: mov %o0, %i0 libmhash.so.2`mutils_word32nswap+0x24: ret libmhash.so.2`mutils_word32nswap+0x28: restore %g0, 0, %o0 libmhash.so.2`mutils_word32nswap+0x2c: mov %l4, %i0 libmhash.so.2`mutils_word32nswap+0x30: cmp %i1, 0 libmhash.so.2`mutils_word32nswap+0x34: bleu,pn %icc,+0x264 libmhash.so.2`mutils_word32nswap+0x38: cmp %i1, 4 libmhash.so.2`mutils_word32nswap+0x3c: bl,pn %icc,+0x200 libmhash.so.2`mutils_word32nswap+0x40: mov %i0, %l6 libmhash.so.2`mutils_word32nswap+0x44: sra %i1, 0, %o0 libmhash.so.2`mutils_word32nswap+0x48: sub %i0, %l4, %l7 libmhash.so.2`mutils_word32nswap+0x4c: srax %l7, 0x3f, %g4 libmhash.so.2`mutils_word32nswap+0x50: sethi %hi(0x7ffffc00), %l1 libmhash.so.2`mutils_word32nswap+0x54: srl %i1, 0, %l2 libmhash.so.2`mutils_word32nswap+0x58: xor %l7, %g4, %g5 libmhash.so.2`mutils_word32nswap+0x5c: sllx %o0, 2, %g3 libmhash.so.2`mutils_word32nswap+0x60: add %l1, 0x3fe, %l0 libmhash.so.2`mutils_word32nswap+0x64: sub %g5, %g4, %g2 libmhash.so.2`mutils_word32nswap+0x68: sub %l2, %l0, %o2 libmhash.so.2`mutils_word32nswap+0x6c: srlx %o2, 0x3f, %o1 libmhash.so.2`mutils_word32nswap+0x70: cmp %g2, %g3 libmhash.so.2`mutils_word32nswap+0x74: clr %g1 libmhash.so.2`mutils_word32nswap+0x78: and %o1, 1, %i4 libmhash.so.2`mutils_word32nswap+0x7c: movge %xcc,1, %g1 libmhash.so.2`mutils_word32nswap+0x80: btst %g1, %i4 libmhash.so.2`mutils_word32nswap+0x84: be,a,pn %icc,+0x1bc libmhash.so.2`mutils_word32nswap+0x88: clr %l2 libmhash.so.2`mutils_word32nswap+0x8c: sethi %hi(0xfc00), %l5 libmhash.so.2`mutils_word32nswap+0x90: add %i1, 0, %g4 libmhash.so.2`mutils_word32nswap+0x94: add %l5, 0x300, %l3 libmhash.so.2`mutils_word32nswap+0x98: add %i1, -1, %i2 libmhash.so.2`mutils_word32nswap+0x9c: clr %l1 libmhash.so.2`mutils_word32nswap+0xa0: sethi %hi(0xff000000), %l2 libmhash.so.2`mutils_word32nswap+0xa4: sethi %hi(0xff0000), %l5 libmhash.so.2`mutils_word32nswap+0xa8: cmp %g4, 5 libmhash.so.2`mutils_word32nswap+0xac: bl,pn %icc,+0x13c libmhash.so.2`mutils_word32nswap+0xb0: sub %i2, 3, %o7 libmhash.so.2`mutils_word32nswap+0xb4: add %l1, 1, %l1 libmhash.so.2`mutils_word32nswap+0xb8: ld [%l4], %g5 libmhash.so.2`mutils_word32nswap+0xbc: add %l4, 4, %l4 libmhash.so.2`mutils_word32nswap+0xc0: and %g5, %l5, %o1 > ::regs %g0 = 0x0000000000000000 %l0 = 0x000000007ffffffe %g1 = 0x0000000000000001 %l1 = 0x0000000000000001 %g2 = 0x00000001801059c3 %l2 = 0x00000000ff000000 %g3 = 0x0000000000000080 %l3 = 0x000000000000ff00 %g4 = 0x0000000000000020 %l4 = 0xffffffff7fffefcd %g5 = 0x00000001801059c3 %l5 = 0x0000000000ff0000 %g6 = 0x0000000000000000 %l6 = 0x0000000100104990 %g7 = 0x0000000000000000 %l7 = 0x00000001801059c3 %o0 = 0x0000000000000020 %i0 = 0x0000000100104990 %o1 = 0x0000000000000001 %i1 = 0x0000000000000020 %o2 = 0xffffffff80000022 %i2 = 0x000000000000001f %o3 = 0x000000007f7e7e7e %i3 = 0x000000006c24cc00 %o4 = 0x00000000007e0000 %i4 = 0x0000000000000001 %o5 = 0x0000000100104990 %i5 = 0x020754a1616aa27a %o6 = 0xffffffff7fffe151 %i6 = 0xffffffff7fffe201 %o7 = 0x000000000000001c %i7 = 0x00007ffeff0190ec libmhash.so.2`havalTransform3+0xc %ccr = 0x00 xcc=nzvc icc=nzvc %y = 0x0000000000000000 %pc = 0x00007ffeff0046b0 libmhash.so.2`mutils_word32nswap+0xb8 %npc = 0x00007ffeff0046b4 libmhash.so.2`mutils_word32nswap+0xbc %sp = 0xffffffff7fffe151 %fp = 0xffffffff7fffe201 %asi = 0x82 %fprs = 0x07 So, the problem is that 0xffffffff7fffefcd is not 32 bit aligned, but I can't see it in the source why this should be the case. Maybe I'm just too blind today and someone other can have a look? Everything has been committed to pkg/libmhash. Best regards -- Dago From rmacduff at opencsw.org Wed Jun 10 19:03:43 2009 From: rmacduff at opencsw.org (Ross Macduff) Date: Wed, 10 Jun 2009 13:03:43 -0400 Subject: [csw-maintainers] gsed 4.2 in testing Message-ID: <1244653354-sup-1474@frog.chass.utoronto.ca> Hello, gsed 4.2 is in testing: gsed-4.2,REV=2009.06.10-SunOS5.8-i386-CSW.pkg.gz gsed-4.2,REV=2009.06.10-SunOS5.8-sparc-CSW.pkg.gz Best, Ross From dam at opencsw.org Wed Jun 10 22:29:03 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 10 Jun 2009 22:29:03 +0200 Subject: [csw-maintainers] gsed 4.2 in testing In-Reply-To: <1244653354-sup-1474@frog.chass.utoronto.ca> References: <1244653354-sup-1474@frog.chass.utoronto.ca> Message-ID: <87834F2F-1E74-437F-BEC0-2186A0EC2653@opencsw.org> Hi Ross, Am 10.06.2009 um 19:03 schrieb Ross Macduff: > gsed 4.2 is in testing: > > gsed-4.2,REV=2009.06.10-SunOS5.8-i386-CSW.pkg.gz > gsed-4.2,REV=2009.06.10-SunOS5.8-sparc-CSW.pkg.gz Looks good. Thanks for the quick response! Best regards -- Dago From mwatters at opencsw.org Thu Jun 11 06:59:42 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 10 Jun 2009 23:59:42 -0500 Subject: [csw-maintainers] another bug in upstream. Message-ID: <4A308F3E.9020306@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I received an upgrade notice for GD ( pasted below ) I am the GD maintainer, however all the links to the download location are for the pm_gd to which I am not the maintainer. ( I believe it was Alex who is retired.) Further investigation showed me that I do in fact have the latest "stable" release of GD packaged, but the perl module is what is in need of upgrade. GD-2.43.tar.gz is the filename of pm_gd gd-2.0.35.tar.gz is the filename of gd note the regex for upstream watch should be the same. I have no solution off hand to this bug, but I will take a look when I get a few spare cycles. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Hello dear GD maintainer, The upstream notification job has detected the availability of new files for GD. The following upstream file(s): GD-2.43.tar.gz is/are available at the following url(s): http://search.cpan.org/CPAN/authors/id/L/LD/LDS/ ftp://ftp.nrc.ca/pub/CPAN/authors/id/L/LD/LDS/ ftp://ftp.nas.nasa.gov/pub/perl/CPAN/authors/id/L/LD/LDS/ http://mirrors.ibiblio.org/pub/mirrors/CPAN/authors/id/L/LD/LDS/ ftp://cpan.pair.com/pub/CPAN/authors/id/L/LD/LDS/ http://mirrors.kernel.org/cpan/authors/id/L/LD/LDS/ Please consider updating your package. - -- Kindest regards upstream notification job - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkowjz4ACgkQLrhmsXMSLxcZzQCgjY0dRLLSaGvrVnFFd13FG7kH 458AoJkUUd5vdjv9sis7uhxniiNbNZRY =fC8f -----END PGP SIGNATURE----- From bonivart at opencsw.org Thu Jun 11 15:13:43 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 11 Jun 2009 15:13:43 +0200 Subject: [csw-maintainers] /testing clamav 0.95.2 Message-ID: <625385e30906110613v59a347f1xd24f33ae77e2c406@mail.gmail.com> A new version of Clam AntiVirus is in testing. http://mirror.opencsw.org/testing/clamav-0.95.2,REV=2009.06.11-SunOS5.8-sparc-CSW.pkg.gz http://mirror.opencsw.org/testing/clamav-0.95.2,REV=2009.06.11-SunOS5.8-i386-CSW.pkg.gz http://mirror.opencsw.org/testing/libclamav-0.95.2,REV=2009.06.11-SunOS5.8-sparc-CSW.pkg.gz http://mirror.opencsw.org/testing/libclamav-0.95.2,REV=2009.06.11-SunOS5.8-i386-CSW.pkg.gz -- /peter From skayser at opencsw.org Fri Jun 12 01:55:49 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 12 Jun 2009 01:55:49 +0200 Subject: [csw-maintainers] Bugtracker (Mantis) patched for performance, feedback welcome Message-ID: <4A319985.8040904@opencsw.org> Hi everyone, i have patched a Mantis test installation today to improve the long page load times we have been seeing. A couple of features that we don't use as of yet (sub-projects, private projects, product versions) have been de-facto deactivated as a result of this. The changes are targeted at data display only and should be non-disruptive. The patched version is available at [1], if some of you guys could have a look and check whether something obvious feels wrong to you i would appreciate it. Heads up!: it works with our live bug data and sends out bug update notifications, so don't just close or delete random bugs. I will be on vacation until Sunday. Pending some feedback and finishing touches i aim to patch the productive installation early next week. If you spot pages that still feel overly sluggish to you, let me know. Note that this is a first attempt at improving performance. In case we might need one of the now missing features, we can always revisit the changes. Sebastian [1] http://opencsw.org/mantis-test/ From bwalton at opencsw.org Fri Jun 12 02:04:48 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 11 Jun 2009 20:04:48 -0400 Subject: [csw-maintainers] Bugtracker (Mantis) patched for performance, feedback welcome In-Reply-To: <4A319985.8040904@opencsw.org> References: <4A319985.8040904@opencsw.org> Message-ID: <1244765051-sup-8822@ntdws12.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Thu Jun 11 19:55:49 -0400 2009: > i have patched a Mantis test installation today to improve the long page > load times we have been seeing. A couple of features that we don't use > as of yet (sub-projects, private projects, product versions) have been > de-facto deactivated as a result of this. The changes are targeted at > data display only and should be non-disruptive. +1. It looks good to me and it's a _lot_ faster. Nice work! :) Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From ihsan at dogan.ch Fri Jun 12 16:38:16 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Fri, 12 Jun 2009 16:38:16 +0200 Subject: [csw-maintainers] Bugtracker (Mantis) patched for performance, feedback welcome In-Reply-To: <1244765051-sup-8822@ntdws12.chass.utoronto.ca> References: <4A319985.8040904@opencsw.org> <1244765051-sup-8822@ntdws12.chass.utoronto.ca> Message-ID: <4A326858.4010405@dogan.ch> Ben Walton wrote: >> i have patched a Mantis test installation today to improve the long page >> load times we have been seeing. A couple of features that we don't use >> as of yet (sub-projects, private projects, product versions) have been >> de-facto deactivated as a result of this. The changes are targeted at >> data display only and should be non-disruptive. > > +1. It looks good to me and it's a _lot_ faster. Nice work! :) Wow, it's a rocket now. Thanks Sebastian. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at dogan.ch Fri Jun 12 16:44:11 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Fri, 12 Jun 2009 16:44:11 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp 2009 In-Reply-To: <4A11B3AF.7090709@opencsw.org> References: <4A11B3AF.7090709@opencsw.org> Message-ID: <4A3269BB.4090604@dogan.ch> Ihsan Dogan wrote: > The OpenCSW Summer Camp 2009 will take place in Oslo/Norway. To figure > out when it should take place, I've create a Doodle poll. Please fill > out the link bellow: > > http://www.doodle.com/6zr9kmdr8p9vv69p The summer camp will take place on the weekend of 22/23 August 2009 in Olso/Norway. Thanks to all, who participated with the poll. @Trygve: Could you please arrange the location and provide a list of recommended hotels? Next task: - Define the topics... Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Fri Jun 12 17:41:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 12 Jun 2009 11:41:47 -0400 Subject: [csw-maintainers] exim maintainer Message-ID: <1244821262-sup-9225@ntdws12.chass.utoronto.ca> Hi Markus, Are you still out there? If so, we've got a bug to file against the exim package. If not, we may consider taking it over. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From rmacduff at opencsw.org Fri Jun 12 18:10:08 2009 From: rmacduff at opencsw.org (Ross Macduff) Date: Fri, 12 Jun 2009 12:10:08 -0400 Subject: [csw-maintainers] Exim and SMF Message-ID: <1244821559-sup-5458@frog.chass.utoronto.ca> Hello, During a recent install of Exim on a Solaris 10 box it was noticed that the postinstall script doesn't register it with SMF. It looks like this is because we have autoenable_deamons set to 0 in our csw.conf. It would be nice to have it register the service, yet leave it disabled. Is this something that could be done? Would making use of cswclassutils solve this problem? Is this worthy of a bug report? Thanks, Ross From phil at bolthole.com Fri Jun 12 18:20:44 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 12 Jun 2009 09:20:44 -0700 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <1244821559-sup-5458@frog.chass.utoronto.ca> References: <1244821559-sup-5458@frog.chass.utoronto.ca> Message-ID: <20090612162044.GB93732@bolthole.com> On Fri, Jun 12, 2009 at 12:10:08PM -0400, Ross Macduff wrote: > Hello, > > During a recent install of Exim on a Solaris 10 box it was noticed that > the postinstall script doesn't register it with SMF. It looks like this > is because we have autoenable_deamons set to 0 in our csw.conf. It > would be nice to have it register the service, yet leave it disabled. > Ummmm.. out of curiosity.. WHY would that be nice? From rmacduff at opencsw.org Fri Jun 12 18:32:08 2009 From: rmacduff at opencsw.org (Ross Macduff) Date: Fri, 12 Jun 2009 12:32:08 -0400 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <20090612162044.GB93732@bolthole.com> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> Message-ID: <1244823762-sup-4624@frog.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 12 12:20:44 -0400 2009: > On Fri, Jun 12, 2009 at 12:10:08PM -0400, Ross Macduff wrote: > > Hello, > > > > During a recent install of Exim on a Solaris 10 box it was noticed that > > the postinstall script doesn't register it with SMF. It looks like this > > is because we have autoenable_deamons set to 0 in our csw.conf. It > > would be nice to have it register the service, yet leave it disabled. > > > > Ummmm.. out of curiosity.. WHY would that be nice? > Perhaps it isn't desirable to have it start up right after installing it, yet you don't want to go through the steps necessary to register it with SMF (I know it's only one step...). That way, it is only a matter of enabling it when it's ready to go. It seems that cswclassutils has this type of behaviour in mind based on what I read here: http://wiki.opencsw.org/cswclassutils-package#toc3 So if use_smf isn't set to 'no' and autoenable_daemons is set to 'no', then it won't start but is ready to be enabled. Best, Ross From trygvis at opencsw.org Fri Jun 12 18:42:08 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Fri, 12 Jun 2009 18:42:08 +0200 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <20090612162044.GB93732@bolthole.com> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> Message-ID: <4A328560.4070805@opencsw.org> Philip Brown wrote: > On Fri, Jun 12, 2009 at 12:10:08PM -0400, Ross Macduff wrote: >> Hello, >> >> During a recent install of Exim on a Solaris 10 box it was noticed that >> the postinstall script doesn't register it with SMF. It looks like this >> is because we have autoenable_deamons set to 0 in our csw.conf. It >> would be nice to have it register the service, yet leave it disabled. >> > > Ummmm.. out of curiosity.. WHY would that be nice? It would be nice to have it registered so that all the user would have to do was to enable it, just like the regular Solaris services. -- Trygve From trygvis at opencsw.org Fri Jun 12 18:43:39 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Fri, 12 Jun 2009 18:43:39 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp 2009 In-Reply-To: <4A3269BB.4090604@dogan.ch> References: <4A11B3AF.7090709@opencsw.org> <4A3269BB.4090604@dogan.ch> Message-ID: <4A3285BB.4000705@opencsw.org> Ihsan Dogan wrote: > Ihsan Dogan wrote: > >> The OpenCSW Summer Camp 2009 will take place in Oslo/Norway. To figure >> out when it should take place, I've create a Doodle poll. Please fill >> out the link bellow: >> >> http://www.doodle.com/6zr9kmdr8p9vv69p > > The summer camp will take place on the weekend of 22/23 August 2009 in > Olso/Norway. Thanks to all, who participated with the poll. > > @Trygve: > Could you please arrange the location and provide a list of recommended > hotels? I'll try to ask around next week. I think I've found a nice place for us to work during the day too. The place is the same that the JUG that I lead use so it should be good enough for us :) -- Trygve From bwalton at opencsw.org Fri Jun 12 18:47:10 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 12 Jun 2009 12:47:10 -0400 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <4A328560.4070805@opencsw.org> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <4A328560.4070805@opencsw.org> Message-ID: <1244825151-sup-6189@ntdws12.chass.utoronto.ca> Excerpts from Trygve Laugst?l's message of Fri Jun 12 12:42:08 -0400 2009: > > Ummmm.. out of curiosity.. WHY would that be nice? > > It would be nice to have it registered so that all the user would have > to do was to enable it, just like the regular Solaris services. Yes, this is our position. If the package is updated, it should be moved to use class utils, which would get this behaviour 'for free.' This was the reason for my ping to Markus earlier today. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From phil at bolthole.com Fri Jun 12 19:11:26 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 12 Jun 2009 10:11:26 -0700 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <1244823762-sup-4624@frog.chass.utoronto.ca> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> Message-ID: <20090612171126.GE95242@bolthole.com> On Fri, Jun 12, 2009 at 12:32:08PM -0400, Ross Macduff wrote: > Excerpts from Philip Brown's message of Fri Jun 12 12:20:44 -0400 2009: > > Ummmm.. out of curiosity.. WHY would that be nice? > > > > Perhaps it isn't desirable to have it start up right after installing > it, yet you don't want to go through the steps necessary to register it > with SMF (I know it's only one step...). That way, it is only a matter > of enabling it when it's ready to go. Ah, got it. So to phrase it more in the general case: When something requires hand-chosen configuration, it can be nice to be able to install a buncha things in batch mode, but actually configure it to start properly, later, when batch install is all done. However, for simple services, its not so much of an issue. Either you want it started, or you dont. From bwalton at opencsw.org Fri Jun 12 19:19:12 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 12 Jun 2009 13:19:12 -0400 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <20090612171126.GE95242@bolthole.com> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> <20090612171126.GE95242@bolthole.com> Message-ID: <1244827079-sup-1294@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 12 13:11:26 -0400 2009: > However, for simple services, its not so much of an issue. > Either you want it started, or you dont. And in either case, it should be one command away...svcadm or /etc/init.d/. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 12 19:22:16 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 12 Jun 2009 10:22:16 -0700 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <1244827079-sup-1294@ntdws12.chass.utoronto.ca> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> <20090612171126.GE95242@bolthole.com> <1244827079-sup-1294@ntdws12.chass.utoronto.ca> Message-ID: <20090612172216.GH95242@bolthole.com> On Fri, Jun 12, 2009 at 01:19:12PM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Fri Jun 12 13:11:26 -0400 2009: > > However, for simple services, its not so much of an issue. > > Either you want it started, or you dont. > > And in either case, it should be one command away...svcadm or > /etc/init.d/. yeah.... what I think would benefit us, would be if the cswinitsmf class, came with an additional util script, to enable things after the fact. From bwalton at opencsw.org Fri Jun 12 21:19:08 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 12 Jun 2009 15:19:08 -0400 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <20090612172216.GH95242@bolthole.com> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> <20090612171126.GE95242@bolthole.com> <1244827079-sup-1294@ntdws12.chass.utoronto.ca> <20090612172216.GH95242@bolthole.com> Message-ID: <1244834220-sup-7183@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 12 13:22:16 -0400 2009: > yeah.... what I think would benefit us, would be if the cswinitsmf class, > came with an additional util script, to enable things after the fact. It doesn't really need to, unless I'm misunderstanding you. After cswinitsmf is done its job, the service can be started with a simple command after placing appropriate config files in place. As far as I can see, the manifest gets imported and any init scripts installed. The state of the service after that depends on the csw.conf settings. Am I missing your point? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 12 22:24:27 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 12 Jun 2009 13:24:27 -0700 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <1244834220-sup-7183@ntdws12.chass.utoronto.ca> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> <20090612171126.GE95242@bolthole.com> <1244827079-sup-1294@ntdws12.chass.utoronto.ca> <20090612172216.GH95242@bolthole.com> <1244834220-sup-7183@ntdws12.chass.utoronto.ca> Message-ID: <20090612202427.GI95242@bolthole.com> On Fri, Jun 12, 2009 at 03:19:08PM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Fri Jun 12 13:22:16 -0400 2009: > > yeah.... what I think would benefit us, would be if the cswinitsmf class, > > came with an additional util script, to enable things after the fact. > > It doesn't really need to, unless I'm misunderstanding you. After > cswinitsmf is done its job, the service can be started with a simple > command after placing appropriate config files in place. As far as I > can see, the manifest gets imported and any init scripts installed. > The state of the service after that depends on the csw.conf settings. > > Am I missing your point? yes, i think so :-) such a script has multiple uses. One,is like the exim case mentioned. Otheres, are when you nfs-share out /opt/csw. Or, replicate it. Most people on this list wouldnt think of doing that sort of thing. But a lot of large places, DO do that sort of thing. there are a few other reasons too i think, but I'm writing in a hurry :) From bwalton at opencsw.org Fri Jun 12 22:31:20 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 12 Jun 2009 16:31:20 -0400 Subject: [csw-maintainers] Exim and SMF In-Reply-To: <20090612202427.GI95242@bolthole.com> References: <1244821559-sup-5458@frog.chass.utoronto.ca> <20090612162044.GB93732@bolthole.com> <1244823762-sup-4624@frog.chass.utoronto.ca> <20090612171126.GE95242@bolthole.com> <1244827079-sup-1294@ntdws12.chass.utoronto.ca> <20090612172216.GH95242@bolthole.com> <1244834220-sup-7183@ntdws12.chass.utoronto.ca> <20090612202427.GI95242@bolthole.com> Message-ID: <1244838650-sup-8959@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 12 16:24:27 -0400 2009: > yes, i think so :-) It happens! :) > such a script has multiple uses. > One,is like the exim case mentioned. Otheres, are when you nfs-share out > /opt/csw. > Or, replicate it. Ok, that makes sense now. Thanks for clarifying. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From maciej at opencsw.org Sat Jun 13 13:04:40 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Sat, 13 Jun 2009 12:04:40 +0100 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt Message-ID: In order to support sparse zones, paths such as /opt/csw/var/... must become /var/opt/csw/... - otherwise applications in non-global zones won't be able to write there. Paths are being set in the $(DIRPATHS), as a space-separated list. What's the proper way of tweaking it? (say, define sysconfdir as /etc/opt/csw) I guess I could add sysconfdir to NODIRPATHS and then append my own sysconfdir, but I'm not sure if it's the way Dago intended. Maciej From maciej at opencsw.org Sat Jun 13 15:06:59 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Sat, 13 Jun 2009 14:06:59 +0100 Subject: [csw-maintainers] File from pkgroot not present on disk after installation Message-ID: Today's quandary: The file mpd.conf.CSW is present in work/pkgroot: http://dpaste.com/54904/ It's also present in the prototype and in resulting pkgmap after installation: http://dpaste.com/54911/ ...but the file is not there in the disk! (after installing the package) Do you have any ideas why might that be? More info: It's mGAR v2, rev. 5237, the Makefile is here: http://dpaste.com/54926/ Maciej From maciej at opencsw.org Sat Jun 13 15:32:43 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Sat, 13 Jun 2009 14:32:43 +0100 Subject: [csw-maintainers] File from pkgroot not present on disk after installation In-Reply-To: References: Message-ID: Found the problem: cswcpsampleconf != cswsampleconf in SPKG_CLASSES From dam at opencsw.org Sat Jun 13 21:29:02 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sat, 13 Jun 2009 21:29:02 +0200 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: References: Message-ID: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> Hi Maciej, Am 13.06.2009 um 13:04 schrieb Maciej (Matchek) Blizinski: > In order to support sparse zones, paths such as /opt/csw/var/... must > become /var/opt/csw/... - otherwise applications in non-global zones > won't be able to write there. Paths are being set in the $(DIRPATHS), > as a space-separated list. What's the proper way of tweaking it? (say, > define sysconfdir as /etc/opt/csw) > > I guess I could add sysconfdir to NODIRPATHS and then append my own > sysconfdir, but I'm not sure if it's the way Dago intended. Well, in gar.conf.mk the definitions for the directories are conditional, that means you can safely set sysconfdir = /etc/opt/csw localstatedir = /var/opt/csw and everything should be fine. We could make it the default, but I am afraid that this may break existing build descriptions. Best regards -- Dago From phil at bolthole.com Sat Jun 13 22:04:34 2009 From: phil at bolthole.com (Philip Brown) Date: Sat, 13 Jun 2009 13:04:34 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> Message-ID: <20090613200434.GA19773@bolthole.com> On Sat, Jun 13, 2009 at 09:29:02PM +0200, Dagobert Michelsen wrote: > Well, in gar.conf.mk the definitions for the directories are > conditional, > that means you can safely set > sysconfdir = /etc/opt/csw > localstatedir = /var/opt/csw > and everything should be fine. We could make it the default, but I am > afraid that this may break existing build descriptions. > > i think we should make it the default. From bwalton at opencsw.org Sat Jun 13 22:31:51 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sat, 13 Jun 2009 16:31:51 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090613200434.GA19773@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> Message-ID: <1244925073-sup-2633@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Sat Jun 13 16:04:34 -0400 2009: > > sysconfdir = /etc/opt/csw > > localstatedir = /var/opt/csw > > i think we should make it the default. I second this. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From maciej at opencsw.org Sun Jun 14 15:53:54 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Sun, 14 Jun 2009 14:53:54 +0100 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> Message-ID: On Sat, Jun 13, 2009 at 8:29 PM, Dagobert Michelsen wrote: > Well, in gar.conf.mk the definitions for the directories are conditional, > that means you can safely set > sysconfdir = /etc/opt/csw > localstatedir = /var/opt/csw > and everything should be fine. We could make it the default, but I am > afraid that this may break existing build descriptions. You could grep for /opt/csw/var instances in all the build files and replace them accordingly. I see two main issues: - potential problems will be seen at runtime, not at the time of packaging - already installed programs will have their data in /opt/csw/var, some form of data migration might be necessary Perhaps focus on just few important packages would be enough. Postfix and cups come to mind. Maciej From ihsan at dogan.ch Sun Jun 14 16:01:12 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Sun, 14 Jun 2009 16:01:12 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp 2009 In-Reply-To: <4A3285BB.4000705@opencsw.org> References: <4A11B3AF.7090709@opencsw.org> <4A3269BB.4090604@dogan.ch> <4A3285BB.4000705@opencsw.org> Message-ID: <4A3502A8.5020903@dogan.ch> Am 12.6.2009 18:43 Uhr, Trygve Laugst?l schrieb: >>> The OpenCSW Summer Camp 2009 will take place in Oslo/Norway. To figure >>> out when it should take place, I've create a Doodle poll. Please fill >>> out the link bellow: >>> >>> http://www.doodle.com/6zr9kmdr8p9vv69p >> >> The summer camp will take place on the weekend of 22/23 August 2009 in >> Olso/Norway. Thanks to all, who participated with the poll. >> >> @Trygve: >> Could you please arrange the location and provide a list of recommended >> hotels? > > I'll try to ask around next week. > > I think I've found a nice place for us to work during the day too. The > place is the same that the JUG that I lead use so it should be good > enough for us :) Thanks a lot Trygve. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Sun Jun 14 16:02:20 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 14 Jun 2009 10:02:20 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> Message-ID: <1244988012-sup-5133@ntdws12.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizinski's message of Sun Jun 14 09:53:54 -0400 2009: > You could grep for /opt/csw/var instances in all the build files and > replace them accordingly. I see two main issues: > > - potential problems will be seen at runtime, not at the time of packaging > - already installed programs will have their data in /opt/csw/var, > some form of data migration might be necessary If this is announced loudly enough, all maintainers should take precautions at the next update of affected packages. Could a class script mitigate some of the pain here? The script would either install the new file or iff the file exists in the old location, copy that in instead? There are some corner cases that would need to be handled, so it's not as simple as described, but I think it could be done. Thoughts? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From ihsan at opencsw.org Sun Jun 14 16:11:18 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 14 Jun 2009 16:11:18 +0200 Subject: [csw-maintainers] Summer Camp 2009 (Please book your flights) Message-ID: <4A350506.1000804@opencsw.org> Hello, As already mentioned, the date and the location is fixed. I've updated the wiki as well, but the information is not complete. --> http://wiki.opencsw.org/summercamp-2009 The information for accommodation will be available soon. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From Murray.Jensen at csiro.au Sun Jun 14 18:34:16 2009 From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au) Date: Mon, 15 Jun 2009 02:34:16 +1000 Subject: [csw-maintainers] /testing: nasm 2.05.01 Message-ID: <21894.1244997256@gerd> The following files have been put into /testing ... 9aea1d1df09e7f51525d8a9fee95b0ee nasm-2.05.01,REV=2009.06.14-SunOS5.8-i386-CSW.pkg.gz 0fa0d0eebd39d15f6c7ae7336764b4fa nasm-2.05.01,REV=2009.06.14-SunOS5.8-sparc-CSW.pkg.gz I haven't even run them yet - let me know if they work. Cheers! Murray... From ihsan at opencsw.org Sun Jun 14 20:02:26 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 14 Jun 2009 20:02:26 +0200 Subject: [csw-maintainers] exim maintainer In-Reply-To: <1244821262-sup-9225@ntdws12.chass.utoronto.ca> References: <1244821262-sup-9225@ntdws12.chass.utoronto.ca> Message-ID: <4A353B32.2000607@opencsw.org> Am 12.6.2009 17:41 Uhr, Ben Walton schrieb: > Are you still out there? If so, we've got a bug to file against the > exim package. If not, we may consider taking it over. He should be still available. At least he logs in from time to time on the mail server. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From william at wbonnet.net Mon Jun 15 00:27:43 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 15 Jun 2009 00:27:43 +0200 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config Message-ID: <4A35795F.8000005@wbonnet.net> Hi I am building some Xfce packages and i have the following error : checking for pkg-config... /opt/csw/bin/pkg-config checking for pkg-config >= 0.9.0... 0.23 checking for gtk+-2.0 >= 2.10.0... not found *** The required package gtk+-2.0 was not found on your system. *** Please install gtk+-2.0 (atleast version 2.10.0) or adjust *** the PKG_CONFIG_PATH environment variable if you *** installed the package in a nonstandard prefix so that *** pkg-config is able to find it. I checked for gtk+ libs, and they are up to date, version is superior to 2.10.0 (2.12.0). It seems the arguments passed to pkg-config are not correct. Any one already faced this ? cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From mwatters at opencsw.org Mon Jun 15 00:38:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 14 Jun 2009 17:38:26 -0500 Subject: [csw-maintainers] OpenCSW is now listed on subversion's binary download page. Message-ID: <4A357BE2.3080001@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 http://subversion.tigris.org/getting.html - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAko1e+EACgkQLrhmsXMSLxdQXwCgwVHuKXRGCBQqiB9JG7a8d+Sy s7UAniekUhnHMhwAtkMhGiTQHRPPcFcI =5OMg -----END PGP SIGNATURE----- From phil at bolthole.com Mon Jun 15 06:02:21 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 14 Jun 2009 21:02:21 -0700 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <4A35795F.8000005@wbonnet.net> References: <4A35795F.8000005@wbonnet.net> Message-ID: <20090615040221.GD12886@bolthole.com> On Mon, Jun 15, 2009 at 12:27:43AM +0200, William Bonnet wrote: > Hi > > I am building some Xfce packages and i have the following error : > > checking for pkg-config... /opt/csw/bin/pkg-config > checking for pkg-config >= 0.9.0... 0.23 > checking for gtk+-2.0 >= 2.10.0... not found well it's worse than that. pkg-config --modversion gtk+-2.0 Package xrender was not found in the pkg-config search path. Package 'xrender', required by 'cairo', not found I'm guiessing with the recent rebilds of the xrender package, someone removed the CSWxrender package from the build machines, but forgot to put it back. From dam at opencsw.org Mon Jun 15 11:32:51 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 15 Jun 2009 11:32:51 +0200 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <20090615040221.GD12886@bolthole.com> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> Message-ID: <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> Hi Phil, Am 15.06.2009 um 06:02 schrieb Philip Brown: > On Mon, Jun 15, 2009 at 12:27:43AM +0200, William Bonnet wrote: >> Hi >> >> I am building some Xfce packages and i have the following error : >> >> checking for pkg-config... /opt/csw/bin/pkg-config >> checking for pkg-config >= 0.9.0... 0.23 >> checking for gtk+-2.0 >= 2.10.0... not found > > well it's worse than that. > > pkg-config --modversion gtk+-2.0 > Package xrender was not found in the pkg-config search path. > Package 'xrender', required by 'cairo', not found > > I'm guiessing with the recent rebilds of the xrender package, someone > removed the CSWxrender package from the build machines, but forgot > to put > it back. No. In fact, gtk+-2.0 requires cairo, which requires xrender >= 0.6. I removed the .pc-files for the legacy .so as there should be no compiles against th legacy library. This breaks cairo-xlib-xrender, cairo and xft, all of which I am going to release soon. Best regards -- Dago From pfelecan at opencsw.org Mon Jun 15 16:02:49 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 15 Jun 2009 16:02:49 +0200 Subject: [csw-maintainers] contact maintainer page Message-ID: I'm trying to use the site's mailer: http://www.opencsw.org/eml.php?devname=pfelecan After entering my e-mail address it shows me a page with the text "Debug: part 1". Nice but not useful if somebody tries to write me something through this... -- Peter From dam at opencsw.org Mon Jun 15 16:40:18 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 15 Jun 2009 16:40:18 +0200 Subject: [csw-maintainers] New in testing: libxft2 and libcairo Message-ID: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> Hi, new packages for libxft and libcairo are in testing: libcairo-1.8.6,REV=2009.06.15-SunOS5.8-i386-CSW.pkg.gz libcairo-1.8.6,REV=2009.06.15-SunOS5.8-sparc-CSW.pkg.gz libcairo_devel-1.8.6,REV=2009.06.15-SunOS5.8-i386-CSW.pkg.gz libcairo_devel-1.8.6,REV=2009.06.15-SunOS5.8-sparc-CSW.pkg.gz libcairo_doc-1.8.6,REV=2009.06.15-SunOS5.8-all-CSW.pkg.gz libxft2-2.1.13,REV=2009.06.15-SunOS5.8-i386-CSW.pkg.gz libxft2-2.1.13,REV=2009.06.15-SunOS5.8-sparc-CSW.pkg.gz libxft2_devel-2.1.13,REV=2009.06.15-SunOS5.8-i386-CSW.pkg.gz libxft2_devel-2.1.13,REV=2009.06.15-SunOS5.8-sparc-CSW.pkg.gz I am installing them now on build8st. William: Do you mind giving them a try with your compilation problem? Best regards -- Dago From phil at bolthole.com Mon Jun 15 16:51:08 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 07:51:08 -0700 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> Message-ID: <20090615145107.GD34818@bolthole.com> On Mon, Jun 15, 2009 at 11:32:51AM +0200, Dagobert Michelsen wrote: >> I'm guiessing with the recent rebilds of the xrender package, someone >> removed the CSWxrender package from the build machines, but forgot to >> put >> it back. > > No. In fact, gtk+-2.0 requires cairo, which requires xrender >= 0.6. I > removed > the .pc-files for the legacy .so as there should be no compiles against > the legacy library. But.... it was working before. How did it break now?? we dont have a "new" cairo. so its a mystery why cairo is complaning now, other than "we dont have xrender installed any more". Seems like you could just install xrender again and it would be happy From phil at bolthole.com Mon Jun 15 16:59:06 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 07:59:06 -0700 Subject: [csw-maintainers] contact maintainer page In-Reply-To: References: Message-ID: <20090615145906.GE34818@bolthole.com> On Mon, Jun 15, 2009 at 04:02:49PM +0200, Peter FELECAN wrote: > I'm trying to use the site's mailer: > http://www.opencsw.org/eml.php?devname=pfelecan > > After entering my e-mail address it shows me a page with the text > "Debug: part 1". Nice but not useful if somebody tries to write me > something through this... posix_getpwnam() [a php call] seems to be broken somehow on the website now. I have asked Ihsan to look into it. Until that is resolved, the mail form doesnt work :-( From william at wbonnet.net Mon Jun 15 16:59:17 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 15 Jun 2009 16:59:17 +0200 Subject: [csw-maintainers] New in testing: libxft2 and libcairo In-Reply-To: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> References: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> Message-ID: <4A3661C5.5030602@wbonnet.net> Hi Dago > William: Do you mind giving them a try with your compilation problem? Thanks a lot I'm testing it. Cheers W. From skayser at opencsw.org Mon Jun 15 17:04:12 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 15 Jun 2009 17:04:12 +0200 (CEST) Subject: [csw-maintainers] contact maintainer page In-Reply-To: <20090615145906.GE34818@bolthole.com> References: <20090615145906.GE34818@bolthole.com> Message-ID: <55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de> Philip Brown wrote: > On Mon, Jun 15, 2009 at 04:02:49PM +0200, Peter FELECAN wrote: >> I'm trying to use the site's mailer: >> http://www.opencsw.org/eml.php?devname=pfelecan >> >> After entering my e-mail address it shows me a page with the text >> "Debug: part 1". Nice but not useful if somebody tries to write me >> something through this... > > posix_getpwnam() [a php call] > seems to be broken somehow on the website now. Might be related to the recent php5 modules split-off? [1] Sebastian [1] http://lists.opencsw.org/pipermail/announce/2009-May/000059.html From mwatters at opencsw.org Mon Jun 15 17:22:11 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 15 Jun 2009 10:22:11 -0500 Subject: [csw-maintainers] contact maintainer page In-Reply-To: <55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de> References: <20090615145906.GE34818@bolthole.com> <55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de> Message-ID: <4A366723.3050806@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Sebastian Kayser wrote: > Philip Brown wrote: >> On Mon, Jun 15, 2009 at 04:02:49PM +0200, Peter FELECAN wrote: >>> I'm trying to use the site's mailer: >>> http://www.opencsw.org/eml.php?devname=pfelecan >>> >>> After entering my e-mail address it shows me a page with the text >>> "Debug: part 1". Nice but not useful if somebody tries to write me >>> something through this... >> posix_getpwnam() [a php call] >> seems to be broken somehow on the website now. > > Might be related to the recent php5 modules split-off? [1] Installing the following should fix php5_posix 5.2.9,REV=2009.04.29 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko2ZyMACgkQLrhmsXMSLxcNhACbBXdVFDyrBIcG+33PFeBX6aCf 6vYAoKu0QSLYN+M8f2+Wa1lrllPLTx3D =U4et -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 15 17:39:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 15 Jun 2009 10:39:26 -0500 Subject: [csw-maintainers] New in testing: libxft2 and libcairo In-Reply-To: <4A3661C5.5030602@wbonnet.net> References: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> <4A3661C5.5030602@wbonnet.net> Message-ID: <4A366B2E.9010503@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 William Bonnet wrote: > Hi Dago >> William: Do you mind giving them a try with your compilation problem? > Thanks a lot > > I'm testing it. > > Cheers > W. > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I am trying to build pycairo against it, can you please install python_devel on build8st. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko2ay4ACgkQLrhmsXMSLxegOgCdGcDZllvG7q1yQY0Q/t5uR8yO j2EAnRcHC7+Sgb3T4mDbRQ0k+htKSMA9 =KNfh -----END PGP SIGNATURE----- From dam at opencsw.org Mon Jun 15 17:42:11 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 15 Jun 2009 17:42:11 +0200 Subject: [csw-maintainers] New in testing: libxft2 and libcairo In-Reply-To: <4A366B2E.9010503@opencsw.org> References: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> <4A3661C5.5030602@wbonnet.net> <4A366B2E.9010503@opencsw.org> Message-ID: <0A545E97-991F-47D8-9DD0-45A58F349165@opencsw.org> Hi Mike, Am 15.06.2009 um 17:39 schrieb Mike Watters: > I am trying to build pycairo against it, can you please install > python_devel on > build8st. Done. Best regards -- Dago From william at wbonnet.net Mon Jun 15 18:33:51 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 15 Jun 2009 18:33:51 +0200 Subject: [csw-maintainers] New in testing: libxft2 and libcairo In-Reply-To: <4A3661C5.5030602@wbonnet.net> References: <987862FF-5D58-4511-9EF1-5CCF88EBBCD0@opencsw.org> <4A3661C5.5030602@wbonnet.net> Message-ID: <4A3677EF.8050103@wbonnet.net> Hi > I'm testing it. It seems to fix my problems :) Thanks a lot. Now i have to install all dependencies on the build8*t machines (to recompile Xfce 4.6.1 on these boxes). Dago can you check that i can sudo pkgrm and pkgadd on the tests machines please ? (i never tried ...). cheers W. From william at wbonnet.net Mon Jun 15 18:35:38 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 15 Jun 2009 18:35:38 +0200 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <20090615145107.GD34818@bolthole.com> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> <20090615145107.GD34818@bolthole.com> Message-ID: <4A36785A.7040801@wbonnet.net> Philip Brown a ?crit : > On Mon, Jun 15, 2009 at 11:32:51AM +0200, Dagobert Michelsen wrote: > >>> I'm guiessing with the recent rebilds of the xrender package, someone >>> removed the CSWxrender package from the build machines, but forgot to >>> put >>> it back. >>> >> No. In fact, gtk+-2.0 requires cairo, which requires xrender >= 0.6. I >> removed >> the .pc-files for the legacy .so as there should be no compiles against >> the legacy library. >> > > But.... it was working before. How did it break now?? > > we dont have a "new" cairo. so its a mystery why cairo is complaning now, > other than "we dont have xrender installed any more". > > Seems like you could just install xrender again and it would be happy > It now works on the test machines :) From mwatters at opencsw.org Mon Jun 15 18:42:42 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 15 Jun 2009 11:42:42 -0500 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <4A36785A.7040801@wbonnet.net> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> <20090615145107.GD34818@bolthole.com> <4A36785A.7040801@wbonnet.net> Message-ID: <4A367A02.4090801@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 William Bonnet wrote: > Philip Brown a ?crit : >> On Mon, Jun 15, 2009 at 11:32:51AM +0200, Dagobert Michelsen wrote: >> >>>> I'm guiessing with the recent rebilds of the xrender package, someone >>>> removed the CSWxrender package from the build machines, but forgot >>>> to put >>>> it back. >>>> >>> No. In fact, gtk+-2.0 requires cairo, which requires xrender >= 0.6. >>> I removed >>> the .pc-files for the legacy .so as there should be no compiles >>> against the legacy library. >>> >> >> But.... it was working before. How did it break now?? >> >> we dont have a "new" cairo. so its a mystery why cairo is complaning now, >> other than "we dont have xrender installed any more". >> >> Seems like you could just install xrender again and it would be happy >> > > It now works on the test machines :) > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers DOH, What happened to libcairo on build8st? I finally get everything ready to build pycairo, type make build and get CAIRO_PS_LEVEL_2 undefined... I check the version of libcairo again and it has reverted to 1.4.10 ;-( guess I need to be faster next time... ;-) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko2egIACgkQLrhmsXMSLxez7ACfc1XO33DqSPRcBR9dRb2bXYCc lr4AoOGWIMnoXBYVr46dBoxfEb1Y+rts =Is1H -----END PGP SIGNATURE----- From phil at bolthole.com Mon Jun 15 19:02:41 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 10:02:41 -0700 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <4A36785A.7040801@wbonnet.net> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> <20090615145107.GD34818@bolthole.com> <4A36785A.7040801@wbonnet.net> Message-ID: <20090615170241.GN95242@bolthole.com> On Mon, Jun 15, 2009 at 06:35:38PM +0200, William Bonnet wrote: > Philip Brown a ?crit : >> Seems like you could just install xrender again and it would be happy >> > > It now works on the test machines :) well, that's nice and alll.. but we need our standard build machines functional as well!! From bwalton at opencsw.org Mon Jun 15 19:45:54 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 15 Jun 2009 13:45:54 -0400 Subject: [csw-maintainers] versioning /etc Message-ID: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> Hi Phil and Peter, I wondered what both you think about supporting a tool like etckeeper[1] from pkg-get and pkgutil. With the recent discussion about defaulting to /etc/opt/csw for our stuff, I'm considering packaging it up. The tools would need to grow the idea of pre- and post- install hooks, so that etckeeper could be driven to snapshot /etc before and after packages are manipulated with the high-level tools. These hooks may be useful for other things too...it seems that apt, yum and co support a generic hook mechanism for things like this to plug in to. I've been dabbling etckeeper a bit on a few systems and I like aspects of it, although I'm still not ready to do a wholesale rollout. Thoughts? Thanks -Ben [1] http://joey.kitenet.net/code/etckeeper/ -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Mon Jun 15 19:55:11 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 10:55:11 -0700 Subject: [csw-maintainers] versioning /etc In-Reply-To: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> Message-ID: <20090615175511.GV95242@bolthole.com> On Mon, Jun 15, 2009 at 01:45:54PM -0400, Ben Walton wrote: > > Hi Phil and Peter, > > I wondered what both you think about supporting a tool like > etckeeper[1] from pkg-get and pkgutil. With the recent discussion > about defaulting to /etc/opt/csw for our stuff, I'm considering > packaging it up. > > The tools would need to grow the idea of pre- and post- install hooks, > so that etckeeper could be driven to snapshot /etc before and after > packages are manipulated with the high-level tools. These hooks may > be useful for other things too...it seems that apt, yum and co support > a generic hook mechanism for things like this to plug in to. I think that sounds rather overly broad. Not to mention, it sorta assumes that "you run an install tool, and you DONT KNOW WHAT IT DOES, so you use this tool to make a copy of everything". If one of our packages falls in the category of "we dont know what it does in etc", it's a bad package, and needs to be fixed!! If you want to package up etckeeper... great! That being said... As far as integrating it in our install tools, however, I dont think it's a good idea. It's extra complication, and it encourages sloppy package creation. From bwalton at opencsw.org Mon Jun 15 20:03:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 15 Jun 2009 14:03:47 -0400 Subject: [csw-maintainers] versioning /etc In-Reply-To: <20090615175511.GV95242@bolthole.com> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <20090615175511.GV95242@bolthole.com> Message-ID: <1245088612-sup-5510@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Mon Jun 15 13:55:11 -0400 2009: > I think that sounds rather overly broad. > > Not to mention, it sorta assumes that > > "you run an install tool, and you DONT KNOW WHAT IT DOES, so you > use this tool to make a copy of everything". No, it's not to mitigate this at all... > If one of our packages falls in the category of > "we dont know what it does in etc", > it's a bad package, and needs to be fixed!! Agreed, but again, that's not the purpose of this tool. > As far as integrating it in our install tools, however, I dont think > it's a good idea. You wouldn't be integrating it _specifically_, but rather providing a framework for other things to be transparently integrated, thus allowing a site to leverage this tool or others that may find running code pre and post package actions to be useful. etckeeper would simply leverage these hooks. Nobody would be forced to use it, as absence of the package (and thus the hooks it would install) would see pkg-get/pkgutil operate as it had historically. > It's extra complication, and it encourages sloppy package creation. Nope. The point of etckeeper is to version /etc...it's not to mitigate sloppy packaging. I think it's fairly common to version parts of /etc these days, whether you use some vcs or cp -p. This wraps it up into a nice package and allows for a single repo for all of /etc instead of one for apache, one for foo, one for...The beauty of this is that it can help in cases where a new version of a package breaks your existing install by trampling a config file (incompatible changes, etc)...or if you want to see exactly what it changed. It goes beyond CSW packages, to everything that lives in /etc. Again, as I mentioned, I'm not totally sold on this particular package myself, but the idea of pkg-get/pkgutil providing hooks for other things can be completely independent of this... -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Mon Jun 15 20:52:34 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 11:52:34 -0700 Subject: [csw-maintainers] versioning /etc In-Reply-To: <1245088612-sup-5510@ntdws12.chass.utoronto.ca> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <20090615175511.GV95242@bolthole.com> <1245088612-sup-5510@ntdws12.chass.utoronto.ca> Message-ID: <20090615185234.GX95242@bolthole.com> On Mon, Jun 15, 2009 at 02:03:47PM -0400, Ben Walton wrote: > Nope. The point of etckeeper is to version /etc...it's not to > mitigate sloppy packaging. [...] > The beauty > of this is that it can help in cases where a new version of a package > breaks your existing install by trampling a config file (incompatible > changes, etc)...or if you want to see exactly what it changed. Hmm. that sounds potentially useful. > It > goes beyond CSW packages, to everything that lives in /etc. And that sounds... like a lot of work. At a very high level, do you have architectural suggestions as to how this sort of integration would function? From rupert at opencsw.org Mon Jun 15 21:08:21 2009 From: rupert at opencsw.org (rupert THURNER) Date: Mon, 15 Jun 2009 21:08:21 +0200 Subject: [csw-maintainers] versioning /etc In-Reply-To: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> Message-ID: <6af4270906151208q757f503bs1fb2022bb0f068a4@mail.gmail.com> On Mon, Jun 15, 2009 at 19:45, Ben Walton wrote: > > Hi Phil and Peter, > > I wondered what both you think about supporting a tool like > etckeeper[1] from pkg-get and pkgutil. ?With the recent discussion > about defaulting to /etc/opt/csw for our stuff, I'm considering > packaging it up. > > The tools would need to grow the idea of pre- and post- install hooks, > so that etckeeper could be driven to snapshot /etc before and after > packages are manipulated with the high-level tools. ?These hooks may > be useful for other things too...it seems that apt, yum and co support > a generic hook mechanism for things like this to plug in to. > > I've been dabbling etckeeper a bit on a few systems and I like aspects > of it, although I'm still not ready to do a wholesale rollout. > > Thoughts? this does not mean that the default /opt/csw/etc goes away, isn't it? because we enjoyed a lot that one mountpoint is sufficient to run csw. rupert. From bwalton at opencsw.org Mon Jun 15 21:31:35 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 15 Jun 2009 15:31:35 -0400 Subject: [csw-maintainers] versioning /etc In-Reply-To: <20090615185234.GX95242@bolthole.com> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <20090615175511.GV95242@bolthole.com> <1245088612-sup-5510@ntdws12.chass.utoronto.ca> <20090615185234.GX95242@bolthole.com> Message-ID: <1245092338-sup-6372@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Mon Jun 15 14:52:34 -0400 2009: > > It goes beyond CSW packages, to everything that lives in /etc. > > And that sounds... like a lot of work. > At a very high level, do you have architectural suggestions as to how this > sort of integration would function? It actually shouldn't be that bad. My suggestion would be the tried and true .d directory. I'd suggest something like: /etc/opt/csw/pkg-hooks/pre.d /etc/opt/csw/pkg-hooks/post.d (no difference between -i and -u is required, I don't think). pkg-get/pkgutil would then do: ---> before doing work cd /etc/opt/csw/pkg-hooks/pre.d/ for f in *; do [ -x $f ] && ./$f done ---> after doing work cd /etc/opt/csw/pkg-hooks/post.d/ for f in *; do [ -x $f ] && ./$f done An alternate sort order could be applied if required...this is just a simplistic approach. There could (possibly) be a user settable option in csw.conf as to how non-zero exits would be handled, thus allowing for various hooks to prevent a run if some trigger is detected, etc. Then, etckeeper (just as an example) would, as part of it's package, drop a file named 01-etckeeper in pkg-hooks/pre.d/ with a script like: #!/bin/sh /opt/csw/bin/etckeeper preinstall A similar post install hook would be placed in post.d. How does that sound? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Mon Jun 15 21:39:22 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 12:39:22 -0700 Subject: [csw-maintainers] versioning /etc In-Reply-To: <1245092338-sup-6372@ntdws12.chass.utoronto.ca> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <20090615175511.GV95242@bolthole.com> <1245088612-sup-5510@ntdws12.chass.utoronto.ca> <20090615185234.GX95242@bolthole.com> <1245092338-sup-6372@ntdws12.chass.utoronto.ca> Message-ID: <20090615193922.GA95242@bolthole.com> On Mon, Jun 15, 2009 at 03:31:35PM -0400, Ben Walton wrote: > It actually shouldn't be that bad. My suggestion would be the tried > and true .d directory. I'd suggest something like: > > /etc/opt/csw/pkg-hooks/pre.d > /etc/opt/csw/pkg-hooks/post.d > Hmmm. interesting idea. still pondering it. but i'll make the comment now, that I think we shouldnt abbreviate. so, "preinstall.d", or simply "preinstall", vs "pre.d" > (no difference between -i and -u is required, I don't think). not so sure about that one either. From bwalton at opencsw.org Mon Jun 15 21:48:16 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 15 Jun 2009 15:48:16 -0400 Subject: [csw-maintainers] versioning /etc In-Reply-To: <20090615193922.GA95242@bolthole.com> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <20090615175511.GV95242@bolthole.com> <1245088612-sup-5510@ntdws12.chass.utoronto.ca> <20090615185234.GX95242@bolthole.com> <1245092338-sup-6372@ntdws12.chass.utoronto.ca> <20090615193922.GA95242@bolthole.com> Message-ID: <1245095107-sup-9601@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Mon Jun 15 15:39:22 -0400 2009: > interesting idea. > still pondering it. but i'll make the comment now, that I think we shouldnt > abbreviate. > > so, "preinstall.d", or simply "preinstall", vs "pre.d" Ok, as long as it's clear what the directory is for, the name isn't important. Since it's meant for computer consumption, longer names are fine with me too. > > (no difference between -i and -u is required, I don't think). > > not so sure about that one either. My argument for the simplification is that there is no underlying concept in pkg{add,rm} of -i being different from -u. There wouldn't be anything inherently wrong with providing more options up front though. There is also the wrinkle that a -u can imply -i for other packages, but maybe that's "just one of those things"... -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Mon Jun 15 22:31:32 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 15 Jun 2009 22:31:32 +0200 Subject: [csw-maintainers] gtk+2.0 not found by pkg-config In-Reply-To: <4A367A02.4090801@opencsw.org> References: <4A35795F.8000005@wbonnet.net> <20090615040221.GD12886@bolthole.com> <6D6D753E-79E2-435D-848C-D6C8848CC7F2@opencsw.org> <20090615145107.GD34818@bolthole.com> <4A36785A.7040801@wbonnet.net> <4A367A02.4090801@opencsw.org> Message-ID: <8BE61299-F4BB-4992-960D-D30953C4BA8F@opencsw.org> Hi Mike, Am 15.06.2009 um 18:42 schrieb Mike Watters: > William Bonnet wrote: >> It now works on the test machines :) > > DOH, What happened to libcairo on build8st? > I finally get everything ready to build pycairo, type make build > and get > CAIRO_PS_LEVEL_2 undefined... I check the version of libcairo > again and it has > reverted to 1.4.10 ;-( > > guess I need to be faster next time... ;-) :-) Hopefully cairo will be released soon, so you should be ok on all machines then. Best regards -- Dago From mwatters at opencsw.org Mon Jun 15 22:35:03 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 15 Jun 2009 15:35:03 -0500 Subject: [csw-maintainers] lcms now in testing Message-ID: <4A36B077.6030108@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 "Little cms intends to be a small-footprint, speed optimized color management engine in open source form." --> http://www.littlecms.com/ pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u lcms Feedback is always welcome. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko2sHcACgkQLrhmsXMSLxeoKwCdHmuCLc/zkwVn8Q1nzJuOgqbf 7W8AoMkrYOfRt0tXGbItP42VMyNpH1W/ =7JRL -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 15 22:47:23 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 15 Jun 2009 15:47:23 -0500 Subject: [csw-maintainers] lcms now in testing In-Reply-To: <4A36B077.6030108@opencsw.org> References: <4A36B077.6030108@opencsw.org> Message-ID: <4A36B35B.5080405@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Rodger, Sorry for popping this in testing, I missed you owning it. (I looked for liblcms and not lcms ) I will remove from testing. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko2s1sACgkQLrhmsXMSLxfI2gCgnkqiSScb3Rm1E4H+zSuaVy5Y 5qwAnjMaXHPvGyWab7Ep70SCJ5h5/vhb =+9dH -----END PGP SIGNATURE----- From hson at opencsw.org Tue Jun 16 00:34:40 2009 From: hson at opencsw.org (=?ISO-8859-1?Q?Roger_H=E5kansson?=) Date: Tue, 16 Jun 2009 00:34:40 +0200 Subject: [csw-maintainers] lcms now in testing In-Reply-To: <4A36B35B.5080405@opencsw.org> References: <4A36B077.6030108@opencsw.org> <4A36B35B.5080405@opencsw.org> Message-ID: <4A36CC80.4060103@opencsw.org> Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Rodger, > Sorry for popping this in testing, I missed you owning it. > (I looked for liblcms and not lcms ) > > I will remove from testing. No problem, I just took over it from the previous "owner" due to the fact that imagemagick needs it. But I'll push out a new package ASAP. From phil at bolthole.com Tue Jun 16 00:58:17 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 15 Jun 2009 15:58:17 -0700 Subject: [csw-maintainers] versioning /etc In-Reply-To: <6af4270906151208q757f503bs1fb2022bb0f068a4@mail.gmail.com> References: <1245087696-sup-9325@ntdws12.chass.utoronto.ca> <6af4270906151208q757f503bs1fb2022bb0f068a4@mail.gmail.com> Message-ID: <20090615225817.GI95242@bolthole.com> On Mon, Jun 15, 2009 at 09:08:21PM +0200, rupert THURNER wrote: > this does not mean that the default /opt/csw/etc goes away, isn't it? nope. From william at wbonnet.net Tue Jun 16 09:42:22 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 16 Jun 2009 09:42:22 +0200 Subject: [csw-maintainers] Is maintainer still active Message-ID: <4A374CDE.6020006@wbonnet.net> Hi I was wondering ... is Nathan Coraor still active ? I don't remember to have heard from him over the last month, and i opened two upgrade request against two packages he maintains. I am asking this just in case it is known Nathan is retired :) cheers W. From nate at blastwave.org Tue Jun 16 16:05:36 2009 From: nate at blastwave.org (Nate Coraor) Date: Tue, 16 Jun 2009 10:05:36 -0400 Subject: [csw-maintainers] Is maintainer still active In-Reply-To: <4A374CDE.6020006@wbonnet.net> References: <4A374CDE.6020006@wbonnet.net> Message-ID: <4A37A6B0.4070506@blastwave.org> William Bonnet wrote: > Hi > > I was wondering ... is Nathan Coraor still active ? I don't remember to > have heard from him over the last month, and i opened two upgrade > request against two packages he maintains. > > I am asking this just in case it is known Nathan is retired :) I've tried to get back into it a few times, but yes, I'm retired. --nate From phil at bolthole.com Tue Jun 16 18:22:28 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 16 Jun 2009 09:22:28 -0700 Subject: [csw-maintainers] Is maintainer still active In-Reply-To: <4A37A6B0.4070506@blastwave.org> References: <4A374CDE.6020006@wbonnet.net> <4A37A6B0.4070506@blastwave.org> Message-ID: <20090616162227.GA49403@bolthole.com> On Tue, Jun 16, 2009 at 10:05:36AM -0400, Nate Coraor wrote: > > I've tried to get back into it a few times, but yes, I'm retired. > noted From william at wbonnet.net Tue Jun 16 18:28:27 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 16 Jun 2009 18:28:27 +0200 Subject: [csw-maintainers] Is maintainer still active In-Reply-To: <4A37A6B0.4070506@blastwave.org> References: <4A374CDE.6020006@wbonnet.net> <4A37A6B0.4070506@blastwave.org> Message-ID: <4A37C82B.9090606@wbonnet.net> Hi Nate >> >> I am asking this just in case it is known Nathan is retired :) > > I've tried to get back into it a few times, but yes, I'm retired. Thanks for answering Nate. So, at short term, i propose to update three of your packages (perl module Glib and its depends). I need some fresh version to build Xfce 4.6.1. I hope you will find time to get back in the project, in the next weeks or months cheers W. From skayser at opencsw.org Tue Jun 16 21:54:39 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 16 Jun 2009 21:54:39 +0200 Subject: [csw-maintainers] Bugtracker (Mantis) patched for performance, feedback welcome In-Reply-To: <4A326858.4010405@dogan.ch> References: <4A319985.8040904@opencsw.org> <1244765051-sup-8822@ntdws12.chass.utoronto.ca> <4A326858.4010405@dogan.ch> Message-ID: <4A37F87F.1020900@opencsw.org> Ihsan Dogan wrote: > Ben Walton wrote: > >>> i have patched a Mantis test installation today to improve the long page >>> load times we have been seeing. A couple of features that we don't use >>> as of yet (sub-projects, private projects, product versions) have been >>> de-facto deactivated as a result of this. The changes are targeted at >>> data display only and should be non-disruptive. >> +1. It looks good to me and it's a _lot_ faster. Nice work! :) > > Wow, it's a rocket now. Thanks Sebastian. Thanks for the feedback. Our productive installation is now also tweaked. Happy bug squashing ;) Sebastian From william at wbonnet.net Tue Jun 16 21:59:31 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 16 Jun 2009 21:59:31 +0200 Subject: [csw-maintainers] Bugtracker (Mantis) patched for performance, feedback welcome In-Reply-To: <4A319985.8040904@opencsw.org> References: <4A319985.8040904@opencsw.org> Message-ID: <4A37F9A3.1050209@wbonnet.net> Hi Sebastian > i have patched a Mantis test installation today to improve the long page > load times we have been seeing. A couple of features that we don't use > as of yet (sub-projects, private projects, product versions) have been > de-facto deactivated as a result of this. The changes are targeted at > data display only and should be non-disruptive. > Thanks ! cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Tue Jun 16 22:31:59 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 16 Jun 2009 22:31:59 +0200 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1244925073-sup-2633@ntdws12.chass.utoronto.ca> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> Message-ID: <4A38013F.60500@opencsw.org> Ben Walton wrote: > Excerpts from Philip Brown's message of Sat Jun 13 16:04:34 -0400 2009: > >>> sysconfdir = /etc/opt/csw >>> localstatedir = /var/opt/csw >> i think we should make it the default. > > I second this. Can we ponder about the /etc stuff a bit? From working with other people using OpenCSW i know that there is always some kind of confusion as to where to find configuration files. Most are in /opt/csw/etc, but some are in /etc/opt/csw. Our file layout standards page [1] says that machine-specific files should go to /etc/opt/csw, but even i as a maintainer sometimes find it difficult to decide whether to place them there or not. Furthermore IIRC there have been discussions on the list about issues with sparse zone installations and configuration files beneath /opt. >From that perspective i would second this change also (only one place to look for configs, smoother integration with zones), but then there has been Rupert who raised some concerns about not being able to just share out /opt/csw any more [2]. Is this common? Is it a requirement that we can break in order to eventually just have /etc/opt/csw or how would it work? Sebastian [1] http://opencsw.org/standards/layout [2] http://lists.opencsw.org/pipermail/maintainers/2009-June/002914.html From ihsan at opencsw.org Tue Jun 16 23:23:47 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Tue, 16 Jun 2009 23:23:47 +0200 Subject: [csw-maintainers] Summer Camp 2009 Message-ID: <4A380D63.6020706@opencsw.org> Hello, I've updated the wiki page with the attendees and the hotel. I've booked my flight and hotel already and it would be great if we could stay all together in one hotel. --> http://wiki.opencsw.org/summercamp-2009 PS: Topics and schedule will follow... Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From phil at bolthole.com Tue Jun 16 23:24:24 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 16 Jun 2009 14:24:24 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <4A38013F.60500@opencsw.org> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> Message-ID: <20090616212424.GD51834@bolthole.com> On Tue, Jun 16, 2009 at 10:31:59PM +0200, Sebastian Kayser wrote: > Our file layout standards page [1] says that machine-specific files > should go to /etc/opt/csw, but even i as a maintainer sometimes find it > difficult to decide whether to place them there or not. when in doubt, discuss it on the maintainers list. > Furthermore IIRC > there have been discussions on the list about issues with sparse zone > installations and configuration files beneath /opt. they should be treated pretty much in the same way as if /opt was NFS-shared between separate hosts. Sometimes, it is DESIRABLE to have the configs globally in /opt/csw/etc. Ideally, a program could be configured to first chec /opt/csw/etc, and then override with any local configs in /etc/opt/csw But if not, then it's up to the maintainer to decide. From mwatters at opencsw.org Tue Jun 16 23:58:18 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 16 Jun 2009 16:58:18 -0500 Subject: [csw-maintainers] squid now in testing Message-ID: <4A38157A.2080002@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: Changed the default Cache location to /var/opt/csw/squid from /opt/csw/var - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko4FXoACgkQLrhmsXMSLxdkGACgvkknL3YgYzKDxVxJYmYqLQSm hYUAoNfPacS5YNoZqGmYUjA5Q7VvO/Hj =mDxZ -----END PGP SIGNATURE----- From mwatters at opencsw.org Wed Jun 17 00:04:28 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 16 Jun 2009 17:04:28 -0500 Subject: [csw-maintainers] maintainer for gmp? Message-ID: <4A3816EC.3060103@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Is anyone working on gmp? I am updating libmpfr, and would like to compile it 64bit. if not, I will rebuild gmp - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko4FusACgkQLrhmsXMSLxdE6ACgsekxESdj/NQqQgMMErVa+Oth 3ykAoL6tj6rnkN9df3jeCrX918WEAJO6 =E9Jx -----END PGP SIGNATURE----- From william at wbonnet.net Wed Jun 17 00:07:56 2009 From: william at wbonnet.net (William Bonnet) Date: Wed, 17 Jun 2009 00:07:56 +0200 Subject: [csw-maintainers] Anyone for Eclipse Message-ID: <4A3817BC.6030101@wbonnet.net> Hi, I am current maintainer of Eclipse, and i am a bit overloaded ;) Eclipse is a very important package, which should be kept up to date. So if anyone is interested in having this package, i'll be happy to step down. Otherwise i'll update it ... not before September (for a real target date). cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Wed Jun 17 00:08:56 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 17 Jun 2009 00:08:56 +0200 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090616212424.GD51834@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> Message-ID: <4A3817F8.50202@opencsw.org> Philip Brown wrote: > On Tue, Jun 16, 2009 at 10:31:59PM +0200, Sebastian Kayser wrote: >> Our file layout standards page [1] says that machine-specific files >> should go to /etc/opt/csw, but even i as a maintainer sometimes find it >> difficult to decide whether to place them there or not. > > when in doubt, discuss it on the maintainers list. > >> Furthermore IIRC >> there have been discussions on the list about issues with sparse zone >> installations and configuration files beneath /opt. > > they should be treated pretty much in the same way as if /opt was NFS-shared > between separate hosts. > > Sometimes, it is DESIRABLE to have the configs globally in /opt/csw/etc. Does "desirable" go hand in hand with the core notion of this thread to change the default $(sysconfdir) to /etc/opt/csw? Have all configuration files in /etc/opt/csw eventually unless a maintainer decides otherwise for his packages? > Ideally, a program could be configured to first chec /opt/csw/etc, and then > override with any local configs in /etc/opt/csw Ok, this is easy for daemons via their init scripts. Not so easy for most other programs. They read one config file location and usually a dot file. > But if not, then it's up to the maintainer to decide. And this leaves the user at the point that i mentioned initially. "Mhh, there are configuration files in two places. Where is the one for package foo located?". Not one of my top concerns right now, just something that i have observed and wanted to share while we are talking about configuration file locations. Sure, they can look it up, but it lends itself to confusion. Sebastian From skayser at opencsw.org Wed Jun 17 00:32:41 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 17 Jun 2009 00:32:41 +0200 Subject: [csw-maintainers] Summer Camp 2009 In-Reply-To: <4A380D63.6020706@opencsw.org> References: <4A380D63.6020706@opencsw.org> Message-ID: <4A381D89.3080506@opencsw.org> Ihsan Dogan wrote: > I've updated the wiki page with the attendees and the hotel. I've booked > my flight and hotel already and it would be great if we could stay all > together in one hotel. > > --> http://wiki.opencsw.org/summercamp-2009 > > > PS: Topics and schedule will follow... For the topics, how about each topic "owner" fills in some information on the respective topic (similar to how it was done for the meeting Zurich [1]) so that the topic itself is prepared and others can give it a thought (and come up with possibly helpful ideas) in advance? Especially if there is something that you might need others to think about a bit first in order to have a good discussion. Sebastian [1] http://wiki.opencsw.org/suggestions From dam at opencsw.org Wed Jun 17 09:14:22 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 17 Jun 2009 09:14:22 +0200 Subject: [csw-maintainers] libtool on build8st downgraded References: <4A388CC6.4090905@opencsw.org> Message-ID: <5AB3EAB5-DBEC-4CA0-A6E0-45D155E821F8@opencsw.org> Hi, Anfang der weitergeleiteten E-Mail: > K?nntest du nicht auf build8st das alte Libtool installieren? FYI: libtool on build8st has been downgraded temporarily to 1.5.24. Best regards -- Dago From maciej at opencsw.org Wed Jun 17 09:37:24 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Wed, 17 Jun 2009 08:37:24 +0100 Subject: [csw-maintainers] Broken links on Trac timeline page Message-ID: I've noticed that the links on the Trac timeline page[1] use IP addresses instead of DNS names. The IP currently present in the links[2] times out. Pages can still be reached by replacing the IP address with the domain name: s/172\.29\.29\.21/sourceforge.net/, perhaps it is a retired IP address? Maciej [1] https://sourceforge.net/apps/trac/gar/timeline [2] https://172.29.29.21/apps/trac/gar/wiki/GarRepository?version=7 From ihsan at opencsw.org Wed Jun 17 10:09:25 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Wed, 17 Jun 2009 10:09:25 +0200 Subject: [csw-maintainers] libtool on build8st downgraded In-Reply-To: <5AB3EAB5-DBEC-4CA0-A6E0-45D155E821F8@opencsw.org> References: <4A388CC6.4090905@opencsw.org> <5AB3EAB5-DBEC-4CA0-A6E0-45D155E821F8@opencsw.org> Message-ID: <4A38A4B5.80106@opencsw.org> Dagobert Michelsen wrote: >> K?nntest du nicht auf build8st das alte Libtool installieren? > FYI: libtool on build8st has been downgraded temporarily to 1.5.24. > Thanks Dago. Ihsa -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Wed Jun 17 11:57:04 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 17 Jun 2009 11:57:04 +0200 Subject: [csw-maintainers] maintainer for gmp? In-Reply-To: <4A3816EC.3060103@opencsw.org> References: <4A3816EC.3060103@opencsw.org> Message-ID: <459B7BA3-8F64-4300-BE69-06C9F7D96E65@opencsw.org> Hi Mike, Am 17.06.2009 um 00:04 schrieb Mike Watters: > Is anyone working on gmp? > I am updating libmpfr, and would like to compile it 64bit. > if not, I will rebuild gmp I did some work on it in the past, but failed x86 assembly optimization. Apart from that the package should be usable. Please take it over and rebuild it :-) Best regards -- Dago From skayser at opencsw.org Wed Jun 17 12:08:19 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 17 Jun 2009 12:08:19 +0200 (CEST) Subject: [csw-maintainers] Broken links on Trac timeline page In-Reply-To: References: Message-ID: <57452.78.140.68.9.1245233299.squirrel@ssl.skayser.de> Maciej (Matchek) Blizinski wrote: > I've noticed that the links on the Trac timeline page[1] use IP > addresses instead of DNS names. The IP currently present in the > links[2] times out. Pages can still be reached by replacing the IP > address with the domain name: s/172\.29\.29\.21/sourceforge.net/, > perhaps it is a retired IP address? After briefly checking with the guys on #sourceforge i just created a ticket for the issue. https://sourceforge.net/apps/trac/sourceforge/ticket/1374 Sebastian From phil at bolthole.com Wed Jun 17 15:47:45 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 17 Jun 2009 06:47:45 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <4A3817F8.50202@opencsw.org> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> Message-ID: <20090617134745.GA46139@bolthole.com> On Wed, Jun 17, 2009 at 12:08:56AM +0200, Sebastian Kayser wrote: > Philip Brown wrote: > > Sometimes, it is DESIRABLE to have the configs globally in /opt/csw/etc. > > Does "desirable" go hand in hand with the core notion of this thread to > change the default $(sysconfdir) to /etc/opt/csw? Have all configuration > files in /etc/opt/csw eventually unless a maintainer decides otherwise > for his packages? "defaults" are usually set to be whatever is used most commonly. my non-quantified belief is that most of our packages that have config files, are/should have them in /etc/opt/csw If so, then the default should point to there. > And this leaves the user at the point that i mentioned initially. "Mhh, > there are configuration files in two places. Where is the one for > package foo located?". a good point. I'll update our User guide to mention to check both locations. From skayser at opencsw.org Wed Jun 17 20:12:58 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 17 Jun 2009 20:12:58 +0200 Subject: [csw-maintainers] Broken links on Trac timeline page In-Reply-To: <57452.78.140.68.9.1245233299.squirrel@ssl.skayser.de> References: <57452.78.140.68.9.1245233299.squirrel@ssl.skayser.de> Message-ID: <4A39322A.9020303@opencsw.org> Sebastian Kayser wrote: > Maciej (Matchek) Blizinski wrote: >> I've noticed that the links on the Trac timeline page[1] use IP >> addresses instead of DNS names. The IP currently present in the >> links[2] times out. Pages can still be reached by replacing the IP >> address with the domain name: s/172\.29\.29\.21/sourceforge.net/, >> perhaps it is a retired IP address? > > After briefly checking with the guys on #sourceforge i just created a > ticket for the issue. > > https://sourceforge.net/apps/trac/sourceforge/ticket/1374 Update: I haven't had any SF feedback on the ticket, but for me the links on the timeline page now contain a domain name again and are working fine. Thanks for the heads up Maciej. Sebastian From skayser at opencsw.org Thu Jun 18 01:45:28 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 18 Jun 2009 01:45:28 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? Message-ID: <4A398018.7050003@opencsw.org> Hi, our perl 5.8.8 contains Storable.pm 2.15 as a core module. Now i would like to package POE [1] which requires Storable.pm >= 2.16. What would be the options to deliver a Storable.pm >= 2.16 that POE can use? >From looking at @INC our /opt/csw/lib/perl/csw comes after the core module locations (which makes sense somehow), so i suppose it won't be as easy as just to package the current Storable.pm (2.20) from CPAN. Sebastian [1] http://search.cpan.org/dist/POE/ From bwalton at opencsw.org Thu Jun 18 01:59:19 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 17 Jun 2009 19:59:19 -0400 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <4A398018.7050003@opencsw.org> References: <4A398018.7050003@opencsw.org> Message-ID: <1245282863-sup-6225@ntdws12.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Wed Jun 17 19:45:28 -0400 2009: > our perl 5.8.8 contains Storable.pm 2.15 as a core module. Now i would > like to package POE [1] which requires Storable.pm >= 2.16. What would > be the options to deliver a Storable.pm >= 2.16 that POE can use? Woudl updating to 5.10.0 (the current stable) get you what you need? > >From looking at @INC our /opt/csw/lib/perl/csw comes after the core > module locations (which makes sense somehow), so i suppose it won't be > as easy as just to package the current Storable.pm (2.20) from CPAN. If you bundle the module with the package and force the package to use the file from your package (by munging @INC), you could likely stay with the current perl...this gets really ugly, really fast though. It's more work, but updating perl is likely the best option...Peter? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From skayser at opencsw.org Thu Jun 18 02:11:22 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 18 Jun 2009 02:11:22 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <1245282863-sup-6225@ntdws12.chass.utoronto.ca> References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> Message-ID: <4A39862A.5030309@opencsw.org> Ben Walton wrote: > Excerpts from Sebastian Kayser's message of Wed Jun 17 19:45:28 -0400 2009: >> our perl 5.8.8 contains Storable.pm 2.15 as a core module. Now i would >> like to package POE [1] which requires Storable.pm >= 2.16. What would >> be the options to deliver a Storable.pm >= 2.16 that POE can use? > > Woudl updating to 5.10.0 (the current stable) get you what you need? According to the POE dependency tree, yes. See http://deps.cpantesters.org/?module=POE;perl=latest Sebastian From mwatters at opencsw.org Thu Jun 18 03:59:29 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 17 Jun 2009 20:59:29 -0500 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <1245282863-sup-6225@ntdws12.chass.utoronto.ca> References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> Message-ID: <4A399F81.8010100@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Sebastian Kayser's message of Wed Jun 17 19:45:28 -0400 2009: >> our perl 5.8.8 contains Storable.pm 2.15 as a core module. Now i would >> like to package POE [1] which requires Storable.pm >= 2.16. What would >> be the options to deliver a Storable.pm >= 2.16 that POE can use? if moving to 5.10.0 is not possible. It is technically be possible to replace the Storable.pm in core with a "new" version. I don't know what the policy would be on this, but you would need a "pre-install" that does a backup of the existing files. and a post-remove that puts them all back -- only if the core perl package has "not" been removed. this may be more hassle then hacking @INC - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAko5n4EACgkQLrhmsXMSLxcJ9wCdGU3cCXfjx8YIGnFOP5Wq4Fvp ilIAoJnAfk5JMziF1RTJRg/ayUd1S2eV =sfVx -----END PGP SIGNATURE----- From mwatters at opencsw.org Thu Jun 18 04:48:41 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 17 Jun 2009 21:48:41 -0500 Subject: [csw-maintainers] maintainer for gmp? In-Reply-To: <459B7BA3-8F64-4300-BE69-06C9F7D96E65@opencsw.org> References: <4A3816EC.3060103@opencsw.org> <459B7BA3-8F64-4300-BE69-06C9F7D96E65@opencsw.org> Message-ID: <4A39AB09.9030604@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > I did some work on it in the past, but failed x86 assembly optimization. > Apart from that the package should be usable. > Please take it over and rebuild it :-) using Studio 11 I can't get it to compile on any platform ;-( it compiles fine using gcc4 currently looking through the forums for patches that may assist. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAko5qwkACgkQLrhmsXMSLxe/BQCeNnid01ukmhSVJst2JJYef9V1 mHYAn3AnE65dE4uF8Qgw/Ywax3p4rs+3 =3MAf -----END PGP SIGNATURE----- From bonivart at opencsw.org Thu Jun 18 09:22:27 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 18 Jun 2009 09:22:27 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <4A399F81.8010100@opencsw.org> References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> <4A399F81.8010100@opencsw.org> Message-ID: <625385e30906180022x60b998fcl8bd0660c02d4ac8d@mail.gmail.com> On Thu, Jun 18, 2009 at 3:59 AM, Mike Watters wrote: > It is technically be possible to replace the Storable.pm in core with a "new" > version. This is what I usually do. > I don't know what the policy would be on this, but you would need a > "pre-install" that does a backup of the existing files. and a post-remove that > puts them all back -- only if the core perl package has "not" been removed. Please explain this further, is this specifically for this module or in general? -- /peter From pfelecan at opencsw.org Thu Jun 18 10:26:49 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 18 Jun 2009 10:26:49 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <4A399F81.8010100@opencsw.org> (Mike Watters's message of "Wed\, 17 Jun 2009 20\:59\:29 -0500") References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> <4A399F81.8010100@opencsw.org> Message-ID: Mike Watters writes: > Ben Walton wrote: >> Excerpts from Sebastian Kayser's message of Wed Jun 17 19:45:28 -0400 2009: >>> our perl 5.8.8 contains Storable.pm 2.15 as a core module. Now i would >>> like to package POE [1] which requires Storable.pm >= 2.16. What would >>> be the options to deliver a Storable.pm >= 2.16 that POE can use? > > if moving to 5.10.0 is not possible. > > It is technically be possible to replace the Storable.pm in core with a "new" > version. > I don't know what the policy would be on this, but you would need a > "pre-install" that does a backup of the existing files. and a post-remove that > puts them all back -- only if the core perl package has "not" been removed. > > this may be more hassle then hacking @INC Maybe, but it's an ugly kludge doomed to fail if somebody out of our control needs the module in the current form. Subsidiary question, how comes that we are still with Perl 5.8.8? -- Peter From bonivart at opencsw.org Thu Jun 18 10:33:47 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 18 Jun 2009 10:33:47 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> <4A399F81.8010100@opencsw.org> Message-ID: <625385e30906180133w7cf717e2s98760085eae14d85@mail.gmail.com> On Thu, Jun 18, 2009 at 10:26 AM, Peter FELECAN wrote: > Subsidiary question, how comes that we are still with Perl 5.8.8? 1. I haven't had time to work on it. 2. Demand has been zero. -- /peter From dam at opencsw.org Thu Jun 18 15:04:42 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 18 Jun 2009 15:04:42 +0200 Subject: [csw-maintainers] maintainer for gmp? In-Reply-To: <4A39AB09.9030604@opencsw.org> References: <4A3816EC.3060103@opencsw.org> <459B7BA3-8F64-4300-BE69-06C9F7D96E65@opencsw.org> <4A39AB09.9030604@opencsw.org> Message-ID: Hi Mike, Am 18.06.2009 um 04:48 schrieb Mike Watters: >> I did some work on it in the past, but failed x86 assembly >> optimization. >> Apart from that the package should be usable. >> Please take it over and rebuild it :-) > > using Studio 11 I can't get it to compile on any platform ;-( > it compiles fine using gcc4 > > currently looking through the forums for patches that may assist. Ok, last time I tried it compiled nicely without assemblers optimization. Please note that Solaris 8 w/Studio 11 is best effort, but Solaris 9 w/Studio 12 is baseline. Best regards -- Dago From mwatters at opencsw.org Thu Jun 18 16:22:59 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 18 Jun 2009 09:22:59 -0500 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <625385e30906180022x60b998fcl8bd0660c02d4ac8d@mail.gmail.com> References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> <4A399F81.8010100@opencsw.org> <625385e30906180022x60b998fcl8bd0660c02d4ac8d@mail.gmail.com> Message-ID: <4A3A4DC3.4010402@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter Bonivart wrote: > On Thu, Jun 18, 2009 at 3:59 AM, Mike Watters wrote: >> It is technically be possible to replace the Storable.pm in core with a "new" >> version. > > This is what I usually do. > >> I don't know what the policy would be on this, but you would need a >> "pre-install" that does a backup of the existing files. and a post-remove that >> puts them all back -- only if the core perl package has "not" been removed. > > Please explain this further, is this specifically for this module or in general? > Just in general. when I replace a "core piece" of software with a newer version, I like to backup and restore the original if the new version is removed. as long as you know the install package names for all software involved it is fairly easily accomplished with installf/removef. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko6TcMACgkQLrhmsXMSLxdREgCfUP6M1JKfIaHPBokM9maRSVd0 CUIAniHVL8ClDh4LNTf+3NfslkDgoOqL =Dg9Y -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu Jun 18 16:32:23 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 18 Jun 2009 10:32:23 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090617134745.GA46139@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> Message-ID: <1245335357-sup-8252@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed Jun 17 09:47:45 -0400 2009: > On Wed, Jun 17, 2009 at 12:08:56AM +0200, Sebastian Kayser wrote: > > Philip Brown wrote: > > > Sometimes, it is DESIRABLE to have the configs globally in /opt/csw/etc. > > > > Does "desirable" go hand in hand with the core notion of this thread to > > change the default $(sysconfdir) to /etc/opt/csw? Have all configuration > > files in /etc/opt/csw eventually unless a maintainer decides otherwise > > for his packages? > > "defaults" are usually set to be whatever is used most commonly. > my non-quantified belief is that most of our packages that have config > files, are/should have them in /etc/opt/csw > If so, then the default should point to there. I'm combining threads here, since they are of overlapping subject matter. > On Mon, Jun 15, 2009 at 09:08:21PM +0200, rupert THURNER wrote: > > this does not mean that the default /opt/csw/etc goes away, isn't it? > > nope. ...but, if the GAR default is changed, it will gradually become less prominent. Lets have a real discussion about this, weighing pros and cons of the move from a default of /opt/csw/etc to a default of /etc/opt/csw. Pros: 1. A single /etc tree (eventually), and not including some of the sfw stuff, etc. 2. Still maintains a partition between system packages and csw packages that might overlap. Cons: 1. For people sharing out a single install (possibly per $arch) with NFS, /etc/opt/csw may make things more difficult, although I'd argue that symlinks could work around this quite easily. [For people in this situation, would creating symlinks for you?] 2. Potentially painful for maintainers during the transition. 3. User confusion during the transition, although I'd argue it's already a confusing situation. Personally, I really dislike having multiple etc directories...it feels wrong to me, but that is likely due to my Linux roots where the FHS has been in place for a long time and it isn't carrying some of the legacy aspects that make solaris attractive in other ways. I also dislike having /opt/csw/apache2/etc and would prefer /etc/opt/csw/apache2, for example. Should this discussion be taken to users@ to 'poll the audience'? I'm curious what others think. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Thu Jun 18 18:04:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 18 Jun 2009 11:04:51 -0500 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245335357-sup-8252@ntdws12.chass.utoronto.ca> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> Message-ID: <4A3A65A3.7000003@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Philip Brown's message of Wed Jun 17 09:47:45 -0400 2009: >> On Wed, Jun 17, 2009 at 12:08:56AM +0200, Sebastian Kayser wrote: >>> Philip Brown wrote: >>>> Sometimes, it is DESIRABLE to have the configs globally in /opt/csw/etc. >>> Does "desirable" go hand in hand with the core notion of this thread to >>> change the default $(sysconfdir) to /etc/opt/csw? Have all configuration >>> files in /etc/opt/csw eventually unless a maintainer decides otherwise >>> for his packages? >> "defaults" are usually set to be whatever is used most commonly. >> my non-quantified belief is that most of our packages that have config >> files, are/should have them in /etc/opt/csw >> If so, then the default should point to there. > > I'm combining threads here, since they are of overlapping subject matter. > > >> On Mon, Jun 15, 2009 at 09:08:21PM +0200, rupert THURNER wrote: >>> this does not mean that the default /opt/csw/etc goes away, isn't it? >> nope. > > ...but, if the GAR default is changed, it will gradually become less > prominent. Lets have a real discussion about this, weighing pros and > cons of the move from a default of /opt/csw/etc to a default of > /etc/opt/csw. > > Pros: > > 1. A single /etc tree (eventually), and not including some of the sfw > stuff, etc. a single etc tree can be accomplished with either directory > > 2. Still maintains a partition between system packages and csw > packages that might overlap. > > Cons: > > 1. For people sharing out a single install (possibly per $arch) with > NFS, /etc/opt/csw may make things more difficult, although I'd > argue that symlinks could work around this quite easily. [For > people in this situation, would creating symlinks for you?] symlinks may or may not work if you are trying to run on zones from a single installation on a global zone. > > 2. Potentially painful for maintainers during the transition. yes, but I wouldn't consider that a con per se > > 3. User confusion during the transition, although I'd argue it's > already a confusing situation. not just confusing for the users, confusing for the maintainers as well, at least for me. -- but that is not saying much ;-) > > Personally, I really dislike having multiple etc directories...it > feels wrong to me, but that is likely due to my Linux roots where the > FHS has been in place for a long time and it isn't carrying some of > the legacy aspects that make solaris attractive in other ways. I also > dislike having /opt/csw/apache2/etc and would prefer > /etc/opt/csw/apache2, for example. I agree with that, anything that has it's own etc tree should be consolidated to whatever "new standard" location is chosen. > > Should this discussion be taken to users@ to 'poll the audience'? I think it is a good Idea to take it to the users. ultimately they are the ones driving this project, without them our work is moot. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko6ZaIACgkQLrhmsXMSLxcV/gCffjgNpSNKCvBYOJZfpYXFHz61 CYsAoMDYLB/xMounpQG61d+D52sndjBu =ByWA -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu Jun 18 18:18:34 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 18 Jun 2009 12:18:34 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <4A3A65A3.7000003@opencsw.org> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <4A3A65A3.7000003@opencsw.org> Message-ID: <1245341535-sup-5196@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Thu Jun 18 12:04:51 -0400 2009: > > 1. A single /etc tree (eventually), and not including some of the sfw > > stuff, etc. > a single etc tree can be accomplished with either directory Not if you take single to mean 'all etc content' Ultimately, I'd like the CSW etc to be nested in the system /etc, which is my 'definition' of single. If you take single to mean single within CSW, then yes, you're right. > > 1. For people sharing out a single install (possibly per $arch) with > > NFS, /etc/opt/csw may make things more difficult, although I'd > > argue that symlinks could work around this quite easily. [For > > people in this situation, would creating symlinks for you?] > symlinks may or may not work if you are trying to run on zones from a single > installation on a global zone. Yes, zones certainly complicate things. As we are zone-free here, I'd appreciate others chiming in here. What is the best way to handle this wrt zones. Does the best option for zone support help or hinder the idea of nfs sharing as a goal? > > 2. Potentially painful for maintainers during the transition. > yes, but I wouldn't consider that a con per se Fair enough. Any change will see some short-term (or long, depending on package update cycle) issues, I guess. It does get tricky though in some cases. Hypothetically take the libxml and docbook packages. If docbook moved etc's first, it would still have to reference the old locations, since libxml is rooted in the old location. When libxml is updated, docbook would need to be updated too, to handle the new location. Shall we start a new thread on users@ then? Any objections? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 18 18:46:51 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 18 Jun 2009 09:46:51 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245335357-sup-8252@ntdws12.chass.utoronto.ca> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> Message-ID: <20090618164651.GG63121@bolthole.com> On Thu, Jun 18, 2009 at 10:32:23AM -0400, Ben Walton wrote: > I'm combining threads here, since they are of overlapping subject matter. Unfortunately, you quoted too much, and overly long emails tend to get zero replies. I'm going to try to shorten it a lot in the interests of more focused discussion:-) > 1. For people sharing out a single install (possibly per $arch) with > NFS, /etc/opt/csw may make things more difficult, although I'd > argue that symlinks could work around this quite easily. [For > people in this situation, would creating symlinks for you?] For people with a shared /opt (whether that be NFS or zones), the same issues still come into play. the issue of, "should configuration for software X, be unique to each box, or globally shared?" You dont see to address that type of issue at all. > Personally, I really dislike having multiple etc directories...it > feels wrong to me, but that is likely due to my Linux roots Or bsd. yes, i understand your background :-) Personally, I think /etc/opt/VENDOR makes a lot of sense. I also think that, in the rare cases global configs are desired (and it really is quite rare) /opt/csw/etc makes sense for those also. > I also > dislike having /opt/csw/apache2/etc and would prefer > /etc/opt/csw/apache2, for example. I think most everyone dislikes that. it is legacy configuration, and should really be changed to use /etc/opt/csw/apache2, in my opinion. > Should this discussion be taken to users@ to 'poll the audience'? I think we should have the discussion on the maintainers list first, to pre-air out any controversial issues. A fully public, large sized mailing list, is never a good place to have initial "exploratory" type technical discussions. Better to distill the core issues first. From mwatters at opencsw.org Thu Jun 18 19:18:30 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 18 Jun 2009 12:18:30 -0500 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090618164651.GG63121@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> Message-ID: <4A3A76E6.6060406@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > For people with a shared /opt (whether that be NFS or zones), the same > issues still come into play. the issue of, "should configuration for > software X, be unique to each box, or globally shared?" > > You dont see to address that type of issue at all. Actually, and correct me if I am wrong, I think "global vs unique" is a moot point if you have a shared /opt (NFS or zones.) I believe you will get a global config regardless of /opt/csw/etc or /etc/opt/csw. Example: shared /opt via NFS and linking /etc/opt/csw to /opt/csw/etc you get the same result as if you just put it in /opt/csw/etc. if you change it to suit one box, everything that has the NFS mount will get changed. the alternative is to instruct the user to copy the contents of /etc/opt/csw from the shared server to each NFS mount, but that IMHO defeats the purpose of a "shared" installation. if we want a unique etc for each box, I say we: * make the default /etc/opt/csw. * create a package "CSWcswshared" that has a script with a config, that will allow the user to pick and choose what packages he/she wants local configs for and copy those to /etc/opt/csw/{catalogname} -- skipping files that already exist. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko6duYACgkQLrhmsXMSLxc4DQCfeb4SVqCYoqwhD60IJ7GUyyf7 EVMAoJRhpv+b2JTMW6b737er59DVaEe6 =EiYA -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu Jun 18 19:21:41 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 18 Jun 2009 13:21:41 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090618164651.GG63121@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> Message-ID: <1245344654-sup-7349@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 18 12:46:51 -0400 2009: > For people with a shared /opt (whether that be NFS or zones), the same > issues still come into play. the issue of, "should configuration for > software X, be unique to each box, or globally shared?" > > You dont see to address that type of issue at all. I think that's driven by the software, not the packaging. While some things (eg: sudo) allow for a single, globally shared configuration and other things (eg: apache) can use includes to share a global template exist, many packages don't offer this flexibility outside of environment variables or other hacks. For the people doing nfs/zone sharing of /opt/csw, how are local configurations being handled presently? Something like apache is simple to handle, but other things aren't as easy. It seems like reconciling some of these things would only work if the software allowed for config1 -> global, config2 -> local (as Phil mentioned previously)...I haven't seen much of that though. Even if this were to be handled by some packages, I'd vote for /etc/opt/csw/$foo and /etc/opt/csw/global/$foo. If we have /etc/opt/csw at all, that means machine local or a second mount of some sort anyway, so why not go all the way? > Personally, I think /etc/opt/VENDOR makes a lot of sense. > I also think that, in the rare cases global configs are desired > (and it really is quite rare) /opt/csw/etc makes sense for those > also. Can you point out one of these rare cases so that we have a good example for discussion? I think sudo is (possibly) a good choice, but only because it _can_ be global, not because it typically is (at least in my experience). > I think we should have the discussion on the maintainers list first, > to pre-air out any controversial issues. A fully public, large > sized mailing list, is never a good place to have initial > "exploratory" type technical discussions. Better to distill the core > issues first. Ok. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 18 19:35:16 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 18 Jun 2009 10:35:16 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245344654-sup-7349@ntdws12.chass.utoronto.ca> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> Message-ID: <20090618173516.GA59068@bolthole.com> On Thu, Jun 18, 2009 at 01:21:41PM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Thu Jun 18 12:46:51 -0400 2009: > > For people with a shared /opt (whether that be NFS or zones), the same > > issues still come into play. the issue of, "should configuration for > > software X, be unique to each box, or globally shared?" > > ... > > I think that's driven by the software, not the packaging. i agree. > Even if this were to be handled by some packages, I'd vote for > /etc/opt/csw/$foo and /etc/opt/csw/global/$foo. anything in /etc, is by definition, NOT global :-} "global/shared" stuff, is under /opt/csw. Whether that be because it is NFS-shared, or lofs-shared across zones. > Can you point out one of these rare cases so that we have a good > example for discussion? I think sudo is (possibly) a good choice, but > only because it _can_ be global, not because it typically is (at least > in my experience). Funnily enough, at my place of work, the sudo config is indeed a global one across all machines. however, it is replicated out :-) I'm not sure of many "global-only config" examples. I can think of one or two good "global AND local" examples. For example... pkg-get :-) I think generally, examples of things that benefit from a global configuration, are things that are commonly used by users, and just dont work right without a detailed etc file. For example, netpbm? or xpdf? It is not likely that there will be any machine-local configuration tweaks needed. It is quite likely that a site-wide configuration is the best style of config, for those type of things. Or to put it another way, things that get some kind of conf file installed at pkgadd time, and the user will probably never bother to change, because the config is just so standard. Another example, might be mailcap files. From william at wbonnet.net Fri Jun 19 00:19:44 2009 From: william at wbonnet.net (William Bonnet) Date: Fri, 19 Jun 2009 00:19:44 +0200 Subject: [csw-maintainers] pixman 0.15.12 is in testing Message-ID: <4A3ABD80.1030700@wbonnet.net> Hi I have update pixman to the latest version. It is available from testing Feedback will be welcomed cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Fri Jun 19 01:02:39 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 19 Jun 2009 01:02:39 +0200 Subject: [csw-maintainers] Perl experts wanted: Possible to update Storable.pm (core module)? In-Reply-To: <625385e30906180133w7cf717e2s98760085eae14d85@mail.gmail.com> References: <4A398018.7050003@opencsw.org> <1245282863-sup-6225@ntdws12.chass.utoronto.ca> <4A399F81.8010100@opencsw.org> <625385e30906180133w7cf717e2s98760085eae14d85@mail.gmail.com> Message-ID: <4A3AC78F.9010502@opencsw.org> Peter Bonivart wrote: > On Thu, Jun 18, 2009 at 10:26 AM, Peter FELECAN wrote: >> Subsidiary question, how comes that we are still with Perl 5.8.8? > > 1. I haven't had time to work on it. > 2. Demand has been zero. Demand++ ;) Would a perl update require us to re-package modules that contain compiled .so s or would you expect any other major breakage? Sebastian From bwalton at opencsw.org Fri Jun 19 01:30:55 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 18 Jun 2009 19:30:55 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090618173516.GA59068@bolthole.com> References: <2E1730AF-9F21-4669-8C67-F1A31A101942@opencsw.org> <20090613200434.GA19773@bolthole.com> <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> Message-ID: <1245367441-sup-4715@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 18 13:35:16 -0400 2009: > > Even if this were to be handled by some packages, I'd vote for > > /etc/opt/csw/$foo and /etc/opt/csw/global/$foo. > > anything in /etc, is by definition, NOT global :-} "global/shared" > stuff, is under /opt/csw. Whether that be because it is NFS-shared, > or lofs-shared across zones. Ok, but if we're already in the position of straddling shared vs non-shared, which presumably sees the nfs/zones people need to make arrangements for sharing the things living in /etc already, why not go all the way? > Funnily enough, at my place of work, the sudo config is indeed a > global one across all machines. however, it is replicated out :-) We'd use cfengine for this task too. > I'm not sure of many "global-only config" examples. I can think of > one or two good "global AND local" examples. For example... pkg-get > :-) I think it's safe to say that not many software authors make this consideration...nor should they be expected to, imo. Maybe the emergence of zones changes the game some, but nfs sharing of software volumes must be on the decline by now? > Or to put it another way, things that get some kind of conf file > installed at pkgadd time, and the user will probably never bother to > change, because the config is just so standard. Again though, if we already have /etc/opt/csw in use, why not go all the way? I understand that things that never need to be fiddled with _could_ live in /opt/csw/etc, but since the local version exists and will cause problems (presumably) for the nfs/zones people, solving it for all packages should be the same work as solving it for half the packages. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 19 01:49:07 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 18 Jun 2009 16:49:07 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245367441-sup-4715@ntdws12.chass.utoronto.ca> References: <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> Message-ID: <20090618234907.GA52737@bolthole.com> On Thu, Jun 18, 2009 at 07:30:55PM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Thu Jun 18 13:35:16 -0400 2009: > > anything in /etc, is by definition, NOT global :-} "global/shared" > > stuff, is under /opt/csw. Whether that be because it is NFS-shared, > > or lofs-shared across zones. > > Ok, but if we're already in the position of straddling shared vs > non-shared, which presumably sees the nfs/zones people need to make > arrangements for sharing the things living in /etc already, why not go > all the way? you make this "go all the way" statement multiple times, but I dont understand what you mean by it. So I will condense and rephrase what i was saying previously: MOST configs, probably belong in /etc/opt/csw. However, there are a few configs, that rarely change, and are presumed to be standardized site-wide. For those things, there is no harm(and sometimes, an actual benefit) in having the configs live in /opt/csw/etc. Things like xpdf configs, and so on, that almost never change, and are presumed normally identical across all machines at a site. .......... Furthermore, in the even that a site administrator decides it is important to hack a local override for the config file, our packages can smoothly and easily handle it, if deployed in a clean manner. if /opt/csw/etc/someglobal.conf is usually copied in via our cswcpsampleconf class action, it will only be copied in or updated, if the site has not created something of their own there. So, if the site admin replaced it, with /opt/csw/etc/someglobal.conf -> /etc/opt/csw/someglobal.conf then that symlink would be left along, and what is normally a globally shared config file, becomes a machine-local config file, for that site only. From mwatters at opencsw.org Fri Jun 19 02:08:28 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 18 Jun 2009 19:08:28 -0500 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090618234907.GA52737@bolthole.com> References: <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> Message-ID: <4A3AD6FC.8030705@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > you make this "go all the way" statement multiple times, but I dont > understand what you mean by it. I have to agree with Ben here, and we should "go all the way" with using /etc/opt/csw. what "I" mean by "go all the way" is no longer use /opt/csw/etc or /opt/csw/$foo/etc and everything gets put in /etc/opt/csw or /etc/opt/csw/$foo > So, if the site admin replaced it, with > /opt/csw/etc/someglobal.conf -> /etc/opt/csw/someglobal.conf I am not sure that is the case. the way you have it linked above... if the admin links the version from /opt/csw/etc to /etc/opt/csw that link would be true for every host. if the /etc/opt/csw file does not exist on a particular host, the config would be non-existent. > then that symlink would be left along, and what is normally a globally > shared config file, becomes a machine-local config file, for that site > only. the link would be left, as well as a someglobal.conf.CSW file which the apps are not currently configured to use. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAko61vsACgkQLrhmsXMSLxcz4wCg0ZmdH0r239M7303r1GuaoM1J LDAAoN9sIpiM5zpnKr9u3+3fteCw4DoM =Hy1F -----END PGP SIGNATURE----- From bwalton at opencsw.org Fri Jun 19 02:40:15 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 18 Jun 2009 20:40:15 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090618234907.GA52737@bolthole.com> References: <1244925073-sup-2633@ntdws12.chass.utoronto.ca> <4A38013F.60500@opencsw.org> <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> Message-ID: <1245370024-sup-1205@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 18 19:49:07 -0400 2009: Ok, let me try to rephrase too. I agree with your points, but I still think a single etc is the way to go. People sharing /opt/csw are going to need to handle configuration differently than others with everything local to the machine (where local means not nfs shared and if in a zone, a non-sparse zone). If the config is in /opt/csw/etc and they need it customized per nfs client (or zone), they'll be symlinking it to somewhere on the box or mounting another directory onto /opt/csw/etc. This means that every box using packages from /opt/csw will be required to have this config maintained locally somehow. Also, because some packages installed on the host for /opt/csw have placed config files in /etc/opt/csw, they'll be symlinking the client /etc/opt/csw to some location in the share, rsyncing it from the host, or otherwise taking additional care to make sure all clients have the required files. IOW, they're making accommodations in _both_ directions, ignoring the sites that only shove a few gnu tools or something onto a share. Not to ignore your point about site standard, glacially changing files...For sites not sharing their /opt/csw, this is moot. It doesn't matter where they live, so it might as well be /etc/opt/csw. For sites that are sharing, I think (hope?) that I've demonstrated that they're already doing extra work anyway, so xpdf.conf and the like wouldn't add burden. Does that make more sense? I think we mostly agree, we're just not seeing eye-to-eye on small semantic points. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 19 02:47:22 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 18 Jun 2009 17:47:22 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <4A3AD6FC.8030705@opencsw.org> References: <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> Message-ID: <20090619004722.GA64424@bolthole.com> On Thu, Jun 18, 2009 at 07:08:28PM -0500, Mike Watters wrote: s> > So, if the site admin replaced it, with > > /opt/csw/etc/someglobal.conf -> /etc/opt/csw/someglobal.conf > > I am not sure that is the case. the way you have it linked above... > if the admin links the version from /opt/csw/etc to /etc/opt/csw > that link would be true for every host. yes, i was saying if the site admin decides, "no i want this app configured locally instead of globally", he still technically has that option, relatively easily. Lest anyone misunderstand me, i will reiterate that I am still completely in favour of the DEFAULT for demons and things, being "configs go in /etc/opt/csw". I am just trying to leave open the door, for certain applications to be allowed to have "global" configs in /opt/csw/etc, if it makese sense to do so. Something like xpdf, I think, it makes sense to do so. You can just NFS-share out /opt/csw everywhere, and it will work correctly. If on the other hand, it looks for configs in /etc/opt..... you share out /opt/csw... and then it breaks, until you go and have to copy in something to /etc/opt/csw on to **Every Single Machine*** you want to run it on. Not very large-scale-install friendly. From phil at bolthole.com Fri Jun 19 02:52:02 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 18 Jun 2009 17:52:02 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245370024-sup-1205@ntdws12.chass.utoronto.ca> References: <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <1245370024-sup-1205@ntdws12.chass.utoronto.ca> Message-ID: <20090619005202.GB64424@bolthole.com> On Thu, Jun 18, 2009 at 08:40:15PM -0400, Ben Walton wrote: > ... > People sharing /opt/csw are going to need to handle configuration > differently than others with everything local to the machine (where > local means not nfs shared and if in a zone, a non-sparse zone). > > If the config is in /opt/csw/etc and they need it customized per nfs > client (or zone), they'll be symlinking it to somewhere on the box or > mounting another directory onto /opt/csw/etc. There is a missing piece here. Site that share out /opt/csw, tend to do it for things that are easily sharable. things like xpdf **firefox** You want firefox to have all its default configs in /etc? Individually? For every machine? I dooont think sooo.... You want global configs for a beast like that. If global configs are then "ok" for something like that, then they should be allowed in certain other targetted situations also. > IOW, they're making accommodations in _both_ directions, ignoring the > sites that only shove a few gnu tools or something onto a share. btw, I dont think we should ignore "sites that only shove a few gnu tools onto a share" either. > Does that make more sense? I think we mostly agree, we're just not > seeing eye-to-eye on small semantic points. I think we're actually disagreeing on a rare-case, but conceptually large issue. Hopefully, the firefox example, shows the issue better ;-) From mwatters at opencsw.org Fri Jun 19 19:11:10 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 19 Jun 2009 12:11:10 -0500 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090619005202.GB64424@bolthole.com> References: <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <1245370024-sup-1205@ntdws12.chass.utoronto.ca> <20090619005202.GB64424@bolthole.com> Message-ID: <4A3BC6AE.9070102@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > There is a missing piece here. > Site that share out /opt/csw, tend to do it for things that are easily > sharable. > things like > xpdf > **firefox** > > You want firefox to have all its default configs in /etc? Individually? > For every machine? > > I dooont think sooo.... > You want global configs for a beast like that. Firefox is a different beast, and by default does not store anything in *etc* the global configs for firefox are stored in .*/share and the "local configs" are stored in the user's home directory under .mozilla I would say that firefox is not an issue in sharing /opt. Software like xpdf, I don't think it matters where the configs are located since they normally don't change. ~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~ the whole problem here, I believe, is how do we make CSW packages respect both local configs and global configs without hacking up every piece of software we package. or how do we tell the users to change their sharing schema to allow the same locations across the board. The easiest but least optimal solution I believe is to put everything in /opt/csw/ and link anything back to /etc that needs to be there... i.e. init scripts on solaris 8/9. for smf, /opt/csw/var/manifest and /opt/csw/var/methods This would solve the issue, although I reiterate this is not the best solution, but I am not sure there is a "best solution" with this problem. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAko7xq0ACgkQLrhmsXMSLxeZLwCgxd9a6cNvkYnLM468JFhKwl5C xjQAn3DPH7f7bHjzxSxFn9MY+zQ4kWGP =Ksfv -----END PGP SIGNATURE----- From bwalton at opencsw.org Fri Jun 19 19:20:44 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 19 Jun 2009 13:20:44 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090619004722.GA64424@bolthole.com> References: <20090616212424.GD51834@bolthole.com> <4A3817F8.50202@opencsw.org> <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> Message-ID: <1245431722-sup-3548@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 18 20:47:22 -0400 2009: > Lest anyone misunderstand me, i will reiterate that I am still > completely in favour of the DEFAULT for demons and things, being > "configs go in /etc/opt/csw". Understood. > You can just NFS-share out /opt/csw everywhere, and it will work > correctly. If on the other hand, it looks for configs in > /etc/opt..... you share out /opt/csw... and then it breaks, until > you go and have to copy in something to /etc/opt/csw on to **Every > Single Machine*** you want to run it on. > Not very large-scale-install friendly. Large-scale sites are exactly the same types of installations that would be using cfengine/puppet/etc to manage things like this. Do you agree that having two etc directories for an installation without nfs/zone sharing is pointless? Do you agree that if everything were in /etc/opt/csw and a site wanted to share their csw install, a move on the server of /etc/opt/csw to /opt/csw/etc and a symlink of /etc/opt/csw -> /opt/csw/etc on the all machines consuming this particular /opt/csw would handle things just fine? (Packages wanting/needing local customization could still symlink individual files to somewhere else as you noted.) I realize that Rupert raised concerns about this, due to restrictions in his environment, so it might be nice to get his thoughts (or the thoughts of others in a shared-csw setup) on this. I'm not in the position of having to deal with something like this. It would also be nice, but difficult to determine, how many sites actually use csw like this. 1%? 5%? More? Less? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 19 19:34:19 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 19 Jun 2009 10:34:19 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245431722-sup-3548@ntdws12.chass.utoronto.ca> References: <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> Message-ID: <20090619173419.GA77530@bolthole.com> On Fri, Jun 19, 2009 at 01:20:44PM -0400, Ben Walton wrote: >.... > Do you agree that having two etc directories for an installation > without nfs/zone sharing is pointless? Why even ask that question? So you can say, "well then, lets forget about sharing and optimize for the non-shared case"? > > Not very large-scale-install friendly. > > Large-scale sites are exactly the same types of installations that > would be using cfengine/puppet/etc to manage things like this. Some do, some dont. Also, using cfengine or puppet to handle this, would merely be a hack around a core problem that the package provider (ie: us) is not providing proper large-install aware packages in the first place! > Do you agree that if everything were in /etc/opt/csw and a site wanted > to share their csw install, a move on the server of /etc/opt/csw to > /opt/csw/etc and a symlink of /etc/opt/csw -> /opt/csw/etc on the all > machines consuming this particular /opt/csw would handle things just > fine? (Packages wanting/needing local customization could still > symlink individual files to somewhere else as you noted.) That is ugly. Besides, you may just as equally say, "everything should use /opt/csw/etc, and people who want local configs, should symlink /opt/csw/etc ->/etc/opt/csw" > It would also be nice, but difficult to determine, how many sites > actually use csw like this. 1%? 5%? More? Less? > I dont think this is a good question to base decisions on. Again, it has behind it, the mindset and assumption of, "well, most of our 'customers' are not large scale, so lets ignore what suits large scale". Forgive me if I'm wrong, but your questions seem to come from a basis of, [I dont care about large installs, i just want to do whats easiest/tidiest for me] is that accurate? From phil at bolthole.com Fri Jun 19 19:40:24 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 19 Jun 2009 10:40:24 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <4A3BC6AE.9070102@opencsw.org> References: <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <1245370024-sup-1205@ntdws12.chass.utoronto.ca> <20090619005202.GB64424@bolthole.com> <4A3BC6AE.9070102@opencsw.org> Message-ID: <20090619174024.GB77530@bolthole.com> On Fri, Jun 19, 2009 at 12:11:10PM -0500, Mike Watters wrote: > > You want firefox to have all its default configs in /etc? Individually? > > For every machine? > > > > I dooont think sooo.... > > You want global configs for a beast like that. > > Firefox is a different beast, and by default does not store anything in *etc* > the global configs for firefox are stored in .*/share and the "local configs" > are stored in the user's home directory under .mozilla but it could. And in some ways, it "should". It just doesnt at the moment :-) > > I would say that firefox is not an issue in sharing /opt. > > Software like xpdf, I don't think it matters where the configs are located > since they normally don't change. So, you seem to be agreeing with my premise, that SOME programs are fine in having their configs on the global /opt/csw, when they dont change. > the whole problem here, I believe, is how do we make CSW packages respect both > local configs and global configs without hacking up every piece of software we > package. We simply cant do that for *every* package. more below. > .... > This would solve the issue, although I reiterate this is not the best solution, > but I am not sure there is a "best solution" with this problem. I believe that is the core truth here. There is no "one size fits all 'best' solution". What we CAN do, is try to make the "best" choice, on a package-by-package basis. with certain guidelines. The most relevant ones being, "treat /opt/csw/* as read-only after install. If something is likely to need significant changing after initial install, then it belongs in /etc/opt/csw or /var/opt/csw" From bwalton at opencsw.org Fri Jun 19 20:03:38 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 19 Jun 2009 14:03:38 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090619173419.GA77530@bolthole.com> References: <20090617134745.GA46139@bolthole.com> <1245335357-sup-8252@ntdws12.chass.utoronto.ca> <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> <20090619173419.GA77530@bolthole.com> Message-ID: <1245433062-sup-2821@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 19 13:34:19 -0400 2009: > > Do you agree that having two etc directories for an installation > > without nfs/zone sharing is pointless? > > Why even ask that question? > So you can say, "well then, lets forget about sharing and optimize for the > non-shared case"? No, just a staring point. > > > Not very large-scale-install friendly. > > > > Large-scale sites are exactly the same types of installations that > > would be using cfengine/puppet/etc to manage things like this. > > Some do, some dont. > Also, using cfengine or puppet to handle this, would merely be a hack > around a core problem that the package provider (ie: us) is not providing > proper large-install aware packages in the first place! I disagree here. The problem isn't that things aren't packaged nicely for this situation, it's that software isn't written to support this situation. If most packages provided support for $localsysconf and $globalsysconf, sourcing global and then overlaying local, I'd say that maintaining 2 etc directories made perfect sense. Given that this certainly isn't the case, I don't think 2 etc's is good. > > Do you agree that if everything were in /etc/opt/csw and a site wanted > > to share their csw install, a move on the server of /etc/opt/csw to > > /opt/csw/etc and a symlink of /etc/opt/csw -> /opt/csw/etc on the all > > machines consuming this particular /opt/csw would handle things just > > fine? (Packages wanting/needing local customization could still > > symlink individual files to somewhere else as you noted.) > > That is ugly. Uglier than having etc directories all over the place? At best, I think we can say that there isn't an optimal solution for consumers of CSW in a setup like this. > Besides, you may just as equally say, "everything should use /opt/csw/etc, > and people who want local configs, should symlink /opt/csw/etc > ->/etc/opt/csw" But then we're still scattering etc directories around the system. To be honest, I'd prefer that all CSW etc be consolidated into a single location, even if the chosen location doesn't see it become a subset of the system etc tree. > > It would also be nice, but difficult to determine, how many sites > > actually use csw like this. 1%? 5%? More? Less? > > > > I dont think this is a good question to base decisions on. > Again, it has behind it, the mindset and assumption of, > "well, most of our 'customers' are not large scale, so lets ignore what > suits large scale". Here, you're making the assumption that all large scale installations would share /opt/csw. I suspect this isn't the case, although please correct me if you think otherwise. The point of this question is not to marginalize sites using CSW like this, but to simply keep in perspective this size of the audience that may or may not benefit from keeping 2 etc directories. > Forgive me if I'm wrong, but your questions seem to come from a basis of, > [I dont care about large installs, i just want to do whats easiest/tidiest > for me] No, I do care. While I think homing things under /etc/ is the better choice, I don't want to discriminate against sites that share /opt/csw. I'm trying to argue that if we've already arrived at the decision that /etc/opt/csw is the better _default_, then we're not going to be discriminating any further by moving everything there. > is that accurate? No. I think we'll end up agreeing to disagree here. I'm in favour of changing the default location, as are you. Lets just carry forward from this decision. We can always look at things when more packages are living in /etc/opt/csw anyway. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 19 22:55:51 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 19 Jun 2009 13:55:51 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245433062-sup-2821@ntdws12.chass.utoronto.ca> References: <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> <20090619173419.GA77530@bolthole.com> <1245433062-sup-2821@ntdws12.chass.utoronto.ca> Message-ID: <20090619205551.GC12157@bolthole.com> On Fri, Jun 19, 2009 at 02:03:38PM -0400, Ben Walton wrote: > I think we'll end up agreeing to disagree here. I'm in favour of > changing the default location, as are you. Lets just carry forward > from this decision. We can always look at things when more packages > are living in /etc/opt/csw anyway. ok, sounds like a working agreement. I think you are proposing, "lets focus on moving what is easily movable, from /opt/csw/etc, to /etc/opt/csw, and then review what is left after that point". I am very happy to move forward on that premise :-) From skayser at opencsw.org Sat Jun 20 02:00:46 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Sat, 20 Jun 2009 02:00:46 +0200 Subject: [csw-maintainers] Heads up: freenode has disabled mibbit access Message-ID: <4A3C26AE.1070601@opencsw.org> Hi, freenode has blocked access via mibbit starting yesterday [1], so don't be surprised when your mibbit sessions ends with "Terminated" soon after you hit "Connect". There is an alternative, a freenode-operated webchat based on qwebirc that can be reached at [2]. Phil, could you please adjust our main page. Sebastian [1]http://blog.freenode.net/2009/06/new-freenode-webchat-and-why-to-use-it/ [2]http://webchat.freenode.net/?channels=opencsw From dam at opencsw.org Sat Jun 20 10:37:37 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sat, 20 Jun 2009 10:37:37 +0200 Subject: [csw-maintainers] We'll always have Paris Message-ID: <42FADF04-9E67-43EC-B4F5-FAA5D9F3CAD0@opencsw.org> Hi, I'll be to a training in Paris next week. Maybe someone wants to have a beer in the evening? William? Best regards -- Dago From william at wbonnet.net Sat Jun 20 11:03:52 2009 From: william at wbonnet.net (William Bonnet) Date: Sat, 20 Jun 2009 11:03:52 +0200 Subject: [csw-maintainers] We'll always have Paris In-Reply-To: <42FADF04-9E67-43EC-B4F5-FAA5D9F3CAD0@opencsw.org> References: <42FADF04-9E67-43EC-B4F5-FAA5D9F3CAD0@opencsw.org> Message-ID: <4A3CA5F8.2040808@wbonnet.net> Hi > I'll be to a training in Paris next week. Maybe someone > wants to have a beer in the evening? William? Yes for sure :) Which days are you staying in Paris ? cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From ihsan at opencsw.org Sat Jun 20 12:56:47 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 20 Jun 2009 12:56:47 +0200 Subject: [csw-maintainers] Heads up: freenode has disabled mibbit access In-Reply-To: <4A3C26AE.1070601@opencsw.org> References: <4A3C26AE.1070601@opencsw.org> Message-ID: <4A3CC06F.10503@opencsw.org> Am 20.6.2009 2:00 Uhr, Sebastian Kayser schrieb: > freenode has blocked access via mibbit starting yesterday [1], so don't > be surprised when your mibbit sessions ends with "Terminated" soon after > you hit "Connect". There is an alternative, a freenode-operated webchat > based on qwebirc that can be reached at [2]. Phil, could you please > adjust our main page. I just updated the website. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sat Jun 20 12:56:48 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 20 Jun 2009 12:56:48 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp: Schedule & Topics Message-ID: <4A3CC070.306@opencsw.org> Hello, I've updated the Wiki with the schedule: http://wiki.opencsw.org/summercamp-2009 This is actually only a draft, we still can change it. @Trygve: Do you think the Schedule is realistic? I didn't know how much time we need from the hotel to the place where it takes place, so I've just wrote something. There are also a few topics that we couldn't discuss in Zurich. Even if it's mainly a technical meeting, it would be really nice if we could talk also about the topics that have been suggested for Zurich. --> http://wiki.opencsw.org/suggestions The topics for now are: - Stable collection - Release process - Tiered packages - Download stats - New website - License discussion and choice I would suggest, that we make a list with the topics that we have to discuss and set priorities. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sat Jun 20 20:48:42 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 20 Jun 2009 20:48:42 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp: Schedule & Topics In-Reply-To: <4A3CC070.306@opencsw.org> References: <4A3CC070.306@opencsw.org> Message-ID: <4A3D2F0A.2040008@opencsw.org> Am 20.6.2009 12:56 Uhr, Ihsan Dogan schrieb: > - Release process I was just pointed out that we have discussed the release process already and that it's already mentioned on the wiki: http://wiki.opencsw.org/release-process I would say, we remove this topic then. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sun Jun 21 15:06:52 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 21 Jun 2009 15:06:52 +0200 Subject: [csw-maintainers] /testing clamav 0.95.2 In-Reply-To: <625385e30906110613v59a347f1xd24f33ae77e2c406@mail.gmail.com> References: <625385e30906110613v59a347f1xd24f33ae77e2c406@mail.gmail.com> Message-ID: <4A3E306C.2050806@opencsw.org> Hello Peter, Am 11.6.2009 15:13 Uhr, Peter Bonivart schrieb: > A new version of Clam AntiVirus is in testing. > > http://mirror.opencsw.org/testing/clamav-0.95.2,REV=2009.06.11-SunOS5.8-sparc-CSW.pkg.gz > http://mirror.opencsw.org/testing/clamav-0.95.2,REV=2009.06.11-SunOS5.8-i386-CSW.pkg.gz > http://mirror.opencsw.org/testing/libclamav-0.95.2,REV=2009.06.11-SunOS5.8-sparc-CSW.pkg.gz > http://mirror.opencsw.org/testing/libclamav-0.95.2,REV=2009.06.11-SunOS5.8-i386-CSW.pkg.gz Thanks Peter, works fine. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From trygvis at opencsw.org Sun Jun 21 15:25:58 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 21 Jun 2009 15:25:58 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp: Schedule & Topics In-Reply-To: <4A3CC070.306@opencsw.org> References: <4A3CC070.306@opencsw.org> Message-ID: <4A3E34E6.2060400@opencsw.org> Ihsan Dogan wrote: > Hello, > > I've updated the Wiki with the schedule: > http://wiki.opencsw.org/summercamp-2009 > > This is actually only a draft, we still can change it. > > @Trygve: Do you think the Schedule is realistic? I didn't know how much > time we need from the hotel to the place where it takes place, so I've > just wrote something. I think that should be ok. The traffic planner say we use 32 minutes from the hotel to the (currently planned) location. -- Trygve From william at wbonnet.net Sun Jun 21 16:33:52 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 21 Jun 2009 16:33:52 +0200 Subject: [csw-maintainers] Using X11 from Sun against using our own version of X11 Message-ID: <4A3E44D0.1060406@wbonnet.net> Hi During the last weeks there was some discussion about reasons and advantages of using our own X11 packages against continuing to use Sun's X11 libs. The work of packaging our own version was started a few months ago when work on FF3 started. It pulled a lot of dependencies because of compilation issues with current libs (either outdated packages, or incompatible Sun's packages including X11). Here is some detailed information about the problem we are facing with using X11 from Sun. This looks like a good reason to start to use now our own X11 packages. I have started FF3 compilation on a freshly installed Solaris 9 box (thanks Dago ;) ), build9st. the conclusion i reached is that i cannot compile latest xrender on this box using Sun's X11. I encountered several problems. Here is the list of what i have done. Start with FF3 compilation -> it requires requires a recent cairo Recompile Cairo -> it requires renderdev Recompile our recent version of xrender -> configure fails, x11.pc is missing This is a know Solaris 9 know bug http://forums.sun.com/thread.jspa?threadID=5204312&tstart=0 http://bugs.opensolaris.org/view_bug.do?bug_id=6571762 The fix is to create by hand /usr/lib/pkgconfig/x11.pc. I did it and resumed compilation -> several *proto.pc are missing. Installation of xproto, kbproto and inputproto from our repository (also missing in Solaris 9) Recompile our recent version of xrender -> This time configure success -> Compilation fails "Picture.c", line 29: cannot find include file: I checked for this file and it is not present in Sun's packages (either Sol 8, 9 or 10 but it seems it is in opensolaris). This missing include file looks to me like a showstopper. Many recent software will use this include, and i have no guarantee it is the only missing file. Thus i can imagine porting effort will be heavy, and it is surely a good reason to use our own, recent version of (Xorg) X11 libs. cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From rupert at opencsw.org Sun Jun 21 20:04:22 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 21 Jun 2009 20:04:22 +0200 Subject: [csw-maintainers] recompiled mod_wsgi now in testing Message-ID: <6af4270906211104q5ffc4648tb2b9a6a3cda67d7@mail.gmail.com> recompiled mod_wsgi against python-2.6.2, is now in testing. rupert. From skayser at opencsw.org Sun Jun 21 22:06:24 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 21 Jun 2009 22:06:24 +0200 Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online Message-ID: <4A3E92C0.8050500@opencsw.org> Hi, to bring some more life to our IRC channel, i have extended an IRC bot framework with a couple of CSW-specific plugins today and hooked a bot named "ceeswie" up to #opencsw for testing purposes. You can either query her with commands or trigger her in the channel by prefixing @ to your commands. Give the following commands a try: @pkginfo trac @buglist trac @fsearch ssl.h There are also some straight-forward URL builders available (no validity checking) @url pkginfo trac @url gar trac @url bugs trac @url bug 3659 The help system (@help) is off-the-shelf and not yet adjusted. The bot is just a glimpse of what can be done. I had the idea recently, when someone came on the channel to request help with his trac installation that went bad on pkg upgrades. Instead of going off to the bug tracker to check for possible bugs (or referring the user there), the bot enables us to query for bugs right away in the channel. It's convenient and fun after all ;) Give it a try, let me know what you think. Expect the one or the other hickup, it's rather basic and doesn't yet handle all exceptions. If someone has prior experience in coding such bots and wants to roll an alternative one, feel free. Feature requests very welcome! Sebastian From maciej at opencsw.org Mon Jun 22 08:08:25 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Mon, 22 Jun 2009 07:08:25 +0100 Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online In-Reply-To: <4A3E92C0.8050500@opencsw.org> References: <4A3E92C0.8050500@opencsw.org> Message-ID: On Sun, Jun 21, 2009 at 9:06 PM, Sebastian Kayser wrote: > to bring some more life to our IRC channel, i have extended an IRC bot > framework Which one? > You can either query her with commands or trigger her in the channel by > prefixing @ to your commands. Give the following commands a try: > > ?@pkginfo trac > ?@buglist trac > ?@fsearch ssl.h I tried them - way cool! > Give it a try, let me know what you think. Expect the one or the other > hickup, it's rather basic and doesn't yet handle all exceptions. If > someone has prior experience in coding such bots and wants to roll an > alternative one, feel free. Feature requests very welcome! A suggestion: would you mind putting your code into the code repository? [1] A feature that comes to mind would be a wiki search: @wiki summer camp http://wiki.opencsw.org/summercamp-2009 - Summer Camp 2009 Maciej [1] http://opencsw.svn.sourceforge.net/viewvc/opencsw/ From dam at opencsw.org Mon Jun 22 08:54:14 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 22 Jun 2009 08:54:14 +0200 Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online In-Reply-To: References: <4A3E92C0.8050500@opencsw.org> Message-ID: <3E5D1D33-B325-47C9-9121-4030BC896ADB@opencsw.org> Hi Maciej, Am 22.06.2009 um 08:08 schrieb Maciej (Matchek) Blizinski: >> Give it a try, let me know what you think. Expect the one or the >> other >> hickup, it's rather basic and doesn't yet handle all exceptions. If >> someone has prior experience in coding such bots and wants to roll an >> alternative one, feel free. Feature requests very welcome! > > A suggestion: would you mind putting your code into the code > repository? [1] YOu can check it in here: http://opencsw.svn.sourceforge.net/viewvc/opencsw/ > A feature that comes to mind would be a wiki search: > > @wiki summer camp > http://wiki.opencsw.org/summercamp-2009 - Summer Camp 2009 and @bugs trac lists bugs, severity etc. Bets regards -- Dago From dam at opencsw.org Mon Jun 22 08:59:16 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 22 Jun 2009 08:59:16 +0200 Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online In-Reply-To: <3E5D1D33-B325-47C9-9121-4030BC896ADB@opencsw.org> References: <4A3E92C0.8050500@opencsw.org> <3E5D1D33-B325-47C9-9121-4030BC896ADB@opencsw.org> Message-ID: Hi, Am 22.06.2009 um 08:54 schrieb Dagobert Michelsen: > and > @bugs trac > lists bugs, severity etc. Ups, never mind. -- Dago From maciej at opencsw.org Mon Jun 22 12:42:08 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Mon, 22 Jun 2009 11:42:08 +0100 Subject: [csw-maintainers] Summer Camp 2009 (Please book your flights) In-Reply-To: <4A350506.1000804@opencsw.org> References: <4A350506.1000804@opencsw.org> Message-ID: On Sun, Jun 14, 2009 at 3:11 PM, Ihsan Dogan wrote: > Hello, > > As already mentioned, the date and the location is fixed. I've updated > the wiki as well, but the information is not complete. > > --> http://wiki.opencsw.org/summercamp-2009 > > The information for accommodation will be available soon. Due to a small mixup, I ended up with an extra, paid reservation for 1 person for 2 nights (Fri/Sat and Sat/Sun). I can't change or cancel it since it was a 'WebDeal' kind of reservation. The price was ?186, I'm willing to let it go for ?120. (I've called Thon and verified with them that it's okay to check a different person in.) Maciej From Darin.Perusich at cognigencorp.com Mon Jun 22 15:17:45 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Mon, 22 Jun 2009 09:17:45 -0400 Subject: [csw-maintainers] Solaris 9 system for test installs Message-ID: <4A3F8479.1000606@cognigencorp.com> Is there a Solaris 9 system in the build farm I can use to install a package? I received a a bug about installation on Sol9 and I don't have a Sol9 host anymore to test the installation. -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From skayser at opencsw.org Mon Jun 22 21:08:10 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 22 Jun 2009 21:08:10 +0200 (CEST) Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online In-Reply-To: References: <4A3E92C0.8050500@opencsw.org> Message-ID: <59133.194.246.122.22.1245697690.squirrel@ssl.skayser.de> Maciej (Matchek) Blizinski schrieb: > On Sun, Jun 21, 2009 at 9:06 PM, Sebastian Kayser > wrote: >> to bring some more life to our IRC channel, i have extended an IRC bot >> framework > > Which one? supybot, based on python. http://sourceforge.net/projects/supybot/ >> You can either query her with commands or trigger her in the channel by >> prefixing @ to your commands. Give the following commands a try: >> >> ??@pkginfo trac >> ??@buglist trac >> ??@fsearch ssl.h > > I tried them - way cool! :) >> Give it a try, let me know what you think. Expect the one or the other >> hickup, it's rather basic and doesn't yet handle all exceptions. If >> someone has prior experience in coding such bots and wants to roll an >> alternative one, feel free. Feature requests very welcome! > > A suggestion: would you mind putting your code into the code repository? > [1] Sure, will do so tomorrow. > A feature that comes to mind would be a wiki search: > > @wiki summer camp > http://wiki.opencsw.org/summercamp-2009 - Summer Camp 2009 Noted. Could also do that one for the GAR reference or maybe even a complete GAR wiki search. Thanks. Sebastian From maciej at opencsw.org Tue Jun 23 00:29:27 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Mon, 22 Jun 2009 23:29:27 +0100 Subject: [csw-maintainers] Summer Camp 2009 (Please book your flights) In-Reply-To: References: <4A350506.1000804@opencsw.org> Message-ID: On Mon, Jun 22, 2009 at 11:42 AM, Maciej (Matchek) Blizinski wrote: > Due to a small mixup, I ended up with an extra, paid reservation for 1 > person for 2 nights (Fri/Sat and Sat/Sun). This has been sorted out. Thanks! Maciej From bwalton at opencsw.org Tue Jun 23 01:56:19 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 22 Jun 2009 19:56:19 -0400 Subject: [csw-maintainers] Fwd: [bug-notifications] [mysql5 0003722]: Request for update to MySQL 5.1 Message-ID: <1245714826-sup-5508@ntdws12.chass.utoronto.ca> Hi All, I haven't had the time to work on mysql5 as I thought I might. The current GAR recipe builds and passes tests for i386 and sparc v8. It fails tests on both v9 and x86_64. I forget which is which now, but on one of the 64-bit platforms, the failed test is a simple 'example plugin' test. All others pass. The failures on the other platform are more serious. If anyone does have the time to take this, I'd appreciate it. If not, I'll see what I can do, but it likely won't be fast. Thanks -Ben --- Begin forwarded message from Mantis Bug Tracker --- From: Mantis Bug Tracker To: bug-notifications Date: Mon, 22 Jun 2009 16:46:07 -0400 Subject: [bug-notifications] [mysql5 0003722]: Request for update to MySQL 5.1 The following issue has been SUBMITTED. ====================================================================== http://www.opencsw.org/mantis/view.php?id=3722 ====================================================================== Reported By: bryanwgraham Assigned To: ====================================================================== Project: mysql5 Issue ID: 3722 Category: upgrade Reproducibility: always Severity: minor Priority: normal Status: new ====================================================================== Date Submitted: 2009-06-22 22:46 CEST Last Modified: 2009-06-22 22:46 CEST ====================================================================== Summary: Request for update to MySQL 5.1 Description: It's out. I'm told this is how you request a new version. ====================================================================== --- End forwarded message --- -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Tue Jun 23 04:29:43 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 22 Jun 2009 22:29:43 -0400 Subject: [csw-maintainers] Solaris 9 system for test installs In-Reply-To: <4A3F8479.1000606@cognigencorp.com> References: <4A3F8479.1000606@cognigencorp.com> Message-ID: <1245724130-sup-6322@ntdws12.chass.utoronto.ca> Excerpts from Darin Perusich's message of Mon Jun 22 09:17:45 -0400 2009: Hi Darin, > Is there a Solaris 9 system in the build farm I can use to install a > package? I received a a bug about installation on Sol9 and I don't have > a Sol9 host anymore to test the installation. I believe there is now a build9st box. You can either mail requests to buildfarm at opencsw or ask Dago for sudo access on that box. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Tue Jun 23 19:32:49 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 23 Jun 2009 12:32:49 -0500 Subject: [csw-maintainers] updating berkeleydb to 4.7.25 on build8st and build8xt Message-ID: <4A4111C1.3080602@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I am going to update to the latest berkeley DB package that Dago put in testing. on the solaris 8 test boxes. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpBEcEACgkQLrhmsXMSLxeLoACgx+nYguH0TpgKS1zCDOzVNywo IAgAnR57650z9joRHRwjJbWdnHc18b+I =Vs9c -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue Jun 23 19:34:28 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 23 Jun 2009 12:34:28 -0500 Subject: [csw-maintainers] updating berkeleydb to 4.7.25 on build8st and build8xt In-Reply-To: <4A4111C1.3080602@opencsw.org> References: <4A4111C1.3080602@opencsw.org> Message-ID: <4A411224.7060405@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > I am going to update to the latest berkeley DB package that Dago put in > testing. on the solaris 8 test boxes. > > Nevermind, someone beat me to it ... ;-) _______________________________________________ maintainers mailing list maintainers at lists.opencsw.org https://lists.opencsw.org/mailman/listinfo/maintainers - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpBEiQACgkQLrhmsXMSLxeS/gCfb4/3sXReuFe5mbtq25C7V7oQ FqEAoJ8omHt5XJuAfveMnxc0Wdr+sSft =eBia -----END PGP SIGNATURE----- From phil at bolthole.com Tue Jun 23 20:12:57 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 23 Jun 2009 11:12:57 -0700 Subject: [csw-maintainers] updating berkeleydb to 4.7.25 on build8st and build8xt In-Reply-To: <4A4111C1.3080602@opencsw.org> References: <4A4111C1.3080602@opencsw.org> Message-ID: <20090623181257.GA56726@bolthole.com> On Tue, Jun 23, 2009 at 12:32:49PM -0500, Mike Watters wrote: > > I am going to update to the latest berkeley DB package that Dago put in > testing. on the solaris 8 test boxes. becuase someone may not be aware that we even HAVE "test boxes", a clarification: we have test boxes, that are separate from our build boxes. the test boxes are build8st build8xt (this is a bit silly... I personally think they should more clearly be named, such as "test8s" and "test8x", but the above, is what they are named at the moment) From skayser at opencsw.org Tue Jun 23 22:17:44 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 23 Jun 2009 22:17:44 +0200 (CEST) Subject: [csw-maintainers] IRC-Bot for #opencsw ... prototype version online In-Reply-To: References: <4A3E92C0.8050500@opencsw.org> Message-ID: <56108.194.246.122.22.1245788264.squirrel@ssl.skayser.de> Maciej (Matchek) Blizinski wrote: > On Sun, Jun 21, 2009 at 9:06 PM, Sebastian Kayser > wrote: >> to bring some more life to our IRC channel, i have extended an IRC bot >> framework > > A suggestion: would you mind putting your code into the code repository? > [1] Done. Please do note: i am not a programmer and touched python for the first time last weekend. That being said, feel free to polish things up. Sebastian > [1] http://opencsw.svn.sourceforge.net/viewvc/opencsw/ From william at wbonnet.net Wed Jun 24 07:53:25 2009 From: william at wbonnet.net (William Bonnet) Date: Wed, 24 Jun 2009 07:53:25 +0200 Subject: [csw-maintainers] Thunderbird 2.0.0.22 is in testing Message-ID: <4A41BF55.9060808@wbonnet.net> Hi all I have updated Thunderbird package to the latest release. It is available from testing. Feedbacks are welcome cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From bonivart at opencsw.org Wed Jun 24 14:32:15 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 24 Jun 2009 14:32:15 +0200 Subject: [csw-maintainers] Minor web site change? Message-ID: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> On the web site we have this written: "CSW (Community SoftWare) is a free, volunteer effort, to provide quality downloadable binary packages for the currently fully supported versions of Sun Solaris (Currently, 8, 9, and 10)." Maybe we should rephrase that somehow now that Solaris 8 is not fully supported any more. -- /peter From pfelecan at opencsw.org Wed Jun 24 15:27:19 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 24 Jun 2009 15:27:19 +0200 Subject: [csw-maintainers] updating berkeleydb to 4.7.25 on build8st and build8xt In-Reply-To: <20090623181257.GA56726@bolthole.com> (Philip Brown's message of "Tue\, 23 Jun 2009 11\:12\:57 -0700") References: <4A4111C1.3080602@opencsw.org> <20090623181257.GA56726@bolthole.com> Message-ID: Philip Brown writes: > build8st > build8xt > > (this is a bit silly... I personally think they should more clearly be > named, such as "test8s" and "test8x", but the above, is what they are > named at the moment) +1 although they are really test build systems but I'm not proposing to name them testbuild8s, but wait, it's not so long... -- Peter From pfelecan at opencsw.org Wed Jun 24 15:28:50 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 24 Jun 2009 15:28:50 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> (Peter Bonivart's message of "Wed\, 24 Jun 2009 14\:32\:15 +0200") References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> Message-ID: Peter Bonivart writes: > On the web site we have this written: > > "CSW (Community SoftWare) is a free, volunteer effort, to provide > quality downloadable binary packages for the currently fully supported > versions of Sun Solaris (Currently, 8, 9, and 10)." > > Maybe we should rephrase that somehow now that Solaris 8 is not fully > supported any more. What about: ..."versions of Sun Solaris (when possible 8, but always 9, and 10)." -- Peter From bwalton at opencsw.org Wed Jun 24 15:49:05 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 24 Jun 2009 09:49:05 -0400 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> Message-ID: <1245851300-sup-4398@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Wed Jun 24 09:28:50 -0400 2009: > > "CSW (Community SoftWare) is a free, volunteer effort, to provide > > quality downloadable binary packages for the currently fully supported > > versions of Sun Solaris (Currently, 8, 9, and 10)." > > > ..."versions of Sun Solaris (when possible 8, but always 9, and 10)." Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From bonivart at opencsw.org Wed Jun 24 15:55:30 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 24 Jun 2009 15:55:30 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <1245851300-sup-4398@ntdws12.chass.utoronto.ca> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> Message-ID: <625385e30906240655t7278640br542b6ff0d0757afc@mail.gmail.com> On Wed, Jun 24, 2009 at 3:49 PM, Ben Walton wrote: > Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." "versions 9 and 10 of Sun Solaris (and 8, when possible) for both Sparc and x86." ??? -- /peter From bwalton at opencsw.org Wed Jun 24 16:05:10 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 24 Jun 2009 10:05:10 -0400 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <625385e30906240655t7278640br542b6ff0d0757afc@mail.gmail.com> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <625385e30906240655t7278640br542b6ff0d0757afc@mail.gmail.com> Message-ID: <1245852281-sup-2721@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Wed Jun 24 09:55:30 -0400 2009: > On Wed, Jun 24, 2009 at 3:49 PM, Ben Walton wrote: > > Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." > > "versions 9 and 10 of Sun Solaris (and 8, when possible) for both > Sparc and x86." +1 -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From pfelecan at opencsw.org Wed Jun 24 16:11:50 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 24 Jun 2009 16:11:50 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <625385e30906240655t7278640br542b6ff0d0757afc@mail.gmail.com> (Peter Bonivart's message of "Wed\, 24 Jun 2009 15\:55\:30 +0200") References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <625385e30906240655t7278640br542b6ff0d0757afc@mail.gmail.com> Message-ID: Peter Bonivart writes: > On Wed, Jun 24, 2009 at 3:49 PM, Ben Walton wrote: >> Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." > > "versions 9 and 10 of Sun Solaris (and 8, when possible) for both > Sparc and x86." +1 -- Peter From james at opencsw.org Wed Jun 24 16:30:22 2009 From: james at opencsw.org (James Lee) Date: Wed, 24 Jun 2009 14:30:22 GMT Subject: [csw-maintainers] Minor web site change? In-Reply-To: <1245851300-sup-4398@ntdws12.chass.utoronto.ca> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> Message-ID: <20090624.14302200.1757446011@gyor.oxdrove.co.uk> On 24/06/09, 14:49:05, Ben Walton wrote regarding Re: [csw-maintainers] Minor web site change?: > Excerpts from Peter FELECAN's message of Wed Jun 24 09:28:50 -0400 2009: > > > "CSW (Community SoftWare) is a free, volunteer effort, to provide > > > quality downloadable binary packages for the currently fully supported > > > versions of Sun Solaris (Currently, 8, 9, and 10)." > > > > > ..."versions of Sun Solaris (when possible 8, but always 9, and 10)." > Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." Except this isn't true, or at best it's misleading. Currently there are no 5.9 packages so 5.9 support is by 5.8 packages, thus 5.8 is as fully supported as 5.9 (a statement of truth about the present). The statement needs to distinguish the aims (CSW's commitment to support Sun supported systems now and in future) and the offerings (what is currently happening). James. From bonivart at opencsw.org Wed Jun 24 16:42:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 24 Jun 2009 16:42:14 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <20090624.14302200.1757446011@gyor.oxdrove.co.uk> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <20090624.14302200.1757446011@gyor.oxdrove.co.uk> Message-ID: <625385e30906240742u3e44cb91u6ee86fee4eebf6f@mail.gmail.com> On Wed, Jun 24, 2009 at 4:30 PM, James Lee wrote: > The statement needs to distinguish the aims (CSW's commitment to support > Sun supported systems now and in future) and the offerings (what is > currently happening). Any suggestions? :-) -- /peter From ihsan at dogan.ch Wed Jun 24 16:46:15 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Wed, 24 Jun 2009 16:46:15 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <1245851300-sup-4398@ntdws12.chass.utoronto.ca> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> Message-ID: <4A423C37.2010101@dogan.ch> Ben Walton wrote: >>> "CSW (Community SoftWare) is a free, volunteer effort, to provide >>> quality downloadable binary packages for the currently fully supported >>> versions of Sun Solaris (Currently, 8, 9, and 10)." >> ..."versions of Sun Solaris (when possible 8, but always 9, and 10)." > > Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." That might be confusing. I would not mention Solaris 8 together with 9 and 10. It should be clearly mentioned, that the Software might run on Solaris 8, but Solaris 9 is the lowest supported version. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Wed Jun 24 17:06:59 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 24 Jun 2009 11:06:59 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090619205551.GC12157@bolthole.com> References: <20090618164651.GG63121@bolthole.com> <1245344654-sup-7349@ntdws12.chass.utoronto.ca> <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> <20090619173419.GA77530@bolthole.com> <1245433062-sup-2821@ntdws12.chass.utoronto.ca> <20090619205551.GC12157@bolthole.com> Message-ID: <1245855797-sup-5350@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 19 16:55:51 -0400 2009: > ok, sounds like a working agreement. I think you are proposing, > "lets focus on moving what is easily movable, from /opt/csw/etc, to > /etc/opt/csw, and then review what is left after that point". > > I am very happy to move forward on that premise :-) Yes, lets do this. I think we should notify users@ of this upcoming change, though. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Wed Jun 24 18:38:55 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 24 Jun 2009 09:38:55 -0700 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <1245855797-sup-5350@ntdws12.chass.utoronto.ca> References: <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> <20090619173419.GA77530@bolthole.com> <1245433062-sup-2821@ntdws12.chass.utoronto.ca> <20090619205551.GC12157@bolthole.com> <1245855797-sup-5350@ntdws12.chass.utoronto.ca> Message-ID: <20090624163855.GC54688@bolthole.com> On Wed, Jun 24, 2009 at 11:06:59AM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Fri Jun 19 16:55:51 -0400 2009: > > ok, sounds like a working agreement. I think you are proposing, > > "lets focus on moving what is easily movable, from /opt/csw/etc, to > > /etc/opt/csw, and then review what is left after that point". > > > > I am very happy to move forward on that premise :-) > > Yes, lets do this. I think we should notify users@ of this upcoming > change, though. it's not a policy change. it would be bringing packages in line with existing guidelines, on a package-by-package basis. the maintainers of each affected package, should make the relevant announcement for their package, when the time comes From bwalton at opencsw.org Wed Jun 24 18:44:11 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 24 Jun 2009 12:44:11 -0400 Subject: [csw-maintainers] Adjusting $(DIRPATHS) for sparse zones support with shared /opt In-Reply-To: <20090624163855.GC54688@bolthole.com> References: <20090618173516.GA59068@bolthole.com> <1245367441-sup-4715@ntdws12.chass.utoronto.ca> <20090618234907.GA52737@bolthole.com> <4A3AD6FC.8030705@opencsw.org> <20090619004722.GA64424@bolthole.com> <1245431722-sup-3548@ntdws12.chass.utoronto.ca> <20090619173419.GA77530@bolthole.com> <1245433062-sup-2821@ntdws12.chass.utoronto.ca> <20090619205551.GC12157@bolthole.com> <1245855797-sup-5350@ntdws12.chass.utoronto.ca> <20090624163855.GC54688@bolthole.com> Message-ID: <1245861741-sup-1660@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed Jun 24 12:38:55 -0400 2009: > the maintainers of each affected package, should make the relevant > announcement for their package, when the time comes Ok, this works for me. I think it will actually have a higher impact, since it will be at 'time of delivery' instead of something that might be read and forgotten. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From maciej at opencsw.org Wed Jun 24 19:07:17 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Wed, 24 Jun 2009 18:07:17 +0100 Subject: [csw-maintainers] Debugging with Sun Studio and mGAR Message-ID: I'm currently working on debugging an issue with cups-polld. At the present stage, I'm about to learn how to build all the binaries with debugging symbols and run them under the control of Sun Studio. I'm guessing I'll be using a 'project with existing sources', pointing at the work/${package}-${version} directory and not calling 'gmake clean' not to nuke the compiled objects or binaries. Has anyone here done this kind of work before? If so, do you have any tips or know any gotchas about the process? Maciej From mwatters at opencsw.org Wed Jun 24 19:08:07 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 24 Jun 2009 12:08:07 -0500 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <4A423C37.2010101@dogan.ch> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> Message-ID: <4A425D77.3020103@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ihsan Dogan wrote: > Ben Walton wrote: > >>>> "CSW (Community SoftWare) is a free, volunteer effort, to provide >>>> quality downloadable binary packages for the currently fully supported >>>> versions of Sun Solaris (Currently, 8, 9, and 10)." >>> ..."versions of Sun Solaris (when possible 8, but always 9, and 10)." >> Or: "versions 9 and 10 of Sun Solaris (and 8, when possible)." > > That might be confusing. I would not mention Solaris 8 together with 9 > and 10. It should be clearly mentioned, that the Software might run on > Solaris 8, but Solaris 9 is the lowest supported version. > > > > > Ihsan > I know baseline is solaris 9 but solaris 8 is "best effort", but when will we drop the "best attempt"? I think we should leave everything as-is unless we are no longer going to create packages specifically for solaris 8. I suggest the following. as we roll out new versions of the packages, no longer attempt solaris 8 compile. compile for solaris 9+ only. Freeze the existing solaris 8 software catalog and keep for those users that want to run solaris 8 unsupported by sun. if the users want an upgrade to the software in the solaris 8 catalog, they can fill a bug report or request on the mailing list and we will apply the "best effort" to create that package and update the catalog. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpCXXcACgkQLrhmsXMSLxdf1gCfQ8C+1ZDHaj1H75HWlcmILsDc dC0An2CPwYnbzUQ0ZMpzhzzhMP+H86JY =On6N -----END PGP SIGNATURE----- From william at wbonnet.net Wed Jun 24 20:35:07 2009 From: william at wbonnet.net (William Bonnet) Date: Wed, 24 Jun 2009 20:35:07 +0200 Subject: [csw-maintainers] updating berkeleydb to 4.7.25 on build8st and build8xt In-Reply-To: References: <4A4111C1.3080602@opencsw.org> <20090623181257.GA56726@bolthole.com> Message-ID: <4A4271DB.1080409@wbonnet.net> Hi >> build8st >> build8xt >> >> (this is a bit silly... I personally think they should more clearly be >> named, such as "test8s" and "test8x", but the above, is what they are >> named at the moment) >> > > +1 although they are really test build systems but I'm not proposing to > name them testbuild8s, but wait, it's not so long... > I like test8s and test8x. Names and not ambiguous and shorter to type ;) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Wed Jun 24 21:36:43 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 24 Jun 2009 21:36:43 +0200 (CEST) Subject: [csw-maintainers] Debugging with Sun Studio and mGAR In-Reply-To: References: Message-ID: <54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de> Maciej (Matchek) Blizinski wrote: > I'm currently working on debugging an issue with cups-polld. At the > present stage, I'm about to learn how to build all the binaries with > debugging symbols and run them under the control of Sun Studio. I guess it doesn't fully cover your answer, but regarding GAR, you can set GARFLAVOR to DBG if you want your binaries built with debugging symbols. This will also skip the step where GAR strips all the binaries. See [1]. Of course this requires your program to honour CFLAGS correctly, because that is what GARFLAVOR changes. I used it to build packages that i wanted to debug with dbx later on (for example because i had seen crashes). Sebastian [1] http://sourceforge.net/apps/trac/gar/wiki/GAR%20Variable%20Reference#CompilationSettings From bwalton at opencsw.org Thu Jun 25 04:46:51 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 24 Jun 2009 22:46:51 -0400 Subject: [csw-maintainers] package hooks Message-ID: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> Hi All, I thought I'd get the discussion of hooks for the package tools going by posting a bit of code. This is completely untested, but is encapsulates the vision I have of how hooks could be implemented. [Please excuse my rusty perl.] The pkgutil patch implements pre/post install/remove hook capabilities that allow the admin (or other packages) to have code run at various points during the management of CSW packages. It does not differentiate between install and upgrade and it implements 'pre' and 'post' in the batch sense, not the per package sense. More granularity could be had by running hooks at each invocation of pkgadd/pkgrm, but I think that might be overkill. Thoughts? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: pkgutil_cswhooks.patch Type: application/octet-stream Size: 1823 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Thu Jun 25 09:01:17 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 09:01:17 +0200 Subject: [csw-maintainers] Debugging with Sun Studio and mGAR In-Reply-To: <54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de> References: <54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de> Message-ID: Hi Sebastian, Am 24.06.2009 um 21:36 schrieb Sebastian Kayser: > Maciej (Matchek) Blizinski wrote: >> I'm currently working on debugging an issue with cups-polld. At the >> present stage, I'm about to learn how to build all the binaries with >> debugging symbols and run them under the control of Sun Studio. > > I guess it doesn't fully cover your answer, but regarding GAR, you > can set > GARFLAVOR to DBG if you want your binaries built with debugging > symbols. > This will also skip the step where GAR strips all the binaries. See > [1]. Yes, and you can set it transienstly with gmake clean && GARFLAVOR=DBG gmake package Best regards -- Dago From dam at opencsw.org Thu Jun 25 09:03:39 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 09:03:39 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <4A425D77.3020103@opencsw.org> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> Message-ID: <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Hi Mike, Am 24.06.2009 um 19:08 schrieb Mike Watters: > I suggest the following. > as we roll out new versions of the packages, no longer attempt > solaris 8 > compile. compile for solaris 9+ only. Why? Usually it is no different packaging for Solaris 9 than for Solaris 8. Why drop Solaris 8 without the need? If there is *any* problem on Solaris 8 it is of course ok to just release for Solaris 9. Best regards -- Dago From bonivart at opencsw.org Thu Jun 25 09:54:32 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 25 Jun 2009 09:54:32 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <625385e30906250054x3cc31a8fqdd2c4278009d195b@mail.gmail.com> On Thu, Jun 25, 2009 at 9:03 AM, Dagobert Michelsen wrote: > Why? Usually it is no different packaging for Solaris 9 than for > Solaris 8. Why drop Solaris 8 without the need? If there is *any* > problem on Solaris 8 it is of course ok to just release for Solaris 9. We still need to rephrase the claim that Solaris 8 is fully supported by Sun. Maybe just remove the part about Sun..? "...provide quality downloadable binary packages for Sun Solaris 8, 9 and 10, both Sparc and x86." -- /peter From bonivart at opencsw.org Thu Jun 25 10:31:10 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 25 Jun 2009 10:31:10 +0200 Subject: [csw-maintainers] package hooks In-Reply-To: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> Message-ID: <625385e30906250131n4daa51car2a44bbaaf0f80e4c@mail.gmail.com> On Thu, Jun 25, 2009 at 4:46 AM, Ben Walton wrote: > I thought I'd get the discussion of hooks for the package tools going > by posting a bit of code. ?This is completely untested, but is > encapsulates the vision I have of how hooks could be implemented. > [Please excuse my rusty perl.] Looks nice, thanks for the patch! I'll add it to 1.7. :-) > The pkgutil patch implements pre/post install/remove hook capabilities > that allow the admin (or other packages) to have code run at various > points during the management of CSW packages. And these hooks will be provided by packages like etckeeper? -- /peter From james at opencsw.org Thu Jun 25 11:16:59 2009 From: james at opencsw.org (James Lee) Date: Thu, 25 Jun 2009 09:16:59 GMT Subject: [csw-maintainers] Minor web site change? In-Reply-To: <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <20090625.9165900.3503357047@gyor.oxdrove.co.uk> On 25/06/09, 08:03:39, Dagobert Michelsen wrote regarding Re: [csw-maintainers] Minor web site change?: > > I suggest the following. > > as we roll out new versions of the packages, no longer attempt > > solaris 8 > > compile. compile for solaris 9+ only. > Why? Usually it is no different packaging for Solaris 9 than for > Solaris 8. Why drop Solaris 8 without the need? Studio 12. > If there is *any* > problem on Solaris 8 it is of course ok to just release for Solaris 9. Except it's not. If a single useful package is missing it scuppers your whole installation. Due to the strong interdependence of the CSW system we can't realistically pick and choose and offer partial support. Further, we should remain egalitarian and not decide what is important to others, i.e. if just one package is missing we no longer support Solaris 8. James. From james at opencsw.org Thu Jun 25 11:17:00 2009 From: james at opencsw.org (James Lee) Date: Thu, 25 Jun 2009 09:17:00 GMT Subject: [csw-maintainers] Minor web site change? In-Reply-To: <625385e30906240742u3e44cb91u6ee86fee4eebf6f@mail.gmail.com> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <20090624.14302200.1757446011@gyor.oxdrove.co.uk> <625385e30906240742u3e44cb91u6ee86fee4eebf6f@mail.gmail.com> Message-ID: <20090625.9170000.106433457@gyor.oxdrove.co.uk> On 24/06/09, 15:42:14, Peter Bonivart wrote regarding Re: [csw-maintainers] Minor web site change?: > On Wed, Jun 24, 2009 at 4:30 PM, James Lee wrote: > > The statement needs to distinguish the aims (CSW's commitment to support > > Sun supported systems now and in future) and the offerings (what is > > currently happening). > Any suggestions? :-) CSW (Community SoftWare) is a free, volunteer effort, to provide quality downloadable Sparc and x86 binary packages for Sun Solaris. CSW is committed to supporting version of Solaris fully supported by Sun (at present 9 and 10) [1]. CSW's packages currently also work on Solaris 8 however users are urged to migrate to Solaris 9 or 10 as the Solaris 8 packages may be dropped soon. [2] [1] http://www.sun.com/software/solaris/releases.jsp (or similar) [2] Reference/link to time table and methodologies for announcements. James. From glaw at opencsw.org Thu Jun 25 11:39:45 2009 From: glaw at opencsw.org (Gary Law) Date: Thu, 25 Jun 2009 10:39:45 +0100 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: 2009/6/25 Dagobert Michelsen : > Why? Usually it is no different packaging for Solaris 9 than for > Solaris 8. Why drop Solaris 8 without the need? If there is *any* > problem on Solaris 8 it is of course ok to just release for Solaris 9. The problem for me is testing. Rolling out something that compiles is all very well and good, but if it's not tested on Sol8 then it might be buggy, and might in fact be worse than the old Sol8 version we have in the archive. Finding the hardware/virtualisation kit for testing for Sol8 is hard enough now, and it's only going to get worse. Gary -- Gary Law glaw at opencsw.org From glaw at opencsw.org Thu Jun 25 11:40:35 2009 From: glaw at opencsw.org (Gary Law) Date: Thu, 25 Jun 2009 10:40:35 +0100 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <20090625.9170000.106433457@gyor.oxdrove.co.uk> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <20090624.14302200.1757446011@gyor.oxdrove.co.uk> <625385e30906240742u3e44cb91u6ee86fee4eebf6f@mail.gmail.com> <20090625.9170000.106433457@gyor.oxdrove.co.uk> Message-ID: 2009/6/25 James Lee : > CSW (Community SoftWare) is a free, volunteer effort, to provide > quality downloadable Sparc and x86 binary packages for Sun Solaris. > > CSW is committed to supporting version of Solaris fully supported > by Sun (at present 9 and 10) [1]. ?CSW's packages currently also work > on Solaris 8 however users are urged to migrate to Solaris 9 or 10 > as the Solaris 8 packages may be dropped soon. [2] > > [1] http://www.sun.com/software/solaris/releases.jsp > (or similar) > > [2] Reference/link to time table and methodologies for announcements. +1 -- Gary Law glaw at opencsw.org From dam at opencsw.org Thu Jun 25 11:47:05 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 11:47:05 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: Hi Gary, Am 25.06.2009 um 11:39 schrieb Gary Law: > 2009/6/25 Dagobert Michelsen : >> Why? Usually it is no different packaging for Solaris 9 than for >> Solaris 8. Why drop Solaris 8 without the need? If there is *any* >> problem on Solaris 8 it is of course ok to just release for Solaris >> 9. > > The problem for me is testing. Rolling out something that compiles is > all very well and good, but if it's not tested on Sol8 then it might > be buggy, and might in fact be worse than the old Sol8 version we have > in the archive. Finding the hardware/virtualisation kit for testing > for Sol8 is hard enough now, and it's only going to get worse. This will be mitigated when the ESX farm has been upgraded to vSphere 4. It should then be possible to assign dedicated package install testing machines for all operating systems. The other reasons however stay very valid. Best regards -- Dago From pfelecan at opencsw.org Thu Jun 25 14:12:24 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 25 Jun 2009 14:12:24 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <625385e30906250054x3cc31a8fqdd2c4278009d195b@mail.gmail.com> (Peter Bonivart's message of "Thu\, 25 Jun 2009 09\:54\:32 +0200") References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> <625385e30906250054x3cc31a8fqdd2c4278009d195b@mail.gmail.com> Message-ID: Peter Bonivart writes: > On Thu, Jun 25, 2009 at 9:03 AM, Dagobert Michelsen wrote: >> Why? Usually it is no different packaging for Solaris 9 than for >> Solaris 8. Why drop Solaris 8 without the need? If there is *any* >> problem on Solaris 8 it is of course ok to just release for Solaris 9. > > We still need to rephrase the claim that Solaris 8 is fully supported > by Sun. Maybe just remove the part about Sun..? > > "...provide quality downloadable binary packages for Sun Solaris 8, 9 > and 10, both Sparc and x86." We just agreed on a rephrasing which seems alright: "versions 9 and 10 of Sun Solaris (and 8, when possible) for both Sparc and x86." What's the issue with that? -- Peter From bwalton at opencsw.org Thu Jun 25 15:18:35 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 09:18:35 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <625385e30906250131n4daa51car2a44bbaaf0f80e4c@mail.gmail.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <625385e30906250131n4daa51car2a44bbaaf0f80e4c@mail.gmail.com> Message-ID: <1245934781-sup-9640@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Thu Jun 25 04:31:10 -0400 2009: > On Thu, Jun 25, 2009 at 4:46 AM, Ben Walton wrote: > > I thought I'd get the discussion of hooks for the package tools going > > by posting a bit of code. ?This is completely untested, but is > > encapsulates the vision I have of how hooks could be implemented. > > [Please excuse my rusty perl.] > > Looks nice, thanks for the patch! I'll add it to 1.7. :-) Cool! :) Are there points of discussion before it gets 'adopted' though? Should install be differentiated from upgrade? Should the hooks run per-package? Are there any other useful points that should be hooked? Is the path I picked (/etc/opt/csw/pkg-hooks) acceptable? Since this system would (presumably) be implemented by any csw utility that is going to add/remove packages, we should all be happy with the initial choices up front...not to say we couldn't change things later, but smart choices are definitely better. > > The pkgutil patch implements pre/post install/remove hook capabilities > > that allow the admin (or other packages) to have code run at various > > points during the management of CSW packages. > > And these hooks will be provided by packages like etckeeper? Etckeeper will likely be the first to leverage the hooks. On my Ubuntu and RHEL systems, there are various other things that hook into the provided plugin systems. With RHEL5, Redhat moved away from their up2date system and simply provides a yum plugin (rhnplugin) that hooks into the facilities provided by yum. There is also a downloadonly yum plugin that I've installed. Things like the Ubuntu update-notifier use the hooks to maintain a timestamp so that if you've just done a manual package update, it won't bother checking again, etc. As long as the system is flexible enough, there are lots of creative things that can be done. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pfelecan at opencsw.org Thu Jun 25 15:26:45 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 25 Jun 2009 15:26:45 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> (Dagobert Michelsen's message of "Thu\, 25 Jun 2009 09\:03\:39 +0200") References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: Dagobert Michelsen writes: > Hi Mike, > > Am 24.06.2009 um 19:08 schrieb Mike Watters: >> I suggest the following. >> as we roll out new versions of the packages, no longer attempt >> solaris 8 >> compile. compile for solaris 9+ only. > > Why? Usually it is no different packaging for Solaris 9 than for > Solaris 8. Why drop Solaris 8 without the need? If there is *any* > problem on Solaris 8 it is of course ok to just release for Solaris 9. This can be true for pre Solaris 8 packages also. In my opinion it's a question of policy. More than 1 year ago we had this discussion, in a more extreme conditions, and the promise was made that when Solaris 8 is no more supported by SUN we drop also the support. Let it be known that I'm really reluctant to use the SUN compiler until we cannot use the last version of it, i.e. SUN Studio 12. May I suggest to have a ballot on this issue? -- Peter From pfelecan at opencsw.org Thu Jun 25 15:31:54 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 25 Jun 2009 15:31:54 +0200 Subject: [csw-maintainers] issue with mail sent by mantis Message-ID: The mail delivered by mantis has a systematic issue, detected by Amavis and others: X-Virus-Scanned: amavisd-new at opencsw.org X-Amavis-Alert: BAD HEADER SECTION, Duplicate header field: "Message-ID" Can this be corrected? -- Peter From dam at opencsw.org Thu Jun 25 15:38:13 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 15:38:13 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <5954B1F4-D33A-4B36-A5B4-3CBFA484F488@opencsw.org> Hi Peter, Am 25.06.2009 um 15:26 schrieb Peter FELECAN: > This can be true for pre Solaris 8 packages also. In my opinion it's a > question of policy. More than 1 year ago we had this discussion, in a > more extreme conditions, and the promise was made that when Solaris > 8 is > no more supported by SUN we drop also the support. Let it be known > that > I'm really reluctant to use the SUN compiler until we cannot use the > last version of it, i.e. SUN Studio 12. May I suggest to have a ballot > on this issue? No need to. If you don't want to package for Solaris 8 it is perfectly ok to start with Solaris 9 and Sun Studio 12. I was just saying that it is still ok to package for Solaris 8 and the catalog will still receive updates (but not all). Best regards -- Dago PS: Are you in for the beer tonight with William? From mwatters at opencsw.org Thu Jun 25 15:39:24 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 25 Jun 2009 08:39:24 -0500 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <4A437E0C.4070101@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter FELECAN wrote: > Dagobert Michelsen writes: > >> Hi Mike, >> >> Am 24.06.2009 um 19:08 schrieb Mike Watters: >>> I suggest the following. >>> as we roll out new versions of the packages, no longer attempt >>> solaris 8 >>> compile. compile for solaris 9+ only. >> Why? Usually it is no different packaging for Solaris 9 than for >> Solaris 8. Why drop Solaris 8 without the need? If there is *any* >> problem on Solaris 8 it is of course ok to just release for Solaris 9. > > This can be true for pre Solaris 8 packages also. In my opinion it's a > question of policy. More than 1 year ago we had this discussion, in a > more extreme conditions, and the promise was made that when Solaris 8 is > no more supported by SUN we drop also the support. Let it be known that > I'm really reluctant to use the SUN compiler until we cannot use the > last version of it, i.e. SUN Studio 12. May I suggest to have a ballot > on this issue? I prefer Studio 12, Sun made enhancements to that version to make it more gnu friendly. gnu apps that would not compile using studio 11 now have a great chance of compiling and running with studio 12. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpDfgsACgkQLrhmsXMSLxfs3gCgqEHWt+U/4KM2GLCSjHnhWgDH AjMAn1CvfkOwALgTLIliQJM6+99EDsEu =YLZE -----END PGP SIGNATURE----- From dam at opencsw.org Thu Jun 25 15:39:49 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 15:39:49 +0200 Subject: [csw-maintainers] package hooks In-Reply-To: <1245934781-sup-9640@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <625385e30906250131n4daa51car2a44bbaaf0f80e4c@mail.gmail.com> <1245934781-sup-9640@ntdws12.chass.utoronto.ca> Message-ID: <617118FB-4D91-4691-B4CE-0F19B5C97A2C@opencsw.org> Hi Ben, Am 25.06.2009 um 15:18 schrieb Ben Walton: > Are there points of discussion before it gets 'adopted' though? > Should install be differentiated from upgrade? Should the hooks run > per-package? Are there any other useful points that should be hooked? I guess we should talk about arguments passed, like the package name of the package currently modified. Additionally we should have a package "update hook" (useful for nginx, which allows uninterrupted update) and a "package purge hook" which brings the system to a pristine state before package installation. Best regards -- Dago From dam at opencsw.org Thu Jun 25 15:41:55 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 25 Jun 2009 15:41:55 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <4A437E0C.4070101@opencsw.org> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> <4A437E0C.4070101@opencsw.org> Message-ID: Hi Mike, Am 25.06.2009 um 15:39 schrieb Mike Watters: > I prefer Studio 12, Sun made enhancements to that version to make it > more gnu > friendly. gnu apps that would not compile using studio 11 now have > a great > chance of compiling and running with studio 12. Again, this is perfectly ok. All I'm asking for is to try Solaris 8 and Studio 11. If there is any issue, just change the one line in GAR for the compiler change, login to build9 and retry. BTW, should I change the default compiler for Solaris 9 and 10 to Studio 12? Best regards -- Dago From bwalton at opencsw.org Thu Jun 25 15:43:08 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 09:43:08 -0400 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <1245936905-sup-7739@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Thu Jun 25 09:26:45 -0400 2009: > last version of it, i.e. SUN Studio 12. May I suggest to have a ballot > on this issue? I think a vote is in order too. Personally, I'll hate to lose sol8 support as I'm still running a few 8 boxes that depend on CSW. That being said, I think I'd still vote to drop the support, since keeping it allows people to continue hobbling along on the same crutch forever. Losing official support from CSW does not preclude individual sites building packages against the last released packages for 8 either, so just like running an unsupported OS, people can continue to run an unsupported CSW. Maybe somebody will spawn a CSW Legacy project ala the now defunct Fedora Legacy project. [I'm hoping to be completely moved away from 8 by the end of the summer, but we'll see.] Just my $0.02 CDN. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 25 16:17:11 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 25 Jun 2009 07:17:11 -0700 Subject: [csw-maintainers] package hooks In-Reply-To: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> Message-ID: <20090625141711.GA25001@bolthole.com> On Wed, Jun 24, 2009 at 10:46:51PM -0400, Ben Walton wrote: > > Hi All, > > I thought I'd get the discussion of hooks for the package tools going > by posting a bit of code. This is completely untested, but is > encapsulates the vision I have of how hooks could be implemented. > [Please excuse my rusty perl.] its always nice to have someone willing to code. However, this is an API we're talking about. It's usually best for an API to be driven by a specification, not just "well look at the code". So.. how about writing up some docs for the API now? :-) (I sugest putting it on the wiki) From bwalton at opencsw.org Thu Jun 25 16:31:39 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 10:31:39 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <617118FB-4D91-4691-B4CE-0F19B5C97A2C@opencsw.org> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <625385e30906250131n4daa51car2a44bbaaf0f80e4c@mail.gmail.com> <1245934781-sup-9640@ntdws12.chass.utoronto.ca> <617118FB-4D91-4691-B4CE-0F19B5C97A2C@opencsw.org> Message-ID: <1245939157-sup-8479@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Thu Jun 25 09:39:49 -0400 2009: Hi Dago, > I guess we should talk about arguments passed, like the package name > of the package currently modified. Additionally we should have a Ok, so you think the hooks should be run per-package instead of per-batch then? This is heavier, but more granular. On a more macro level, the package info could be provided to hooks through a defined 'protocol' whereby the tool would create a set of files (/var/opt/csw/lib/pkg-hooks/{installs,removes,upgrades}) that would be available to the hooks. As long as these files were created prior to the batch call, the hooks could use this as required. If all hooks are made per-package instead of per-batch, the protocol could simply be a single argument passed to the hook. That's much easier from an implementation perspective simply because there are no filesystem objects to look after. This brings up the issue of hook exit status though. The patch I posted will abort the whole (batch) operation if any hook exits non-zero. Should per-package hooks have the ability to do the same? Should the batch hooks have this ability? My thinking is yes to both, but I'm interested in alternate views on this. > package "update hook" (useful for nginx, which allows uninterrupted > update) and a "package purge hook" which brings the system to a > pristine state before package installation. Ok, so I'm taking this as: If a package is being upgraded (pkgrm followed immediately by pkgadd) that a preupdate and postupdate hook should be called? If a package is being removed (without a new version being added) that there should be a hook called after pkgrm named purge? This would allow for batch hooks as well as per-package hooks. In this case then, should there be a pre/post upgrade batch hook too? This is something I wasn't sure of...since the underlying tools don't have any notion of an upgrade, our distinction would be mostly artificial...not to say useless though. This is exactly the type of discussion I was hoping for. It's good to define the needs up front. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Thu Jun 25 16:36:05 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 10:36:05 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <20090625141711.GA25001@bolthole.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> Message-ID: <1245940343-sup-1963@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 25 10:17:11 -0400 2009: > its always nice to have someone willing to code. > However, this is an API we're talking about. It's usually best for an API > to be driven by a specification, not just "well look at the code". I agree. It was just a good way to grab eyeballs and start the discussion. I wasn't expecting Peter to grab it so quickly, which is why I prefaced it with 'completely untested...' > So.. how about writing up some docs for the API now? :-) I'm willing to do this, but I wanted to make sure it was a community discussion, since it will be a fundamental portion of the package tools. I'll start the wiki document and modify as the discussion goes. When we're happy, I'll update the patch. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 25 17:36:52 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 25 Jun 2009 08:36:52 -0700 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> Message-ID: <20090625153652.GC13603@bolthole.com> On Thu, Jun 25, 2009 at 10:39:45AM +0100, Gary Law wrote: > 2009/6/25 Dagobert Michelsen : > > Why? Usually it is no different packaging for Solaris 9 than for > > Solaris 8. Why drop Solaris 8 without the need? If there is *any* > > problem on Solaris 8 it is of course ok to just release for Solaris 9. > > The problem for me is testing. Rolling out something that compiles is > all very well and good, but if it's not tested on Sol8 then it might > be buggy, and might in fact be worse than the old Sol8 version we have > in the archive. Finding the hardware/virtualisation kit for testing > for Sol8 is hard enough now, and it's only going to get worse. for people who run 8, even having "untested packages" is better than "having no packages". From maciej at opencsw.org Thu Jun 25 17:43:41 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Thu, 25 Jun 2009 16:43:41 +0100 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <20090625153652.GC13603@bolthole.com> References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> <20090625153652.GC13603@bolthole.com> Message-ID: On Thu, Jun 25, 2009 at 4:36 PM, Philip Brown wrote: > for people who run 8, even having "untested packages" is better than > "having no packages". Perhaps having separate levels of 'testedness' for each Solaris major version would bring more clarity? "This package has been tested on $(TESTED_VERSIONS)." Maciej From maciej at opencsw.org Thu Jun 25 17:58:08 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Thu, 25 Jun 2009 16:58:08 +0100 Subject: [csw-maintainers] gmake hangs on the buildfarm Message-ID: maciej at build10s [build10s]:~/src/cups/trunk > gmake -d package GNU Make 3.81 Copyright (C) 2006 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. This program built for sparc-sun-solaris2.8 Reading makefiles... Reading makefile `Makefile'... Reading makefile `gar/category.mk' (search path) (no ~ expansion)... Reading makefile `gar/categories/net/category.mk' (search path) (no ~ expansion)... Reading makefile `gar/gar.mk' (search path) (no ~ expansion)... Reading makefile `gar/gar.conf.mk' (search path) (no ~ expansion)... Reading makefile `/home/maciej/.garrc' (search path) (don't care) (no ~ expansion)... Reading makefile `gar/gar.lib.mk' (search path) (no ~ expansion)... ...and it hangs there. Is anyone else observing this? Maciej From ihsan at dogan.ch Thu Jun 25 18:08:33 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Thu, 25 Jun 2009 18:08:33 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> <4A437E0C.4070101@opencsw.org> Message-ID: <4A43A101.7000509@dogan.ch> Dagobert Michelsen wrote: > BTW, should I change the default compiler for Solaris 9 and 10 to Studio > 12? Yes, that would make sense. I don't there is any reason to keep the default compiler on Studio 11. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at dogan.ch Thu Jun 25 18:10:44 2009 From: ihsan at dogan.ch (Ihsan Dogan) Date: Thu, 25 Jun 2009 18:10:44 +0200 Subject: [csw-maintainers] issue with mail sent by mantis In-Reply-To: References: Message-ID: <4A43A184.8040202@dogan.ch> Peter FELECAN wrote: > The mail delivered by mantis has a systematic issue, detected by Amavis > and others: > > X-Virus-Scanned: amavisd-new at opencsw.org > X-Amavis-Alert: BAD HEADER SECTION, Duplicate header field: "Message-ID" Looks like a Mantis issue. Sebastian? Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From pfelecan at opencsw.org Thu Jun 25 18:32:05 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 25 Jun 2009 18:32:05 +0200 Subject: [csw-maintainers] Minor web site change? In-Reply-To: <5954B1F4-D33A-4B36-A5B4-3CBFA484F488@opencsw.org> (Dagobert Michelsen's message of "Thu\, 25 Jun 2009 15\:38\:13 +0200") References: <625385e30906240532p526d4336i94f0f6180cc4ebe3@mail.gmail.com> <1245851300-sup-4398@ntdws12.chass.utoronto.ca> <4A423C37.2010101@dogan.ch> <4A425D77.3020103@opencsw.org> <0C725D18-39EE-4394-8EF0-A6C84A829873@opencsw.org> <5954B1F4-D33A-4B36-A5B4-3CBFA484F488@opencsw.org> Message-ID: Dagobert Michelsen writes: > PS: Are you in for the beer tonight with William? I wished to but William forgot to tell me... although I'm thirsty. -- Peter From bwalton at opencsw.org Thu Jun 25 18:54:46 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 12:54:46 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <20090625141711.GA25001@bolthole.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> Message-ID: <1245948672-sup-1293@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 25 10:17:11 -0400 2009: > So.. how about writing up some docs for the API now? :-) Done: http://wiki.opencsw.org/package-hooks This is rough and definitely needs comment, thought and most likely improvement. I've already made some changes from what I posted in the patch last night. Notably, I've incorporated the comments (as I interpreted them) from Dago and I also changed /etc/opt/csw/pkg-hooks to /var/opt/csw/libexec/pkg-hooks, as I felt that was a better location. Lets talk it through. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 25 19:54:07 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 25 Jun 2009 10:54:07 -0700 Subject: [csw-maintainers] package hooks In-Reply-To: <1245948672-sup-1293@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245948672-sup-1293@ntdws12.chass.utoronto.ca> Message-ID: <20090625175407.GH13603@bolthole.com> On Thu, Jun 25, 2009 at 12:54:46PM -0400, Ben Walton wrote: > Excerpts from Philip Brown's message of Thu Jun 25 10:17:11 -0400 2009: > > So.. how about writing up some docs for the API now? :-) > > Done: http://wiki.opencsw.org/package-hooks it's a good start. I think it would benefit, first and foremost, by an explaination of why this exists. ie: what is it trying to achieve, that cannot be done via the existing mechanisms of preinstall/postinstall scripts. Additionally, an explicit handling of how they are delivered, and more explicit naming, would probably be a good idea. for example, you dont explicitly say that these "hooks" are always delivered as part of a CSW package. (are they?) Also, while you specify a directory of /var/opt/csw/libexec/pkg-hooks, you dont specify or propose any naming standard for the actual files that live in there. (ie: possible a PKG based prefix or postfix to the name?) Additionally.. i actually do NOT think that /var/opt is appropriate. /var/opt is for data that will be changed /etc/opt is for scripts and configs these are scripts, therefore, they belong under /etc/opt. Now, the scripts themselves, might use /var/opt/XXX for scratch space or state info for themselves. But it doesnt make sense to me, that the scripts themselves would be under /var/opt From bwalton at opencsw.org Thu Jun 25 21:25:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 15:25:47 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <20090625175407.GH13603@bolthole.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245948672-sup-1293@ntdws12.chass.utoronto.ca> <20090625175407.GH13603@bolthole.com> Message-ID: <1245957871-sup-1485@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 25 13:54:07 -0400 2009: > I think it would benefit, first and foremost, by an explaination of why > this exists. > ie: what is it trying to achieve, that cannot be done via the existing > mechanisms of preinstall/postinstall scripts. Updated. > Additionally, an explicit handling of how they are delivered, and more > explicit naming, would probably be a good idea. > for example, you dont explicitly say that these "hooks" are always > delivered as part of a CSW package. (are they?) Updated. > Also, while you specify a directory of /var/opt/csw/libexec/pkg-hooks, > you dont specify or propose any naming standard for the actual files that > live in there. (ie: possible a PKG based prefix or postfix to the > name?) Good point. Updated. > Additionally.. i actually do NOT think that /var/opt is appropriate. > > /var/opt is for data that will be changed > /etc/opt is for scripts and configs Ok. I was thinking they were a 'libexec-ish' type thing, but I think you're correct. I've reverted to the old path. Thanks for the feedback. More welcome, as always. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu Jun 25 23:13:19 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 25 Jun 2009 14:13:19 -0700 Subject: [csw-maintainers] package hooks In-Reply-To: <1245957871-sup-1485@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245948672-sup-1293@ntdws12.chass.utoronto.ca> <20090625175407.GH13603@bolthole.com> <1245957871-sup-1485@ntdws12.chass.utoronto.ca> Message-ID: <20090625211319.GD71096@bolthole.com> On Thu, Jun 25, 2009 at 03:25:47PM -0400, Ben Walton wrote: > Thanks for the feedback. More welcome, as always. ah, that helps a lot. http://wiki.opencsw.org/package-hooks Some more thoughts: "The package tools shall pass, at most, a single argument to each hook that is run. " I think putting arbitrary limits on that, is unneccessary, and possibly harmful. It's usually better to define a MINIMUM set of arguments that is passed, but allow room for future expansion. Hmm... was going to comment on something else, but will think about it a bit more first.... From skayser at opencsw.org Fri Jun 26 00:01:22 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 26 Jun 2009 00:01:22 +0200 (CEST) Subject: [csw-maintainers] issue with mail sent by mantis In-Reply-To: <4A43A184.8040202@dogan.ch> References: <4A43A184.8040202@dogan.ch> Message-ID: <60284.194.246.122.22.1245967282.squirrel@ssl.skayser.de> Ihsan Dogan wrote: > Peter FELECAN wrote: > >> The mail delivered by mantis has a systematic issue, detected by Amavis >> and others: >> >> X-Virus-Scanned: amavisd-new at opencsw.org >> X-Amavis-Alert: BAD HEADER SECTION, Duplicate header field: >> "Message-ID" > > Looks like a Mantis issue. Sebastian? I have just issued a bug notification to myself and can't see such a duplicate header. Might be that there is something on my side normalizing things. Questions: Are others seeing this issue also? Peter, is every bug notification flagged with this alert on your side (the "systematic" makes believe yes, just want to make sure)? Sebastian From skayser at opencsw.org Fri Jun 26 00:24:32 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 26 Jun 2009 00:24:32 +0200 (CEST) Subject: [csw-maintainers] issue with mail sent by mantis In-Reply-To: <60284.194.246.122.22.1245967282.squirrel@ssl.skayser.de> References: <4A43A184.8040202@dogan.ch> <60284.194.246.122.22.1245967282.squirrel@ssl.skayser.de> Message-ID: <50970.194.246.122.22.1245968672.squirrel@ssl.skayser.de> Sebastian Kayser wrote: > Ihsan Dogan wrote: >> Peter FELECAN wrote: >> >>> The mail delivered by mantis has a systematic issue, detected by Amavis >>> and others: >>> >>> X-Virus-Scanned: amavisd-new at opencsw.org >>> X-Amavis-Alert: BAD HEADER SECTION, Duplicate header field: >>> "Message-ID" >> >> Looks like a Mantis issue. Sebastian? > > I have just issued a bug notification to myself and can't see such a > duplicate header. Might be that there is something on my side normalizing > things. > > Questions: Are others seeing this issue also? Peter, is every bug > notification flagged with this alert on your side (the "systematic" makes > believe yes, just want to make sure)? Ok, now i can see it on my side also. Forget about the questions, i will try to take care of it. Sebastian From bonivart at opencsw.org Fri Jun 26 00:51:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Fri, 26 Jun 2009 00:51:14 +0200 Subject: [csw-maintainers] package hooks In-Reply-To: <1245940343-sup-1963@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245940343-sup-1963@ntdws12.chass.utoronto.ca> Message-ID: <625385e30906251551n7c67da54qca80d28ef42bb1bd@mail.gmail.com> On Thu, Jun 25, 2009 at 4:36 PM, Ben Walton wrote: > ?When we're happy, I'll update the patch. If you want to, I can add your Sourceforge account to the pkgutil project, that way you can check in your changes directly into the repository. http://sourceforge.net/project/memberlist.php?group_id=250920 -- /peter From bwalton at opencsw.org Fri Jun 26 01:58:01 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 19:58:01 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <20090625211319.GD71096@bolthole.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245948672-sup-1293@ntdws12.chass.utoronto.ca> <20090625175407.GH13603@bolthole.com> <1245957871-sup-1485@ntdws12.chass.utoronto.ca> <20090625211319.GD71096@bolthole.com> Message-ID: <1245974162-sup-2859@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu Jun 25 17:13:19 -0400 2009: > http://wiki.opencsw.org/package-hooks [Leaves URL included...] > "The package tools shall pass, at most, a single argument to each hook that > is run. " > I think putting arbitrary limits on that, is unneccessary, and possibly > harmful. It's usually better to define a MINIMUM set of arguments that is > passed, but allow room for future expansion. The point with that was to make things a little more rigid, but I've altered the language as you suggest. I also added some thoughts (in the same section) about stdout/stderr from hook scripts. > Hmm... was going to comment on something else, but will think about it a > bit more first.... Fire when ready. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Fri Jun 26 01:59:24 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 25 Jun 2009 19:59:24 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <625385e30906251551n7c67da54qca80d28ef42bb1bd@mail.gmail.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245940343-sup-1963@ntdws12.chass.utoronto.ca> <625385e30906251551n7c67da54qca80d28ef42bb1bd@mail.gmail.com> Message-ID: <1245974288-sup-2717@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Thu Jun 25 18:51:14 -0400 2009: > If you want to, I can add your Sourceforge account to the pkgutil > project, that way you can check in your changes directly into the > repository. > > http://sourceforge.net/project/memberlist.php?group_id=250920 Sure, as long as you trust my perl. It's been several years now since I've written any volume of it. When we've hashed out the standard formally, I'll add support for it to pkgutil as the reference implementation. Sound good? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bonivart at opencsw.org Fri Jun 26 09:53:31 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Fri, 26 Jun 2009 09:53:31 +0200 Subject: [csw-maintainers] package hooks In-Reply-To: <1245974288-sup-2717@ntdws12.chass.utoronto.ca> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245940343-sup-1963@ntdws12.chass.utoronto.ca> <625385e30906251551n7c67da54qca80d28ef42bb1bd@mail.gmail.com> <1245974288-sup-2717@ntdws12.chass.utoronto.ca> Message-ID: <625385e30906260053s5e6d90a4yc89596143ebfda5e@mail.gmail.com> On Fri, Jun 26, 2009 at 1:59 AM, Ben Walton wrote: > Sure, as long as you trust my perl. ?It's been several years now since > I've written any volume of it. ?When we've hashed out the standard > formally, I'll add support for it to pkgutil as the reference > implementation. ?Sound good? Sounds very good. :-) I have added you to the project now. -- /peter From glaw at opencsw.org Fri Jun 26 14:23:02 2009 From: glaw at opencsw.org (Gary Law) Date: Fri, 26 Jun 2009 13:23:02 +0100 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) Message-ID: 2009/6/25 Philip Brown : > > for people who run 8, even having "untested packages" is better than > "having no packages". I'd disagree if they're displacing tested, known good, older Sol 8 versions. I'd suggest we archive what's available for Sol8, publish a health warning to users about no security or other patches, and move on. Given we support the 'supported versions of Solaris', why no OpenSolaris support? This is now a supported Solaris version, according to Sun, and has been since Nov 2008 AFAIK. Gary -- Gary Law glaw at opencsw.org From dam at opencsw.org Fri Jun 26 14:46:33 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 26 Jun 2009 14:46:33 +0200 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) In-Reply-To: References: Message-ID: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> Hi Gary, Am 26.06.2009 um 14:23 schrieb Gary Law: > Given we support the 'supported versions of Solaris', why no > OpenSolaris support? This is now a supported Solaris version, > according to Sun, and has been since Nov 2008 AFAIK. Yes, we should support OpenSolaris as there is support from Sun available. I am currently planning to set up LDoms on the T5220 for OpenSolaris sparc and get some extra capacity on VMware for OpenSolaris x86, but this all will take some time. Best regards -- Dago From phil at bolthole.com Fri Jun 26 15:46:22 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 26 Jun 2009 06:46:22 -0700 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) In-Reply-To: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> References: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> Message-ID: <20090626134622.GA34937@bolthole.com> On Fri, Jun 26, 2009 at 02:46:33PM +0200, Dagobert Michelsen wrote: > Yes, we should support OpenSolaris as there is support from > Sun available. I am currently planning to set up LDoms on the > T5220 for OpenSolaris sparc and get some extra capacity on > VMware for OpenSolaris x86, but this all will take some time. Do we have anyone willing to step up and make an IPS repository out of our packages? From mwatters at opencsw.org Fri Jun 26 15:54:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 26 Jun 2009 08:54:22 -0500 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) In-Reply-To: <20090626134622.GA34937@bolthole.com> References: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> <20090626134622.GA34937@bolthole.com> Message-ID: <4A44D30E.4040701@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > Do we have anyone willing to step up and make an IPS repository out of our > packages? > I am willing to "help" setup the IPS repo, but I currently don't have the time to champion this. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpE0w4ACgkQLrhmsXMSLxdbkQCfWTNQ9ihOuJrUDZEJUG/yunmD NGAAoI1bwuin2xeh4IELgj7tewX+9+ED =uG2+ -----END PGP SIGNATURE----- From schwindt at dfki.uni-kl.de Fri Jun 26 18:50:54 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Fri, 26 Jun 2009 18:50:54 +0200 Subject: [csw-maintainers] Zope in the package repository Message-ID: <200906261650.n5QGotw2008790@dfki.uni-kl.de> As I have the pleasure to transfer an old zope instance I gave Zope 3.3.1 a shot pkg-get -a | grep zope zope 3.3.1 I vaguely remembered problems with python 2.5 and was proven right. This one goes "boom". Even Zope 3.4 fails with python 2.6, it even fails with /usr/bin/python. Maybe the package should be pulled off the catalog ? I am thinking of building python 2.4.6 together with zope 3.4. Would it be a good idea to package both in /opt/csw/zope3 ? Nicolai From bwalton at opencsw.org Fri Jun 26 19:00:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 26 Jun 2009 13:00:47 -0400 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <200906261650.n5QGotw2008790@dfki.uni-kl.de> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> Message-ID: <1246035332-sup-1101@ntdws12.chass.utoronto.ca> Excerpts from Nicolai Schwindt's message of Fri Jun 26 12:50:54 -0400 2009: > Even Zope 3.4 fails with python 2.6, it even fails with > /usr/bin/python. Yes, Zope is rather stinky in this regard. > Maybe the package should be pulled off the catalog ? > > I am thinking of building python 2.4.6 together with zope 3.4. > Would it be a good idea to package both in /opt/csw/zope3 ? There is a 'Unified Installer' for Plone that bundles the required (and tested combinations of) python, python egg libraries, zope and plone install a single installation point. Is there something similar for just Zope? If so, that might be the way to go in this case. It's similar to what you're proposing, but if the work and testing is already done, it should save you some time. We're using the Plone bundled option here and it works well for our needs... HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 26 19:25:18 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 26 Jun 2009 10:25:18 -0700 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <1246035332-sup-1101@ntdws12.chass.utoronto.ca> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> <1246035332-sup-1101@ntdws12.chass.utoronto.ca> Message-ID: <20090626172518.GB26196@bolthole.com> On Fri, Jun 26, 2009 at 01:00:47PM -0400, Ben Walton wrote: > > There is a 'Unified Installer' for Plone that bundles the required > (and tested combinations of) python, python egg libraries, zope and > plone install a single installation point. Is there something similar > for just Zope? If so, that might be the way to go in this case. It's > similar to what you're proposing, but if the work and testing is > already done, it should save you some time. a special "local python", just for zope?! thats... disgusting :-( What's wrong with python, that it cant support slightlyi older versions? and what's wrong with Zope, that it doesnt work with recent perl versions? From bwalton at opencsw.org Fri Jun 26 19:31:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 26 Jun 2009 13:31:47 -0400 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <20090626172518.GB26196@bolthole.com> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> <1246035332-sup-1101@ntdws12.chass.utoronto.ca> <20090626172518.GB26196@bolthole.com> Message-ID: <1246037257-sup-2162@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 26 13:25:18 -0400 2009: > a special "local python", just for zope?! > > thats... disgusting :-( I completely agree. > What's wrong with python, that it cant support slightlyi older > versions? > and what's wrong with Zope, that it doesnt work with recent perl versions? Zope 'monkey patched' into some python functionality that has changed across the versions of python. I don't know why it hasn't been resolved yet, given how long python 2.5 has been around for. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Fri Jun 26 19:37:10 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 26 Jun 2009 10:37:10 -0700 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <1246037257-sup-2162@ntdws12.chass.utoronto.ca> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> <1246035332-sup-1101@ntdws12.chass.utoronto.ca> <20090626172518.GB26196@bolthole.com> <1246037257-sup-2162@ntdws12.chass.utoronto.ca> Message-ID: <20090626173710.GD26196@bolthole.com> On Fri, Jun 26, 2009 at 01:31:47PM -0400, Ben Walton wrote: > Zope 'monkey patched' into some python functionality that has changed > across the versions of python. I don't know why it hasn't been > resolved yet, given how long python 2.5 has been around for. http://wiki.zope.org/zope3/Zope3UsingPython25 The project is currently completed and the final code uploaded can be obtained from the following link. http://google-summer-of-code-2007-zope.googlecode.com/files/Nikhil_N.tar.gz From phil at bolthole.com Fri Jun 26 19:42:30 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 26 Jun 2009 10:42:30 -0700 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <20090626173710.GD26196@bolthole.com> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> <1246035332-sup-1101@ntdws12.chass.utoronto.ca> <20090626172518.GB26196@bolthole.com> <1246037257-sup-2162@ntdws12.chass.utoronto.ca> <20090626173710.GD26196@bolthole.com> Message-ID: <20090626174230.GE26196@bolthole.com> On Fri, Jun 26, 2009 at 10:37:10AM -0700, Philip Brown wrote: > On Fri, Jun 26, 2009 at 01:31:47PM -0400, Ben Walton wrote: > > Zope 'monkey patched' into some python functionality that has changed > > across the versions of python. I don't know why it hasn't been > > resolved yet, given how long python 2.5 has been around for. > > > http://wiki.zope.org/zope3/Zope3UsingPython25 > > The project is currently completed and the final code uploaded can be > obtained from the following link. > http://google-summer-of-code-2007-zope.googlecode.com/files/Nikhil_N.tar.gz also... http://mail.zope.org/pipermail/zope-tests/2009-June/011951.html So there seems to have been some work done, even in the Zope main "trunk"code. just not released i guess? From bwalton at opencsw.org Fri Jun 26 19:42:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 26 Jun 2009 13:42:47 -0400 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: <20090626173710.GD26196@bolthole.com> References: <200906261650.n5QGotw2008790@dfki.uni-kl.de> <1246035332-sup-1101@ntdws12.chass.utoronto.ca> <20090626172518.GB26196@bolthole.com> <1246037257-sup-2162@ntdws12.chass.utoronto.ca> <20090626173710.GD26196@bolthole.com> Message-ID: <1246038075-sup-4078@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri Jun 26 13:37:10 -0400 2009: > http://wiki.zope.org/zope3/Zope3UsingPython25 > > The project is currently completed and the final code uploaded can be > obtained from the following link. > http://google-summer-of-code-2007-zope.googlecode.com/files/Nikhil_N.tar.gz Ok. This has changed since I last checked. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From schwindt at dfki.uni-kl.de Fri Jun 26 20:04:01 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Fri, 26 Jun 2009 20:04:01 +0200 Subject: [csw-maintainers] Zope in the package repository In-Reply-To: Your message of "Fri, 26 Jun 2009 10:42:30 PDT." <20090626174230.GE26196@bolthole.com> Message-ID: <200906261804.n5QI42qH009352@dfki.uni-kl.de> [...] > > http://wiki.zope.org/zope3/Zope3UsingPython25 I found this one too, an I am tryin to build right now .) > > The project is currently completed and the final code uploaded can be > > obtained from the following link. > > http://google-summer-of-code-2007-zope.googlecode.com/files/Nikhil_N.tar.gz > > also... > > http://mail.zope.org/pipermail/zope-tests/2009-June/011951.html This is the svn trunk - I'd rather wait on a stable. I personally dislike python somewhat and hate zope .) Had enough trouble with them already. What zope calls final would be alpha with me - so imagine what I'd think of ther svn trunk. At our site I will have to even use an zope 2.11 because I found projects still using old Zope Products. But I will also look into the possibilities of upgrade the existing zope in csw to some reasonable version Nicolai From bwalton at opencsw.org Fri Jun 26 22:12:57 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 26 Jun 2009 16:12:57 -0400 Subject: [csw-maintainers] package hooks In-Reply-To: <625385e30906260053s5e6d90a4yc89596143ebfda5e@mail.gmail.com> References: <1245897357-sup-1627@ntdws12.chass.utoronto.ca> <20090625141711.GA25001@bolthole.com> <1245940343-sup-1963@ntdws12.chass.utoronto.ca> <625385e30906251551n7c67da54qca80d28ef42bb1bd@mail.gmail.com> <1245974288-sup-2717@ntdws12.chass.utoronto.ca> <625385e30906260053s5e6d90a4yc89596143ebfda5e@mail.gmail.com> Message-ID: <1246047141-sup-6274@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Fri Jun 26 03:53:31 -0400 2009: > I have added you to the project now. Great. This will also give me a chance to try out the git-svn bridge, since there are no svn:externals involved! :) Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From rupert at opencsw.org Sat Jun 27 13:40:50 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 27 Jun 2009 13:40:50 +0200 Subject: [csw-maintainers] libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive (there is only .so ...) In-Reply-To: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> References: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> Message-ID: <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> i tried to compile serf, and get: "test/test_context.c", line 635: warning: statement not reached "test/test_context.c", line 809: warning: statement not reached "buckets/ssl_buckets.c", line 1267: warning: argument #2 is incompatible with prototype: ? ? ? ?prototype: pointer to pointer to unsigned char : "/opt/csw/include/openssl/x509.h", line 939 ? ? ? ?argument : pointer to pointer to char /opt/csw/apache2/share/build/libtool --silent --mode=link /opt/studio/SOS11/SUNWspro/bin/cc -xarch=v8 -L/opt/csw/lib -L/opt/csw/bdb44/lib -R/opt/csw/apache2/lib -L/opt/csw/apache2/lib -R/opt/csw/apache2/lib -L/opt/csw/apache2/lib -rpath /opt/csw/lib -o libserf-0.la buckets/aggregate_buckets.lo buckets/request_buckets.lo context.lo buckets/buckets.lo buckets/simple_buckets.lo buckets/file_buckets.lo buckets/mmap_buckets.lo buckets/socket_buckets.lo buckets/response_buckets.lo buckets/headers_buckets.lo buckets/allocator.lo buckets/dechunk_buckets.lo buckets/deflate_buckets.lo buckets/limit_buckets.lo buckets/ssl_buckets.lo buckets/barrier_buckets.lo buckets/chunk_buckets.lo -laprutil-1 -lldap -llber -ldb-4.4 -lexpat -liconv ?-lapr-1 -luuid -lsendfile -lrt -lsocket ?-lpthread -ldl -lm ?-lz -lssl -lcrypto grep: can't open /opt/csw/bdb44/lib/libdb-4.4.la /opt/csw/bin/gsed: can't read /opt/csw/bdb44/lib/libdb-4.4.la: No such file or directory libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive gmake[2]: *** [libserf-0.la] Error 1 gmake[2]: Leaving directory `/home/rupert/mgar/pkg/libserf/trunk/work/build-isa-sparcv8/serf-0.3.0' gmake[1]: *** [build-work/build-isa-sparcv8/serf-0.3.0/Makefile] Error 2 gmake[1]: Leaving directory `/home/rupert/mgar/pkg/libserf/trunk' gmake: *** [merge-isa-sparcv8] Error 2 how could one avoid that? rupert. From rupert at opencsw.org Sat Jun 27 14:00:56 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 27 Jun 2009 14:00:56 +0200 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <4A180558.70601@opencsw.org> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> <6af4270905221809u232fdfe7l7f55df4705a0434@mail.gmail.com> <4A180558.70601@opencsw.org> Message-ID: <6af4270906270500nbc25e89rcc96612455fa6c50@mail.gmail.com> hi mike, do you have an example how you put such plugins best into a plugin pack, especially: * multiple downloads * installs then i could give it a try. rupert. On Sat, May 23, 2009 at 16:16, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi rupert, > rupert THURNER wrote: >> >> that is great to hear! i even would appreciate if the most important >> plugins make it into trac as well (not enabled by default), e.g. >> >> easy_install http://trac-hacks.org/svn/tracwysiwygplugin/0.11/ >> easy_install http://trac-hacks.org/svn/ticketdeleteplugin/0.11/ >> easy_install http://trac-hacks.org/svn/ticketchangeplugin/0.11/ >> easy_install http://trac-hacks.org/svn/customfieldadminplugin/0.11/ >> easy_install http://trac-hacks.org/svn/tagsplugin/tags/0.6/ >> easy_install http://svn.ipd.uka.de/repos/javaparty/JP/trac/plugins/tracnav-0.11/ >> easy_install http://trac-hacks.org/svn/sectioneditplugin/0.11/ >> easy_install http://trac-hacks.org/svn/extracturlplugin/0.11/ >> easy_install http://trac-hacks.org/svn/serversideredirectplugin/0.11/ >> easy_install http://trac-hacks.org/svn/masterticketsplugin/0.11/ >> easy_install http://trac-hacks.org/svn/autoupgradeplugin/0.11/ >> > > I am working on updating Trac in the repository. ( after I get python > re-released as separate packages.) ?I will create a second package for trac, > trac_plugin_pack that contains the eggs you listed above. > > also, > your last post on the google group forum link in the ticket. > the '-G' is gone because you export LDSHARED=/opt/SUNWspro/bin/cc > ''' > compiler.linker_so > ['/opt/studio/SOS11/SUNWspro/bin/cc', '-G'] > ''' > means that then setting for compiler.linker_so is > /opt/studio/SOS11/SUNWspro/bin/cc -G > > os.environ settings override what python finds in its config. > export LDSHARED='/opt/SUNWspro/bin/cc -G' > would have saved the -G > > you can easily play with this using the following python script > =============================================================== > #! /usr/bin/env python > > from distutils.ccompiler import new_compiler > from distutils.sysconfig import customize_compiler > > compiler = new_compiler(None,None,None,None) > customize_compiler(compiler) > > print """ > *** Compiler Settings for Python *** > > ? ?Type: %s > ?Linker: %s > Compiler: %s > """ % (compiler.compiler_type, compiler.linker_so, compiler.compiler) > ====================================================================== > > $ ./pycompile.py > > *** Compiler Settings for Python *** > > ? ?Type: unix > ?Linker: ['/opt/studio/SOS11/SUNWspro/bin/cc', '-G'] > Compiler: ['/opt/studio/SOS11/SUNWspro/bin/cc', '-DNDEBUG', '-O'] > > you can set various environment variables and see how that affects those settings. > > - -- > Thanks, > Mike > > "Any intelligent fool can make things bigger, more complex, > and more violent. ?It takes a touch of genius -- and a lot of courage -- > to move in the opposite direction." > > * Albert Einstein 1879 - 1955 > ? ?US German-born Theoretical Physicist > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.9 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iEYEARECAAYFAkoYBVgACgkQLrhmsXMSLxdLswCg4qlEUAdZIpxfr6B/0UpQhQHj > UW4AoLk0dKQ3J5rsmkEwylnShAaWUtrh > =GqGq > -----END PGP SIGNATURE----- > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers > From bwalton at opencsw.org Sat Jun 27 15:13:56 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sat, 27 Jun 2009 09:13:56 -0400 Subject: [csw-maintainers] libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive (there is only .so ...) In-Reply-To: <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> References: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> Message-ID: <1246108151-sup-8443@ntdws12.chass.utoronto.ca> Excerpts from rupert THURNER's message of Sat Jun 27 07:40:50 -0400 2009: > buckets/barrier_buckets.lo buckets/chunk_buckets.lo -laprutil-1 -lldap > -llber -ldb-4.4 -lexpat -liconv ?-lapr-1 -luuid -lsendfile -lrt > -lsocket ?-lpthread -ldl -lm ?-lz -lssl -lcrypto > grep: can't open /opt/csw/bdb44/lib/libdb-4.4.la > /opt/csw/bin/gsed: can't read /opt/csw/bdb44/lib/libdb-4.4.la: No such > file or directory You can try the libtool fixup tools that Mike added a few months back. You'll want to set STRIP_LIBTOOL to 1 (or any value, really) in your GAR Makefile. I hit a situation where this didn't help, but that was for a package that was doing things 'their own way' instead of the more standard way you'll typically encounter. If this variable (and the gar/bin/fixlibtool script) doesn't help in your case, let me know and I'll show you how I fixed it for my oddball. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Sat Jun 27 17:11:35 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 27 Jun 2009 10:11:35 -0500 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270906270500nbc25e89rcc96612455fa6c50@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> <6af4270905221809u232fdfe7l7f55df4705a0434@mail.gmail.com> <4A180558.70601@opencsw.org> <6af4270906270500nbc25e89rcc96612455fa6c50@mail.gmail.com> Message-ID: <4A4636A7.8040602@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > hi mike, > > do you have an example how you put such plugins best into a plugin > pack, especially: > * multiple downloads > * installs > > then i could give it a try. > > rupert. > I don't have an example unfortunately. basically you would have to write a custom Makefile. downloading all the software you want to bundle, building and installing each one in $(DESTDIR) the merge and package should be relatively normal once everything is installed. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkpGNqYACgkQLrhmsXMSLxfuawCeLYqj5aKBIFY+/JlJoduqqTf5 RzgAmwUtancgZXxy0DJepU68IULol0ue =g96R -----END PGP SIGNATURE----- From dam at opencsw.org Sat Jun 27 21:27:47 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sat, 27 Jun 2009 21:27:47 +0200 Subject: [csw-maintainers] /testing: nasm 2.05.01 In-Reply-To: <21894.1244997256@gerd> References: <21894.1244997256@gerd> Message-ID: Hi Murray, Am 14.06.2009 um 18:34 schrieb : > The following files have been put into /testing ... > > 9aea1d1df09e7f51525d8a9fee95b0ee nasm-2.05.01,REV=2009.06.14- > SunOS5.8-i386-CSW.pkg.gz > 0fa0d0eebd39d15f6c7ae7336764b4fa nasm-2.05.01,REV=2009.06.14- > SunOS5.8-sparc-CSW.pkg.gz > > I haven't even run them yet - let me know if they work. Cheers! Looks good. Please update to 2.06 and relase it :-) Best regards -- Dago From rupert at opencsw.org Sun Jun 28 21:14:26 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 28 Jun 2009 21:14:26 +0200 Subject: [csw-maintainers] libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive (there is only .so ...) In-Reply-To: <1246108151-sup-8443@ntdws12.chass.utoronto.ca> References: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> <1246108151-sup-8443@ntdws12.chass.utoronto.ca> Message-ID: <6af4270906281214rd80c747hc04f69b442ea41a2@mail.gmail.com> On Sat, Jun 27, 2009 at 15:13, Ben Walton wrote: > Excerpts from rupert THURNER's message of Sat Jun 27 07:40:50 -0400 2009: >> buckets/barrier_buckets.lo buckets/chunk_buckets.lo -laprutil-1 -lldap >> -llber -ldb-4.4 -lexpat -liconv ?-lapr-1 -luuid -lsendfile -lrt >> -lsocket ?-lpthread -ldl -lm ?-lz -lssl -lcrypto >> grep: can't open /opt/csw/bdb44/lib/libdb-4.4.la >> /opt/csw/bin/gsed: can't read /opt/csw/bdb44/lib/libdb-4.4.la: No such >> file or directory > > You can try the libtool fixup tools that Mike added a few months > back. ?You'll want to set STRIP_LIBTOOL to 1 (or any value, really) in > your GAR Makefile. ?I hit a situation where this didn't help, but that > was for a package that was doing things 'their own way' instead of the > more standard way you'll typically encounter. > > If this variable (and the gar/bin/fixlibtool script) doesn't help in > your case, let me know and I'll show you how I fixed it for my > oddball. i did set it, but got the error anyway. i also found http://trac.macports.org/ticket/19751 which says: serf links to lib_apr, which depends on bdb. apr-1.3.7 fixes this, and dynamically links to bdb. but this would mean we'd need libapr, libaprutil as separate package i guess? rupert. From william at wbonnet.net Sun Jun 28 21:55:52 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 28 Jun 2009 21:55:52 +0200 Subject: [csw-maintainers] another bug in upstream. In-Reply-To: <4A308F3E.9020306@opencsw.org> References: <4A308F3E.9020306@opencsw.org> Message-ID: <4A47CAC8.5080603@wbonnet.net> Hi Mike > I received an upgrade notice for GD ( pasted below ) > Thanks for the bug report. I'll try to fix it in the next days. I am currently changing several things to uwatch to handle this kind of problem, and a adding a few missing features. > I am the GD maintainer, however all the links to the download location are for > the pm_gd to which I am not the maintainer. ( I believe it was Alex who is > retired.) > > Further investigation showed me that I do in fact have the latest "stable" > release of GD packaged, but the perl module is what is in need of upgrade. > GD-2.43.tar.gz is the filename of pm_gd > gd-2.0.35.tar.gz is the filename of gd > > note the regex for upstream watch should be the same. > > I have no solution off hand to this bug, but I will take a look when I get a > few spare cycles. the problem is not the regexp used by uwatch, but the same Phil already reported before. Both package have the same GARNAME. The perl module should be renamed to have a different GARNAME (actually pm_gd garname is GD, and gd garname is gd). Since pm_gd is orphaned, i'll fix this package in the next days. cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From william at wbonnet.net Sun Jun 28 22:16:35 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 28 Jun 2009 22:16:35 +0200 Subject: [csw-maintainers] Adding a new category to mantis ? Message-ID: <4A47CFA3.6040608@wbonnet.net> Hi I would to propose a small evolution to mantis. To add a category for GAR build description issues. Actually the following categories are available : Other Packaging Upgrade Regular Use IMHO "Packaging" category is different of a possible "Build description" category. I do the following distinction between these cat?gories. Packaging can keep track of problem with "package content", and build description of problem about prackage creation process. For exemple, a missing config file or an invalid permission on a file is a packaging issue. A missing uwatch regexp or a duplicated garname is a build description issue. "Build description" for category name is just a proposal. Other names are welcomed :) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From rupert at opencsw.org Sun Jun 28 23:06:28 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 28 Jun 2009 23:06:28 +0200 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) In-Reply-To: <20090626134622.GA34937@bolthole.com> References: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> <20090626134622.GA34937@bolthole.com> Message-ID: <6af4270906281406o60495920lbaa4c8c5789bdb75@mail.gmail.com> On Fri, Jun 26, 2009 at 15:46, Philip Brown wrote: > On Fri, Jun 26, 2009 at 02:46:33PM +0200, Dagobert Michelsen wrote: >> Yes, we should support OpenSolaris as there is support from >> Sun available. I am currently planning to set up LDoms on the >> T5220 for OpenSolaris sparc and get some extra capacity on >> VMware for OpenSolaris x86, but this all will take some time. > > Do we have anyone willing to step up and make an IPS repository out of our > packages? you mean: 1.setting up a repository: http://blogs.sun.com/halton/entry/setup_a_ips_repository 2. http://blogs.sun.com/migi/resource/creating_ips_repo_part_5.html use solaris.py ? (out of http://blogs.sun.com/migi/entry/create_your_own_opensolaris_ips2) or just do something like: for cswpkg in `ls -1 /home/wherever-opencsw-packages-are` do pkgsend -s http://pkg.opencsw.org:10000 send $cswpkg done From phil at bolthole.com Sun Jun 28 23:51:26 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 28 Jun 2009 14:51:26 -0700 Subject: [csw-maintainers] Supported Versions (was Re: Minor web site change?) In-Reply-To: <6af4270906281406o60495920lbaa4c8c5789bdb75@mail.gmail.com> References: <1B1252BA-18EB-4687-9CCD-A09760B0AD0F@opencsw.org> <20090626134622.GA34937@bolthole.com> <6af4270906281406o60495920lbaa4c8c5789bdb75@mail.gmail.com> Message-ID: <20090628215126.GB83796@bolthole.com> On Sun, Jun 28, 2009 at 11:06:28PM +0200, rupert THURNER wrote: > On Fri, Jun 26, 2009 at 15:46, Philip Brown wrote: > > Do we have anyone willing to step up and make an IPS repository out of our > > packages? > > you mean: > > 1.setting up a repository: > http://blogs.sun.com/halton/entry/setup_a_ips_repository > 2. http://blogs.sun.com/migi/resource/creating_ips_repo_part_5.html > use solaris.py ? > (out of http://blogs.sun.com/migi/entry/create_your_own_opensolaris_ips2) > > or just do something like: > > for cswpkg in `ls -1 /home/wherever-opencsw-packages-are` > do > pkgsend -s http://pkg.opencsw.org:10000 send $cswpkg > done I meant primarily taking on the responsability of converting packages, on an ongoing basis. This is not a trivial task: Quite a few of our packages will need massaging, for them to work under IPS. Additionally, this person would then have de-facto responsability, of giving feedback to the "mainline group" (ie: the maintainers at large :) on things that do not translate well to IPS mechanisms If anyone is interested, please email me privately, and i will discuss more details there. From dam at opencsw.org Mon Jun 29 14:09:00 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 29 Jun 2009 14:09:00 +0200 Subject: [csw-maintainers] Firmware update on BO Buildfarm 2009/06/30 Message-ID: Hi, the T5220 at the BO farm carrying login, build*s, web and mirror will be down tomorrow, on 2009/06/30 from 10:00 CEST to 12:00 CEST (CEST = UTC+2). I am going to update the firmware of the T5220 to prepare it for Logical Domains as prerequisite for an OpenSolaris build host. There will be an additional announcement when the update is finished. You can still use build8s.go.opencsw.org during the downtime. The SVN repository is not affected as it is hosted at SourceForge. Best regards -- Dago From bonivart at opencsw.org Mon Jun 29 15:12:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 29 Jun 2009 15:12:14 +0200 Subject: [csw-maintainers] Duplicate CSWlibgnet2rt in catalog Message-ID: <625385e30906290612l3c07f602k7b9c9856ba67d20e@mail.gmail.com> It seems that CSWlibgnet2rt has changed its catalogname from libgnet2rt to libgnet2_rt but the old entry is still there: libgnet2_rt 2.0.8,REV=2009.05.27 CSWlibgnet2rt libgnet2_rt-2.0.8,REV=2009.05.27-SunOS5.8-sparc-CSW.pkg.gz 61d14c836f09148fd79c1ae86b53e434 78126 CSWcommon|CSWgcc3corert|CSWggettextrt|CSWglib2|CSWiconv none libgnet2rt 2.0.7 CSWlibgnet2rt libgnet2rt-2.0.7-SunOS5.8-sparc-CSW.pkg.gz ee590915e6ee972869c32ee9a372dc47 99258 CSWcommon|CSWggettext|CSWglib2|CSWiconv none -- /peter From rmacduff at opencsw.org Mon Jun 29 17:16:58 2009 From: rmacduff at opencsw.org (Ross Macduff) Date: Mon, 29 Jun 2009 11:16:58 -0400 Subject: [csw-maintainers] gsed 4.2.1 in testing Message-ID: <1246288551-sup-4211@frog.chass.utoronto.ca> Hello, gsed 4.2.1 is now in testing. Best, Ross From mwatters at opencsw.org Mon Jun 29 17:23:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 29 Jun 2009 10:23:51 -0500 Subject: [csw-maintainers] pycairo 1.8.0 now in testing Message-ID: <4A48DC87.7070006@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 you can get it here: pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u pycairo Feedback always welcome. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpI3IYACgkQLrhmsXMSLxdP7QCfbeVU7llZsXacKp7Ejvs3ym1y JDsAoKKwteXvihKNE3MzBXiUe3aJzvjx =iQq+ -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 29 17:48:44 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 29 Jun 2009 10:48:44 -0500 Subject: [csw-maintainers] anyone updating libglade2? Message-ID: <4A48E25C.20309@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 if not, I will do it, once pycairo (in testing) is released it will be the last dependency I need to update pygtk. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpI4lwACgkQLrhmsXMSLxdHugCfT5/WM93dw6YWQbTz9tOmGmk8 cqwAmwf95q59EqQfqs8AYJRy4sGCLzcf =6PhQ -----END PGP SIGNATURE----- From mwatters at opencsw.org Mon Jun 29 18:10:30 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 29 Jun 2009 11:10:30 -0500 Subject: [csw-maintainers] anyone updating libglade2? In-Reply-To: <4A48E25C.20309@opencsw.org> References: <4A48E25C.20309@opencsw.org> Message-ID: <4A48E776.4040200@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > if not, I will do it, once pycairo (in testing) is released it will be the last > dependency I need to update pygtk. > well, last dependency besides gtk ;-) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpI53YACgkQLrhmsXMSLxdVxwCbByC5pS7czeK/adnplppPjhma ttkAoMU7Q8ZaUcY3DGQ0vVTTbzzMI4ka =lNYv -----END PGP SIGNATURE----- From benny at opencsw.org Mon Jun 29 18:48:59 2009 From: benny at opencsw.org (Benjamin von Mossner) Date: Mon, 29 Jun 2009 18:48:59 +0200 Subject: [csw-maintainers] pen: initial commit Message-ID: <20090629164859.GA73690@vonmossner.de> Hi maintainers, i just commited my first package (pen) :-) a quite tiny one, but nevertheless. If anyone feels about checking my Makefile and/or the package itself, please see svn and/or /home/testing. Any feedback is welcome. Cheers, benny -- /"\ ASCII RIBBON CAMPAIGN | Benjamin von Mossner \ / AGAINST HTML MAIL | benny at vonmossner.de X / \ multiple exclamation marks are a sure sign of a diseased mind From benny at vonmossner.de Mon Jun 29 18:48:59 2009 From: benny at vonmossner.de (Benjamin von Mossner) Date: Mon, 29 Jun 2009 18:48:59 +0200 Subject: [csw-maintainers] pen: initial commit Message-ID: <20090629164859.GA73690@vonmossner.de> Hi maintainers, i just commited my first package (pen) :-) a quite tiny one, but nevertheless. If anyone feels about checking my Makefile and/or the package itself, please see svn and/or /home/testing. Any feedback is welcome. Cheers, benny -- /"\ ASCII RIBBON CAMPAIGN | Benjamin von Mossner \ / AGAINST HTML MAIL | benny at vonmossner.de X / \ multiple exclamation marks are a sure sign of a diseased mind From phil at bolthole.com Mon Jun 29 19:11:25 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 29 Jun 2009 10:11:25 -0700 Subject: [csw-maintainers] Duplicate CSWlibgnet2rt in catalog In-Reply-To: <625385e30906290612l3c07f602k7b9c9856ba67d20e@mail.gmail.com> References: <625385e30906290612l3c07f602k7b9c9856ba67d20e@mail.gmail.com> Message-ID: <20090629171125.GC98636@bolthole.com> On Mon, Jun 29, 2009 at 03:12:14PM +0200, Peter Bonivart wrote: > It seems that CSWlibgnet2rt has changed its catalogname from > libgnet2rt to libgnet2_rt but the old entry is still there: > Urk. thanks for catching that. From bwalton at opencsw.org Mon Jun 29 19:19:42 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 29 Jun 2009 13:19:42 -0400 Subject: [csw-maintainers] libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive (there is only .so ...) In-Reply-To: <6af4270906281214rd80c747hc04f69b442ea41a2@mail.gmail.com> References: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> <1246108151-sup-8443@ntdws12.chass.utoronto.ca> <6af4270906281214rd80c747hc04f69b442ea41a2@mail.gmail.com> Message-ID: <1246295816-sup-8970@ntdws12.chass.utoronto.ca> Excerpts from rupert THURNER's message of Sun Jun 28 15:14:26 -0400 2009: > but this would mean we'd need libapr, libaprutil as separate package > i guess? I thought that someone was working on this already...? As for your issues, see the Makefile for openjade. Specifically, the post-configure-modulated step. That's how I 'broke' libtool there...not sure if it's helpful for your case or not, since your situation is different. HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From mwatters at opencsw.org Mon Jun 29 20:03:02 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 29 Jun 2009 13:03:02 -0500 Subject: [csw-maintainers] libtool: link: `/opt/csw/bdb44/lib/libdb-4.4.la' is not a valid libtool archive (there is only .so ...) In-Reply-To: <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> References: <6af4270906270440h4eca2eeei77ea5bd8e1da1531@mail.gmail.com> <6af4270906270440y4b3a1961kcad94d8288d2b2f0@mail.gmail.com> Message-ID: <4A4901D6.50402@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > > how could one avoid that? > the main issue is the Makefile generated by configre is using "libtool" from /opt/csw/apache2/share/build/libtool which does "NOT" get modified by my strip. Here is a patch to the Makefile currently in svn that will allow you to build and test successfully libserf if you cut and paste, please take care to watch for wrapping I have Thunderbird configured to wrap at 100 characters. $ svn diff Makefile Index: Makefile =================================================================== - --- Makefile (revision 5414) +++ Makefile (working copy) @@ -23,13 +23,19 @@ # UPSTREAM_MASTER_SITES = DEPENDS = server/apache2rt - -STRIP_LIBTOOL = 1 CONFIGURE_ARGS = $(DIRPATHS) - -CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 - -CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2 - -#CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config - -#CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config +CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config +CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config TEST_TARGET = check include gar/category.mk + +EXTRA_LDFLAGS = -R/opt/csw/bdb44/lib +APR_LIBTOOL = $(shell /opt/csw/apache2/bin/apr-config --apr-libtool) +post-configure-modulated: + gcp $(APR_LIBTOOL) $(WORKSRC) + perl -i -pne 's|$(APR_LIBTOOL)|$(abspath $(WORKSRC)/libtool)|' \ + $(WORKSRC)/Makefile + $(GARBIN)/fixlibtool $(WORKSRC) + @$(MAKECOOKIE) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpJAdUACgkQLrhmsXMSLxeBXQCgmTxH6hTrxyY2TrT6eSMj+S4t c+4Anj0nU/L6fyoBYYX4VqFjMCGd4iRo =qeGL -----END PGP SIGNATURE----- From skayser at opencsw.org Mon Jun 29 23:20:19 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 29 Jun 2009 23:20:19 +0200 (CEST) Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") In-Reply-To: <4A2403D0.4090109@opencsw.org> References: <4A2403D0.4090109@opencsw.org> Message-ID: <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> Sebastian Kayser wrote: > a couple of RFEs for the "GAR package status" [1] page. Can we get: > > * an "last updated" field at the top of the page > * the numbers / stats right at the top of the page (instead of further > down) > * a contact address for the maintainer of the page on the page (for > queries like this one and to get packages that are in GAR but not > listed as such fixed) > > Ah, and some sort of monthly historical data would be cool also. So that > we can see some progress. > > Sebastian > > [1] http://www.opencsw.org/package-gar-status.html Anyone taking care of the GAR status page? Sebastian From william at wbonnet.net Mon Jun 29 23:41:26 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 29 Jun 2009 23:41:26 +0200 Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") In-Reply-To: <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> References: <4A2403D0.4090109@opencsw.org> <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> Message-ID: <4A493506.9050204@wbonnet.net> Hi Sebastian > Sebastian Kayser wrote: > >> a couple of RFEs for the "GAR package status" [1] page. Can we get: >> >> * an "last updated" field at the top of the page >> * the numbers / stats right at the top of the page (instead of further >> down) >> * a contact address for the maintainer of the page on the page (for >> queries like this one and to get packages that are in GAR but not >> listed as such fixed) >> >> Ah, and some sort of monthly historical data would be cool also. So that >> we can see some progress. >> >> Sebastian >> >> [1] http://www.opencsw.org/package-gar-status.html >> > > Anyone taking care of the GAR status page? > I'm currently working on this page and uwatch. Rewriting it in Perl and improving things. I will modify the page by the end of the week, and uwatch next week. uwatch have several bugs to fix (problem with garname used to retrieve package instead of catalog name). I would like to merge some information from uwatch to this page. cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From dam at opencsw.org Mon Jun 29 23:44:39 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 29 Jun 2009 23:44:39 +0200 Subject: [csw-maintainers] New in testing: gtk2 2.16.1 Message-ID: Hi, there is now a current gtk2 in testing: gtk2-2.16.1,REV=2009.06.29-SunOS5.8-i386-CSW.pkg.gz gtk2-2.16.1,REV=2009.06.29-SunOS5.8-sparc-CSW.pkg.gz gtk2_devel-2.16.1,REV=2009.06.29-SunOS5.8-i386-CSW.pkg.gz gtk2_devel-2.16.1,REV=2009.06.29-SunOS5.8-sparc-CSW.pkg.gz gtk2_doc-2.16.1,REV=2009.06.29-SunOS5.8-i386-CSW.pkg.gz gtk2_doc-2.16.1,REV=2009.06.29-SunOS5.8-sparc-CSW.pkg.gz Feedback as always welcome! (I know, doc should be archall) Best regards -- Dago From william at wbonnet.net Mon Jun 29 23:47:01 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 29 Jun 2009 23:47:01 +0200 Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") In-Reply-To: <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> References: <4A2403D0.4090109@opencsw.org> <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> Message-ID: <4A493655.1090206@wbonnet.net> Hi opps i hit the send key too fast :) Is there any other RFE to come ? I have added a wiki page for this RFE http://wiki.opencsw.org/gar-package-status-rfe cheers W. From skayser at opencsw.org Tue Jun 30 00:31:59 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 30 Jun 2009 00:31:59 +0200 (CEST) Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") In-Reply-To: <4A493655.1090206@wbonnet.net> References: <4A2403D0.4090109@opencsw.org> <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> <4A493655.1090206@wbonnet.net> Message-ID: <51059.217.227.50.24.1246314719.squirrel@ssl.skayser.de> William Bonnet wrote: > opps i hit the send key too fast :) > > Is there any other RFE to come ? > > I have added a wiki page for this RFE > http://wiki.opencsw.org/gar-package-status-rfe Thanks for the status update and thanks for taking care of it. No additional RFEs from my side - at least for now. ;) Sebastian From william at wbonnet.net Tue Jun 30 00:49:54 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 30 Jun 2009 00:49:54 +0200 Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka "Which packages are in GAR") In-Reply-To: <51059.217.227.50.24.1246314719.squirrel@ssl.skayser.de> References: <4A2403D0.4090109@opencsw.org> <50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de> <4A493655.1090206@wbonnet.net> <51059.217.227.50.24.1246314719.squirrel@ssl.skayser.de> Message-ID: <4A494512.9040104@wbonnet.net> Hi Sebastian > Thanks for the status update and thanks for taking care of it. No > additional RFEs from my side - at least for now. ;) > Page content is (supposed to be) updated everyday. I will add modification date tomorrow. The rest of the evolutions will come with the perl version. I'm working on this since yesterday, i am also pluging the database backend. Actaully this page is generated by an ugly shell script (generating html...). cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From car at opencsw.org Tue Jun 30 11:08:52 2009 From: car at opencsw.org (Chris Reece) Date: Tue, 30 Jun 2009 21:08:52 +1200 Subject: [csw-maintainers] tcsh 6.16.00 in testing Message-ID: As per subject, current version of tcsh in: /home/testing/tcsh-6.16.00,REV=2009.06.30-SunOS5.8-i386-CSW.pkg.gz /home/testing/tcsh-6.16.00,REV=2009.06.30-SunOS5.8-sparc-CSW.pkg.gz Works on my only test host, would appreciate any additional testing. Cheers, Chris. From bonivart at opencsw.org Tue Jun 30 13:36:47 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 30 Jun 2009 13:36:47 +0200 Subject: [csw-maintainers] tcsh 6.16.00 in testing In-Reply-To: References: Message-ID: <625385e30906300436v1eed482eq4c9f5354ed6415e7@mail.gmail.com> On Tue, Jun 30, 2009 at 11:08 AM, Chris Reece wrote: > As per subject, current version of tcsh in: > > /home/testing/tcsh-6.16.00,REV=2009.06.30-SunOS5.8-i386-CSW.pkg.gz > /home/testing/tcsh-6.16.00,REV=2009.06.30-SunOS5.8-sparc-CSW.pkg.gz > > Works on my only test host, would appreciate any additional testing. It was years ago I used tcsh but it installs/removes cleanly and seems to work properly during my short test on a Solaris 9 Sparc and a Solaris 10 x86. Man page also works. -- /peter From dam at opencsw.org Tue Jun 30 13:45:37 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 30 Jun 2009 13:45:37 +0200 Subject: [csw-maintainers] Firmware update on BO Buildfarm 2009/06/30 In-Reply-To: References: Message-ID: Hi, Am 29.06.2009 um 14:09 schrieb Dagobert Michelsen: > the T5220 at the BO farm carrying login, build*s, web and mirror > will be down tomorrow, on 2009/06/30 from 10:00 CEST to 12:00 CEST > (CEST = UTC+2). I am going to update the firmware of the T5220 > to prepare it for Logical Domains as prerequisite for an > OpenSolaris build host. There will be an additional announcement > when the update is finished. Unfortunately I had an unplanned customer visit and the update wasn't done. I will schedule it for a later time. Best regards -- Dago From ja at opencsw.org Tue Jun 30 15:10:28 2009 From: ja at opencsw.org (Juergen Arndt) Date: Tue, 30 Jun 2009 15:10:28 +0200 Subject: [csw-maintainers] Fwd: [csw-pkgrequests] package request In-Reply-To: References: <20090630115854.1C68B8754@mail.opencsw.org> Message-ID: Hi, I'll try to build packages for the munin server and node. Could take some days, please be patient :) Juergen >> Von: mcb at uib.no >> Datum: 30. Juni 2009 13:58:54 MESZ >> An: pkgrequests at lists.opencsw.org >> Betreff: [csw-pkgrequests] package request >> >> Munin: http://munin.projects.linpro.no/ >> >> I would appreciate packages for both munin and munin-node. >> >> What is Munin? >> Munin the monitoring tool surveys all your computers and remembers what >> it saw. It presents all the information in graphs through a web >> interface. Its emphasis is on plug and play capabilities. After >> completing a installation a high number of monitoring plugins will be >> playing with no more effort. >> >> Using Munin you can easily monitor the performance of your computers, >> networks, SANs, applications, weather measurements and whatever comes >> to mind. It makes it easy to determine "what's different today" when a >> performance problem crops up. It makes it easy to see how you're doing >> capacity-wise on any resources. >> >> Munin uses the excellent RRDTool (written by Tobi Oetiker) and the >> framework is written in Perl, while plugins may be written in any >> language. Munin has a master/node architecture in which the master >> connects to all the nodes at regular intervals and asks them for data. >> It then stores the data in RRD files, and (if needed) updates the >> graphs. One of the main goals has been ease of creating new plugins >> (graphs). >> _______________________________________________ >> pkgrequests mailing list >> pkgrequests at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/pkgrequests > -- Juergen Arndt From mwatters at opencsw.org Tue Jun 30 18:19:16 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 30 Jun 2009 11:19:16 -0500 Subject: [csw-maintainers] New in testing: gtk2 2.16.1 In-Reply-To: References: Message-ID: <4A4A3B04.2040701@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Feedback as always welcome! I have successfully compiled and tested libglade2 v2.6.4, pidgin v2.5.7 against the version in testing. +1 on the release. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpKOwQACgkQLrhmsXMSLxcCCQCfSJ20DzwMjfzWUGNxjzbWT3Yz CvQAoNl7l3cHrmXEXStl7L4WqufJOxWP =DE5/ -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue Jun 30 19:57:24 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 30 Jun 2009 12:57:24 -0500 Subject: [csw-maintainers] New in testing: gtk2 2.16.1 In-Reply-To: <4A4A3B04.2040701@opencsw.org> References: <4A4A3B04.2040701@opencsw.org> Message-ID: <4A4A5204.7070705@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > Dagobert Michelsen wrote: > >> Feedback as always welcome! > Successfully compiled and tested pygtk against 2.16.1 in testing. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpKUgQACgkQLrhmsXMSLxff+QCgvA9nuS4NtF2O+lE6k01+X8rn Dx8AoNJRTWgoLDCdVaRY/3MYrJ34MMiZ =dK3f -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue Jun 30 20:28:42 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 30 Jun 2009 13:28:42 -0500 Subject: [csw-maintainers] New in testing: gtk2 2.16.1 In-Reply-To: <4A4A5204.7070705@opencsw.org> References: <4A4A3B04.2040701@opencsw.org> <4A4A5204.7070705@opencsw.org> Message-ID: <4A4A595A.3010304@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 While playing with pygtk, I found a "nicety" for the package install. have a post install that runs: /opt/csw/bin/gdk-pixbuf-query-loaders >/opt/csw/etc/gtk-2.0/gdk-pixbuf.loaders I am going to file a feature bug against the pango package to request the same for /opt/csw/bin/pango-querymodules >/opt/csw/etc/pango/pango.modules - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkpKWVkACgkQLrhmsXMSLxf+MACg5924oC6sIGEB3ZxnLq3CEXvB 930An0teOX8d1T3iI/33diKsJxfM+jNe =/U7x -----END PGP SIGNATURE----- From car at opencsw.org Tue Jun 30 23:57:29 2009 From: car at opencsw.org (Chris Reece) Date: Wed, 1 Jul 2009 09:57:29 +1200 (NZST) Subject: [csw-maintainers] tcsh 6.16.00 in testing Message-ID: <15954.131.203.105.1.1246399049.squirrel@webmail.jessies.org> On Tue, June 30, 2009 11:36 pm, Peter Bonivart wrote: > It was years ago I used tcsh but it installs/removes cleanly and seems > to work properly during my short test on a Solaris 9 Sparc and a Solaris 10 > x86. Man page also works. Thanks, Peter, much obliged. My only remaining problem is that I can't log in to www.opencsw.org, which the documentation suggests is the location of /home/newpkgs. None of my keys are accepted. Or, rather, there are "Too many authentication failures" after all of my keys have been tried. I can log in to login.opencsw.org, and from there to build8s and build8x.