From maciej at opencsw.org Sat Mar 1 13:22:31 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sat, 1 Mar 2014 12:22:31 +0000 Subject: New check: files needed but missing Message-ID: Hello maintainers, While debugging a piece of checkpkg code, I've found out that there was a bit of logic previously missed, that I needed to patch up. Checkpkg uses a concept of needed files: if there is a symlink in a package, that symlink needs the target file. Checkpkg will look if anything provides the target file; it could be the same package as the symlink, or another package. If it's another package, a dependency must be declared. The problematic part was when no package was reported to provide the needed file. Previously, checkpkg would ignore this situation. This would cause it to e.g. not report dangling symlinks. I've closed that hole in this commit: https://sourceforge.net/p/gar/code/23092/tree//csw/mgar/gar/v2/lib/python/checkpkg_lib.py?diff=52a9da7b5fcbc97e0c48c34f:23091 This means you might see new error tags in your packages, the error tag name is: file-needed-but-no-package-satisfies-it This check should only catch real problems in packages. If this check misfires for you, please let me know and we'll see how to make it better. Maciej -------------- next part -------------- An HTML attachment was scrubbed... URL: From maciej at opencsw.org Sat Mar 1 19:52:45 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sat, 1 Mar 2014 18:52:45 +0000 Subject: New code deployment imminent In-Reply-To: <20140227082108.00000d46@isg-1505> References: <20140227082108.00000d46@isg-1505> Message-ID: TL;DR all critical stuff is working again Hello maintainers, I've spent a good chunk of today unbreaking various bits of our infrastructure after the code update. Highlights: - fixed a bug in dependency detection in checkpkg - fixed catalog generation - tested full package pipeline from being built to being released In the meantime, Nikolai is helping debug the Apache problem. Maciej -------------- next part -------------- An HTML attachment was scrubbed... URL: From maciej at opencsw.org Sun Mar 2 12:35:29 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sun, 2 Mar 2014 11:35:29 +0000 Subject: Promoting kiel to stable Message-ID: We haven't integrated from unstable into kiel for ages. How about marking it as stable? Maciej -------------- next part -------------- An HTML attachment was scrubbed... URL: From dam at opencsw.org Sun Mar 2 12:59:58 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 2 Mar 2014 12:59:58 +0100 Subject: Promoting kiel to stable In-Reply-To: References: Message-ID: <944436E2-5E89-4D51-8B0D-72E798329D5F@opencsw.org> Hi Maciej, Am 02.03.2014 um 12:35 schrieb Maciej (Matchek) Blizi?ski : > We haven't integrated from unstable into kiel for ages. How about marking it as stable? Is anybody using Kiel? Are there open bugs fixed in unstable only? Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From maciej at opencsw.org Sun Mar 2 15:06:37 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sun, 2 Mar 2014 14:06:37 +0000 Subject: Promoting kiel to stable In-Reply-To: <944436E2-5E89-4D51-8B0D-72E798329D5F@opencsw.org> References: <944436E2-5E89-4D51-8B0D-72E798329D5F@opencsw.org> Message-ID: 2014-03-02 11:59 GMT+00:00 Dagobert Michelsen : > Is anybody using Kiel? Are there open bugs fixed in unstable only? User counts are hard to tell, as we all know too well. I also don't have a good overview of bug counts. But I'm reading the mailing list with mantis updates, and I don't see any major issues that would prevent us from taking a snapshot of unstable and marking it as testing. The kiel-to-stable promotion would only impact systems that subscribe to the testing and stable symlinks. The default for pkgutil is "testing", so all people who haven't touched their mirror= setting, would be affected. The kiel catalog is already getting old, and dublin is even older. Let's not let them get stale. I'll send an announcement, and flip symlinks in a week. Maciej From maciej at opencsw.org Sun Mar 2 15:41:22 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sun, 2 Mar 2014 14:41:22 +0000 Subject: 94 old _stub packages have been removed from unstable Message-ID: I've used Carsten's script that finds old _stub packages, and fed the output to our safe_remove_pkg.py. It found 94 packages that could be removed from unstable, and I did so. Some of you will receive emails about these stubs being removed. Please remove the definitions of these stub packages from your build recipes. Please don't re-upload these _stub packages to the catalog! Maciej From pfelecan at opencsw.org Sun Mar 2 16:00:10 2014 From: pfelecan at opencsw.org (Peter FELECAN) Date: Sun, 02 Mar 2014 16:00:10 +0100 Subject: Promoting kiel to stable In-Reply-To: ("Maciej (Matchek) =?utf-8?Q?Blizi=C5=84ski=22's?= message of "Sun, 2 Mar 2014 14:06:37 +0000") References: <944436E2-5E89-4D51-8B0D-72E798329D5F@opencsw.org> Message-ID: "Maciej (Matchek) Blizi?ski" writes: > 2014-03-02 11:59 GMT+00:00 Dagobert Michelsen : >> Is anybody using Kiel? Are there open bugs fixed in unstable only? > > User counts are hard to tell, as we all know too well. I also don't > have a good overview of bug counts. But I'm reading the mailing list > with mantis updates, and I don't see any major issues that would > prevent us from taking a snapshot of unstable and marking it as > testing. > > The kiel-to-stable promotion would only impact systems that subscribe > to the testing and stable symlinks. The default for pkgutil is > "testing", so all people who haven't touched their mirror= setting, > would be affected. > > The kiel catalog is already getting old, and dublin is even older. > Let's not let them get stale. I'll send an announcement, and flip > symlinks in a week. You're very right. Please do. -- Peter From maciej at opencsw.org Tue Mar 4 01:02:09 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Tue, 4 Mar 2014 00:02:09 +0000 Subject: New code deployment imminent In-Reply-To: References: <20140227082108.00000d46@isg-1505> Message-ID: I've discovered a bug in package uploads, I'm investigating. Sorry about that, please don't upload for now. From maciej at opencsw.org Tue Mar 4 01:23:58 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Tue, 4 Mar 2014 00:23:58 +0000 Subject: New code deployment imminent In-Reply-To: References: <20140227082108.00000d46@isg-1505> Message-ID: Bug found and fixed. https://sourceforge.net/p/gar/code/23125/ Feel free to upload again! From maciej at opencsw.org Tue Mar 4 01:31:07 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Tue, 4 Mar 2014 00:31:07 +0000 Subject: New code deployment imminent In-Reply-To: References: <20140227082108.00000d46@isg-1505> Message-ID: 2014-03-04 0:23 GMT+00:00 Maciej (Matchek) Blizi?ski : > Bug found and fixed. > > https://sourceforge.net/p/gar/code/23125/ > > Feel free to upload again! I forgot to mention: this is a client side fix, so you must update your sources (mgar up --all), otherwise csw-upload-pkg will not do what you expect it to! One more thing, I've created /opt/csw/bin/csw-upload-pkg on login, so you can just type 'csw-upload-pkg' on login and it will work for you, but /opt/csw/bin/csw-upload-pkg is only a wrapper which runs csw-upload-pkg from your GAR sources. Maciej From dam at opencsw.org Tue Mar 4 08:19:56 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 4 Mar 2014 08:19:56 +0100 Subject: New code deployment imminent In-Reply-To: References: <20140227082108.00000d46@isg-1505> Message-ID: Hi Maciej, Am 04.03.2014 um 01:02 schrieb Maciej (Matchek) Blizi?ski : > I've discovered a bug in package uploads, I'm investigating. Sorry > about that, please don't upload for now. I uploaded a new version of mbuffer yesterday which is completely gone now: http://www.opencsw.org/packages/mbuffer/ Is this related? Should I reupload everything again? Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From dam at opencsw.org Tue Mar 4 09:00:28 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 4 Mar 2014 09:00:28 +0100 Subject: OSQA needs updating: MySQL experts question In-Reply-To: References: <01F766C9-FF0F-47F3-8675-64A8BB8DB594@opencsw.org> <145637A2-310D-4EA2-8CB7-D60417A00C76@opencsw.org> Message-ID: <37252AD3-7499-4EC5-85D5-A8A6973EA797@opencsw.org> Hi Carsten, Am 28.02.2014 um 09:38 schrieb Carsten Grzemba : > It seems to be that we have to upgrade MySQL. Did you manage to do an upgrade? Regardless of that let me put big thanks on you, osqa is already much faster! :-) Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From dam at opencsw.org Tue Mar 4 09:34:34 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 4 Mar 2014 09:34:34 +0100 Subject: Direct binding and libumem Message-ID: <384D3FDA-B0C1-4BAD-833B-CE25739AA0E7@opencsw.org> Hi folks, I just encountered an interesting case: when compiling the latest xjobs (which does not have funky linker magic at all) I got the following checkpkg result: > CHECKPKG_OVERRIDES_CSWxjobs += no-direct-binding|/opt/csw/bin/xjobs|is|not|directly|bound|to|soname|libumem.so.1 Inspection of the build process showed that -B direct is correctly passed. Using the excellent documentation at the checkpkg error tag wiki at http://wiki.opencsw.org/checkpkg-error-tags#toc31 showed the following issue: > dam at unstable10s [unstable10s]:/home/dam/mgar/pkg/xjobs/trunk > $(mgar show-buildsys | cut -f1)/bin/check_db_symbols work/solaris10-sparc/pkgroot/opt/csw/bin/xjobs > Use of uninitialized value $library_path in string eq at /home/dam/mgar/pkg/.buildsys/v2/bin/check_db_symbols line 64, <$ldd> line 4. > Use of uninitialized value $soname in hash element at /home/dam/mgar/pkg/.buildsys/v2/bin/check_db_symbols line 67, <$ldd> line 4. > > Library Directly bound Not directly bound Not directly bindable > libumem.so.1 0 0 3 > libc.so.1 74 0 2 > libm.so.2 2 0 0 It seems that some libraries contain only not directly bindable symbols, essentially these should be excluded from the test. @Yann: Do you think this is possible? Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From maciej at opencsw.org Tue Mar 4 09:39:00 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Tue, 4 Mar 2014 08:39:00 +0000 Subject: New code deployment imminent In-Reply-To: References: <20140227082108.00000d46@isg-1505> Message-ID: 2014-03-04 7:19 GMT+00:00 Dagobert Michelsen : > I uploaded a new version of mbuffer yesterday which is completely gone now: > http://www.opencsw.org/packages/mbuffer/ > Is this related? Should I reupload everything again? Please re-run csw-upload-pkg against your packages. This is the symptom of the bug I fixed yesterday. Maciej From dam at opencsw.org Tue Mar 4 14:06:22 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 4 Mar 2014 14:06:22 +0100 Subject: What happened to bison? In-Reply-To: <3fdcv60KW9zx2@mail.opencsw.org> References: <3fdcv60KW9zx2@mail.opencsw.org> Message-ID: Hi folks, I just got this email: Am 04.03.2014 um 13:28 schrieb Catalog update notifier: ... > You no longer maintain packages: > - bison-3.0.2,REV=2014.01.07-SunOS5.10-i386-CSW.pkg.gz > In catalogs: > - bratislava: i386 (5.10, 5.11) > > - bison-3.0.2,REV=2014.01.07-SunOS5.10-sparc-CSW.pkg.gz > In catalogs: > - bratislava: sparc (5.10, 5.11) > > - bison-2.5,REV=2011.05.16-SunOS5.9-i386-CSW.pkg.gz > In catalogs: > - bratislava: i386 (5.9) > > - bison-2.5,REV=2011.05.16-SunOS5.9-sparc-CSW.pkg.gz > In catalogs: > - bratislava: sparc (5.9) It is perfectly ok because Maciej kindly rebuild it, but now the view is missing: http://www.opencsw.org/packages/bison/ Any idea what happened here? Best regards -- Dago From bwalton at opencsw.org Tue Mar 4 15:21:33 2014 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 4 Mar 2014 14:21:33 +0000 Subject: What happened to bison? In-Reply-To: References: <3fdcv60KW9zx2@mail.opencsw.org> Message-ID: This could be a result of the web automation not catching up with the catalog changes... Or we could be pointed at the catalog that was just deprecated. I'll take a look tonight. The web view only presents what mysql records so I suspect it's an issue in the back end processes. Thanks -Ben On Mar 4, 2014 1:06 PM, "Dagobert Michelsen" wrote: > Hi folks, > > I just got this email: > > Am 04.03.2014 um 13:28 schrieb Catalog update notifier: > ... > > You no longer maintain packages: > > - bison-3.0.2,REV=2014.01.07-SunOS5.10-i386-CSW.pkg.gz > > In catalogs: > > - bratislava: i386 (5.10, 5.11) > > > > - bison-3.0.2,REV=2014.01.07-SunOS5.10-sparc-CSW.pkg.gz > > In catalogs: > > - bratislava: sparc (5.10, 5.11) > > > > - bison-2.5,REV=2011.05.16-SunOS5.9-i386-CSW.pkg.gz > > In catalogs: > > - bratislava: i386 (5.9) > > > > - bison-2.5,REV=2011.05.16-SunOS5.9-sparc-CSW.pkg.gz > > In catalogs: > > - bratislava: sparc (5.9) > > It is perfectly ok because Maciej kindly rebuild it, but now the view is > missing: > http://www.opencsw.org/packages/bison/ > > Any idea what happened here? > > > Best regards > > -- Dago -------------- next part -------------- An HTML attachment was scrubbed... URL: From bwalton at opencsw.org Tue Mar 4 21:22:37 2014 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 4 Mar 2014 20:22:37 +0000 Subject: Fwd: [Webmaster Tools] http://www.opencsw.org/: Increase in server errors In-Reply-To: References: Message-ID: FYI, the apache issue is affecting googlebot's ability to Crawl us... Thanks -Ben ---------- Forwarded message ---------- From: Date: Mar 3, 2014 8:10 AM Subject: [Webmaster Tools] http://www.opencsw.org/: Increase in server errors To: Cc: [image: Google Logo] http://www.opencsw.org/: Increase in server errors Google detected a significant increase in the number of URLs we could not access. It is likely that your server had an internal error or was busy when attempting to process these requests. ------------------------------ *Recommended action* - Check the Crawl Errorspage in Webmaster Tools. - Check your scripts and script permissions. - Examine the log files on the server for your site for scripts or pages that might be crashing. - Consider addressing the load on your server. Learn more in our Help Center. ------------------------------ Got feedback? Leave it here. Be sure to include this message ID: [WMT-92516] *Google Inc.* 1600 Amphitheatre Parkway Mountain View, CA 94043 | Unsubscribe. -------------- next part -------------- An HTML attachment was scrubbed... URL: From yann at pleiades.fr.eu.org Wed Mar 5 00:00:08 2014 From: yann at pleiades.fr.eu.org (Yann Rouillard) Date: Wed, 5 Mar 2014 00:00:08 +0100 Subject: Direct binding and libumem In-Reply-To: <384D3FDA-B0C1-4BAD-833B-CE25739AA0E7@opencsw.org> References: <384D3FDA-B0C1-4BAD-833B-CE25739AA0E7@opencsw.org> Message-ID: Hi Dago, Yes, that's a known special case that happens in some rare cases. In your case, the binary uses only malloc, free, realloc and that set of symbols can't be directly bound to, wether it is in libumem or libc. I suppose this is done on purpose to allow to change the memory allocator. I already excluded some native libraries and I will add libumem to the list. Even better would be to check if the linked symbols can't be directly bound to but that would make the test more complex and I don't think it's worth it for now. Yann 2014-03-04 9:34 GMT+01:00 Dagobert Michelsen : > Hi folks, > > I just encountered an interesting case: when compiling the latest xjobs > (which does > not have funky linker magic at all) I got the following checkpkg result: > > > CHECKPKG_OVERRIDES_CSWxjobs += > no-direct-binding|/opt/csw/bin/xjobs|is|not|directly|bound|to|soname|libumem.so.1 > > > Inspection of the build process showed that -B direct is correctly passed. > Using the excellent documentation at the checkpkg error tag wiki at > http://wiki.opencsw.org/checkpkg-error-tags#toc31 > showed the following issue: > > > dam at unstable10s [unstable10s]:/home/dam/mgar/pkg/xjobs/trunk > $(mgar > show-buildsys | cut -f1)/bin/check_db_symbols > work/solaris10-sparc/pkgroot/opt/csw/bin/xjobs > > Use of uninitialized value $library_path in string eq at > /home/dam/mgar/pkg/.buildsys/v2/bin/check_db_symbols line 64, <$ldd> line 4. > > Use of uninitialized value $soname in hash element at > /home/dam/mgar/pkg/.buildsys/v2/bin/check_db_symbols line 67, <$ldd> line 4. > > > > Library Directly bound Not directly bound Not > directly bindable > > libumem.so.1 0 0 > 3 > > libc.so.1 74 0 > 2 > > libm.so.2 2 0 > 0 > > > It seems that some libraries contain only not directly bindable symbols, > essentially these should be excluded from the test. > @Yann: Do you think this is possible? > > > Best regards > > -- Dago > > -- > "You don't become great by trying to be great, you become great by wanting > to do something, > and then doing it so hard that you become great in the process." - xkcd > #896 > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From maciej at opencsw.org Wed Mar 5 02:18:10 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Wed, 5 Mar 2014 01:18:10 +0000 Subject: [Webmaster Tools] http://www.opencsw.org/: Increase in server errors In-Reply-To: References: Message-ID: This one could a different issue, it could be problem with our OSQA installation. Carsten is working on it. From maciej at opencsw.org Wed Mar 5 02:43:05 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Wed, 5 Mar 2014 01:43:05 +0000 Subject: What happened to bison? In-Reply-To: References: <3fdcv60KW9zx2@mail.opencsw.org> Message-ID: 2014-03-04 14:21 GMT+00:00 Ben Walton : > The web view only presents what mysql records so I suspect it's an issue in > the back end processes Every time I try to find that code, I fail. I know it's not in the mgar/pkg repo, so I usually start with the second one, where we keep our website and a bunch of tools: https://sourceforge.net/p/opencsw/code/HEAD/tree/ ...then I look at the subdirectories one by one, but I can't find what I'm looking for. Maciej From dam at opencsw.org Wed Mar 5 08:04:01 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 5 Mar 2014 08:04:01 +0100 Subject: Catalog broken again Message-ID: <71F41C49-4AF5-4AAD-9B3C-DBCD8F7203F0@opencsw.org> Hi folks, there is another cyclic dependency which prohibits a catalog push: > + chkcat -e catalog > Use of uninitialized value within %pkgdep in string eq at /opt/csw/bin/chkcat line 115. > Use of uninitialized value in split at /opt/csw/bin/chkcat line 115. > > ERROR! Dependency CSWlibadmsslutil0 of package CSW389-admin is missing. > > ERROR! Dependency CSWlibadmsslutil0 of package CSW389-adminutil-dev is missing. > > ERROR! Dependency CSWlibadmsslutil0 of package CSWlibds-admin-serv0 is missing. Carsten, would you mind having a look? Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From bwalton at opencsw.org Wed Mar 5 08:43:59 2014 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 5 Mar 2014 07:43:59 +0000 Subject: [Webmaster Tools] http://www.opencsw.org/: Increase in server errors In-Reply-To: References: Message-ID: OK. I didn't look at the info behind the link. I should move the ownership of this to an opencsw address. Any volunteers? Thanks -Ben On Mar 5, 2014 1:19 AM, "Maciej (Matchek) Blizi?ski" wrote: > This one could a different issue, it could be problem with our OSQA > installation. > > Carsten is working on it. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From grzemba at contac-dt.de Wed Mar 5 09:47:23 2014 From: grzemba at contac-dt.de (Carsten Grzemba) Date: Wed, 05 Mar 2014 09:47:23 +0100 Subject: [Webmaster Tools] http://www.opencsw.org/: Increase in server errors In-Reply-To: References: Message-ID: It was URI /communtiy (OSCA) related. I have done some updates at robots.txt and will observe it. Carsten Am 05.03.14 schrieb Ben Walton : > > OK. I didn't look at the info behind the link. I should move the ownership of this to an opencsw address. Any volunteers? > > Thanks > > -Ben > On Mar 5, 2014 1:19 AM, "Maciej (Matchek) Blizi?ski" wrote: > > > This one could a different issue, it could be problem with our OSQA > > > > installation. > > > > > > > > Carsten is working on it. > > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From maciej at opencsw.org Wed Mar 5 21:52:10 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Wed, 5 Mar 2014 20:52:10 +0000 Subject: [syslog_ng 0005155]: please upgrade to 3.5 In-Reply-To: <63152d13f73f38408a8ae79188b35264@www.opencsw.org> References: <63152d13f73f38408a8ae79188b35264@www.opencsw.org> Message-ID: Hey guys, I've received this request to upgrade syslog-ng. Is there anyone here who uses it? If so, could you take over the package? Maciej Em 05/03/2014 04:17, "Mantis Bug Tracker via bug-notifications" < bug-notifications at lists.opencsw.org> escreveu: > > The following issue has been SUBMITTED. > ====================================================================== > https://www.opencsw.org/mantis/view.php?id=5155 > ====================================================================== > Reported By: cyclop5 > Assigned To: > ====================================================================== > Project: syslog_ng > Issue ID: 5155 > Category: upgrade > Reproducibility: have not tried > Severity: minor > Priority: normal > Status: new > ====================================================================== > Date Submitted: 2014-03-05 05:17 CET > Last Modified: 2014-03-05 05:17 CET > ====================================================================== > Summary: please upgrade to 3.5 > Description: > upstream is at version 3.5.3. > ====================================================================== > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From dam at opencsw.org Mon Mar 10 19:43:50 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 10 Mar 2014 19:43:50 +0100 Subject: Outage on thursday and friday Message-ID: Hi folks, we will be moving to a new office with all our desks and hardware. This includes the opencsw buildfarm, so there will be outages on thursday and friday as we need to physically move the hardware and also coordinate the transfer of our ip network to the new facility. Sorry for the inconvenience ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From dam at opencsw.org Thu Mar 13 11:19:49 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 13 Mar 2014 11:19:49 +0100 Subject: Farm down Message-ID: Hi folks, the farm is now being shut down. I expect it to be fully functional again tomorrow during the day. I?ll announce the availability again. Sorry for the inconvenience ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From maciej at opencsw.org Thu Mar 13 11:30:33 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 13 Mar 2014 10:30:33 +0000 Subject: Farm down In-Reply-To: References: Message-ID: Everybody, save your buffers! :) From bonivart at opencsw.org Thu Mar 13 13:21:37 2014 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 13 Mar 2014 13:21:37 +0100 Subject: Farm down In-Reply-To: References: Message-ID: On Thu, Mar 13, 2014 at 11:30 AM, Maciej (Matchek) Blizi?ski wrote: > Everybody, save your buffers! :) # sync ; sync ; sync From wilbury at opencsw.org Thu Mar 13 13:25:07 2014 From: wilbury at opencsw.org (Juraj Lutter) Date: Thu, 13 Mar 2014 13:25:07 +0100 Subject: Farm down In-Reply-To: References: Message-ID: <5321A3A3.7070502@opencsw.org> On 03/13/14 13:21, Peter Bonivart wrote: > On Thu, Mar 13, 2014 at 11:30 AM, Maciej (Matchek) Blizi?ski > wrote: >> Everybody, save your buffers! :) > # sync ; sync ; sync lockfs -w -- Juraj Lutter From dam at opencsw.org Thu Mar 13 19:36:16 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 13 Mar 2014 19:36:16 +0100 Subject: isaexec and Tcl/Tk In-Reply-To: <201402181734.s1IHYeY7020891@login.bo.opencsw.org> References: <52B30C42.3020702@opencsw.org> <79B2C570-50AB-4FA0-A27E-4BDA74BBDB8C@opencsw.org> <52E1905C.9020000@opencsw.org> <6056C3B3-3741-43A0-982F-2675F5A8454F@opencsw.org> <52FBB9FF.4090704@opencsw.org> <206DEDE1-D249-4B5A-9152-4FF5E5B67583@opencsw.org> <201402131414.s1DEExeO006421@login.bo.opencsw.org> <7F76D98A-5096-4919-BABA-5779058F2BEB@opencsw.org> <201402181326.s1IDQr9I002823@login.bo.opencsw.org> <6EF52ED5-5716-4160-942B-F0B40F458C12@opencsw.org> <1B361E20-9A20-46D3-8027-501734258DCA@opencsw.org> <201402181734.s1IHYeY7020891@login.bo.opencsw.org> Message-ID: <7F3AEDA6-E9EB-4722-B0CD-59714480DCB8@opencsw.org> Hi Oliver, Am 18.02.2014 um 18:34 schrieb Oliver Kiddle : > Dagobert Michelsen wrote: >> >> If you want you can try one of these: >> http://mirror.opencsw.org/opencsw/allpkgs/alternatives-1.3.30c%2cREV%3d2010.02.18-SunOS5.8-i386-CSW.pkg.gz >> http://mirror.opencsw.org/opencsw/allpkgs/alternatives-1.3.30c%2cREV%3d2010.02.18-SunOS5.8-sparc-CSW.pkg.gz >> >> They should work correctly on the provided packages even when in one file. > > I can't get them to do anything. Perhaps they need some database or > something setup but if all I do is install them, nothing happens when I > try to configure alternatives. The contents of /var/lib/alternatives on > my Redhat based system don't seem quite similar enough to what we have > in /opt/csw/share/alternatives. > > Note that the alternatives/trunk/files/README file in subversion says: > Phil decided that a from-scratch, CSW-custom implementation was > needed, because the debian one was hugely bloated, and the redhat > smaller one, did not play nicely with NFS-shared /opt/csw So please > dont go getting ideas that we can migrated back to redhat,etc in the > future! :) we tried, and it failed. > > Having GAR create multiple alternatives files would seem easiest to me. > If you prefer, I can modify the existing script to copy relevant entries > to /etc/opt/csw/alternatives instead of a symlink. I think it can cope > with things like package removals and upgrades with a little extra > checking. If you get the RedHat system working, I'm not opposed to it > either: I just suspect it is more work that way. As obivously neither me nor you had the impetus to fix it please go ahead and just make the adjustments to solve your immediate issue. I do feel that alternatives should be redone in a more thorough way, but I don?t want to delay that for an infinite time if it makes your life more difficult :-) Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From dam at opencsw.org Fri Mar 14 11:08:08 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 14 Mar 2014 11:08:08 +0100 Subject: Fwd: New package request : Motif: 2.1.0 and GCC: package 3.4.2 References: <38742609-3221-4604-8D44-86054968D080@opencsw.org> Message-ID: <44EC5375-192C-4CB5-976B-BFA5D04BD9E8@opencsw.org> Hi folks, I just received this request: Anfang der weitergeleiteten Nachricht: > Hi senthilkumar, > > Am 14.03.2014 um 09:24 schrieb via pkgrequests : >> A new package request has been received from senthilkumar (mailto:skandasamy at ctepl.com). Motif: 2.1.0 and GCC: package 3.4.2 is requested to be added to our catalog. >> >> Here is the attached message : >> >> I want this two packages >> >> Motif: 2.1.0 > > This is probably some work, I?ll forward the request to the maintainers list, maybe someone > is interested in the task. Maybe anyone is interested in packaging up MOTIF? Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From grzemba at contac-dt.de Fri Mar 14 15:56:50 2014 From: grzemba at contac-dt.de (Carsten Grzemba) Date: Fri, 14 Mar 2014 15:56:50 +0100 Subject: OSQA local passwords In-Reply-To: References: Message-ID: Hi, the new OSQA has an other hash algorithm for local stores passwords so that it seems to be that all migrated users can only login with external authentication providers if they have this setup already. So I guess I have to set a password for all how have no external providers. Carsten -------------- next part -------------- An HTML attachment was scrubbed... URL: From jh at opencsw.org Fri Mar 14 16:03:19 2014 From: jh at opencsw.org (Jan Holzhueter) Date: Fri, 14 Mar 2014 16:03:19 +0100 Subject: OSQA local passwords In-Reply-To: References: Message-ID: <53231A37.4090304@opencsw.org> Am 14.03.14 15:56, schrieb Carsten Grzemba: > Hi, > > the new OSQA has an other hash algorithm for local stores passwords so > that it seems to be that all migrated users can only login with > external authentication providers if they have this setup already. > > So I guess I have to set a password for all how have no external providers. doesn password reset work? http://www.opencsw.org/community/account/tempsignin/ Greetings Jan From grzemba at contac-dt.de Fri Mar 14 16:11:55 2014 From: grzemba at contac-dt.de (Carsten Grzemba) Date: Fri, 14 Mar 2014 16:11:55 +0100 Subject: OSQA local passwords In-Reply-To: References: <53231A37.4090304@opencsw.org> Message-ID: Am 14.03.14 schrieb Jan Holzhueter : > Am 14.03.14 15:56, schrieb Carsten Grzemba: > > Hi, > > > > the new OSQA has an other hash algorithm for local stores passwords so > > that it seems to be that all migrated users can only login with > > external authentication providers if they have this setup already. > > > > So I guess I have to set a password for all how have no external providers. > > doesn password reset work? > > http://www.opencsw.org/community/account/tempsignin/ > > Ah.., this link produce the very long exception in the log ;-) -------------- next part -------------- An HTML attachment was scrubbed... URL: From slowfranklin at opencsw.org Fri Mar 14 22:10:36 2014 From: slowfranklin at opencsw.org (slowfranklin) Date: Fri, 14 Mar 2014 22:10:36 +0100 Subject: Kerberos on unstable11x Message-ID: <3FC818E0-817E-469D-AE95-488EAB48F77A@opencsw.org> %%Resend with correct sender Hi Dago, can you please apply the Kerberos fix to unstable11s as well: -bash-4.2$ hostname unstable11s -bash-4.2$ krb5-config --libs gssapi -L/opt/csw/lib -R/opt/csw/lib -mcpu=v9 -Wa,-L/opt/csw/lib -lintl -L/opt/csw/lib -lgssapi_krb5 -lkrb5 -lk5crypto -lcom_err -bash-4.2$ mgar package ... checking for gss_acquire_cred... no configure: error: GSSAPI: required function gss_acquire_cred missing /home/slowfranklin/opencsw/.buildsys/v2/gar//gar.lib.mk:835: recipe for target 'configure-work/solaris11-sparc/build-isa-sparcv8plus/netatalk-3.1.1/configure' failed gmake[1]: *** [configure-work/solaris11-sparc/build-isa-sparcv8plus/netatalk-3.1.1/configure] Error 1 gmake[1]: Leaving directory '/home/slowfranklin/opencsw/netatalk/trunk' gar/gar.mk:198: recipe for target 'merge-isa-sparcv8plus' failed gmake: *** [merge-isa-sparcv8plus] Error 2 -bash-4.2$ Thanks! -slow From dam at opencsw.org Sat Mar 15 13:13:09 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Sat, 15 Mar 2014 13:13:09 +0100 Subject: Farm fully functional again Message-ID: <9C5AEFB8-4A7A-421B-9930-8709A2406B12@opencsw.org> Hi folks, the farm should be fully functional again. Please let me know if you encounter any issues. Best regards -- Dago From dam at opencsw.org Sat Mar 15 13:13:40 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Sat, 15 Mar 2014 13:13:40 +0100 Subject: Kerberos on unstable11x In-Reply-To: <3FC818E0-817E-469D-AE95-488EAB48F77A@opencsw.org> References: <3FC818E0-817E-469D-AE95-488EAB48F77A@opencsw.org> Message-ID: <81494B3E-99D5-4C85-ABA1-FBDE545D1996@opencsw.org> Hi Slow, Am 14.03.2014 um 22:10 schrieb slowfranklin: > %%Resend with correct sender > > Hi Dago, > > can you please apply the Kerberos fix to unstable11s as well: > > -bash-4.2$ hostname > unstable11s > -bash-4.2$ krb5-config --libs gssapi > -L/opt/csw/lib -R/opt/csw/lib -mcpu=v9 -Wa,-L/opt/csw/lib -lintl -L/opt/csw/lib -lgssapi_krb5 -lkrb5 -lk5crypto -lcom_err > -bash-4.2$ mgar package > ... > checking for gss_acquire_cred... no > configure: error: GSSAPI: required function gss_acquire_cred missing > /home/slowfranklin/opencsw/.buildsys/v2/gar//gar.lib.mk:835: recipe for target 'configure-work/solaris11-sparc/build-isa-sparcv8plus/netatalk-3.1.1/configure' failed > gmake[1]: *** [configure-work/solaris11-sparc/build-isa-sparcv8plus/netatalk-3.1.1/configure] Error 1 > gmake[1]: Leaving directory '/home/slowfranklin/opencsw/netatalk/trunk' > gar/gar.mk:198: recipe for target 'merge-isa-sparcv8plus' failed > gmake: *** [merge-isa-sparcv8plus] Error 2 > -bash-4.2$ > > Thanks! > -slow Sure, done. Are you making a special version for Solaris 11? Best regards -- Dago From slowfranklin at opencsw.org Sat Mar 15 13:45:41 2014 From: slowfranklin at opencsw.org (slowfranklin) Date: Sat, 15 Mar 2014 13:45:41 +0100 Subject: Kerberos on unstable11x In-Reply-To: <81494B3E-99D5-4C85-ABA1-FBDE545D1996@opencsw.org> References: <3FC818E0-817E-469D-AE95-488EAB48F77A@opencsw.org> <81494B3E-99D5-4C85-ABA1-FBDE545D1996@opencsw.org> Message-ID: Am 15.03.2014 um 13:13 schrieb Dagobert Michelsen : > Hi Slow, > > Am 14.03.2014 um 22:10 schrieb slowfranklin: >> %%Resend with correct sender >> >> Hi Dago, >> >> can you please apply the Kerberos fix to unstable11s as well: >> >> -bash-4.2$ hostname >> unstable11s >> -bash-4.2$ krb5-config --libs gssapi >> -L/opt/csw/lib -R/opt/csw/lib -mcpu=v9 -Wa,-L/opt/csw/lib -lintl -L/opt/csw/lib -lgssapi_krb5 -lkrb5 -lk5crypto -lcom_err >> -bash-4.2$ mgar package >> ... >> checking for gss_acquire_cred... no >> configure: error: GSSAPI: required function gss_acquire_cred missing >> /home/slowfranklin/opencsw/.buildsys/v2/gar//gar.lib.mk:835: recipe for target 'configure-work/solaris11-sparc/build-isa-sparcv8plus/netatalk-3.1.1/configure' failed >> gmake[1]: *** [configure-work/solaris11-sparc/build-isa-sparcv8plus/netatalk-3.1.1/configure] Error 1 >> gmake[1]: Leaving directory '/home/slowfranklin/opencsw/netatalk/trunk' >> gar/gar.mk:198: recipe for target 'merge-isa-sparcv8plus' failed >> gmake: *** [merge-isa-sparcv8plus] Error 2 >> -bash-4.2$ >> >> Thanks! >> -slow > > Sure, done. Thanks! In the meantime I helped myself with defining an alias to a modified copy of the krb-config script. Was somewhat surprised though to see that this method can be used to bypass PASS as set by the buildenv. :) > Are you making a special version for Solaris 11? yes. Netatalk on Solaris 11 is built with mDNS support which is not available on Solaris 10. -slow From raos at opencsw.org Sat Mar 15 17:35:08 2014 From: raos at opencsw.org (Rafael Ostertag) Date: Sat, 15 Mar 2014 17:35:08 +0100 Subject: OpenCSW Winter Camp 2014 Message-ID: <20140315163508.GA20130@bender.opencsw.org> Hi guys I closed the poll OpenCSW Winter Camp 2014. The earliest possible date for the Winter Camp 2014 is April 4 to April 6. So, I suggest the camp is taking place on this date. Let me know if this date is *inconvenient* to you. The next best date would be May 9 to May 11. Further, let me know if you need a hotel room. I will then forward the list to one of our assistants which will take care of the reservation and send a confirmation email. cheers rafi From pfelecan at opencsw.org Sat Mar 15 19:47:11 2014 From: pfelecan at opencsw.org (Peter FELECAN) Date: Sat, 15 Mar 2014 19:47:11 +0100 Subject: ISA specific dependencies Message-ID: I'm struggling to provide a multi ISA guile which will enable a 64 bit libopts build. The issue that I encounter is that the runtime dependencies for 32 and 64 bit ISAs are different. I looked for directions in our build documentation and in the current recipes and I didn't found a solution. There is the possibility to write: RUNTIME_DEP_PKGS_CSWlibguile2-0-22_sparc += CSWlibltdl7 But how can I write a more specific one for sparcv9 or amd64? Thank you in advance -- Peter From dam at opencsw.org Sat Mar 15 23:19:13 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Sat, 15 Mar 2014 23:19:13 +0100 Subject: ISA specific dependencies In-Reply-To: References: Message-ID: <01D3812A-3D24-4F16-AE93-AFEF13FCE5FD@opencsw.org> Hi Peter, Am 15.03.2014 um 19:47 schrieb Peter FELECAN : > I'm struggling to provide a multi ISA guile which will enable a 64 bit > libopts build. > > The issue that I encounter is that the runtime dependencies for 32 and > 64 bit ISAs are different. > > I looked for directions in our build documentation and in the current > recipes and I didn't found a solution. > > There is the possibility to write: > > RUNTIME_DEP_PKGS_CSWlibguile2-0-22_sparc += CSWlibltdl7 > > But how can I write a more specific one for sparcv9 or amd64? This would be only relevant if you split 32/64 between packages which is quite uncommon. You can always write things like RUNTIME_DEP_PKGS_-32 += CSWfoo RUNTIME_DEP_PKGS_ += $(RUNTIME_DEP_PKGS_-$(MEMORYMODEL)) but I would advise against it. Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From dam at opencsw.org Sat Mar 15 23:24:29 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Sat, 15 Mar 2014 23:24:29 +0100 Subject: OpenCSW Winter Camp 2014 In-Reply-To: <20140315163508.GA20130@bender.opencsw.org> References: <20140315163508.GA20130@bender.opencsw.org> Message-ID: Hi Rafi, Am 15.03.2014 um 17:35 schrieb Rafael Ostertag : > Hi guys > > I closed the poll OpenCSW Winter Camp 2014. > > The earliest possible date for the Winter Camp 2014 is April 4 to April 6. So, > I suggest the camp is taking place on this date. > > Let me know if this date is *inconvenient* to you. The next best date would be > May 9 to May 11. > > Further, let me know if you need a hotel room. I will then forward the list to > one of our assistants which will take care of the reservation and send a > confirmation email. I would need two rooms from Friday until Monday for me and Daniel who again will join the camp! Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From maciej at opencsw.org Sun Mar 16 13:52:41 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sun, 16 Mar 2014 12:52:41 +0000 Subject: Faster catalog generation In-Reply-To: References: Message-ID: 2013-05-15 9:40 GMT+01:00 Maciej (Matchek) Blizi?ski : > I spent some time yesterday working on the buildfarm[1], and managed > to remove one bottleneck and cut down the catalog generation time from > ~80 minutes to ~32 minutes. New catalog generation time: 8 minutes. The catalog is now generated once every hour, and the generation is lazy, meaning that if there are no changes to a given catalog, no new index files ("catalog" and "description") are generated. So this change both shortens the time you have to wait for new packages to be pushed out, and reduces the file churn on the mirrors. I also updated the projects page to indicate which projects are complete: http://wiki.opencsw.org/projects-available Maciej From pfelecan at opencsw.org Sun Mar 16 19:41:56 2014 From: pfelecan at opencsw.org (Peter FELECAN) Date: Sun, 16 Mar 2014 19:41:56 +0100 Subject: ISA specific dependencies References: <01D3812A-3D24-4F16-AE93-AFEF13FCE5FD@opencsw.org> Message-ID: Dagobert Michelsen writes: > Hi Peter, > > Am 15.03.2014 um 19:47 schrieb Peter FELECAN : >> I'm struggling to provide a multi ISA guile which will enable a 64 bit >> libopts build. >> >> The issue that I encounter is that the runtime dependencies for 32 and >> 64 bit ISAs are different. >> >> I looked for directions in our build documentation and in the current >> recipes and I didn't found a solution. >> >> There is the possibility to write: >> >> RUNTIME_DEP_PKGS_CSWlibguile2-0-22_sparc += CSWlibltdl7 >> >> But how can I write a more specific one for sparcv9 or amd64? > > This would be only relevant if you split 32/64 between packages which is > quite uncommon. You can always write things like > RUNTIME_DEP_PKGS_-32 += CSWfoo > RUNTIME_DEP_PKGS_ += $(RUNTIME_DEP_PKGS_-$(MEMORYMODEL)) > but I would advise against it. The issue is that when the 64 bit component are built the dependencies are declared as missing, I add them and when the 32 bit component are built, the same dependencies are declared in surplus! The immediate correction would be to declare these dependencies as missing and as surplus ? But that is not what I whish. Here is the flip-flop from the build log: When building without the dependencies: * Dependency issues of CSWguile: * CSWlibintl8 is needed by CSWguile, because: * - opt/csw/bin/sparcv9/guile needs the libintl.so.8 soname * - opt/csw/bin/sparcv9/guile needs the libintl.so.8 soname * - opt/csw/bin/guile needs the libintl.so.8 soname * RUNTIME_DEP_PKGS_CSWguile += CSWlibintl8 * Dependency issues of CSWlibguilereadline-v18-18: * CSWlibffi5 is needed by CSWlibguilereadline-v18-18, because: * - opt/csw/lib/libguilereadline-v-18.so.18.0.0 needs the libffi.so.5 soname * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libffi.so.5 soname * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libffi.so.5 soname * RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibffi5 * CSWlibgc1 is needed by CSWlibguilereadline-v18-18, because: * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libgc.so.1 soname * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libgc.so.1 soname * - opt/csw/lib/libguilereadline-v-18.so.18.0.0 needs the libgc.so.1 soname * RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibgc1 * CSWlibgmp10 is needed by CSWlibguilereadline-v18-18, because: * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libgmp.so.10 soname * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libgmp.so.10 soname * - opt/csw/lib/libguilereadline-v-18.so.18.0.0 needs the libgmp.so.10 soname * RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibgmp10 * CSWlibltdl7 is needed by CSWlibguilereadline-v18-18, because: * - opt/csw/lib/libguilereadline-v-18.so.18.0.0 needs the libltdl.so.7 soname * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libltdl.so.7 soname * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libltdl.so.7 soname * RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibltdl7 * CSWlibncurses5 is needed by CSWlibguilereadline-v18-18, because: * - opt/csw/lib/libguilereadline-v-18.so.18.0.0 needs the libncurses.so.5 soname * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libncurses.so.5 soname * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libncurses.so.5 soname * RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibncurses5 * Dependency issues of CSWguilelib12: * If you don't know of any reasons to include these dependencies, you might remove them: * ? CSWlibguile2-0-22 * ? CSWlibguilereadline-v18-18 # Checkpkg suggests adding the following lines to the GAR recipe: # This is a summary; see above for details. RUNTIME_DEP_PKGS_CSWguile += CSWlibintl8 RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibltdl7 RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibffi5 RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibncurses5 RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibgmp10 RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibgc1 If any of the reported errors were false positives, you can override them pasting the lines below to the GAR recipe. CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += missing-dependency|CSWlibltdl7 CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += missing-dependency|CSWlibffi5 CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += missing-dependency|CSWlibncurses5 CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += missing-dependency|CSWlibgmp10 CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += missing-dependency|CSWlibgc1 CHECKPKG_OVERRIDES_CSWguile += missing-dependency|CSWlibintl8 When building with the dependencies: * Dependency issues of CSWlibguilereadline-v18-18: * If you don't know of any reasons to include these dependencies, you might remove them: * ? CSWlibffi5 * ? CSWlibgc1 * ? CSWlibgmp10 * ? CSWlibltdl7 * ? CSWlibncurses5 * Dependency issues of CSWguile: * If you don't know of any reasons to include these dependencies, you might remove them: * ? CSWlibintl8 * Dependency issues of CSWguilelib12: * If you don't know of any reasons to include these dependencies, you might remove them: * ? CSWlibguile2-0-22 * ? CSWlibguilereadline-v18-18 If any of the reported errors were false positives, you can override them pasting the lines below to the GAR recipe. CHECKPKG_OVERRIDES_CSWguile += surplus-dependency|CSWlibintl8 CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += surplus-dependency|CSWlibgc1 CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += surplus-dependency|CSWlibltdl7 CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += surplus-dependency|CSWlibffi5 CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += surplus-dependency|CSWlibncurses5 CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += surplus-dependency|CSWlibgmp10 Something is rotten and frankly I don't wish to add overrides as it falsifies the dependencies. The build logs are available on the build farm in the ~pfelecan/logs as guile-surplus and guile-missing. -- Peter From dam at opencsw.org Sun Mar 16 21:28:08 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 16 Mar 2014 21:28:08 +0100 Subject: ISA specific dependencies In-Reply-To: References: <01D3812A-3D24-4F16-AE93-AFEF13FCE5FD@opencsw.org> Message-ID: Hi Peter, Am 16.03.2014 um 19:41 schrieb Peter FELECAN: > Dagobert Michelsen writes: >> Am 15.03.2014 um 19:47 schrieb Peter FELECAN : >>> I'm struggling to provide a multi ISA guile which will enable a 64 bit >>> libopts build. >>> >>> The issue that I encounter is that the runtime dependencies for 32 and >>> 64 bit ISAs are different. >>> >>> I looked for directions in our build documentation and in the current >>> recipes and I didn't found a solution. >>> >>> There is the possibility to write: >>> >>> RUNTIME_DEP_PKGS_CSWlibguile2-0-22_sparc += CSWlibltdl7 >>> >>> But how can I write a more specific one for sparcv9 or amd64? >> >> This would be only relevant if you split 32/64 between packages which is >> quite uncommon. You can always write things like >> RUNTIME_DEP_PKGS_-32 += CSWfoo >> RUNTIME_DEP_PKGS_ += $(RUNTIME_DEP_PKGS_-$(MEMORYMODEL)) >> but I would advise against it. > > The issue is that when the 64 bit component are built the dependencies > are declared as missing, I add them and when the 32 bit component are > built, the same dependencies are declared in surplus! The immediate > correction would be to declare these dependencies as missing and as > surplus ? But that is not what I whish. > > Here is the flip-flop from the build log: > > When building without the dependencies: > > * Dependency issues of CSWguile: > * CSWlibintl8 is needed by CSWguile, because: > * - opt/csw/bin/sparcv9/guile needs the libintl.so.8 soname > * - opt/csw/bin/sparcv9/guile needs the libintl.so.8 soname > * - opt/csw/bin/guile needs the libintl.so.8 soname > * RUNTIME_DEP_PKGS_CSWguile += CSWlibintl8 > * Dependency issues of CSWlibguilereadline-v18-18: > * CSWlibffi5 is needed by CSWlibguilereadline-v18-18, because: > * - opt/csw/lib/libguilereadline-v-18.so.18.0.0 needs the libffi.so.5 soname > * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the > libffi.so.5 soname > * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the > libffi.so.5 soname > * RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibffi5 > * CSWlibgc1 is needed by CSWlibguilereadline-v18-18, because: > * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libgc.so.1 > soname > * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the libgc.so.1 > soname > * - opt/csw/lib/libguilereadline-v-18.so.18.0.0 needs the libgc.so.1 soname > * RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibgc1 > * CSWlibgmp10 is needed by CSWlibguilereadline-v18-18, because: > * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the > libgmp.so.10 soname > * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the > libgmp.so.10 soname > * - opt/csw/lib/libguilereadline-v-18.so.18.0.0 needs the libgmp.so.10 > soname > * RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibgmp10 > * CSWlibltdl7 is needed by CSWlibguilereadline-v18-18, because: > * - opt/csw/lib/libguilereadline-v-18.so.18.0.0 needs the libltdl.so.7 > soname > * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the > libltdl.so.7 soname > * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the > libltdl.so.7 soname > * RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibltdl7 > * CSWlibncurses5 is needed by CSWlibguilereadline-v18-18, because: > * - opt/csw/lib/libguilereadline-v-18.so.18.0.0 needs the libncurses.so.5 > soname > * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the > libncurses.so.5 soname > * - opt/csw/lib/sparcv9/libguilereadline-v-18.so.18.0.0 needs the > libncurses.so.5 soname > * RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibncurses5 > * Dependency issues of CSWguilelib12: > * If you don't know of any reasons to include these dependencies, you might > remove them: > * ? CSWlibguile2-0-22 > * ? CSWlibguilereadline-v18-18 > > # Checkpkg suggests adding the following lines to the GAR recipe: > # This is a summary; see above for details. > RUNTIME_DEP_PKGS_CSWguile += CSWlibintl8 > RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibltdl7 > RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibffi5 > RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibncurses5 > RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibgmp10 > RUNTIME_DEP_PKGS_CSWlibguilereadline-v18-18 += CSWlibgc1 > If any of the reported errors were false positives, you can override them > pasting the lines below to the GAR recipe. > CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += missing-dependency|CSWlibltdl7 > CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += missing-dependency|CSWlibffi5 > CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += missing-dependency|CSWlibncurses5 > CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += missing-dependency|CSWlibgmp10 > CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += missing-dependency|CSWlibgc1 > CHECKPKG_OVERRIDES_CSWguile += missing-dependency|CSWlibintl8 > > When building with the dependencies: > > * Dependency issues of CSWlibguilereadline-v18-18: > * If you don't know of any reasons to include these dependencies, you might > remove them: > * ? CSWlibffi5 > * ? CSWlibgc1 > * ? CSWlibgmp10 > * ? CSWlibltdl7 > * ? CSWlibncurses5 > * Dependency issues of CSWguile: > * If you don't know of any reasons to include these dependencies, you might > remove them: > * ? CSWlibintl8 > * Dependency issues of CSWguilelib12: > * If you don't know of any reasons to include these dependencies, you might > remove them: > * ? CSWlibguile2-0-22 > * ? CSWlibguilereadline-v18-18 > If any of the reported errors were false positives, you can override them > pasting the lines below to the GAR recipe. > CHECKPKG_OVERRIDES_CSWguile += surplus-dependency|CSWlibintl8 > CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += surplus-dependency|CSWlibgc1 > CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += surplus-dependency|CSWlibltdl7 > CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += surplus-dependency|CSWlibffi5 > CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += surplus-dependency|CSWlibncurses5 > CHECKPKG_OVERRIDES_CSWlibguilereadline-v18-18 += surplus-dependency|CSWlibgmp10 > > Something is rotten and frankly I don't wish to add overrides as it > falsifies the dependencies. This looks wrong regarding the checks. IMHO when a dependency is given (regardless if it is 32 and/or 64 bit) the package should be specified and no surplus dependency should be thrown. However, I could think of a useful test that the linked libraries between files of the same name in different ISA subdirs are the same or must be overridden if not. Wrong linkages are a problem when doing 32/64. @Maciej, @Yann: What do you think? > The build logs are available on the build farm in the ~pfelecan/logs as > guile-surplus and guile-missing. Best regards -- Dago From maciej at opencsw.org Mon Mar 17 22:38:36 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Mon, 17 Mar 2014 21:38:36 +0000 Subject: Python coding project: csw-upload-pkg logic issues Message-ID: If you upload a set of packages for 1 Solaris release only, all is good. If you upload with something like "csw-upload-pkg * " and in your set of packages there is a mix of 5.9 and 5.10 packages, the uploader will doubly assign packages to 5.10 and 5.11 catalogs. This logic needs to be corrected. This link points to the specific place of code with a comment describing the issue and suggests which way to go: https://buildfarm.opencsw.org/source/xref/opencsw/csw/mgar/gar/v2/lib/python/csw_upload_pkg.py#163 There is also a set of tests: https://buildfarm.opencsw.org/source/xref/opencsw/csw/mgar/gar/v2/lib/python/csw_upload_pkg_test.py As usual, I'm offering advice and code reviews; if you mail me your patch before committing, I'll reply with a review. It's easy to test your changes, you just modify the file in your buildtree (~/opencsw/.buildsys) and run it. Or you can rewrite the whole thing in Perl. ;-) Any takers? Maciej -------------- next part -------------- An HTML attachment was scrubbed... URL: From maciej at opencsw.org Mon Mar 17 22:47:41 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Mon, 17 Mar 2014 21:47:41 +0000 Subject: Python coding project: csw-upload-pkg logic issues In-Reply-To: References: Message-ID: TL;DR everybody, do not run csw-upload-pkg with a mix of 5.9 and 5.10 packages 2014-03-17 21:38 GMT+00:00 Maciej (Matchek) Blizi?ski : > in your set of packages there is a mix of 5.9 and 5.10 packages, the uploader will doubly assign packages to 5.10 and 5.11 catalogs. To clarify: some of your 5.9 packages may end up in 5.10 and/or 5.11 catalogs even though you've given csw-upload-pkg native 5.10 and/or 5.11 packages, which are clearly a better fit. Until this problem is fixed, please do not mix 5.9, 5.10, and/or 5.11 packages in a single invocation of csw-upload-pkg. Instead, give it 5.9 packages in the first run, then 5.10 in the second run, and so on. Maciej From pfelecan at opencsw.org Tue Mar 18 20:02:26 2014 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 18 Mar 2014 20:02:26 +0100 Subject: Python coding project: csw-upload-pkg logic issues In-Reply-To: ("Maciej (Matchek) =?utf-8?Q?Blizi=C5=84ski=22's?= message of "Mon, 17 Mar 2014 21:47:41 +0000") References: Message-ID: "Maciej (Matchek) Blizi?ski" writes: > TL;DR everybody, do not run csw-upload-pkg with a mix of 5.9 and 5.10 > packages Who's still bothering with 5.9 which was EOLd a long time ago? There are so much mysteries in this world! -- Peter From pfelecan at opencsw.org Tue Mar 18 20:06:10 2014 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 18 Mar 2014 20:06:10 +0100 Subject: ISA specific dependencies In-Reply-To: (Dagobert Michelsen's message of "Sun, 16 Mar 2014 21:28:08 +0100") References: <01D3812A-3D24-4F16-AE93-AFEF13FCE5FD@opencsw.org> Message-ID: Dagobert Michelsen writes: >> Something is rotten and frankly I don't wish to add overrides as it >> falsifies the dependencies. > > This looks wrong regarding the checks. IMHO when a dependency is given (regardless > if it is 32 and/or 64 bit) the package should be specified and no surplus dependency > should be thrown. > > However, I could think of a useful test that the linked libraries between files of the > same name in different ISA subdirs are the same or must be overridden if not. Wrong > linkages are a problem when doing 32/64. > > @Maciej, @Yann: What do you think? What's the wisdom offered by my dear colleagues? Sometime, in the middle term, I'll override the dependencies if there is no other way. Till then I'm devoting myself to other, cleaner, endeavours. -- Peter From maciej at opencsw.org Tue Mar 18 20:51:14 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Tue, 18 Mar 2014 19:51:14 +0000 Subject: Python coding project: csw-upload-pkg logic issues In-Reply-To: References: Message-ID: The specific release number does not matter. The same problem happens with the 5.10/5.11 mix. -------------- next part -------------- An HTML attachment was scrubbed... URL: From dam at opencsw.org Wed Mar 19 04:05:35 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 19 Mar 2014 04:05:35 +0100 Subject: Libpq5 missing Message-ID: <8B5A99C5-55F7-41D0-970C-229829AA12D9@opencsw.org> Hi Folks, Libpq5 IS missing, could the one who removed it please reupload it again? http://buildfarm.opencsw.org/catalog-generation.log Best regards -- Dago From raos at opencsw.org Wed Mar 19 08:22:06 2014 From: raos at opencsw.org (Rafael Ostertag) Date: Wed, 19 Mar 2014 08:22:06 +0100 Subject: Libpq5 missing In-Reply-To: <8B5A99C5-55F7-41D0-970C-229829AA12D9@opencsw.org> References: <8B5A99C5-55F7-41D0-970C-229829AA12D9@opencsw.org> Message-ID: <20140319072206.GJ21815@bender.opencsw.org> Hi Dago On Wed, Mar 19, 2014 at 04:05:35AM +0100, Dagobert Michelsen wrote: > Hi Folks, > > Libpq5 IS missing, could the one who removed it please reupload it again? Due to a glitch while uploading postgresql 8.4 on Sunday, the catalog was broken, i.e. postgresql 8.4 got removed from the catalog. I suspect this being the culprit. However, I just uploaded postgresql 8.4 Yesterday, so there should be a libpq5. If it helps, I can respin the whole Postgres family and re-upload everything again... cheers rafi From maciej at opencsw.org Wed Mar 19 10:11:45 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Wed, 19 Mar 2014 09:11:45 +0000 Subject: Libpq5 missing In-Reply-To: <20140319072206.GJ21815@bender.opencsw.org> References: <8B5A99C5-55F7-41D0-970C-229829AA12D9@opencsw.org> <20140319072206.GJ21815@bender.opencsw.org> Message-ID: Rafi, we need to look why these packages vanish. This is the second time it happens, so there is a repeatable failure mode that bit us. Maciej From raos at opencsw.org Wed Mar 19 10:16:58 2014 From: raos at opencsw.org (Rafael Ostertag) Date: Wed, 19 Mar 2014 10:16:58 +0100 Subject: Libpq5 missing In-Reply-To: References: <8B5A99C5-55F7-41D0-970C-229829AA12D9@opencsw.org> <20140319072206.GJ21815@bender.opencsw.org> Message-ID: <20140319091657.GL21815@bender.opencsw.org> Hi Maciej On Wed, Mar 19, 2014 at 09:11:45AM +0000, Maciej (Matchek) Blizi??ski wrote: > Rafi, we need to look why these packages vanish. This is the second > time it happens, so there is a repeatable failure mode that bit us. Let me know if I can be of assistance. cheers rafi > > Maciej > From dam at opencsw.org Sun Mar 23 23:01:40 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 23 Mar 2014 23:01:40 +0100 Subject: New package request : Postgis In-Reply-To: <201403231014.s2NAEZ5J028193@www.opencsw.org> References: <201403231014.s2NAEZ5J028193@www.opencsw.org> Message-ID: Hi Andreas, Am 23.03.2014 um 11:14 schrieb via pkgrequests : > A new package request has been received from Andreas Hirner (mailto:andreas.hirner at dlr.de). Postgis is requested to be added to our catalog. > > Here is the attached message : > > Hello OpenCsw, > > it would be great to have Postgis (http://postgis.net/) support for Postgresql. Especially since most dependencies like geos and gdal are already included. Sounds interesting. Did you already compile it on Solaris? Maybe you are interested in maintaining a package? @Rafi: If the extension is not too big you may also be interested in building it. Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From maciej at opencsw.org Mon Mar 24 00:27:35 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sun, 23 Mar 2014 23:27:35 +0000 Subject: ISA specific dependencies In-Reply-To: References: <01D3812A-3D24-4F16-AE93-AFEF13FCE5FD@opencsw.org> Message-ID: 2014-03-16 20:28 GMT+00:00 Dagobert Michelsen : > This looks wrong regarding the checks. IMHO when a dependency is given (regardless > if it is 32 and/or 64 bit) the package should be specified and no surplus dependency > should be thrown. I agree. One reason is enough to do this. I need to reread the code to figure out what is exactly going on in checkpkg's head, this outcome surprises me. I thought it was implement in a way that makes this always consistent. One thing that could be the cause is that checkpkg currently does not understand 32/64 bit libraries: it thinks that anything that has name "foo.so.1" will satisfy a binary with "NEEDED foo.so.1". It does not look at the file type but only at the file existence. On the plus side, we now have the file type data in the database, so we have the data necessary to implement this. I don't know if this is the cause, but on the surface it looks like it could be. Maciej From raos at opencsw.org Mon Mar 24 08:07:07 2014 From: raos at opencsw.org (Rafael Ostertag) Date: Mon, 24 Mar 2014 08:07:07 +0100 Subject: New package request : Postgis In-Reply-To: References: <201403231014.s2NAEZ5J028193@www.opencsw.org> Message-ID: <20140324070707.GS21815@bender.opencsw.org> Hi On Sun, Mar 23, 2014 at 11:01:40PM +0100, Dagobert Michelsen wrote: > Hi Andreas, > > Am 23.03.2014 um 11:14 schrieb via pkgrequests : > > A new package request has been received from Andreas Hirner (mailto:andreas.hirner at dlr.de). Postgis is requested to be added to our catalog. > > > > Here is the attached message : > > > > Hello OpenCsw, > > > > it would be great to have Postgis (http://postgis.net/) support for Postgresql. Especially since most dependencies like geos and gdal are already included. > > Sounds interesting. Did you already compile it on Solaris? > Maybe you are interested in maintaining a package? > > @Rafi: If the extension is not too big you may also be interested in building it. Yes, indeed. I will give it a shot by the end of the week. cheers rafi From dam at opencsw.org Mon Mar 24 09:09:03 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 24 Mar 2014 09:09:03 +0100 Subject: Outage of buildfarm Message-ID: <895BE077-39CB-4506-BC23-3A9B84B6CBD9@opencsw.org> Hi folks, we had a power outage last night at around 22, so all buildfarm services have been down again. Unfortunately the UPS was not cabled yet but is going to be connected tomorrow morning, so there will be another downtime from 9 UTC to probably 12 (or whenever the electrician finishes the work in the power distribution). Sorry for the inconvenience ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From dam at opencsw.org Mon Mar 24 09:45:27 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 24 Mar 2014 09:45:27 +0100 Subject: Signing Message-ID: Hi, could please one of you gentlemen restart the signing? Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From dam at opencsw.org Mon Mar 24 16:03:47 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 24 Mar 2014 16:03:47 +0100 Subject: Downtime again Message-ID: <41A3A74B-FBF3-4693-AFD6-F3031A575073@opencsw.org> Hi folks, due to the connection and recabling of our UPS we will shut down all servers this evening. They will hopefully go back online tomorrow morning. Sorry for the inconvenience ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From jh at opencsw.org Tue Mar 25 19:08:40 2014 From: jh at opencsw.org (=?windows-1252?Q?Jan_Holzh=FCter?=) Date: Tue, 25 Mar 2014 19:08:40 +0100 Subject: Downtime again In-Reply-To: <41A3A74B-FBF3-4693-AFD6-F3031A575073@opencsw.org> References: <41A3A74B-FBF3-4693-AFD6-F3031A575073@opencsw.org> Message-ID: <5331C628.8060404@opencsw.org> Hi, Am 24.03.14 16:03, schrieb Dagobert Michelsen: > Hi folks, > > due to the connection and recabling of our UPS we will shut down all servers this evening. > They will hopefully go back online tomorrow morning. > sorry it took longer then expected. But we are back now. Everything should work again. If not let me now. Please enable cswsign again if someone with the key is still online :) Greetings Jan From raos at opencsw.org Tue Mar 25 21:13:46 2014 From: raos at opencsw.org (Rafael Ostertag) Date: Tue, 25 Mar 2014 21:13:46 +0100 Subject: Winter Camp 2014: Wiki Page Message-ID: <20140325201346.GW21815@bender.opencsw.org> Hi guys I finally managed to create a wiki page for the Winter Camp 2014 http://wiki.opencsw.org/wintercamp-2014 cheers rafi From dam at opencsw.org Wed Mar 26 21:51:11 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 26 Mar 2014 21:51:11 +0100 Subject: Catalog generation checks Message-ID: Hi Maciej, hi Jan, the new checks for the catalog generation are now in place. The check is in fact configured as ?passive? in Nagios, that means the status is delivered during the generation to Nagios. If the generation passes to the end everything is fine, if there is an error the script TRAPs out and sends an error. Additionally, there is a freshness check that makes sure a status is reported at least every 2 hours. Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From maciej at opencsw.org Wed Mar 26 22:24:09 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Wed, 26 Mar 2014 21:24:09 +0000 Subject: Catalog generation checks In-Reply-To: References: Message-ID: Looks like something's wrong with the generation script; the output looks wrong, the generation didn't happen. -------------- next part -------------- An HTML attachment was scrubbed... URL: From maciej at opencsw.org Thu Mar 27 01:20:45 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 27 Mar 2014 00:20:45 +0000 Subject: Changes on the website Message-ID: A guy showed up on the IRC channel tonight and expressed concern that the packages go stale and the website is not updated. We only chatted for a short while but I managed to figure out that this impression was based on the age of our php5 package and the "Latest News" section on the website. I couldn't do anything about the PHP package, but I could do something about the website. I logged on to the server and commented out the "news" and "updates" sections. We're updating the website infrequently enough that it makes no sense to subdivide our post stream in any way. The changes have been made live on the web server. I don't know what we do to push the changes to subversion. The htdocs directory is not a subversion working copy, so... how does this work? Does anybody know? Maciej -------------- next part -------------- An HTML attachment was scrubbed... URL: From maciej at opencsw.org Thu Mar 27 01:29:06 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 27 Mar 2014 00:29:06 +0000 Subject: Our PHP 5.x package is getting old Message-ID: Hello fellow maintainers, Our PHP package is aging. http://www.opencsw.org/packages/php5/ shows that the package is from February 2013. Do any us maintainers care about the PHP package enough to try and update it? Maciej -------------- next part -------------- An HTML attachment was scrubbed... URL: From jh at opencsw.org Thu Mar 27 08:35:37 2014 From: jh at opencsw.org (Jan Holzhueter) Date: Thu, 27 Mar 2014 08:35:37 +0100 Subject: Our PHP 5.x package is getting old In-Reply-To: References: Message-ID: <5333D4C9.6040509@opencsw.org> Am 27.03.14 01:29, schrieb Maciej (Matchek) Blizi?ski: > Hello fellow maintainers, > > Our PHP package is aging. http://www.opencsw.org/packages/php5/ shows > that the package is from February 2013. Do any us maintainers care about > the PHP package enough to try and update it? Dago is working on it :) Greetings Jan From dam at opencsw.org Thu Mar 27 08:55:35 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 27 Mar 2014 08:55:35 +0100 Subject: Changes on the website In-Reply-To: References: Message-ID: <0920DD5E-638D-407F-B8EB-735261184BEE@opencsw.org> Hi Maciej, Am 27.03.2014 um 01:20 schrieb Maciej (Matchek) Blizi?ski : > A guy showed up on the IRC channel tonight and expressed concern that the packages go stale and the website is not updated. We only chatted for a short while but I managed to figure out that this impression was based on the age of our php5 package and the "Latest News" section on the website. > > I couldn't do anything about the PHP package, but I could do something about the website. I logged on to the server and commented out the "news" and "updates" sections. We're updating the website infrequently enough that it makes no sense to subdivide our post stream in any way. > > The changes have been made live on the web server. I don't know what we do to push the changes to subversion. The htdocs directory is not a subversion working copy, so... how does this work? Does anybody know? Good, now if anybody releases something cool please everybody make a post about it! Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From dam at opencsw.org Thu Mar 27 08:59:22 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 27 Mar 2014 08:59:22 +0100 Subject: Catalog generation checks In-Reply-To: References: Message-ID: Hi Maciej, Am 26.03.2014 um 22:24 schrieb Maciej (Matchek) Blizi?ski : > Looks like something's wrong with the generation script; the output looks wrong, the generation didn't happen. Can you please be more specific? I just added sending the status on success and failure and the report looks good to me: http://buildfarm.opencsw.org/catalog-generation.log Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From dam at opencsw.org Thu Mar 27 09:59:58 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 27 Mar 2014 09:59:58 +0100 Subject: Our PHP 5.x package is getting old In-Reply-To: <5333D4C9.6040509@opencsw.org> References: <5333D4C9.6040509@opencsw.org> Message-ID: Hi folks, Am 27.03.2014 um 08:35 schrieb Jan Holzhueter : > Am 27.03.14 01:29, schrieb Maciej (Matchek) Blizi?ski: >> Our PHP package is aging. http://www.opencsw.org/packages/php5/ shows >> that the package is from February 2013. Do any us maintainers care about >> the PHP package enough to try and update it? > > Dago is working on it :) Currently I?m stuck: > dam at unstable10s [unstable10s]:/home/dam/mgar/pkg/php5/branches/php-5.5.x/work/solaris10-sparc/build-isa-sparcv8plus-sapi-cgi/php-5.5.10 > /home/dam/mgar/pkg/php5/branches/php-5.5.x/work/solaris10-sparc/build-isa-sparcv8plus-sapi-cgi/php-5.5.10/sapi/cli/php -n -dshort_open_tag=0 -dopen_basedir= -derror_reporting=1803 -dmemory_limit=-1 -ddetect_unicode=0 pear/install-pear-nozlib.phar -d "/opt/csw/php5/lib/php" -b "/opt/csw/php5/bin" -dp a -ds a > zsh: 16636 bus error (core dumped) -n -dshort_open_tag=0 -dopen_basedir= -derror_reporting=1803 -d -b -dp > dam at unstable10s [unstable10s]:/home/dam/mgar/pkg/php5/branches/php-5.5.x/work/solaris10-sparc/build-isa-sparcv8plus-sapi-cgi/php-5.5.10 > pstack core > core 'core' of 16636: /home/dam/mgar/pkg/php5/branches/php-5.5.x/work/solaris10-sparc/build- > 000a6500 compare_opcodes (c1d7ad, 0, ffbfe5b8, ffbfd3d8, c1d789, c1d7ad) + 1178 > 000a6944 auto_possessify (c1d767, 0, ffbfe5b8, 64, ffbfd3d8, 661eb4) + 284 > 000b0114 php_pcre_compile2 (b9, 0, 20000, 0, c1d6c0, 96372c) + d90 > 000d4138 pcre_get_compiled_regex_cache (a5e4ac, 2f, 2f, 2f, 0, 961064) + 444 > 000d4680 php_do_pcre_match (3, dde418, ffbfe770, 961064, ffbfe774, 0) + 7c > 005acaa0 zend_do_fcall_common_helper_SPEC (9ee560, 100, 100, 961064, d646cc, 9e4798) + 320 > 005ac448 execute_ex (9ee560, 9e45f0, 1, 961064, 8358c, 0) + 3c > 00533ce4 zend_execute_scripts (8, 0, ffbff3c8, 1, 9e4714, ffbfe8ac) + 678 > 0043eb94 php_execute_script (9e4f30, ffbff3c8, 9e4f30, 961064, 1, 83c00) + 32c > 005ff608 do_cli (961064, 8, 0, 83c00, 961064, 83c00) + d3c > 00600684 main (10, ffbff674, ffbff6b8, 963000, 96be44, 96be44) + 660 > 0004d9c0 _start (0, 0, 0, 0, 0, 0) + 108 Everything is committed, if you have a great idea let me know. Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From maciej at opencsw.org Thu Mar 27 10:04:01 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 27 Mar 2014 09:04:01 +0000 Subject: Catalog generation checks In-Reply-To: References: Message-ID: It looks good again. When I looked earlier there was a 'false' command which stopped the process. Our scripts run with 'set -e' so we need to make sure all commands are successful or followed by "|| true". Maybe one of the new monitoring commands failed and that stopped the script? I will save the log if I see thus problem again. -------------- next part -------------- An HTML attachment was scrubbed... URL: From dam at opencsw.org Thu Mar 27 10:12:56 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 27 Mar 2014 10:12:56 +0100 Subject: Catalog generation checks In-Reply-To: References: Message-ID: <20A019A5-B273-4B0E-AEF7-68FFBC03F47C@opencsw.org> Hi Maciej, Am 27.03.2014 um 10:04 schrieb Maciej (Matchek) Blizi?ski : > It looks good again. > > When I looked earlier there was a 'false' command which stopped the process. > I deliberately added ?false? as command while I was developing to see if the failure state is correctly propagated :-) > Our scripts run with 'set -e' so we need to make sure all commands are successful or followed by "|| true". Maybe one of the new monitoring commands failed and that stopped the script? I will save the log if I see thus problem again. > Everything should be fine now. Best regards ? Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2418 bytes Desc: not available URL: From maciej at opencsw.org Thu Mar 27 10:17:25 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 27 Mar 2014 09:17:25 +0000 Subject: Faster catalog generation In-Reply-To: References: Message-ID: New change: - The unstable catalog is updated every 10 minutes - Other catalogs are updated every 6 hours -------------- next part -------------- An HTML attachment was scrubbed... URL: From raos at opencsw.org Thu Mar 27 13:55:42 2014 From: raos at opencsw.org (Rafael Ostertag) Date: Thu, 27 Mar 2014 13:55:42 +0100 Subject: Our PHP 5.x package is getting old In-Reply-To: References: <5333D4C9.6040509@opencsw.org> Message-ID: <20140327125542.GA1133@bender.opencsw.org> Hi Dago On Thu, Mar 27, 2014 at 09:59:58AM +0100, Dagobert Michelsen wrote: > Hi folks, > > Am 27.03.2014 um 08:35 schrieb Jan Holzhueter : > > Am 27.03.14 01:29, schrieb Maciej (Matchek) Blizi??ski: > >> Our PHP package is aging. http://www.opencsw.org/packages/php5/ shows > >> that the package is from February 2013. Do any us maintainers care about > >> the PHP package enough to try and update it? > > > > Dago is working on it :) > > Currently I???m stuck: > > > dam at unstable10s [unstable10s]:/home/dam/mgar/pkg/php5/branches/php-5.5.x/work/solaris10-sparc/build-isa-sparcv8plus-sapi-cgi/php-5.5.10 > /home/dam/mgar/pkg/php5/branches/php-5.5.x/work/solaris10-sparc/build-isa-sparcv8plus-sapi-cgi/php-5.5.10/sapi/cli/php -n -dshort_open_tag=0 -dopen_basedir= -derror_reporting=1803 -dmemory_limit=-1 -ddetect_unicode=0 pear/install-pear-nozlib.phar -d "/opt/csw/php5/lib/php" -b "/opt/csw/php5/bin" -dp a -ds a > > zsh: 16636 bus error (core dumped) -n -dshort_open_tag=0 -dopen_basedir= -derror_reporting=1803 -d -b -dp > > > dam at unstable10s [unstable10s]:/home/dam/mgar/pkg/php5/branches/php-5.5.x/work/solaris10-sparc/build-isa-sparcv8plus-sapi-cgi/php-5.5.10 > pstack core > > core 'core' of 16636: /home/dam/mgar/pkg/php5/branches/php-5.5.x/work/solaris10-sparc/build- > > 000a6500 compare_opcodes (c1d7ad, 0, ffbfe5b8, ffbfd3d8, c1d789, c1d7ad) + 1178 > > 000a6944 auto_possessify (c1d767, 0, ffbfe5b8, 64, ffbfd3d8, 661eb4) + 284 > > 000b0114 php_pcre_compile2 (b9, 0, 20000, 0, c1d6c0, 96372c) + d90 > > 000d4138 pcre_get_compiled_regex_cache (a5e4ac, 2f, 2f, 2f, 0, 961064) + 444 > > 000d4680 php_do_pcre_match (3, dde418, ffbfe770, 961064, ffbfe774, 0) + 7c > > 005acaa0 zend_do_fcall_common_helper_SPEC (9ee560, 100, 100, 961064, d646cc, 9e4798) + 320 > > 005ac448 execute_ex (9ee560, 9e45f0, 1, 961064, 8358c, 0) + 3c > > 00533ce4 zend_execute_scripts (8, 0, ffbff3c8, 1, 9e4714, ffbfe8ac) + 678 > > 0043eb94 php_execute_script (9e4f30, ffbff3c8, 9e4f30, 961064, 1, 83c00) + 32c > > 005ff608 do_cli (961064, 8, 0, 83c00, 961064, 83c00) + d3c > > 00600684 main (10, ffbff674, ffbff6b8, 963000, 96be44, 96be44) + 660 > > 0004d9c0 _start (0, 0, 0, 0, 0, 0) + 108 > > Everything is committed, if you have a great idea let me know. Not a great idea, but... SIGBUS on SPARC is usually connected to miss aligned memory access. So, if feasible, you could play around with -xmemalign, e.g. -xmemalign=8i. You may want to read the man page of cc for implications of -xmemalign. cheers rafi > > > Best regards > > ??? Dago > > -- > "You don't become great by trying to be great, you become great by wanting to do something, > and then doing it so hard that you become great in the process." - xkcd #896 > From maciej at opencsw.org Sat Mar 29 16:29:15 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sat, 29 Mar 2014 15:29:15 +0000 Subject: List of packages that need to be rebuilt Message-ID: One of the lesser known URLs that we have is the one with obsolete packages; packages that can be rebuilt and packages that can be removed. It's powered by Carsten's script. Until today the output of the script was only a list of catalognames. I've spent some time today on making the output more comprehensive. Here's the result (for 5.10 intel): http://buildfarm.opencsw.org/obsolete-pkgs/PkgsToRebuildFrom_unstable_SunOS5.10_i386.html What you see on the page, is a list of packages to be rebuilt. Next to each package, there is a list of its dependencies, with the obsolete dependencies highlighted in red. These are the ones we need to remove from the RUNTIME_DEPS list when rebuilding. The list is grouped by maintainer name. My suggestion is to look at them and rebuild, without rebuilding a newer version. Improving package dependencies alone is already helpful, and upgrades can be handled separately. Maciej -------------- next part -------------- An HTML attachment was scrubbed... URL: From maciej at opencsw.org Sat Mar 29 17:24:34 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sat, 29 Mar 2014 16:24:34 +0000 Subject: Caches going stale Message-ID: Hi maintainers, The script which detects obsolete packages is periodically running on the buildfarm. It uses various bits of code, like the RevDeps class from the safe_remove_pkgs script. It saves various bits of data on local disk to speed up execution. https://sourceforge.net/p/gar/code/HEAD/tree/csw/mgar/gar/v2/lib/python/find_obsolete_pkgs.py https://sourceforge.net/p/gar/code/HEAD/tree/csw/mgar/gar/v2/lib/python/safe_remove_package.py Since these scripts are now left for themselves for extended periods of time, we need to make sure they don't start outputting wrong information because of stale caches. We need to think what to do with this. Here are some ideas: - periodically purge the caches (delete files from disk), and let the scripts will regenerate them - modify the programs to refresh the caches at as necessary - stop using any caches and let the scripts run slowly until we find a better solution Is anybody interested in tackling this? Maciej -------------- next part -------------- An HTML attachment was scrubbed... URL: From dam at opencsw.org Sun Mar 30 22:15:30 2014 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 30 Mar 2014 22:15:30 +0200 Subject: Interesting incompatibility between Solaris 10 and Solaris 11 Message-ID: Hi folks, I just received a bug report about CSWrcs not working on Solaris 11: https://www.opencsw.org/mantis/view.php?id=5159 The problem is immediately reproducable. What makes it interesting is when I compile specifically on Solaris 11 it works. From my understanding every binary from Solaris 10 which adheres to the published API should work OOTB on Solaris 11 and almost every time I looked it does so. The stack of the corefile looks like this: unstable11s% pstack core core 'core' of 11856: ./rcs ci ci ff2de060 memcpy%sun4v-hwcap2 (5d460, 4b324, 5ce67, 23518, 0, 8bb40) + df4 0002e514 ???????? (ffbffc01, 0, ffbffc01, 2, 49be4, 2) 0002e9d8 pairnames (1, ffbffb0c, 2c2f8, 0, 0, 2) + 3b8 0001a86c ???????? (ffbffbfe, 2, ffbffb08, 2f, ff0000, 2f000000) 000224e4 ???????? (ffbffa90, 1a058, ffbffbfe, 2, ffbffb08, 46d45) 00022b44 main (3, ffbffb04, ffbffb14, 5c800, 1a058, ffbffbfe) + 204 000153f0 _start (0, 0, 0, 0, 0, 0) + 108 I haven't investigated further, but just wanted to let you know about this strange issue. Best regards -- Dago From maciej at opencsw.org Mon Mar 31 02:11:50 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Mon, 31 Mar 2014 01:11:50 +0100 Subject: Caches going stale In-Reply-To: References: Message-ID: 2014-03-29 16:24 GMT+00:00 Maciej (Matchek) Blizi?ski : > > - periodically purge the caches (delete files from disk), and let the scripts will regenerate them > - modify the programs to refresh the caches at as necessary > - stop using any caches and let the scripts run slowly until we find a better solution It bit me directly this morning while I was cleaning up the catalogs from *_stub packages. I went for the option of no caching. The script detecting obsolete packages runs slower now, but gives correct (up to date) results. Maciej From maciej at opencsw.org Mon Mar 31 02:21:45 2014 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Mon, 31 Mar 2014 01:21:45 +0100 Subject: How to stop re-uploads of *_stub packages Message-ID: If you got en email that a *_stub owned by you has been dropped from the catalog -- it was me, and your stub is no longer necessary, so you can delete it from your build recipe. We have multiple recipes defining *_stub packages. When you rename a package, you add a stub (OBSOLETED_BY) - it's simple and good when introducing a new *_stub package, but we are now cleaning our catalogs from unused *_stub packages, and I'm thinking that we (maintainers) will re-upload obsolete stubs just because we don't know they can be removed. This in principle can be done in checkpkg. You can download the state of the catalog, simulate adding/replacing the packages, and seeing if the newly added _stub package has any reverse dependencies. If so, we can throw an error. This would add about 30s-40s to the checkpkg running time. This would require additional code written for checkpkg, because the check functions currently have no access to the catalog as a whole. This feature is not a must, but it would be cool to save constant deletes and reuploads of *_stub packages to the catalog. The positive side is that when we have access to the whole catalog, we can also run cycle detection and dangling dependency detection. This is something that Peter Bonivart was suggesting for a long time, but it has only recently became feasible / practical. Should I write it up as a coding project? Maciej From Joerg.Schilling at fokus.fraunhofer.de Mon Mar 31 11:11:36 2014 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Mon, 31 Mar 2014 11:11:36 +0200 Subject: Interesting incompatibility between Solaris 10 and Solaris 11 In-Reply-To: References: Message-ID: <53393148.AcUNOjsY6aueZY9r%Joerg.Schilling@fokus.fraunhofer.de> Dagobert Michelsen wrote: > Hi folks, > > I just received a bug report about CSWrcs not working on Solaris 11: > https://www.opencsw.org/mantis/view.php?id=5159 > > The problem is immediately reproducable. What makes it interesting is when I compile > specifically on Solaris 11 it works. From my understanding every binary from Solaris 10 This trace contains a hint that the problem may be related to memory corruption. If the malloc() algorithm changes, this may cause a program to fail that otherwise did just manage to survive. 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 james at opencsw.org Mon Mar 31 11:41:58 2014 From: james at opencsw.org (James Lee) Date: Mon, 31 Mar 2014 10:41:58 +0100 Subject: Our PHP 5.x package is getting old In-Reply-To: References: <5333D4C9.6040509@opencsw.org> Message-ID: <53393866.90704@opencsw.org> On 27/03/2014 08:59, Dagobert Michelsen wrote: > >> dam at unstable10s [unstable10s]:/home/dam/mgar/pkg/php5/branches/php-5.5.x/work/solaris10-sparc/build-isa-sparcv8plus-sapi-cgi/php-5.5.10 > pstack core >> core 'core' of 16636: /home/dam/mgar/pkg/php5/branches/php-5.5.x/work/solaris10-sparc/build- >> 000a6500 compare_opcodes (c1d7ad, 0, ffbfe5b8, ffbfd3d8, c1d789, c1d7ad) + 1178 >> 000a6944 auto_possessify (c1d767, 0, ffbfe5b8, 64, ffbfd3d8, 661eb4) + 284 >> 000b0114 php_pcre_compile2 (b9, 0, 20000, 0, c1d6c0, 96372c) + d90 >> 000d4138 pcre_get_compiled_regex_cache (a5e4ac, 2f, 2f, 2f, 0, 961064) + 444 >> 000d4680 php_do_pcre_match (3, dde418, ffbfe770, 961064, ffbfe774, 0) + 7c >> 005acaa0 zend_do_fcall_common_helper_SPEC (9ee560, 100, 100, 961064, d646cc, 9e4798) + 320 >> 005ac448 execute_ex (9ee560, 9e45f0, 1, 961064, 8358c, 0) + 3c >> 00533ce4 zend_execute_scripts (8, 0, ffbff3c8, 1, 9e4714, ffbfe8ac) + 678 >> 0043eb94 php_execute_script (9e4f30, ffbff3c8, 9e4f30, 961064, 1, 83c00) + 32c >> 005ff608 do_cli (961064, 8, 0, 83c00, 961064, 83c00) + d3c >> 00600684 main (10, ffbff674, ffbff6b8, 963000, 96be44, 96be44) + 660 >> 0004d9c0 _start (0, 0, 0, 0, 0, 0) + 108 > Everything is committed, if you have a great idea let me know. PHP build adds -xmemalign=8s to the compile options. The 's' says throws a bus error if a memory misalignment is detected. The 'i' xmemalign option say to me it will work but might be slower. The misalignment is in compare_opcodes which is part of prce 8.34 and the misalignment is right after the comment "Compare 4 bytes to improve speed". What it's doing is: function(char *cp) { int *ip; ip = (int *) cp; ... *ip ...; // handle in 4 byte chucks which is not 8 byte aligned when "cp % 8 != 0". On the Sparc and Intel chips I've tested it doesn't seem to make a difference but it must depends on the chip design and perhaps there are some where the speed difference is greater. Workaround: remove or change to 'i' xmemalign from PHP's configure: sparc*) if test "$SUNCC" = "yes"; then CFLAGS="$CFLAGS -xmemalign=8s" fi Fix: http://bugs.exim.org/show_bug.cgi?id=1427 James.