From grzemba at contac-dt.de Thu Jan 7 11:08:32 2016 From: grzemba at contac-dt.de (Carsten Grzemba) Date: Thu, 07 Jan 2016 11:08:32 +0100 Subject: the future of cswclassutils in IPS Message-ID: Hi, for IPS packages we have to do some rework related postinstall/preremove and the class action utils. Some issues can handle IPS itself, e.g. create user group, preserve configuration. Other we have to rewrite. Should we keep the cswclassutil package with scripts are still usefull but install the scripts in a new /opt/csw location. Or should we create for every script an own package. One example would be cswpycompile. This script/package would be a dependency for every Python package and on install package it has to trigger a SMF for compiling. Any thoughts -------------- next part -------------- An HTML attachment was scrubbed... URL: From dam at opencsw.org Thu Jan 7 11:45:19 2016 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 7 Jan 2016 11:45:19 +0100 Subject: the future of cswclassutils in IPS In-Reply-To: References: Message-ID: <5A740C9B-F8F0-4E71-B6F3-B93FB5B9EE20@opencsw.org> Hi Carsten, Am 07.01.2016 um 11:08 schrieb Carsten Grzemba : > for IPS packages we have to do some rework related postinstall/preremove and the class action utils. > Some issues can handle IPS itself, e.g. create user group, preserve configuration. Other we have to rewrite. > Should we keep the cswclassutil package with scripts are still usefull but install the scripts in a new /opt/csw location. > Or should we create for every script an own package. > One example would be cswpycompile. This script/package would be a dependency for every Python package and on install package it has to trigger a SMF for compiling. I tend to remove the dependency to CSWcswclassutils and include whats needed in the package. 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: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From grzemba at contac-dt.de Thu Jan 7 12:04:01 2016 From: grzemba at contac-dt.de (Carsten Grzemba) Date: Thu, 07 Jan 2016 12:04:01 +0100 Subject: the future of cswclassutils in IPS In-Reply-To: <5A740C9B-F8F0-4E71-B6F3-B93FB5B9EE20@opencsw.org> References: <5A740C9B-F8F0-4E71-B6F3-B93FB5B9EE20@opencsw.org> Message-ID: On 07.01.16 11:45, Dagobert Michelsen wrote: > > Hi Carsten, > > Am 07.01.2016 um 11:08 schrieb Carsten Grzemba : > > for IPS packages we have to do some rework related postinstall/preremove and the class action utils. > > Some issues can handle IPS itself, e.g. create user group, preserve configuration. Other we have to rewrite. > > Should we keep the cswclassutil package with scripts are still usefull but install the scripts in a new /opt/csw location. > > Or should we create for every script an own package. > > One example would be cswpycompile. This script/package would be a dependency for every Python package and on install package it has to trigger a SMF for compiling. > > I tend to remove the dependency to CSWcswclassutils and include whats needed in the package. > But it is not so elegant if every Python package provide its own pycompile script with a own FB-SMF. > > > > > 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 dam at opencsw.org Fri Jan 8 10:07:57 2016 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 8 Jan 2016 10:07:57 +0100 Subject: the future of cswclassutils in IPS In-Reply-To: References: <5A740C9B-F8F0-4E71-B6F3-B93FB5B9EE20@opencsw.org> Message-ID: <568EF32B-9E85-42B3-9646-8D3E47AFFB1A@opencsw.org> Hallo Carsten, Am 07.01.2016 um 12:04 schrieb Carsten Grzemba : > On 07.01.16 11:45, Dagobert Michelsen wrote: >> Am 07.01.2016 um 11:08 schrieb Carsten Grzemba : >> > for IPS packages we have to do some rework related postinstall/preremove and the class action utils. >> > Some issues can handle IPS itself, e.g. create user group, preserve configuration. Other we have to rewrite. >> > Should we keep the cswclassutil package with scripts are still usefull but install the scripts in a new /opt/csw location. >> > Or should we create for every script an own package. >> > One example would be cswpycompile. This script/package would be a dependency for every Python package and on install package it has to trigger a SMF for compiling. >> >> I tend to remove the dependency to CSWcswclassutils and include whats needed in the package. > > But it is not so elegant if every Python package provide its own pycompile script with a own FB-SMF. True, I haven?t investigated the details for IPS in such cases, if you are willing to implement it I certainly won?t go against your educated decision. 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: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From rmottola at opencsw.org Fri Jan 8 14:28:37 2016 From: rmottola at opencsw.org (Riccardo Mottola) Date: Fri, 8 Jan 2016 14:28:37 +0100 Subject: aspell - new solaris9 packages Message-ID: <568FB985.7030703@opencsw.org> Hi, I would need an aspell package on solaris9. I have enabled solaris9 as a platform. They built just fine. May I upload them? They are the same version of the solaris 10 one, there hasn't been a stable upstream release. Yann, is that fine for you? do you want to that yourelf? I'd just upload the sol9 ones. Riccardo From dam at opencsw.org Fri Jan 8 14:34:05 2016 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 8 Jan 2016 14:34:05 +0100 Subject: aspell - new solaris9 packages In-Reply-To: <568FB985.7030703@opencsw.org> References: <568FB985.7030703@opencsw.org> Message-ID: <1FFCEC48-BF5A-440D-8106-261D3DB0E53C@opencsw.org> Hi Riccardo, Am 08.01.2016 um 14:28 schrieb Riccardo Mottola : > I would need an aspell package on solaris9. > > I have enabled solaris9 as a platform. They built just fine. > > May I upload them? They are the same version of the solaris 10 one, there hasn't been a stable upstream release. > > Yann, is that fine for you? do you want to that yourelf? I'd just upload the sol9 ones. AFAIK Yann has been busy with other things lately, so if you don?t hear within a couple of days I guess it is ok to release. @Yann: I hope this is ok for you? :-) 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: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From grzemba at contac-dt.de Mon Jan 11 09:53:14 2016 From: grzemba at contac-dt.de (Carsten Grzemba) Date: Mon, 11 Jan 2016 09:53:14 +0100 Subject: the future of cswclassutils in IPS In-Reply-To: <568EF32B-9E85-42B3-9646-8D3E47AFFB1A@opencsw.org> References: <5A740C9B-F8F0-4E71-B6F3-B93FB5B9EE20@opencsw.org> <568EF32B-9E85-42B3-9646-8D3E47AFFB1A@opencsw.org> Message-ID: In the case of pycompile the /usr/share/pkg/transforms/documentation could be a template for us. It contains postprocessing actuators for man-index, texinfo, etc. On 08.01.16 10:08, Dagobert Michelsen wrote: > > Hallo Carsten, > > Am 07.01.2016 um 12:04 schrieb Carsten Grzemba : > > On 07.01.16 11:45, Dagobert Michelsen wrote: > >> Am 07.01.2016 um 11:08 schrieb Carsten Grzemba : > >> > for IPS packages we have to do some rework related postinstall/preremove and the class action utils. > >> > Some issues can handle IPS itself, e.g. create user group, preserve configuration. Other we have to rewrite. > >> > Should we keep the cswclassutil package with scripts are still useful but install the scripts in a new /opt/csw location. > >> > Or should we create for every script an own package. > >> > One example would be cswpycompile. This script/package would be a dependency for every Python package and on install package it has to trigger a SMF for compiling. > >> > >> I tend to remove the dependency to CSWcswclassutils and include whats needed in the package. > > > > But it is not so elegant if every Python package provide its own pycompile script with a own FB-SMF. > > True, I haven?t investigated the details for IPS in such cases, if you are willing to > implement it I certainly won?t go against your educated decision. > > > 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 rmottola at opencsw.org Tue Jan 12 12:41:49 2016 From: rmottola at opencsw.org (Riccardo Mottola) Date: Tue, 12 Jan 2016 12:41:49 +0100 Subject: aspell - new solaris9 packages In-Reply-To: <1FFCEC48-BF5A-440D-8106-261D3DB0E53C@opencsw.org> References: <568FB985.7030703@opencsw.org> <1FFCEC48-BF5A-440D-8106-261D3DB0E53C@opencsw.org> Message-ID: <5694E67D.7050406@opencsw.org> Hi, Dagobert Michelsen wrote: > AFAIK Yann has been busy with other things lately, so if you don?t hear > within a couple of days I guess it is ok to release. > > @Yann: I hope this is ok for you?:-) I take the silence as OK! After all, it is really a minor rebuild and extension to solaris 9. No new version, No patches, no nothing. Riccardo From rmottola at opencsw.org Wed Jan 13 11:47:27 2016 From: rmottola at opencsw.org (Riccardo Mottola) Date: Wed, 13 Jan 2016 11:47:27 +0100 Subject: aspell package upload problems Message-ID: <56962B3F.9030005@opencsw.org> Hi, I essentially just rebuild the aspell packages, by extenging them to solaris 9. build&package went fine, but upload failed (I hope it wasn't that one that broke the catalog). I see this error on solaris 10 http://buildfarm.opencsw.org/pkgdb/srv4/6d5e0bbc711cdd4216e8543e4ab2f890/ What's wrong here? I see only two errors, both overridden. This error on solaris 9: http://buildfarm.opencsw.org/pkgdb/srv4/365c69846f4a1ebc93e0bff077899f28/ This I understand even less: it looks as if the lib package collides with ... the package itself? However the pkgmap looks like the one in Solaris 10, thus the package looks fine to me. Is that a collision against another existing package perhaps? Riccardo From dam at opencsw.org Wed Jan 13 11:53:53 2016 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 Jan 2016 11:53:53 +0100 Subject: aspell package upload problems In-Reply-To: <56962B3F.9030005@opencsw.org> References: <56962B3F.9030005@opencsw.org> Message-ID: Hi Riccardo, > Am 13.01.2016 um 11:47 schrieb Riccardo Mottola : > > Hi, > > I essentially just rebuild the aspell packages, by extenging them to solaris 9. build&package went fine, but upload failed (I hope it wasn't that one that broke the catalog). > > I see this error on solaris 10 > http://buildfarm.opencsw.org/pkgdb/srv4/6d5e0bbc711cdd4216e8543e4ab2f890/ > > What's wrong here? I see only two errors, both overridden. This should be no problem during upload. > This error on solaris 9: > http://buildfarm.opencsw.org/pkgdb/srv4/365c69846f4a1ebc93e0bff077899f28/ > > This I understand even less: it looks as if the lib package collides with ... the package itself? However the pkgmap looks like the one in Solaris 10, thus the package looks fine to me. > > Is that a collision against another existing package perhaps? These packages collide: CSWaspell CSWlibaspell15 Do you build and upload both at the same time? THis is requires for checkpkg to notice 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: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From rmottola at opencsw.org Wed Jan 13 12:03:29 2016 From: rmottola at opencsw.org (Riccardo Mottola) Date: Wed, 13 Jan 2016 12:03:29 +0100 Subject: aspell package upload problems In-Reply-To: References: <56962B3F.9030005@opencsw.org> Message-ID: <56962F01.8010000@opencsw.org> Hi, Dagobert Michelsen wrote: > These packages collide: > CSWaspell CSWlibaspell15 > > Do you build and upload both at the same time? THis is requires for checkpkg to notice it. this time it worked... just to be sure I reversed the command line parameters. Now I wait for the confirmation mail before bothering you :) Riccardo From dam at opencsw.org Tue Jan 19 09:57:31 2016 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 Jan 2016 09:57:31 +0100 Subject: Python 2.7.11 Message-ID: <727E91B3-1506-4245-8EFD-4DC2823A7DEC@opencsw.org> Hi Jubal, I noticed you were working on a current Python 2.7: http://buildfarm.opencsw.org/experimental.html#jubal We just received an additional bug report for an update: https://www.opencsw.org/mantis/view.php?id=5265 Is the package in good condition? Would you mind bumping it to 2.7.11? 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: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From dbossert at opencsw.org Tue Jan 19 10:06:24 2016 From: dbossert at opencsw.org (Daniel Bossert) Date: Tue, 19 Jan 2016 10:06:24 +0100 Subject: Introduction to opencsw / First steps / Getting started Message-ID: <569DFC90.8070804@opencsw.org> Hi all My evening school ends soon and in February I have good time to deep into opencsw. Is there a person who have time to help me with compiling e.g. zabbix agent with teamviewer/skype? From 8th - 13th February I have additionally time during the day; otherwise it will be in the evening. Kind regards Daniel -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 842 bytes Desc: OpenPGP digital signature URL: From dam at opencsw.org Tue Jan 19 10:33:38 2016 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 Jan 2016 10:33:38 +0100 Subject: Python 2.7.11 In-Reply-To: <487100EC-3366-4E47-8B37-942259FF96E1@opencsw.org> References: <727E91B3-1506-4245-8EFD-4DC2823A7DEC@opencsw.org> <487100EC-3366-4E47-8B37-942259FF96E1@opencsw.org> Message-ID: <2910C0CD-2555-4B44-84A7-6A2877B4BFEC@opencsw.org> Hu Jubal, Am 19.01.2016 um 10:19 schrieb Jubal Skaggs : > Ah yes! Thank you Dago, > > I was able to test it out on sparc and x86 and everything has worked well for me. I was ready to bring it up to the next level and get it into "testing" but fear got the best of me. I'd be happy to bump it up to 2.7.11, can you give me some guidance as to how to promote the package and what kind of review process there is for something in place for the promotion? I'd hate to push something out that didn't meet opencsw standards. First commit your changes. The rule is ?commit early? so changes do not get lost. ?svn status? must give an empty result, when you have that UNCOMMITED goes away: http://buildfarm.opencsw.org/experimental.html#jubal Please do not distribute any stuff that as UNCOMMITTED in the name as it is not reproducable. Just bump to 2.7.11, commit and put new packages in the experimental directory, then ask people to test (e.g. on IRC and maintainers@ and users@). If there is no negative feedback in a couple of days you can upload the packages with csw-upload-pkg ? from ?login?. The promotion to testing is automatic after 14 days without reported bugs: http://buildfarm.opencsw.org/package-promotions/promote-packages.html 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: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From jh at opencsw.org Wed Jan 20 15:55:13 2016 From: jh at opencsw.org (Jan Holzhueter) Date: Wed, 20 Jan 2016 15:55:13 +0100 Subject: Reboot T5220 (Buildfram downtime) Message-ID: <569F9FD1.6070909@opencsw.org> Hi, we need to reboot the T5220 again. (Still some performance problems we hope to improve) So this will be another downtime for the buildfram as login etc are affected too. I will do this tomorrow morning. Sorry for the inconvenience Greetings Jan From dbossert at opencsw.org Wed Jan 20 15:56:54 2016 From: dbossert at opencsw.org (Daniel Bossert) Date: Wed, 20 Jan 2016 15:56:54 +0100 Subject: Reboot T5220 (Buildfram downtime) In-Reply-To: <569F9FD1.6070909@opencsw.org> References: <569F9FD1.6070909@opencsw.org> Message-ID: <569FA036.8090603@opencsw.org> Hi Jan Could you please provide time in UTC when the server will be down? That could help other people which are in another timezone than you :-) Kind regards Daniel Am 20.01.16 um 15:55 schrieb Jan Holzhueter: > Hi, > we need to reboot the T5220 again. > (Still some performance problems we hope to improve) > So this will be another downtime for the buildfram as login etc are > affected too. > I will do this tomorrow morning. > Sorry for the inconvenience > > Greetings > Jan > -------------- next part -------------- A non-text attachment was scrubbed... Name: 0x62A70B36.asc Type: application/pgp-keys Size: 3115 bytes Desc: not available URL: From jh at opencsw.org Wed Jan 20 16:01:45 2016 From: jh at opencsw.org (Jan Holzhueter) Date: Wed, 20 Jan 2016 16:01:45 +0100 Subject: Reboot T5220 (Buildfram downtime) In-Reply-To: <569FA036.8090603@opencsw.org> References: <569F9FD1.6070909@opencsw.org> <569FA036.8090603@opencsw.org> Message-ID: <569FA159.3060308@opencsw.org> Hi, Am 20.01.16 um 15:56 schrieb Daniel Bossert: > Hi Jan > > Could you please provide time in UTC when the server will be down? > That could help other people which are in another timezone than you :-) sure sorry. sometime on 21. Jan. 2015 between 8-11 UTC :) Depending on my workload. Greetings Jan From jh at opencsw.org Thu Jan 21 09:52:15 2016 From: jh at opencsw.org (Jan Holzhueter) Date: Thu, 21 Jan 2016 09:52:15 +0100 Subject: Reboot T5220 (Buildfram downtime) In-Reply-To: <569FA159.3060308@opencsw.org> References: <569F9FD1.6070909@opencsw.org> <569FA036.8090603@opencsw.org> <569FA159.3060308@opencsw.org> Message-ID: <56A09C3F.4070302@opencsw.org> Hi, reboot is done. All should be back to normal. I hope the performance problem is fixed now. If you have any problems let me know. Greetings Jan From dam at opencsw.org Mon Jan 25 11:28:23 2016 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 25 Jan 2016 11:28:23 +0100 Subject: Buildfarm IPS server for experimental IPS packages Message-ID: <1639CC4E-5C5D-4C00-BB1B-E1B77EB64B57@opencsw.org> Hi folks, I setup an experimental IPS repo server on the buildfarm that behaves similar to /home/experimental for SVR4-packages. * Creating a new repo On a Solaris 11 host (like unstable11s or unstable11x) use cd /home/ips pkgrepo create The repository can now receive packages from inside the buildfarm. * Using the repo The IPS repo is accessible at https://buildfarm.opencsw.org/ips/ which gives a list about the available repository. The list is currently static and rebuild every 10 minutes. Please note that Safari is not able to browse the repo for some technical reason, Firefox and Chrome work fine, though. dam at unstable11s [unstable11s]:/home/ips > pkgrepo list -s http://buildfarm.opencsw.org/ips/dam HERAUSGEBER NAME O VERSION opencsw lsof 4.89,5.11:20151229T152446Z opencsw lsof 4.88,5.11:20150322T130102Z opencsw lsof 4.88,5.11:20150302T160752Z opencsw lsof 4.88,5.11:20150301T212923Z opencsw sudo 1.8.12,5.11:20150309T110254Z Technically there is a forwarder from the ?web? zone to the ?ipsrepo? zone which runs on the m3000. From there another Apache is getting the requests and distrbutes them to the local repo server instances which each run on a different tcp port. At the moment the repos are not cleaned up, but I wanted to give you an early heads up. Just let me know if you encounter any issues. @Carsten: What do you think about publising build IPS-packages directly into /home/ips/ ? 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: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From grzemba at contac-dt.de Mon Jan 25 17:09:56 2016 From: grzemba at contac-dt.de (Carsten Grzemba) Date: Mon, 25 Jan 2016 17:09:56 +0100 Subject: Buildfarm IPS server for experimental IPS packages In-Reply-To: <1639CC4E-5C5D-4C00-BB1B-E1B77EB64B57@opencsw.org> References: <1639CC4E-5C5D-4C00-BB1B-E1B77EB64B57@opencsw.org> Message-ID: On 25.01.16 11:28, Dagobert Michelsen wrote: > > Hi folks, > > I setup an experimental IPS repo server on the buildfarm that behaves similar > to /home/experimental for SVR4-packages. > > > * Creating a new repo > > On a Solaris 11 host (like unstable11s or unstable11x) use > cd /home/ips > pkgrepo create > The repository can now receive packages from inside the buildfarm. > > > * Using the repo > > The IPS repo is accessible at > https://buildfarm.opencsw.org/ips/ > which gives a list about the available repository. The list is currently static and > rebuild every 10 minutes. Please note that Safari is not able to browse the repo > for some technical reason, Firefox and Chrome work fine, though. > > dam at unstable11s [unstable11s]:/home/ips > pkgrepo list -s http://buildfarm.opencsw.org/ips/dam > HERAUSGEBER NAME O VERSION > opencsw lsof 4.89,5.11:20151229T152446Z > opencsw lsof 4.88,5.11:20150322T130102Z > opencsw lsof 4.88,5.11:20150302T160752Z > opencsw lsof 4.88,5.11:20150301T212923Z > opencsw sudo 1.8.12,5.11:20150309T110254Z > > Technically there is a forwarder from the ?web? zone to the ?ipsrepo? zone > which runs on the m3000. From there another Apache is getting the requests and > distrbutes them to the local repo server instances which each run on a different > tcp port. > > > At the moment the repos are not cleaned up, but I wanted to give you an early > heads up. Just let me know if you encounter any issues. > > @Carsten: What do you think about publising build IPS-packages directly into > /home/ips/ ? > Yes, it is a good starting point. Finally we have to publish in a common repository. For that we have to discuss some things, mainly package naming and checking. > > > > > 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 dam at opencsw.org Mon Jan 25 17:11:35 2016 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 25 Jan 2016 17:11:35 +0100 Subject: Buildfarm IPS server for experimental IPS packages In-Reply-To: References: <1639CC4E-5C5D-4C00-BB1B-E1B77EB64B57@opencsw.org> Message-ID: <52EAECFE-3629-462D-B413-18613D4B78F3@opencsw.org> Hi Carsten Am 25.01.2016 um 17:09 schrieb Carsten Grzemba : > On 25.01.16 11:28, Dagobert Michelsen wrote: >> @Carsten: What do you think about publising build IPS-packages directly into >> /home/ips/ ? > > Yes, it is a good starting point. Finally we have to publish in a common repository. For that we have to discuss some things, mainly package naming and checking. Sure. I forgot to mention that the repo is of course directly accessible from the outside just the same as experimental/. 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: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From wilbury at opencsw.org Tue Jan 26 17:31:52 2016 From: wilbury at opencsw.org (Juraj Lutter) Date: Tue, 26 Jan 2016 17:31:52 +0100 Subject: OpenSSL Message-ID: <56A79F78.2040905@opencsw.org> Hi, I don't know if you noticed, but FYI: http://www.securityweek.com/openssl-patch-high-severity-vulnerability /j -------------- next part -------------- An HTML attachment was scrubbed... URL: From grzemba at contac-dt.de Wed Jan 27 15:55:57 2016 From: grzemba at contac-dt.de (Carsten Grzemba) Date: Wed, 27 Jan 2016 15:55:57 +0100 Subject: Install SMF's in standard locations Message-ID: Some questions 1. In Solaris11 is an SMF 'manifest-import' which can triggered if a new SMF is installed. But this SMF covers only the standard locations for SMF /lib/svc/manifest and /var/svc/manifest. I would prefer to use this SMF and install our SMF's also in standard location /var/svc/manifest. 2. Package Repositories: How want we handle our concept of catalogs unstable, testing, ... Want we cover this with different repositories/publishers? 3. Package naming: Our SVR4 packages use a "flat" namespace. IPS use a hirachical one but this is only cosmetics. There are aome lists of the relation of SVR4 and IPS names for some application like this: https://github.com/MrStaticVoid/spec-files-extra/blob/master/experimental/packagenames.sort_newnames.ts Should we keep the flat namespace and prefix with opencsw or also establish such hierachic? 4. Packages for x86 and Sparc: IPS can deliver both architectures in one package. Has anyone some experience in building dual-arch IPS packages. Have we made the mgar platform target mandatory for that? Any thoughts Carsten -------------- next part -------------- An HTML attachment was scrubbed... URL: From dam at opencsw.org Wed Jan 27 16:10:17 2016 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 Jan 2016 16:10:17 +0100 Subject: Install SMF's in standard locations In-Reply-To: References: Message-ID: Hi Carsten, Am 27.01.2016 um 15:55 schrieb Carsten Grzemba : > Some questions > > 1. In Solaris11 is an SMF 'manifest-import' which can triggered if a new SMF is installed. But this SMF covers only the standard locations for SMF /lib/svc/manifest and /var/svc/manifest. I would prefer to use this SMF and install our SMF's also in standard location /var/svc/manifest. I would also go with default locations. > 2. Package Repositories: How want we handle our concept of catalogs unstable, testing, ... Want we cover this with different repositories/publishers? I am not sure about this one, also not about the drawbacks of either solution. It should be possible to mix and match packages from different catalogs. > 3. Package naming: Our SVR4 packages use a "flat" namespace. IPS use a hirachical one but this is only cosmetics. There are aome lists of the relation of SVR4 and IPS names for some application like this: > https://github.com/MrStaticVoid/spec-files-extra/blob/master/experimental/packagenames.sort_newnames.ts > Should we keep the flat namespace and prefix with opencsw or also establish such hierachic? library/ sounds useful, I would make that one automatic. Regarding the others I tend to mimic the classification Oracle uses. > 4. Packages for x86 and Sparc: IPS can deliver both architectures in one package. Has anyone some experience in building dual-arch IPS packages. Have we made the mgar platform target mandatory for that? Right, I thought about that too, it should be fairly easy as GAR can already build multiple trees for one package, however we don?t have one for Sparc *and* x86 in one recipe. There are some adjustments in gar.conf.mk needed. But this is definitely needed. Having one tree for Sparc and x86 is very important IMHO. 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: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From laurent at opencsw.org Wed Jan 27 16:51:36 2016 From: laurent at opencsw.org (Laurent Blume) Date: Wed, 27 Jan 2016 16:51:36 +0100 Subject: OpenSSL In-Reply-To: <56A79F78.2040905@opencsw.org> References: <56A79F78.2040905@opencsw.org> Message-ID: <56A8E788.1080801@opencsw.org> Le 2016/01/26 17:31 +0100, Juraj Lutter a ?crit: > Hi, > > I don't know if you noticed, but FYI: > > http://www.securityweek.com/openssl-patch-high-severity-vulnerability > > /j > Do not panic. The ?high severity vulnerability? from the title impacts only 1.0.2, which is not in OpenCSW. There is only a low one for 1.0.1. Next time, feel free to sum up the situation instead of merely posting a fear-inducing link. Laurent From pfelecan at opencsw.org Wed Jan 27 19:21:42 2016 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 27 Jan 2016 19:21:42 +0100 Subject: Install SMF's in standard locations In-Reply-To: (Carsten Grzemba's message of "Wed, 27 Jan 2016 15:55:57 +0100") References: Message-ID: Carsten Grzemba writes: > Packages for x86 and Sparc: IPS can deliver both architectures in one package. Can or must ? If the former, we can just deliver mon architectural packages as we do today. I don't see where is the advantage of delivering multi-architectural packages. Is the transport/storage cost so low that we can ignore it now ? -- Peter From dam at opencsw.org Wed Jan 27 21:22:11 2016 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 Jan 2016 21:22:11 +0100 Subject: Install SMF's in standard locations In-Reply-To: References: Message-ID: Hi Peter, Am 27.01.2016 um 19:21 schrieb Peter FELECAN : > Carsten Grzemba writes: > >> Packages for x86 and Sparc: IPS can deliver both architectures in one package. > > Can or must ? If the former, we can just deliver mon architectural > packages as we do today. I don't see where is the advantage of > delivering multi-architectural packages. Is the transport/storage cost > so low that we can ignore it now ? Well, an IPS-package is usually not a blob but installed into a repository server. A client retreives only the files he needs. This may add development files for a library, localization files for the locales you like, optimized and debug versions or just the correct ISA. It is easier for the user if we provide combined packages and I don?t see any drawbacks (apart from people who manually extract p5p-archives and transfer these to the clients which would be very non-ips-lile). 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: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From pfelecan at opencsw.org Wed Jan 27 23:11:37 2016 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 27 Jan 2016 23:11:37 +0100 Subject: Install SMF's in standard locations In-Reply-To: (Dagobert Michelsen's message of "Wed, 27 Jan 2016 21:22:11 +0100") References: Message-ID: Dagobert Michelsen writes: > Hi Peter, > > Am 27.01.2016 um 19:21 schrieb Peter FELECAN : >> Carsten Grzemba writes: >> >>> Packages for x86 and Sparc: IPS can deliver both architectures in one package. >> >> Can or must ? If the former, we can just deliver mon architectural >> packages as we do today. I don't see where is the advantage of >> delivering multi-architectural packages. Is the transport/storage cost >> so low that we can ignore it now ? > > Well, an IPS-package is usually not a blob but installed into a repository > server. A client retreives only the files he needs. This may add development > files for a library, localization files for the locales you like, optimized > and debug versions or just the correct ISA. It is easier for the user if > we provide combined packages and I don?t see any drawbacks (apart from > people who manually extract p5p-archives and transfer these to the clients > which would be very non-ips-lile). Well, this shows my ignorance in terms of IPS packaging. The glimpse you offer here makes me think that IPS is a different beast from any packaging system that I'm used to. When I have some time to spend on it I'll read about in more detail to understand it. Until then, I'll lurk in these discussions. -- Peter From grzemba at contac-dt.de Thu Jan 28 08:28:53 2016 From: grzemba at contac-dt.de (Carsten Grzemba) Date: Thu, 28 Jan 2016 08:28:53 +0100 Subject: Install SMF's in standard locations In-Reply-To: References: Message-ID: On 27.01.16 23:12, Peter FELECAN wrote: > > Dagobert Michelsen writes: > > > Hi Peter, > > > > Am 27.01.2016 um 19:21 schrieb Peter FELECAN : > >> Carsten Grzemba writes: > >> > >>> Packages for x86 and Sparc: IPS can deliver both architectures in one package. > >> > >> Can or must ? If the former, we can just deliver mon architectural > >> packages as we do today. I don't see where is the advantage of > >> delivering multi-architectural packages. Is the transport/storage cost > >> so low that we can ignore it now ? > > > > Well, an IPS-package is usually not a blob but installed into a repository > > server. A client retreives only the files he needs. This may add development > > files for a library, localization files for the locales you like, optimized > > and debug versions or just the correct ISA. It is easier for the user if > > we provide combined packages and I don?t see any drawbacks (apart from > > people who manually extract p5p-archives and transfer these to the clients > > which would be very non-ips-lile). > > Well, this shows my ignorance in terms of IPS packaging. The glimpse you > offer here makes me think that IPS is a different beast from any > packaging system that I'm used to. When I have some time to spend on it > I'll read about in more detail to understand it. Until then, I'll lurk > in these discussions. > -- > An other option is to provide two different repositories Sparc and x86. It could be an option for an interim period. Perhaps there is also an advantage for products which take very long to build like Qt. Does anyone know if it possible to merge Sparc and x86 packages later? -------------- next part -------------- An HTML attachment was scrubbed... URL: From ihsan at opencsw.org Sat Jan 30 22:31:01 2016 From: ihsan at opencsw.org (=?UTF-8?B?xLBoc2FuwqBEb8SfYW4=?=) Date: Sat, 30 Jan 2016 22:31:01 +0100 Subject: Disk needed for mail server Message-ID: <56AD2B95.8090602@opencsw.org> Hi, One of the disks on the OpenCSW has failed: pool: rpool state: DEGRADED status: One or more devices has experienced an unrecoverable error. An attempt was made to correct the error. Applications are unaffected. action: Determine if the device needs to be replaced, and clear the errors using 'zpool clear' or replace the device with 'zpool replace'. see: http://www.sun.com/msg/ZFS-8000-9P scan: scrub repaired 2.64M in 1h10m with 0 errors on Sun Jan 18 04:55:11 2015 config: NAME STATE READ WRITE CKSUM rpool DEGRADED 0 0 0 mirror-0 DEGRADED 0 0 0 c1t0d0s0 ONLINE 0 0 0 c1t1d0s0 DEGRADED 0 0 23 too many errors Does anybody has or is willing to sponsor a 73 GB disk for a T2000 machine? Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From laurent at opencsw.org Sat Jan 30 23:54:17 2016 From: laurent at opencsw.org (Laurent Blume) Date: Sat, 30 Jan 2016 23:54:17 +0100 Subject: Disk needed for mail server In-Reply-To: <56AD2B95.8090602@opencsw.org> References: <56AD2B95.8090602@opencsw.org> Message-ID: <56AD3F19.7030605@opencsw.org> Hello, That column could indicate something else than the disk. Have you checked its SMART data to be sure? Otherwise, there was a bunch of HP hardware going to the trash not long ago. I think I can still put my hands on some, probably nothing that small, but you don't mind 146 or 300, right? SAS, 2.5", 10k. Laurent Le 30/01/2016 22:31, ?hsan Do?an a ?crit : > Hi, > > One of the disks on the OpenCSW has failed: > > pool: rpool > state: DEGRADED > status: One or more devices has experienced an unrecoverable error. An > attempt was made to correct the error. Applications are unaffected. > action: Determine if the device needs to be replaced, and clear the errors > using 'zpool clear' or replace the device with 'zpool replace'. > see: http://www.sun.com/msg/ZFS-8000-9P > scan: scrub repaired 2.64M in 1h10m with 0 errors on Sun Jan 18 > 04:55:11 2015 > config: > > NAME STATE READ WRITE CKSUM > rpool DEGRADED 0 0 0 > mirror-0 DEGRADED 0 0 0 > c1t0d0s0 ONLINE 0 0 0 > c1t1d0s0 DEGRADED 0 0 23 too many errors > > Does anybody has or is willing to sponsor a 73 GB disk for a T2000 machine? > > > > > Ihsan > From dam at opencsw.org Sun Jan 31 12:04:17 2016 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 31 Jan 2016 12:04:17 +0100 Subject: Disk needed for mail server In-Reply-To: <56AD3F19.7030605@opencsw.org> References: <56AD2B95.8090602@opencsw.org> <56AD3F19.7030605@opencsw.org> Message-ID: Hi, Am 30.01.2016 um 23:54 schrieb Laurent Blume : > That column could indicate something else than the disk. Have you checked its SMART data to be sure? > > Otherwise, there was a bunch of HP hardware going to the trash not long ago. I think I can still put my hands on some, probably nothing that small, but you don't mind 146 or 300, right? > SAS, 2.5", 10k. I have also a pack of 146GB disks in the shelve, but if you can get 300GB from Laurent that is probably the better choice :-) Best regards ? Dago > > Laurent > > Le 30/01/2016 22:31, ?hsan Do?an a ?crit : >> Hi, >> >> One of the disks on the OpenCSW has failed: >> >> pool: rpool >> state: DEGRADED >> status: One or more devices has experienced an unrecoverable error. An >> attempt was made to correct the error. Applications are unaffected. >> action: Determine if the device needs to be replaced, and clear the errors >> using 'zpool clear' or replace the device with 'zpool replace'. >> see: http://www.sun.com/msg/ZFS-8000-9P >> scan: scrub repaired 2.64M in 1h10m with 0 errors on Sun Jan 18 >> 04:55:11 2015 >> config: >> >> NAME STATE READ WRITE CKSUM >> rpool DEGRADED 0 0 0 >> mirror-0 DEGRADED 0 0 0 >> c1t0d0s0 ONLINE 0 0 0 >> c1t1d0s0 DEGRADED 0 0 23 too many errors >> >> Does anybody has or is willing to sponsor a 73 GB disk for a T2000 machine? >> >> >> >> >> Ihsan >> > -- "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: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From ihsan at opencsw.org Sun Jan 31 12:22:33 2016 From: ihsan at opencsw.org (=?UTF-8?B?xLBoc2FuwqBEb8SfYW4=?=) Date: Sun, 31 Jan 2016 12:22:33 +0100 Subject: Disk needed for mail server In-Reply-To: <56AD3F19.7030605@opencsw.org> References: <56AD2B95.8090602@opencsw.org> <56AD3F19.7030605@opencsw.org> Message-ID: <56ADEE79.1080008@opencsw.org> Hi, Am 30.01.2016 um 23:54 schrieb Laurent Blume: > That column could indicate something else than the disk. Have you > checked its SMART data to be sure? I haven't, but iostat -Ene does show errors on this disk: c1t1d0 Soft Errors: 29 Hard Errors: 59 Transport Errors: 9 Vendor: FUJITSU Product: MAV2073RCSUN72G Revision: 0301 Serial No: 0545S00UEJ Size: 73.41GB <73407865856 bytes> Media Error: 58 Device Not Ready: 0 No Device: 1 Recoverable: 29 Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sun Jan 31 12:22:37 2016 From: ihsan at opencsw.org (=?UTF-8?B?xLBoc2FuwqBEb8SfYW4=?=) Date: Sun, 31 Jan 2016 12:22:37 +0100 Subject: Disk needed for mail server In-Reply-To: References: <56AD2B95.8090602@opencsw.org> <56AD3F19.7030605@opencsw.org> Message-ID: <56ADEE7D.5060503@opencsw.org> Hi, Am 31.01.2016 um 12:04 schrieb Dagobert Michelsen: >> That column could indicate something else than the disk. Have you checked its SMART data to be sure? >> >> Otherwise, there was a bunch of HP hardware going to the trash not long ago. I think I can still put my hands on some, probably nothing that small, but you don't mind 146 or 300, right? >> SAS, 2.5", 10k. > > I have also a pack of 146GB disks in the shelve, but if you can get > 300GB from Laurent that is probably the better choice :-) If they are original Sun disks, then I would prefer these, as they are much less of trouble. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From laurent at opencsw.org Sun Jan 31 16:21:08 2016 From: laurent at opencsw.org (Laurent Blume) Date: Sun, 31 Jan 2016 16:21:08 +0100 Subject: Disk needed for mail server In-Reply-To: <56ADEE7D.5060503@opencsw.org> References: <56AD2B95.8090602@opencsw.org> <56AD3F19.7030605@opencsw.org> <56ADEE7D.5060503@opencsw.org> Message-ID: <56AE2664.3090709@opencsw.org> Le 2016/01/31 12:22 +0100, ?hsan do?an a ?crit: > If they are original Sun disks, then I would prefer these, as they are > much less of trouble. Oh really? Why? T2000 have buggy SAS controllers? Then by all means, change the server, not the disks. There's no such things as original Sun disks anyway, Sun never manufactured any disk, only relabelled other brands, the same as HP does. Laurent