devel July 2019

devel@lists.centos.org

33 participants
19 discussions

CentOS 8 Artwork needed
by Rich Bowen 31 Jan '20

31 Jan '20

To all the artists and designers here, we need your help. With RHEL 8 beta released, we need to start producing artwork for CentOS 8. It's time for the CentOS Artwork SIG, and anyone else that's interested in contributing to this effort, to start working on everything that will be needed for the new release. What we know we need is anaconda artwork, and new desktop backgrounds to replace the current '7' themed stuff. As we dig, we may find more things that need to be updated to the new 8 theme. Additionally, we will need to update some of the assets listed here: https://wiki.centos.org/ArtWork/Identity ... and produce new assets for 8 to be listed here: https://wiki.centos.org/ArtWork/Brand/Logo And, finally, we need someone to step up to lead this effort. While we don't yet know for certain when RHEL 8 will officially release, it's safe to assume that it's soon, and we need your help. -- Rich Bowen - rbowen(a)redhat.com @CentOSProject // @rbowen 859 351 9166

10 30

Using mirrorlist in centos-release-* packages
by Anssi Johansson 12 Sep '19

12 Sep '19

Hi, as mentioned in https://lists.centos.org/pipermail/centos-devel/2018-September/016949.html mirrorlist.centos.org has been updated to produce mirror lists for SIG content as well. Some centos-release-* packages have already been modified to use mirrorlist.centos.org instead of mirror.centos.org, but many still point to mirror.centos.org. Most of mirror.centos.org hosts are also used for seeding the 600+ external mirrors we have. By directing some of their download traffic to external mirrors we can offer faster sync speeds for those external mirrors, and for people downloading individual rpms from mirror.centos.org. Second, most of those external mirrors offer faster download speeds to end users than what could be achieved by downloading from mirror.centos.org, so the users will benefit from this change as well. The better geographical coverage does not hurt either. mirror.centos.org nodes are donated servers, and it would be kind to the sponsors to use their bandwidth sparingly. The .repo files should be modified so that the current baseurl line gets commented out and a mirrorlist entry gets added, like this: mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch… #baseurl=http://mirror.centos.org/$contentdir/$releasever/storage/$basearch/gluster-4.1/ Note that mirrorlist.c.o can automatically choose between main and altarch content based on the arch parameter. The repo names are constructed from the repository path by stripping away the architecture and changing any slashes to dashes: sclo/x86_64/rh -> sclo-rh, sclo/x86_64/sclo -> sclo-sclo, paas/x86_64/openshift-origin311 -> paas-openshift-origin311, storage/x86_64/gluster-5 -> storage-gluster-5 etc. You can use 'curl' to check what mirrorlist outputs for your repository. New repos are added to mirrorlist.centos.org automatically once the repository hits mirror.centos.org. That script is currently run every three hours. I guess it could be run more frequently (it's fairly lightweight), but you should give the external mirrors a few hours to sync before announcing your release anyway. You can either submit an update to your current centos-release-* file now, or wait until you need to create a new centos-release-* package for other reasons such as releasing softwarepackage-(version+1) which uses different paths. My recommendation would be to submit an update to your current centos-release-* package, so that you would not need to worry about this when you eventually need to publish a new one for other reasons. We released an updated centos-release package for CentOS AltArch before 7.6.1810 was released for this exact reason, ie. we didn't need to worry about this change at 7.6.1810 release time. Other people (such as bstinson, arrfab and hughesjr) can probably help with creating and publishing an updated centos-release-* package, but the usual tagging procedure to extras will apply. The SIG Guide https://wiki.centos.org/SIGGuide should have all the needed details for this (if not, complain to bstinson). It is highly recommended to actually test your new centos-release-* package from buildlogs before you ask that package to be signed and pushed to extras. The holiday season is fast approaching so (I) don't expect much activity regarding these changes in the following days. I'm hoping that the SIGs would take a look at this in early 2019, though. Thanks to those SIGs who have already made these changes to their centos-release-* packages! If you have questions, feel free to ask either here or on #centos-devel. Thanks for your time!

1 2

rh-ruby26
by Marko Bevc 13 Aug '19

13 Aug '19

Hi. Not sure this is the right mailing list, but do we have a status when those packages would be available for CentOS 7? Or can someone point me in the right direction? Thanks. Cheers, Marko

2 2

Community build services and EPEL8
by Thomas Oulevey 31 Jul '19

31 Jul '19

Hello folks, In the planning of next Community build services, a new question came up. In the past we never allowed EPEL repositories to be used as external repositories for SIGs. Doing so, would allow to build against EPEL packages and not duplicate the work for maintainers in EPEL/SIGs. However it would mean also building against a moving target which can be difficult if EPEL policies are the same as today. As EPEL 8 is also in the making it would be good to hear what SIGs think about this specific issue. Let us know ! -- Thomas Oulevey

17 48

RHEL 8.1 Beta
by Thomas Stephen Lee 30 Jul '19

30 Jul '19

Hi, https://www.redhat.com/en/blog/red-hat-enterprise-linux-81-beta-now-availab… RHEL 8.1 Beta released. -- Lee

4 4

qemu-kvm-ev issues creating a VM
by Stefano Mingoli 24 Jul '19

24 Jul '19

Hello, not sure if it's the right place to post. I have a VMWare Workstation 15.1 VM based on Centos 7.6 (Kenel 3.10.0-957) with the CentOS-QEMU-EV repository enabled. Version of qemu is: 10:2.12.0-18.el7_6.5.1 I'm having an issue with qemu-kvm-ev, basically it looks like I'm not able to create any VM. I'm getting a "NMI watchdog: BUG: soft lockup - CPU#0 stuck for 22s! [kworker/0:0H:5]" and the installation will not proceed. Everything works fine if I use the standard qemu-kvm B R Stefano Mingoli

2 1

[SCLo SIG] new sclo-php*-php-pecl-redis5 packages available for testing
by Remi Collet 23 Jul '19

23 Jul '19

Yet another package available in centos-sclo-sclo-testing This extension provides an API for communicating with Redis servers. https://pecl.php.net/package/redis For rh-php72 (EL-7 only) * sclo-php72-php-pecl-redis5-5.0.1-1 For rh-php71 (EL-7 only) * sclo-php71-php-pecl-redis5-5.0.1-1 For rh-php70 * sclo-php70-php-pecl-redis5-5.0.1-1 Notices: - this extension is already available, as sclo-php7#-php-pecl-redis (version 3.1.6) and php7#-php-pecl-redis4 version 4.3.0. But as version 5 introduce some API changes, we choose to provide it as a separate package, giving users the choice to install v3, v4 or v5 - both versions cannot be installed simultaneously. - new collections (e.g. PHP 7.3, 7.4...) will only provide the v5 package. - upstream Changelog: https://pecl.php.net/package-changelog.php?package=redis Remi. P.S.1: https://wiki.centos.org/SpecialInterestGroup/SCLo P.S.2: you can follow activity on sclo-php* packages on https://blog.remirepo.net/post/2017/02/23/Additional-PHP-packages-for-RHSCL

1 0

Recreating the centos.org website
by jatin 16 Jul '19

16 Jul '19

Recently fedora changed their website design and is now having alot more professional look and feel. centos.org hasnt changed alot. With the release of CentOS 8 it will be great to present it with a new centos.org. Just a suggestion... A bit about myself - Jatin Garg Nation- India DOB - 18th December 2001 Notable contributions to CentOS project - https://github.com/jatin-cbs/Lutris-RHEL-CentOS7 making lutris wine gaming accessible on RHEL, CentOS to users creating sh scripts. Contribution to Vulkan, Unreal Engine to optimize performance and wiki improvements for CentOS.

1 0

Rpm post scriplet failed with exit status 1 but return status of rpm is 0
by Maheshwari, Shagun 15 Jul '19

15 Jul '19

Hi, On CentOS 6, when we try to install any rpm it fails with scriplet failed with exit status 1 but return status of rpm is also 1 which is correct. Example: But On CentOS 7, we are not getting return status of rpm as 1 it is coming 0, which is installing the RPM. rpm -Uvh /home/seldridg/rpmbuild/RPMS/noarch/test-1.0-1.el6.noarch.rpm; echo $? Preparing... ########################################### [100%] 1:test ########################################### [100%] Test Script called failure seen while executing test-script error: %post(test-1.0-1.el6.noarch) scriptlet failed, exit status 1 1 <-- verified this works as expected I checked the source code of rpm package for CentOS6/7 and found that one patch is applicable in CentOS6 rpm package and because of this any failure is coming during scriplet failure it treats warning as an error and giving rpm status as a 1 and not installing that RPM in your system. rpm -Uvh /home/seldridg/rpmbuild/RPMS/noarch/test-1.0-1.el7.noarch.rpm; echo $? Preparing... ########################################### [100%] 1:test ########################################### [100%] Test Script called failure seen while executing test-script error: %post(test-1.0-1.el7.noarch) scriptlet failed, exit status 1 0 <-- verified this works as expected Below is the patch: +++ b/rpm-4.8.x-strict-script-errors.patch @@ -0,0 +1,25 @@ +diff -up rpm-4.8.0/lib/psm.c.strict-script rpm-4.8.0/lib/psm.c +--- rpm-4.8.0/lib/psm.c.strict-script 2013-08-05 11:09:40.290136043 +0300 ++++ rpm-4.8.0/lib/psm.c 2013-08-05 11:12:18.465137945 +0300 +@@ -517,7 +517,9 @@ static rpmRC runLuaScript(rpmpsm psm, He + if (rc != RPMRC_OK) { + if ((stag != RPMTAG_PREIN && stag != RPMTAG_PREUN && + stag != RPMTAG_VERIFYSCRIPT && stag != RPMTAG_PRETRANS)) { +- warn_only = 1; ++ if (!rpmExpandNumeric("%{?_strict_script_errors}")) { ++ warn_only = 1; ++ } + } + } + umask(oldmask); +@@ -774,7 +776,9 @@ static rpmRC runScript(rpmpsm psm, Heade + } else { + /* filter out "regular" error exits from non-pre scriptlets */ + if ((stag != RPMTAG_PREIN && stag != RPMTAG_PREUN && stag != RPMTAG_VERIFYSCRIPT && stag != RPMTAG_PRETRANS)) { +- warn_only = 1; ++ if (!rpmExpandNumeric("%{?_strict_script_errors}")) { ++ warn_only = 1; ++ } + } + rpmlog(warn_only ? RPMLOG_WARNING : RPMLOG_ERR, + _("%s scriptlet failed, exit status %d\n"), I tried to modify the code according to this but not getting success. Please help me on this. I want the same behavior of CentOS 6 on CentOS 7. Regards, Shagun

4 4

EPEL 8
by Thomas Stephen Lee 12 Jul '19

12 Jul '19

Hi, It looks like Fedora EPEL people are having issues with the "modules" architecture. You all are probably busy with CentOS 8. If anyone has spare time, take a look. https://lists.fedoraproject.org/archives/list/epel-devel@lists.fedoraprojec… thanks -- Lee

5 9

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

devel July 2019