CI-users November 2016

ci-users@lists.centos.org

12 participants
10 discussions

Fix for the Jenkins browser memory leak on cico?
by Laurentiu Pancescu 24 Jan '17

24 Jan '17

Hi there, My proposed fix for Jenkins causing memory leaks in all major browsers has now been accepted upstream. [1] Is there any chance of patching ci.centos.org to include it? Regards, Laurențiu [1] https://github.com/jenkinsci/jenkins/pull/2539

4 5

Pre-Release 7.1611 Content is now Active in CI
by Brian Stinson 25 Nov '16

25 Nov '16

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Folks, We just finished seeding CentOS Linux 7.1611 pre-release content onto the ci.centos.org mirror. We are working to reinstall the existing duffy pool with the new packages, and we should be completely switched over in the next hour or so. Going forward from there, you may test your projects against what will become the next CentOS Linux 7 release! If you have feedback or questions, please let us know here, or find us in #centos-devel on IRC. Cheers! - -- Brian Stinson CentOS CI Infrastructure Team -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQIcBAEBAgAGBQJYNRxPAAoJEIMGvNKzCweMO0YQAKqn6zcVVHDIwvUmfJD3t+vO vJ0ECgaPTlTdXDPqGvTGeEAGalCkJAJXQaNA6Ant218EQHmGAU85JuvFWo+Q19x0 nr7SwOL9ZSiHzfBIfd8V4RO4mCV35NI/jlUPePVyO6VNkHFFBsSR2qm1Mxc7xs5Q Xo4ul9vwEbZBS67lwTYApe4wgTfKHGraa8li9+pz9uAKGJbJtbFPfTnhQbU6ucL0 +BX42oogNNeyVqstJlrN1gGfn5mFaGfK3YOkoqbwGx7sEnAnP0HRmHcCZ755+eHb fQ/mwXMJMN+2/0sUwo6Q0R9WnHqQDYZlxRUOHc7+dWH3EshbGywQYtCz7IStTIF0 pHH4aGhE90woLaKt9MGXmPQ+MAVRg4h1ix71bcUUFeJ6f00Ev7IEsavI0YfungA9 u8edv4Xv5kExfQpi2tXS2BggsW5BxGOTCuOjnPNqAKiMID/88kkfo/YtJVpE+3yk Ylbb34jWhfwEOJwTa4JgWRJB/g6FXP3k0CublU2gecWcRofKzB2PfTW8D9XbIwkN JAdpR0LwG0+Z5AWoy2tRCIqeJKdrr0xdnLANnZI0427fnzA4TmtRvn6cQ06dFKtn Zs3wedi2PglZLmj3tWhussl45WDbbYoSoGXPY+YpwHgf8WheDRSVzYZJhQ6LPImA WrXzA1r4kMx7utzpsus4 =j4Ys -----END PGP SIGNATURE-----

3 2

Problems in cloud-sig jobs after enabling 7.1611
by Alfredo Moralejo Alonso 24 Nov '16

24 Nov '16

Hi, After getting new 7.3 packages in the jobs running in cloud-sig we have started getting error "qemu-kvm: CPU feature arat not found" while launching VMs in intel hardware. According to information found in https://bugzilla.redhat.com/show_bug.cgi?id=1371617, the root cause for this is using libvirt from 7.3 with qemu-kvm-ev-2.3.0 from virt-sig. As there is not qemu-kvm-ev-2.6.0 available yet, this is making our jobs to fail whenever they run in any intel server and the known workaround is not so easy to implement. Is there anyway we could disable cr repo at kickstart time for specific jobs?, i think it's risky to enable in every job when there are dependencies with packages from SIGs which are not included in cr repo. Best regards, Alfredo

2 1

centpkg
by Julien Pivotto 20 Nov '16

20 Nov '16

Hi, It looks like centpkg is not installed in the jenkins nodes... is there any reason for this? -- (o- Julien Pivotto //\ Config Management SIG V_/_ https://frama.link/cfgmgmt

1 0

Re: [Ci-users] [CentOS-devel] Config Management SIG Jenkins Job Builder
by Brian Stinson 20 Nov '16

20 Nov '16

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Nov 19 15:45, Julien Pivotto wrote: > Hi all, > > I have started to bootstrap the CI jobs for the Jenkins Job Builder > > Job is at https://ci.centos.org/job/sig-configmanagement-jenkins-jobs/ > > Git repo is at > https://github.com/centos-sig-configmanagement/jenkins-jobs > > Jobs definitions are in jobs/ > > -- > (o- Julien Pivotto > //\ Config Management SIG > V_/_ https://frama.link/cfgmgmt > _______________________________________________ > CentOS-devel mailing list > CentOS-devel(a)centos.org > https://lists.centos.org/mailman/listinfo/centos-devel Nice work so far! Do you need to pin a specific version of JJB? If not, could you please use the one installed on the agent machines? Virtualenvs are perfect if you need them, but they become quite expensive in terms of disk space when the number of jobs grows. Cheers, - --Brian -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQIcBAEBAgAGBQJYMJlYAAoJEIMGvNKzCweMx2oP/3uNKgghECevEcCd61ZTVhlF IZ8w8J5FbK1esm4TNQQaZWHMpz5pPgWqRXgplyj9IQPs5fJYzh1oRDkbx3ILHW1Z 1W2Lxz0rsXN2ildfJEn+7a+T2vXzEejikr/XkrT5Dy8Fi00EG+wbg3lgkGgOlk+N 2PzI/V/LYV6jrBq3I/+GJ/WHPKbUp1UAy1EA336EE7Y9utH4s4A0tDJV0szfTxXC JLf8v1l3hxCCD16xRhFhigRrUxxeVrufz80vTOHATIpbjrV6oPw05R2WU11Gmtd7 GpYu8e3iMgtbPRl3F11wPunE+ISaHwVn7DFqayltiE1BZwYc8N9+4lhBnsgBrWxB Iu/zwyCW6Wv7JP4lBd4k+1kXHoXekVGswbOW0z4TdjzpTRCWMtvOr6Pp60a6Yoil gDBh02+aL9pW7xxRtA8uOmOC/SBkx0F28M1N6EBz/o9e+wkRr6jdne3lso6kWrk2 XGVqJczrj5uGhLYUV7Aga7W1U5hn0VTMniSkQjNMagWG1RYr7x1GMGNYWMmHX/Ic TzJhO24dlM+B+gm7Fd4RyUNLUgxPwExWvDwl44PBaEOBVSE7bQatN5+mLUAiUDIH dwDrmwumIiJRiLLYpGgIm++6tOHBlWbsUcCVNMy20UGrK+0X0p4BbapwXdAMewMJ 5wd3q/tv9V+9hbeIEpKC =nAKs -----END PGP SIGNATURE-----

2 1

getting 7.1611 content to the CI Tenants ahead of release
by Karanbir Singh 17 Nov '16

17 Nov '16

hi guys What I would really like is to have the CentOS Linux 7.1611 ( based on RHEL 7.3 sources ) be available and deployed on the CI nodes ahead of when we GA - so that SIGs and other people who are invested with us, have a chance to validate their content and also prep anything that might be needed ahead of GA for 7.1611 Towards that goal - there are 2 stages we can do this in. 1) When CR/ repo content is ready, we enable that content directly in the CI nodes. The CR repo represents the next-release content, but shipped as a repo in the existing release tree. This allows folks to 'yum update' and get the content ahead of GA. In our case, on the CI nodes, we would run the yum upgrade as a part of the install process, so while the deployment would have been with the previous installer, the booted machine would be running the new kernel + new packages. And any yum operation from here on, would include the newer content. 2) this stage would come a few days down the road, when we have the installer in place and the new CI deployments are then run with the new installer Since the installer is typically the last thing that we build, stage2 might be a few more days out, but we should be able to start seeding up stage1 above, in the next day or two. I will announce it here once this content is in place. What is important is that when things break with the new code, please let us know - this list is fine, we will track here, but also the centos-devel list is a good place to talk about it. For real time chat, #centos-devel on irc.freenode.net Does anyone have thoughts around this ? -- Karanbir Singh +44-207-0999389 | http://www.karan.org/ | twitter.com/kbsingh GnuPG Key : http://www.karan.org/publickey.asc

2 1

Duffy outage around 1 am UTC?
by Nigel Babu 17 Nov '16

17 Nov '16

Hello, For the last 2 days Gluster nightly rpm jobs have been failing because of Duffy outages around 1 am UTC[1]. Is there a planned downtime around 1 am or did we get unlucky twice in a row? [1]: https://ci.centos.org/view/Gluster/job/gluster_build-rpms/1908/console -- nigelb

2 1

We are temporarily disabling access from cli.jar ci.centos.org
by Brian Stinson 11 Nov '16

11 Nov '16

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi All, Due to an upcoming security announcement, we are disabling CLI access over the network in ci.centos.org. If you are downloading/using the cli.jar please let us know. The REST API (and configuring jobs via JJB) will remain unaffected. Cheers! - -- Brian Stinson CentOS CI Infrastructure Team -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQIcBAEBAgAGBQJYJgCbAAoJEIMGvNKzCweM2YIQALW3S/8B9GIGumIhn0X9GQMa A9s6u5RsGpKfTIgcUHG+D05DwgnyvswYviYdE1nXd0KRbymgmQ0O4nptev+vi3RE RibiXM3J7bCKzoe8/AwgH33njIjdfLp3UJpp58NADCrRmnMCnNqlB1OMC4Pq8npP s5dVROvgU0wnxFo2XkpN0x7D2lp7ujfmHbgjYZA0Xjqinnfq0Uws9627AsYMKeuk bAlHzNZ+tygfyz7GBkSDeutds3JeiWGQ197fOdbn1d2IlK4tsMuJgrJykpmy1lCP /KkrcnyI2GIAbhSKFxVS2yzHS0GNN/Xo7xOa6gkvjzLXSVoPdxy8ZuxnJ2AtkPP+ DD3zoUuufhzCXxF47C22XZx6lJjo7M81LzTp7DjPmA8DMRrOIA2Lx3+ESjeaODSw 95lkcC6f2ZHa9heDe6fVyp5XDXvUFWE0v+Owvb1iV+Zn5Zaqatqt42gBEkE2hKol MJwr3vdXkcFP9aa3XZlmx0XM6sXGsHG50X0+DXLp2pp6d9xRkalJV9cHUZznezjE sTMvMohlM5XO6kaqMgpERC2HIRq43ywt0hoSScCBHN9QfWNFn3znsanKaJy2htTQ 9QAMjvZBQQvvO7vOoV9N6FiV3UfOC2n+93NF+ulbRqwmYF50MIGp86bIilIV3xjf zlkK69R9IaIr2snRlpMx =KXlk -----END PGP SIGNATURE-----

1 0

No space left on device
by Martin Kudlej 04 Nov '16

04 Nov '16

Hi all, could you please free some space in your home directories? $ df -h . Filesystem Size Used Avail Use% Mounted on /dev/mapper/vg_slave01-home 11G 11G 44K 100% /home Thank you a lot! -- Best Regards, Martin Kudlej. IRC: mkudlej at #brno, #gluster, #storage-qa, #rhs, #rh-ceph, #usm-meeting @ redhat #tendrl-devel @ freenode

2 2

Re: [Ci-users] stuck devdoc build on ci.centos
by Karanbir Singh 03 Nov '16

03 Nov '16

Hi, + ci-users, someone there might have an idea about this, On 02/11/16 07:12, Max Rydahl Andersen wrote: > Can you help by killing this job and then I'll see when it starts > tomorrow if I can figure out why it get stuck before it is git > pushing the result. > > I've cleared that out, and the next job has run through. > > Thanks - it got stuck again, but this time I ran it in > |/usr/bin/timeout| so it would kill itself. > > No idea why, but it is the same place when running |git log| > > |+ git log -n1 WARNING: terminal is not fully functional | > > Now rerunning without that all since it was there for information only. > > But any idea why it would get stuck on that ? unsure, as a reference to the ci-users list : this is w.r.t the early jobs on https://ci.centos.org/view/Devtools/job/devtools-almighty-devdoc-build-mast… -- Karanbir Singh, Project Lead, The CentOS Project, London, UK Red Hat Ext. 8274455 | DID: 0044 207 009 4455

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

CI-users November 2016