6:59 a.m.
The x86_64 job[1] has a downstream project[2] taking care of deploying the
build to the repos
while the ppc64le[3] hasn't.
Result is that ovirt-master-snapshot is not receiving updates for ppc64le
and is currently failing repository closure:
*00:02:47.636* Error: Package:
vdsm-gluster-4.20.0-612.git378d837.el7.centos.noarch
(ovirt-master-snapshot)*00:02:47.637* Requires: vdsm =
4.20.0-612.git378d837.el7.centos*00:02:47.637* Installing:
vdsm-4.18.999-925.git195c44e.el7.centos.ppc64le
(ovirt-master-snapshot)*00:02:47.637* vdsm =
4.18.999-925.git195c44e.el7.centos*00:02:47.637* Error: Package:
vdsm-hook-ethtool-options-4.20.0-612.git378d837.el7.centos.noarch
(ovirt-master-snapshot)*00:02:47.637* Requires: vdsm =
4.20.0-612.git378d837.el7.centos*00:02:47.637* Installing:
vdsm-4.18.999-925.git195c44e.el7.centos.ppc64le
(ovirt-master-snapshot)*00:02:47.637* vdsm =
4.18.999-925.git195c44e.el7.centos
Please fix vdsm job for triggering deploy-to-ovirt_experimental_master
[1] http://jenkins.ovirt.org/job/vdsm_master_build-artifacts-el7-x86_64/
[2] http://jenkins.ovirt.org/job/deploy-to-ovirt_experimental_master/
[3] http://jenkins.ovirt.org/job/vdsm_master_build-artifacts-el7-ppc64le/
--
SANDRO BONAZZOLA
ASSOCIATE MANAGER, SOFTWARE ENGINEERING, EMEA ENG VIRTUALIZATION R&D
Red Hat EMEA <https://www.redhat.com/>
<https://red.ht/sig>
TRIED. TESTED. TRUSTED. <https://redhat.com/trusted>
1:54 a.m.
We made the PPC64 build_artifacts jobs not invoke deploy to experimental a
long while ago.
We did this because we have a race condition between the build_artifacts
jobs for a given commit for different platforms. This can and does cause
various issues, but having the Fedora build invoke deploy-to-experimental
and OST before the EL7 build is done is mostly harmless.
With PPC64 the issue is that when PPC finishes building before x86_64 we
end up with 'noarch' packages in the repos that are newer then the
equivalent x86_64. This in turn causes the experimental repos to be broken
and OST to fail until the x86_64 build is done.
Fixing this will require creating some sort of a synchronization mechanism
between the build artifacts jobs. This is not easy to do unfortunately.
I made a ticket for us to track various possible solutions:
https://ovirt-jira.atlassian.net/browse/OVIRT-1315
On 11 April 2017 at 14:59, Sandro Bonazzola <sbonazzo(a)redhat.com> wrote:
The x86_64 job[1] has a downstream project[2] taking care of
deploying the
build to the repos
while the ppc64le[3] hasn't.
Result is that ovirt-master-snapshot is not receiving updates for ppc64le
and is currently failing repository closure:
*00:02:47.636* Error: Package: vdsm-gluster-4.20.0-612.git378d837.el7.centos.noarch
(ovirt-master-snapshot)*00:02:47.637* Requires: vdsm =
4.20.0-612.git378d837.el7.centos*00:02:47.637* Installing:
vdsm-4.18.999-925.git195c44e.el7.centos.ppc64le (ovirt-master-snapshot)*00:02:47.637*
vdsm = 4.18.999-925.git195c44e.el7.centos*00:02:47.637* Error: Package:
vdsm-hook-ethtool-options-4.20.0-612.git378d837.el7.centos.noarch
(ovirt-master-snapshot)*00:02:47.637* Requires: vdsm =
4.20.0-612.git378d837.el7.centos*00:02:47.637* Installing:
vdsm-4.18.999-925.git195c44e.el7.centos.ppc64le (ovirt-master-snapshot)*00:02:47.637*
vdsm = 4.18.999-925.git195c44e.el7.centos
Please fix vdsm job for triggering deploy-to-ovirt_experimental_master
[1] http://jenkins.ovirt.org/job/vdsm_master_build-artifacts-el7-x86_64/
[2] http://jenkins.ovirt.org/job/deploy-to-ovirt_experimental_master/
[3] http://jenkins.ovirt.org/job/vdsm_master_build-artifacts-el7-ppc64le/
--
SANDRO BONAZZOLA
ASSOCIATE MANAGER, SOFTWARE ENGINEERING, EMEA ENG VIRTUALIZATION R&D
Red Hat EMEA <https://www.redhat.com/>
<https://red.ht/sig>
TRIED. TESTED. TRUSTED. <https://redhat.com/trusted>
--
Barak Korren
bkorren(a)redhat.com
RHCE, RHCi, RHV-DevOps Team
https://ifireball.wordpress.com/
2593
days inactive
2594
days old
1 comments
2 participants
participants (2)
-
Barak Korren -
Sandro Bonazzola