Barak Korren commented on OVIRT-1918:
I'm not sure that the bisection really works then, since we have still packages
released 2 months ago not yet landing on tested. I don't recall to have seen
any job triggered automatically after a change-queue-tester failure doing
It works. 'change-queue-tester' itself gets re-triggered by 'change-queue' with
a different set of patches after failure. Only failures that were narrowed down
to a single patch are reported to infra-list.
You can see the state of the queue an the bisection status in a graphical
display on the status page of the 'change-queue' job.
If you see something that wasn't moved to tested it can be because:
# the HEAD itself that was tested is really broken
# some build job failed because of an infra issue (e.g s390x/fcraw issue)
# OST itself was broken at the time of the test
"ci re-merge please" is for handling the last 2 reasons, essentially [~dron] or
the infra owner is supposed to be monitoring the CQ reports, detecting theses
cases and issuing "ci re-merge please" as needed.
> Single project tests in change-queue-tester
> Key: OVIRT-1918
> URL: https://ovirt-jira.atlassian.net/browse/OVIRT-1918
> Project: oVirt - virtualization made easy
> Issue Type: By-EMAIL
> Reporter: sbonazzo
> Assignee: infra
> Please change change-queue-tester jobs for testing a single project at a
> time (ok for multiple patches at once, but from the same project).
> Right now if multiple patches are merged between a change-queue-tester
> execution and the next one are done, all of them will be tested in the next
> run, even if the changes are in different projects.
> So, supposing project A and B are completely fine, a failure in project C
> can still prevent changes in A and B to go ahead through the pipeline and
> get released.
> This cause major headaches at least to me, requiring me to go again over
> all the HEAD of the projects not being published and run yet again "ci
> re-merge please" again and again and again until I'm lucky enough to have
> nobody else merging patches or having all the ttested patches pass at once.
> I understand the need to reduce the amount of executions of the job since
> it takes an hour to execute but right now it's stealing days of execution
> for getting a patch landing on tested repo.
> SANDRO BONAZZOLA
> ASSOCIATE MANAGER, SOFTWARE ENGINEERING, EMEA ENG VIRTUALIZATION R&D
> Red Hat EMEA <https://www.redhat.com/>
> TRIED. TESTED. TRUSTED. <https://redhat.com/trusted>
This message was sent by Atlassian Jira
Infra mailing list