CQ: Test also on tested (was: Re: [CQ]: 97694, 4 (ovirt-engine) failed "ovirt-master" system tests, but isn't the failure root cause)
On Thu, Feb 14, 2019 at 9:56 AM oVirt Jenkins <jenkins(a)ovirt.org> wrote:
A system test invoked by the "ovirt-master" change queue including change
97694,4 (ovirt-engine) failed. However, this change seems not to be the root
cause for this failure. Change 97726,5 (ovirt-engine) that this change depends
on or is based on, was detected as the cause of the testing failures.
This change had been removed from the testing queue. Artifacts built from this
change will not be released until either change 97726,5 (ovirt-engine) is fixed
and this change is updated to refer to or rebased on the fixed version, or this
change is modified to no longer depend on it.
For further details about the change see:
https://gerrit.ovirt.org/#/c/97694/4
This change is (obviously) not the cause for current failure. Current is due to:
https://bugzilla.redhat.com/show_bug.cgi?id=1677162
I also have this idea:
Perhaps, while CQ is bisecting trying to find the change that broke
it, if it's the first one, CQ will also re-test the clean tested repo,
with no changes applied at all? Then, if it fails:
1. We know for sure that it's not due to that first change, but most
likely an infra issue, or something external (like current, which is
in CentOS).
2. We have clear reproducer logs/data with no "noise" from our own
changes, that can be useful for further debugging the external/infra
issue.
Makes sense?
--
Didi