[release][masakari] - [Release-job-failures] Release of openstack/masakari-monitors for ref refs/tags/(7.0.1|8.0.2|9.0.1) failed

Radosław Piliszek radoslaw.piliszek at gmail.com
Fri Nov 13 12:54:06 UTC 2020

On Fri, Nov 13, 2020 at 1:46 PM Herve Beraud <hberaud at redhat.com> wrote:
> Le ven. 13 nov. 2020 à 13:17, Radosław Piliszek <radoslaw.piliszek at gmail.com> a écrit :
>> My question is whether we could detect such issues more proactively?
> Good question... I don't think that this kind of a check should be hosted on the release management side.
> Here our issue is more an environment issue, It's due to the fact that we decided to run the job `publish-openstack-artifacts` on ubuntu-focal, so to detect this kind of issue early I think that jobs on the projects side should rely on ubuntu-focal too. I'm not a Zuul expert and I don't know if it is possible to add a condition based on the branches (per example) to choose the right env (focal or not). By following this way I guess it could be possible to fail early on the project side, and it could allow you to fix this kind of issue well before proposing a new release. In other words, if projects's jobs env are aligned with release management our job we could catch them early.

The thing is Focal is not a target platform for Train/Ussuri so
projects should not just switch to it for their CI.

> We have other scenarios to consider. These are when no patches haven't been merged since a while on projects side, and where a release is proposed, in this case even if environments are aligned, in this case we aren't able to detect this kind of issue, and they will continue to occur during publication, as in this case.

In Masakari Monitors' case it was the previous one. The CI was green
as long as one didn't move it to Focal.
Surely, there will be cases where this particular scenario applies but
then it is really up to the project team.

> It could be worth it to involve the infra team to see how to fail early.

Good call, could use their expertise.
I shall repost this thread.


More information about the openstack-discuss mailing list