[Gluster-infra] Zuul?

Jeff Darcy jdarcy at redhat.com
Fri Sep 2 16:12:00 UTC 2016


> We already only merge after NetBSD-regression and CentOS-regression have
> voted
> back. All I'm changing is that you don't need to do the merge manually or do
> Verified +1 for regression to run.. Zuul will run the tests after you get
> Code-Review +2 and merge it for you with patches ordered correctly.

The problem is that some reviewers (including myself) might not even look at
a patch until it already has CentOS+1 and NetBSD+1.  Reviewing code, having
it fail regressions, reviewing a substantially new version, having *that*
fail regressions, etc. tends to be very frustrating for both authors and
reviewers.  Fighting with the regression tests *prior* to review can still
be very frustrating for authors, but at least it doesn't frustrate reviewers
as much and doesn't contribute to author/reviewer animosity (apparently a
real problem in this group) as much.

That said, it would be nice to have *something* as a gate between +2 and
merge - certainly a build, and at least a few basic tests (more than smoke
does IMO).  If Zuul can help us avoid broken builds due to improper merge
order, which seem to be the most common kind of broken builds, I'm all for
it.


More information about the Gluster-infra mailing list