[prev in list] [next in list] [prev in thread] [next in thread] 

List:       cassandra-dev
Subject:    Re: State of Unit tests (1 out of 100 passes in trunk)
From:       Josh McKenzie <jmckenzie () apache ! org>
Date:       2016-07-21 15:54:41
Message-ID: CAPH+nLPp0sPEgRFLd788mik7RnsCYTgVzuuyVJXgiBO3eF2bbg () mail ! gmail ! com
[Download RAW message or body]


Yeah, unfortunately right now we're in clean-up / debt mode when it comes
to shoring up the quality of tests in the code-base, as well as the
stability of CI.

Once we're consistently all-green it should be much easier to figure out
who to point fingers at when things fail. :)

On Thu, Jul 21, 2016 at 11:47 AM, Sylvain Lebresne <sylvain@datastax.com>
wrote:

> On Thu, Jul 21, 2016 at 5:42 PM, Jeremiah D Jordan <
> jeremiah.jordan@gmail.com> wrote:
>
> > > Josh, add me to the "test fixers" queue, as well. However, I think the
> > > authors of patches that break the build should also be on the hook for
> > > fixing problems, as well.
> >
> > +1 I have always been a fan of "you broke it, you fix it"
> >
> >
> Problem is, we're still kind of in fixing the backlog mode, so we don't
> know who break something, not without basically understanding the issue (at
> which point writing the fix is the easiest part usually). In fact, some of
> those are flaky test where the problem is sometime just a test issue, so
> the one to blame would be test author but you don't really know that
> without looking again.
>
> Whenever we do get to a consistent clean green dashboard, then we might be
> able to quickly find out who break thing and assign to me. I don't really
> think anyone disagree with the "you broke it, you fix it" otherwise.
>


[prev in list] [next in list] [prev in thread] [next in thread]