On 12/01/2015 16:58, [email protected] wrote:
I've just started experimenting with the Gerrit Trigger Plugin. It's great, but
I have one problem:
If the build job is UNSTABLE (i.e. if any tests fail) Jenkins will post this status back
to Gerrit, and flag the Gerrit patchset as not passing. That strategy only works if the
normal condition is "all tests pass".
Your logic is the wrong way around.
You want to mark a build with the flaky test as stable - so this would
be almost the same for a normal job with the complication that Gerrit
patchsets are not linear...
In my case, there is a test case, unrelated to the patchset under test, that
has been failing for some time. I'd like the Gerrit trigger to only mark the
patchset being tested if introduces a NEW test failure. Is there a way to
achieve this?
Take a look at https://github.com/samsta/quarantine this may be the only
way of ignoring your bad test.
Note it forks the junit plugin and is probably missing features fixes in
that junit plugin.
(Yes I do now that having failing tests that have not been fixed is poor
practice, but in a large test suite, that is sometimes the case).
In this case I really would just recommend removing the test and raising
a defect for the team to fix it and tracking it that way - it will
reduce the headache (given you are not currently fixing it) and it gets
the team to accept that they can't push changes to the branch when the
code is broken.
/James
--
You received this message because you are subscribed to the Google Groups "Jenkins
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/jenkinsci-users/54B4066B.5010807%40teilo.net.
For more options, visit https://groups.google.com/d/optout.
