[Buildbot-devel] Buildbot-0.7.12rc2

[Axel Hecht]

I don't think it's testNoMerges, tbh. The test alone runs fine, even though
I'm able to reproduce the failure if running the whole test suite. It
complains about the reactor not being empty, so I blame a previous test for
not cleaning up. My uneducated guess would be test_maildir, though I have no
clue why, that test didn't change at all.

trial doesn't let me run a subset of tests in order, so it's hard to figure
out which combination of test cases actually triggers it.

Axel

2010/1/11 Charles Lepple <clepple at gmail.com>

> Actually, the other failed test in the original thread was due to a
> mismatch, but the testNoMerges failure mentioned below was using the same
> source for both /sw/bin/buildbot (first in the path) and the source
> directory.
>
> Basically, I built rc2 once, telling it to install and skip the test phase.
> The second time, I told it to build and test (and due to the test failure,
> it stopped there).
>
> The test is invoked by "PYTHONPATH=. trial buildbot.test" from the source
> directory. Is there another way I should be doing this? That fails the same
> way as if I just run 'trial buildbot.test' from the command line, so I think
> testNoMerges is not currently tripping over the installed/not-installed
> issue.
>
> On Jan 10, 2010, at 7:25 PM, Axel Hecht wrote:
>
> I bet this is once again a mismatch of installed buildbot vs test buildbot.
>
> My comment in the other thread applies, I don't really know of a good way
> to test this code.
>
> Axel
>
> 2010/1/10 Charles Lepple <clepple at gmail.com>
>
>> On Sun, Jan 10, 2010 at 1:26 PM,  <exarkun at twistedmatrix.com> wrote:
>> > On 05:59 pm, clepple at gmail.com wrote:
>> >>
>> >> On Jan 9, 2010, at 11:16 AM, Dustin J. Mitchell wrote:
>> >>>
>> >>> Here's release candidate 2.  We don't have any open bugs for the
>> >>> release branch right now, so with luck this will become the official
>> >>> release.  Please check it out!
>> >>
>> >> Dustin,
>> >>
>> >> Per the discussion here, I re-tested on Fink/python2.6 with buildbot
>> >> installed:
>> >>
>> >>
>> http://thread.gmane.org/gmane.comp.python.buildbot.devel/5227/focus=5229
>> >>
>> >> That gets rid of the buildbot.test.test_limitlogs.MasterLogs.testLog
>> >> failure.
>> >>
>> >> I am still seeing the following:
>> >>
>> >>  [ERROR]:
>> buildbot.test.test_mergerequests.MergeRequestsTest.testNoMerges
>> >>
>> >>  Traceback (most recent call last):
>> >>  Failure: twisted.internet.error.ProcessTerminated: A process has ended
>> >> with a probable error condition: process ended with exit code 1.
>> >>
>> >> I extracted what is probably the relevant portion of test.log
>> (attached),
>> >> and I don't see any obvious problems with the test results.
>> >>
>> >> Ideas?
>> >
>> > It might help to instrument the test to log the process and arguments it
>> is
>> > trying to run, and any stdout or stderr it receives.  This will often
>> give a
>> > clue as to why its exiting in an unexpected way.
>>
>> Jean-Paul,
>>
>> Unfortunately, that is well beyond the limits of what I know about the
>> internals of Buildbot (the processes appear to be implicitly created
>> as part of the test master/slave arrangement). It looks like Axel
>> wrote most of this test, so I'm CC'ing him.
>>
>> --
>> - Charles Lepple
>>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://buildbot.net/pipermail/devel/attachments/20100111/d80669ab/attachment.html>