launchpad-dev team mailing list archive

[Jonathan Lange <jml@xxxxxxxxxxxxx>]

On Wed, Jan 12, 2011 at 7:28 AM, Benji York <benji.york@xxxxxxxxxxxxx> wrote:
...
> Error in test test_couldNotListenTac
> (canonical.launchpad.daemons.tests.test_tachandler.TacTestSetupTestCase)
> Traceback (most recent call last):
>  File "/usr/lib/python2.6/unittest.py", line 279, in run
>    testMethod()
>  File "/srv/buildbot/slaves/launchpad/lucid-devel/build/lib/canonical/launchpad/daemons/tests/test_tachandler.py",
> line 45, in test_couldNotListenTac
>    self.assertRaises(TacException, CouldNotListenTac().setUp)
>  File "/usr/lib/python2.6/unittest.py", line 336, in failUnlessRaises
>    callableObj(*args, **kwargs)
>  File "/srv/buildbot/slaves/launchpad/lucid-devel/build/lib/canonical/launchpad/daemons/tachandler.py",
> line 140, in setUp
>    two_stage_kill(pid)
>  File "/srv/buildbot/slaves/launchpad/lucid-devel/build/lib/canonical/launchpad/daemons/tachandler.py",
> line 56, in two_stage_kill
>    _kill_may_race(pid, SIGTERM)
>  File "/srv/buildbot/slaves/launchpad/lucid-devel/build/lib/canonical/launchpad/daemons/tachandler.py",
> line 38, in _kill_may_race
>    os.kill(pid, signal_number)
> OSError: [Errno 1] Operation not permitted
>
> This looks spurious to me, so I'm resubmitting in the hopes that the
> update takes this time.

Michael Hudson-Doyle figured this one out. It's a stale pid file on
the build slave, something called 'cannotlisten.pid', that is now
referring to a process spawned by a different run of the test suite.

We don't know how this happened, but the buildbot should run the tests
successfully again if a LOSA finds and deletes the file.

Note that since this failure, new failures have been introduced into
the test suite by r12181. I'll roll that back once the rollout is
done, or find someone else to do it.

jml