launchpad-dev team mailing list archive

[Jonathan Lange <jml@xxxxxxxxxxxxx>]

On Wed, Oct 27, 2010 at 9:31 AM, Deryck Hodge
<deryck.hodge@xxxxxxxxxxxxx> wrote:
> rockstar has a branch that keeps disappearing from ec2.  It's a large
> lazr-js refactor.  So I ran the windmill layer locally, and it dies
> at:
>
>  Ran 11 tests with 6 failures and 0 errors in 12 minutes 13.872 seconds.
> Running lp.code.windmill.testing.CodeWindmillLayer tests:
>  Tear down lp.bugs.windmill.testing.BugsWindmillLayer in 7.799 seconds.
>  Set up lp.code.windmill.testing.CodeWindmillLayerwindmill.bin.shell_objects:
> INFO     ['/tmp/tmprIqfXM.mozrunner/mozrunner-firefox', '-profile',
> '/tmp/tmprIqfXM.mozrunner',
> 'http://code.launchpad.dev:8085/windmill-serv/start.html']
>  in 5.843 seconds.
> Traceback (most recent call last):
>  File "./bin/test", line 275, in <module>
>    testrunner.run([])
...
>  File "/home/deryck/projects/launchpad/lp-branches/rockstar-js-refresh/lib/canonical/testing/layers.py",
> line 462, in check
>    "Test didn't reset the socket default timeout.")
> canonical.testing.layers.LayerIsolationError: Test didn't reset the
> socket default timeout.
> No handlers could be found for logger "lazr.smtptest"
> firefox-bin: Fatal IO error 11 (Resource temporarily unavailable) on X
> server :99.0.
>
>
> So perhaps something similar is happening with the WindmillLayer in
> all these hanging/disappearing branches?
>

One way to check this is to add a test to
lib/devscripts/ec2test/tests/test_remote.py that spawns a custom test
that explodes during layer set up, and then see if any email gets sent
out.

jml