I just triggered a rerun, seeing a crash with in https://travis-ci.org/mozilla-b2g/gaia/jobs/14610294
test_launch_everything_me_app (test_everythingme_launch_app.TestEverythingMeLaunchApp) ... ERROR Dont think its solved yet On 27 November 2013 21:19, Jonathan Griffin <jgrif...@mozilla.com> wrote: > Gaia tests on TBPL haven't been showing any increase in intermittents. > > Travis is really a bad UI for sheriffing these kinds of problems, and > we're not dumping gecko.log from Travis which doesn't help either. We > should dump gecko.log at the end of the run so we can see what sorts of > errors the build is sending to stdout/stderr. > > Some digging shows a few details. We haven't seen the apparent crash in > the last few hours; all the failures more recent than about 7 hours ago are > single-test failures, mostly of tests that we're not running in TBPL > (yet). Probably, these are flakey tests, and certainly, we should be using > the same manifest in TBPL as we do in Travis so we can use TBPL to catch > flakey test issues. > > The crash occurred several times when the tests were using the build > http://ftp.mozilla.org/pub/mozilla.org/b2g/tinderbox-builds/mozilla-central-linux64_gecko/1385546575/b2g-28.0a1.multi.linux-x86_64.tar.bz2, > and a few earlier builds, but haven't been seen with later builds...maybe > this problem has already resolved itself? > > Jonathan > > > On 11/27/2013 12:58 PM, Dale Harvey wrote: > > mentioning marionette was just a stab in the dark, I was more saying > that it seems like something handwavey infrastructurey than specific > component failures, something crashy when going through the b2g code path > seems feasible > > If anyone better than me at reading tbpl could see if there was any point > in the last day in which gaia tests got less stable, that would be useful > information > > > > On 27 November 2013 20:47, Jonathan Griffin <jgrif...@mozilla.com> wrote: > >> Yes, that landed on Saturday, so if the problems you're seeing date back >> that far, that's a potential culprit. >> >> Jonathan >> >> >> On 11/27/2013 11:42 AM, Gareth Aye wrote: >> >> Either way this is not necessarily a marionette server issue. >> >> >> On Wed, Nov 27, 2013 at 2:42 PM, Gareth Aye <gareth....@gmail.com> wrote: >> >>> http://hg.mozilla.org/mozilla-central/rev/1fd1596ffb9f also? >>> >>> >>> On Wed, Nov 27, 2013 at 2:39 PM, Jonathan Griffin >>> <jgrif...@mozilla.com>wrote: >>> >>>> The only Marionette change to have gone in recently is: >>>> >>>> http://hg.mozilla.org/mozilla-central/rev/9cc147e1d222 >>>> >>>> Jonathan >>>> >>>> >>>> On 11/27/2013 11:28 AM, Gareth Aye wrote: >>>> >>>>> Also fwiw I don't understand how this could have landed if the python >>>>> tests >>>>> are running on tbpl... >>>>> >>>>> On Wed, Nov 27, 2013 at 2:16 PM, Gareth Aye <gareth....@gmail.com> >>>>> wrote: >>>>> >>>>> Zac, >>>>>> >>>>>> I'm a bit busy with other work right now. Would you mind bisecting >>>>>> gecko >>>>>> to find the culprit patch for us? >>>>>> >>>>>> >>>>>> On Wed, Nov 27, 2013 at 2:10 PM, Zac Campbell <zcampb...@mozilla.com >>>>>> >wrote: >>>>>> >>>>>> On 27/11/13 18:29, Dale Harvey wrote: >>>>>>> >>>>>>> Tracking bug is: >>>>>>>> https://bugzilla.mozilla.org/show_bug.cgi?id=943980 >>>>>>>> >>>>>>>> Been seeing test failures on travis for the last 6 hours ( >>>>>>>> https://travis-ci.org/mozilla-b2g/gaia/builds) so we have closed >>>>>>>> the >>>>>>>> tree >>>>>>>> >>>>>>>> The failures wildly vary and all seem to be intermittent, so this is >>>>>>>> likely >>>>>>>> due to a new b2g build / marionette changes, currently debugging >>>>>>>> (although >>>>>>>> I will be afk for ~90minutes) so if anyone else wants to help debug >>>>>>>> that >>>>>>>> would be great :) >>>>>>>> >>>>>>>> Will reopen as soon as we get consistent passing builds >>>>>>>> >>>>>>>> Cheers >>>>>>>> Dale >>>>>>>> _______________________________________________ >>>>>>>> dev-b2g mailing list >>>>>>>> dev-b2g@lists.mozilla.org >>>>>>>> https://lists.mozilla.org/listinfo/dev-b2g >>>>>>>> >>>>>>>> wrt the gaia-ui-tests failures: >>>>>>> "Connection Refused" does look like the whole b2g/gecko is crashing. >>>>>>> The >>>>>>> console trace is similar to when we run these tests on device and the >>>>>>> device crashes. >>>>>>> >>>>>>> Also the one about: >>>>>>> TypeError: app is undefined ; stacktrace: execute_async_script >>>>>>> @gaia_test.py, line 74 >>>>>>> inline javascript, line 251 >>>>>>> src: " let result = {frame: app.frame.firstChild," >>>>>>> >>>>>>> I have witnessed on device and this is when b2g fails to start up. >>>>>>> The >>>>>>> device was stuck on the splash screen, but it is conscious enough >>>>>>> for it to >>>>>>> be stopped and restarted (and thus the test suite continue). >>>>>>> >>>>>>> >>>>>>> There was one test failing intermittently earlier in the day: >>>>>>> test_edit_contact.py >>>>>>> I've prepared a fix so it doesn't stop you from re-opening the tree: >>>>>>> https://github.com/mozilla-b2g/gaia/pull/14136 >>>>>>> >>>>>>> >>>>>>> Good luck, >>>>>>> Zac >>>>>>> _______________________________________________ >>>>>>> dev-b2g mailing list >>>>>>> dev-b2g@lists.mozilla.org >>>>>>> https://lists.mozilla.org/listinfo/dev-b2g >>>>>>> >>>>>>> >>>>>> >>>>>> -- >>>>>> Best, >>>>>> Gareth >>>>>> >>>>>> >>>>>> >>>>> >>>> >>> >>> >>> -- >>> Best, >>> Gareth >>> >>> >> >> >> -- >> Best, >> Gareth >> >> >> > > _______________________________________________ dev-b2g mailing list dev-b2g@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-b2g
