Starting on inbound cset https://hg.mozilla.org/integration/mozilla-inbound/rev/082542b01af8 there have been extremely frequent (~two thirds of the runs) linux32-only mochitest-5 failures: https://tbpl.mozilla.org/?tree=Mozilla-Inbound&jobname=mochitests-5 That cset was a backout, so not really sure what to make of this. I've clobbered, but it hasn't helped. Filing this so we have something to point at whilst waiting for retriggers. Is currently holding up the merge to m-c in time for the uplift. On https://hg.mozilla.org/integration/mozilla-inbound/rev/082542b01af8: -> [Orange] https://tbpl.mozilla.org/php/getParsedLog.php?id=13567653&tree=Mozilla-Inbound -> [Orange] https://tbpl.mozilla.org/php/getParsedLog.php?id=13568016&tree=Mozilla-Inbound -> [Green] https://tbpl.mozilla.org/php/getParsedLog.php?id=13568994&tree=Mozilla-Inbound On https://hg.mozilla.org/integration/mozilla-inbound/rev/b85bf85ebc06: -> [Orange] https://tbpl.mozilla.org/php/getParsedLog.php?id=13570071&tree=Mozilla-Inbound -> [Orange] https://tbpl.mozilla.org/php/getParsedLog.php?id=13570491&tree=Mozilla-Inbound -> [Green] https://tbpl.mozilla.org/php/getParsedLog.php?id=13570856&tree=Mozilla-Inbound There are 3 assertions: -> ###!!! ASSERTION: Oops! You're asking for a weak reference to an object that doesn't support that.: 'factoryPtr', file nsWeakReference.cpp, line 77 -> ###!!! ASSERTION: Profile change cancellation.: 'Error', file ../../../toolkit/xre/nsXREDirProvider.cpp, line 817 -> Assertion failure: !connections[i]->ConnectionReady(), at ../../../storage/src/mozStorageService.cpp:853

Found a few related bugs... > -> ###!!! ASSERTION: Oops! You're asking for a weak reference to an object > that doesn't support that.: 'factoryPtr', file nsWeakReference.cpp, line 77 Bug 622218 > -> ###!!! ASSERTION: Profile change cancellation.: 'Error', file > ../../../toolkit/xre/nsXREDirProvider.cpp, line 817 Bug 715137 > -> Assertion failure: !connections[i]->ConnectionReady(), at > ../../../storage/src/mozStorageService.cpp:853 Bug 764030

Ok, so it went away for one push (two runs since I retriggered), then came back again on the next push: https://tbpl.mozilla.org/php/getParsedLog.php?id=13574665&tree=Mozilla-Inbound Doesn't seem quite as frequent as first appeared, but still ideally don't want to let it off inbound.

Due to coalescling, we've only had a handful of M5 runs since the above. The failure rate does look slightly less drastic at least. I think we should wait a little bit longer just to make sure, before merging inbound to m-c.

https://tbpl.mozilla.org/php/getParsedLog.php?id=13578489&tree=Mozilla-Inbound https://tbpl.mozilla.org/php/getParsedLog.php?id=13585775&tree=Mozilla-Inbound

Whilst I'm still reluctant to let this migrate out of inbound, it is now seeming to not be frequent enough for us to actually track down a range - so I think we have no other choice.

https://tbpl.mozilla.org/php/getParsedLog.php?id=13604972&tree=Mozilla-Inbound https://tbpl.mozilla.org/php/getParsedLog.php?id=13605237&tree=Mozilla-Inbound

https://tbpl.mozilla.org/php/getParsedLog.php?id=15461444&tree=Mozilla-Beta (10.8, but that's pretty close to linux32)

(In reply to Ed Morley [:edmorley UTC+1] from comment #31) > https://tbpl.mozilla.org/php/getParsedLog.php?id=15619235&tree=Mozilla- > Inbound OS X

Also m-oth: https://tbpl.mozilla.org/php/getParsedLog.php?id=15723232&tree=Mozilla-Inbound

bug 795703 should remove the weakref assertion.

https://tbpl.mozilla.org/php/getParsedLog.php?id=16209642&tree=Mozilla-Inbound https://tbpl.mozilla.org/php/getParsedLog.php?id=16208838&tree=Mozilla-Inbound

Resolving WFM keyword:intermittent-failure bugs last modified >3 months ago, whose whiteboard contains none of: {random,disabled,marked,fuzzy,todo,fails,failing,annotated,time-bomb,leave open} There will inevitably be some false positives; for that (and the bugspam) I apologise. Filter on orangewfm.