The following crash has been seen while running stability tests. [@ mozilla::a11y::DocAccessible::ProcessContentInserted | mozilla::a11y::NotificationController::ContentInsertion::Process | mozilla::a11y::NotificationController::WillRefresh | nsRefreshDriver::Tick ] Maybe this is related to bug 1145724. Cafbot will upload minidump and extra file (with logs).

Hi Mike, Do you know who can take a look at this?

Observed on: Device: msm8909 Gonk Version: AU_LINUX_GECKO_LF.BR.1.2.3.00.00.00.000.110 Moz BuildID: 20150322002503 Manifest: https://www.codeaurora.org/cgit/quic/lf/b2g/manifest/tree/caf_AU_LINUX_GECKO_LF.BR.1.2.3.00.00.00.000.110.xml?h=release Gecko Version: 37.0 Gaia: http://git.mozilla.org/?p=releases/gaia.git;a=commit;h=44c62060581fde8de1e12e94cf55e9673b401a47 Gecko: http://git.mozilla.org/?p=releases/gecko.git;a=commit;h=c36e800ae5e5e7e790129e18edaef202d8d3dbff Patches: bug 1133147, bug 1139469, bug 1145724, bug 1142770, bug 1146987

is there a link at mxr to see where it crashes (the dump is not applied to trunk)

Hi David, Can you assign someone on your team to look into this issue? Thanks, Mike

Alex it looks like either container or aContainer must be null? (in DocAccessible::ProcessContentInserted)

(In reply to David Bolter [:davidb] from comment #9) > Alex it looks like either container or aContainer must be null? (in > DocAccessible::ProcessContentInserted) what is changeset the dump is for? it's quite different from nightly sources.

Observed on: Device: msm8909 Gonk Version: AU_LINUX_GECKO_LF.BR.1.2.3.00.00.00.000.113 Moz BuildID: 20150323002504 Manifest: https://www.codeaurora.org/cgit/quic/lf/b2g/manifest/tree/caf_AU_LINUX_GECKO_LF.BR.1.2.3.00.00.00.000.113.xml?h=release Gecko Version: 37.0 Gaia: http://git.mozilla.org/?p=releases/gaia.git;a=commit;h=7f367fc98ffdd183f21d2cdfe20556ab877ece34 Gecko: http://git.mozilla.org/?p=releases/gecko.git;a=commit;h=3970b1e517b6447237421c1f9e7845c00bde0838 Patches: bug 1143694, bug 1146987, bug 1145724, bug 1133147, bug 1139469, bug 1142770

(In reply to alexander :surkov from comment #10) > (In reply to David Bolter [:davidb] from comment #9) > > Alex it looks like either container or aContainer must be null? (in > > DocAccessible::ProcessContentInserted) > what is changeset the dump is for? it's quite different from nightly sources. It's Gecko 37, which is currently in the mozilla-beta repository.

With null pointer check, bug 1145724 has not crashed in latest test runs (80+ hours). But this crash is now seen 8 time so far. This bug is now the biggest obstacle to meeting stability goals. Can we perform similar null check to prevent the crash?

Comment on attachment 8584545 [details] [diff] [review] patch Review of attachment 8584545 [details] [diff] [review]: ----------------------------------------------------------------- r=me, looks like the right place for the check. Thanks!

btw, if you can reproduce it then could you enable "tree" a11y logging (http://asurkov.blogspot.ca/2012/10/debugging-firefox-accessibility.html) and file output here. It'd be interesting to know exactly why the container is not in DOM tree longer.

Greg re: your comment 15: > With null pointer check, bug 1145724 has not crashed in latest test runs > (80+ hours). That's great news. > ...But this crash is now seen 8 time so far. This bug is now the > biggest obstacle to meeting stability goals. Can we perform similar null > check to prevent the crash? Does the proposed check in attachment 8584545 [details] [diff] [review] suffice? Mike

(In reply to Mike Lee [:mlee] from comment #19) > Does the proposed check in attachment 8584545 [details] [diff] [review] > suffice? it should be

I'm going to apply the patch today, hopefully we can see the benefit from it by Monday. Thanks for the quick turnaround. To enable logging, I tried |export A11YLOG=focus,tree,DOMEvents;| but it didn't seem to have any effect for me. But I also don't see Screen Reader option in setting menu.

We need this fixed for v2.2

surkov/:dbolter, can you please fill the details here to consider uplift for b2g37 branch ? Thanks! NOTE: Please see https://wiki.mozilla.org/Release_Management/B2G_Landing to better understand the B2G approval process and landings. [Approval Request Comment] Bug caused by (feature/regressing bug #): User impact if declined: Testing completed: Risk to taking this patch (and alternatives if risky): String or UUID changes made by this patch:

(In reply to Greg Grisco from comment #21) > I'm going to apply the patch today, hopefully we can see the benefit from it > by Monday. Thanks for the quick turnaround. > > To enable logging, I tried |export A11YLOG=focus,tree,DOMEvents;| but it > didn't seem to have any effect for me. But I also don't see Screen Reader > option in setting menu. Greg, was anyone able to verify the fix?

(In reply to David Bolter [:davidb] from comment #25) > Greg, was anyone able to verify the fix? Hi David, we didn't see the crash in first 60 hours of test, so it looks good so far. Probably best if we wait one more day before claiming victory though. I'll keep my ni? and report back tomorrow.

(In reply to bhavana bajaj [:bajaj] from comment #24) > surkov/:dbolter, can you please fill the details here to consider uplift for [Approval Request Comment] Bug caused by (feature/regressing bug #): unknown User impact if declined: crashes Testing completed: yes Risk to taking this patch (and alternatives if risky): low, might affect on screen readers String or UUID changes made by this patch: no

We're still not seeing this reproduce, thanks!