CrashKill/2012-12-03

General

Any New/spiked Windows 8 crashes after the latest release
For Reference: Holiday coverage in December
- https://docs.google.com/spreadsheet/ccc?key=0Ait_RA8Il0WWdFhxT2NBQXRBdm5NaWJOWFllOE5uS0E#gid=0

PostgreSQL upgrade scheduled for Wednesday evening, will involve a short downtime window for processing and the UI.
Processing is now running the exploitability analysis.

Flash 11.5 crashes
- Marcia: Have we sorted out all the symbols issues? I remember a thread where we were missing symbols last week for some of the new beta versions.
- KaiRo: Yes, all sorted out for the current versions, and should that happen again, we know who to contact.

New crashes
- bug 817391 crash in gfxTeeSurface::`.* deleting destructor
- bug 816976 [10.8] crash in glcGetIOAccelService
- bug 817568 crash in StringCchCopyNW - correlated to Dealply

bug 770238 (Yandex-related js::gc::Arena::finalize) is still high, Yandex update didn't help, they are working on a minimized testcase, the bug might be on our side.

We had a spike on the 29th, which was promptly fixed but we still see crashes from it. By-build-date reports show the one-day spike nicely, though.

bug 815684 is adding some logcat info to crash reports for JB, we hope that will help us diagnose some high-volume crashes.
Nightly and Aurora has a spike yesterday. Not sure exactly why at this moment in time, nor whether or not it will or will not go down.
- lots of corrupt signatures and flash crashes.
bug 775142 is fixed, needs to be pushed to aurora.
bug 791441 affects nightly and aurora only, we are guessing that it has something to do with updating.

We seem to have almost no testing on weekends. For example, yesterday, only a single crash report from a device came in! :(
- Single crash was likely marcia since she was using the phone yesterday and crashed
bug 811335 seems to be moving forward with the help of Julian Seward.
bug 814078 is still waiting for feedback/review.