CrashKill/2013-07-22

General

Socorro

• Shipped 53 bugs
• Added a third database master to give us n+1
• Repos merged: broke some things, froze trees. Now open again
  • Crash classifier slipped

Flash

Desktop

Beta

• bug 877164 (nsNodeUtils::LastRelease) looks high but that's mostly dupes. Not too sure what to do with it.
• bug 860641 (RelevantKnowledge) sadly is still trickling down only slowly, and is still high.

Aurora

• bug 886295 (nsPrintEngine::PrePrintPage, PDF printing) is #16 on Aurora - for a PDF- and printing-related crash, that seems high.

Trunk

• bug 896200 (nsFontFaceLoader::Cancel, DestroyIterator) is the new #1 crash on trunk, needs owner (jfkthame has taken a look though).
• bug 896225 (gfxFcFontSet::SortPreferredFonts) is another new, unowned, high-volume crash.

Mobile

• crash in _JNIEnv::CallStaticVoidMethod | mozilla::AndroidBridge::SendThumbnail - bug 890590
  • Have steps now
• Bug 871290 - crash in js::ion::Assembler::TraceJumpRelocations - bug 871290
  • needinfo'ed to mrosenberg for 14 days

B2G

• 1.0.1 Buri builds are no longer accessible; they moved over to 1.1
  • bug 881431 track the no crash reports issue; fixed in later build, not in the release build
  • bug 894409 is marked fixed, need to work with partners.
• bug 891824, bug 891027 fixed, ZTE symbols should be back up

nightly

• bug 880818 - SkiaGL canvas crashes when loading font
• bug 878366 - Crash [@ mozilla::dom::ContentChild::ProcessingError]

release

• bug 895246 - lots of dupe signatures happening on the Spain build - any idea why?
  • Some of the uptimes are different; how are the duplicates counted for? ie. if there are two of the same crashes, etc. are they both marked dup?
  • are the signatures based on the time of the upload or the time of the crash?
• two crashes that are marked fixed but still occur. Need to investigate why:
  • bug 838215 - crash in mozilla::layout::RenderFrameParent::ContentReceivedTouch
  • bug 833964 - crash [@ mozilla::layers::AsyncPanZoomController::ContentReceivedTouch]
    • Follow up: seems ok. looks like 1.0.1 crashes that are fixed in 1.1

Notes