Difference between revisions of "NightlyCrashTriage/2017Q4"

From MozillaWiki
Jump to: navigation, search
(11-8 Nightly)
Line 1: Line 1:
 +
== 2017-11-10 - mccr8 ==
 +
11-8 Nightly
 +
* {{Bug|1257387}} - This OOM signature spiked on Nightly, but almost entirely from a single installation. Commented.
 +
* Otherwise, just a lot of single installation crashes.
 +
 
== 2017-11-09 - marcia ==
 
== 2017-11-09 - marcia ==
 
Builds from Nov 6/7th
 
Builds from Nov 6/7th

Revision as of 18:08, 10 November 2017

2017-11-10 - mccr8

11-8 Nightly

  • bug 1257387 - This OOM signature spiked on Nightly, but almost entirely from a single installation. Commented.
  • Otherwise, just a lot of single installation crashes.

2017-11-09 - marcia

Builds from Nov 6/7th

  • Didn't see any recent activity, so adding builds up to the 7th
  • Filed bug 1415151 - Crash in av1_temporal_filter_apply_sse2 - Looks like a signature morph
  • Filed bug 1415754 - Crash in rx::d3d11::DynamicCastComObject<T>
  • Filed bug 1415884 - Crash in SkBitmap::SkBitmap
  • Filed bug 1414853 - Crash in nsDisplayImageContainer::ConfigureLayer
  • Filed bug 1415945 - Crash in ShortLivedStringBuffer<T>::Destroy
  • Added a few signatures to existing bugs

2017-11-06 - marcia

Nov 04/5th Nightly

  • Filed bug 1414853 Crash in nsDisplayImageContainer::ConfigureLayer
  • Filed bug 1414877 Crash in mozilla::gfx::ReadSkImage - turns out it is a dupe of another bug
  • Filed bug 1414936 Mac printing crash in <name omitted> | mozilla::layout::PrintTranslator::TranslateRecording
  • bug 1413500 is quite high on Android

2017-11-06 - njn

Nov 03 Nightly

  • Filed bug 1414686 - Crash in nsCanvasFrame::Reflow.
  • Commented on some bugs.

2017-11-05 - jseward

11-02 Nightly

  • Windows 20171102100041 #17 [webrtc::Merge::SignalScaling]: filed bug 1414632
  • Android 20171102100039 #3[ <name omitted> | js::TraceGenericPointerRoot]: filed bug 1414634

2017-11-02 - mccr8

11-1 Nightly

  • Nothing interesting. A number of new crashes, but they were all single installation.

2017-11-01 - jchen

10-31 Nightly

  • Android #1 mozilla::gl::GLBlitHelper::BlitImage: commented in bug 1413500

2017-11-01 - kanru

Oct 30 Nightly

  • It seems 10/30 nightly was late so there weren't many crashes.

2017-10-30 - marcia

Oct 28/29 Nightly (covering for Ting)

  • Filed bug 1412848 Crash in memcpy | mozilla::gfx::RecordedScaledFontCreationByIndex::Record<T>
  • Filed bug 1412853 Fennec Crash in compiler_builtins::int::udiv::__udivsi3::__aeabi_uidiv
  • The second build on the 29th doesn't seem to show any crashes: http://bit.ly/2yXP8Mw
  • Skipped several single install crashes such as:
    • mozilla::dom::UIEvent::GetMovementPoint
    • vbufbackend_gecko_ia2.dll@0x1cae0
  • Still seeing 10.13 Mac signatures such as @0x7fff414b73b0. In 20171029102300 there are 8 overall crashes, 6 unique signatures.

2017-10-30 - njn

Oct 27 Nightly

  • Ignored various single-installation crashes.
  • Commented on some existing bugs.

2017-10-29 - jseward

Oct 26 Nightly

  • Win 20171026221945 #16 [PR_Write | mozilla::layout::PRFileDescStream::write]: filed bug 1412643
  • Win 20171026221945 #25 [mozilla::places::History::KeyClass::`scalar deleting destructor]: filed bug 1412647

2017-10-26 - mccr8

10-25 Nightly

  • bug 1411685 - Crash in <T>::operator() | mozilla::detail::RunnableFunction<T>::Run. Commented on bug.
  • bug 1412059 - Crash in nsTArray_Impl<T>::IndexOfFirstElementGt<T> | `anonymous namespace::AxisPartition::InsertCoord. Filed.
  • nsCOMPtr<T>::nsCOMPtr<T> | nsXULPopupManager::ExecuteMenu. Ignored this single installation crash.

2017-10-25 - jchen

10-24 Nightly

  • Windows #10 nsHtml5TreeOpExecutor::ContinueInterruptedParsingAsync: single installation

2017-10-25 - kanru

Oct 23 Nightly


2017-10-23 - marcia

Oct 21/22 Nightly (covering for Ting)

  • Fair amount of single installation crashes, but nothing significant to note
  • bug 1400637 has pretty high crash volume and has been nominated for tracking

2017-10-23 - njn

Oct 20 Nightly

  • Not a lot going on.

2017-10-22 - jseward

Oct 19 Nightly

  • Win 20171019100107: [mozilla::dom::XPathExpression::EvaluateWithContext]: 79 crashes, but only 1 installation. Ignored.
  • Linux 20171019100107: [mozalloc_abort | abort | audioipc_server::Server::poll]: I would have filed this, but got locked out of bugzilla for some reason.
  • Nothing else of interest.

2017-10-19 - mccr8

10-18-2017

  • bug 1410212 - stylo: Crash in NoteDirtyElement. Previous crash fixed, but more still showing up, so I filed a new bug.

2017-10-18 - kanru

Oct 16 Nightly

  • Nothing of interest.

2017-10-16 - marcia

October 14/15 Nightly (covering for Ting for 10/15)

  • Most of the new bugs are on file such as bug 1408638
  • New signature EnterJit appears to be another manifestion of an earlier JIT signature: bug 1408766
  • Ignored single install signature xul.dll@0x2fba378 | mozilla::layers::ShadowLayerForwarder::EndTransaction
  • Ignored diagnostic logging signature hashglobe::diagnostic::DiagnosticHashMap<T>::report_corruption<T>

2017-10-16 - njn

Oct 13 Nightly

  • Pretty quiet. I commented on a few bugs.

2017-10-14 - jseward

Oct 12 Nightly

  • (windows) [mozilla::mscom::Interceptor::GetInitialInterceptorForIID] bug 1399557: still happening despite being closed. Commented.
  • (osx): filed bug 1408629: Crash in shutdownhang | libsystem_kernel.dylib@0x1be7e
  • There seem to be a number of single-instance crashes to do with threads at shutdown on OSX.
  • Nothing else of note.

2017-10-13 - mccr8

Oct 11 Nightly (looks like jseward did this to? or maybe mislabeled his entry)

  • filed: bug 1408445 - Crash in java.lang.IllegalArgumentException: codec does not support type at android.os.Parcel.readException(Parcel.java).
  • Nothing else of interest that hadn't already been fixed.

2017-10-12 - jseward

Oct 11 Nightly

  • Windows: [arena_salloc | moz_xrealloc | nsTArray_base<T>::EnsureCapacity<T> | nsTArray_Impl<T>::AppendElement<T> | mozilla::net::CacheIOThread::DispatchInternal]: filed bug 1408075

2017-10-11 - jchen

10-10 Nightly

  • Windows #7 mozilla::mscom::Interceptor::GetInitialInterceptorForIID: NI'ed in bug 1399557
  • Windows #12 mozilla::PrioritizedEventQueue<T>::HasReadyEvent: single installation
  • Windows #18 GetPropertyOperation: single installation

2017-10-10 - marcia

Oct 8/9 Nightly

  • Filed DOM/Service Workers crash bug 1407225. Fix is in the works
  • Filed new Android crash bug 1407266. Affects OnePlus devices only

2017-10-09 - marcia

Oct 7 Nightly

  • Filed bug 1406925 Crash in @0x0 | mozilla::wr::RenderDXGITextureHostOGL::Lock
  • Ignored several single install crashes:
    • RtlVirtualUnwind | CrashReporter::JitExceptionHandler
    • geckoservo::glue::Servo_GetComputedKeyframeValues
    • Android java.lang.IllegalArgumentException: radius must be > 0 at android.graphics.RadialGradient.<init>(RadialGradient.java)

2017-10-09 - njn

Oct 6 Nightly

  • Filed bug 1406812 - Crash in android.content.ActivityNotFoundException: No Activity found to handle Intent { act=android.intent.action.VIEW dat= } at android.app.Instrumentation.checkStartActivityResult(Instrumentation.java).

2017-10-06 - jseward

10-05 Nightly

  • Nothing much interesting.
  • Win 20171005220204: #8 [mozilla::mscom::Interceptor::GetInitialInterceptorForIID]: still happening quite often despite bug 1399557 being closed. Commented.

2017-10-05 - mccr8

10-4 Nightly

  • bug 1406245 - Crash in CoreVideo@0x2d6d. OS X 10.13 specific crash.
  • bug 1229384 - pages_commit. This is the top signature, by a large margin. This is an existing signature, but not in this volume. If it keeps up, somebody should look into it. Maybe there's a new leak or some new memory usage spike? It doesn't look that bad on 10-5 so far.

2017-10-04 - jchen

10-03 Nightly

  • Windows #3 RtlVirtualUnwind | CrashReporter::JitExceptionHandler: single installation
  • Windows #7 sxwmon64.dll@0x4b1ca: single installation
  • Windows #10 mozilla::ipc::Endpoint<T>::Bind: single installation
  • OS X #3 @0x0 | mozilla::dom::ContentChild::SetProcessName: single installation
  • Linux @0x0 | libfreshwrapper-flashplayer.so@*: single installation
  • Android #12 java.lang.IndexOutOfBoundsException: setSpan (43 ... 45) ends beyond length 44 at android.text.SpannableStringBuilder.checkRange(SpannableStringBuilder.java): filed bug 1405846
  • Android #13 java.lang.NullPointerException: Attempt to invoke virtual method void android.graphics.Bitmap.setHasAlpha(boolean) on a null object reference at android.graphics.Bitmap.createBitmap(Bitmap.java): filed bug 1405848

2017-10-04 - njn

Oct 02 Nightly

  • Filed bug 1405518 - Crash in mozilla::layers::ClientLayerManager::ForwardTransaction.
  • Filed bug 1405521 - Crash in nsGlobalWindow::ClearDocumentDependentSlots: MOZ_CRASH(Unhandlable OOM while clearing document dependent slots.)


2017-10-03 - ting

10-01-10 Nightly

  • Windows
    • #9 [js::jit::IonBuilder::visitTest] single installation, ignored
    • #11 [CanAttachNativeGetProp] single installation, ignored
    • #14 [nsStorageInputStream::Available] single installation, ignored
    • #19 [RtlpWaitOnCriticalSection | RtlpEnterCriticalSectionContended | RtlEnterCriticalSection | _cairo_win32_font_face_hash_table_lock] filed bug 1405198
    • #20 [js::ConstraintTypeSet::addType] single installation, ignored
    • #21 [JSScript::maybeSweepTypes] linked to bug 1380216
  • Android
    • #4 [java.lang.IllegalStateException: Trying to start session but it is already started at org.mozilla.gecko.telemetry.measurements.SessionMeasurements.recordSessionStart(SessionMeasurements.java)] filed bug 1405192
  • Linux
    • #3 [mozilla::ReflowInput::ReflowInput] single installation, ignored

10-01-22 Nightly

  • Mac
    • #5 [@0x0 | nsSSLSocketProvider::NewSocket] single installation, ignored

2017-10-02 - marcia

Sept 30/October 1

  • Filed new webrender crash bug 1404949
  • Filed bug 1404961 Present on Nightly and 57 beta.
  • Filed Memory allocator crash bug 1405062
  • Residual crashes from landing of OMTP. Will wait for the reland to file any additional crashes

2017-10-02 - njn

Sep 29 Nightly

  • Lots of new gfx crashes due to OMTP being enabled (bug 1403935); it was then disabled again a day or two later.
  • Commented on various bugs.