1615241 - Crash in [@ AsyncShutdownTimeout | profile-change-teardown | Extension shutdown: [e-mail],Extension shutdown: [e-mail],Extension shutdown: [e-mail],Extension shutdown: [e-mail]]

Status: RESOLVED DUPLICATE of bug 1464938

(Toolkit :: Add-ons Manager, defect)

Description

[Alexandru Trif, Desktop Test Engineering [:atrif]]

This bug is for crash report bp-f18222a7-7983-4f23-b950-d3d160200213.

Top 10 frames of crashing thread:

0 mozglue.dll mozalloc_abort memory/mozalloc/mozalloc_abort.cpp:33
1 xul.dll Abort xpcom/base/nsDebugImpl.cpp:442
2 xul.dll NS_DebugBreak xpcom/base/nsDebugImpl.cpp
3 xul.dll nsDebugImpl::Abort xpcom/base/nsDebugImpl.cpp:134
4 xul.dll XPTC__InvokebyIndex 
5  @0x1bf1451f 
6 xul.dll trunc 
7 xul.dll trunc 
8 xul.dll trunc 
9 xul.dll static XPCWrappedNative::CallMethod js/xpconnect/src/XPCWrappedNative.cpp:1150

Unfortunately, I have no STR for this. The crash reporter was suddenly displayed. This happened to me two times today.
Also, I don't know the exact component for this bug. Please feel free to change it. Thanks!

Comment 1

[Daniel Bodea [:danibodea]]

According to the crash signature, it appears that this occurs on Windows 7, 8 and 10, and on Release 72 and Release 73, but mostly on Windows 10 and Firefox Release v73.0and on an amd64 build architecture. Unfortunately, it does not have steps to reproduce, so a regression window can not be attempted.

I would like to attempt to provide a regression range if STR is provided. Thanks.
We need someone to read the crash report and give us some feedback on its reproduction in order to continue the investigation.

I am moving this but to the (Toolkit) Crash Reporting component in an attempt to find someone who can have an opinion on this bug's reproduction and supposedly find its real component. Please help and thank you.

Comment 2

[BugBot [:suhaib / :marco/ :calixte]]

Bugbug thinks this bug is a regression, but please revert this change in case of error.

Comment 3

[Ryan VanderMeulen [:RyanVM]]

This looks like a pretty big spike :\

Comment 4

[[:philipp]]

going through some user comments for the signature, users are curiously describing a startup crash (with symptoms like error message 0xc0000005 which we appear to have fixed in 73.0.1 in bug 1614885). among many others:

• bp-96c85329-7583-4803-a92d-f304c0200218: "FireFox.exe-Appication error The application was unable to start correctly (Oxc0000005)"
• bp-2c2f4d75-0223-42e1-92ab-689530200218: "Suddenly Firefox quit working - it will not allow new tabs to load, then won't allow pre-set tabs at start up to load. Screen stays completely white (blank)...just sits there. I did restarts and nothing improved."
• bp-3ff5d1cf-94b2-4276-a6c6-1b2300200218: "when I launch Mozilla nothing happens and does not allow me to search anything.Page remains blank. I restarted browser several times and been unsuccessful."

so perhaps this signature was a side effect of bug 1614885 - in the affected state back then i also noticed that firefox.exe wasn't properly shutting down after closing the UI, so after 60 seconds when an artificial crash is triggered maybe it resulted in this signature.

in 73.0 [@ AsyncShutdownTimeout | profile-change-teardown | Extension shutdown: [e-mail],Extension shutdown: [e-mail],Extension shutdown: [e-mail],Extension shutdown: [e-mail] ] accounted for 3.6% of browser crashes, in early data from 73.0.1 it's just 0.3%.

Comment 5

[[:philipp]]

the volume here is no longer in a range that's concerning after 73.0.1, so we don't have to worry about this signature in particular anymore.