1275162 - Crash in shutdownhang | CacheIOThread::Shutdown | CacheFileIOManager::Shutdown

Status: RESOLVED WONTFIX

(Core :: Networking: Cache, defect, P3)

Description

[Ting-Yu Chou [:ting] (away)]

This bug was filed from the Socorro interface and is 
report bp-22667b2e-2405-43ab-b515-dcfd22160523.
=============================================================

There are 27 crashes at #1 [shutdownhang | ntdll.dll@0xa5164] of Nightly 20160522030240, this is one of them.

Comment 1

[Honza Bambas (:mayhemer)]

According the report, this seems like a machine with a rootkit or something.  Here is an example report having the same ntdll version, but different (correct) debug id for it.  I can confirm correctness locally too.

Hard to say what to do here.  Probably INVALID?

Comment 2

[Patrick McManus [:mcmanus]]

more likely a dup of bug 1275167 with nsAppShell::ProcessNextNativeEvent(bool) on the top of the main thread stack.

Comment 3

[Michal Novotny [:michal]]

I checked all reports for version 49.0a1 with this signature and almost all belong to bug 1275167. But I found 2 cache related that should be IMO fixed by bug 1271701:

https://crash-stats.mozilla.com/report/index/46947dfc-d040-4653-b27b-764f52160521#allthreads
https://crash-stats.mozilla.com/report/index/11737207-da60-4648-a0ee-0c40e2160522#allthreads

Honza, do you have any idea why the fix didn't help here?

Comment 4

[Kyle Huey (Exited; not receiving bugmail, old account, do not use)]

(In reply to Patrick McManus [:mcmanus] from comment #2)
> more likely a dup of bug 1275167 with
> nsAppShell::ProcessNextNativeEvent(bool) on the top of the main thread stack.

Have you actually downloaded the minidump?  None of the minidumps in bug 1275167 have ProcessNextNativeEvent on the stack in Visual Studio.

Comment 5

[Patrick McManus [:mcmanus]]

I don't have privs for the minidump - that's what crash stats shows for a stack. you're saying its bogus? Haven't seen it before and I've read a lot of crash stats for (related) hangs..

Comment 6

[Kyle Huey (Exited; not receiving bugmail, old account, do not use)]

(In reply to Patrick McManus [:mcmanus] from comment #5)
> I don't have privs for the minidump - that's what crash stats shows for a
> stack. you're saying its bogus? Haven't seen it before and I've read a lot
> of crash stats for (related) hangs..

None of the links in this bug have ProcessNextNativeEvent on the stack in the Socorro page, and the ones in bug 1275167 that have it in Socorro do *not* have it in VS (although the stacks are generally otherwise accurate).

Comment 7

[Ting-Yu Chou [:ting] (away)]

The stack shows it gets stuck at CacheIOThread::Shutdown for waiting CacheIOThread termination, not like bug 1275167 waiting in nsEventQueue::GetEvent. So I don't think this is a dup of bug 1275167.

Comment 8

[Michal Novotny [:michal]]

(In reply to Ting-Yu Chou [:ting] from comment #7)
> The stack shows it gets stuck at CacheIOThread::Shutdown for waiting
> CacheIOThread termination, not like bug 1275167 waiting in
> nsEventQueue::GetEvent. So I don't think this is a dup of bug 1275167.

But there is no cache IO thread in the report.

Comment 9

[Honza Bambas (:mayhemer)]

(In reply to Michal Novotny (:michal) from comment #3)
> I checked all reports for version 49.0a1 with this signature and almost all
> belong to bug 1275167. But I found 2 cache related that should be IMO fixed
> by bug 1271701:
> 
> https://crash-stats.mozilla.com/report/index/46947dfc-d040-4653-b27b-
> 764f52160521#allthreads
> https://crash-stats.mozilla.com/report/index/11737207-da60-4648-a0ee-
> 0c40e2160522#allthreads
> 
> Honza, do you have any idea why the fix didn't help here?

Probably because we are hanging for a long time inside OpenFile or FileRead call?

Comment 10

[Michal Novotny [:michal]]

I found only 2 crashes for 49.0a1 with this stack:

https://crash-stats.mozilla.com/report/index/77943e31-38c1-454f-8df8-fc16c2160523#allthreads
https://crash-stats.mozilla.com/report/index/22667b2e-2405-43ab-b515-dcfd22160523#allthreads

Putting to backlog for now until we have more reports with some information about the cache IO thread.

Comment 11

[Firefox Bug Husbandry Bot]

Bulk change to priority: https://bugzilla.mozilla.org/show_bug.cgi?id=1399258

Comment 12

[Firefox Bug Husbandry Bot]

Bulk change to priority: https://bugzilla.mozilla.org/show_bug.cgi?id=1399258

Comment 13

[Sylvestre Ledru [:Sylvestre]]

Closing because no crash reported since 12 weeks.