Intermittent test_browserElement_oop_getWebManifest.html | application terminated with exit code 11 after Assertion failure: !mCanSend, at /builds/slave/fx-team-lx-d-00000000000000000/build/src/gfx/layers/ipc/CompositorChild.cpp:398

RESOLVED FIXED

Status

()

defect
RESOLVED FIXED
4 years ago
3 years ago

People

(Reporter: philor, Assigned: pchang)

Tracking

({assertion, intermittent-failure})

Firefox Tracking Flags

(Not tracked)

Details

(Whiteboard: [gfx-noted])

Assignee

Updated

4 years ago
Whiteboard: [gfx-noted]
Assignee

Updated

4 years ago
Assignee: nobody → howareyou322
Comment hidden (Intermittent Failures Robot)
Assignee

Comment 2

3 years ago
After checking this, the watchdog detected system hung and kill the chrome process. Therefore, it caused assertion on child side because it didn't go through normal termination flow.


 11:28:24     INFO -  Hit MOZ_CRASH(Shutdown too long, probably frozen, causing a crash.) at /builds/slave/fx-team-lx-d-00000000000000000/build/src/toolkit/components/terminator/nsTerminator.cpp:149
 11:28:53     INFO -  #01: mozilla::::RunWatchdog [toolkit/components/terminator/nsTerminator.cpp:149]
 11:28:53     INFO -  #02: _pt_root [nsprpub/pr/src/pthreads/ptthread.c:215]
 11:28:53     INFO -  #03: libpthread.so.0 + 0x6d4c
 11:28:53     INFO -  Assertion failure: !mCanSend, at /builds/slave/fx-team-lx-d-00000000000000000/build/src/gfx/layers/ipc/CompositorChild.cpp:398
 11:28:53     INFO -  #01: mozilla::layers::CompositorChild::ActorDestroy(mozilla::ipc::IProtocolManager<mozilla::ipc::IProtocol>::ActorDestroyReason) [gfx/layers/ipc/CompositorChild.cpp:398]
 11:28:53     INFO -  #02: mozilla::layers::PCompositorChild::DestroySubtree(mozilla::ipc::IProtocolManager<mozilla::ipc::IProtocol>::ActorDestroyReason) [obj-firefox/ipc/ipdl/PCompositorChild.cpp:1384]
 11:28:53     INFO -  #03: mozilla::layers::PCompositorChild::OnChannelError() [obj-firefox/ipc/ipdl/PCompositorChild.cpp:1294]
 11:28:53     INFO -  #04: mozilla::ipc::MessageChannel::NotifyMaybeChannelError() [ipc/glue/MessageChannel.cpp:1848]
Assignee

Comment 3

3 years ago
(In reply to peter chang[:pchang][:peter] from comment #2)
> After checking this, the watchdog detected system hung and kill the chrome
> process. Therefore, it caused assertion on child side because it didn't go
> through normal termination flow.
> 
> 
>  11:28:24     INFO -  Hit MOZ_CRASH(Shutdown too long, probably frozen,
> causing a crash.) at
> /builds/slave/fx-team-lx-d-00000000000000000/build/src/toolkit/components/
> terminator/nsTerminator.cpp:149
>  11:28:53     INFO -  #01: mozilla::::RunWatchdog
> [toolkit/components/terminator/nsTerminator.cpp:149]
>  11:28:53     INFO -  #02: _pt_root [nsprpub/pr/src/pthreads/ptthread.c:215]
>  11:28:53     INFO -  #03: libpthread.so.0 + 0x6d4c
>  11:28:53     INFO -  Assertion failure: !mCanSend, at
> /builds/slave/fx-team-lx-d-00000000000000000/build/src/gfx/layers/ipc/
> CompositorChild.cpp:398
>  11:28:53     INFO -  #01:
> mozilla::layers::CompositorChild::ActorDestroy(mozilla::ipc::
> IProtocolManager<mozilla::ipc::IProtocol>::ActorDestroyReason)
> [gfx/layers/ipc/CompositorChild.cpp:398]
>  11:28:53     INFO -  #02:
> mozilla::layers::PCompositorChild::DestroySubtree(mozilla::ipc::
> IProtocolManager<mozilla::ipc::IProtocol>::ActorDestroyReason)
> [obj-firefox/ipc/ipdl/PCompositorChild.cpp:1384]
>  11:28:53     INFO -  #03:
In my opinion, it is normal behavior to invoke ActorDestroy in child side since chrom process got hang problem. But we don't need to raise assertion in child side, especially in graphic code. Someone has to analyze the call stacks generated from system hang.

I will create another bug to remove the assertion.

> mozilla::layers::PCompositorChild::OnChannelError()
> [obj-firefox/ipc/ipdl/PCompositorChild.cpp:1294]
>  11:28:53     INFO -  #04:
> mozilla::ipc::MessageChannel::NotifyMaybeChannelError()
> [ipc/glue/MessageChannel.cpp:1848]
Assignee

Updated

3 years ago
Blocks: 1250418
Comment hidden (Intermittent Failures Robot)
Assignee

Updated

3 years ago
Flags: needinfo?(philringnalda)
See Also: → 1253883
Flags: needinfo?(philringnalda)
Assignee

Comment 7

3 years ago
Ryan, thanks for the info. I will close this case.

The problem in Cache2 I/O thread was checked in bug 1253883. Case closed.
Status: NEW → RESOLVED
Closed: 3 years ago
Resolution: --- → FIXED
Comment hidden (Intermittent Failures Robot)
You need to log in before you can comment on or make changes to this bug.