Closed Bug 1268993 Opened 3 years ago Closed 3 years ago

Intermittent test_presentation_1ua_sender_and_receiver_oop.html | Test timed out and three more timeouts breaking the entire run

Categories

(Core :: DOM: Core & HTML, defect, P3)

49 Branch
defect

Tracking

()

RESOLVED FIXED
mozilla52
Tracking Status
firefox49 --- wontfix
firefox50 --- fixed
firefox51 --- fixed
firefox52 --- fixed

People

(Reporter: KWierso, Assigned: kershaw)

References

(Blocks 1 open bug)

Details

(Keywords: intermittent-failure)

Attachments

(1 file, 1 obsolete file)

Bulk assigning P3 to all open intermittent bugs without a priority set in Firefox components per bug 1298978.
Priority: -- → P3
Hi Tommy, the failure rate gets higher since last week. Mind taking a look? Thanks.
Flags: needinfo?(kuoe0)
Summary: Intermittent test_presentation_1ua_sender_and_receiver_oop.html | Test timed out. → Intermittent test_presentation_1ua_sender_and_receiver_oop.html | Test timed out and three more timeouts breaking the entire run
Hi Hsin-Yi, Kershaw will take a look on this intermittent failure.
Flags: needinfo?(kuoe0) → needinfo?(kechang)
(In reply to Tommy Kuo [:KuoE0] from comment #6)
> Hi Hsin-Yi, Kershaw will take a look on this intermittent failure.

I will take a look.
Flags: needinfo?(kechang)
FWIW, I hit this in a debug build and got this stack:

 [task 2016-09-22T16:58:00.336213Z] 16:58:00     INFO -  Assertion failure: mRawPtr != 0 (You can't dereference a NULL nsCOMPtr with operator->().), at /home/worker/workspace/build/src/obj-firefox/dist/include/nsCOMPtr.h:747
 [task 2016-09-22T16:58:30.115302Z] 16:58:30     INFO -  #01: mozilla::dom::PresentationBuilderChild::RecvOnIceCandidate [dom/presentation/ipc/PresentationBuilderChild.cpp:98]
 [task 2016-09-22T16:58:30.118164Z] 16:58:30     INFO -  #02: mozilla::dom::PPresentationBuilderChild::OnMessageReceived [obj-firefox/ipc/ipdl/PPresentationBuilderChild.cpp:350]
 [task 2016-09-22T16:58:30.120096Z] 16:58:30     INFO -  #03: mozilla::ipc::MessageChannel::DispatchAsyncMessage [ipc/glue/MessageChannel.h:623]
 [task 2016-09-22T16:58:30.121966Z] 16:58:30     INFO -  #04: mozilla::ipc::MessageChannel::DispatchMessage [ipc/glue/MessageChannel.cpp:1602]
 [task 2016-09-22T16:58:30.123655Z] 16:58:30     INFO -  #05: mozilla::ipc::MessageChannel::OnMaybeDequeueOne [ipc/glue/MessageChannel.cpp:1556]
 [task 2016-09-22T16:58:30.126304Z] 16:58:30     INFO -  #06: mozilla::detail::RunnableMethodImpl<bool (mozilla::ipc::MessageChannel::*)(), false, true>::Run [xpcom/glue/nsThreadUtils.h:767]
 [task 2016-09-22T16:58:30.127959Z] 16:58:30     INFO -  #07: mozilla::ipc::MessageChannel::DequeueTask::Run [ipc/glue/MessageChannel.h:567]
 [task 2016-09-22T16:58:30.130005Z] 16:58:30     INFO -  #08: nsThread::ProcessNextEvent [xpcom/threads/nsThread.cpp:1059]
 [task 2016-09-22T16:58:30.131910Z] 16:58:30     INFO -  #09: NS_ProcessNextEvent [xpcom/glue/nsThreadUtils.cpp:290]
 [task 2016-09-22T16:58:30.133744Z] 16:58:30     INFO -  #10: mozilla::ipc::MessagePump::Run [ipc/glue/MessagePump.cpp:124]
 [task 2016-09-22T16:58:30.135562Z] 16:58:30     INFO -  #11: MessageLoop::RunInternal [ipc/chromium/src/base/message_loop.cc:233]
 [task 2016-09-22T16:58:30.137168Z] 16:58:30     INFO -  #12: MessageLoop::Run [ipc/chromium/src/base/message_loop.cc:490]
 [task 2016-09-22T16:58:30.139079Z] 16:58:30     INFO -  #13: nsBaseAppShell::Run [widget/nsBaseAppShell.cpp:158]
 [task 2016-09-22T16:58:30.140932Z] 16:58:30     INFO -  #14: XRE_RunAppShell [toolkit/xre/nsEmbedFunctions.cpp:875]
 [task 2016-09-22T16:58:30.142721Z] 16:58:30     INFO -  #15: mozilla::ipc::MessagePumpForChildProcess::Run [ipc/glue/MessagePump.cpp:269]
 [task 2016-09-22T16:58:30.144365Z] 16:58:30     INFO -  #16: MessageLoop::RunInternal [ipc/chromium/src/base/message_loop.cc:233]
 [task 2016-09-22T16:58:30.146341Z] 16:58:30     INFO -  #17: MessageLoop::Run [ipc/chromium/src/base/message_loop.cc:490]
 [task 2016-09-22T16:58:30.148259Z] 16:58:30     INFO -  #18: XRE_InitChildProcess [toolkit/xre/nsEmbedFunctions.cpp:709]
[task 2016-09-22T16:58:30.251183Z] 16:58:30 INFO - #19: content_process_main [ipc/contentproc/plugin-container.cpp:200]
Attached patch Add a null check (obsolete) — Splinter Review
This patch should be able to fix the crash.

@sc, could you take a look? Thanks.
Assignee: nobody → kechang
Attachment #8794038 - Flags: feedback?(schien)
Comment on attachment 8794038 [details] [diff] [review]
Add a null check

Directly ask for a review, since it's a tiny change.
Attachment #8794038 - Flags: feedback?(schien) → review?(bugs)
Attachment #8794038 - Flags: review?(bugs) → review+
Keywords: checkin-needed
https://hg.mozilla.org/mozilla-central/rev/71f595be2820
Status: NEW → RESOLVED
Closed: 3 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla52
Please request Aurora/Beta approval on this when you get a chance.
Flags: needinfo?(kechang)
It will fix the null-dereference crash, but will still cause the child process to be killed because the method returns false. I'm not convinced that this will be an improvement for the intermittent failures...
(In reply to Josh Matthews [:jdm] from comment #18)
> It will fix the null-dereference crash, but will still cause the child
> process to be killed because the method returns false. I'm not convinced
> that this will be an improvement for the intermittent failures...

No, it only returns false when |mBuilder| is not null and |OnIceCandidate| fails. I think this patch should be able to fix the intermittent failure.
Flags: needinfo?(kechang)
Whoops, you're right. My mistake!
Comment on attachment 8794138 [details] [diff] [review]
Null check for mBuilder, r=smaug

Approval Request Comment
[Feature/regressing bug #]: bug 1228474
[User impact if declined]: It causes intermittent failures.
[Describe test coverage new/current, TreeHerder]: Covered by existing unit tests.
[Risks and why]: Low, since this is just a tiny change.
[String/UUID change made/needed]: N/A
Attachment #8794138 - Flags: approval-mozilla-beta?
Attachment #8794138 - Flags: approval-mozilla-aurora?
Comment on attachment 8794138 [details] [diff] [review]
Null check for mBuilder, r=smaug

The patch looks simple enough, Aurora51+, Beta50+
Attachment #8794138 - Flags: approval-mozilla-beta?
Attachment #8794138 - Flags: approval-mozilla-beta+
Attachment #8794138 - Flags: approval-mozilla-aurora?
Attachment #8794138 - Flags: approval-mozilla-aurora+
Blocks: 1307041
Component: DOM → DOM: Core & HTML
You need to log in before you can comment on or make changes to this bug.