Closed Bug 1399601 Opened 7 years ago Closed 6 years ago

Intermittent test_refresh_firefox.py TestFirefoxRefresh.testReset | application crashed [@ RunWatchdog]

Categories

(Firefox :: Migration, defect, P3)

Product:
Firefox
Component:
Migration
Type:
defect

Tracking

()

Status:
RESOLVED DUPLICATE of bug 1425323
Tracking Flags:
Tracking Status
firefox56 --- unaffected
firefox57 + fix-optional
firefox58 --- wontfix
firefox59 --- ?

People

(Reporter: intermittent-bug-filer, Unassigned)

References

(Blocks 1 open bug)

Details

(4 keywords)

Crash Data

Filed by: hskupin [at] gmail.com

https://treeherder.mozilla.org/logviewer.html#?job_id=130580283&repo=try

https://queue.taskcluster.net/v1/task/HXxEKU5oSySyl0JxLAHB-A/runs/0/artifacts/public/logs/live_backing.log

[task 2017-09-13T08:51:32.790132Z] 08:51:32     INFO - Crash reason:  SIGSEGV
[task 2017-09-13T08:51:32.790853Z] 08:51:32     INFO - Crash address: 0x0
[task 2017-09-13T08:51:32.791383Z] 08:51:32     INFO - Process uptime: not available
[task 2017-09-13T08:51:32.791673Z] 08:51:32     INFO - 
[task 2017-09-13T08:51:32.792249Z] 08:51:32     INFO - Thread 27 (crashed)
[task 2017-09-13T08:51:32.792934Z] 08:51:32     INFO -  0  libxul.so!RunWatchdog [nsTerminator.cpp:a73cc4e08bf5 : 160 + 0x0]
[task 2017-09-13T08:51:32.793289Z] 08:51:32     INFO -     rax = 0x0000000000000000   rdx = 0x0000000000000000
[task 2017-09-13T08:51:32.793860Z] 08:51:32     INFO -     rcx = 0x00007ff28e0692ad   rbx = 0x00007ff24f2674e8
[task 2017-09-13T08:51:32.796891Z] 08:51:32     INFO -     rsi = 0x00007ff28e338770   rdi = 0x00007ff28e337540
[task 2017-09-13T08:51:32.796959Z] 08:51:32     INFO -     rbp = 0x00007ff24a465ec0   rsp = 0x00007ff24a465eb0
[task 2017-09-13T08:51:32.797064Z] 08:51:32     INFO -      r8 = 0x00007ff28e338770    r9 = 0x00007ff24a466700
[task 2017-09-13T08:51:32.797125Z] 08:51:32     INFO -     r10 = 0x0000000000000012   r11 = 0x0000000000000000
[task 2017-09-13T08:51:32.797232Z] 08:51:32     INFO -     r12 = 0x000000000000003f   r13 = 0x00000000000014c2
[task 2017-09-13T08:51:32.798058Z] 08:51:32     INFO -     r14 = 0x00007ff24a466700   r15 = 0x00007ff24a466670
[task 2017-09-13T08:51:32.798109Z] 08:51:32     INFO -     rip = 0x00007ff27f753590
[task 2017-09-13T08:51:32.798202Z] 08:51:32     INFO -     Found by: given as instruction pointer in context
[task 2017-09-13T08:51:32.798258Z] 08:51:32     INFO -  1  libnspr4.so!_pt_root [ptthread.c:a73cc4e08bf5 : 216 + 0x7]
[task 2017-09-13T08:51:32.798312Z] 08:51:32     INFO -     rbx = 0x00007ff250094800   rbp = 0x00007ff24a465f10
[task 2017-09-13T08:51:32.799102Z] 08:51:32     INFO -     rsp = 0x00007ff24a465ed0   r12 = 0x0000000000000001
[task 2017-09-13T08:51:32.799891Z] 08:51:32     INFO -     r13 = 0x00000000000014c2   r14 = 0x00007ff24a466700
[task 2017-09-13T08:51:32.800712Z] 08:51:32     INFO -     r15 = 0x00007ff24a466670   rip = 0x00007ff28d8d16e6
[task 2017-09-13T08:51:32.801419Z] 08:51:32     INFO -     Found by: call frame info
[task 2017-09-13T08:51:32.802104Z] 08:51:32     INFO -  2  libpthread-2.23.so + 0x76ba
[task 2017-09-13T08:51:32.802868Z] 08:51:32     INFO -     rbx = 0x0000000000000000   rbp = 0x0000000000000000
[task 2017-09-13T08:51:32.803407Z] 08:51:32     INFO -     rsp = 0x00007ff24a465f20   r12 = 0x0000000000000000
[task 2017-09-13T08:51:32.804057Z] 08:51:32     INFO -     r13 = 0x00007ffd1916420f   r14 = 0x00007ff24a4669c0
[task 2017-09-13T08:51:32.804703Z] 08:51:32     INFO -     r15 = 0x00007ffd191642a0   rip = 0x00007ff28eff06ba
[task 2017-09-13T08:51:32.805379Z] 08:51:32     INFO -     Found by: call frame info
[task 2017-09-13T08:51:32.806027Z] 08:51:32     INFO -  3  libc-2.23.so + 0x1073dd
[task 2017-09-13T08:51:32.806717Z] 08:51:32     INFO -     rsp = 0x00007ff24a465fc0   rip = 0x00007ff28e0793dd
[task 2017-09-13T08:51:32.806915Z] 08:51:32     INFO -     Found by: stack scanning

Looks like there is something blocking us from shutdown, and Firefox gets killed.
I see the following in the log:

[task 2017-09-15T08:17:53.934Z] 08:17:53     INFO -  1505463473929	Marionette	DEBUG	Received observer notification "xpcom-shutdown"
[task 2017-09-15T08:17:53.955Z] 08:17:53     INFO -  [Parent 4712, Main Thread] WARNING: 'NS_FAILED(rr->RetargetDeliveryTo(sts))', file /builds/worker/workspace/build/src/dom/fetch/FetchDriver.cpp, line 661
[task 2017-09-15T08:17:53.956Z] 08:17:53     INFO -  [Parent 4712, Main Thread] WARNING: 'NS_FAILED(rv)', file /builds/worker/workspace/build/src/dom/fetch/FetchConsumer.cpp, line 516
[task 2017-09-15T08:17:53.957Z] 08:17:53     INFO -  [Parent 4712, Main Thread] WARNING: Retargeting failed: file /builds/worker/workspace/build/src/dom/fetch/FetchConsumer.cpp, line 517
[task 2017-09-15T08:17:54.038Z] 08:17:54     INFO -  [Parent 4712, Main Thread] WARNING: A runnable was posted to a worker that is already shutting down!: file /builds/worker/workspace/build/src/dom/workers/WorkerPrivate.cpp, line 2958
[task 2017-09-15T08:17:54.040Z] 08:17:54     INFO -  [Parent 4712, Main Thread] WARNING: Could not dispatch ConsumeBodyRunnable: file /builds/worker/workspace/build/src/dom/fetch/FetchConsumer.cpp, line 224
[task 2017-09-15T08:18:21.499Z] 08:18:21     INFO -  JavaScript error: resource://gre/modules/osfile/osfile_async_front.jsm, line 410: Error: OS.File has been shut down. Rejecting post to stat
[task 2017-09-15T08:18:57.334Z] 08:18:57     INFO -  Hit MOZ_CRASH(Shutdown too long, probably frozen, causing a crash.) at /builds/worker/workspace/build/src/toolkit/components/terminator/nsTerminator.cpp:160
[task 2017-09-15T08:18:57.336Z] 08:18:57     INFO -  #01: ???[/builds/worker/workspace/build/application/firefox/libnspr4.so +0x286e6]
[task 2017-09-15T08:18:57.337Z] 08:18:57     INFO -  #02: ???[/lib/x86_64-linux-gnu/libpthread.so.0 +0x76ba]
[task 2017-09-15T08:18:57.338Z] 08:18:57     INFO -  #03: clone[/lib/x86_64-linux-gnu/libc.so.6 +0x1073dd]
[task 2017-09-15T08:18:57.340Z] 08:18:57     INFO -  #04: ??? (???:???)

Andrea, could this be one more issue with FetchConsumer and which is causing a shutdown hang?
Flags: needinfo?(amarchesini)
Those hangs started on September 13th on autoland if it helps.
status-firefox56: --- → unaffected
status-firefox57: --- → affected
Keywords: regression, regressionwindow-wanted
This seems more related to worker shutting down. I'm planning to change this part soon.
Flags: needinfo?(amarchesini)
Hey baku, this is currently a P5 but triage managers felt it might be more important. Can you update the priority and let us know when you expect to get to this?
Flags: needinfo?(amarchesini)
I would say P3. overholt?
Flags: needinfo?(amarchesini) → needinfo?(overholt)
Priority: P5 → P3
Since it's something baku's likely to work on in the next few months, let's go with P2. (The component here seems odd)
Flags: needinfo?(overholt)
Priority: P3 → P2
status-firefox57: affected → fix-optional
OS: Unspecified → All
Priority: P2 → P3
Hardware: Unspecified → All
Version: unspecified → Trunk
This bug looks like a dupe of bug 1425323 now. Here a link to the stack:

https://treeherder.mozilla.org/logviewer.html#?repo=mozilla-inbound&job_id=151425042&lineNumber=47965

Andrea, I assume sometimes in December the quota manager got moved from the main thread to it's own thread?
Flags: needinfo?(amarchesini)
> Andrea, I assume sometimes in December the quota manager got moved from the
> main thread to it's own thread?

Yes, but the shutdown notifications are received on the main-thread. I agree. it's a dup of bug 1425323.
Flags: needinfo?(amarchesini)
Status: NEW → RESOLVED
Closed: 6 years ago
Resolution: --- → DUPLICATE
You need to log in before you can comment on or make changes to this bug.