Filed by: bclary [at] mozilla.com https://treeherder.mozilla.org/logviewer.html#?job_id=210227205&repo=autoland https://queue.taskcluster.net/v1/task/SeJG07XUTKqcbDl_nZq0xg/runs/0/artifacts/public/logs/live_backing.log

From https://treeherder.mozilla.org/logviewer.html#?job_id=210192676&repo=autoland&lineNumber=5166 [task 2018-11-06T23:06:28.078Z] 23:06:28 INFO - GECKO(1777) | Assertion failure: aError != NS_ERROR_DOM_MEDIA_END_OF_STREAM, at /builds/worker/workspace/build/src/dom/media/MediaDecoderStateMachine.cpp:932 [task 2018-11-06T23:07:53.648Z] 23:07:53 INFO - GECKO(1777) | #01: mozilla::MozPromise<RefPtr<mozilla::AudioData>, mozilla::MediaResult, true>::ThenValue<mozilla::MediaDecoderStateMachine::LoopingDecodingState::HandleEndOfAudio()::{lambda()#1}::operator()() const::{lambda(RefPtr<mozilla::AudioData>)#1}, mozilla::MediaDecoderStateMachine::LoopingDecodingState::HandleEndOfAudio()::{lambda()#1}::operator()() const::{lambda(mozilla::MediaResult const&)#1}>::DoResolveOrRejectInternal(mozilla::MozPromise<RefPtr<mozilla::AudioData>, mozilla::MediaResult, true>::ResolveOrRejectValue&) [xpcom/threads/MozPromise.h:791] [task 2018-11-06T23:07:53.649Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.650Z] 23:07:53 INFO - GECKO(1777) | #02: mozilla::MozPromise<RefPtr<mozilla::AudioData>, mozilla::MediaResult, true>::ThenValueBase::ResolveOrRejectRunnable::Run() [mfbt/RefPtr.h:69] [task 2018-11-06T23:07:53.650Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.651Z] 23:07:53 INFO - GECKO(1777) | #03: mozilla::AutoTaskDispatcher::TaskGroupRunnable::Run() [mfbt/ThreadLocal.h:177] [task 2018-11-06T23:07:53.652Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.653Z] 23:07:53 INFO - GECKO(1777) | #04: mozilla::TaskQueue::Runner::Run() [xpcom/threads/TaskQueue.cpp:236] [task 2018-11-06T23:07:53.654Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.655Z] 23:07:53 INFO - GECKO(1777) | #05: nsThreadPool::Run() [xpcom/threads/nsThreadPool.cpp:0] [task 2018-11-06T23:07:53.656Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.658Z] 23:07:53 INFO - GECKO(1777) | #06: non-virtual thunk to nsThreadPool::Run() [xpcom/threads/nsThreadPool.cpp:0] [task 2018-11-06T23:07:53.660Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.661Z] 23:07:53 INFO - GECKO(1777) | #07: nsThread::ProcessNextEvent(bool, bool*) [xpcom/threads/nsThread.cpp:1233] [task 2018-11-06T23:07:53.662Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.663Z] 23:07:53 INFO - GECKO(1777) | #08: NS_ProcessNextEvent(nsIThread*, bool) [xpcom/threads/nsThreadUtils.cpp:530] [task 2018-11-06T23:07:53.665Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.667Z] 23:07:53 INFO - GECKO(1777) | #09: mozilla::ipc::MessagePumpForNonMainThreads::Run(base::MessagePump::Delegate*) [ipc/glue/MessagePump.cpp:334] [task 2018-11-06T23:07:53.668Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.670Z] 23:07:53 INFO - GECKO(1777) | #10: MessageLoop::Run() [ipc/chromium/src/base/message_loop.cc:598] [task 2018-11-06T23:07:53.672Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.674Z] 23:07:53 INFO - GECKO(1777) | #11: nsThread::ThreadFunc(void*) [xpcom/threads/nsThread.cpp:507] [task 2018-11-06T23:07:53.675Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.976Z] 23:07:53 INFO - GECKO(1777) | #12: _pt_root [nsprpub/pr/src/pthreads/ptthread.c:204] [task 2018-11-06T23:07:53.978Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.980Z] 23:07:53 INFO - GECKO(1777) | #13: libpthread.so.0 + 0x76ba [task 2018-11-06T23:07:53.984Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.987Z] 23:07:53 INFO - GECKO(1777) | #14: libc.so.6 + 0x10741d [task 2018-11-06T23:07:53.989Z] 23:07:53 INFO - [task 2018-11-06T23:07:53.991Z] 23:07:53 INFO - GECKO(1777) | #15: ??? (???:???)

I've started triggering this on linux64 debug M-sw-e10s with some patches recently, [1]. It's not the patches themselves though, as I'm getting this on linux64 debug non-opt locally too, with the command: > ./mach test dom/media/test/test_looping_eventsOrder.html --setpref=dom.serviceWorkers.parent_intercept=true --verify It doesn't happen on the first iteration but typically the second. Alastor, dom/media/test/test_looping_eventsOrder.html is your test. Can you take a look? [1] https://treeherder.mozilla.org/#/jobs?repo=try&tier=1%2C2%2C3&selectedJob=217059107&revision=e6c4cfa97edcd9cdb18c69275fbbb987a20b470d

I believe this is the same issue as bug1508501, but now I still didn't figure out the root cause yet.

FWIW you can use my patches as linked in comment 4 to repro this locally. This is currently blocking bug 1512958 which is a 65 regression.

OK, I'll continue to investigate this bug today.

If we can't get any sample anymore, the resource might have been closed, so we should change state to 'completedState'.

Please request Beta approval on this when you get a chance.

Comment on attachment 9032569 [details] Bug 1505250 - change state to 'completedState' if we can't get any sample anymore. [Beta/Release Uplift Approval Request] Feature/Bug causing the regression: Bug 1499903 User impact if declined: can't fix the intermittent test case fail. Is this code covered by automated tests?: No Has the fix been verified in Nightly?: No Needs manual test from QE?: No If yes, steps to reproduce: no List of other uplifts needed: no Risk to taking this patch: Low Why is the change risky/not risky? (and alternatives if risky): This patch is used to remove the assertion and changes the internal state transition which is trivial. String changes made/needed: no

Comment on attachment 9032569 [details] Bug 1505250 - change state to 'completedState' if we can't get any sample anymore. [Triage Comment] Needed for the uplift of bug 1512958. Approved for 65.0b8.