Filed by: rvandermeulen [at] mozilla.com https://treeherder.mozilla.org/logviewer.html#?job_id=3734995&repo=mozilla-aurora https://archive.mozilla.org/pub/firefox/tinderbox-builds/mozilla-aurora-linux-pgo/1475547840/mozilla-aurora_ubuntu32_vm_test_pgo-cppunit-bm05-tests1-linux32-build2.txt.gz [ RUN ] MediaPipelineTest.TestAudioSendMux /builds/slave/m-aurora-lx-000000000000000000/build/src/media/webrtc/signaling/test/mediapipeline_unittest.cpp:453: Failure Value of: p2_.GetAudioRtpCountReceived() Actual: 233 Expected: p1_.GetAudioRtpCountSent() Which is: 970 [ FAILED ] MediaPipelineTest.TestAudioSendMux (11146 ms)

Not sure this is a great check as written.

We stopped building this test on trunk via bug 1322707. However, 52 continues to be affected (and is destined to be our next ESR).

If disabling the entire test isn't desirable, can we maybe skip just the one problematic subtest?

Dan, can you look at this? thanks

Try run here: https://treeherder.mozilla.org/#/jobs?repo=try&revision=44a5cb4d2f19833bbf70f0e044259afac8bb4858

Comment on attachment 8831749 [details] [diff] [review] Disable TestAudioSendMux Approval Request Comment [Feature/Bug causing the regression]: Not sure, we're working around an intermittent failure. [User impact if declined]: None [Is this code covered by automated tests?]: Yes. [Has the fix been verified in Nightly?]: This test is completely disabled in aurora and nightly. [Needs manual test from QE? If yes, steps to reproduce]: No. [List of other uplifts needed for the feature/fix]: None. [Is the change risky?]: No. [Why is the change risky/not risky?]: We're disabling an intermittent test. There is a slight risk that we could introduce a regression that this test would have caught, but since this test is disabled on aurora and currently disabled on nightly while it is being rewrittent, I think we're ok with that risk. [String changes made/needed]: None

Comment on attachment 8831749 [details] [diff] [review] Disable TestAudioSendMux test-only changes don't need approval, thanks Dan!

Bah, it just moved to TestAudioSendNoMux. Guess that's not surprising given bug 1317044, in retrospect. https://treeherder.mozilla.org/logviewer.html#?job_id=73368952&repo=mozilla-beta

If we see any more problems after this, I'll just disable the test completely.

Bug resolution tracks status on trunk.

New failure: https://treeherder.mozilla.org/logviewer.html#?job_id=75172990&repo=mozilla-beta&lineNumber=2625

Just disable the test completely. Ryan said he would be willing to take the gtest version of this test once it is ready if it ends up being less prone to intermittent failures.