Filed by: philringnalda [at] gmail.com https://treeherder.mozilla.org/logviewer.html#?job_id=105630854&repo=try https://archive.mozilla.org/pub/firefox/try-builds/philringnalda@gmail.com-82073c8c7c40563e8830e286ab041cbb3000f1ee/try-macosx64-debug/try_yosemite_r7-debug_test-mochitest-media-e10s-bm106-tests1-macosx-build2920.txt.gz

Mass change P3->P4 to align with new Mozilla triage process.

On December 1st, this failure started reoccurring. In the last 7 days, there are 44 failures. Main platform: android-6-0-armv8-api16 and some of the failures are on android-4-4-armv7-api16. Most of the failures are on opt, a few of them on debug. An example of a recent log file: https://treeherder.mozilla.org/logviewer.html#?repo=mozilla-inbound&job_id=150290370&lineNumber=901 And the relevant part of the log: 2017-12-07 04:57:58,458 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests TEST-INFO | started process screentopng 2017-12-07 04:57:58,458 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests Unable to init server: Could not connect: Connection refused 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests (screentopng:31369): Gdk-WARNING **: cannot open display: 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests TEST-INFO | screentopng: exit 1 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests org.mozilla.fennec_aurora still alive after SIGABRT: waiting... 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests org.mozilla.fennec_aurora still alive after SIGABRT: waiting... 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests org.mozilla.fennec_aurora still alive after SIGABRT: waiting... 2017-12-07 04:57:58,460 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests TEST-UNEXPECTED-FAIL | dom/media/tests/mochitest/test_selftest.html | application timed out after 330 seconds with no output 2017-12-07 04:57:58,460 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests INFO | automation.py | Application ran for: 0:17:38.033886 2017-12-07 04:57:58,460 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests INFO | zombiecheck | Reading PID log: /tmp/tmptGkDJHpidlog 2017-12-07 04:57:58,460 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests /data/tombstones does not exist; tombstone check skipped 2017-12-07 04:57:58,460 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests Stopping web server 2017-12-07 04:57:58,460 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests Stopping web socket server 2017-12-07 04:57:58,461 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests Stopping ssltunnel 2017-12-07 04:57:58,461 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests websocket/process bridge listening on port 48901 2017-12-07 04:57:58,461 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests Stopping websocket/process bridge 2017-12-07 04:57:58,461 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests leakcheck | refcount logging is off, so leaks can't be detected! 2017-12-07 04:57:58,461 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests runtests.py | Running tests: end. :drno As you are the triage owner of this component, could you please take a look at this? Thank you!

To me it looks like the test framework is aeehh confused...: 2017-12-07 04:57:58,456 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests 367 INFO TEST-START | dom/media/tests/mochitest/test_selftest.html 2017-12-07 04:57:58,457 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests 368 INFO TEST-OK | dom/media/tests/mochitest/test_selftest.html | took 728ms 2017-12-07 04:57:58,457 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests 369 INFO TEST-START | Shutdown 2017-12-07 04:57:58,457 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests 370 INFO Passed: 33497 2017-12-07 04:57:58,457 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests 371 INFO Failed: 1 2017-12-07 04:57:58,457 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests 372 INFO Todo: 1388 2017-12-07 04:57:58,458 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests 373 INFO Mode: non-e10s 2017-12-07 04:57:58,458 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests 374 INFO Slowest: 328617ms - /tests/dom/media/tests/mochitest/test_peerConnection_transceivers.html 2017-12-07 04:57:58,458 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests 375 INFO SimpleTest FINISHED 2017-12-07 04:57:58,458 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests Browser unexpectedly found running. Killing... 2017-12-07 04:57:58,458 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests TEST-INFO | started process screentopng 2017-12-07 04:57:58,458 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests Unable to init server: Could not connect: Connection refused 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests (screentopng:31369): Gdk-WARNING **: cannot open display: 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests TEST-INFO | screentopng: exit 1 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests org.mozilla.fennec_aurora still alive after SIGABRT: waiting... 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests org.mozilla.fennec_aurora still alive after SIGABRT: waiting... 2017-12-07 04:57:58,459 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests org.mozilla.fennec_aurora still alive after SIGABRT: waiting... 2017-12-07 04:57:58,460 nexus-6p-05 INFO UnitTestJob mozilla-inbound 20171206200232 opt api-23 android-api-16 mochitest-dom-media-tests TEST-UNEXPECTED-FAIL | dom/media/tests/mochitest/test_selftest.html | application timed out after 330 seconds with no output Right there it says test_selftest.html finished successfully and took 728ms. But the test which failed in that run is the new transceivers test. Which would also explain why the failure rate has spiked recently. Byron: could you please have a look? Maybe we need to make the new transceivers test smaller because Android is to slow (again)?

I don't think the problem is that the transceivers test is too slow. I'm pretty sure there's a small flaw in the timing of onnegotiationneeded. More on that soon over in bug 1421775.