+++ This bug was initially created as a clone of Bug #690989 +++ This regressed some time on Thursday. Currently bisecting it down on Try. Note that there is now an uncaught exception as well. I do see some XPConnect patches in the regression range. https://tbpl.mozilla.org/php/getParsedLog.php?id=26902442&tree=Mozilla-Inbound Rev4 MacOSX Lion 10.7 mozilla-inbound debug test mochitest-1 on 2013-08-22 18:07:52 PDT for push b4b8ecfae683 slave: talos-r4-lion-060 18:34:49 INFO - 144143 INFO TEST-PASS | /tests/content/html/content/test/test_fullscreen-api.html | [rollback] Should have left full-screen entirely 18:34:49 INFO - [Parent 1263] WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result 0x80040111: file ../../../../content/events/src/nsContentEventHandler.cpp, line 94 18:34:49 INFO - System JS : ERROR (null):0 18:34:49 INFO - uncaught exception: 2147746065 18:34:49 INFO - 144144 INFO TEST-PASS | /tests/content/html/content/test/test_fullscreen-api.html | [rollback] FSE should be e('fse') 18:34:49 INFO - 144145 ERROR TEST-UNEXPECTED-FAIL | /tests/content/html/content/test/test_fullscreen-api.html | [rollback] FSE should be e('fse-inner') - got null, expected [object HTMLDivElement] 18:34:49 INFO - 144146 ERROR TEST-UNEXPECTED-FAIL | /tests/content/html/content/test/test_fullscreen-api.html | uncaught exception - TypeError: opener.finish is not a function at http://mochi.test:8888/tests/content/html/content/test/file_fullscreen-rollback.html:49 18:34:49 INFO - JavaScript error: http://mochi.test:8888/tests/content/html/content/test/file_fullscreen-rollback.html, line 49: opener.finish is not a function etc... 18:39:57 INFO - 144147 ERROR TEST-UNEXPECTED-FAIL | /tests/content/html/content/test/test_fullscreen-api.html | Test timed out.

Try bisection points to bug 903891 as the culprit.

(In reply to Ryan VanderMeulen [:RyanVM UTC-4] from comment #1) > Try bisection points to bug 903891 as the culprit. Maybe not. A backout push to Try was still reproducing. Running more tests on the earlier bisection runs.

I don't understand what I'm seeing at this point. Try run for inbound rev 581251a392cd - NONE https://tbpl.mozilla.org/?tree=Try&rev=1408d94e7a1c Try run for inbound rev 89e249527629 - FAILS https://tbpl.mozilla.org/?tree=Try&rev=e187c53fb515 Which points to bug 896087 as the culprit. https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=581251a392cd&tochange=89e249527629 How in the world does that make sense? But as you can see, I can't get this to happen on 581251a392cd no matter how much I retrigger, and it occurred on 89e249527629 without too much difficulty. Will attempt a Try run with bug 896087 backed out next.

Backing out bug 896087 didn't help. I'm at a loss. There are pushes from Thursday that I cannot reproduce this on no matter how many times I try, but backing out anything in between "known good" and "known bad" doesn't seem to make it go away. Really hoping someone else can make more sense of all this.

Is this at all reproducible locally?

Test disabled on OS X 10.7 and 10.8 for bug 900453 comment 19: remote: https://hg.mozilla.org/integration/mozilla-inbound/rev/bbd8b29a4261

Tweaking summary to avoid false positive with another unrelated failure mode.

Bulk assigning P3 to all open intermittent bugs without a priority set in Firefox components per bug 1298978.

We have enabled this test on all desktop platforms since long ago, and there aren't any new report. Closing as WORKSFORME.