Closed Bug 1126052 Opened 7 years ago Closed 7 years ago

Turn test_SeekTwice_mp4.html back on


(Core :: Audio/Video, defect, P2)




Tracking Status
firefox38 --- fixed


(Reporter: bholley, Assigned: bholley)


(Blocks 1 open bug)



(1 file)

The tests in bug 1121692 went intermittently orange, and got disabled. These were the first and only in-tree tests for the MSE code that we're planning to ship on youtube. I think figuring out why they're timing out is pretty important.
Blocks: 1120030
So, one of the problems here is that we don't have an h264 story on winxp (duh), and don't have a way to mark the test as skip-if in the winxp case. I'll file a bug about that.

However, this did go orange at least once on win7 with my logging machinery. The log is here:

I've pruned the log of unrelated stuff locally (especially other decoders that get shutdown off a timer gc), and will upload that.
Attached file pruned log
This *should* work:
skip-if = (os == "win" && os_version == "5.1")
Depends on: 1126465
Priority: -- → P2
bug 1126465 fixes one of the failure modes, but as [1] shows, not all (I'd seen this second failure mode before on the original push too).

Pushing with more logging to see if I can suss out the second failure:

Comment 5 wasn't able to reproduce any failure in the test with logging (and that was a week ago, which is an eternity in the MSE codebase).

I just did another set of pushes with the same results. The first just turns the test back on, the second does so and activates the logging:

So far those look pretty green. Ryan, are you comfortable with me turning the test back on, or are there any other steps I should take?
Flags: needinfo?(ryanvm)
Are we going to leave the logging on when it's re-enabled? If so, go for it. Try doesn't lie (usually) :)
Flags: needinfo?(ryanvm)
Pushed per IRC conversation with RyanVM:
Closed: 7 years ago
Flags: in-testsuite+
Resolution: --- → FIXED
Target Milestone: --- → mozilla38
You need to log in before you can comment on or make changes to this bug.