Closed Bug 1840914 Opened 2 years ago Closed 2 years ago

High frequency [tier 2] wmfme dom/media/test/< test> | Test timed out!

Categories

(Core :: Audio/Video: Playback, defect, P5)

defect

Tracking

()

RESOLVED FIXED
117 Branch
Tracking Status
firefox-esr102 --- unaffected
firefox-esr115 --- unaffected
firefox114 --- unaffected
firefox115 --- unaffected
firefox116 --- wontfix
firefox117 --- fixed

People

(Reporter: intermittent-bug-filer, Assigned: alwu)

References

(Regression)

Details

(Keywords: intermittent-failure, intermittent-testcase, regression, Whiteboard: [retriggered])

Attachments

(2 files)

Filed by: sstanca [at] mozilla.com
Parsed log: https://treeherder.mozilla.org/logviewer?job_id=420956555&repo=mozilla-central
Full log: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/a8C1eWu0TCytxEh1hMNQTA/runs/0/artifacts/public/logs/live_backing.log


[task 2023-06-28T17:36:55.412Z] 17:36:55     INFO - TEST-PASS | dom/media/test/test_buffered.html | [finished bug482461-theora.ogv-12 t=1.587] Length of array should match number of running tests 
[task 2023-06-28T17:36:55.412Z] 17:36:55     INFO - Buffered messages finished
[task 2023-06-28T17:36:55.414Z] 17:36:55     INFO - TEST-UNEXPECTED-FAIL | dom/media/test/test_buffered.html | Test timed out! 
[task 2023-06-28T17:36:55.414Z] 17:36:55     INFO -     SimpleTest.ok@SimpleTest/SimpleTest.js:424:16
[task 2023-06-28T17:36:55.414Z] 17:36:55     INFO -     onTimeout@dom/media/test/manifest.js:2380:9
[task 2023-06-28T17:36:55.414Z] 17:36:55     INFO -     setTimeout handler*MediaTestManager/this.started@dom/media/test/manifest.js:2387:57
[task 2023-06-28T17:36:55.414Z] 17:36:55     INFO -     startTest@dom/media/test/test_buffered.html:106:11
[task 2023-06-28T17:36:55.414Z] 17:36:55     INFO -     MediaTestManager/this.nextTest@dom/media/test/manifest.js:2461:12
[task 2023-06-28T17:36:55.414Z] 17:36:55     INFO -     MediaTestManager/this.finished@dom/media/test/manifest.js:2433:12
[task 2023-06-28T17:36:55.414Z] 17:36:55     INFO -     testBuffered@dom/media/test/test_buffered.html:69:11
[task 2023-06-28T17:36:55.414Z] 17:36:55     INFO -     EventListener.handleEvent*onfetched@dom/media/test/test_buffered.html:102:7
[task 2023-06-28T17:36:55.414Z] 17:36:55     INFO -     loaded@dom/media/test/test_buffered.html:82:23
[task 2023-06-28T17:36:55.415Z] 17:36:55     INFO -     EventListener.handleEvent*fetch@dom/media/test/test_buffered.html:88:7
[task 2023-06-28T17:36:55.415Z] 17:36:55     INFO -     startTest@dom/media/test/test_buffered.html:107:8
[task 2023-06-28T17:36:55.415Z] 17:36:55     INFO -     MediaTestManager/this.nextTest@dom/media/test/manifest.js:2461:12
[task 2023-06-28T17:36:55.415Z] 17:36:55     INFO -     MediaTestManager/this.finished@dom/media/test/manifest.js:2433:12
[task 2023-06-28T17:36:55.415Z] 17:36:55     INFO -     testBuffered@dom/media/test/test_buffered.html:69:11
[task 2023-06-28T17:36:55.415Z] 17:36:55     INFO -     EventListener.handleEvent*onfetched@dom/media/test/test_buffered.html:102:7
[task 2023-06-28T17:36:55.415Z] 17:36:55     INFO -     loaded@dom/media/test/test_buffered.html:82:23
[task 2023-06-28T17:36:55.415Z] 17:36:55     INFO -     EventListener.handleEvent*fetch@dom/media/test/test_buffered.html:88:7
[task 2023-06-28T17:36:55.415Z] 17:36:55     INFO -     startTest@dom/media/test/test_buffered.html:107:8
[task 2023-06-28T17:36:55.416Z] 17:36:55     INFO -     MediaTestManager/this.nextTest@dom/media/test/manifest.js:2461:12
[task 2023-06-28T17:36:55.416Z] 17:36:55     INFO -     MediaTestManager/this.finished@dom/media/test/manifest.js:2433:12
[task 2023-06-28T17:36:55.416Z] 17:36:55     INFO -     testBuffered@dom/media/test/test_buffered.html:69:11
[task 2023-06-28T17:36:55.416Z] 17:36:55     INFO -     EventListener.handleEvent*onfetched@dom/media/test/test_buffered.html:102:7
[task 2023-06-28T17:36:55.416Z] 17:36:55     INFO -     loaded@dom/media/test/test_buffered.html:82:23
[task 2023-06-28T17:36:55.416Z] 17:36:55     INFO -     EventListener.handleEvent*fetch@dom/media/test/test_buffered.html:88:7
[task 2023-06-28T17:36:55.416Z] 17:36:55     INFO -     startTest@dom/media/test/test_buffered.html:107:8
[task 2023-06-28T17:36:55.416Z] 17:36:55     INFO -     MediaTestManager/this.nextTest@dom/media/test/manifest.js:2461:12
[task 2023-06-28T17:36:55.416Z] 17:36:55     INFO -     MediaTestManager/this.runTests/<@dom/media/test/manifest.js:2356:12
[task 2023-06-28T17:36:55.416Z] 17:36:55     INFO -     Async*MediaTestManager/this.runTests@dom/media/test/manifest.js:2355:19
[task 2023-06-28T17:36:55.416Z] 17:36:55     INFO -     @dom/media/test/test_buffered.html:112:9
[task 2023-06-28T17:36:55.417Z] 17:36:55     INFO - split.webm-7 timed out!
[task 2023-06-28T17:36:55.419Z] 17:36:55     INFO - [finished split.webm-7] remaining= 
[task 2023-06-28T17:36:55.419Z] 17:36:55     INFO - TEST-PASS | dom/media/test/test_buffered.html | [finished split.webm-7 t=180.611] Length of array should match number of running tests 
[task 2023-06-28T17:36:55.421Z] 17:36:55     INFO - Finished at Wed Jun 28 2023 17:36:55 GMT+0000 (Coordinated Universal Time) (1687973815.306s)
[task 2023-06-28T17:36:55.422Z] 17:36:55     INFO - Running time: 180.668s
[task 2023-06-28T17:36:55.422Z] 17:36:55     INFO - GECKO(3288) | MEMORY STAT | vsize 2103990MB | vsizeMaxContiguous 67850303MB | residentFast 124MB | heapAllocated 10MB
[task 2023-06-28T17:36:55.423Z] 17:36:55     INFO - TEST-OK | dom/media/test/test_buffered.html | took 180776ms

Hello Karl! This seems to be frequent after your latest changes reached Central, as it can be seen here. Could you please take a look at this?

PS There are several lines of failure :
TEST-UNEXPECTED-FAIL | dom/media/test/test_load_candidates.html | Test timed out!
TEST-UNEXPECTED-FAIL | dom/media/test/test_reset_src.html | Test timed out!
TEST-UNEXPECTED-FAIL | dom/media/test/test_arraybuffer.html | Test timed out!
TEST-UNEXPECTED-FAIL | dom/media/test/test_buffered.html | Test timed out!

Thank you!

Flags: needinfo?(karlt)

Set release status flags based on info from the regressing bug 1829068

Summary: High frequency [tier 2] dom/media/test/test_buffered.html | Test timed out! → High frequency [tier 2] dom/media/test/< test> | Test timed out!
Flags: needinfo?(karlt)
Regressed by: 1839938
No longer regressed by: 1829068
Summary: High frequency [tier 2] dom/media/test/< test> | Test timed out! → High frequency [tier 2] wmfme dom/media/test/< test> | Test timed out!
Blocks: 1796871

:alwu, since you are the author of the regressor, bug 1839938, could you take a look?

For more information, please visit BugBot documentation.

Flags: needinfo?(alwu)
Whiteboard: [retriggered]

Set release status flags based on info from the regressing bug 1839938

Still debugging this one, but didn't understand why bug1839938 can cause these regression...

I found that sometime the media engine couldn't create dcomp surface handle successfully. I've tried to add a retry mechanism, but once that issue appears, all function calls to create the dcomp handle would fail as well. That might be caused by limited resources controlled by DXVA/media engine or other window frameworks, so when we run tests too frequently and parallelly, there won't be enough resource to create the handle.

The workaround I found is to disable the parallel test for media engine test, which seems help for the media engine (or whatever controls the resource) to properly release the old resource so that we can always create the handle successfully.

Flags: needinfo?(alwu)

There have been 33 total failures](https://treeherder.mozilla.org/intermittent-failures/bugdetails?startday=2023-07-01&endday=2023-07-08&tree=trunk&bug=1840914) in the last 7 days.
There are:

  • 6 failures on windows11-64-2009-asan-qr opt
  • 3 failures on windows11-64-2009-ccov-qr opt
  • 14 failures n windows11-64-2009-qr opt and debug
  • 10 failures on windows11-64-2009-shippable-qr opt

Recent failure log.

The intermittent failure is due to the media engine couldn't create a
dcomp surface handle sucessfully.

The error didn't show much information to us, and one possible reason is
that running too much test parallelly consumes hardware resources very
fast so that there won't be enough resource to create a new dcomp
handle.

I've tried to call GetVideoSwapchainHandle() later multiple times if
we fail to create the handle at the first try. However, it didn't work.
Once we got that error, all following calls for
GetVideoSwapchainHandle() would be error as well.

This seems the media engine internal issue, which we couldn't solve in
our end. Therefore, we workaround this problem by disabling the parallel
test when running the media engine tests.

Depends on D183182

Assignee: nobody → alwu
Status: NEW → ASSIGNED
Pushed by alwu@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/cb5af72a426a part1 : add log and marker to capture the failure of updating video size. r=media-playback-reviewers,chunmin https://hg.mozilla.org/integration/autoland/rev/1c909f267af6 part2 : disable parallel test for media engine tests. r=media-playback-reviewers,chunmin
Flags: needinfo?(alwu)
Pushed by alwu@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/79d3f546d9fc part1 : add log and marker to capture the failure of updating video size. r=media-playback-reviewers,chunmin https://hg.mozilla.org/integration/autoland/rev/1098fbc31aff part2 : disable parallel test for media engine tests. r=media-playback-reviewers,chunmin
Status: ASSIGNED → RESOLVED
Closed: 2 years ago
Resolution: --- → FIXED
Target Milestone: --- → 117 Branch

The patch landed in nightly and beta is affected.
:alwu, is this bug important enough to require an uplift?

  • If yes, please nominate the patch for beta approval.
  • If no, please set status-firefox116 to wontfix.

For more information, please visit BugBot documentation.

Flags: needinfo?(alwu)
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: