Closed Bug 573161 Opened 16 years ago Closed 15 years ago

These are not the media test failures you are looking for

Categories

(Core :: Audio/Video, defect)

Product:
Core
Component:
Audio/Video
Platform:
x86
Windows 7
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED FIXED

People

(Reporter: kinetik, Unassigned)

References

Details

(Keywords: intermittent-failure, Whiteboard: [fixed by 623444?])

s: talos-r3-fed-006 49152 ERROR TEST-UNEXPECTED-FAIL | /tests/content/media/test/test_progress.html | Test timed out. 49162 ERROR TEST-UNEXPECTED-FAIL | /tests/content/media/test/test_reactivate.html | Test timed out. 49194 ERROR TEST-UNEXPECTED-FAIL | /tests/content/media/test/test_replay_metadata.html | Test timed out. 49214 ERROR TEST-UNEXPECTED-FAIL | /tests/content/media/test/test_seek.html | Test timed out. 49215 ERROR TEST-UNEXPECTED-FAIL | (SimpleTest/TestRunner.js) | 4 test timeouts, giving up. 49216 ERROR TEST-UNEXPECTED-FAIL | (SimpleTest/TestRunner.js) | Skipping 80 remaining tests. TEST-UNEXPECTED-FAIL | Shutdown | application timed out after 330 seconds with no output PROCESS-CRASH | Shutdown | application crashed (minidump found) Thread 0 (crashed) TEST-UNEXPECTED-FAIL | automationutils.processLeakLog() | missing output line for total leaks!
There are a few suspects in the window: me@kylehuey.com – Thu Jun 17 18:11:22 2010 PDT Bug 508082 - Raw decoder mgregan@mozilla.com – Thu Jun 17 17:48:12 2010 PDT Bug 572034 - Linux and Mac64 Y'CbCr changes Bug 555121 - Vorbis 1.3.1 All of these were green after landing, but this particular pattern of intermittent orange (where all four tests fail, and only on the Fedora machines) has only started since these patches landed. Failures have been at: Rev3 Fedora 12 mozilla-central debug test mochitests-1/5 on 2010/06/17 20:02:43 Rev3 Fedora 12 mozilla-central debug test mochitests-1/5 on 2010/06/17 21:24:33 Rev3 Fedora 12 mozilla-central debug test mochitests-1/5 on 2010/06/18 12:13:29 Rev3 Fedora 12 mozilla-central debug test mochitests-1/5 on 2010/06/18 13:27:28
Blocks: 438871
Whiteboard: [orange]
I think we should back out bug 572034, as that's the most specific to Linux. Then try backing out bug 555121. I'm having trouble working out how any of these changes could've caused this, though, so these backouts are blind.
Summary: Intermittant failure of test_progress.html, test_reactivate.html, test_replay_metadata.html, and test_seek.html on Fedora 12 machines (talos-r3-fed-006) → Intermittent failure of test_progress.html, test_reactivate.html, test_replay_metadata.html, and test_seek.html on Fedora 12 machines (talos-r3-fed-006)
Backed out bug 572034, let's see what happens.
Backed out bug 555121.
Still happening with both of mine backed out: http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1276932985.1276936919.14691.gz Kyle, can you try backing out the raw decoder? Alternatively, we can try debugging the failures directly, but I'm not going to be able to do that until Monday NZST.
(In reply to comment #8) > Still happening with both of mine backed out: > http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1276932985.1276936919.14691.gz > > Kyle, can you try backing out the raw decoder? > > Alternatively, we can try debugging the failures directly, but I'm not going to > be able to do that until Monday NZST. I'll back out later today.
So it's been all quiet since I backed out my patch. Matthew, you want to try relanding your patches?
There's a lot of spew from nsContentUtils::DispatchTrustedEvent right before 3 of the 4 timeouts. WARNING: NS_ENSURE_TRUE(docEvent && target) failed: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3157 WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result 0x80070057: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3193 WARNING: NS_ENSURE_TRUE(docEvent && target) failed: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3157 WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result 0x80070057: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3193 WARNING: NS_ENSURE_TRUE(docEvent && target) failed: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3157 WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result 0x80070057: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3193 WARNING: NS_ENSURE_TRUE(docEvent && target) failed: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3157 WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result 0x80070057: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3193 WARNING: NS_ENSURE_TRUE(docEvent && target) failed: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3157 WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result 0x80070057: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3193 WARNING: NS_ENSURE_TRUE(docEvent && target) failed: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3157 WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result 0x80070057: file /builds/slave/mozilla-central-linux-debug/build/content/base/src/nsContentUtils.cpp, line 3193 It's also odd that all of these failures happen after ogg videos.
When I run these tests locally they don't time out, but I do get INFO Error: Unable to restore focus, expect failures and timeouts.
I relanded the raw decoder a few days ago and we haven't seen this yet. Marking WFM
Status: NEW → RESOLVED
Closed: 16 years ago
Resolution: --- → WORKSFORME
Status: RESOLVED → REOPENED
Resolution: WORKSFORME → ---
Filed bug 600458 for comment 36 comment 37 and comment 38, since it's patently obvious that they are not this bug.
Summary: Intermittent failure of test_progress.html, test_reactivate.html, test_replay_metadata.html, and test_seek.html on Fedora 12 machines (talos-r3-fed-006) → Intermittent timeout in test_progress.html, test_reactivate.html, test_replay_metadata.html, and test_seek.html on Linux
(In reply to comment #47) > ehsan%mozilla.com > http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1290981667.1290984079.21600.gz > Rev3 Fedora 12x64 mozilla-central debug test mochitests-1/5 on 2010/11/28 > 14:01:07 > > s: talos-r3-fed64-008 > 57898 ERROR TEST-UNEXPECTED-FAIL | /tests/content/media/test/test_seek.html | > Test timed out. It's possible this is a regression from the landing of bug 610570, so I think we should open a new bug specifically for this, until we figure it out. Also apart from this latest starring, there's been no occurrences of one of these failures in the last month, and we've checked in a bunch of fixes to resolve random video orange recently, so I think we should close this bug. It's an unspecific catch all for a number of failures which shouldn't happen any more.
Filed Bug 615134 for the failure in comment 47. The failures this bug was filed for should be fixed so I'm closing this bug. If failures occur in the other tests which this bug was filed for, please file a new bug for them.
Status: REOPENED → RESOLVED
Closed: 16 years ago → 15 years ago
Resolution: --- → FIXED
We weren't 100% sure about comment 50. Let's see if it happens again. May need to file a new bug.
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Whiteboard: [orange] → [orange][fixed by 623444?]
Adjusting the summary to reflect the current bugs which should be starred as this bug about four particular tests timing out all together once upon a time. Bug numbers are cheap.
Status: REOPENED → RESOLVED
Closed: 15 years ago → 15 years ago
Resolution: --- → FIXED
Summary: Intermittent timeout in test_progress.html, test_reactivate.html, test_replay_metadata.html, and test_seek.html on Linux → These are not the media test failures you are looking for
Whiteboard: [orange][fixed by 623444?] → [fixed by 623444?]
You need to log in before you can comment on or make changes to this bug.