Closed Bug 725911 Opened 8 years ago Closed 8 years ago

Nearly constant mochitest-robocop test_bug720538 | Checking bottom-left corner of viewport - Color rgba(255,255,255,255) not close enough to expected rgb(0,128,0)

Categories

(Testing :: Mochitest, defect)

ARM
Android
defect
Not set

Tracking

(Not tracked)

RESOLVED WORKSFORME

People

(Reporter: philor, Assigned: kats)

References

Details

(Keywords: intermittent-failure)

Attachments

(1 file)

https://tbpl.mozilla.org/php/getParsedLog.php?id=9220469&tree=Mozilla-Inbound
Android Tegra 250 mozilla-inbound opt test robocop on 2012-02-09 14:06:57 PST for push 779ca542b71f

7 INFO TEST-PASS | test_bug720538 | Checking page background the right of the iframe - Color rgba(0,130,0,255)  close enough to expected rgb(0,128,0)
8 INFO TEST-UNEXPECTED-FAIL | test_bug720538 | Checking bottom-left corner of viewport - Color rgba(255,255,255,255) not close enough to expected rgb(0,128,0)
9 INFO TEST-END | test_bug720538 | finished in 22248ms

43 INFO TEST-PASS | test_bug720538 | Checking page background the right of the iframe - Color rgba(0,130,0,255)  close enough to expected rgb(0,128,0)

44 INFO TEST-UNEXPECTED-FAIL | test_bug720538 | Checking bottom-left corner of viewport - Color rgba(255,255,255,255) not close enough to expected rgb(0,128,0)

45 INFO TEST-END | test_bug720538 | finished in 22248ms
I've been running this patch on try all morning (https://tbpl.mozilla.org/?tree=Try&rev=d9db4e89b110) and it hasn't failed yet (at least, not in the right way). I might need to just land the patch and wait for it to fail on inbound to diagnose the error.
Attachment #596157 - Flags: review?(jmaher)
Landed on inbound, hopefully the next time it fails like this we'll have more info on why.

https://hg.mozilla.org/integration/mozilla-inbound/rev/8716f10ab139
Whiteboard: [orange] → [orange], [inbound], [leave open after inbound merge]
hmm the tests are green, but there are no results:
https://tbpl.mozilla.org/php/getParsedLog.php?id=9263099&tree=Mozilla-Inbound&full=1

seems that we might be hitting that OOM error.  Our log parsing/error handling needs to be updated to be more robust I guess.
This hasn't failed since I put in the above patch. Should the bug be closed?
I don't know if it has passed either, I see a lot of tests which fail due to OOM.
Depends on: 728955
Whiteboard: [orange], [inbound], [leave open after inbound merge] → [orange]
So even after bug 731637 landed, I'm still seeing the following tests not run to completion:

testPanCorrectness
test_bug720538
testOverscroll
testAxisLocking

The TEST-END marker for these tests aren't in the logs on tbpl; in some cases all of the assertions also don't run. Clearly the test is dying somehow and it's not getting picked up by the framework. This should be fixed.
I believe bug 734221 will be the fix we need for this.  It seems that sutagent needs a little updating.
Depends on: 734221
This works now. I see it running to completion on an unrelated try run (https://tbpl.mozilla.org/?tree=Try&rev=145347dde805).
Assignee: nobody → bugmail.mozilla
Status: NEW → RESOLVED
Closed: 8 years ago
Resolution: --- → WORKSFORME
Whiteboard: [orange]
You need to log in before you can comment on or make changes to this bug.