Closed Bug 1478106 Opened 7 years ago Closed 7 years ago

Intermittent /css/css-transitions/properties-value-001.html | outline-color color(rgba) / events - assert_equals: Expected TransitionEnd events triggered on .transition expected "outline-color:0.5s" but got ""

Categories

(Core :: CSS Parsing and Computation, defect, P5)

defect

Tracking

()

RESOLVED FIXED
mozilla63
Tracking Status
firefox63 --- fixed

People

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

References

Details

(Keywords: intermittent-failure)

Attachments

(1 file, 1 obsolete file)

Over the last 7 days there are 30 failures on this bug. These happen on linux64. Here is the most recent log example: https://treeherder.mozilla.org/logviewer.html#?job_id=191055916&repo=mozilla-inbound&lineNumber=2104 [task 2018-07-31T07:16:03.194Z] 07:16:03 INFO - TEST-PASS | /css/css-transitions/properties-value-001.html | outline-offset length(cm) / values [task 2018-07-31T07:16:03.194Z] 07:16:03 INFO - TEST-UNEXPECTED-FAIL | /css/css-transitions/properties-value-001.html | outline-offset length(cm) / events - assert_equals: Expected TransitionEnd events triggered on .transition expected "outline-offset:0.5s" but got "" [task 2018-07-31T07:16:03.194Z] 07:16:03 INFO - assertExpectedEventsFunc/<@http://web-platform.test:8000/css/css-transitions/support/generalParallelTest.js:207:13 [task 2018-07-31T07:16:03.195Z] 07:16:03 INFO - Test.prototype.step@http://web-platform.test:8000/resources/testharness.js:1538:20
Flags: needinfo?(mreavy)
Hi Sean -- Was anything new landed in the last 2 weeks to trigger this? Thanks.
Flags: needinfo?(mreavy) → needinfo?(svoisen)
Hiro - Do you possibly have any insight here?
Flags: needinfo?(svoisen) → needinfo?(hikezoe)
I know this test is originally flaky. :/ But interesting, there is no suspicious changes making this failure more frequent, AFAICT. Now it's time to investigate what's going on there. Keep NI to me, CCing Brian since, as far as I know, he has been rewriting these test cases.
The reason why the test is originally flaky is that the test uses setTimeout() [1] to tell the transitions in question finished. I wonder other UAs are not suffering from this kind of failures? The reason why the test started failing frequently is that, given that all failures on ASAN build (as per the link in comment 1), we started doing something heavy load on ASAN build, I guess. We should just skip this test on ASAN too until Brian makes the test more sane. (We've already skipped on debug build) [1] https://hg.mozilla.org/mozilla-central/file/937c64cef999/testing/web-platform/tests/css/css-transitions/support/runParallelAsyncHarness.js#l117
Assignee: nobody → hikezoe
Status: NEW → ASSIGNED
Flags: needinfo?(hikezoe)
(In reply to Hiroyuki Ikezoe (:hiro) (PTO on Aug 20) from comment #6) > Keep NI to me, CCing Brian since, as far as I know, he has been rewriting > these test cases. Yeah, I haven't rewritten this one yet. I wouldn't waste time debugging it and would just disable the test for now.
(Sorry, wrote that comment before I saw yours.)
Skip properties-value-001.html on ASAN build. I'd like to change the link in properties-value-001.html.ini to a bug for fixing this test. Do we already have a bug for that?
Skip properties-value-001.html on ASAN build
Comment on attachment 8999039 [details] Bug 1478106 - Skip properties-value-001.html on ASAN build. r?birtles Brian Birtles (:birtles) has approved the revision.
Attachment #8999039 - Flags: review+
Attachment #8999033 - Attachment is obsolete: true
Pushed by hikezoe@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/ab427e770c87 Skip properties-value-001.html on ASAN build. r=birtles
Status: ASSIGNED → RESOLVED
Closed: 7 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla63
See Also: → 1486434
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: