Closed Bug 1156451 Opened 9 years ago Closed 9 years ago

Intermittent test_transitions_events.html | absolutely any damn failure at all

Categories

(Core :: CSS Parsing and Computation, defect)

Product:
Core
Component:
CSS Parsing and Computation
Platform:
x86
Linux
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED INVALID

People

(Reporter: RyanVM, Unassigned)

References

Details

(Keywords: intermittent-failure) 

First instance was starred into bug 1018502 last week.
https://bugzilla.mozilla.org/show_bug.cgi?id=1018502#c67

12:06:59 INFO - 814 INFO TEST-START | layout/style/test/test_transitions_events.html
12:06:59 INFO - ++DOMWINDOW == 42 (0x8bdfa400) [pid = 1864] [serial = 2606] [outer = 0x984e0600]
12:06:59 INFO - [Parent 1864] WARNING: Performance Entry buffer size maximum reached!: file /builds/slave/m-in-lx-d-00000000000000000000/build/src/dom/base/nsPerformance.cpp, line 593
12:07:00 INFO - 815 INFO TEST-PASS | layout/style/test/test_transitions_events.html | A valid string reason is expected
12:07:00 INFO - 816 INFO TEST-PASS | layout/style/test/test_transitions_events.html | Reason cannot be empty
12:07:00 INFO - 817 INFO TEST-FAIL | layout/style/test/test_transitions_events.html | The author of the test has indicated that flaky timeouts are expected. Reason: untriaged
12:07:00 INFO - 818 INFO TEST-FAIL | layout/style/test/test_transitions_events.html | The author of the test has indicated that flaky timeouts are expected. Reason: untriaged
12:07:00 INFO - 819 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 820 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 821 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 822 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 823 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 824 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 825 INFO TEST-PASS | layout/style/test/test_transitions_events.html | transitionend on body on root
12:07:00 INFO - 826 INFO TEST-PASS | layout/style/test/test_transitions_events.html | propertyName for transitionend on body
12:07:00 INFO - 827 INFO TEST-PASS | layout/style/test/test_transitions_events.html | elapsedTime for transitionend on body
12:07:00 INFO - 828 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 829 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 830 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 831 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 832 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 833 INFO TEST-PASS | layout/style/test/test_transitions_events.html | undefined assertion name
12:07:00 INFO - 834 INFO TEST-UNEXPECTED-FAIL | layout/style/test/test_transitions_events.html | unexpected event on SPAN element with id 'six' elapsedTime=0.5 propertyName='color' - expected PASS
12:07:00 INFO - 835 INFO TEST-UNEXPECTED-FAIL | layout/style/test/test_transitions_events.html | unexpected event on SPAN element with id 'five' elapsedTime=0.5 propertyName='color' - expected PASS
Why do we bother putting the failure part in the bug summary?
Summary: Intermittent test_transitions_events.html | unexpected event on SPAN element with id 'six' elapsedTime=0.5 propertyName='color' → Intermittent test_transitions_events.html | absolutely any damn failure at all
Brian says he'll look into this.
Flags: needinfo?(bbirtles)
For my own reference, I suspect the regressions to bug 1148949 might have the same cause. Perhaps bug 1155823?
Actually, this is likely bug 1141607. The failures pick up in frequency with the changeset for that bug which also modified the failing test file: test_transitions_events.html
Flags: needinfo?(bbirtles) → needinfo?(bzbarsky)
We seem to get an extra transitionend firing on "four" on Android only.
No idea... I did add an explicit outline-color style for #four, precisely because I was afraid of issues like this.  So there should be no extra transitions happening on #four, I'd think.
Flags: needinfo?(bzbarsky)
One thing I noticed while auditing -- the patch in bug 1141607 should have added a started_test() call -- as, probably, should some other changes to the test in the past.  I think we're two short for the listener on #one, and also two short for the listener on the root.  But we're two over for #seven since that listener doesn't call finished_test (and also doesn't check for duplicate events on ::after).

It might be worth seeing if adding the necessary calls makes failures that are currently intermittent more reliable.
The patches in bug 1161021 (in particular, patch 2) should fix this variant of the errors here:

> 93 INFO TEST-UNEXPECTED-FAIL |
> layout/style/test/test_transitions_events.html | transitionend on four on
> root - Result logged after SimpleTest.finish()
> 94 INFO TEST-UNEXPECTED-FAIL |
> layout/style/test/test_transitions_events.html | propertyName for
> transitionend on four - Result logged after SimpleTest.finish()
> 95 INFO TEST-UNEXPECTED-FAIL |
> layout/style/test/test_transitions_events.html | elapsedTime for
> transitionend on four - Result logged after SimpleTest.finish()
> 96 INFO TEST-UNEXPECTED-FAIL |
> layout/style/test/test_transitions_events.html | computed style for
> transitionend on four (end of reverse transition) - Result logged after
> SimpleTest.finish()

which I think are the only active ones.
All failures from comment 3 to the present (except for the unrelated test_value_storage.html failures in comment 38 and comment 71) should be fixed by bug 1161021.
Component: DOM: Events → CSS Parsing and Computation
Given that the problem represented by the bulk of the comments is fixed, I don't think this bug is useful anymore.  If comment 0 or comment 1/comment 2 come back, they should get their own bugs.
Status: NEW → RESOLVED
Closed: 9 years ago
Resolution: --- → INVALID
You need to log in before you can comment on or make changes to this bug.