If you think a bug might affect users in the 57 release, please set the correct tracking and status flags for Release Management.

Frequent Win7-only test_accessiblecaret_selection_mode.py AccessibleCaretSelectionModeTestCase.test_long_press_to_select_when_partial_visible_word_is_selected | AssertionError: 'aaaaaaaa' != u'' when Gecko 57 merges to Beta on 2017-09-14

NEW
Unassigned

Status

()

Core
Selection
18 days ago
2 days ago

People

(Reporter: RyanVM, Unassigned)

Tracking

({intermittent-failure})

Trunk
Unspecified
Windows 7
intermittent-failure
Points:
---
Dependency tree / graph

Firefox Tracking Flags

(firefox57+ affected)

Details

(Reporter)

Description

18 days ago
[Tracking Requested - why for this release]: Frequent test failure on the next merge day.

https://treeherder.mozilla.org/logviewer.html#?job_id=127931611&repo=try

TEST-UNEXPECTED-FAIL | test_accessiblecaret_selection_mode.py AccessibleCaretSelectionModeTestCase.test_long_press_to_select_when_partial_visible_word_is_selected | AssertionError: 'aaaaaaaa' != u''

I haven't had a chance to bisect this yet, but it started in the last couple days and is on Win7 only.

To reproduce the build configuration in question, the patches below should suffice.
https://hg.mozilla.org/try/rev/70cb6436d870bc25344b98e8ec3ed1be9cf9afc6
https://hg.mozilla.org/try/rev/30a74bd46d8aa778a3ab706c4c77b567e5355a6d
(Reporter)

Comment 1

17 days ago
Try bisection confirms this was caused by bug 1394881. Failure rate is ~20% on opt builds.
Blocks: 1394881
Flags: needinfo?(ato)
Keywords: intermittent-failure
Summary: test_accessiblecaret_selection_mode.py is going to fail frequently on Windows 7 when Gecko 57 merges to Beta on 2017-09-20 → Frequent Win7-only test_accessiblecaret_selection_mode.py AccessibleCaretSelectionModeTestCase.test_long_press_to_select_when_partial_visible_word_is_selected | AssertionError: 'aaaaaaaa' != u'' when Gecko 57 merges to Beta on 2017-09-20

Comment 2

16 days ago
1 failures in 939 pushes (0.001 failures/push) were associated with this bug in the last 7 days.   

Repository breakdown:
* mozilla-central: 1

Platform breakdown:
* windows7-32: 1

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1396231&startday=2017-08-28&endday=2017-09-03&tree=all
(Reporter)

Comment 3

13 days ago
I've verified on m-c tip that this is still happening around the same frequency. Note that per Sylvestre's email today, the date of the 57 merge to Beta is now a week from today (14-Sep) rather than the 20th as was previously the case.
Summary: Frequent Win7-only test_accessiblecaret_selection_mode.py AccessibleCaretSelectionModeTestCase.test_long_press_to_select_when_partial_visible_word_is_selected | AssertionError: 'aaaaaaaa' != u'' when Gecko 57 merges to Beta on 2017-09-20 → Frequent Win7-only test_accessiblecaret_selection_mode.py AccessibleCaretSelectionModeTestCase.test_long_press_to_select_when_partial_visible_word_is_selected | AssertionError: 'aaaaaaaa' != u'' when Gecko 57 merges to Beta on 2017-09-14

Comment 4

9 days ago
4 failures in 924 pushes (0.004 failures/push) were associated with this bug in the last 7 days.   

Repository breakdown:
* mozilla-inbound: 2
* try: 1
* autoland: 1

Platform breakdown:
* windows7-32: 4

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1396231&startday=2017-09-04&endday=2017-09-10&tree=all
tracking-firefox57: ? → +

Updated

8 days ago
See Also: → bug 1399204
I’ve not been meaning to ignore this, but it’s unclear to me how
and where to reproduce this.  According to https://bugzil.la/1399204
this is now also hit as an intermittent on trunk (m-i currently,
soon m-c?).  But what is unclear to me is what the two patches
linked to in comment #0 changes Marionette’s behaviour?

How would I proceed in reproducing this reliably on m-c?
Depends on: 1399076
Flags: needinfo?(ato)
(Reporter)

Comment 6

7 days ago
The patches in #c0 change the Firefox build configuration to mimic how it'll be built once uplifted to Beta. Branding changes, ifdef changes, etc. For whatever reason, those changes combined with bug 1394881 (confirmed with bisection on Try and lots of retriggers) made this test start failing on Win7 around 20% of the time.
(Reporter)

Comment 7

7 days ago
Also, if you think bug 1399076 may have helped here, I'd be happy to run another Try push and see?
Flags: needinfo?(ato)
(Reporter)

Comment 8

7 days ago
I did a new round of simulations off m-c tip this morning and verified that it's still affected.
Flags: needinfo?(ato)

Comment 9

6 days ago
29 failures in 247 pushes (0.117 failures/push) were associated with this bug yesterday.   

Repository breakdown:
* try: 29

Platform breakdown:
* windows7-32: 18
* windows7-32-nightly: 7
* windows7-32-devedition: 4

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1396231&startday=2017-09-13&endday=2017-09-13&tree=all
29 failures in 1032 pushes (0.028 failures/push) were associated with this bug in the last 7 days.    

Repository breakdown:
* try: 29

Platform breakdown:
* windows7-32: 17
* windows7-32-nightly: 8
* windows7-32-devedition: 4

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1396231&startday=2017-09-11&endday=2017-09-17&tree=all
You need to log in before you can comment on or make changes to this bug.