Closed Bug 1395469 Opened 5 years ago Closed 5 years ago

Permafail test_submit_unencrypted_info_warning.py TestSubmitUnencryptedInfoWarning.test_submit_unencrypted_info_warning | UnknownException: TypeError: win is null

Categories

(Testing :: Marionette, defect, P1)

Version 3
defect

Tracking

(firefox56 unaffected, firefox57 fixed)

RESOLVED FIXED
mozilla57
Tracking Status
firefox56 --- unaffected
firefox57 --- fixed

People

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

References

Details

(Keywords: intermittent-failure, Whiteboard: [stockwell fixed:other])

Attachments

(1 file)

Filed by: hskupin [at] mozilla.com

https://treeherder.mozilla.org/logviewer.html#?job_id=127302159&repo=autoland

https://queue.taskcluster.net/v1/task/UXc_E2snTfea9TkGNoQfmw/runs/0/artifacts/public/logs/live_backing.log

The signature of the test failure as reported on bug 1392854 has been changed since the patch on bug 
1394881 landed.
Andreas, please have a look. The issues are still not fixed.

Here is how it looks now:

[task 2017-08-31T06:10:36.510759Z] 06:10:36     INFO - TEST-UNEXPECTED-ERROR | test_submit_unencrypted_info_warning.py TestSubmitUnencryptedInfoWarning.test_submit_unencrypted_info_warning | UnknownException: TypeError: win is null
[task 2017-08-31T06:10:36.511638Z] 06:10:36     INFO - stacktrace:
[task 2017-08-31T06:10:36.512463Z] 06:10:36     INFO - 	element.isStale@chrome://marionette/content/element.js:639:7
[task 2017-08-31T06:10:36.513423Z] 06:10:36     INFO - 	get@chrome://marionette/content/element.js:181:9
[task 2017-08-31T06:10:36.514317Z] 06:10:36     INFO - 	isElementEnabled@chrome://marionette/content/listener.js:1484:12
[task 2017-08-31T06:10:36.515168Z] 06:10:36     INFO - 	dispatch/</req<@chrome://marionette/content/listener.js:515:16
[task 2017-08-31T06:10:36.516005Z] 06:10:36     INFO - 	async*dispatch/<@chrome://marionette/content/listener.js:513:16
Flags: needinfo?(ato)
Priority: P5 → P1
Summary: Intermittent test_submit_unencrypted_info_warning.py TestSubmitUnencryptedInfoWarning.test_submit_unencrypted_info_warning | UnknownException: TypeError: win is null → Permafail test_submit_unencrypted_info_warning.py TestSubmitUnencryptedInfoWarning.test_submit_unencrypted_info_warning | UnknownException: TypeError: win is null
I would like to see a specific Marionette unit test which covers this scenario.
A permanent failure is an improvement, we’re making progress on
the matter here.

I suspect that XUL elements’ owner global is not a transparent
proxy in the same way that an HTML element’s WindowProxy is.  That
is, when a ChromeWindow disappears, a XUL elements’ owner global
will be reset to null, whereas an HTML element’s will always be a
WindowProxy pointing to the current browsing context’s Window.
Assignee: nobody → ato
Status: NEW → ASSIGNED
Flags: needinfo?(ato)
OS: Unspecified → All
Hardware: Unspecified → All
HTML elements' ownerGlobal is a WindowProxy that indirects operations
to the current browsing context.  However, for XUL elements this global
is ChromeWindow which gets reset to null when the window is discarded.

For the same reason, since win indirects to the current browsing context's
associated window, we need to test that the element's node document is
not the active document.

Finally test that its shadow-including root is a document.

MozReview-Commit-ID: EYh18P8DbcN
Comment on attachment 8903131 [details] [diff] [review]
Check discarded ChromeWindow in staleness test. r?automatedtester

https://bugzil.la/1394881 has not landed on m-c yet so mozreview
prevents me from submitting a patch there.  This will apply on top
of m-c once the dependent patch lands, and I’ve confirmed locally
that it fixes the Firefox UI test fallout.
Attachment #8903131 - Flags: review?(dburns)
Pushed by atolfsen@mozilla.com:
https://hg.mozilla.org/integration/mozilla-inbound/rev/f2124b88c59d
Check discarded ChromeWindow in staleness test. r=me
I landed this as r=me and will do a rare request for a
post-integration review because the sheriffs were concerned with
leaving this perma-fail in the tree over the weekend.
Whiteboard: [stockwell needswork]
(In reply to Andreas Tolfsen ‹:ato› from comment #8)
> I landed this as r=me and will do a rare request for a
> post-integration review because the sheriffs were concerned with
> leaving this perma-fail in the tree over the weekend.

Just as a side note... I was around the whole day. Why didn't you just ask me, or Maja, or James? We all could have done that review request. It's not always David who has to review everything.
https://hg.mozilla.org/mozilla-central/rev/f2124b88c59d
Status: ASSIGNED → RESOLVED
Closed: 5 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla57
Whiteboard: [stockwell needswork] → [stockwell fixed:other]
Attachment #8903131 - Flags: review?(dburns) → review+
You need to log in before you can comment on or make changes to this bug.