Closed Bug 1710069 Opened 3 years ago Closed 3 years ago

[tb perma] docshell/test/unit/test_subframe_stop_after_parent_error.js | xpcshell return code: 0

Categories

(Thunderbird :: Upstream Synchronization, defect, P5)

Tracking

(thunderbird91+ fixed)

RESOLVED FIXED
92 Branch
Tracking Status
thunderbird91 + fixed

People

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

References

Details

(Keywords: intermittent-failure, Whiteboard: [stockwell disable-recommended])

Attachments

(1 file)

Summary: [tb perma] docshell/test/unit/test_subframe_stop_after_parent_error.js | xpcshell return code: 0 → [tb] docshell/test/unit/test_subframe_stop_after_parent_error.js | xpcshell return code: 0

From https://treeherder.mozilla.org/logviewer?job_id=339254902&repo=comm-central&lineNumber=4362:
[task 2021-05-10T10:58:27.335Z] 10:58:27 INFO - PID 5142 | \x07[Parent 5142, Main Thread] ###!!! ASSERTION: Attempt to load a non-authorised load in the parent process: http://example.com/top.html: 'false', file /builds/worker/checkouts/gecko/netwerk/ipc/ParentProcessDocumentChannel.cpp:66

Summary: [tb] docshell/test/unit/test_subframe_stop_after_parent_error.js | xpcshell return code: 0 → [tb perma] docshell/test/unit/test_subframe_stop_after_parent_error.js | xpcshell return code: 0

The failing part of the test is testRemoveFrameAfterErrorPage which will not load the errorpage, just keep showing about:blank https://searchfox.org/mozilla-central/rev/048146721b3c1a568257afb4325c7a8d94334e2d/docshell/test/unit/test_subframe_stop_after_parent_error.js#114

Kris is there anything we can do about this failure? It must be something to do with the config difference between Firefox and Thunderbird, but I can't figure out what. I do know that it goes away with browser.tabs.remote.autostart set to false, but that's hardly a solution.

Flags: needinfo?(kmaglione+bmo)

If we can't find a solution this week I'd like to turn the test off for Thunderbird so we don't have to put up with it for an entire ESR cycle.

BTW if anyone's wondering why the failure count has dropped, we've largely given up marking it.

I have very little hope of this test passing on Thunderbird before 91 moves to beta. Therefore I'm turning it off to reduce the noise on what will become comm-beta and comm-esr91.

Assignee: nobody → geoff
Status: NEW → ASSIGNED

(In reply to Geoff Lankow (:darktrojan) from comment #21)

Kris is there anything we can do about this failure? It must be something to do with the config difference between Firefox and Thunderbird, but I can't figure out what. I do know that it goes away with browser.tabs.remote.autostart set to false, but that's hardly a solution.

I believe that is the cause of this issue. Firefox always runs xpcshell tests with browser.tabs.remote.autostart = false, even though we don't ship Firefox with e10s disabled.

Flags: needinfo?(geoff)

Then I think all we can do is disable the test (or fix it to work with e10s). We can't stop Thunderbird tests running with the default preferences, and we can't override e10s in the manifest.

Flags: needinfo?(geoff)

Alternatively the test could check browser.tabs.remote.autostart and bail out if it's true.

See Also: → 1168178
Pushed by geoff@darktrojan.net:
https://hg.mozilla.org/integration/autoland/rev/056ab3b94275
Disable test_subframe_stop_after_parent_error.js on Thunderbird until after 91. r=nika
Pushed by csabou@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/052cf3de0adb
Fix syntax to unbust decision task. a=bustage-fix
Status: ASSIGNED → RESOLVED
Closed: 3 years ago
Resolution: --- → FIXED
Target Milestone: --- → 92 Branch
Flags: needinfo?(kmaglione+bmo)

Comment on attachment 9230410 [details]
Bug 1710069 - Disable test_subframe_stop_after_parent_error.js on Thunderbird until after 91. r=kmag,nika

Beta/Release Uplift Approval Request

  • User impact if declined: Just disables a test that fails when run in Thunderbird. We'd like to clear the failing test from TaskCluster so that failures we're actually concerned about aren't missed.

Please note that I screwed up the original patch so two changes landed.

  • Is this code covered by automated tests?: Yes
  • Has the fix been verified in Nightly?: No
  • Needs manual test from QE?: No
  • If yes, steps to reproduce:
  • List of other uplifts needed: None
  • Risk to taking this patch: Low
  • Why is the change risky/not risky? (and alternatives if risky):
  • String changes made/needed:
Attachment #9230410 - Flags: approval-mozilla-beta?

Comment on attachment 9230410 [details]
Bug 1710069 - Disable test_subframe_stop_after_parent_error.js on Thunderbird until after 91. r=kmag,nika

Low risk in early betas, approved for uplift in 91 beta 4, thanks

Attachment #9230410 - Flags: approval-mozilla-beta? → approval-mozilla-beta+

Build bustage with this error:

Intermittent manifestparser.expression.ParseError: could not parse: || appname == 'thunderbird'
https://hg.mozilla.org/releases/mozilla-beta/rev/faa5a9ebd3e1

Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Status: REOPENED → RESOLVED
Closed: 3 years ago3 years ago
Resolution: --- → FIXED

Comment on attachment 9230410 [details]
Bug 1710069 - Disable test_subframe_stop_after_parent_error.js on Thunderbird until after 91. r=kmag,nika

Clearing the approval flag to get this off the needs-uplift radar.

Attachment #9230410 - Flags: approval-mozilla-beta+
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: