browser/components/sessionstore/test/browser_async_remove_tab.js fails with SHIP+BFCache
Categories
(Firefox :: Session Restore, defect, P2)
Tracking
()
| Fission Milestone | M8 |
| Tracking | Status | |
|---|---|---|
| firefox-esr78 | --- | unaffected |
| firefox89 | --- | unaffected |
| firefox90 | --- | unaffected |
| firefox91 | --- | affected |
People
(Reporter: smaug, Unassigned)
References
(Regression)
Details
(Keywords: regression)
https://searchfox.org/mozilla-central/rev/fbc78630d6d435b9b9647a4ac8a5def888f85f04/browser/components/sessionstore/test/browser_async_remove_tab.js#109 fails every now and then.
The test was disabled until last week, bug 1715264
One may need to use --repeat <some number> to trigger the failure. Fails locally with an opt linux build.
|
||
Updated•3 years ago
|
|
||
Updated•3 years ago
|
|
||
Updated•3 years ago
|
|
|
||
Comment 1•3 years ago
|
||
Set release status flags based on info from the regressing bug 1715264
Failure in recent try push: https://treeherder.mozilla.org/jobs?repo=try&revision=a2ad2dd6c66300716faa7e737df6af960a71adc8
Should be the same root cause as bug 1716444, which is a race in the final tab state flush (we appear to destroy frame loaders more often with BFCache enabled).
|
|
||
Comment 4•3 years ago
|
||
same reasoning as https://bugzilla.mozilla.org/show_bug.cgi?id=1716444#c3 so M8
Sometimes we do a tab state flush on a replaced browsing context. We aren't updating tab state because of bug 1705689, but we still send the "tab state flush complete" notification, which is causing things to get into a bad state.
Description
•