Bug 171234
| Summary: | REGRESSION (r215620): Multiple worker LayoutTests are flaky timeouts | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Ryan Haddad <ryanhaddad> |
| Component: | New Bugs | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED CONFIGURATION CHANGED | ||
| Severity: | Normal | CC: | ap, jlewis3, keith_miller, webkit-bug-importer |
| Priority: | P2 | Keywords: | InRadar |
| Version: | WebKit Nightly Build | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| See Also: | https://bugs.webkit.org/show_bug.cgi?id=170976 | ||
Ryan Haddad
LayoutTest http/tests/websocket/tests/hybi/workers/worker-reload.html is a flaky timeout
https://webkit-test-results.webkit.org/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=http%2Ftests%2Fwebsocket%2Ftests%2Fhybi%2Fworkers%2Fworker-reload.html
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Ryan Haddad
*** Bug 171237 has been marked as a duplicate of this bug. ***
Ryan Haddad
I'm seeing flakiness with the following tests:
fast/workers/worker-exception-during-navigation.html
http/tests/websocket/tests/hybi/workers/worker-reload.html
fetch/fetch-worker-crash.html
All of them seem are timing out very frequently after http://trac.webkit.org/changeset/215620. I cannot reproduce the timeouts with r215619.
Ryan Haddad
Same goes for fast/workers/worker-exception-during-navigation.html
Ryan Haddad
(In reply to Ryan Haddad from comment #3)
> Same goes for fast/workers/worker-exception-during-navigation.html
Heh, I already pasted that one in comment 2.
Radar WebKit Bug Importer
<rdar://problem/31827333>
Ryan Haddad
I think we should roll out this change (or rather, changes) because of its impact on tests. It isn't immediately clear what changes need to go, since it was rolled out and back in once already.
Ryan Haddad
(In reply to Ryan Haddad from comment #6)
> I think we should roll out this change (or rather, changes) because of its
> impact on tests. It isn't immediately clear what changes need to go, since
> it was rolled out and back in once already.
I did some digging and I now see that there have been follow-up fixes that stop this from being a clean revert.
Alexey Proskuryakov
I wonder if Yusuke fixed this in bug 171319.
Matt Lewis
(In reply to Alexey Proskuryakov from comment #8)
> I wonder if Yusuke fixed this in bug 171319.
After 3 builds, looks like http/tests/websocket/tests/hybi/workers/worker-reload.html isn't flaky but will keep a watch on it.
Matt Lewis
Checked the flakinesses dashboard for all test mentioned in this bug. Issue seems to be resolved.