Bug 150577 - http/tests/contentextensions/async-xhr-onerror.html is flaky on mac-wk2
Summary: http/tests/contentextensions/async-xhr-onerror.html is flaky on mac-wk2
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebKit Misc. (show other bugs)
Version: WebKit Nightly Build
Hardware: Mac Unspecified
: P2 Normal
Assignee: Alex Christensen
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-10-26 15:33 PDT by Ryan Haddad
Modified: 2015-11-09 13:25 PST (History)
3 users (show)

See Also:


Attachments
Patch (1.85 KB, patch)
2015-11-06 13:18 PST, Alex Christensen
no flags Details | Formatted Diff | Diff
Patch (3.79 KB, patch)
2015-11-09 12:35 PST, Alex Christensen
no flags Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Ryan Haddad 2015-10-26 15:33:09 PDT
http/tests/contentextensions/async-xhr-onerror.html is flaky on mac-wk2

<http://webkit-test-results.webkit.org/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=http%2Ftests%2Fcontentextensions%2Fasync-xhr-onerror.html>

--- /Volumes/Data/slave/mavericks-debug-tests-wk2/build/layout-test-results/http/tests/contentextensions/async-xhr-onerror-expected.txt
+++ /Volumes/Data/slave/mavericks-debug-tests-wk2/build/layout-test-results/http/tests/contentextensions/async-xhr-onerror-actual.txt
@@ -5,6 +5,7 @@
 Asynchronous onreadystatechange status: 0, readyState:1, responseText: 
 Finished runTest. Waiting for callbacks
 Asynchronous onreadystatechange status: 0, readyState:4, responseText: 
+timeout - this should not be called.
 onerror was called.
 Asynchronous onreadystatechange status: 0, readyState:4, responseText: 
 onerror was called.
Comment 1 Ryan Haddad 2015-11-05 09:20:14 PST
Marked as flaky in <https://trac.webkit.org/r192059>
Comment 2 Alex Christensen 2015-11-06 13:18:49 PST
Created attachment 264954 [details]
Patch
Comment 3 Alex Christensen 2015-11-06 15:23:54 PST
This should be landed along with marking the test as not flaky any more.
Comment 4 Darin Adler 2015-11-07 16:33:47 PST
Comment on attachment 264954 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=264954&action=review

> Source/WebCore/xml/XMLHttpRequest.cpp:786
> +            if (m_timeoutTimer.isActive())
> +                m_timeoutTimer.stop();

Why do we need the "!isActive" check here? Calling stop should be harmless if the timer is not active.
Comment 5 Alex Christensen 2015-11-09 11:57:01 PST
(In reply to comment #4)
> Comment on attachment 264954 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=264954&action=review
> 
> > Source/WebCore/xml/XMLHttpRequest.cpp:786
> > +            if (m_timeoutTimer.isActive())
> > +                m_timeoutTimer.stop();
> 
> Why do we need the "!isActive" check here? Calling stop should be harmless
> if the timer is not active.
It was done elsewhere in this file.  I'll remove it elsewhere.
Comment 6 Alex Christensen 2015-11-09 12:35:32 PST
Created attachment 265081 [details]
Patch
Comment 7 WebKit Commit Bot 2015-11-09 13:25:47 PST
Comment on attachment 265081 [details]
Patch

Clearing flags on attachment: 265081

Committed r192175: <http://trac.webkit.org/changeset/192175>
Comment 8 WebKit Commit Bot 2015-11-09 13:25:52 PST
All reviewed patches have been landed.  Closing bug.