Bug 161249

Summary:	REGRESSION (r204601): Many new flaky timeouts started on 8/18/16
Product:	WebKit	Reporter:	Ryan Haddad <ryanhaddad>
Component:	New Bugs	Assignee:	Nobody <webkit-unassigned>
Status:	NEW ---
Severity:	Normal	CC:	ap, eric.carlson
Priority:	P2
Version:	WebKit Nightly Build
Hardware:	Unspecified
OS:	Unspecified
See Also:	https://bugs.webkit.org/show_bug.cgi?id=161208 https://bugs.webkit.org/show_bug.cgi?id=161210 https://bugs.webkit.org/show_bug.cgi?id=161157 https://bugs.webkit.org/show_bug.cgi?id=161164 https://bugs.webkit.org/show_bug.cgi?id=161279 https://bugs.webkit.org/show_bug.cgi?id=161144 https://bugs.webkit.org/show_bug.cgi?id=160652 https://bugs.webkit.org/show_bug.cgi?id=161141 https://bugs.webkit.org/show_bug.cgi?id=161205

Description Ryan Haddad 2016-08-26 09:14:10 PDT

There have been a large number of flaky tests that all started timing out around 8/18/16.

Comment 1 Alexey Proskuryakov 2016-08-29 10:28:37 PDT

This is very likely caused by <http://trac.webkit.org/r204601>, even though there is no obvious relationship to the failing tests.

The failures are rare, but cross-referencing multiple tests and multiple bots, I found that there are no timeouts on these before r204601, and then there are 2-3 failures for each revision.

The timeout happens in the same worker that ran hls/hls-accessiblity-describes-video-menu.html, only 2-3 minutes later.

Comment 2 Alexey Proskuryakov 2016-08-29 14:02:12 PDT

Ryan skipped hls/hls-accessiblity-describes-video-menu.html in r205124 to see if it helps.

Comment 3 Alexey Proskuryakov 2016-08-31 14:52:50 PDT

Well, that did it. Still a mystery why this test causes timeouts later.