Bug 58545

Summary: REGRESSION (r83764): webkitpy.layout_tests.port.chromium_win_unittest.ChromiumWinTest.test_default_worker_model fails on apple-windows-13
Product: WebKit Reporter: Adam Roben (:aroben) <aroben>
Component: Tools / TestsAssignee: Nobody <webkit-unassigned>
Severity: Normal CC: dpranke
Priority: P2 Keywords: MakingBotsRed, PlatformOnly
Version: 528+ (Nightly build)   
Hardware: PC   
OS: Windows XP   
URL: http://build.webkit.org/builders/Windows%20XP%20Debug%20%28Tests%29/builds/27597/steps/webkitpy-test/logs/stdio
Description Flags
Patch aroben: review+

Description Adam Roben (:aroben) 2011-04-14 08:21:38 PDT
webkitpy.layout_tests.port.chromium_win_unittest.ChromiumWinTest.test_default_worker_model has been failing on apple-windows-13 since somewhere between r83760 and r83769. It's not failing on any of our other Windows slaves.

http://build.webkit.org/builders/Windows%20XP%20Debug%20%28Tests%29/builds/27554 did not fail
http://build.webkit.org/builders/Windows%20XP%20Debug%20%28Tests%29/builds/27557 failed
Comment 1 Adam Roben (:aroben) 2011-04-14 08:22:46 PDT
r83764 seems almost certainly to blame: http://trac.webkit.org/changeset/83764
Comment 2 Dirk Pranke 2011-04-14 09:12:09 PDT
Whoops.  Yeah, that would do it. The fix is simple (to wrap the check in an 'if multiprocessing_is_available:' block) but I won't be able to get to it until later this morning.
Comment 3 Adam Roben (:aroben) 2011-04-14 10:44:46 PDT
I guess maybe apple-windows-13 has a different version of Python from the other bots?
Comment 4 Adam Roben (:aroben) 2011-04-14 10:46:27 PDT
Indeed, apple-windows-13 has 2.5.2, apple-windows-14 has 2.6.5.

I really wish all slaves on a particular builder were clones of each other!
Comment 5 Dirk Pranke 2011-04-14 11:50:25 PDT
Created attachment 89617 [details]
Comment 6 Dirk Pranke 2011-04-14 11:54:01 PDT
Committed r83880: <http://trac.webkit.org/changeset/83880>