Both centos machines fail a few times a day on this test: http://tinderbox.mozilla.org/showlog.cgi?log=Mozilla2/1215070076.1215075802.8487.gz&fulltext=1 Looking at the testcase there is a setTimeout in there: https://bugzilla.mozilla.org/attachment.cgi?id=255989&action=diff Would like to remove that dependency or up the timeout to try and increase the reliability of this test.

Test runs noticeably faster locally too.

Comment on attachment 328038 [details] [diff] [review] v1.0 We should also get this on the branch for those tests.

Pushed to mozilla-central: http://hg.mozilla.org/mozilla-central/index.cgi/rev/7392f9bdbea8 And to CVS: Checking in toolkit/content/tests/chrome/bug331215_window.xul; new revision: 1.2; previous revision: 1.1

So far, looks like the new version has only failed one of three runs on mozilla-central qm-centos5-03, only one of three on qm-centos5-moz2-01, and only two of four on cn-sea-qm-centos5-01. Is that better than before?

Ugh. Is it failing in a different manner?

Inconveniently enough, someone seems to have disappeared the past for the mozilla-central boxes, but my memory of them was that they were just like the SeaMonkey box's ongoing failure, "*** 455 ERROR FAIL | Findfield status attribute should have been 'notfound' after entering ltest | | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_bug331215.xul"

The failure before was a timeout, so that's something new. New bug please?

If it turns out that all this "oh, can you file that bug for me?" stuff is because filing bugs costs money, and the bills only go out every eight years, I'm gonna be testy. Bug 443763.