629444 - [Debug Linux] Long buildbot timeout fires instead of shorter mochitest harness timeout, at least for mochitests-4 test_value_cloning.html

Status: NEW

(SeaMonkey :: Testing Infrastructure, defect)

Description

[Serge Gautherie (:sgautherie)]

(Noticed while checking bug 558705 cases.)

It seems that, for some time, long buildbot timeout fires instead of shorter mochitest harness timeout, at least for this test.

I could try and guess that this is related to our switch to buildbot v0.8,
but I have no (searched for) evidence of that at all:
it could be anything else (like a Tracemonkey issue or whatever).

http://tinderbox.mozilla.org/showlog.cgi?log=SeaMonkey/1296101281.1296106770.1629.gz
Linux comm-central-trunk debug test mochitests-4/5 on 2011/01/26 20:08:01
{
64831 INFO TEST-PASS | /tests/layout/style/test/test_value_cloning.html | computed values should match when cloning -moz-transform-origin: bottom - "50% 100%" should equal "50% 100%"
++DOMWINDOW == 96 (0xd66b784) [serial = 2991] [outer = 0xd01bcc0]

command timed out: 5400 seconds elapsed, killing pid 19277
}
(5,400 s = 1,5 hours.)

Comment 1

[Serge Gautherie (:sgautherie)]

Fwiw, this was reported once (only) in bug 558705 comment 69:
{
http://tinderbox.mozilla.org/showlog.cgi?log=MozillaTry/1288759735.1288765196.881.gz
Rev3 MacOSX Leopard 10.5.8 tryserver debug test mochitests-4/5 on 2010/11/02
21:48:55
}

Comment 2

[Shawn Wilsher :sdwilsh]

Happened on mozilla-central:
Rev3 Fedora 12 mozilla-central debug test mochitests-4/5 on 2011/03/31 11:15:15 
http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1301595315.1301601172.21763.gz&fulltext=1#err0

Comment 3

[Marco Bonardo [:mak]]

http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1305111100.1305116992.21309.gz#err0