486970 - [SeaMonkey, Linux] mochitest-plain times out and mochitest-chrome fails since bug 485036 landing

Status: VERIFIED FIXED

(SeaMonkey :: Release Engineering, defect)

Description

[Serge Gautherie (:sgautherie)]

{
http://tinderbox.mozilla.org/showlog.cgi?log=SeaMonkey/1238938947.1238945640.4337.gz
Linux comm-central unit test on 2009/04/05 06:42:27

*** 76875 INFO Running /tests/toolkit/content/tests/widgets/test_menubar.xul...
Document http://localhost:8888/tests/toolkit/content/tests/widgets/window_menubar.xul loaded successfully

command timed out: 300 seconds without output, killing pid 19660
process killed by signal 9

[then]

*** 5411 INFO Error: Unable to restore focus, expect failures and timeouts.
*** 5412 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_bug360437.xul | Test timed out.

*** 5419 INFO Error: Unable to restore focus, expect failures and timeouts.
*** 5420 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_bug418874.xul | t1 value correct following input - got "", expected "1"
*** 5421 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_bug418874.xul | undo correctly enabled when emptyText was not changed through property - got false, expected true
*** 5423 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_bug418874.xul | t2 value correct following input - got "", expected "2"
*** 5424 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_bug418874.xul | undo correctly enabled when emptyText explicitly changed through property - got false, expected true

*** 5440 INFO Error: Unable to restore focus, expect failures and timeouts.
*** 5442 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_findbar.xul | testNormalFind: find field is not focused
*** 5443 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_findbar.xul | testNormalFind: failed to find 'text tes'
*** 5450 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_findbar.xul | Case Sensitive label was not correctly updated
*** 5451 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_findbar.xul | Test timed out.

*** 5454 INFO Error: Unable to restore focus, expect failures and timeouts.
*** 5455 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_keys.xul | Test timed out.
*** 5458 INFO Error: Unable to restore focus, expect failures and timeouts.
*** 5459 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_largemenu.xul | Test timed out.
*** 5460 ERROR TEST-UNEXPECTED-FAIL | chrome://mochikit/content/chrome/toolkit/content/tests/chrome/test_largemenu.xul | Too many test timeouts, giving up.
}

I'm not backing it out just yet, as it passes on Linux Firefox 3.5.
What do you think ?


PS: I guess this tells us which tests are "fragile" to this kind of failure ;->

Comment 1

[Matthew Gregan [:kinetik]]

Weird.  I'd try to argue that there's no way that change could cause this, but the SeaMonkey tinderbox is pretty damning in this case.  Still, that doesn't explain why the change has been fine on trunk and 1.9.1.  Is there any way to back this change out just on SeaMonkey, or is that impossible due to the way it tracks the 1.9.1 branch?

Comment 2

[Serge Gautherie (:sgautherie)]

(In reply to comment #1)
> Weird.  I'd try to argue that there's no way that change could cause this, but
> the SeaMonkey tinderbox is pretty damning in this case.  Still, that doesn't
> explain why the change has been fine on trunk and 1.9.1.

I concur with all that.

> Is there any way to back this change out just on SeaMonkey,
> or is that impossible due to the way it tracks the 1.9.1 branch?

(Hg) Backout would be 1.9.1; SeaMonkey specific checkin action would mean #ifdef'ing.
Atm, I prefer to wait for KaiRo (or Callek) to have a look at the machine first: he might (want to) try a local backout to confirm.

Comment 3

[Robert Kaiser]

Grr, I had the cron jobs for restarting metacity in the wrong place (main crontab with user-crontab syntax) and so they didn't work. The test failures were due to a crashed window manager and it not coming up again. Fixed the crontab now, metacity is up again, Linux should go green.

Comment 4

[Serge Gautherie (:sgautherie)]

B.Fixed, box is back to green :-)