[Bug 7479] New: Occasional lockup at boot on nforce2, XP3000+, 1.5 GB system, occurs 10-30 % of time at bootups

[bugzilla-daemon at bugzilla.ubuntu.com]

Please do not reply to this email.  You can add comments at
http://bugzilla.ubuntu.com/show_bug.cgi?id=7479
Ubuntu | linux

           Summary: Occasional lockup at boot on nforce2, XP3000+, 1.5 GB
                    system, occurs 10-30 % of time at bootups
           Product: Ubuntu
           Version: unspecified
          Platform: Other
        OS/Version: other
            Status: NEW
          Severity: normal
          Priority: P2
         Component: linux
        AssignedTo: fabbione at ubuntu.com
        ReportedBy: alankila at bel.fi
         QAContact: kernel-team at lists.ubuntu.com


I seem to have a real horde of this kind of weird kernel crashing trouble on my
system. I'd be *sure* to mistrust my hardware if these crashes were random, but
this one has started to happen during the "Starting hotplug..." phase at the
bootup. I've seen it happen three times while I've been watching the system boot
up, and maybe twice when I've been doing something else to come to a crashed
console. Of course, in those times I can't be _sure_ that it's really crashed at
that time...

It is important to note that the system does _not_ crash after it is up and
running. I can play UT2004 or Doom3 or lazy about the web all day and never
experience a hiccup. (This after I disabled inotify; I have reported previous
crashes that were inotify related, and some odd problems that looked like memory
corruption with 2.6.10-3 that I think may have been inotify related, too.)

I think I've been seeing these crashes for the past 2 weeks or so, maybe. I
don't remember, as I don't always get around to writing a bug report immediately
when it happens, you know, Ubuntu being a work in progress, hoping that somebody
else notices, so I hope it would get fixed without me earning the reputation of
a tedious whiner in the process. But this has been going on and nobody seems to
know anything, so I'm making these noises now.

In each case the kernel has been totally dead, and usually with this Oops trace
(I lose the part that scrolls off the screen, the system is totally dead and
will no longer respond):

detach_pid
__unhash_process
release_task
wait_task_zombie
do_signal
do_wait

I wrote that part of the call stack down on paper maybe a week ago, intending to
google around to see if it were widely reported. I just dug that paper up from
the chaos...

I realize this is pretty little to go on with. Today the crash was different,
the crux of the matter was that the EIP was in the console printk routine and
that it was in some kind of recursive printing loop and the oops data ended with
"In interrupt handler, not syncing". Then it was locked solid as usual. It's a
pity I didn't take a digicam picture. I'll try to reboot a few times and disable
the splash mode so that I might get to see what module it is loading when it
dies. (But if the oops happens immediately afterwards I can't scroll up and see
the name of the module, so I'll try not to blink. :-/)

Stay tuned...

-- 
Configure bugmail: http://bugzilla.ubuntu.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.