CentOS - Mar 2011 - BUG: soft lockup CPU stuck for 10seconds (Server went down)

Hello,

Today my server stopped responding.
i went to the console and on the screen there were a continuous loop of the
following info shown on the screen:

BUG: soft lockup - CPU#0 stuck for 10s! [java:13959]

and alot of other information.
ii've took a screen shot of the info shown , you can find it under the
following url: http://img585.imageshack.us/i/img00012201103070833.jpg/
and had to hard reset for it to be back up and running.

i tried googling with no luck for direct relevant info.
so hoping you can help out

Thanks,

--Roland
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.centos.org/pipermail/centos/attachments/20110307/31aa0347/attachment-0002.html>

Am 07.03.2011 08:31, schrieb Roland RoLaNd:
> 
> Hello,
> 
> Today my server stopped responding.
> i went to the console and on the screen there were a continuous loop of the
following info shown on the screen:
> 
> BUG: soft lockup - CPU#0 stuck for 10s! [java:13959]
> 
> and alot of other information.
> ii've took a screen shot of the info shown , you can find it under the
following url: http://img585.imageshack.us/i/img00012201103070833.jpg/
> and had to hard reset for it to be back up and running.
> 
> i tried googling with no luck for direct relevant info.
> so hoping you can help out
> 
> Thanks,
> 
> --Roland
A good reason why to run CentOS 5.4 (at least its kernel)?

Running remotely vulnerable Oracle Java which can be affected by a DoS
throug a web application?

Alexander

On Mon, 07 Mar 2011 09:31:42 +0200
Roland RoLaNd wrote:
> BUG: soft lockup - CPU#0 stuck for 10s! [java:13959]
> i tried googling with no luck for direct relevant info.
The first google result for the above string takes me here:

http://bugs.centos.org/view.php?id=3582

Which in turn contains a reference that takes me here:

https://bugzilla.redhat.com/show_bug.cgi?id=484590

And it appears that this issue  was fixed with kernel version 2.6.18-164.

-- 
MELVILLE THEATRE ~ Melville Sask ~ www.melvilletheatre.com
www.creekfm.com - FIFTY THOUSAND WATTS of POW WOW POWER!

----- Original Message -----
| Hello,
| 
| Today my server stopped responding.
| i went to the console and on the screen there were a continuous loop
| of the following info shown on the screen:
| 
| BUG: soft lockup - CPU#0 stuck for 10s! [java:13959]
| 
| and alot of other information.
| ii've took a screen shot of the info shown , you can find it under the
| following url: http://img585.imageshack.us/i/img00012201103070833.jpg/
| and had to hard reset for it to be back up and running.
| 
| i tried googling with no luck for direct relevant info.
| so hoping you can help out
| 
| Thanks,
| 
| --Roland
| 
| _______________________________________________
| CentOS mailing list
| CentOS at centos.org
| http://lists.centos.org/mailman/listinfo/centos

This is likely due to thread deadlock.  What is the load on the machine at the
time that this error occurs?  I've seen this very error when running The
Mathworks Distributed Computing Toolbox server.  The machine would become very
"unsettled" when it tried to run on more than four 2 CPUs.  How many
CPUs are in this system?

-- 
James A. Peltier
IT Services - Research Computing Group
Simon Fraser University - Burnaby Campus
Phone   : 778-782-6573
Fax     : 778-782-3045
E-Mail  : jpeltier at sfu.ca
Website : http://www.sfu.ca/itservices
          http://blogs.sfu.ca/people/jpeltier

On 07/03/11 08:31, Roland RoLaNd wrote:
> Hello,
> 
> Today my server stopped responding.
> i went to the console and on the screen there were a continuous loop of the
> following info shown on the screen:
> 
> BUG: soft lockup - CPU#0 stuck for 10s! [java:13959]
> 
> and alot of other information.
> ii've took a screen shot of the info shown , you can find it under the
> following url: http://img585.imageshack.us/i/img00012201103070833.jpg/
> and had to hard reset for it to be back up and running.
> 
> i tried googling with no luck for direct relevant info.
> so hoping you can help out
Some real kernel developers might have better insight on why this happens.
 But this hits APIC timers during a syscall.  I would probably try to boot
the box with 'noapic' in the kernel command line, to see if this
improves
things or not.

Do you see the "soft lockup - CPU#0" always?  or does it also happen
to
other CPUs as well?  And if it does, is the java process running on more CPUs?


kind regards,

David Sommerseth