Hi David,

David Rhodes Clymer wrote:

> Jul 27 13:49:16 scylla kernel: [18446743878.972008] BUG: soft lockup - CPU#1 stuck for 17163091968s! [transfer_log_fi:9956]

Seems like a timekeeping problem --- that uptime is more than 500
years.

[...]
> Pid: 9956, comm: transfer_log_fi Not tainted 2.6.32-5-amd64 #1 PowerEdge R710

Good.

[...]
> [18446743878.972212] Call Trace:

Not that I expect it to help, but do you happen to have the call trace
in your logs?

Thanks,
Jonathan


--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: http://lists.debian.org/20120728062906.GA4739@burratino

On Sat, Jul 28, 2012 at 2:29 AM, Jonathan Nieder <jrnieder@gmail.com> wrote:

Hi David,



David Rhodes Clymer wrote:



> Jul 27 13:49:16 scylla kernel: [18446743878.972008] BUG: soft lockup - CPU#1 stuck for 17163091968s! [transfer_log_fi:9956]



Seems like a timekeeping problem --- that uptime is more than 500

years.



[...]

> Pid: 9956, comm: transfer_log_fi Not tainted 2.6.32-5-amd64 #1 PowerEdge R710



Good.



[...]

> [18446743878.972212] Call Trace:



Not that I expect it to help, but do you happen to have the call trace

in your logs?


Unfortunately, no. There was none. I thought it the missing call trace was very odd, then found this:*http://damntechnology.blogspot.com/2010/04/linux-crash-debug-tips-i-have-soft.html

That person claims that if it is a userspace process, then the kernel can't provide a traceback. So, I don't know if that means this isn't a kernel bug at all, or if it is a scheduling bug, or what.

-davidc
--
David Clymer


VistaShare




866-828-4782, ext. 828

www.VistaShare.com

*




* www.facebook.com/vistashare



* www.twitter.com/vistashare

David Clymer wrote:
> On Sat, Jul 28, 2012 at 2:29 AM, Jonathan Nieder <jrnieder@gmail.com> wrote:
>> David Rhodes Clymer wrote:

>>> Jul 27 13:49:16 scylla kernel: [18446743878.972008] BUG: soft lockup - CPU#1 stuck for 17163091968s! [transfer_log_fi:9956]
[...]
>>> [18446743878.972212] Call Trace:
>>
>> Not that I expect it to help, but do you happen to have the call trace
>> in your logs?
>
> Unfortunately, no. There was none. I thought it the missing call trace was
> very odd, then found this:
> http://damntechnology.blogspot.com/2010/04/linux-crash-debug-tips-i-have-soft.html

Right. Good catch.

> That person claims that if it is a userspace process, then the kernel can't
> provide a traceback. So, I don't know if that means this isn't a kernel bug
> at all, or if it is a scheduling bug, or what.

A jump in time like that is almost certainly a kernel bug.

Thanks,
Jonathan


--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: http://lists.debian.org/20120728142728.GB1603@burratino