[Ocfs2-users] heartbeat write timeout

[Sunil Mushran]

<<<<<<<<<<<<<<
I was able to reproduce this behaviour on SuSE 10.1 beta 6 and 8. Furthermore,
ocfs2 (also 1.2) on the very same system but installed with CentOS 4.3 does not
make the kernel panic.
>>>>>>>>>>>>>>>

Is this true? Well, I imagine, considering you wrote it. But I would still
like a confirmation as this narrows down the problem. This means this
is SLES10 specific.

Ahh... check out the following bug. This appears to be an iscsi issue
on sles10.

Bug 157463 - iSCSI initiator unstable

Stephan A. Rickauer wrote:
> Sunil Mushran wrote:
>   
>> This looks very similar to what we were seeing with cfq.
>>
>> What type of load are you running when this happens?
>>     
>
> Stress test using bonnie++, writing/reading on one node only.
>
>   
>> What is the full kernel version?
>>     
>
> Testing with SuSE 10.1 beta's and rc1, currently 2.6.16-20-default (rc1)
>
>   
>> Are there are other messages during that time?
>>     
>
> Nothing.
>
>   
>> What type of storage is this?
>>     
>
> It's a Transtec Provigo 1100F, iSCSI, which is now called 410F. In case
> you need to know more about it, I can connect you to the guys writing
> the firmware for it, they are somewhere in the States. It is basically a
> kernel 2.4 based linux system.
>
> Additionally, I have filed a bug at novell.com/SuSE:
> https://bugzilla.novell.com/show_bug.cgi?id=167289 (may be of interest,
> since I explain our setup there in more detail).
>
>
> Thanks for your help.
>
>   
> ------------------------------------------------------------------------
>
> _______________________________________________
> Ocfs2-users mailing list
> Ocfs2-users at oss.oracle.com
> http://oss.oracle.com/mailman/listinfo/ocfs2-users
>