[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [Linux-cluster] GFS on Ubuntu hardy: BUG soft lockup - CPU stuck



Well just for the record... After trying again to mount the GFS on the
second and having the exact same problem, I did:

* A gfs_fsck, which found many inode errors.
* Paid more attention so that when the second node comes up, the fc
switch port is already up.

After that I got up the cluster two times without any problem (had a
scheduled shutdown in between).
Unfortunaly, I didn't have the time to rule one of those out, or do
any more tests.

Thanks for  the anwsers...


Diego Morales




On Thu, Apr 9, 2009 at 9:38 AM, Virginian <virginian blueyonder co uk> wrote:
> This could be a kernel bug. A quick google search revealed this:
>
> http://www.linuxquestions.org/questions/linux-server-73/bug-soft-lockup-cpu3-stuck-for-10s-648946/
>
> Probably worth googling some more but I don't think this is a RHCS issue to
> be honest.
>
>
> ----- Original Message ----- From: "Diego Morales" <dgmorales gmail com>
> To: "linux clustering" <linux-cluster redhat com>
> Sent: Wednesday, April 08, 2009 8:23 PM
> Subject: Re: [Linux-cluster] GFS on Ubuntu hardy: BUG soft lockup - CPU
> stuck
>
>
>> Well, it's a Supermicro server hardware (seems a little like "generic
>> server hardware with glorified name", never had seen one before), Xeon
>> 3Ghz, two CPUs + HT, 8GB RAM.
>>
>> Well, maybe. We recently had some overheating problems and power
>> outages on that room too.
>>
>> But the lockup only happened when doing gfs mounting/umounting. And
>> those machines were up using the gfs file system for several days with
>> no problem.
>>
>>
>> On Wed, Apr 8, 2009 at 2:47 PM, Virginian <virginian blueyonder co uk>
>> wrote:
>>>
>>> I'm not entirely sure that the CPU lockup message is actually caused by
>>> RHCS.What hardware are you running on, how many CPUs, how much RAM? I've
>>> see
>>> this soft lockup error before on IBM 3850 machines running RHEL 5 (but
>>> not
>>> RHCS) I think....
>>>
>>
>> --
>> Linux-cluster mailing list
>> Linux-cluster redhat com
>> https://www.redhat.com/mailman/listinfo/linux-cluster
>>
>
>
> --
> Linux-cluster mailing list
> Linux-cluster redhat com
> https://www.redhat.com/mailman/listinfo/linux-cluster
>


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]