[Linux-cluster] CMAN panicing...
Patrick Caulfield
pcaulfie at redhat.com
Mon Dec 11 14:52:53 UTC 2006
HAWKER, Dan wrote:
>
> Hi All,
>
> Have been testing a RH cluster for a month or so and was working fine.
> Predictably enough, now that its in production, its playing up a bit.
>
> Have 3 nodes setup as a 3-node cluster. Am only using it for GFS. We don't
> run any *real* cluster apps. The nodes are Gonzo, Kermit & Voyager. The
> shared storage is an EMC iSCSI unit, each node has a Qlogic HBA inside.
>
> They seem to be having similar error messages. Almost as if they are
> simultaneously fencing eachother off and hence causing problems.
>
> I have attached a jpg screen grab of Kermit, (Voyager auto-rebooted when
> this happened) and the /var/log/messages from the node that survived.
>
> Any ideas what is causing this, and more importantly a direction I can aim
> at to fix it...
>
Can you tell me which version of cluster suite this is please ?
patrick
More information about the Linux-cluster
mailing list