[Linux-cluster] Two node cluster without quorum disk

Spencer Parker sjpark at rondaandspencer.info
Wed Apr 22 18:30:11 UTC 2009


I have this same problem with the 2 node cluster that I am working on.  I
have mysql running with an NFS that holds all of the data for it.  If I
unplug one of the node to test the network, it will move the service to the
other node...no problem.  When that node comes back up, it will then kill
the services on the node that it failed back too.  This goes on and on
resulting in a loop.  They'll start up and then begin to fence each other.
One comes up...starts the services...then it dies...the other one takes
over...starts the services and then dies...over and over again.

I have found that I need to start one node up...and let it run for a short
period of time...then start up the second node...and sometimes it won't
start the loop and sometimes it will.

2009/4/22 Jakub Suchy <jakub.suchy at enlogit.cz>

> > > I tried this, but when I killed the switch for the cluster traffic,
> > > BOTH nodes got fenced! Can I avoid this?
>
> Hi,
> i accidentally deleted previous messages so I am not sure if you have
> sent your cluster.conf? Please resend it...
>
>
> Jakub
>
> --
> Linux-cluster mailing list
> Linux-cluster at redhat.com
> https://www.redhat.com/mailman/listinfo/linux-cluster
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/linux-cluster/attachments/20090422/d9281cc0/attachment.htm>


More information about the Linux-cluster mailing list