On 2008-11-22T08:33:31, Andrew Beekhof <[EMAIL PROTECTED]> wrote:
> > i only use master/slave. i tested master/master without heartbeat and
> > ocfs2. it caused troubles like ocfs2 fencing a node and problems with
> > consistency of the drbd device after that fencing.
> I'm guessing you weren't using the "userspace stack" functionality of
> OCFS2 - otherwise this would never have happened.
> This is the perfect example of internal split-brain that occurs when
> applications use a different messaging/membership layer than the rest
> of the cluster.
For completeness, my guess would be that this is related more to the
latency incurred by drbd8 replicated writes, in particular while drbd8
is figuring out that the node has gone down.
ocfs2 sees the latencies, figures out the node has gone down, and fences
itself.
_Consistency_ issues however should _never_ arise, please report these
to the drbd developers!
Regards,
Lars
--
Teamlead Kernel, SuSE Labs, Research and Development
SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg)
"Experience is the name everyone gives to their mistakes." -- Oscar Wilde
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
