[Pacemaker] Few questions

Dejan Muhamedagic dejanmm at fastmail.fm
Wed Feb 11 10:44:09 UTC 2009


Hi,

On Tue, Feb 10, 2009 at 12:09:01PM +0530, Glory Smith wrote:
> >
> >
> > >
> > > As you said if CRM decides a node should be fenced it will be fenced and
> > it
> > > has nothing to do with sbd.
> >
> > Right. The decision itself has nothing to do with sbd. sbd is
> > just an executioner.
> >
> > > so in this case cluster must need an another
> > > stonith to provide fencing of errant node.  is my understanding is
> > correct.
> >
> > No.
> >
> > > can you explain a bit more , how CRM decide that a node should be fence ?
> > is
> > > it when node become unreachable.??
> >
> > Yes. Also if a resource can't be stopped on that node.
> >
> > Thanks,
> >
> 
> Hi Dejan,
> 
> Well i  tried to configure sbd  on two nodes cluster.  it is not working
> properly.  need your view here. In split-brain situation both nodes are
> trying to stonith each other and they have written  "Reset" to each other's
> slot also. (nodes name are tejas and toddy)
> 
> tejas:~ # sbd -d /dev/sdd list
> 0       tejas reset   toddy
> 1       toddy  reset   tejas
> 
> Syslog is also saying that they have reset each other successfully.
> 
> Feb 10 10:02:54 tejas stonithd: [9195]: info: Succeeded to STONITH the node
> toddy: optype=RESET. whodoit: tejas
> 
> Feb 10 10:03:55 toddy stonithd: [9195]: info: Succeeded to STONITH the node
> tejas: optype=RESET. whodoit: toddy

This is expected. Only writing a "poison pill", which you see
listed above, is enough to consider the node fenced.

> But both nodes didn't reset actually.  can you tell me why it happened.

Nope. Did you follow http://www.linux-ha.org/SBD_Fencing? Is the
sbd daemon running?

Thanks,

Dejan


>  Thanks,

> _______________________________________________
> Pacemaker mailing list
> Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker





More information about the Pacemaker mailing list