[Linux-HA] Another question on suicide STONITH (and multiple STONITH devices at a time)

Andrew Beekhof beekhof at gmail.com
Fri Oct 5 03:06:26 MDT 2007

On 10/5/07, Dominik Klein <dk at in-telegence.net> wrote:
> Hi
> I'm currently fiddling with a cluster of xen domains on different
> physical machines.
> Physical Machine 1 runs virtual Machine A
> Physical Machine 2 runs virtual Machine B
> A and B are my HA nodes.
> I modified the xen-stonith plugin from [1] to work with 2 dom0 hosts and
> configured stonith to use this plugin. It works great.
> One thing this can not accomplish is a node reboot if dom0 loses all
> network connections (3, so I have some redundancy here). Total loss of
> network connectivity (in dom0 !!) causes splitbrain.
> I think suicide might help here. It should be done in the node if no
> ping hosts are available.

/lazy paste from another thread a few minutes ago

the inherent problem with a suicide plugin is that you're relying on a
sick node to shoot itself... depending on exactly how sick it is, that
may or may not be impossible.

wrt. to ssh... the "meatware" plugin would arguably be better a better
option, at least from a reliability perspective, as it will work in
all cases (providing the admin is prepared to get up at 3am or
whenever the node goes bad :-)

> So here are my questions:
> Can I configure multiple stonith devices (extern/xen and suicide) at a time?

yes, but i've no idea if/how stonith would use them

> Could someone share some cib.xml snippet that says "if no ping hosts are
> reachable, commit suicide"?

instead, use pingd to make your resource's stop when connectivity is lost


> Regards
> Dominik
> [1] http://etbe.coker.com.au/2007/06/24/xen-and-heartbeat/
> _______________________________________________
> Linux-HA mailing list
> Linux-HA at lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems

More information about the Linux-HA mailing list