I have a special case here, both nodes are in a dmz and I can't ping them through the gateway. Isn't using the stonith:ssh an option?
Thanks On 11.02.2010 15:22, Marian Marinov wrote: > Hello, > you may consider configuring two pingd resources, one to the default gw of > both > nodes and one to ping the remote node trough the gw. > > Then any of the nodes can say that it is alive if it has ping to the GW and to > the other machine. > You can then decide what to do with the master/slave resources. > > Note, that generally it is advised to use at least 3 nodes to be sure that > there is no split brain and every node knows which is the master. > > Regards, > Marian > > On Thursday 11 February 2010 15:49:26 Frank Lazzarini wrote: > >> Hi, >> >> I have the following setup, I have two nodes in a cluster, all working >> fine with the resources. If i take one down the other node becomes >> master. But what if I pull the plug on the communication interface? As I >> understood it I need to use STONITH but how can one node shoot the other >> in the head if it doesn't see the node anymore? Or would I need node >> fencing. >> >> My question is actually just to get some advice on how to deal with a >> situation like this. >> >> >> >> [node1]-------X-----[node2] >> eth1 >> >> >> If eth1 is down ( say the cross link cable is plugged ) how can I say >> that only node1 will be master? >> >> >> Thanks for any answers ... >> _______________________________________________ >> Linux-HA mailing list >> [email protected] >> http://lists.linux-ha.org/mailman/listinfo/linux-ha >> See also: http://linux-ha.org/ReportingProblems >> >> > > > > _______________________________________________ > Linux-HA mailing list > [email protected] > http://lists.linux-ha.org/mailman/listinfo/linux-ha > See also: http://linux-ha.org/ReportingProblems _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
