The relevant metric is nodes * resources. When the compressed CIB size exceeds corosync's maximum message size, thats when the cluster is going to experience trouble.
On Fri, Jul 24, 2009 at 6:36 PM, <li...@unimitated.com> wrote: > What is the general maximum number of nodes I can realistically support in a > cluster with pacemaker and openais? If I wanted to do something like 100 > nodes supporting a set of resources in a reliable network, could I make that > work or will I have a lot of problems keeping things up, failing, etc.? > > Thanks for any insight. > > _______________________________________________ > Pacemaker mailing list > Pacemaker@oss.clusterlabs.org > http://oss.clusterlabs.org/mailman/listinfo/pacemaker > _______________________________________________ Pacemaker mailing list Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker
