[Pacemaker] How to configure the openais.conf

[lepace]

Hi,
First, thanks everyone for your suggestions
I consider I made a wrong design, I am a newbie,I have some question below:
1. what does the quorum, cluster-glue,resource-agent,openais,pacemaker or corosync? In RHCS, a failover domain can be defined, does the failover domain use a less time to vote to select a node? How it realize it?
2.I have 64 nodes for example, and construct 4 HA clusters,each have 16 nodes.My question is if most nodes in a cluster fails ,and the services on these nodes have been migrated,but maybe the remaining nodes can not afford the so many services.How can I solve this problem and I do not want to interrupt the services for a long time.Can I migrate some of service to nodes in another HA cluster automately?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20091110/e627eb07/attachment-0001.html>