During the start of a cluster containing 5 nodes of HotRod Server, I receive a couple of warnings like this:
2010-11-04 13:55:06,721 WARN [org.infinispan.remoting.InboundInvocationHandlerImpl] (OOB-7,wp-334) Cache named [___defaultcache] exists but isn't in a state to handle invocations. Its state is INSTANTIATED.
2010-11-04 13:55:06,722 WARN [org.infinispan.remoting.InboundInvocationHandlerImpl] (OOB-6,wp-334) Cache named [___defaultcache] exists but isn't in a state to handle invocations. Its state is INSTANTIATED.

And eventually an exception:

2010-11-04 13:56:19,199 ERROR [org.infinispan.distribution.JoinTask] (Rehasher-wp-334) Caught exception!
java.lang.ClassCastException: org.infinispan.remoting.responses.RequestIgnoredResponse cannot be cast to org.infinispan.remoting.responses.SuccessfulResponse
at org.infinispan.distribution.JoinTask.updateTopologyInfo(JoinTask.java:162)
at org.infinispan.distribution.JoinTask.broadcastNewCh(JoinTask.java:157)
at org.infinispan.distribution.JoinTask.performRehash(JoinTask.java:103)
at org.infinispan.distribution.RehashTask.call(RehashTask.java:53)
at org.infinispan.distribution.RehashTask.call(RehashTask.java:33)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:303)
at java.util.concurrent.FutureTask.run(FutureTask.java:138)
at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)
at java.lang.Thread.run(Thread.java:619)

By the way, I'm trying to answer the question - is it possiblle to start mid size cluster (let's say 20 nodes) at once - every node at the very same time. Have you ever tried this scenario?