I have re-installed my cluster with only one instance per worker node and still have the same problem. Everything seems to work but session replication. Snooping the interfaces, I am getting heartbeat/health messages (XML formatted) from each of the node elements, but I don't see anything indicating replication of session variables and the like. I see some Shoal logs on the instances, but they look like success messages.
I do have two WARNING messages on startup. I'm not sure if they are related to my issue. One seems to be related to the cluster trying to use__CallFlowPool. I have seen some references online that suggest this is not a problem, but will try to make it go away. My other warning looks like a jms connect exception. I'm not intentionally using JMS, but maybe that is how the replication is working. Does anyone know where to look to indicated even in the slightest that memory replication is working, trying to work, disabled, etc.? I am now fairly positive that things are configured correctly based on the several slide shows and web searches I have seen. Thanks in advance.
Timestamp
Dec 31, 2008 15:18:42.537
Log Level
WARNING
Logger
javax.enterprise.resource.resourceadapter
Name-Value Pairs
_ThreadID=10;_ThreadName=main;__CallFlowPool;_RequestID=c67c3ed5-5ab6-4dd8-b919-e287455678f3;
Record Number
909
Message ID
RAR5005
Complete Message
Error in accessing XA resource with JNDI name [__CallFlowPool] for recovery
and
Timestamp
Dec 31, 2008 15:18:39.017
Log Level
WARNING
Logger
javax.jms
Name-Value Pairs
_ThreadID=10;_ThreadName=main;_RequestID=c67c3ed5-5ab6-4dd8-b919-e287455678f3;
Record Number
897
Message ID
[C4003]
Complete Message
Error occurred on connection creation [localhost:37676]. - cause: java.net.ConnectException: Connection refused
[Message sent by forum member 'jrobey' (jrobey)]
http://forums.java.net/jive/thread.jspa?messageID=323715