[jboss-user] [Clustering/JBoss] - StateTransferTimeout appears to be ignored

[dcolwell]

I have deployed JBoss 4.2.1GA across a two node cluster.   It's configured for asynchronous, non-transactional replication of user session data using UDP.  Application state is exclusively servlet based with no EJBs.  After starting up the first node I fire up the second and have observed it pause from 1 to 10 minutes after logging:

Fetching state (will wait for 30000 milliseconds)

The next line in my server.log file appears successful, just long:

state was retrieved successfully (in 189067 milliseconds)

The 30000 millis is configured by setting a StateTransferTimeout attribute, but different values appear to be equally ignored.  Does anyone have thoughts on what may be causing this?  The amount of user data potentially active on the first node is rather small (< 100k) if any.

Thanks much!


View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4124957#4124957

Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4124957