AFAIK the JBoss6 documentation is not complete and partially wrong (e.g. JMS).
But you'll find most of the concepts and configuration similar in JBoss5 and 6.
If not feel free to ask here .
As I am reading the Jboss5 clustering document.... And in that we have
EachJBoss server instance (node) specifies which cluster (i.e., partition) it joins in the ClusterPartition MBean in the deploy/cluster-service.xml file. As deploy/cluster-service.xml file doesn't exist in JBoss 6. So I am going to point out which xml works as cluster-service.xml in Jboss6.
By default (if you use the all configuration) all instance join in to the same cluster.
You should not change the configuration, instead you should start with run.sh -g <partitionName> -u <mcast address> -m <mcast port>
this will separate the instances.
To answer your question:
look into server/all/deploy/cluster/jgroups-channelfactory.sar/META-INF/jgroups-channelfactory-stacks.xml
there you find the configuration you looking for.
But I want a file in which the following code is written..
<! -- Name of the partition being built-->
<! -- The address used to determine thenode name -->
<! -- Determine if deadlock detection isenabled -->
<! -- Max time (in ms) to wait for statetransfer to complete.
Increase for large states -->
<! -- The JGroups protocol configuration-->
The NodeAddress can be also set by start parameter runn.sh -b <ip>
JGroups protocol (PartitionConfig) should be also in the channelfactory.
For DeadlockDetection and StateTransfer I don't know it, but I suppose this will be a infinispan configuration, see cluster/infinispan...
Maybe here someone other can give you an answer.