I have installed and configurated Cluster in JBoss EAP 6.3 with mod_cluster 1.2.6.
Server 1 (192.168.1.51): jboss 6.3 with node name: node1 domain master
Server 2 (192.168.1.52): jboss 6.3 with node name: node2, slave
Server 3 (192.168.1.53): installed httpd and mod_cluster.
The app is marked with the tag <distributable/> and this is my replication-config in jboss-web.xml
<replication-config>
<cache-name>web.repl</cache-name>
<replication-trigger>SET</replication-trigger>
<replication-granularity>SESSION</replication-granularity>
<use-jk>false</use-jk>
<max-unreplicated-interval>30</max-unreplicated-interval>
<snapshot-mode>INSTANT</snapshot-mode>
<snapshot-interval>1000</snapshot-interval>
</replication-config>
The cluster running fine with the demo app ClusterWebApp.war; when I shutdown the node 1, the attribute mantain its value in session.
But when I deployed my app in the Cluster, the operation is unexpected. Sometimes, Initially the app run normally: autenthicated with Active Directory and Oracle's Database, this running fine. But later app run very slow, throwing the ERROR
JBWEB001018: An exception or error occurred in the container during the request processing: java.lang.RuntimeException: JBAS018060: Exception to acquire ownership of jSAsRbXDgI+x2jnxFHlaFSO2
Caused by: org.jboss.as.clustering.lock.TimeoutException: JBAS010213: You can not acquire locks default-host/ibSuite/jSAsRbXDgI+x2jnxFHlaFSO2 of cluster
Previously to that error, the console show the WARN
WARN [org.jboss.as.clustering.web.infinispan] (Incoming-16,shared=udp) JBAS010325:Possible concurrency problem: id replicated version 25 is less than or equal to the in-memory version for session GR5VUGp4HwPzWvG8ZnI6xuGQ
I have adjusted some parameters in the configuration:
- In the Cache Container web.repl I have set the Lock Timeout from 60000 to
180000. With the value 60000 the behavior is very bad.
- In the web mod_cluster, I have set the Sticky Session and Sticky Session Remove to true.
