CA IDM cluster sync issues

[Mahesh Velde]

Hello all,

 

We are facing an issue while restarting the CA IDM servers which are in cluster. May be because of this issue we are seeing sync issues as well in the cluster. Any one has faced this issue earlier or any solution to solve the cluster issue.

Please find the logs from server.log and we are using Jboss wildfly 8.1 version.

 

2018-08-28 14:09:46,838 WARN  [org.jboss.weld.Event] (MSC service thread 1-5) WELD-000411: Observer method [BackedAnnotatedMethod] com.sun.jersey.server.impl.cdi.CDIExtension.processAnnotatedType(@Observes ProcessAnnotatedType<Object>) receives events for all annotated types. Consider restricting events using @WithAnnotations or a generic type with bounds.
2018-08-28 14:09:50,052 WARN  [org.hornetq.core.server] (ServerService Thread Pool -- 60) HQ222116: unable to start broadcast group bg-group1: java.lang.IllegalStateException: channel is closed
 at org.jgroups.JChannel.checkClosed(JChannel.java:902) [jgroups-3.4.5.Final.jar:3.4.5.Final]
 at org.jgroups.JChannel._preConnect(JChannel.java:522) [jgroups-3.4.5.Final.jar:3.4.5.Final]
 at org.jgroups.JChannel.connect(JChannel.java:284) [jgroups-3.4.5.Final.jar:3.4.5.Final]
 at org.jgroups.JChannel.connect(JChannel.java:275) [jgroups-3.4.5.Final.jar:3.4.5.Final]

 

 

 

2018-08-28 14:09:50,060 WARN  [org.hornetq.core.server] (ServerService Thread Pool -- 60) HQ222117: unable to start cluster connection my-cluster: HornetQInternalErrorException[errorType=INTERNAL_ERROR message=HQ119004: Failed to initialise session factory]
 at org.hornetq.core.client.impl.ServerLocatorImpl.initialise(ServerLocatorImpl.java:390) [hornetq-core-client-2.4.5.Final.jar:]
 at org.hornetq.core.client.impl.ServerLocatorImpl.start(ServerLocatorImpl.java:619) [hornetq-core-client-2.4.5.Final.jar:]
 at org.hornetq.core.server.cluster.impl.ClusterConnectionImpl.activate(ClusterConnectionImpl.java:648) [hornetq-server-2.4.5.Final.jar:]
 at org.hornetq.core.server.cluster.impl.ClusterConnectionImpl.start(ClusterConnectionImpl.java:385) [hornetq-server-2.4.5.Final.jar:]

[Scott Owens]

This issue is related to Jboss wildfly-8.2.1 implementation in cluster mode.  It is my inderstanding this is not caused directly by the CA IM component.

 

Please find the following solution.  Let us know.

 

https://access.redhat.com/solutions/293823

 

Thank you.

[Mahesh Velde]

Thanks for your response. I do not have the partner access to access the above url. Also, I have checked with my team they confirmed me that the configuration are same as mentioned in the link. Is their any other I have to check.

[LarryK]

A common problem with JBoss/Wildfly clustering is cross-cluster interference--i.e multiple clusters, like one for Development, another for Production, but both have the same cluster naming in their configurations. All of the cluster artifacts should be unique to each cluster. It's not sufficient to just have unique cluster IP or node names. Make sure jgroups-channel, broadcast-group name, discovery group name,  etc. are all uniquely named for each cluster. Your application server support team or JBoss/Wildfly communities will be able to provide more focused support for clustering issues.

[Sai Kumar Valluri]

Hi Mahesh,

 

Trying to check if you're able to fix the mentioned issue and what are the steps you followed. 

 

Thanks,

Sai