Hi Guys,
Need your inputs on one strange problem I have in one of my environment.
I have 2 IS Clusters, 2 MWS hosts and 2 DB one for each cluster and one Broker.
Details:
------------------
IS Cluster -1: Host1:5555 & Host2:5555
MWS: MWSHost:8585
Optimize: MWSHost: 12500
-------
IS Cluster -2: Host2:6666 & Host2:6666
MWS: MWSHost:8080
Optimize: MWSHost: 12600
Broker is shared across the environment.
I have created aliases on both clustered IS as Host1 and Host2 pointing to corresponding port.
Now the problem:
· Processes are running fine on Cluster-1, but I can see the process instances on MWS which is incorrect and I am expecting processes instance to be listed only if its executed in that particular cluster. Though entries on MWS-2 shows the status of instances in broker and doesn’t have any step level detail as the steps executed on Cluster-1.
I checked Process audit database on cluster-2 and everything looks fine, ie. WMProcess don’t have any entries for the process ran on Cluster-1, but when see the process tracking database (Optimize) I can see the entries of instances which ran on cluster-1 as well in BAM_PT_Process_Instance table. I think its due to the fact broker is shared and BAM_process_intance trigger is shared and causing this entries to be written in both databases.
Is there any way I can stop this, please provide your thoughts on that.
· My second issue is in case of failed instances, when I am resubmitting the transactions on cluster-1 I can see steps are executing on both cluster’s which is kind of weird and I am kind of clueless why it’s happening. Just let me know if you can think of something here.
PS: All IS have the same code deployed, but triggers for process models are disabled in case where the process are not supposed to execute on specific cluster.
Thanks for reading this long post 
and appreciate your responses and pointers to resolve this problem.
What is the Broker prefix configured in each IS instance?
prefix for both IS on cluser-1 is Prod1
Cluster-2 is Prod2
Hi Reamon,
Did you got any suggestion for me on this 
…
thnx.
To Resolve the problem what I have done till now:
I created one new client group “analays2” on common broker… with same ACL’s as of analysis; created new environment pointed the JNDI setting to use to client group, deployed it. Corrected WmOptimize and WmPrt package setting to use new optimize url and reloaded them.
All of the above steps were fine, but when i tried to restart AE after deployment of new environment… I am getting continous errors (attached below), I can see there is no clients were created for this new client group and seems like errors are complaining for the same that we can’t have the same name for client even under different client grp.
FONT=Tahoma [Analytic Engine v8.0.3.0.JMSCommunication$ReceiverThread.jndiCreateMessageConsumerError] 2010-11-02 14:20:35.168 EDT ERROR: Error creating consumer to destination: queue/rule_kpi[/font]
FONT=Tahoma [Analytic Engine v8.0.3.0.JMSCommunication$ReceiverThread.jndiCreateMessageConsumerError] 2010-11-02 14:20:35.168 EDT ERROR: javax.jms.InvalidDestinationException: [BRM.10.2205] JMS: Cannot connect to queue “rule_kpi” with ClientGroup=analysis2. It already exists with a conflicting value.[/font]
at com.webMethods.jms.protocol.BinaryProtocolHandler.initializeSession(BinaryProtocolHandler.java:1491)
at com.webMethods.jms.impl.WmConnectionImpl.initializeReceiver(WmConnectionImpl.java:1392)
at com.webMethods.jms.impl.WmMessageConsumerImpl.(WmMessageConsumerImpl.java:107)
at com.webMethods.jms.impl.WmQueueReceiverImpl.(WmQueueReceiverImpl.java:34)
at com.webMethods.jms.impl.WmSessionImpl.createConsumer(WmSessionImpl.java:913)
at com.webMethods.jms.impl.WmSessionImpl.createConsumer(WmSessionImpl.java:860)
at sun.reflect.GeneratedMethodAccessor30.invoke(Unknown Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke(Method.java:597)
at org.springframework.jms.connection.CachingConnectionFactory$CachedSessionInvocationHandler.invoke(CachingConnectionFactory
---- Any expert advices from here, i kind of exhausted my ideas on this now----
What’s the intent of using 1 broker for both IS clusters? If the intent is to have segregated IS/MWS/Optimize groups, then create another broker on the Broker Server for the 2nd group.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.