Hazelcast sepearted group of CAS communicating with either other question

[Andy Ng]

Hi all,

I am using CAS 5.3.x, and this problem is about Hazelcast.

We have setup 2 groups of CAS in the same local network, the 2 group shouldn't communicates with each other.

To make sure the group don't communicate, we use the following configuration.

cas.properties for group one

# Configuration of group one

cas.ticket.registry.hazelcast.cluster.instanceName=10.0.0.1 (In the other servers, this value would be 10.0.0.2 and so on)

cas.ticket.registry.hazelcast.cluster.members=10.0.0.1,10.0.0.2,...

cas.ticket.registry.hazelcast.crypto.signing.key=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX (Generated by CAS)

cas.ticket.registry.hazelcast.crypto.encryption.key=XXXXXXXXXXXXX (Generated by CAS)

cas.properties for group two

# Configuration of group two

cas.ticket.registry.hazelcast.cluster.instanceName=10.0.0.3 (In the other servers, this value would be 10.0.0.4 and so on)

cas.ticket.registry.hazelcast.cluster.members=10.0.0.3,10.0.0.4,...

cas.ticket.registry.hazelcast.crypto.signing.key=YYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYYY (Generated by CAS, different than above)

cas.ticket.registry.hazelcast.crypto.encryption.key=YYYYYYYYYYYY (Generated by CAS, different than above)

Today, when one of the server is down it also crashed the server in another group I inspect the server log. then I found that the 2 group is communicating with each other.

See the following:
============================================================================================================

2019-04-02 17:13:19,807 ERROR [com.hazelcast.spi.impl.operationservice.impl.operations.PartitionIteratingOperation] - [10.0.0.1]:5701 [dev] [3.10.6] Member [10.0.0.3]:5701 is currently passive! Operation: c

om.hazelcast.spi.impl.operationservice.impl.operations.PartitionIteratingOperation{serviceName='hz:impl:mapService', identityHash=xxxxxxxxxxxxx, partitionId=-1, replicaIndex=0, callId=15345, invocationTime=xxxxxxxxxxx (TIME), waitTimeout=-1, callTimeout=60000, operationFactory=com.hazelcast.map.impl.operation.SizeOperationFactory@xxxxxxxxxxxxx}

com.hazelcast.core.HazelcastInstanceNotActiveException: Member [10.0.0.3]:5701 is currently passive! Operation: com.hazelcast.spi.impl.operationservice.impl.operations.PartitionIteratingOperation{serviceName

='hz:impl:mapService', identityHash=xxxxxxxxxxxxx, partitionId=-1, replicaIndex=0, callId=15345, invocationTime=xxxxxxxxxxx (TIME), waitTimeout=-1, callTimeout=60000, operationFactory=com.ha

zelcast.map.impl.operation.SizeOperationFactory@xxxxxxxxxxxxx}

        at com.hazelcast.spi.impl.operationservice.impl.OperationRunnerImpl.checkNodeState(OperationRunnerImpl.java:261) ~[hazelcast-3.10.6.jar:3.10.6]

        at com.hazelcast.spi.impl.operationservice.impl.OperationRunnerImpl.run(OperationRunnerImpl.java:179) ~[hazelcast-3.10.6.jar:3.10.6]

        at com.hazelcast.spi.impl.operationservice.impl.OperationRunnerImpl.run(OperationRunnerImpl.java:405) ~[hazelcast-3.10.6.jar:3.10.6]

        at com.hazelcast.spi.impl.operationexecutor.impl.OperationThread.process(OperationThread.java:115) ~[hazelcast-3.10.6.jar:3.10.6]

        at com.hazelcast.spi.impl.operationexecutor.impl.OperationThread.run(OperationThread.java:100) ~[hazelcast-3.10.6.jar:3.10.6]

============================================================================================================

I was very confused as to why the 2 groups of CAS can communicates, above is all of the hazelcast config I added so I didn't enabled multicast.

Is there any problem with my Hazelcast Config?

Thanks in advance!

Cheers!

- Andy

[Andy Ng]

Hi all,

Found the issue, my own problem. Two of the Hazelcast server have set the same instanceName name so they crashed hard. Rookie mistake.

Now fixed and is all good.

Cheers!

- Andy