At the moment if you configure a NoB within OSE like:

A network of broker on Openshift with 3 broker under 1 service.
Those broker are configured like this :

<transportConnectors>
            <transportConnector name="openwire" uri="tcp://0.0.0.0:61616?maximumConnections=1000&amp;wireFormat.maxFrameSize=104857600" updateClusterClients="true" rebalanceClusterClients="true" updateClusterClientsOnRemove="true">
                <publishedAddressPolicy>
                    <publishedAddressPolicy publishedHostStrategy="IPADDRESS" />
                </publishedAddressPolicy>
            </transportConnector>
        </transportConnectors>

And on the client side utilize:

failover:(" + openshiftService + ")?randomize=true&initialReconnectDelay=700

The rebalancing does not work properly, likely due to the openshiftService URI not being new as per this line of code.

Instead what is seen are errors like this on the broker:

Transport Connection to: tcp://10.1.5.1:46684 failed: java.io.EOFException 

And this in a transacted camel route producer:

org.springframework.jms.TransactionRolledBackException: Transaction completion in doubt due to failover. Forcing rollback of TX:ID:dpl-cfg-u1-1-zp6yk-45255-1474550221616-1:361:1; nested exception is javax.jms.TransactionRolledBackException: Transaction completion in doubt due to failover. Forcing rollback of TX:ID:dpl-cfg-u1-1-zp6yk-45255-1474550221616-1:361:1

The desired behaviour is to rebalance when scaling UP or DOWN the brokers, and as well to
rebalance when a client connects or disconnects. e.g.:

If we had 6 consumer and 3 brokers at start we would have :

broker1 :
  - consumer1
  - consumer2
broker2:
  - consumer3
  - consumer4
broker3:
  - consumer5
  - consumer6

And when losing consumer 6/5/4 it would automaticaly rebalance to :

broker1 :
  - consumer1
broker2:
  - consumer2
broker3:
  - consumer3