AMQP connection failover doesn't work when connecting to AMQ Broker via OpenShift routes

As a part of workaround we can add the following dependency 

<dependency>
<groupId>org.apache.qpid</groupId>
<artifactId>qpid-jms-client</artifactId>
<version>0.61.0.redhat-00001</version>
</dependency>

After adding this dependency the failover happens as expected.

Please follow the below steps to reproduce the issue:

Step 1: Deploy AMQ Brokers on openshift and add amqp acceptor

Step 2: Expose this acceptor on the route so that an application outside openshift can connect to it.

Step 3: Use the attached Fuse project and update the amq broker URL as below

failover:(amqps://host1:443,amqps://host2:443)?failover.nested.transport.trustAll=true

Step 4: Ensure that the pom.xml contains Fuse 7.10 bom.

{{Note: Please ensure that there is no }}qpid-jms-client dependency.

Step 5: Start the application and start publishing messages.

Step 6: While the messages are being published delete the amq broker pod on which the application has made the connection and delete the pod.

Expectation: The client application should failover to the other amq broker pod.

Actual: The application keeps trying the same broker pod and there is loss of messages until the pod comes back.

{{}}