This feature adds new ability to use ClusterIP and NodePort services instead of OpenShift Route in DVM. It helps utilize the host network for data transfer when both source and target clusters are within the same network. It is implied that nodes in the target cluster are reachable either via their internal IP addresses or internal DNS names. That said, the default behaviour of using Routes does not change. The user has to set an explicit flag to enable using ClusterIP or NodePort services.

To test, pgaikwad1@redhat.com set source and target to the same cluster to simulate clusters within same network and rhn-engineering-jmontleo tested it in his home environment which has two different OpenShift clusters within the same network. Both tests were successful. We also tested it in customer environment which was successful too.

Upstream doc is available here +https://github.com/migtools/mig-operator/blob/master/docs/usage/RsyncConfiguration.md#choosing-alternate-endpoint-for-data-transfer

Note to QE: You may want to find out whether you can deploy two different clusters within same network. If that's not possible, we can work out an alternative verification strategy. Let pgaikwad1@redhat.com know.