Active-Active cluster should stop / unstop / kill jobs on cluster node

Current Situation

• In an active cluster, we have more than one running member. JOC connects with the first available member. "Stop job" and other commands work only at the current "connected" cluster member.
• When a job runs in the active-active cluster it is distributed on all the cluster nodes and cannot stop/unstop/kill jobs on the cluster nodes.

Desired Behavior

• In active-active cluster stop/unstop/kill jobs should be possible on cluster nodes.