Kubernetes server pod is evicted due to out of memory exception. Any idea how to relaunch the pod.
Can you provide some more details - have you deployed using Couchbase Autonomous Operator? If yes, then Operator should automatically self-heal by launching a new pod and rejoining the cluster. Typically, when a Couchbase instance fails, a failover operation is performed if you have a replica and auto-failover configured. Then when the instance becomes available, a full or delta rebalance operation gets performed.