If you remove a node and then add it back in and the Cluster Service does not startTombstone Stuff when you put a node back inncli disk ls-tombstone-entriesReturns all the disksFor each disk do….ncli disk rm-tombstone-entry serial-number=Check until no entriesncli disk ls-tombstone-entriesCluster start Ring Statusnodetool -h localhost ring Change MTU Sizesudo vi /etc/sysconfig/network-scripts/ifcfg-eth0Add the following lineMTU=”9000″sudo service network restart Cluster balance checkncc health_checks cassandra_checks ring_balance_checkncc health_checks run_all NUTANIX DRsudo ip route add via mtu 1472sudo ip route add via mtu 1472Can we try to manually modify the routing table to specify MTU for a certain subnet from the CVM’s?sudo ip route add DESTINATIONSUBNET/24 via LOCALGATEWAY mtu 1000example: sudo ip route add via mtu 1000