Description of problem: glusterd: 'gluster peer probe <hostname/ip>' - able to add peer which is already part of cluster Version-Release number of selected component (if applicable): 3.4.0.6rhs-1.el6rhs.x86_64 How reproducible: always Steps to Reproduce: 1.from one RHS node peer probe all other RHS node [root@cutlass ~]# gluster peer probe fred.lab.eng.blr.redhat.com peer probe: success [root@cutlass ~]# gluster peer probe fan.lab.eng.blr.redhat.com peer probe: success [root@cutlass ~]# gluster peer probe mia.lab.eng.blr.redhat.com peer probe: success 2.verify peer connection from all peers using 'gluster peer status' 3.from another RHS node again peer probe existing peer [root@fred ~]# gluster peer probe cutlass.lab.eng.blr.redhat.com peer probe: success Actual results: peer probe is successful. Expected results: It should fail with error message saying that peer is already part of cluster Additional info:
Rachana, Notice that this would work even though its part of the cluster because we would update the 'hostname' part of the peer list from that RHS node. To verify, please check 'peer status' before and after the 'peer probe'. Currently its working as designed, so marking it as not a bug.