hi Hareem, When the NODE1 comes back up it will automatically be part of cluster and the self-heal of the files will happen whenever they are accessed. If you want to sync the files manually, follow the instructions in http://europe.gluster.org/community/documentation/index.php/Gluster_3.1:_Triggering_Self-Heal_on_Replicate Pranith. ----- Original Message ----- From: "Hareem Haque" <hareem.haque at gmail.com> To: gluster-users at gluster.org Sent: Tuesday, March 8, 2011 7:37:15 PM Subject: Glusterfs cluster recovery and healing setup. Can someone please kindly help in the following: How best to deal with node failures and cluster repairs. I have a problem. I setup 4 node distribute replicated cluster. Node1-4 Node1: 192.168.1.100 Node2: 192.168.1.101 Node3: 192.168.1.102 Node4: 192.168.1.103 On Node1 is initiated the peer probe command and formed the storage pool (Node1 being hypothetical master). I mounted the glusterfs native client on a machine. Using Node1 is used to mount the glusterfs onto the client machine. Now say NODE1 fails and goes offline. How do i add it back to the cluster and get the cluster back to its 4 node strength. Do i need to do rebalance or something else. Your help is greatly appreciated. Best Regards Hareem. Haque _______________________________________________ Gluster-users mailing list Gluster-users at gluster.org http://gluster.org/cgi-bin/mailman/listinfo/gluster-users