Hi ..., I have a couple of queries on replacing completely failed Gluster servers/nodes, refer - http://community.gluster.org/q/how-do-you-replace-a-gluster-node/ and http://community.gluster.org/q/a-replica-node-has-failed-completely-and-must-be-replaced-with-new-empty-hardware-how-do-i-add-the-new-hardware-and-bricks-back-into-the-replica-pair-and-begin-the-healing-process/ 1. If Gluster is configured in 'distribute' mode, and there is no replication, what happens if a node goes down completely when there is a 'write' going on? What are the chances of corruption? 2. Say I have 2 Nodes in the Gluster Storage, each with an external disk array, can I setup a 3rd one as failover to any 1 of the first 2 nodes, using a backup of '/etc/glusterd', and renaming the 3rd node to that of the failed node? The 3rd failover node will have direct access to the HDDs of both the 1st and 2nd nodes and will serve the same sub-volumes as the failed node. Will this configuration work? If this is possible, I would like to automate this process using scripts+heartbeat/ucarp. Regards, Indivar Nair -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20121021/88bb5d40/attachment.html>