In order to migrate from my current DRBD+OCFS2 2-node setup to a 2-node GlusterFS setup, I would need to do the following: * extract one node from my old DRBD+OCFS2 cluster (easy) * reinstall it as as a new single-brick-based GlusterFS storage server (easy) * rsync data from my old (now one-legged) DRBD+OCFS2 storage to new Gluster storage server (easy) * reinstall second node from DRBD+OCFS2 cluster as a Gluster replica I see contradictory postings regarding the possibility of adding a replica (some links below). What I've tried is this: # on fiori2, the first GlusterFS node: mkdir /vol/brick0/vmimages gluster volume create vmimages fiori2:/vol/brick0/vmimages gluster volume start vmimages mount fiori2:/vol/brick0/vmimages /mnt rsync -a <my-old-drbd+ocfs2-cluster-remaining-node>:/.../ /mnt/ # still on fiori2, try to add the second node's brick to the volume: gluster peer probe torchio2 gluster volume add-brick vmimages replica 2 torchio2:/vol/brick0/vmimages This reports success, but I see no evidence of success; I expected to see a synchronisation process started, but this doesn't happen. Attempts to rebalance, just tell me it is not a distributed volume (correct, it's meant to replicated, but not distributed). I also tried creating the volume specifying that there are (meant to be) two replicas, but providing the name of only one of them: gluster volume create vmimages replica 2 fiori2:/vol/brick0/vmimages but this also failed, telling me I needed to specify more bricks. So my questions are: Is it possible to increase the number of replicas after the volume has been initialised? If yes, then how? Link please. Many thanks! Alexis PS Links: https://github.com/matiu2/rackspace-articles/blob/master/glusterfs/how-to-add-and-remove-a-node.md (right at the bottom) http://servertopic.com/topic/xA0e-glusterfs-how-to-add-replica-brick-to-system (right at the bottom) http://prefetch.net/blog/index.php/2011/11/27/some-interesting-insights-on-the-gluster-replicated-volume-replica-value/ https://www.gluster.org/pipermail/gluster-users/2011-March/006905.html (says you can't) _______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://www.gluster.org/mailman/listinfo/gluster-users