I am not using replicate, so that won't help me. On Feb 4, 2011, at 5:23 PM, Roberto Franchini wrote: > On Fri, Feb 4, 2011 at 3:28 PM, Thomas Wakefield <twake at cola.iges.org> wrote: >> What's the best process for removing a brick from a gluster setup running 3.0 (possibly getting upgraded to 3.1 soon)? >> >> We have 32 bricks, over 8 servers, and need to start thinking about how we will age out the smaller disks in favor of larger disk sizes. > > We a 6 servers cluster in distribute/replicate running on 3.0.5. > Yesterday we drop a node in favor of a a new one. > So we umount and the stopped all the gluster servers, then modify the > client vol file to point the new server instead of the old one and > then restart/remount the cluster. > At the beginning the new node was empty, so we did a ls -lR on a dir > on the gluster storage to see the new node filled with data. > Hope this help, cheers, > RF > -- > Roberto Franchini > http://www.celi.it > http://www.blogmeter.it > http://www.memesphere.it > Tel +39.011.562.71.15 > jabber:ro.franchini at gmail.com skype:ro.franchini
