Vikas Gorur wrote: > Olivier Le Cam wrote: >> Hi - >> >> I already have an NFS server in production which shares Web data for a >> 4-node Apache cluster. I'd like to switch to GlusterFS. >> >> Do I have to copy the files from the NFS storage to a GlusterFS one, >> or may it work if I just install GlusterFS on that server, configuring >> a GFS volume to the existing storage directory (assuming, of course, >> the NFS server is shuuted down and not used anymore), then ls -lR? > What kind of GlusterFS configuration do you intend to use? Distribute, > Replicate, or both? At this step, I just wanna give it a try. I was actually thinking (if this is possible) to install GlusterFS on the current NFS server and switch from regular-NFS to GlusterFS (on both clients and server side). Then, if everything goes fine, add a second replicated brick (forcing synchronization). Then again, if everything is still fine, add two other bricks in order to distribute. Thanks to the "defrag" script, it should be possible to rebalance the files over the distributed storage with an acceptable disturbance. Does it make sense to go that way? In fact, I don't have all the equipment required to switch to a (replicated/distributed) 4-bricks. This is why it came to my mind to do it step-by-step. Thanks and best regards, -- Olivier Le Cam D?partement des Technologies de l'Information et de la Communication Acad?mie de Versailles
