Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Sorry to bring up that old thread again. Can I really take snapshot on node1, do lvscan on node2, mount it there an take the backup? Don't I need CLVM for that? Thanks for clarify. On 14.02.2007 17:51, Graham Wood wrote: > If you run DRBD at the bottom of the stack, it's a little more > complicated. At this point the primary node needs to be more aware of > what's going on, since (I believe) creating the snapshot has an affect > on the initial volume. There is work being done on cluster aware > snapshotting, but I can't remember if it's production ready yet. The > advantage of doing it this way, however, is that you can do all the lvm > work on the primary, and then tell the secondary to backup the created > snapshot - as opposed to needing to run tasks in a co-ordinated fashion > on the two nodes. Any restore can be done back to the device that > you've backed up - since it's on top of DRBD -- Regards, H.D.