Hey Mark, On vr, 2013-08-30 at 13:04 -0500, Mark Chaney wrote: > Full disclosure, I have zero experience with openstack and ceph so far. > > If I am going to use a Ceph RBD cluster to store my kvm instances, how > should I be doing backups? > > 1) I would prefer them to be incremental so that a whole backup doesnt > have to happen every night. Others, correct me if I'm wrong, just starting to play with this feature myself. :-) You create a base-snapshot, which you re-create and back up, say, every month and then you create another snapshot, say, every day. Ceph RBD then allows you to generate a file with just the delta's between the two snapshots. You back up those delta's and then either throw away the daily snapshot, or keep it around for online restores, if you want. In case of catastrophic failure of your RBD-pool, you first recover the base-snapshots and then merge in the most recent daily delta's for that month to get at your most recent state. See: http://ceph.com/dev-notes/incremental-snapshots-with-rbd/ > 2) I would also like the instances to obviously stay online during the > backup No problem. Before starting your backup of an image, you first make a snapshot of it and then backup the snapshot instead. Assuming you use a journaled filesystem on your VMs, your general data should be consistent in almost all cases. Specific applications, like databases, might still not recover very well and may need to be put in a consistent state within the VM before generating the snapshot. Snapshot-generation on RBD is very fast and when finished, you can resume the VM and do your backups off the snapshot. Test what happens if you try to restore and use your backups. Regularly.. :-) > 3) Backups will be stored off the ceph cluster on slower sata drives on > another storage server. Sounds like a good idea. Store it off-site or at least in another location in the building, if possible. Regards, Oliver _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |