On Tue, Jul 01, 2008 at 07:03:26AM -0400, Bryan Kadzban wrote: > If you mean taking a snapshot and leaving it around forever *as* the > backup, then that will work -- but the more backups you have, the slower > LVM gets (currently). This is because of the copy-on-write code: every > time the root logical volume changes, all of the snapshots need to be > updated. If you only have one or two snapshots, this isn't so bad, but > if you had (say) 20, writing to your real root LV would slow to a crawl. > > There is some code hanging around somewhere that I *think* is supposed > to handle this better, but I don't know the status... You may be referring to ddsnap, go to http://code.google.com/p/zumastor/source/browse and look at svn/trunk/ddsnap. And you can check out and build from google code too of course. I find it difficult to tell where this is going though, there doesn't seem to be disagreement between the people who maintain ddsnap and the people who maintain lvm2, but I haven't found any statements about what the way forward is. Any comments from lvm2 maintainers? -- Dan Shearer dan@shearer.org _______________________________________________ linux-lvm mailing list linux-lvm@redhat.com https://www.redhat.com/mailman/listinfo/linux-lvm read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
