> One of the main drawbacks I see to an LVM is the inability to upgrade > a disk. If I want to pull a 120GB drive out of the LVM and drop in a > 300 GB, it might actually be possible, but a massive in the arse to do.
This is actually a straightforward procedure and can be done with your system up and running. I recently had a disk in my LVM began to fail and I needed get all of the data off of it. I connected another drive and added it to the volume group (pvcreate followed by vgextend). Then I got the LVM to move all of the data off of the bad drive (pvmove) and then dropped that drive from my volume group (vgreduce). I was able to do all of this with the filesystem live, mounted and in use. I've had excellent results just using LVM and XFS. I can add drives and grow my filesystem while in use (just a pvcreate, vgextend, lvextend and xfs_growfs) and remove them as above. If one drive fails, I'll use the repair tools and as I don't have the volumes stripped I should only lose the files that were on that drive. But that's fine, as I keep everything I couldn't stand to lose nice and backed up. I thought about going the symbolic link way, but I really wanted one giant pool of space that all of the myth components can easily tap into. Plus, I can move my drives around, switch from one IDE controller to another, etc, and LVM figures it all out without any reconfiguration. Cheers, Jon
_______________________________________________ mythtv-dev mailing list [email protected] http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
