> Turns out that wasn't the only problem I faced in my migration. With 2 >KVM servers, both sharing a volume mounted via NFS for VMs, I migrated >all VMs to the second node, upgraded the first, them moved them all back >to KVM1. Instant disk corruption on all VMs. Boom. I've never used live migration, but it's my understanding that you place the VM's virtual drive file on a sharable file system (as you have done with NFS), and that the server you migrate to has access to the file. So, this doesn't actually involve moving the virtual drive's file, it just hands off the running the VM to another server. Is that correct? If that's so, I don't see how it becomes corrupted. I back up my VM's by shutting down the VM and copying the image file to other locations, and this has never resulted in file system corruption...