[rear-users] Restoring mdadm system to a single-disk destination fails

Les Mikesell lesmikesell at gmail.com
Tue Jul 24 17:19:03 CEST 2012

On Tue, Jul 24, 2012 at 9:10 AM, Dag Wieers <dag at wieers.com> wrote:

> And personally, I don't think spending a lot of effort on the case where
> we restore to smaller devices is time well spent. Because it is totally
> acceptable to require same or larger storage when recovering or migrating.
> And in the case of migrations you can plan ahead and resize filesystems
> and devices prior to migration. (LVM)

Maybe...  I think there are many scenarios where the choice of the
target system will change between the time the backup is made and the
restore is done.  Or maybe you want to clone to a variety of targets,
or spin up a copy in a small test VM.

> But again, all the 'smooth' and 'effortless' recovery is exactly what
> Relax-and-Recover is for. In the best scenario, no intervention is needed
> to restore a server (except for initiating the restore and rebooting the
> system) and in those cases where it is not the case already, we need to
> improve Relax-and-Recover.

That's always great when it works.  When it doesn't the more
hidden/automated the attempted changes were, the harder it is going to
be to fix manually.

   Les Mikesell
     lesmikesell at gmail.com

More information about the rear-users mailing list