Simon,
Assuming you have a copy on a server drive or CD somewhere with the
version of Data Ontap you're running, all you need is boot floppies (for
non-flash units), and a copy of the config files (rc, hosts, exports,
cifsconfig.cfg, lclgroups.cfg etc). Providing you keep a backup of the
config files - many of my customers simply replicate the files in the /etc
directory without all the subdirs to a folder on a couple of other servers. The
config files all used to fit on a floppy.
In a
full blown DR scenario you can then have the root volume online
and ready within a few minutes, unless disks need to be zero'd then it
might take 20-30 minutes. You can than start restoring the rest of the root
volume (home directories, workgroup data etc).
The
only time this wouldn't work is if you used "snap mirror to tape" for bare-metal
restores rather than regular NDMP backups.
regards,
Alan.
The
only benefit as I see it to having the smallest possible vol0 is recovery
time. You can back up and recover the small amount of data in a few minutes on
most tape devices, meaning you are up and running fairly quickly in the event
of losing the volume.
However, there are methods of replicating the /etc
data to another volume, then telling the filer to boot this as vol0. This
would of course mean any NIS maps etc are broken until you change their paths,
but the filer would be up and running quicker than if you were to have to
restore the whole root volume first, and you would not be using two 72Gb disks
for 40mb of data.