Eric,  I didn't phrase that very well.  All we are doing is restoring
the files that were backed up on Server A to Server B, by running the
restore command on Server B and telling ADSM to restore files from
Server A.

In most cases where we have restored the backup of one server
to another server, we were in the process of migrating to a new,
larger server, but couldn't take the old server out of service while
doing that.  (Obviously, one last incremental backup has to be taken
at the last minute just before the old server is replaced with the
new one, and then a final restore of changed/new files must be done,
but that doesn't take long, and by then one is sure that the new
server is properly configured, etc., because it has already been run
as a full image of the old server.)

In this case, one of our servers has a damaged bindery (as the result
of a software failure not related to ADSM).  We are restoring a full
image to a test server so that we can experiment with different
approaches to mending the bindery.  (We have also captured what we
think is the last backup of that bindery before it was damaged.  It
is good to have ADSM saving several levels of "inactive" files.)

Melinda Varian,
Princeton University