The server searches through *available* (having status read-only or
read-write, and checked in a library) copypool volumes in alphabetical
order. You can also override which copy to be used by specifying "restore
stg <primary_pool> copy=<desired_copypool>".
If one of the copies is not checked in or the volume is not in readable
status, there is no choice and server attempts restore from the other
copy.

Zlatko Krastev
IT Consultant






Thomas Denier <Thomas.Denier AT MAIL.TJU DOT EDU>
Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
23.10.2003 22:08
Please respond to "ADSM: Dist Stor Manager"


        To:     ADSM-L AT VM.MARIST DOT EDU
        cc:
        Subject:        Choosing volumes for DR restores


We are in the process of reorganizing some of our storage pools. We
are using 'move nodedata' to move primary storage pool contents to
new storage pools. We are running 'backup stgpool' commands to create
new offsite copies in corresponding new copy storage pools. We are using
'delete volume' and 'backup stgpool' to clean up the copies left behind
in pre-existing copy storage pools. It is not clear that we will be able
to finish the copy storage pool cleanup before our next disaster recovery
test. If we don't finish by then our reconstructed server will have
access to two copies of some client backup files. How does TSM decide
which copy to use in this situation?