I wouldn't do an INCREMENTAL !  It will expire a heck of a bunch of
     files and back up a heck of a bunch not needed....

     NEVER, NEVER, NEVER, NEVER pretend to be another node and perform an
     INCREMENTAL !   (very ugly)

     also this is why you shouldn't let your nodes delete their own backups
     & archives... root on boxA gets mad at root on boxB, root on boxA
     connects to adsm server using boxB's name & password and deletes all
     archives & backups that belong to boxB.  and who gets the blame? you
     guessed it... the adsm administrator ;-)

     OK back to the problem...   ANOTHER thing would be the next time you
     run an incremental on the node that has the backups but not the real
     files.... it will be viewed as the files being deleted and thus begins
     the cycle to flush these now inactive backups from the system


     Do an archive from the owning system as that system, set the retention
     time to what ever you need.... then if you need them on the other box,
     then at that time pretend to be the owning box and retrieve the
     archives using the owning server's name and password.

     laTER
          Dwight




______________________________ Reply Separator _________________________________
Subject: changing node names during backup
Author:  ADSM-L (ADSM-L AT VM.MARIST DOT EDU) at unix,mime
Date:    10/2/97 9:19 AM


Our ADSM server is MVS and the client platform in question in AIX.  We have
reason to want to backup client one's data with client two's machine but let
ADSM think client two is actually client one by changing the node name and
password on client two to that of client one before connecting to the server to
perform the backup.  Then, if we need to restore this data, we will want to use
client one to perform the restore just as if the backup had been taken from
client one.  Has anyone else tried this?  Is there anything I need to watch for
or do special?  Thanks alot for any info you may have.