I have occasionally noticed this with my TDP for Oracle sessions.  The
nodes are in their own domain, management class, no limit on size on the
device class, and space left in the disk pool.

The only reason I ever notice is because we use 4 channels in the scripts,
and if there is a tape write error, the job fails, since one of the
channels reports a write-block error, and it can't restart the session.


I guess the only thing I could offer is whether you validated and activated
the management classes (it's got me before;-)

good luck!

lisa



             William Sherrill
             <billsh AT US.IBM DOT CO
             M>                                                         To
             Sent by: "ADSM:           ADSM-L AT VM.MARIST DOT EDU
             Dist Stor                                                  cc
             Manager"
             <[email protected]                                     Subject
             .EDU>                     Backups bypassing disk pool


             07/30/2003 10:29
             PM


             Please respond to
             "ADSM: Dist Stor
                 Manager"
             <[email protected]
                   .EDU>






     I have some clients (Windows 2000) that are backing up straight to my
tapepool instead of the disk pool.  My server is AIX 5.1.6.2.  and most of
the windows clients are running at 5.1.5.9.  The clients are associated to
a management class that's default management class should cause the backups
to go to the disk storage pool.  There is a management class within the
domain that directs the backups to the tape pool, but I do not have it
included in the dsm.opt file, in fact I do not have any management class
defined, meaning that it should go to the default.  Has anyone experienced
this problem?  Also, I tried an incremental of the c drive with the disk
storage pool empty and it's estimated capacity is 38 gb and it still called
for a tape.  The capacity of the c drive of the client is under 1.5 gb.

Thanks,
Bill Sherrill