----- Eric Thompson wrote: -----

>My thought was that if I set both to 100, then a migration would not
>take place ever.  I would then depend on my secondary pool to accept
>any unexpected "excess data".  (which shouldn't occur, since we size
>our primary storage pools (disk) to handle all data, but occasionally
>we get an unexpected influx that we want to overflow to tape
>temporarily until it can be "looked at").

We have occasionally set up this kind of arranagement unintentionally
when migration processes started with some space still left in the
disk pool but the disk pool filled up completely before TSM could
mount tapes and move significant amounts of data. The results were
decidedly unpleasant. All the servers that were performing backups
at the time ended up queuing up for access to the tape drives. In
some cases backups ended up finishing hours later than usual.