Backup Exec 9.0 has some horrible bugs related to incorrect ordering of SCSI
IDs on drives that are in a tape library. Unless you order the SCSI ID
from lowest to highest in the Device Manager, media will spontaneously start
to declare itself as cleaning media, and drives will just disappear from the
library when you start Backup Exec 9.0. Device Manager will bring the
drives back again, but as soon as you quite BE 9.0 and restart it, the
drives disappear again. Weird unexplained things happen during tape
operations as well, such as jobs will hang indefinitely, tapes will fail to
eject for sometimes hours after finishing an erase, etc.

The only way I seem to be able to get this product to work somewhat reliably
is to use less than seven devices on the HVD SCSI chain. I make the
library SCSI ID=1, and the drives are SCSI ID = 2 through 6. As soon as I
put either library or drives above SCSI ID=7, I start to get the nasty
cleaning tape and disappearing drives. The Adaptec HVD host adapter has no
problem at all seeing all of the drives on the chain, including those above
and below SCSI ID=7. It is Backup Exec that doesn't seem to be able to
handle large numbers of drives.

My question is has anyone successfully configured an HVD SCSI library with
Backup Exec 9.0 using SCSI ID 1 through 15 and a large number of tape
drives? If yes, could you please share your configuration, including are
you using multiple SCSI channels, and which IDs are you occupying on each
channel?

I'm about ready to give up on this product. It seems to be completely hit
or miss, and what is most scary is that the failure modes are insidious and
not obvious. If I got an outright failure code I could at least try
something else. But getting weird relabeling of tapes, and time-wasting
timeouts that have no corresponding error message in any log, is getting to
be more than I can take.

--
Will

NOTE: To reply, CHANGE the username to westes AT earthbroadcast.com