The dedicated tape kind of removes the possibility of jobs remaining queued.

You might want to look at the other fix for this problem, though.  It's 
pretty easy.

You create a class that calls another class.  The class that calls the 
second class actually runs the bpstart_notify, and the second class does 
the backing up and multistreaming.

At 10:50 AM 6/21/2001 -0600, you wrote:
>The last thing in the bpstart_notify is sleep 180 (independent of stream
>number), which gives us plenty of time for our database scripts to complete.
>
>We have a dedicated oracle tape drive and do not run multiple clients in the
>same class.  It seems to me that it would be odd for some of a client's jobs
>to remain in queue at the same point in time that all the client's active
>jobs are completed.  Or at least, I have not seen that unless there was some
>other issue.  Am I missing something?
>I am relatively unfamiliar with NetBackup, I appreciate all the productive
>input.

---
W. Curtis Preston
Principal Consultant for Storage Designs, your storage experts
Webmaster: http://www.backupcentral.com Phone: 760 631 7991