Hello

Recently, we had a situation where jobs from one class (Unix Oracle servers) was
running long for some reason thereby causing the jobs from another class (NT
servers) to queue up.   I suggested that we could adjust the Concurrent number
of jobs per class settings (currently at the default 99) down to a value that
would be a fraction of the number of drives times allowed multiplexing value;
the idea being to prevent any single class from monopolizing all drive resources
and allowing a more equitable use of the drives.   However, this was met with
some opposition from the sysadmin who set up the classes in that he says that he
tried that in the past and all that happened was the jobs did not queue but
rather exited with a drive unavailable code of some sort.

Can someone explain what the behaviour really is or supposed to be and whether
there's anything that might have caused the undesired behaviour of jobs exiting
instead of queuing?

We're running Netbackup Version 3.2GA, Unix media server, Unix Master server, NT
slave servers, ADIC Scalar 1000 tape library.

Regards,
John I Wang
Sr. Systems Engineer
Steverson Information Professionals

---
Enron Broadband Services
3 Allen Center, Room 337C
PH (713) 345-6238