>1. Global Max Jobs per client
>1) this is currently set to the number of drives in the library.

It should be set to 99.  No need to throttle anything here.

>2. Max Jobs per policy
>2) in Netbackup 6.0, the "Limit jobs per policy" box in the policy
>attributes tab is unchecked, which, according to the documentation,
means >that the policy can start unlimited jobs

You got it.

>Jobs per stunit
>3) This is currently set to the number of drives in the library

I think you're confusing things.  You tell it the number of drives and
you tell it the maximum multiplexing setting for each drive.  It's the
latter that I also think should be set to 99.  If you've set both values
to n, you're saying there's n drives that can each have n jobs each
simultaneously.  

>Jobs per sched
>4) I can't find any reference to this setting... where is it found? 

It's the multiplexing setting in every schedule, and it's one of two
places where I think you should throttle things.  (The other place I
like to throttle is the individual max jobs per client you can set with
bpclient.)  Each schedule can have its own mpx setting, so you can have
different mpx settings for full and incremental schedules, or local and
remote backups, for example.

So if:

1. You've got a high number in the global max jobs per client.
2. You leave the max jobs per policy alone (which is unlimited).
3. You put a high number in the max jobs per stunit.
4. You put a reasonable number in the multiplexing setting of your
scheds.

Then you'll get what you want!



_______________________________________________
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu