Hello Nick-

What you are describing is normal behavior for multiplexing.

The scheduler has a strict set of rules for whether or not a
new job can join an active "multiplex group".  These rules
basically boil down to a number of qualities that the new
job has to share with the already-active job:

   -same retention level
   -same volume pool
   etc...

One of these less-obvious rules is that, if inline tape
copy is in use, the new job and the existing job have to
come from similarly configured ITC schedules.  This is best
illustrated by an example:

===========
scheduleA is configured to write ITC copy#1 to tape, copy#2 to disk
scheduleB is configured to write a single copy to tape
scheduleC is configured to write ITC copy#1 to disk, copy#2 to tape
scheduleD is configured to write a single copy to disk
scheduleE is configured to write ITC copy#1 to tape, copy#2 to disk
scheduleF is configured to write ITC copy#1 to disk, copy#2 to tape
scheduleG is configured to write ITC copy#1 to disk, copy#2 to tape, 
copy#3 to tape
scheduleH is configured to write a single copy to disk

Question:
==
Assuming all other aspects of the configuration would allow multiplexing
for these schedules, which of the above schedules would be allowed to
multiplex to the same media?

Answer:
==
A+E can MPX together
C+F can MPX together
B,D,G,H cannot MPX at all in this example
============

HTH
rob



On Feb 2, 2005, at 4:01 AM, Nick Toye wrote:

>
> Can anyone help with the following problem please. We are running
> Netbackup 5.0 MP3 on Solaris 8.
>
> We have 2 policies.
>
> Policy A is for a single client and it's schedules are configured to 
> run
> inline copy, with the primary copy going to a tape library, and copy 2
> going to a disk storage unit.
>
> Policy B is for all other clients and it's schedules are configured to 
> run
> a primary copy to the tape library only.
>
> The retention levels for the primary copys of the respective schedules 
> in
> both policies A and B are the same, and also for copy 2 of policy A.
> Multiplexing is set to the same across both policies, and set in the 
> tape
> storage unit (5 for each of 2 drives).
>
> The problem is as follows. The single job for policy A starts. 10 
> minutes
> later the many jobs for policy B are queued. But the tape that has been
> selected for policy A continues to run only that job, and does not 
> pick up
> the policy B jobs until policy A has finished.
>
> Nick Toye
> Systems Administrator
> ingenta plc
> 1 Riverside Court
> Lower Bristol Road
> BATH
> BA2 3DZ
> Tel: +44 (0)1225 361123
> Email: nick.toye AT ingenta DOT com
>
> **********************************************
> *  www.ingenta.com                           *
> *  Ingenta: Empowering the exchange of       *
> *  academic and professional content online  *
> **********************************************
>
> _______________________________________________
> Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
> http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
>