We need the *test commands because we have a set of reserved slots that we
don't want NBU to manage. Another requirement is knowing the tape volser
in the CAP, so that we can add any tapes not in the db before the move to
a free slot proc. We "pipe" the output of the *test cmd so the "freeze"
isn't noticable.

/Steve
---
On Fri, 16 Feb 2001 anthony.guzzi AT storability DOT com wrote:

> 
> 
> Ahhh!   A question I had myself.   You're right, there don't appear to be
> any  vm commands that tell you this so using the '*test' commands only to
> query the library (and not move anything) is a sensible use  assuming you
> keep in mind that while the '*test' commands are access in the robot,
> NetBackup routines can't.  (Don't want to lock out the NB processes, do
> we.)
> 
> As for determining when the cap becomes full,  keep in mind that you have
> to issue the command for each tape that needs to be ejected. The 'vmchange'
> command will return a status code of 221 when the CAP is full and it can't
> move a tape to the cap.  Now don't everyone go running to your NetBackup
> Troubleshooting Guide to look this status code up because this is not a NB
> 221, but a Media Manager 221.  And again we run into the lack of sufficient
> documentation from Veritas --- Media Manager status codes are not properly
> or completely documented (at least I haven't been able to find a list of
> them I was able to for the NetBackup status codes).
> 

<<snip>>