AW: Reclamation of COPY-POOL-Volumes

Thanks,
but I think it is some more complicated :
because it works with server-to-server connection for every client the
server creates a new logical volume (on the source server) each day a 'copy
stg_pool' is started. On the target server all logical volumes (of all
clients)  of one 'copy stg_pool' session  are stored on one tape until it is
filled up. So I do want collocated data but I do want some reclamation where
the lets say 10% of valid data of more then one volume are stored together
on one new logical volume (e.g. delete 5 volumes create 1 new). 
If this is handeled by putting 'reclamation threshold' to 90% for copy_pool
it's ok, but what it looks like on my server is that he creates one new
logical volume for one reclamed old one (so for the target server
reclamation works, but on the source server the number of logical volumes
increases from day to day).

Christoph

> -----Ursprüngliche Nachricht-----
> Von:  Alex Paschal [SMTP:AlexPaschal AT FREIGHTLINER DOT COM]
> Gesendet am:  Freitag, 22. Juni 2001 20:10
> An:   ADSM-L AT VM.MARIST DOT EDU
> Betreff:      Re: Reclamation of COPY-POOL-Volumes
> 
> For collocated pools, if there are available scratch tapes, a single node
> is
> assigned to a tape.  Only when you no longer have available scratch tapes
> does TSM start "stacking" nodes on a tape.  If you want to minimize tapes
> in
> your copypool, I would suggest either not collocating or setting a low
> enough maxscratch number to force TSM to start stacking multiple nodes on
> a
> tape.  For more informaiton, the Admin Guide has a section describing tape
> selection criteria and precedence.
> 
> Alex
> 
> -----Original Message-----
> From: Christoph Pilgram
> [mailto:Christoph.Pilgram AT BC.BOEHRINGER-INGELHEIM DOT COM]
> Sent: Friday, June 22, 2001 5:33 AM
> To: ADSM-L AT VM.MARIST DOT EDU
> Subject: Reclamation of COPY-POOL-Volumes
> 
> 
> Hy all,
> 
> we have two backup-server, where we write our data to a tape-library. For
> all our tape-pools we have defined one copy-pool which links to the other
> server where the data are written to a separate tape-pool (pool_extern).
> We
> have defined collocation for all our pools. When I now run reclamation on
> the copy_pool I see, that for all copy-pool-volume the server deletes, he
> defines a new one where the rest of the data are transfered.
> This seems to be logical (because the reclaimed volumes are not empty and
> the rest of the data has to be handeled like this because collocation is
> on)
> but what I wanted is to reduce the number of volumes in the copy-pool. How
> can I manage this and what is the real right way to reclaim data in a
> copy-pool (or do I have to set the 'Maximum Scratch Volumes
> Allowed'-Parameter in the Copy-Pool-Definition to unlimited ?).
> 
> Thanks for help
> 
> Christoph