Richard,

as you mention you have to take care of the expiration that archivecontents
will have during their lifecycle. Therefore you have to query every
accountiong period for the actual size of the archives. I do not see any
possibility with perl since the commandline does not show estimate retrieve
size (as the gui does).

Kind regards
Stefan Holzwarth

-----Ursprüngliche Nachricht-----
Von: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] Im Auftrag von
Richard Sims
Gesendet: Montag, 8. August 2005 16:56
An: ADSM-L AT VM.MARIST DOT EDU
Betreff: Re: Select for archive size

On Aug 8, 2005, at 10:31 AM, Stefan Holzwarth wrote:

> Hi *SM-ers,
>
> some time ago i setup a dedicated TSM node for archiving of all
> kind of data
> from different users.
> Each archive is generated with an (internal) unique ID in the
> description
> field under the same "archive-node". Sometimes the different
> archives come
> from the same filespace because we use a central networkshare for
> collection
> of files to archive.
> Now our users should pay for their archives. So i started searching
> for an
> select statement that gives me the size of an single archive (as in
> the gui
> when you estimate retrieve size).
>
> The problem is that the archives table shows no size information
> and the
> dsmc.exe has no estimate retrieve for automation...

Stefan -

There is no inherent capability for listing stored file sizes from
TSM server queries: what it shows, in the Contents table FILE_SIZE is
the Aggregate size; and that equals the file size only if the file is
as large as or larger than an Aggregate. You'll need to acquire sizes
via client-side queries, as for example performed through a Perl script.

The TSM accounting records could be used to charge for Archive and
Retrieve, in that they record username; but there is no like
accounting of deletion/expiration, to balance the books from that
viewpoint.

On a go-forward basis, it may be feasible in your environment to have
all archiving occur through an interface script which would charge
per observed file sizes involved in Archive, Retrieve, and Delete.

    Richard Sims