Hi Mike,
I will, of course, defer to the more real-life experiences of our
customers; but unless you have a very small number of nodes to deal with,
trying to use QUERY BACKUP for each and every node name would be a
relatively challenging effort in coordination. Off-hand, for what you want
to do, I think dumping the BACKUPS table via SELECT or ODBC driver would
be easier. That would get you all node names, file spaces, and file names
(high- and low-level names) in one big file that you can then process as
you wish.
Regards,
Andy.
Andy Raibeck
IBM Software Group
Tivoli Storage Manager Client Development
Internal Notes e-mail: Andrew Raibeck/Tucson/IBM@IBMUS
Internet e-mail: storman AT us.ibm DOT com
The only dumb question is the one that goes unasked.
The command line is your friend.
"Good enough" is the enemy of excellence.
Mike Eggleston <mikee AT MIKEE.ATH DOT CX>
Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
06/01/2004 13:52
Please respond to
"ADSM: Dist Stor Manager"
To
ADSM-L AT VM.MARIST DOT EDU
cc
Subject
Re: report of usage by file type?
Hi Andrew!
On Tue, 01 Jun 2004, Andrew Raibeck wrote:
> In addition to others' remarks:
>
> 1) Of course the admin client is available for Windows. You have to do a
> custom install to get it.
>
> 2) That said, not sure where dsmadmc fits in, since QUERY BACKUP is a
> client command, not admin.
>
> 3) You can not use a client of one type to operate with data of another
> type. Thus you can not use the Unix dsmc program to query backups for
> Windows clients.
>
> 4) There is probably little justification needed (in general) for
> excluding .mp3 files from backup, though I can understand why you'd want
> to try to quantify the savings.
>
> 5) Another approach would be to do a SELECT * FROM BACKUPS from an admin
> client. Be sure to redirect the output to a file on a disk with plenty
of
> space, and consider using the options -comma -dataonly=yes. Do this when
> the server is not too busy and be prepared to wait a long time for it to
> run. But that will get you info for all backup versions in the system
that
> you can then "massage" to get the info you want. Alternatively, you can
> use a Windows ODBC application to load the data via the TSM ODBC driver,
> though for a large TSM server database you'll want an application
capable
> of storing millions of records. With either the admin or ODBC driver,
you
> can get the data all at once, then mine it to your heart's content.
I want to put together a script that can query all filespaces, then query
the files for each filespace. Take those files and summarize the results
into an awk-type associative array, then print the array by extension.
The querying for all current filespaces is where I would use dsmadmc. If
there is another way without having to touch each and every node, then
great.
I also do not want to write a dsm.opt/dsm.sys parser.
I see this as something that can run once a month (or so) to report on
where the money for the tape library is spent. It could be that the tape
library costs 10,000 (just a number) and 10% of the library is used to
store E-Commerce data files, then they're using 1000/month. If however
the boss sees that 5% of the library is storing mp3 files, he might be
somewhat irritated.
So, initially this exercise is to justify creating a more restrictive
inclexcl.list file. As time passes this exercise is to demonstrate what
groups are using what percent of the library.
I have the dsmc.exe file for windows, though no dsmadmc.exe. I understand
the restrictions for one client accessing another client-type's data. My
preference is to do this all on the server using 'query ...' type commands
so as to present the least load and inteferrence to the server.
Does this all help?
Mike
|
