Tape usage is high

[sasikala.m]

Hi,

In my TSM server, the tape is usage is getting increased day by day. But there is no collocate enabled for any stg in my environment.

There are 301 nodes got registered in my tsm server.It contains sap nodes as well.

Please help me out to sort out this, Let me know what are the pssoible steps need to take care for the same.

Much appreciated for the support.

Thanks

 

[fcosta]

Hi friend!

I think the best way to solve this problem is using a good retention strategy.

In my environment helps me a lot. When i made changes in my retention, the tape reclamation turns more efficient, and freed good amount of space.

 

[sasikala.m]

Hi friend!

I think the best way to solve this problem is using a good retention strategy.

In my environment helps me a lot. When i made changes in my retention, the tape reclamation turns more efficient, and freed good amount of space.

Hi,

Thanks for your quick reply,

but in our environment, the retentions are 3,3,60,90..and this is the customer requirements as well

CCIL_FS ACTIVE STANDARD STANDARD 3 3 60 90
CCIL_FS STANDARD STANDARD STANDARD 3 3 60 90
CCIL_SAP ACTIVE DB2_7YR STANDARD 1 0 0 0
CCIL_SAP ACTIVE DB2_DISK STANDARD 1 0 0 0
CCIL_SAP ACTIVE DB2_LOG STANDARD 1 0 0 0
CCIL_SAP ACTIVE DB2_TAPE STANDARD 1 0 0 0
CCIL_SAP ACTIVE SAP_DISK STANDARD 1 0 0 0
CCIL_SAP ACTIVE SAP_LOG STANDARD 1 0 0 0
CCIL_SAP STANDARD DB2_7YR STANDARD 1 0 0 0
CCIL_SAP STANDARD DB2_DISK STANDARD 1 0 0 0
CCIL_SAP STANDARD DB2_LOG STANDARD 1 0 0 0
CCIL_SAP STANDARD DB2_TAPE STANDARD 1 0 0 0
CCIL_SAP STANDARD SAP_DISK STANDARD 1 0 0 0
CCIL_SAP STANDARD SAP_LOG STANDARD 1 0 0 0
INFRA ACTIVE STANDARD STANDARD 3 3 60 90
INFRA STANDARD STANDARD STANDARD 3 3 60 90
ORACLE ACTIVE STANDARD STANDARD 1 0 0 0
ORACLE STANDARD STANDARD STANDARD 1 0 0 0
TEMPLATE- ACTIVE STANDARD STANDARD 3 3 60 90
_FS
TEMPLATE- STANDARD STANDARD STANDARD 3 3 60 90
_FS
TEMPLATE- ACTIVE DB2_7YR STANDARD 1 0 0 0
_SAP
TEMPLATE- ACTIVE DB2_DISK STANDARD 1 0 0 0
_SAP
TEMPLATE- ACTIVE DB2_LOG STANDARD 1 0 0 0
_SAP
TEMPLATE- ACTIVE DB2_TAPE STANDARD 1 0 0 0
_SAP
TEMPLATE- ACTIVE SAP_DISK STANDARD 1 0 0 0
_SAP
TEMPLATE- ACTIVE SAP_LOG STANDARD 1 0 0 0
_SAP
TEMPLATE- STANDARD DB2_7YR STANDARD 1 0 0 0
_SAP
TEMPLATE- STANDARD DB2_DISK STANDARD 1 0 0 0
more... (<ENTER> to continue, 'C' to cancel)
_SAP
TEMPLATE- STANDARD DB2_LOG STANDARD 1 0 0 0
_SAP
TEMPLATE- STANDARD DB2_TAPE STANDARD 1 0 0 0
_SAP
TEMPLATE- STANDARD SAP_DISK STANDARD 1 0 0 0
_SAP
TEMPLATE- STANDARD SAP_LOG STANDARD 1 0 0 0
_SAP

 

[JohanW]

Is there anything in particular that leads you to believe you're using more tapes than you're storing data? Reclamation running? Finishing?

Frankly, it'd probably take a good close look at your environment to optimise the tape usage, unless something's very obviously wrong. Difficult to do over the internet.

 

[sasikala.m]

Space reclamation,migration,expiration all the admin process are running fine.

 

[SleeperGTP]

Space reclamation,migration,expiration all the admin process are running fine.

Do you have a lot of pending tapes?

 

[sasikala.m]

Do you have a lot of pending tapes?

Thanks for your reply,

only this many tapes are in pending status

Volume Name Storage Device Estimated Pct Volume
Pool Name Class Name Capacity Util Status
------------------------ ----------- ---------- --------- ----- --------
CC0021L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending
CC0025L4 COPYPOOL LTOCLASS 0.0 M 0.0 Pending
CC0074L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending
CC0095L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending
CC0102L4 COPYPOOL LTOCLASS 0.0 M 0.0 Pending
CC0108L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending
CC0127L4 COPYPOOL LTOCLASS 0.0 M 0.0 Pending
CC0130L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending
CC0157L4 COPYPOOL LTOCLASS 0.0 M 0.0 Pending
CC0159L4 COPYPOOL LTOCLASS 0.0 M 0.0 Pending
CC0167L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending
CC0260L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending
CC0352L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending
CC0364L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending
CC0405L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending
CC0417L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending
CC0443L4 TAPEPOOL LTOCLASS 0.0 M 0.0 Pending

 

[Phil_02GT]

You should look at q auditocc and find the big nodes. You can pinpoint retention problems.
As an example, DB2 admin who comments his cleanup/expiration scripts... that happened to me.

 

[sasikala.m]

You should look at q auditocc and find the big nodes. You can pinpoint retention problems.
As an example, DB2 admin who comments his cleanup/expiration scripts... that happened to me.

retention settings are 3,3,30,60 only.

Please clearly explain about the DB2 admin who comments his cleanup/expiration scripts

have chosen the big nodes from the audit occupancy, will tht audit occupancy data talks about the tape usage??

 

[Phil_02GT]

audit occ does not show tape usage. But if you find the big nodes you will know which ones consume the most space in your tape pool.

DBA's have to cleanup the old data in TSM. If your company says that they want to retain 30 days worth for DB backup. The DBA should have something in place to clean that up.
What I meant with "it happened to me" is that a DBA stopped doing his cleanup and 1 node ended up taking near 120 TB. After our group asked him to cleanup old stuff, the node went down to 30 TB. That's something to look at.

-Run audit occ and ask yourself if it's normal that these server use so much space in your stg pools
- Check include/excludes
- Check if compression is enabled on your drives.
- etc

 

[DanGiles]

Can we backtrack a bit? You stated "there is no collocate enabled": what do you mean by that? Do you have "collocation=no" set on all your tape pools? If it shows collocation by group (the default) and you do not have collocation groups set up, then TSM will act as if you have set collocation=yes.

The other thing to check is if this tape increase is on the primary storage pools or the copy pools?

 

[sasikala.m]

Can we backtrack a bit? You stated "there is no collocate enabled": what do you mean by that? Do you have "collocation=no" set on all your tape pools? If it shows collocation by group (the default) and you do not have collocation groups set up, then TSM will act as if you have set collocation=yes.

The other thing to check is if this tape increase is on the primary storage pools or the copy pools?

we dont have any storagepools as collocation=group,
onsite and offsite pools has collocation=no.

 

[SleeperGTP]

we dont have any storagepools as collocation=group,
onsite and offsite pools has collocation=no.

Can you post a q stgp f=d?