Diskpool Full

[OKTSMGURU21]

I have a diskpool that is sitting at 99.9 Pct Util..........

What will happen when it hits 100 ?

Will TSM crater ?

How can I avoid this ?

HELP !!!!!

TSM 5.3.3.0
AIX 5.2.8

 

[BBB]

Post the output of: q stg <stgpool> f=d.

It depends if you have caching turned on or off. If caching is off, check your highmig is set to something less than 100% to start migration to the nextstgpool and that nextstgpool is read/write and has enough scratch volumes etc.

If caching is on, this is normal.

 

[tmcquaig]

If you have a nextstgpool with scratch volumes or filling volumes you will be fine. Data that is coming in from your backups will overflow into the nextstgpool which is usually tape.

As BBB said if your caching is on this is normal.

If caching is off, and you have no nextstgpool and your diskpool hits 100% you will have problems. TSM won't crater but you will not backup data.


I think you have caching turned on.

Posting a "q stg <stgpool> f=d" for the diskpool will display the answer.

 

[OKTSMGURU21]

tsm: CORE_TSM>q stg dirtree_d f=d

Storage Pool Name: DIRTREE_D
Storage Pool Type: Primary
Device Class Name: DISK
Estimated Capacity: 8 G
Space Trigger Util: 93.2
Pct Util: 93.2
Pct Migr: 73.1
Pct Logical: 96.4
High Mig Pct: 90
Low Mig Pct: 70
Migration Delay: 0
Migration Continue: Yes
Migration Processes: 1
Reclamation Processes:
Next Storage Pool: INCR_T
Reclaim Storage Pool:
Maximum Size Threshold: No Limit
Access: Read/Write
Description: Directory trees on disk
Overflow Location:
Cache Migrated Files?: Yes
Collocate?:
Reclamation Threshold:
Offsite Reclamation Limit:
Maximum Scratch Volumes Allowed:
Number of Scratch Volumes Used:
Delay Period for Volume Reuse:
Migration in Progress?: No
Amount Migrated (MB): 1,954.58
Elapsed Migration Time (seconds): 309
Reclamation in Progress?:
Last Update by (administrator): CARLOS
Last Update Date/Time: 05/21/03 11:09:37
Storage Pool Data Format: Native
Copy Storage Pool(s):
Continue Copy on Error?:
CRC Data: No
Reclamation Type:

 

[moon-buddy]

The diskpool should off load to the next defined storage pool, if one is defined, which is normally the tape backup. If none is defined, the TSM will cease to operate.

If you don't want to see 99.9 or so percent full after the data has been migrated, turn off caching.

 

[may]

when Caching is ON, the output which you have to surveil is "Pct Migr: xxx"

here you have Pct Migr: 73.1, so you need to migrate the datas to your Next Stgpool to gain some space on your Diskpool

 

[pstacy]

Is there a way to get it to wipe out cached content from DISKPOOL? I once experimented with it one day some time ago and turned it off the next day. What was cached on that night has been slowly clearing out but I'd like to get rid of it completely so that when I do a command like "q stg" I can see actual stats.

We do have some CDP installations in our environment so there can be new content coming to our DISKPOOL at any given moment so I don't want to risk losing any new content.

Thank you.

 

[may]

why not just delete and recreate the diskpool volume once you disabled cache?

 

[pstacy]

why not just delete and recreate the diskpool volume once you disabled cache?

Since our vendor did much of the inital setup of the system, I was hesitant to do anything that might make a mess of things.

It's not a big deal to me. I was just wondering if there might be a simple command I could use for flushing out cache content.

 

[tmcquaig]

You can update the stgpool to cache=no. Then the next migration that hits it will clean the data out of the cash while moving it to next stg pool.

To reenable the cache simply update the stgpool to cache=yes.

 

[PJ]

I'm afraid it won't. The only way to really get rid of cached versions is to delete the volumes (with discardd=y - always a little frightening) and define them back in again.

PJ