pct_utilized not showing correctly

guidocaffa

ADSM.ORG Member
Joined
Jul 11, 2019
Messages
11
Reaction score
0
Points
0
PREDATAR Control23

I'm traying to move data from one tape to another. The target tape pct_utilized is 40%, but I have the following error:

1565006974143.png

ANR2017I Administrator XXXXXX issued command: MOVE DATA 000050L7

ANR1149W Move data process terminated for volume 000050L7 - insufficient space in target storage pool.


This is not the first time we have this issue, but we don't know why it's continue happening.
 
PREDATAR Control23

could you please check if the collocation is enabled for target STGPOOL ?
may be you don't have scratch tapes, check for the tapes availability," q libvol"
there may be many causes for that
 
PREDATAR Control23

Yes, collocation is enabeled. The pool should not be configured in this way, we will modify it and do some tests. I'll keep you informed.
 
PREDATAR Control23

could you please check if the collocation is enabled for target STGPOOL ?
may be you don't have scratch tapes, check for the tapes availability," q libvol"
there may be many causes for that

After adding a new tape to the pool, the pct_utilized has change to the "correct" percent..... I don't know why this is happening or how can I improve this...
 
PREDATAR Control23

If the source tape has data from one node and the target tape has data from another node, thenit may not be possible to use the target since collocation is enabled.

By adding new tape, the move command can now see a "free" tape to write to.

If you want to maximized tape use, specially for primary tapes, trun OFF collocation. This also goes to the copy (DR) tapes.

Just be warned that if you do DR and the DR tapes are needed to do restores for all nodes, you will end up is a waiting game regardless of how many tape drives you have. This is because many node data will be on a single tape. TSM will wait until that node is restored before moving on to the next node to restored.

For fast restores in DR the solution is really collocation for both primary and DR tapes. The price to pay is using more tapes.

This is all up to you.
 
PREDATAR Control23

If the source tape has data from one node and the target tape has data from another node, thenit may not be possible to use the target since collocation is enabled.

By adding new tape, the move command can now see a "free" tape to write to.

If you want to maximized tape use, specially for primary tapes, trun OFF collocation. This also goes to the copy (DR) tapes.

Just be warned that if you do DR and the DR tapes are needed to do restores for all nodes, you will end up is a waiting game regardless of how many tape drives you have. This is because many node data will be on a single tape. TSM will wait until that node is restored before moving on to the next node to restored.

For fast restores in DR the solution is really collocation for both primary and DR tapes. The price to pay is using more tapes.

This is all up to you.

Thanks for your response. What I don understand is why after adding a new tape, the target tape "000042L7" has change the pct_utilized 40,3 -> 97,9. This is the main problem I'm traing to get throught. Between the two pictures no backup where made.

First:
1565050116004.png
Then:
1565050010284.png
 
PREDATAR Control23

Thanks for your response. What I don understand is why after adding a new tape, the target tape "000042L7" has change the pct_utilized 40,3 -> 97,9. This is the main problem I'm traing to get throught. Between the two pictures no backup where made.

First:
View attachment 1580
Then:
View attachment 1579

Did you check that there was no data movement after you added the scratch tape?
 
PREDATAR Control23

Now I'm looking in actlog that the tape was mount...

2019-08-05 13:54:31 ANR1157I Removable volume 000042L7 is required for move process. (SESSION: 520544, PROCESS: 342)
2019-08-05 13:54:32 ANR1176I Moving data for collocation set 1 of 9 on volume 000042L7. (SESSION: 520544, PROCESS: 342)
2019-08-05 13:54:32 ANR1176I Moving data for collocation set 2 of 9 on volume 000042L7. (SESSION: 520544, PROCESS: 342)
 
Top