Diskpool Migrations

[Demond]

I am having a very weird problem with migration jobs. Each morning we flush our diskpools to our primary tape pool. We kick off 6 processes per disk pool (Windows1-DP and AIX1-DP) The migration processes run fine throughout the day, however around 1 or 2pm all but one of my migration processes for AIX1-DP will stop. Even though the pool is set to highmig=5 lowmig=1. The diskpool will be around 30% utilized but the processes will just stop. I have found the the act log that these processes are starting with a "duration=no" option.....I can't find that option anywhere within TSM. Could this be why my jobs are stoping?

 

[moon-buddy]

"duration=no" option.....

Huh? What is your TSM Server version?

 

[hogmaster]

Hi
Check what admin script is starting the daily migration.
YOu can check that by issuing
q sched type=admin format=detail
this will give a list of all the jobs that are scheduled

 

[Demond]

We are running TSM Server v 5.5.4.2, working to implement a new environment that will run TSM Server v6.2.

The admin job that kicks off the migration is below,

Schedule Name: A1_MOVE_DP_TP_START
Description: 8:30 Start moving data from DP to TP
Command: UPD STG AIX1-DP HI=5 LO=1 MIGPR=5
Priority: 5
Start Date/Time: 05/22/10 08:35:00
Duration: 2 Hour(s)
Schedule Style: Classic
Period: 1 Day(s)
Day of Week: Any
Month:
Day of Month:
Week of Month:
Expiration:
Active?: Yes
Last Update by (administrator): DG1M
Last Update Date/Time: 10/15/10 09:49:28
Managing profile:

 

[TMcD]

When you are down to one process is it possible that there is only one client left in the pool. When that last process is running do a "q occ stg=AIX1-DP" and see how many clients have data left in the pool. If there are more than one, are they all in the same collocation group. I'm not sure if TSM will only do one process in that case but I know if there's only one client it will only run one migration process.