Not sure I understand exactly what you're doing (I am mystified by "q scr
cancel"?), but maybe this can help. A common workaround is to have the script
check a return code, then define (or change the start time of) an admin
schedule that invokes a script, which can check a return code, define an admin
schedule that invokes a script, etc.
For instance here is how I prevent a reclaim stgpool from running if the backup
for a particular client (which backs up direct to tape in that pool) is still
running. If the script finds a backup in progress, it reschedules for 2 hours
later, otherwise it runs immediately:
tsm: LOCAL_SERVER1>q script ELPROBLEMO_reclaim_offsite f=raw
/* This script checks to see if ELPROBLEMO is currently backing up before
starting reclamation on the */
/* ELPROBLEMO_OFFSITE stgpool
*/
/* If ELPROBLEMO is not currently backing up, it starts
ELPROBLEMO_reclaim_offite */
/* If ELPROBLEMO is currently backing up, this script reschedules itself in 2
hours */
/* see if ELPROBLEMO is currently backing up */
select * from sessions where client_name='ELPROBLEMO' /* and is holding a
tape? */
if (RC_OK) goto reschedule
/* ELPROBLEMO not running, so start the reclamation */
reclaim stgpool ELPROBLEMO_offsite_pool threshold=65 duration=480 wait=yes
exit
RESCHEDULE:
update schedule resched_ELPROBLEMO_recl_offsite type=admin starttime=now+2
expiration=today
exit
And here is the definition of the resched schedule that got updated:
tsm: LOCAL_SERVER1>q schedule resched_ELPROBLEMO_recl_offsite type=admin f=d
Schedule Name: RESCHED_ELPROBLEMO_RECL_OFFSITE
Description:
Command: reclaim stgpool ELPROBLEMO_OFFSITE_POOL
THRESHOLD=65 DUR=480 wait=yes
Priority: 5
Start Date/Time: 02/18/2011 11:57:56
Duration: 10 Minute(s)
Schedule Style: Classic
Period: One Time
Day of Week: Any
Month:
And you could easily change the RESCHED_ELPROBLEMO_RECL_OFFSITE to invoke
another script that checks again to see if the backup is STILL running, and do
the reschedule again, etc.
W
