Detecting Large number of Locks

[Walden Leverich <walden@xxxxxxxxxxxxxxx>]

Does anyone know of a way to detect (other than polling) when a job is
holding an excessive number of records locks? I know there is a message
(CPI1466 - Job holds large number of locks.) that is sent to QSYSOPR, but it
appears to be sent only when the job holds an excessive number of object
locks. I receive this message during backups, with 30,000 plus locks
allocated to the backup job.
 
My problem is this: We have several new programs that are in production that
use commitment control. Unfortunately, these programs have bugs that result
in some looping. We have had jobs hold over 1.2 million record locks in a
commitment control environment. When these runaway jobs are cancelled the
system must rollback 1.2 million updates, a process that literally takes all
day. 
 
If I could detect a job that holds more than, say 1000 locks, and hold it
pending operator intervention this would solve many problems. Anyone have a
good way of getting this information (again, other than polling all jobs
every few minutes)?
 
-Walden
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---