Iseries object lock monitor and report
Is there a way to setup a collection service and/or report that will provide object lock information for objects that generate contention errors? We have a program that runs unattended that periodically encounters an allocation issue for a variety of files, typically due to an end user's exclusive access. By the time we see the error in the joblog, the contention no longer exists and we have no way of following up with the party that caused the problem. We use WRKOBJLCK to investigate issues presented during our attended monitoring, but there is no way to go back in time to see what caused the contention for an event that happened during our unattended hours.
Do you own Performance Tools? (GO LICPGM, option 10)
If you own the Performance Tools licensed program, and you capture performance data while the problems occur, you can run the Lock Report and see where your contention is. If you don't have Performance Tools, you can still capture performance data, but you will have to query the data yourself, or send it to a service provider or IBM for analysis.
Might be an even easier way, though:
What specific CPF message are you getting? If you check the job log in the job that gets the error, you will often see a diagnostic or escape message that includes the name of the locking job - all you have to do is look at the second level help text (cursor on the message and press F1). If you are seeing this in a printed job log, change the jobd or sbmjob, or schedule entry that starts the job to LOG(4 0 *SECLVL) to force second level message text to be printed to the spooled job log.
- Gary Patterson
If you own the Performance Tools licensed program, and you capture performance data while the problems occur, you can run the Lock Report and see where your contention is. If you don't have Performance Tools, you can still capture performance data, but you will have to query the data yourself, or send it to a service provider or IBM for analysis.
Might be an even easier way, though:
What specific CPF message are you getting? If you check the job log in the job that gets the error, you will often see a diagnostic or escape message that includes the name of the locking job - all you have to do is look at the second level help text (cursor on the message and press F1). If you are seeing this in a printed job log, change the jobd or sbmjob, or schedule entry that starts the job to LOG(4 0 *SECLVL) to force second level message text to be printed to the spooled job log.
- Gary Patterson
ASKER
Gary,
The programs in question are out of our control, part of a packaged solution.
I am in the operations group, looking to help the programming team diagnose the root cause of the lock issue. The problem is trying to figure out what happened after the fact.
Thanks,
Ken
The contention issue appears in the joblog as a "Not able to allocate objects needed for file FCTHFDL1 in library ORCFP member or program device FCTHDL1"
Followed by RPG MSGID RPG1216
Additional Message Information
Message ID . . . . . . : RPG1216
Date sent . . . . . . : 07/01/09 Time sent . . . . . . : 06:05:46
Message . . . . : Error message CPF4128 appeared during OPEN (C S D F).
Cause . . . . . : The RPG program FCT002 in library ORCOP received the
message CPF4128 while doing an implicit OPEN to file FCTHFDL1. See the job
log for a complete description of message CPF4128. If the file has a device
type of SPECIAL, there may be no message in the job log.
Recovery . . . : Enter C to cancel, S to obtain a printout of system
storage, D to obtain an RPG formatted printout of system storage, or F to
obtain a full formatted printout of system storage.
Possible choices for replying to message . . . . . . . . . . . . . . . :
D -- Obtain RPG formatted printout of system storage.
S -- Obtain printout of system storage.
F -- Obtain full formatted printout of system storage.
The programs in question are out of our control, part of a packaged solution.
I am in the operations group, looking to help the programming team diagnose the root cause of the lock issue. The problem is trying to figure out what happened after the fact.
Thanks,
Ken
The contention issue appears in the joblog as a "Not able to allocate objects needed for file FCTHFDL1 in library ORCFP member or program device FCTHDL1"
Followed by RPG MSGID RPG1216
Additional Message Information
Message ID . . . . . . : RPG1216
Date sent . . . . . . : 07/01/09 Time sent . . . . . . : 06:05:46
Message . . . . : Error message CPF4128 appeared during OPEN (C S D F).
Cause . . . . . : The RPG program FCT002 in library ORCOP received the
message CPF4128 while doing an implicit OPEN to file FCTHFDL1. See the job
log for a complete description of message CPF4128. If the file has a device
type of SPECIAL, there may be no message in the job log.
Recovery . . . : Enter C to cancel, S to obtain a printout of system
storage, D to obtain an RPG formatted printout of system storage, or F to
obtain a full formatted printout of system storage.
Possible choices for replying to message . . . . . . . . . . . . . . . :
D -- Obtain RPG formatted printout of system storage.
S -- Obtain printout of system storage.
F -- Obtain full formatted printout of system storage.
ASKER
Gary and Tliotta...The object lock error is CPF4128. We do have performance tools, and do recall a Lock Report--but I don't remember how to setup the collection or run the query.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
THANKS GARY!!
There are many, many possibilities...
Is this "program that runs unattended" one which your site has full control over? I.e., did your company create it, or is it part of a 3rd-party package or otherwise out of your hands?
What do you see in the joblog when the contention causes an error? It might be necessary to show one or more messages before the error message.
And, are you a programmer or are you looking for some operations automation?
Tom