Hi all,
Example :
I've a job that waits for another job & uses an exclusive CR (CR is being used as 'shared' in lots of other jobs). At a certain moment this job gets its
condition and then waits to take the CR exclusive.
Question:
Is there any way (control/m-functionality??) to monitor this job, to check after -for example- 15min that it effectively started (that status is
changed from 'wait resource'to 'executing') ?
(PS = kind of analogue to 'latesub/time'-functionality in postproc, but then for a job that has 'wait resource'-state (blue) ==> the shout needs to be done after the job is (for example) 15min in 'wait resource'-state).
Technical details :
control/m z/os (mainframe) 6.3.01
Control/m em : 6.3.01.300
Thx in advance for support.
Robbie.
Check wait resource submission after are predefined interval
Hi,
To my knowledge this functionality does not exist.
I was confronted in the past with the situation where it was necessary to detect in our environment the jobs in waiting state (Wait Resource) and identify those resources when all the IN conditions were present. The method used for this monitoring is to perform in a cyclic 'script' one loop to execute SQL query in the <CMR_AJF> table and extract the status for each job given parameter from a input file.
If you used Control-M/Server for Distributed Systems you can try to implemented similar method:
1) Create an input file that contents the name of jobs that you want to check
2) Search in your AJF jobs in status <Wait for Resource>
select JOBNAME, ORDERNO from CMR_AJF where STATE='D' and JOBNAME like 'jobs%'
3) Compare the name of jobs that have the status <Wait for Resource> with the contents of the file input.
If matching, then extract the ORDERNO
4) Convert ORDERNO to ORDERID by utility <p_36>
5) Run the utility <ctmwhy> with the ORDERID parameter. This will indicate what resource the job wait.
If you store all results in a logs files, after you can implement an alert by <ctmshout> utility to indicate whether the time (15 minutes) is exceeded between the control cycle for specific jobs.
Just suggestion, perhaps it can give you other ideas
To my knowledge this functionality does not exist.
I was confronted in the past with the situation where it was necessary to detect in our environment the jobs in waiting state (Wait Resource) and identify those resources when all the IN conditions were present. The method used for this monitoring is to perform in a cyclic 'script' one loop to execute SQL query in the <CMR_AJF> table and extract the status for each job given parameter from a input file.
If you used Control-M/Server for Distributed Systems you can try to implemented similar method:
1) Create an input file that contents the name of jobs that you want to check
2) Search in your AJF jobs in status <Wait for Resource>
select JOBNAME, ORDERNO from CMR_AJF where STATE='D' and JOBNAME like 'jobs%'
3) Compare the name of jobs that have the status <Wait for Resource> with the contents of the file input.
If matching, then extract the ORDERNO
4) Convert ORDERNO to ORDERID by utility <p_36>
5) Run the utility <ctmwhy> with the ORDERID parameter. This will indicate what resource the job wait.
If you store all results in a logs files, after you can implement an alert by <ctmshout> utility to indicate whether the time (15 minutes) is exceeded between the control cycle for specific jobs.
Just suggestion, perhaps it can give you other ideas
Best regards
Walty
Walty
-
philmalmaison
- Nouveau
- Posts: 1148
- Joined: 08 Jun 2007 12:00
- Location: Ile de France