cancel
Showing results for 
Search instead for 
Did you mean: 

Job Scheduling for Expected Event Monitoring

former_member655821
Discoverer
0 Kudos

Hi,

I am trying to schedule a job for expected event monitoring "/SAPTRX/EE_MONITOR", but when i schedule the job hourly then the job is getting overlapped with the privious one which is producing dump - DBSQL_DUPLICATE_KEY_ERROR.If we have to run this job hourly only then how can we schedule the job is such a way that each time when job runs it will check  for the completion of privious job.please suggest.

Thanks

Vaibhav

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

Hi Vaibhav,

We had a similar requirement in the previous project but we have done below option as we can't predict how long the job runs.

In my opinion, you can achieve like below.

Step -1:

1. Go to SE38 and create a variant for program RSBTONEJOB2.

2. Add Step 2 Job name, Report /SAPTRX/EE_MONITOR and Variant.


Step - 2.


1.Add Step 1 of program, Report RSBSTONEJOB2 and new Variant.

2. Add program RSBTONEJOB2 and newly added variant

3. Add step 2 and include program /SAPTRX/EE_MONITOR.


Outcome:


Step 1 RSBSTONEJOB2 checks the previous job program /SAPTRX/EE_MONITOR runs and it would cancel the current one without making any dump.


If you have CPS scheduling tool, you can achieve this function, i believe.


Or else, just simply calculate the maximum amount of time it runs in a day.


For example -


9 AM to 12 PM - it runs close 2 hours but runs after 2 PM only one hour.


In that case, you can schedule multiple job name or same name with same program to schedule in different custom time daily.


Regards


GGOPII



Answers (2)

Answers (2)

former_member655821
Discoverer
0 Kudos

Hi Gopi,

Thanks for your suggestion.

I checked with custom functon of Expected Event Monitor, it seems no issue from that prospective,checking with the second option.

Hi Steffen,

Thanks for the refrence document, i hope it will help to reduce the no. of  expected events.

former_member190756
Active Contributor
0 Kudos

Hi Vaibhav,

normally EE_MONITOR should not run over one hour.

How many overdue EE are found by the job? You should maybe run a trace to see what costs so much runtime.

Best regards,

Steffen

former_member655821
Discoverer
0 Kudos

Hi Steffen,

Thanks for suggestion...!!

There are 78.619 entries in one job and some jobs have even more then that.

former_member190756
Active Contributor
0 Kudos

Hi Vaibhav,

does that mean you get per hour over 70.000 Expected Events overdue in your scenario?

What you can check is the following document:

There are some notes available how you can improve performance and also a possibility to parallelize the EE Monitor.

But you should also check if your scenario is setup correctly because so many overdue EE normally make not much sense.

Best regards,

Steffen

Former Member
0 Kudos

Hi Vaibhav,

Two possibilities:

1. Check Custom Activity functions which are assigned in Expected Event Monitor and Optimize and remediate those function codes.

2. You have to check these many events overdue every hour or the batch job setting overdue every time which are previously already overdue set.  You have to look out expected event activity which makes this running longer.

Regards


GGOPII