Long time for execution for scheduled CIF background jobs

[Former Member]

Hi,

we hv schedlued CIF background job to be run daily around 10.30 PST.

there is large variation for the time required for execution of this job.

on 26 Dec, it took approx 48000 seconds while regular average is 120 seconds only.

today, despite past of 6000 seconds, the job is still in ACTIVE stage.

can anyone know why such long delay for such jobs?

how can i reduce its execution time (as one of case in a week)?

rgds/Jay

[Former Member · Accepted solution]

Hi Jayaprakash,

When a CIF job is prolonging, it is getting struck for updation of a or few data into the system. This can be checked via inbound or outbound queues. The sysfail analysis can throw more light. Use transaction codes, smq1,smq2, /n/sapapo/c3, sfg1 etc., to view the application logs created.

Hope this helps

Regards

R. Senthil Mareeswaran.

[Former Member]

please delete obsolete and processed change pointers ,it will improve the performance a lot. You can use bd22 for this and run the jobs making use of parallelization process

[Former Member]

Hi Jay,

A few obvious things to look for:

1. Multiple CIF activation jobs running at the same time

2. Large change in the master data, eg new plant, new Material Masters, new customers, etc etc.

3. Conflicts with other non CIF programs that may be going after the same data

4. Communication degradation between the OLTP and SCM clients

Normally you refer such questions to someone on your Basis team, or perhaps your DBA. They can turn on tracing tools that can track the changes in your environment that may be contributing to the changes in run time.

Regards,

DB49