cancel
Showing results for 
Search instead for 
Did you mean: 

Help with long Job Time Delay

sap_cohort
Active Contributor
0 Kudos

I am running a BW Extraction in our BWQ System and on the R3 QA side the Job Log is showing some kind of very long delay and I could use some Basis Expertise to help solve this.  The Job keeps showing a large gap of time in the Job Log.  This delay started just recently and used to run for 2-3 minutes, but now is showing 2 hours.

Could someone look at this job log and let me know where I might start to track down this problem?  My Basis team said this might be due to heavy job load in QA, but this seems unlikely and excessive.  Thanks for any help moving forward on this. 

Accepted Solutions (1)

Accepted Solutions (1)

former_member227600
Contributor
0 Kudos

Can you please check in systm log (sm21) if any dead lock is happening on that time.

sap_cohort
Active Contributor
0 Kudos

I did an SM21 and did not see any issues at all.  Thanks!

Answers (3)

Answers (3)

Sriram2009
Active Contributor
0 Kudos

Hi Kenneth

1.  As per your log It shows the data load ‘Full’ and also your extraction was developed in ‘Z’ may be because of this its taking
more time. During the time gap you can check the system activity in T code SM66 or SM50 (User id - ALEREMOTE)

2. Kindly go thru the SAP Note 1597364 - FAQ: BW-BCT: Extraction performance in source system

Thanks

Sriram

sap_cohort
Active Contributor
0 Kudos

It looks to me like the extraction is not happening until after the delay.  So I'm thinking it's something on the basis side.  Thanks for the Note I will review.

Former Member
0 Kudos

Hi ,

As its a Bw extractor i have seen similar delay some time in system because there are hold events in the BW chain to collect data . So please check the chain if there is any hold activated in the program code which are called in the extractor.

sap_cohort
Active Contributor
0 Kudos

I have been using this extractor for many years with no issue and there has not been any changes on the BW Development end.  Thanks for your input.

paul_power
Active Contributor
0 Kudos

Hi Kenneth,

Does it happens every time the job is run now?

If so, run the job while an st05 trace is active, also during the long time gap, check what is occuring in SM66/SM50 and monitor CPU usage, disk I/O

After the timegap, you can de-activate the trace in st05 and see what was being done during the time

Regards,

Paul

sap_cohort
Active Contributor
0 Kudos

Thanks for the reply.  The job name is not consistent so I think it would be hard to do this trace.