We have a set of jobs and they are all having a similar problem when scheduled. The steps where they are failing is the S&OP Heuristic (infinite - SOP). When I look at the log there is no detailed provided (e.g. attachment file) . Below is a screen shot of a recent case. What is strange is that I can make a copy of the same job that failed and run it inmediatelly and the job doesnt fail.
Here below is the screen of the same job (ran only 5 minutes after the one the failed) and the details of the second step. This one completed, this step didnt failed and the details in the log Attachement says basically:
2 to 3 months ago these jobs were running (without any changes made) without any problems. If it was a MD issues (which it does not show in the log) I would have thought that the job/step was always going to failed until it was fixed.
I will appreciate if you can help me here. I have looked everywhere in our system and in the SAP community.
Thanks in advance for your help
Hello robvidAlthough it is surprisng that this strange behaviour is, but I am not completely surprised. I have seen such cases few times before. Although the best option would be to raise a OSS with SAP by mentioning the correct CI component from the error log, but here are few things that you can check.1) when this job is failing are thier any parallel jobs of different varient that are scheduled to run within that time frame ? E.g to parallel heusristics using the same S&OP profile but having different Subnetwork? If Yes you need to add a 7 to 10 min delay between those jobs, as the TS objects get locked in many cases.2) Any advanced dissagg operator running in parallel, that touches any of the direct or indirect input / output KF of TS heuristics? If yes please adjust.Let me know if that helps
Thanks for your reply and excellent observation/recommendations
We do have jobs that were running with no gap, that are/were variants with different subnetworks. I will follow your advice and space them.
If there is another job, like a CPI-DS (e.g copying and sending Plan Order to SAP/S4) running at the same time or within a few minutes of each other, could that create a problem as well?. There has been occasions were we had the same failure late at night 22:45 and the only job was a CPI-DS.
I will make the changes and check for your comments and of any other member.
Thanks very much for your advice