Question
Problems with resumption of flow via SLA when cases are being created in bulk
Hello All,
Currently in our project we are using Pega v 7.1.7.
We have a functionality where cases are being created in bulk from an external system. All the cases created first sit in an assignment and then an SLA resumes the cases from that assignment after 20 seconds.
But sometimes we are seeing some of the cases are not being resumed by the SLA. The cases are not erroneous cases. When we are opening up the cases and submitting them they are advancing without any problem. We have analyzed the logs and not found any obvious problem.
We have tried creating cases in UAT environment but everything worked as expected.
Is SLA failing to execute because cases are created in bulk? Please note that out of approx. 2000 created from external system only 50-70 is facing the problem.
Thanks and Regards,
Prayas
Hello,
Thanks for replying . No error information is present on the logs or the Service Level agent instance from the SMA. Even if SLA would have run for the cases that would have been there in the history of the work item. History of the work item also did not show that Service level has run for these items.
Thanks and Regards,
Prayas
Right,
Can you compare both instances (working and not working) to check the difference. Maybe the SLA is under the impression that this isn't the right time to pickup those instances for any reason.
Hello,
There is no difference between the work objects. Exact similar work objects. SLA is set to 20 seconds timer. We have compared work objects from the same batch which have come at same time. We compared the history of the items to double check on the creation time and everything was in order. The ones that were stuck , remained there in the assignment even after 5 days.
Have you checked the SLA queue item for those work objects? Is it in a Broken-Process state or Scheduled? If Scheduled, what is the pyMinimumDateTimeForProcessing?
Hello,
I think you already have the answer to your first question. SLA is surely not failing because cases are created in bulk. As you said only a few have them aren't moving to next step correctly.
I would trace the agent in order to check what could be happening. Anything visible on the SLA instance? Any error/information in there?