E9.2 Scheduler Kernal stops submitting batches after x time

rival

Active Member
Dear All,

We have a strange issue. For years our scheduler was working without problems. We submit 2000 jobs per day into our production environment without a problem.

BUT:

We also schedule some jobs to be submitted into our CRP environment. When we schedule a batch which submits a batch every 10 minutes we see that after 5 or 6 succesful submits the batch is not submitted to the enterprise server from the CRP environment. In the scheduler there is launch failed. Also other jobs are not submitted anymore into the CRP environment.

What we noticed is that when we submit a batch every 5 minutes it keeps working perfectly. ( we did this for 5 days in a row every 5 minutes) When I disable it and start scheduling every 10 minutes we have the issue again.

Production Scheduler kernel log:

5432/5436 Mon Jul 4 09:51:07.011000 jdeknube.c3873
KNT0000106 - JDENET_ReceiveMsg failed. Error = 11

5432/5436 Mon Jul 4 09:51:07.035000 sc_lnch.c1236
Errors submitting the batch job, TEST EVERY 10 MINUTES CRP.

5432/5436 Mon Jul 4 09:51:07.045000 sc_lnch.c258
The job, TEST EVERY 10 MINUTES CRP, had errors when launching.

The job should have been submitted at 09:45 but the error in the log is 9:51. Below is the message in one of the jdenet logs.

<jdenet_n> registered in entry 22
5084/1444 Mon Jul 4 09:45:26.980000 netrcv.c486

10054-Error from Host=<10.140.X.X> in recv returned 10054 (WSAECONNRESET): Connection was reset by peer

We have this (WSAECONNRESET) message in our logs for 2 years but it always worked and we didn't notice an issue. Does someone have an idea what could be wrong and why it works if we schedule the job every 5 minutes?

What is changed from our side is that we running on new hardware ( virtual environment)

Kind regards,

Ron
 
Try changing the schedule in some way - Change the Scheduled State Date /Time.
 
Back
Top