SAP BI4.1 SP01, Adaptive Job server issues

The Adaptive Job server, after taking some jobs from Crystal Report schedules, is going to “Stopping…” mode and does not come back. All other reports at this time are going into “Pending” status. I had to create another job server to release all of them. In a fraction all the Pending jobs Run to success. Even all the jobs have been completed, the first AJS is still in “Stopping…” mode. When I try to stop the SIA, its also going in “Stopping” mode. it is like this for hours…the only way out is to kill the sia.exe process in the task manager.

Now my question is, how to avoid this situation. I cannot afford this to happen on prod. I have created multiple AJS on each cluster, each job server is receiving a load of around 200 Crystal reports…any ideas how to stop this from happening? Thanks!


IronMan (BOB member since 2005-09-09)

Turn logging on and see what it says at the time the server starts stopping. There should at least be a clue in there.


Nick Daniels :uk: (BOB member since 2002-08-15)

My standard question: Are you on a somewhat recent patch level at all?


Andreas :de: (BOB member since 2002-06-20)

Enabled tracing but that did not help, as in order to enable tracing I had to make the servers come back to Active mode. But now I don’t know when it will go in stopping mode.

We have not installed any patches…only sp01.

is there any timout setting for scheduled jobs in the Job server? so that jobs stop when they exceed certain time limit.


IronMan (BOB member since 2005-09-09)

can any one explain what does below options mean in Job server settings?

Max concurrent jobs 5(default)

Max child request 100 (default)

I understand the first option but not the second one (Child request)??

Thanks!


IronMan (BOB member since 2005-09-09)

ok, after looking at the trace file of the job server, I see below text:

“currentLoadToThisChild[1], maxLoadForThisChild[1], maxLoadDetected[true],”

so it seems like as the report is huge, the jobserverchild.exe is reaching its maxload and unable to process futher, so it keeps in Running state.

is there anyway to increase the maxload limit of the child process? :?:


IronMan (BOB member since 2005-09-09)

this issue is now resolved :smiley:
the bottleneck was that this was a huge report which had 250,000 pages and when exported to pdf, it takes forever, almost 14hours to complete the job. the business has decided to cutdown on the no. of pages a report can return.

cannot imagine a quarter million page pdf :crazy_face:
Thanks!


IronMan (BOB member since 2005-09-09)

I know that your post is from 2013, but this is an issue that we are having also and I could not find a newer post for this.

So what was the fix, have the business modify the report, or did you increase the # of Max children from default?


bgodwin64 (BOB member since 2015-12-28)