Autosys Chase Error

Hi All,

We use Autosys to schedule and report job which runs in UNIX machine. Most of the job runs fine. But we have problem with one of the file watcher job . The Job is defined to run from 2nd to 10th Business day of every month starting 06:00 PT to 17:00 PT. It watches for a file in a directory. It's in running state for a long time then suddenly the job fails with chase error.

Chase error being remote agent not available in the machine. Can you please tell me how to trap this issue and avoid it.

This job is running in the same machine where rest of the jobs are running fine.

Thanks,
Lokesh Kumar C

Just spitballing here, but it sound like autosys is like a cron on a central machine, with scripts that ssh over to run batches. Running a batch so long means any network hiccup could kill the ssh session and so the controlling terminal and process of your batch process. The most obvious solution is to have the job start as a nohup background job that terminates itself. Autosys just has to start it successfully to be happy. You need to worry about reruns/restarts if a system is rebooted during the day.

Lokesh,

Filewatcher job continuously runs to look for the file untill and unless you terminate the job by defining the termination in the JIl code. You said that the job should look for the file from 2nd to 10th business day of every month. Does that mean this job is supposed to run every day or when the job starts running on 2nd BD it will be running throughout 10th BD of the month? For second instance, the job will get CHASE_ERROR for sure since there might be network glicthes/server might get reboot etc. which are the main reasons for the CHASE_ERROR.

I would suggest to add a termination by 17:00 hrs PT and start it again on next day by 6 AM PT.

Let me know if it makes sense.

Thanks,
Mohan.

Hi Lokesh,

The file watcher job failure occured due to the server reboot that happened in the server in which the job is running(FILE WATCHER) so at that point Autosys chase agent might have disconnected .
So check once whether the server is up and after that if you restart the job it will run as usual and it will attain success once the file is present in the file watcher path.

Regards,
Manoj

Hi Lokesh,

Can you please share the o/p of below commands,

autorep -J <JOB_NAME> -d
autorep -J <JOB_NAME> -q

Thanks & Regards,
Pradeep Agarwal