In the old days prior to Oracle 12c, whenever a session of a DBMS_JOB gots killed the job was restarting by its own.
After our DBA installed a new version of Oracle 12c and we ported our application into this one we observed the above behavior is not the same.
A long running job was killed, prior 12c the job was restarting after a while on its own or we were forcing to run it via the NEXT_DATE procedure.
With the new version the job is not restarting even if we use the NEXT_DATE procedure or the database is restarted. The DBMS_JOB hangs around and you have to remove and re-create the job so that it is again running.
The problem is that our application was relying on the restart functionality of dbms_jobs.
Is there a specific internal process that was taking care of this restart or is there a parameter in the new database that enables/disables this feature. Or is this a Bug.
Thanks.