Resumption of "Avoid unneccessary creation of new IDs for table SCHEDULER_HISTORY when using a non-distributed job chain with a file order source" when two job chain monitors the same directory with the same filter

Current Situation

• We assume we have two job chains with a file order source where both monitors the same directory with the same filter:

  <file_order_source  directory="test" regex="\.trg"/>
  

• If both job chains are active then all works fine. Two IDs for the table SCHEDULER_HISTORY are created.
• If one of the job chain is stopped then for the stopped job chain unneccessary new IDs for table SCHEDULER_HISTORY are created.
  • It seems further depends on the used microscheduling
  • If old microscheduling is on then ~12 IDs are unneccessary created.
  • If old microscheduling is off then ~3 IDs are unneccessary created.

Maintainer Note

The Jobscheduler behavior is an immediate result of a users' misconfiguration.
We will therefore not change this behavior but ask our users to avoid conflicting file order sources for the same directories.