Restart the Airflow container if the DB health condition repeatedly fails

[rafidka]

Overview

We want to make our AirflowDbReachableCondition result in a container restart if the DB is consistently unreachable. This could be useful if the result if the result of the DB connectivity issue is related to the container, and restarting it would help.

Acceptance Criteria

• The restart should NOT happen immediately after the DB becomes unreachable, to avoid unnecessary restarts.
  • Decide on a reasonable grace period.
• After consistently failing, the container should be terminated.

Additional Info

N/A