When we are applying changes all the time to a cluster, we want to avoid disruption.
- Apply changes to cluster
- Create configurations
- Start services if not running
But don't:
- Update services
- Restart machines with bootc updates
- etc...
(anything disruptive basically..)