Customer has some data handling task and would like to run it on their OpenShift platform, however as the job requires quite a few resources and it is not a long running job. So considering of spending, they expect they could hibernate some worker nodes, so that they could quickly bring up them once they get job request and power it off when job finish

However as the auto repair nature of the machine config operator, nodes being powered off will be removed and new nodes created to replace them, hence currently they could not temporarily shutdown nodes to save cost

Above all, we would like to have two enhancements:

1. Temporarily disable auto repair of machine config operator on selective nodes
2. Hibernate nodes when resource free and power on when requests come in