This is a mirror of upstream issue https://github.com/kubernetes-sigs/security-profiles-operator/issues/1152

If on OCP, the following link might be useful:

https://docs.openshift.com/container-platform/4.11/machine_management/manually-scaling-machineset.html

Although I suspect the best way would be to write a unit test for the controller that handles the finalizes and the node statuses.

Acceptance criteria:

• profiles are still usable even if nodes come and go