Kubernetes Drain Bad Node

For reference, when a node has been determined to be bad/inaccessible etc, this seems like the fastest way to flush it:

kubectl drain NODE_IN_QUESTION --ignore-daemonsets --grace-period=0