Remove node from scheduling

kubectl drain <node_name> --ignore-daemonsets --delete-emptydir-data

Add node back to scheduling

kubectl uncordon <node_name>