Openshift Container Platform 3:11: /usr/bin/docker-current: Error response from daemon: containerd: container did not start before the specified timeout

Issue

A node will mark all of its pods as terminating and get hung while the remaining pods will be stuck at "ContainerCreating" after restarting. There are no messages about the node having a bad condition, such as disk or memory pressure, but all pods will be marked as terminating and not recover. The nodes seem to go into this state during heavy traffic times.