Checking the status of the nodes in my test cluster, I noticed that the master is shown in the NotReady state.
NAME STATUS ROLES runlevl41c.mylabserver.com NotReady master
I described the node to see if I could quickly discover what the problem was.
$ k describe no runlevl41c.mylabserver.com | grep -i conditions -A 10 Conditions: Type Status LastHeartbeatTime LastTransitionTime Reason Message ---- ------ ----------------- ------------------ ------ ------- MemoryPressure False Sun, 12 Jan 2020 04:46:03 +0000 Sun, 12 Jan 2020 01:37:35 +0000 KubeletHasSufficientMemory kubelet has sufficient memory available DiskPressure False Sun, 12 Jan 2020 04:46:03 +0000 Sun, 12 Jan 2020 01:37:35 +0000 KubeletHasNoDiskPressure kubelet has no disk pressure PIDPressure False Sun, 12 Jan 2020 04:46:03 +0000 Sun, 12 Jan 2020 01:37:35 +0000 KubeletHasSufficientPID kubelet has sufficient PID available Ready False Sun, 12 Jan 2020 04:46:03 +0000 Sun, 12 Jan 2020 01:37:35 +0000 KubeletNotReady Failed to initialize CSINodeInfo: error updating CSINode annotation: timed out waiting for the condition; caused by: the server could not find the requested resource
So it appears that the kubelet on this node isn’t happy.