If everyone else is running a cluster with kubeadm on virtual machine i think this lectures is a must have because i have spend same days to solve this error:
error: unable to upgrade connection: pod does not exist
or this
kubectl logs
Error from server (NotFound): the server could not find the requested resource
and i haven't find samethink about it on the kubernetes guide or on the kubeadm issue traker.
If you are running same virtualmachine you had to do on the Master add this at the kubeadm init:
--apiserver-advertise-address=<ip-address>
and on the other nodes in the config file: /etc/systemd/system/kubelet.service.d/10-kubeadm.conf
you had to add this other ambient variables:
Environment="KUBELET_EXTRA_ARGS=--node-ip=<worker IP address>"
so restart the kubelet services and that had permit ot me to connect to pod schedule on other nodes that are virtual machine and are not exposed to the internet but are linked to the real machine where they are running.
This is a little tips from my experience working with kubernetes

@kubernetes/sig-contributor-experience-bugs