After successfully setting up k8s (using `kubeadm init`), the coredns pod
remains in a 'Pending' state:
NAMESPACE     NAME                               READY   STATUS    RESTARTS  
AG
kube-system   coredns-86ccd44ff8-24h9w           0/1     Pending   0         
12
kube-system   coredns-86ccd44ff8-rz5vq           0/1     Pending   0         
12
kube-system   etcd-k8s-head                      1/1     Running   1         
12
kube-system   kube-apiserver-k8s-head            1/1     Running   1         
12
kube-system   kube-controller-manager-k8s-head   1/1     Running   1         
12
kube-system   kube-proxy-qprnj                   1/1     Running   0         
12
kube-system   kube-scheduler-k8s-head            1/1     Running   1         
13

This is due to a not yet installed network plugin.
When installing `kubeadm` thru 'patterns-kubernetes-kubeadm', the package
'flannel-k8s-yaml' is installed alongside which contains a yaml file to set up
Flannel as network plugin.
Running `kubectl apply -f /usr/share/k8s-yaml/flannel/kube-flannel.yaml` using
the supplied file fails, however after:

| clusterrole.rbac.authorization.k8s.io/flannel created
| clusterrolebinding.rbac.authorization.k8s.io/flannel created
| serviceaccount/flannel created
| configmap/kube-flannel-cfg created
| daemonset.apps/kube-flannel-ds created

with the error:

| error: resource mapping not found for name: "psp.flannel.unprivileged"
namespace: "" from "/usr/share/k8s-yaml/flannel/kube-flannel.yaml": no matches
for kind "PodSecurityPolicy" in version "policy/v1beta1"
ensure CRDs are installed first

When using the yaml file from
https://github.com/flannel-io/flannel/blob/master/Documentation/kube-flannel.yml
(wget
https://raw.githubusercontent.com/flannel-io/flannel/master/Documentation/kube-flannel.yml)
it works, however, the flannel pod is still cycling thru a CrashLoopBackOff -
it fails with (from log):
| E0110 11:46:48.988799       1 main.go:332] Error registering network: failed
to acquire lease: node "k8s-head" pod cidr not assigned