metrics-service in kubernetes not working - kubernetes

Question

I'm running kubernetes using an ec2 machine on aws.
Node is in Ubuntu.
my metrics-server version.
wget https://github.com/kubernetes-sigs/metrics-server/releases/download/v0.3.7/components.yaml
components.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
name: metrics-server
namespace: kube-system
labels:
k8s-app: metrics-server
spec:
serviceAccountName: metrics-server
volumes:
# mount in tmp so we can safely use from-scratch images and/or read-only containers
- name: tmp-dir
emptyDir: {}
containers:
- name: metrics-server
image: k8s.gcr.io/metrics-server/metrics-server:v0.3.7
imagePullPolicy: IfNotPresent
args:
- --cert-dir=/tmp
- --secure-port=4443
- --kubelet-preferred-address-type=InternalIP,ExternalIP,Hostname
- --kubelet-insecure-tls
Even after adding args, the error appears.
error :
Error from server (ServiceUnavailable): the server is currently unable to handle the request (get nodes.metrics.k8s.io)
or
error: metrics not available yet
No matter how long I wait, that error appears.
my kops version : Version 1.18.0 (git-698bf974d8)
i use networking calico.
please help...
++
I try to wget https://github.com/kubernetes-sigs/metrics-server/releases/download/v0.5.0/components.yaml
view logs..
kubectl logs -n kube-system deploy/metrics-server
"Failed to scrape node" err="GET "https://172.20.51.226:10250/stats/summary?only_cpu_and_memory=true": bad status code "401 Unauthorized"" node="ip-172-20-51-226.ap-northeast-2.compute.internal"
"Failed probe" probe="metric-storage-ready" err="not metrics to serve"

Answer 1

Download the components.yaml file manually:
wget https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml
Then edit the args section under Deployment:
spec:
containers:
- args:
- --cert-dir=/tmp
- --secure-port=443
- --kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname
- --kubelet-use-node-status-port
- --metric-resolution=15s
add there two more lines:
- --kubelet-insecure-tls=true
- --kubelet-preferred-address-types=InternalIP
kubelet Of 10250 The port uses https agreement , The connection needs to be verified by tls certificate. Adding ,--kubelet-insecure-tls tells it do not verify client certificate.
After this modification just apply the manifest:
kubectl apply -f components.yaml
wait a minute and you will see metrics server pod is up

Answer 2

Last comment is useful.You can edit the deploy directly as well and adding line "--kubelet-insecure-tls=true" its enought for me:
Edit deploy:
$ kubectl edit deployment.apps/metrics-server -n kube-system
Add the line:
- --kubelet-insecure-tls=true
Similar result:
containers:
- args:
- --cert-dir=/tmp
- --secure-port=4443
- --kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname
- --kubelet-use-node-status-port
- --metric-resolution=15s
- --kubelet-insecure-tls=true
And save with ":wq" and enjoy.
~$ kubectl top pods -n kube-system
NAME CPU(cores) MEMORY(bytes)
coredns-6d4b75cb6d-k8dmc 3m 18Mi
coredns-6d4b75cb6d-wxxn6 3m 17Mi
kube-apiserver-k8s-master1 82m 306Mi
kube-apiserver-k8s-master2 65m 247Mi
kube-controller-manager-k8s-master1 32m 47Mi
kube-controller-manager-k8s-master2 4m 19Mi
kube-proxy-9dbgk 1m 9Mi
kube-proxy-bwhdm 1m 14Mi
kube-proxy-fz8v8 1m 15Mi
kube-proxy-vcnrc 1m 9Mi
kube-scheduler-k8s-master1 7m 18Mi
kube-scheduler-k8s-master2 4m 16Mi
metrics-server-79576f7ff-97tpc 6m 15Mi
metrics-server-79576f7ff-qzczp 4m 13Mi
~$ kubectl top nodes
NAME CPU(cores) CPU% MEMORY(bytes) MEMORY%
k8s-master1 318m 15% 1047Mi 55%
k8s-master2 208m 10% 1002Mi 52%
k8s-worker1 30m 3% 804Mi 42%
k8s-worker2 35m 3% 550Mi 29%