Connection to Server Refused

calendar2020-06-27 · 1 min read · k8s kubernetes troubleshooting kubectl doks managed-kubernetes  ·
Share on: linkedincopy

When you see similar error as below:

1user@example:~$ kubectl get nodes
2The connection to the server localhost:8080 was refused - did you specify the right host or port?

Check for the "$KUBECONFIG" if set and "kubectl config view" output. Chances are, both are empty like below:

1user@example:~$ echo $KUBECONFIG

1user@example:~$ kubectl config view
2apiVersion: v1
3clusters: null
4contexts: null
5current-context: ""
6kind: Config
7preferences: {}
8users: null

In this example, we're using the DOKS or DigitalOcean Managed Kubernetes.

To fix the problem: Get the config file and set the value of "$KUBECONFIG" to it.

1user@example:~$ export KUBECONFIG=./k8s-infra-kubeconfig.yaml 
2user@example:~$ echo $KUBECONFIG
3./k8s-infra-kubeconfig.yaml

and output of "kubectl config view" would also have something like below:

 1user@example:~$ kubectl config view
 2apiVersion: v1
 3clusters:
 4- cluster:
 5    certificate-authority-data: DATA+OMITTED
 6    server: https://b91642bd-a291-4200-c1ad-135ec3094be8.k8s.ondigitalocean.com
 7  name: do-nyc1-k8s-infra
 8contexts:
 9- context:
10    cluster: do-nyc1-k8s-infra
11    user: do-nyc1-k8s-infra-admin
12  name: do-nyc1-k8s-infra
13current-context: do-nyc1-k8s-infra
14kind: Config
15preferences: {}
16users:
17- name: do-nyc1-k8s-infra-admin
18  user:
19    token: d44a591ac43024737ad524ffd17039b318d21646b685a140754391ebe23dea6c

Going back to our earlier command that has the error, you should now see like below:

1user@example:~$ kubectl get nodes
2NAME                 STATUS   ROLES    AGE     VERSION
3default-pool-9yhb2   Ready    <none>   6h40m   v1.18.3
4default-pool-9yhbd   Ready    <none>   6h40m   v1.18.3

More troubleshooting: https://github.com/kubernetes/kubernetes/issues/50295