How to fix issue of 'Unable to connect to the server: EOF' Kubernetes - Kubectl

Question

When I try any kubectl command, it always returns:

Unable to connect to the server: EOF

I followed these tutorials:

https://kubernetes.io/docs/tasks/tools/install-kubectl/

https://kubernetes.io/docs/tasks/access-application-cluster/configure-access-multiple-clusters/

But they have not helped me. According to the first link, by default, kubectl configuration is located at

~/.kube/config

But in that path I don't have anything. I don't know if this is causing the issue.

Other thing is when I try to check the kubectl configuration:

M:.kube candres$ kubectl cluster-info
Kubernetes master is running at http://localhost:8080

To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.
Unable to connect to the server: EOF
M:.kube candres$ kubectl cluster-info dump
Unable to connect to the server: EOF

The versions I have installed are:

Kubernetes - kubectl

M:.kube candres$ kubectl version
Client Version: version.Info{Major:"1", Minor:"9", GitVersion:"v1.9.3", GitCommit:"X", GitTreeState:"clean", BuildDate:"2018-02-09T21:51:06Z", GoVersion:"go1.9.4", Compiler:"gc", Platform:"darwin/amd64"}
Unable to connect to the server: EOF

Minikube

M:.kube candres$ minikube version
minikube version: v0.25.0

Docker:

M:.kube candres$ docker version
Client:
 Version:   17.12.0-ce
 API version:   1.35
 Go version:    go1.9.2
 Git commit:    X
 Built: Wed Dec 27 20:03:51 2017
 OS/Arch:   darwin/amd64

Server:
 Engine:
  Version:  17.12.0-ce
  API version:  1.35 (minimum version 1.12)
  Go version:   go1.9.2
  Git commit:   X
  Built:    Wed Dec 27 20:12:29 2017
  OS/Arch:  linux/amd64
  Experimental: true

Does anyone know how to resolve this?

Hi @Carlos, It's quite an odd behaviour that one of the command `kubectl cluster-info` is working. However, others are not working. Could you please run following command `kubectl config view` which will list all of the clusters and contexts. — Suresh Vishnoi, Feb 22 '18 at 13:20
btw, Port number for API Server is 8443 or 443. Why the port number is 8080 ? — Suresh Vishnoi, Feb 22 '18 at 13:21
Hi @SureshVishnoi! When I run that command, that return: `M:.kube candres$ kubectl config view apiVersion: v1 clusters: [] contexts: [] current-context: "" kind: Config preferences: {} users: []` With regard about the port, I dont know why this use port 8080! — Carlos Andres, Feb 22 '18 at 13:25
It means kubectl does not know about k8s API server. IF it does not know about the api server. It won't be able to do anything — Suresh Vishnoi, Feb 22 '18 at 13:27
Hi Following link will give your comprehensive detail [kubeconfig file](https://kubernetes.io/docs/concepts/configuration/organize-cluster-access-kubeconfig/) — Suresh Vishnoi, Feb 22 '18 at 13:29
As you are using minikube , As far as I know, Minikube configure `kubeconfig` file automatically. So You can restart the minikube. You can run `minikube status` to verify it. — Suresh Vishnoi, Feb 22 '18 at 13:33
@SureshVishnoi ! With your help I fix my issue. The issue was than I dont start the minikube. So, later do this: `minikube start` With this, my machine come to download the ISO of minikube. Then `M:.kube candres$ kubectl cluster-info Kubernetes master is running at https://192.168.99.100:8443` Thanks a lot ! — Carlos Andres, Feb 22 '18 at 13:47
Glad it worked. I will post the answer and we can close the question. — Suresh Vishnoi, Feb 22 '18 at 13:48

score 24 · Accepted Answer · edited Jul 13 '20 at 03:31

24

After Minikube is started, kubectl is configured automatically.

minikube start
Starting local Kubernetes cluster...
Kubernetes is available at https://192.168.99.100:8443.
Kubectl is now configured to use the cluster.

You can verify and validate the cluster and context with following commands.

kubectl config view

edited Jul 13 '20 at 03:31

ghitesh

160
2
14

answered Feb 22 '18 at 13:51

Suresh Vishnoi

17,341
8
47
55

score 5 · Answer 2 · answered Jan 31 '19 at 20:11

5

I also had this issue. Be sure to check your config file that is generated by minikube. This file can most likely be found ~/.kube/config. Make sure that you are referencing the right cluster name in the current context you are using. You can see what context you are currently using by: kubectl get current-context. The important thing is that you understand why you are getting this error and as @Suresh Vishnoi stated, kubectl doesn't know about k8s api-server.

answered Jan 31 '19 at 20:11

Nappstir

995
2
20
38

2

Writing this in 03Dec2020 while picking up Kubernetes and `kubectl get current-context` doesn't work. View current context with `kubectl config current-context` instead – Jonalogy Dec 03 '20 at 05:51

score 2 · Answer 3 · answered Oct 30 '21 at 12:07

2

here the steps to my solution

Install minikube:

brew install minikube

Start minikube

minikube start

check again and ✅

kubectl version --short

Client Version: v1.16.6-beta.0

Server Version: v1.22.2

answered Oct 30 '21 at 12:07

Junior Vieira

553
5
5

Flag --short has been deprecated, and will be removed in the future. The --short output will become the default. – Noam Nol Oct 24 '22 at 16:23

score 2 · Answer 4 · answered Jun 23 '22 at 11:40

2

Here is all I had to run:

minikube delete
minikube start

answered Jun 23 '22 at 11:40

Akintunde

3,525
1
21
23

score 1 · Answer 5 · answered Jul 01 '20 at 12:08

1

Just updating Kubectl version to latest version resolve my problem.

answered Jul 01 '20 at 12:08

Mudit Nagpal

11
2

I think this depends of the cluster version? – Eddy Hernandez Aug 20 '20 at 19:24
Worked. Downloaded `kubectl.exe` v1.23.0 (before I had v1.17.4) and then in powershell the familiar `https://microsoft.com/devicelogin` authentication process reappeared. – SushiGuy Feb 02 '22 at 21:13

score 1 · Answer 6 · answered Sep 06 '21 at 22:14

1

If you get a message like this:

You appear to be using a proxy, but your NO_PROXY environment does not include the minikube IP (www.xxx.yyy.zzz).

Then set your environment variable NO_PROXY to the address given before running kubectl. This is probably configurable somewhere, but that's a short quick solution.

answered Sep 06 '21 at 22:14

Ken

11
1

this is the point for me. Here is my config in ~/.zshrc: ``` bash export http_proxy=http://192.168.168.168:1087; export https_proxy=http://192.168.168.168:1087; export no_proxy="10.0.0.0/8,localhost,127.0.0.1,192.168.0.0/16" ``` – terry zhang Dec 04 '21 at 13:43

score 1 · Answer 7 · edited Oct 11 '21 at 08:29

1

reset kubeadm via force

reset kubeadm -f

and then copy config file again

cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
chown $(id -u):$(id -g) $HOME/.kube/config

at last

kubectl init

edited Oct 11 '21 at 08:29

Suraj Rao

29,388
11
94
103

answered Oct 11 '21 at 08:28

Fox

11
2

score 0 · Answer 8 · answered May 02 '22 at 11:31

0

I'm using wsl. So it helps me to synchronize time between windows and linux console.

sudo hwclock --hctosys

answered May 02 '22 at 11:31

Vlad Pavlovski

1,582
1
14
10

score 0 · Answer 9 · answered Aug 02 '22 at 07:06

0

Check your VPN security as well as your anti virus internet security. in case it is ON then we have to make it off. and it worked for me after that.

Try it out this also.

answered Aug 02 '22 at 07:06

Mohit Jariwal

71
1
2

score 0 · Answer 10 · answered Jan 25 '23 at 20:03

0

VPN should solve it, if not, then also try unsetting local env proxy settings.

answered Jan 25 '23 at 20:03

Ajit Surendran

709
7
4

How to fix issue of 'Unable to connect to the server: EOF' Kubernetes - Kubectl

10 Answers10