Kubelet command stops responding for some time intervals

aditya03

Hi

I have created a cluster and deployed nginx in 3 pods.
i am facing issue with my kubelet service.
Sometime it returns an output and sometimes it returns a message
The connection to the server k8scp:6443 was refused - did you specify the right host or port?
It works normally for sometime when I restart the kubelet service.
The swapoff is disabled.
What could be causing this and how to resolve this ?

chrispokorni

Hi @aditya03,

So far I am concerned about the Kubernetes version mismatch between the two nodes - cp and worker. The lab guide in chapter 3 specifically calls for v1.27.1 for both nodes.

What is the version of your kubectl CLI? You can retrieve it with kubectl version -o yaml

Are the two EC2 instances provisioned according to the video guide from the introductory chapter?

Please provide the outputs of the following commands:

kubectl get nodes -o wide
kubectl get pods -A -o wide
journalctl -u kubelet #may require 'sudo'

Regards,
-Chris

aditya03

Hi Chris
I think there is a minor difference in the kubectl versions.
I also followed the video guide to create instances.
PFA the screenshots for all the commands mentioned above.

Thanks

aditya03

++ @chrispokorni

chrispokorni

Hi @aditya03,

It seems you deviated from the installation instructions.

Please fix the following according to the latest lab guide release:
1 - the guest OS should be Ubuntu 20.04 LTS (yours is 22.04 LTS - not yet recommended for the lab environment)
2 - the installed Kubernetes version should be 1.27.1 for all three components: kubeadm, kubelet, and kubectl

Also, while provisioning the EC2 instances make sure to closely follow the demo video on AWS from the introductory chapter, for VPC, SG, and instance size details.

Regards,
-Chris

aditya03

Hi @chrispokorni

I will go through the installation guide once again.
Thanks for resolving the query.

Regards,

• Aditya