CORE STUDY: Kubernetes

Reading Time: 2 minutes

edit: 1/16/2022 – I am putting some basic info first. At some point I will drop in the text for the Kubernetes installation or I should create a post just for that.
edit 1/25/2022 – more notes – a little more organization

This is a place holder. This topic deserves quite a bit of attention. Everything from using Kubernetes to scale computation to creating clusters … or even storage clusters.

We are going to break this discussion into a number of sections.
1 – Theory of Operation
2 – The Installation
3 – Now what – how to I use this thing?

Kubernetes: An Introduction

What is Kubernetes? Put simply Kubernetes is an orchestration system for managing applications. This means managing applications that consist of a collection of containers.

Theory of Operation

We are going to end up with 3 core components.

* Control Plane – the orchestration component responsible for clustering, nodes, pods, services and networking,
* Management Plane – the administrative layer
* Data Plane
– traffic handling layer

Let’s talk about Kublet. Kublet is the node agent for Kubernetes. If you have a server that has Docker installed… then you have a server with docker installed. If you want to leverage the machines resources with Kubernetes then installing the Kublet service is used in part to identify and add the server [node] with the rest.

The Installation

sudo apt-get update
sudo apt-get install -y kubectl
kubectl get nodes

Note if you see something like:

The connection to the server was refused - did you specify the right host or port?

Then certain environment variables haven’t been set – or there are other issues – like making sure computer has the kubelet service started.

sudo kubeadm init --pod-network-cidr=
kubadm token list
kubeadm token list
kubeadm version

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

NOTE: After a restart.

Some care will need to be given after restarting your Kerbernetes server.

swapoff -a
systemctl stop kubelet
systemctl start kubelet
kubectl get nodes