Table of Contents
Was this article helpful?
Try Vultr Today with

$50 Free on Us!

Want to contribute?

You could earn up to $600 by adding new articles!

Deploy a Kubernetes Cluster on Ubuntu 20.04

Author: Thomas Rakwach

Last Updated: Tue, Nov 16, 2021
Kubernetes Ubuntu


Kubernetes, also known as K8s, is an open-source platform for container orchestration. It manages clusters meant to run containerized applications, scaling deployments, and provides a platform for deployment automation. A container is a form of operating system virtualization that isolates code and all its dependencies, so the application runs securely, quickly, and reliably from other computing environments. Due to the deprecation of Docker runtime by Kubernetes, a new API plugin for container runtimes called CRI (Container Runtime Interface) has been introduced, enabling kubelet to use a wide variety of container runtimes. CRI-O is an implementation of the Kubernetes CRI to enable using OCI (Open Container Initiative).

Why Kubernetes?

After deploying a small application in a container, you might need to scale it on many servers. Due to the complexity of this operation, there should be an automation tool responsible for allocating resources to the application on different servers, monitoring the application and servers. This is where Kubernetes comes in. You need at least a master node and a worker node to have a cluster. Later on, you can expand the cluster by adding as many worker nodes as you need. This enables your application's containers to achieve high performance and high availability by distributing it between available hosts within the cluster.

In this article, you will learn how to deploy a Kubernetes cluster on Ubuntu 20.04.


  • Deploy two or more fully updated Vultr Ubuntu 20.04 servers.
  • Create a non-root user with sudo access on all servers.

1. Install CRI-O

Kubernetes requires CRI-O to run. Perform all these steps on all the nodes.

Update the system.

$ sudo apt update

Load the overlay and br_netfilter modules.

$ sudo modprobe overlay

$ sudo modprobe br_netfilter

Create a sysctl config file to enable IP forwarding.

$ sudo nano /etc/sysctl.d/99-kubernetes-cri.conf

Add the following code to the file. Save and close the file.

net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
net.ipv4.ip_forward = 1

Reload sysctl.

$ sudo sysctl --system

Change user to root.

$ sudo -i

Add repository version information.

# export OS=xUbuntu_20.04

# export VERSION=1.18

Add CRI-O repositories.

# sudo echo "deb$OS/ /" > /etc/apt/sources.list.d/devel:kubic:libcontainers:stable.list
# sudo echo "deb$VERSION/$OS/ /" > /etc/apt/sources.list.d/devel:kubic:libcontainers:stable:cri-o:$VERSION.list

# sudo curl -L$VERSION/$OS/Release.key | sudo apt-key add -
# sudo curl -L$OS/Release.key | sudo apt-key add -

Exit root.

# exit

Update the system.

$ sudo apt update

Install CRI-O.

$ sudo apt install cri-o cri-o-runc -y

Find the path of container monitoring utility.

$ which conmon

Edit the CRI-O configuration file /etc/crio/crio.conf.

$ sudo nano /etc/crio/crio.conf

The file should have the following changes:

conmon = "/usr/bin/conmon"

registries = [

Reload the daemon.

$ sudo systemctl daemon-reload

Start the CRI-O service.

$ sudo systemctl start crio

Enable CRI-O service to startup at system boot.

$ sudo systemctl enable crio

Check the CRI-O service status.

$ sudo systemctl status crio

2. Disable Swap Memory

Swap memory needs to be disabled for Kubernetes to run normally. Perform these steps for all the nodes.

Disable the swap memory.

$ sudo swapoff -a

Reboot the system for changes to take effect.

$ sudo reboot

3. Install Kubernetes

Perform these steps on all the nodes.

Install all required packages.

$ sudo apt install -y apt-transport-https curl

Add Kubernetes GPG Key.

$ curl -s | sudo apt-key add -

Add Kubernetes APT Repository.

$ sudo apt-add-repository "deb kubernetes-xenial main"

Install Kubeadm, Kubelet, and Kubectl. This also installs Kubernetes-cni to enable container networking in the cluster.

$ sudo apt install -y kubelet kubeadm kubectl

$ sudo apt-mark hold kubelet kubeadm kubectl

Reload the daemon.

$ sudo systemctl daemon-reload

Restart the CRI-O service.

$ sudo systemctl restart crio

Restart the kubelet service.

$ sudo systemctl restart kubelet

Initialize the master node using kubeadm. This should only be run on the master node. Make sure you copy and save the worker node join command and tokens where you can retrieve them for later use.

$ sudo kubeadm init --pod-network-cidr= --ignore-preflight-errors=all

Create new .kube configuration directory. This should only be run on the master node.

$ sudo mkdir -p $HOME/.kube

Copy the configuration admin.conf from /etc/kubernetes directory to the new .kube directory. This should only be run on the master node.

$ sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config

Change the permissions to the configuration directory. This should only be run on the master node.

$ sudo chown $(id -u):$(id -g) $HOME/.kube/config

Check kubeadm version. This should only be run on the master node.

$ kubeadm version

4. Configure Pod Network

Install the Weave network plugin to allow communication between the master and worker nodes. This should only be run on the master node.

$ kubectl apply -f "$(kubectl version | base64 | tr -d '\n')"

5. Join Worker Node to the Cluster

If you lost the join command, you could find it by running this command on the master node.

$ kubeadm token create --print-join-command

Join worker nodes to master node. Perform this step for worker nodes only. Use the join instructions generated and saved in the kubernetes initialization step. Make sure you add the --ignore-preflight-errors=all flag. For example:

$ sudo kubeadm join --token 2qmwga.qyejfbo9vouiowlt \ --discovery-token-ca-cert-hash  sha256:083a2a20c8de9254100f1b37b4be1999946aee6f34791985c80d9eced9618e94 --ignore-preflight-errors=all

From the master node, check all nodes status.

$ kubectl get nodes

From the master node, verify pod namespaces.

$ kubectl get pods --all-namespaces

6. Deploy Nginx in Kubernetes

Deploy an Nginx application in Kubernetes on the master node using YAML.

Create a pod.

$ kubectl apply -f

Check pod status.

$ kubectl get pods

Check all pods' information.

$ kubectl describe pods


You have successfully learned how to install Container Runtime Interface, Kubernetes cluster, join worker nodes to the cluster and deploy a containerized application on your server. You can now deploy containerized applications and scale them to handle more user traffic.

More Information

For more information on Kubernetes, please see the official documentation.

Want to contribute?

You could earn up to $600 by adding new articles

La versión española del presente sitio web es una traducción realizada únicamente con fines informativos, y la versión inglesa prevalecerá.