How to install Kubernetes with Kubeadm – Ubuntu/Debian

Last updated on December 5th, 2024 at 06:05 pm.

Table of Contents

• Prerequisites
• Edit hosts file
• Control plane
• Worker node(s)
• Installing a container runtime – Containerd
  • Installation prerequisites
    • Verify that the br_netfilter, overlay modules are loaded by running:
  • Installing Containerd
    • Step 1: Installing containerd
    • Step 2: Installing runc
    • Step 3: Installing CNI plugins
    • Step 4 Configuring the systemd cgroup driver 
• Install kubeadm, kubelet and kubectl
• Creating a cluster with kubeadm
• Install CNI – Calico
• Joining your other nodes to the cluster – Workers
• Test your cluster – deploy nginx
• Extra commands
• Update Github Containerd , CNI Network plugins and RunC

Note : I will upload a video with these steps. Check back soon.

Documentation source: https://kubernetes.io/docs/setup/production-environment/tools/kubeadm/install-kubeadm/

Prerequisites

• Linux host
• Minimum 2GB RAM and 2 (v)CPUs
• Public or Private Network connectivity between nodes
• Unique hostname, MAC address, and product_uuid for every node.
  Verify MAC address using : ip link or ifconfig -a . Check product_uuid using :
  sudo cat /sys/class/dmi/id/product_uuid
• Swap is supported since kubelet v1.22. As of v1.28 it is only supported for croup v2 : Check croup by running:
  grep cgroup /proc/filesystems . To be safe, turn off swap by running:
• sudo swapoff -a ; sudo sed -i '/ swap / s/^/#/' /etc/fstab

• Open required ports as below:

src: https://kubernetes.io/docs/reference/networking/ports-and-protocols/

Edit hosts file

Add your IP to your /etc/hosts file with your hostname. The screenshot below is an example on a Virtualbox VM. In a public cloud environment, you should use a fqdn domain as your hostname.

sudo vi /etc/hosts

Add your private IP or public IP depending on which one you want kubelet to identify as your default node IP. On Virtualbox, add the Host-only IP.

Control plane

Although etcd ports are included in the control plane section, you can also host your own etcd cluster externally or on custom ports.

Worker node(s)

Commands in a script:

!/bin/sh

product_uuid=$(sudo cat /sys/class/dmi/id/product_uuid)
sudo swapoff -a
sudo sed -i '/ swap / s/^/#/' /etc/fstab

echo " The Output is:"
echo " UUID: $product_uuid\n"
echo " Hostname: $(hostname)"

Installing a container runtime – Containerd

Documentation source: https://kubernetes.io/docs/setup/production-environment/container-runtimes/

Container runtime is the application responsible for running containers in Kubernetes.

Installation prerequisites

Forwarding IPv4 and letting iptables see bridged traffic:

cat <<EOF | sudo tee /etc/modules-load.d/k8s.conf
overlay
br_netfilter
EOF

sudo modprobe overlay
sudo modprobe br_netfilter

#sysctl params required by setup, params persist across reboots

cat <<EOF | sudo tee /etc/sysctl.d/k8s.conf
net.bridge.bridge-nf-call-iptables = 1
net.bridge.bridge-nf-call-ip6tables = 1
net.ipv4.ip_forward = 1
EOF

#Apply sysctl params without reboot

sudo sysctl --system

Verify that the br_netfilter, overlay modules are loaded by running:

lsmod | grep br_netfilter
lsmod | grep overlay

Verify that the net.bridge.bridge-nf-call-iptables, net.bridge.bridge-nf-call-ip6tables, and net.ipv4.ip_forward system variables are set to 1 in your sysctl config by running the following command:

sudo sysctl net.bridge.bridge-nf-call-iptables net.bridge.bridge-nf-call-ip6tables net.ipv4.ip_forward

cgroupfs or systemd driver?

If your Linux system uses systemd as the init system, then use systemd, there is no need to use 2 different cgroup drivers.

Since v 1.22 kubeadm defaults the cgroup driver to systemd.

Installing Containerd

Documentaion source : https://github.com/containerd/containerd/blob/main/docs/getting-started.md

After installing Containerd, you will also install runc and CNI plugins.

Step 1: Installing containerd

Download the binary for your amd64 or arm64 system from https://github.com/containerd/containerd/releases. Ensure to download the latest releases from the provided link and switch it in place of the one below.

cd /tmp 

#You can replace the version with the latest from github releases page

wget https://github.com/containerd/containerd/releases/download/v1.7.23/containerd-1.7.23-linux-amd64.tar.gz


# Extract the containerd download
sudo tar Cxzvf /usr/local containerd-1.7.23-linux-amd64.tar.gz

# For systemd system users you can add the unit file for containerd using either of the following commands:

sudo wget -O /usr/local/lib/systemd/system/containerd.service https://raw.githubusercontent.com/containerd/containerd/main/containerd.service

# Or if you get the error: No such file or directory this will work (Debian / Ubuntu):

sudo wget -O /usr/lib/systemd/system/containerd.service https://raw.githubusercontent.com/containerd/containerd/main/containerd.service


# Finally reload service daemons and enable containerd
sudo systemctl daemon-reload
sudo systemctl enable --now containerd

Step 2: Installing runc

Download the right version for your system: amd64 or arm64 . Assets download page : https://github.com/opencontainers/runc/releases

Download and install runc

wget https://github.com/opencontainers/runc/releases/download/v1.2.0/runc.amd64
sudo install -m 755 runc.amd64 /usr/local/sbin/runc

Step 3: Installing CNI plugins

Download the cni-plugins—.tgz archive from https://github.com/containernetworking/plugins/releases , and extract it under /opt/cni/bin:

Download and install for amd64

wget https://github.com/containernetworking/plugins/releases/download/v1.6.0/cni-plugins-linux-amd64-v1.6.0.tgz

#Extract it/Install it
sudo mkdir -p /opt/cni/bin
sudo tar Cxzvf /opt/cni/bin cni-plugins-linux-amd64-v1.6.0.tgz

Containerd endpoint: unix:///var/run/containerd/containerd.sock

Step 4 Configuring the systemd cgroup driver

Setting cgroup for containerd for each Virtual machine .
Cgroups (control groups) constrain resources allocated to Linux processes.

Based on how you installed containerd (using the binary as illustrated above), the default config file will not be available; you have to create it manually as shown below:

Switch to root first before running the next set of commands

#Switch to root
sudo su - 

# Create containerd config folder then #Generate a config file
sudo mkdir /etc/containerd && sudo touch /etc/containerd/config.toml && sudo containerd config default > /etc/containerd/config.toml 

# Edit the generated containerd service configuration file /etc/containerd/config.toml:
sudo vi /etc/containerd/config.toml

• Once the file is open find the section shown below and change to this SystemdCgroup = true . Make the value true and save. Easiest way is to search for SystemdCgroup and edit it in the file.

• Save /etc/containerd/config.toml
• Apply the changes by restarting containerd:
  

sudo systemctl restart containerd

You can log out of root or continue with the root shell, it is up to you.

#log out of root
exit

Install kubeadm, kubelet and kubectl

kubeadm: the tool used to bootstrap the Kubernetes cluster.

kubelet: the component that runs on all the machines in your cluster and does things like starting pods and containers.

kubectl: the command line utility used in talking to your cluster.

Update apt package index and install prerequisite apps .

Latest versions for the new stable Kubernetes can be installed by updating the keys and repository details as shown on the Kubernetes documentation here.

sudo apt update
sudo apt install -y apt-transport-https ca-certificates curl gpg

# Download the public signing key for the Kubernetes package repositories:
sudo mkdir -p -m 755 /etc/apt/keyrings

#The version below (v1.30) does not affect the Kubernetes version to be installed
curl -fsSL https://pkgs.k8s.io/core:/stable:/v1.30/deb/Release.key | sudo gpg --dearmor -o /etc/apt/keyrings/kubernetes-apt-keyring.gpg

# Add the Ubuntu Debian K8S repository. This version affects the kubernetes components that will be installed. 
# Change it (v1.30) to the current kubernetes version you'd like to install
echo 'deb [signed-by=/etc/apt/keyrings/kubernetes-apt-keyring.gpg] https://pkgs.k8s.io/core:/stable:/v1.30/deb/ /' | sudo tee /etc/apt/sources.list.d/kubernetes.list

sudo apt update

#install kubeadm kubelet and kubectl and hold their versions to prevent the system from updating them.

sudo apt install -y kubelet kubeadm kubectl
sudo apt-mark hold kubelet kubeadm kubectl

#Enable the kubelet service before running kubeadm:

sudo systemctl enable --now kubelet

You have added apt repository for packages intended for v1.30 . If you want to use the latest Kubernetes, go to the Kubernetes documentation and get the latest repository. Link: https://kubernetes.io/docs/setup/production-environment/tools/kubeadm/install-kubeadm/ . Or just change the version where indicated above.

Creating a cluster with kubeadm

Source docs : https://kubernetes.io/docs/setup/production-environment/tools/kubeadm/create-cluster-kubeadm/

Side note: Before running kubeadm init, you can pull all the images needed to bootstrap a kubeadm cluster:

sudo kubeadm config images pull

Initialize kubeadm cluster: To initialize the control-plane node, run:

sudo kubeadm init --apiserver-advertise-address=192.168.0.10 --pod-network-cidr=192.168.0.0/16 

Change the pod network cidr and your control plane’s IP as needed. You may also leave out these options and the default values will be used.

Install CNI – Calico

Source documentation: https://docs.tigera.io/calico/latest/getting-started/kubernetes/self-managed-onprem/onpremises

#Install the operator on your cluster.
kubectl create -f https://raw.githubusercontent.com/projectcalico/calico/v3.29.1/manifests/tigera-operator.yaml

# Download the custom resource needed to configure Calico.
curl https://raw.githubusercontent.com/projectcalico/calico/v3.29.1/manifests/custom-resources.yaml -O

# Apply the manifest to install Calico.
kubectl create -f custom-resources.yaml

If you wish to customize the Calico install, customize the downloaded custom-resources.yaml manifest locally.

Verify Calico installation in your cluster. Check/Confirm as the pods get created

watch kubectl get pods -n calico-system

Joining your other nodes to the cluster – Workers

SSH to the machine.

Become root (e.g. sudo su -)

Install a runtime if needed. We’ve already installed it up there.

Run the command that was output by kubeadm init. For example:

kubeadm join --token <token> <control-plane-host>:<control-plane-port> --discovery-token-ca-cert-hash sha256:<hash>

If you do not have the token, you can get it by running the following command on the control-plane node. kubeadm token create or print an already existing one with the full join command. Remeber to add sudo when using the join command

kubeadm token create --print-join-command

Tokens expire after 24 hours.You can create a new token by running the following command on the control-plane node:

Test your cluster – deploy nginx

In any folder, create a file called mynginx.yaml . In this file copy and paste the following:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 2
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:latest
        ports:
        - containerPort: 80

---
apiVersion: v1
kind: Service
metadata:
  name: nginx-service
spec:
  selector:
    app: nginx
  type: NodePort
  ports:
    - port: 80
      targetPort: 80
      nodePort: 30000

In the directory with the above file, run:

kubectl apply -f mynginx.yaml

On your local browser visit machine_ip:30000 .

You should be able to see the nginx welcome page.

To delete the deployment do:

kubectl delete -f mynginx.yaml

Extra commands

If you check the Kubelet status (sudo systemctl status kubelet) and it is not running chec journal logs for what might be wrong :

sudo journalctl -f -u kubelet

On WIndows disable Hyperv :

Disable-WindowsOptionalFeature -Online -FeatureName Microsoft-Hyper-V-Hypervisor

Source : https://learn.microsoft.com/en-us/troubleshoot/windows-client/application-management/virtualization-apps-not-work-with-hyper-v

I will upload a video with these steps. Check back soon. You can watch the video for installing Kubernetes via Kubeadm here.

Update Github Containerd , CNI Network plugins and RunC

Copy the following, update the different Github release versions to the current ones. Then run it on each node. Wherever there is wget, ensure you update to the latest version available on Github.

#go to temp directory
cd /tmp

# download the latest version of containerd

wget https://github.com/containerd/containerd/releases/download/v1.7.23/containerd-1.7.23-linux-amd64.tar.gz

#Extract it

sudo tar Cxzvf /usr/local containerd-1.7.23-linux-amd64.tar.gz


sudo systemctl daemon-reload
sudo systemctl enable --now containerd

# Updateing/Installing runc

wget https://github.com/opencontainers/runc/releases/download/v1.2.0/runc.amd64
sudo install -m 755 runc.amd64 /usr/local/sbin/runc


# Cni - Download an extract

wget https://github.com/containernetworking/plugins/releases/download/v1.6.0/cni-plugins-linux-amd64-v1.6.0.tgz

sudo tar Cxzvf /opt/cni/bin cni-plugins-linux-amd64-v1.6.0.tgz

Then reboot all your nodes.