Kubernetes Installation (kubeadm - v1.34)¶

This section describes how to install Kubernetes components and initialize the cluster using kubeadm.

At the end of this section, a single-node control plane will be up and running, ready to accept worker nodes.

Prerequisites¶

Before proceeding, ensure that: - the operating system is prepared (see OS Preparation) - swap is disabled - containerd is running - required kernel modules and sysctl parameters are applied

All commands below must be executed as root or with sudo.

Install Kubernetes packages¶

Add Kubernetes APT repository (v1.34)¶

Install required packages

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

Create keyrings directory (if not present)

sudo mkdir -p -m 755 /etc/apt/keyrings

Add CPG key

curl -fsSL https://pkgs.k8s.io/core:/stable:/v1.34/deb/Release.key \
| sudo gpg --dearmor -o /etc/apt/keyrings/kubernetes-apt-keyring.gpg

Add Kubernetes APT repository

echo "deb [signed-by=/etc/apt/keyrings/kubernetes-apt-keyring.gpg] \
https://pkgs.k8s.io/core:/stable:/v1.34/deb/ /" \
| sudo tee /etc/apt/sources.list.d/kubernetes.list

Update package index

sudo apt-get update

Install Kubernetes components (pinned version)¶

Select the Kubernetes version to install:

apt-cache madison kubeadm

Expected output:

kubeadm | 1.34.3-1.1 | https://pkgs.k8s.io/core:/stable:/v1.34/deb  Packages
kubeadm | 1.34.2-1.1 | https://pkgs.k8s.io/core:/stable:/v1.34/deb  Packages
kubeadm | 1.34.1-1.1 | https://pkgs.k8s.io/core:/stable:/v1.34/deb  Packages
kubeadm | 1.34.0-1.1 | https://pkgs.k8s.io/core:/stable:/v1.34/deb  Packages

Install kubeadm, kubelet and kubectl using the selected version (example: 1.34.3-1.1):

sudo apt-get install -y \
kubelet=1.34.3-1.1 \
kubeadm=1.34.3-1.1 \
kubectl=1.34.3-1.1

Hold package versions

Prevent automatic upgrades to avoid version drift:

sudo apt-mark hold kubelet kubeadm kubectl

Verify installation¶

kubeadm version
kubectl version --client
kubelet --version

Expected output: - all components report v1.34.x

Configure kubelet¶

Kubelet behavior before cluster initialization¶

At this stage, kubelet is installed but the cluster is not yet initialized.

It is expected that:

kubelet is running
kubelet repeatedly fails and retries
the node is not yet registered in the cluster

This behavior is normal and will be resolved automatically after running kubeadm init.

No manual kubelet configuration is required.

Initialize the control plane¶

Run the following command on the control plane node only:

sudo kubeadm init \
--apiserver-advertise-address=192.168.1.53 \
--pod-network-cidr=10.244.0.0/16

The Pod CIDR must match the CNI configuration that will be installed later.

Save the join command¶

At the end of the output, kubeadm prints a kubeadm join command.

⚠️ Save this command — it will be required to join worker nodes later.

Configure kubectl access¶

Configure kubectl for the current user:

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

Verify access:

kubectl get nodes

Verify control plane status¶

Check core components:

kubectl get pods -n kube-system

Expected state:

kube-apiserver running
kube-controller-manager running
kube-scheduler running
etcd running

Some pods may be in Pending state until a CNI is installed — this is normal.