Self-Managed with Kubeadm

Self-Managed Kubernetes Cluster Using kubeadm¶

Overview¶

This guide explains how to build a self-managed Kubernetes cluster using kubeadm combined with the infra-bootstrap automation scripts.

This method follows the real sequence used by cluster operators:

• Promote a machine to become a control plane
• Optionally add more control planes
• Prepare worker nodes
• Join all nodes to the cluster
• Install a CNI plugin
• Validate the cluster

1. Create the First Control Plane¶

Run the following on the machine you want to become the primary control plane:

curl -sL https://raw.githubusercontent.com/ibtisam-iq/infra-bootstrap/main/scripts/kubernetes/K8s-Control-Plane-Init.sh | sudo bash

This script:

• Prepares the node
• Cleans previous kubeadm data
• Starts container runtime + kubelet
• Runs kubeadm init
• Configures kubeconfig
• Produces join commands for workers and additional control planes

After completion, your cluster exists — but with a single control-plane.

2. (Optional) Create Additional Control Planes¶

If you want multi-control-plane (HA) setup:

Run the same script on your additional control plane nodes:

curl -sL https://raw.githubusercontent.com/ibtisam-iq/infra-bootstrap/main/scripts/kubernetes/K8s-Control-Plane-Init.sh | sudo bash

During initialization of the first node, kubeadm prints the control-plane join command, e.g.:

kubeadm join <API-SERVER-IP>:6443 \
  --token <token> \
  --discovery-token-ca-cert-hash sha256:<hash> \
  --control-plane

Copy that and run it on your additional control-plane nodes.

If the token expired:

kubeadm token create --print-join-command

3. Prepare Worker Nodes¶

On every node you want to convert into a worker, run:

curl -sL https://raw.githubusercontent.com/ibtisam-iq/infra-bootstrap/main/scripts/kubernetes/K8s-Node-Init.sh | sudo bash

This script:

• Installs containerd
• Installs kubelet / kubeadm / kubectl
• Applies kernel modules + sysctl
• Enables required services

Each machine becomes Kubernetes-ready.

4. Join Worker Nodes¶

Joining nodes uses the command printed during the first control-plane init:

kubeadm join <API-SERVER-IP>:6443 \
  --token <token> \
  --discovery-token-ca-cert-hash sha256:<hash>

Run this on each worker node.

If the token expires:

kubeadm token create --print-join-command

5. Install a CNI Plugin (Required for Networking)¶

Kubernetes will not schedule pods until a CNI is installed.

To deploy a CNI via infra-bootstrap:

curl -sL https://raw.githubusercontent.com/ibtisam-iq/infra-bootstrap/main/scripts/kubernetes/k8s-cni-setup.sh | bash

This script:

• Prompts you to choose Calico, Flannel, or Weave
• Applies the correct manifest
• Verifies networking readiness

Important: Install the CNI only once, on the first control-plane node.

6. Verify the Cluster¶

Check nodes:¶

kubectl get nodes

Check system pods:¶

kubectl get pods -A

Check control-plane health:¶

kubectl get componentstatuses

Check API server readiness:¶

kubectl get --raw='/readyz?verbose'

If everything is Running/Ready — your Kubernetes cluster is healthy.

Reset / Rebuild (if needed)¶

sudo kubeadm reset -f
sudo systemctl restart containerd
sudo rm -rf ~/.kube

Official Reference¶

Kubeadm Upstream Documentation: https://kubernetes.io/docs/setup/production-environment/tools/kubeadm/