🚀 Bare-Metal Kubernetes Cluster with kubeadm on Ubuntu

Complete Platform for Development, DevOps & Observability (Includes full installation scripts)

This repository provides a full guide and reference architecture for setting up a production-ready Kubernetes cluster on bare-metal Ubuntu servers using kubeadm. It also includes a full ecosystem of development and DevOps tooling—CI/CD, observability, message queues, storage, GitOps, monitoring, and more.

Overview

This repository provides a complete guide and automation scripts for provisioning a bare-metal Kubernetes cluster on Ubuntu using kubeadm, along with a full suite of Dev/DevOps platforms:

• GitOps: ArgoCD

• CI/CD + VCS: Jenkins, GitLab

• Container registry: Harbor

• Message Queue: RabbitMQ

• Databases / Cache: PostgreSQL, Redis

• Object storage: MinIO

• API Gateway / Ingress: Traefik or NGINX Ingress

• Monitoring: Prometheus, Grafana

• Tracing & Telemetry: OpenTelemetry

• Logging (optional): Loki

K8S Cluster Architecture

Component	Count	Notes
Control Plane Nodes	1-3	kube-apiserver, etcd, controller-manager, scheduler
Worker Nodes	2-N	Run workloads with all the listing services
Networking	Calico	CNI, IpTables or eBPF
Storage	Rook & Ceph	Persistent volumes for services

Prerequisites

• Ubuntu Server 22.04/24.04 (all nodes)

• 3+ nodes with:

  • 2+ CPUs

  • 8+ GB RAM minimum

  • 100+ GB disk(each worker node equires one or more raw, dedicated disks for ceph)

Repository Structure

• install: all installation scripts with steps

  • 1-prepare-all-node.sh: install all the necessary packages in all nodes to prepare for Kubernetes installation.
  • 2-1-init-master-control-plane.sh: install k8s in master node
  • 2-2-kubectl-master-control-plane.sh: setup kubectl autocomplete in profile and install helm
  • 2-3-calico-master-control-plane.sh: install calico networking with iptables filter
  • 3-join-control-plane.sh: other control plane node join
  • 4-join-worker.sh: worker node join
  • 5-reset-worker.sh: reset environment and uninstall all packages in worker node
  • 6-reset-master-control-plane.sh: reset environment and uninstall all packages in control plane node

• pem: crt and key file certificate for k8s

• services: all services includes develop, devops and observability

• tf: Terraform configuration files (in progress)