Building a High-Availability Kubernetes Cluster On-Premises: A Step-by-Step Guide

High Availability (HA) is crucial for production-grade Kubernetes environments. This guide details setting up a highly available Kubernetes cluster on-premises using HAProxy and Keepalived across three master nodes.

Prerequisites

• Three servers (physical or virtual) are designated as master nodes.
• The YUM package manager was installed (typical for CentOS/RHEL systems).
• Sudo privileges on all three servers.

Step 1: Install HAProxy and Keepalived

First, install HAProxy and Keepalived on each master node. Execute the following command on each node:

sudo yum install haproxy keepalived -y

To download Keepalived without installing (for offline installation scenarios), use:

sudo yum install --downloadonly --downloaddir=/tmp/ keepalived

Step 2: Configure Keepalived

Script for API Server Health Checks

1. Create the health check script on all master nodes at /etc/keepalived/check_apiserver.sh:

sudo vi /etc/keepalived/check_apiserver.sh

#…