Clusters from Scratch

Step-by-Step Instructions for Building Your First High-Availability Cluster

Abstract

This document provides a step-by-step guide to building a simple high-availability cluster using Pacemaker.

The example cluster will use:

• AlmaLinux 9 as the host operating system

• Corosync to provide messaging and membership services

• Pacemaker 2 as the cluster resource manager

• DRBD as a cost-effective alternative to shared storage

• GFS2 as the cluster filesystem (in active/active mode)

Given the graphical nature of the install process, a number of screenshots are included. However, the guide is primarily composed of commands, the reasons for executing them, and their expected outputs.

Table of Contents

• 1. Introduction
  • 1.1. The Scope of This Document
  • 1.2. What Is Pacemaker?
    • 1.2.1. Cluster Architecture
    • 1.2.2. Pacemaker Architecture
    • 1.2.3. Node Redundancy Designs
• 2. Installation
  • 2.1. Install AlmaLinux 9
    • 2.1.1. Boot the Install Image
    • 2.1.2. Installation Options
    • 2.1.3. Configure Network
    • 2.1.4. Configure Disk
    • 2.1.5. Configure Time Synchronization
    • 2.1.6. Root Password
    • 2.1.7. Finish Install
  • 2.2. Configure the OS
    • 2.2.1. Verify Networking
    • 2.2.2. Login Remotely
    • 2.2.3. Apply Updates
    • 2.2.4. Use Short Node Names
  • 2.3. Repeat for Second Node
  • 2.4. Configure Communication Between Nodes
    • 2.4.1. Configure Host Name Resolution
    • 2.4.2. Configure SSH
• 3. Set up a Cluster
  • 3.1. Simplify Administration With a Cluster Shell
  • 3.2. Install the Cluster Software
  • 3.3. Configure the Cluster Software
    • 3.3.1. Allow cluster services through firewall
    • 3.3.2. Enable pcs Daemon
    • 3.3.3. Configure Corosync
  • 3.4. Explore pcs
• 4. Start and Verify Cluster
  • 4.1. Start the Cluster
  • 4.2. Verify Corosync Installation
  • 4.3. Verify Pacemaker Installation
  • 4.4. Explore the Existing Configuration
• 5. Configure Fencing
  • 5.1. What is Fencing?
  • 5.2. Choose a Fence Device
  • 5.3. Configure the Cluster for Fencing
  • 5.4. Example
• 6. Create an Active/Passive Cluster
  • 6.1. Add a Resource
  • 6.2. Perform a Failover
  • 6.3. Prevent Resources from Moving after Recovery
• 7. Add Apache HTTP Server as a Cluster Service
  • 7.1. Install Apache
  • 7.2. Create Website Documents
  • 7.3. Enable the Apache Status URL
  • 7.4. Configure the Cluster
  • 7.5. Ensure Resources Run on the Same Host
  • 7.6. Ensure Resources Start and Stop in Order
  • 7.7. Prefer One Node Over Another
  • 7.8. Move Resources Manually
• 8. Replicate Storage Using DRBD
  • 8.1. Install the DRBD Packages
  • 8.2. Allocate a Disk Volume for DRBD
  • 8.3. Configure DRBD
  • 8.4. Initialize DRBD
  • 8.5. Populate the DRBD Disk
  • 8.6. Configure the Cluster for the DRBD device
  • 8.7. Configure the Cluster for the Filesystem
  • 8.8. Test Cluster Failover
• 9. Convert Storage to Active/Active
  • 9.1. Install Cluster Filesystem Software
  • 9.2. Configure the Cluster for the DLM
  • 9.3. Create and Populate GFS2 Filesystem
  • 9.4. Reconfigure the Cluster for GFS2
  • 9.5. Clone the Filesystem Resource
  • 9.6. Test Failover
• 10. Configuration Recap
  • 10.1. Final Cluster Configuration
  • 10.2. Node List
  • 10.3. Cluster Options
  • 10.4. Resources
    • 10.4.1. Default Options
    • 10.4.2. Fencing
    • 10.4.3. Service Address
    • 10.4.4. DRBD - Shared Storage
    • 10.4.5. Cluster Filesystem
    • 10.4.6. Apache
• 11. Sample Corosync Configuration
• 12. Further Reading

Index

• Index

• Search Page