NGINX Cookbook

Chapter 12. High-Availability Deployment Modes

Introduction

Fault-tolerant architecture separates systems into identical, independent stacks. Load balancers like NGINX are employed to distribute load, ensuring that what’s provisioned is utilized. The core concepts of high availability are load balancing over multiple active nodes or an active-passive failover. Highly available applications have no single points of failure; every component must use one of these concepts, including the load balancers themselves. For us, that means NGINX. NGINX is designed to work in either configuration: multiple active or active-passive failover. This chapter details techniques on how to run multiple NGINX servers to ensure high availability in your load-balancing tier.

NGINX HA Mode

Problem

You need a highly available load-balancing solution.

Solution

Use NGINX Plus’s highly available (HA) mode with keepalived by installing the nginx-ha-keepalived package from the NGINX Plus repository.

Discussion

The nginx-ha-keepalived package is based on keepalived and manages a virtual IP address exposed to the client. Another process is run on the NGINX server that ensures that NGINX Plus and the keepalived process are running. Keepalived is a process that utilizes the Virtual Router Redundancy Protocol (VRRP), sending small messages often referred to as heartbeats, to the backup server. If the backup server does not receive the heartbeat for three consecutive periods, the backup server initiates the failover, ...