Master Nginx Load Balancing: Step‑by‑Step Configuration Guide

You can place Nginx in front of a web application as a load balancer.

For example, if your enterprise application runs on Apache (or Tomcat), you can deploy a second instance of the application on different servers and let Nginx balance traffic between the two Apache/Tomcat servers.

If you are new to Nginx, it is important to understand the differences between Nginx, Apache, and Nginx architecture.

Nginx supports three types of load balancing:

round-robin – the default algorithm that cycles through servers.

least-connected – sends requests to the server with the fewest active connections.

ip-hash – uses the client’s IP address to consistently route requests to the same server.

1. Define upstream and proxy_pass in the Nginx config file

For load balancing you need to add two directives: upstream and proxy_pass .

Upstream: give a unique name (e.g., the application name) and list all servers that Nginx will balance.

<code>upstream crmdev {
    server 192.168.101.1;
    server 192.168.101.2;
}</code>

If servers listen on non‑default ports, specify the port number:

<code>upstream crmdev {
    server 192.168.101.1:8080;
    server 192.168.101.2:8080;
}</code>

proxy_pass: reference the upstream name in a

location

block inside a

server

block.

<code>server {
    listen 80;
    location / {
        proxy_pass http://crmdev;
    }
}</code>

Note: In this example Nginx itself listens on port 80.

You can also use

proxy_pass

to reverse‑proxy Apache/PHP.

2. Define upstream and proxy_pass in the default Nginx configuration

Usually the definitions belong in the

http

context:

<code>http {
    upstream crmdev {
        server 192.168.101.1:8080;
        server 192.168.101.2:8080;
    }
    server {
        listen 80;
        location / {
            proxy_pass http://crmdev;
        }
    }
}</code>

If you edit the supplied

default.conf

, you do not need the outer

http

block because it is already present.

For HTTPS replace the

http

context with

https

and use

https://

in

proxy_pass

. Otherwise you may see an error like:

<code>Starting nginx: nginx: [emerg] "http" directive is not allowed here in /etc/nginx/conf.d/default.conf:1</code>

3. Set the least‑connected algorithm for minimal connections

Add the

least_conn

directive at the top of the upstream block:

<code>upstream crmdev {
    least_conn;
    server 192.168.101.1:8080;
    server 192.168.101.2:8080;
}</code>

If several servers have similar connection counts, Nginx falls back to weighted round‑robin.

4. Configure a sticky (session‑persistent) algorithm

The round‑robin and least‑connected methods do not guarantee that subsequent requests from the same client go to the same server. For session‑dependent applications, use

ip_hash

:

<code>upstream crmdev {
    ip_hash;
    server 192.168.101.1:8080;
    server 192.168.101.2:8080;
}</code>

For IPv4 the hash uses the first three octets; for IPv6 it uses the full address.

5. Assign weights to individual servers

You can give specific servers a higher weight (default weight is 1). Example with five servers where the third has weight 2:

<code>upstream crmdev {
    server 192.168.101.1:8080;
    server 192.168.101.2:8080;
    server 192.168.101.3:8080 weight 2;
    server 192.168.101.4:8080;
    server 192.168.101.5:8080;
}</code>

This means that out of every six new requests, two will be sent to server 3, while the others receive one each, allowing more load on a more powerful server. Weights can also be used with

least_conn

and

ip_hash

.

6. Configure per‑server timeout options – max_fails and fail_timeout

You can set

max_fails

and

fail_timeout

for each server:

<code>upstream crmdev {
    server 192.168.101.1:8080 max_fails=3 fail_timeout=30s;
    server 192.168.101.2:8080;
    server 192.168.101.3:8080 weight 2;
    server 192.168.101.4:8080;
    server 192.168.101.5:8080;
}</code>

The default

fail_timeout

is 10 seconds; the example sets it to 30 seconds. After the defined number of failures (here 3), the server is marked unavailable for 30 seconds.

The default

max_fails

is 1; the example raises it to 3, meaning three consecutive connection failures trigger the timeout.

7. Add a backup server to the Nginx load‑balancing pool

Mark a server as a backup by adding the

backup

keyword:

<code>upstream crmdev {
    server 192.168.101.1:8080 max_fails=3 fail_timeout=30s;
    server 192.168.101.2:8080;
    server 192.168.101.3:8080 weight 2;
    server 192.168.101.4:8080;
    server 192.168.101.5:8080 backup;
}</code>

The fifth server will only receive traffic if all other four servers are down.