Demo Load Balancer

In this tutorial, you'll learn how to configure Nginx as a load balancer using three common algorithms:

1. Round Robin: Distributes requests evenly across all backend servers.
2. Weighted Round Robin: Assigns traffic proportionally based on server weights.
3. IP Hash: Routes each client IP to the same backend to maintain session persistence.

Load Balancing Algorithm Comparison

Environment Setup

We have three Ubuntu nodes in our test lab:

• nginx: Nginx load balancer (192.230.202.10)
• node01 & node02: Apache backend servers

On each backend, Apache is running and UFW is active but only port 22 is open by default:

# On node01 (repeat on node02)
root@node01:~# systemctl status apache2
● apache2.service - The Apache HTTP Server
   Active: active (running) ...

root@node01:~# ufw status
Status: active

To                         Action    From
--                         ------    ----
22/tcp                     ALLOW     Anywhere
22/tcp (v6)                ALLOW     Anywhere (v6)

# On node01
root@node01:~# ufw allow from 192.230.202.10 proto tcp to any port 80

# Verify the rule
root@node01:~# ufw status
Status: active

To                         Action    From
--                         ------    ----
22/tcp                     ALLOW     Anywhere
192.230.202.10 80/tcp      ALLOW     Anywhere
22/tcp (v6)                ALLOW     Anywhere (v6)

# Repeat on node02
root@node02:~# ufw allow from 192.230.202.10 proto tcp to any port 80

Configuring the Load Balancer

On the nginx node, create or edit your site configuration:

root@nginx:~# cd /etc/nginx/sites-available/
root@nginx:/etc/nginx/sites-available# vim apache-app

Start with a basic server block:

server {
    listen 80;
    server_name apache.example.com;
    root /var/www/html;

    # Default index files
    index index.html index.htm index.nginx-debian.html;

    location / {
        try_files $uri $uri/ =404;
    }
}

1. Round Robin

Add an upstream block for your backend pool:

# Upstream (Round Robin)
upstream apache_example {
    server 192.230.202.12:80;
    server 192.230.202.3:80;
}

server {
    listen 80;
    server_name apache.example.com;
    root /var/www/html;
    index index.html index.htm index.nginx-debian.html;

    location / {
        proxy_pass http://apache_example;
    }
}

Enable the site and reload Nginx:

root@nginx:/etc/nginx/sites-available# nginx -t
root@nginx:/etc/nginx/sites-available# ln -s apache-app /etc/nginx/sites-enabled/
root@nginx:/etc/nginx/sites-available# nginx -s reload

Test with curl to see alternating responses:

root@nginx:~# curl localhost
<p>Served by node02</p>
root@nginx:~# curl localhost
<p>Served by node01</p>

2. Weighted Round Robin

Adjust the upstream block to assign server weights:

# Upstream (Weighted Round Robin)
upstream apache_example {
    server 192.230.202.12:80 weight=10;
    server 192.230.202.3:80  weight=1;
}

Reload Nginx to apply:

root@nginx:~# nginx -s reload

Now about 10× more requests go to the higher-weighted server.

3. IP Hash

Enable ip_hash for session stickiness by client IP:

# Upstream (IP Hash)
upstream apache_example {
    ip_hash;
    server 192.230.202.12:80;
    server 192.230.202.3:80;
}

Reload Nginx:

root@nginx:~# nginx -s reload

You can extend this setup with health checks, SSL termination, or alternative Nginx modules.

Links and References

• Nginx Load Balancing
• UFW Documentation
• Apache HTTP Server Docs