Load Balancers

Load Balancers

A load balancer distributes incoming client requests across a pool of backend servers, preventing any single server from becoming overwhelmed and providing automatic failover when a backend goes down. On Ubuntu, HAProxy is the standard software load balancer — it is fast, stable, well-documented, and handles tens of thousands of concurrent connections. HAProxy is used by companies like GitHub, Stack Overflow, and Reddit for production traffic.

Load balancing concepts

HAProxy installation and configuration

sudo apt update
sudo apt install -y haproxy
sudo nano /etc/haproxy/haproxy.cfg

/etc/haproxy/haproxy.cfg — HTTP load balancer example

global
    log /dev/log local0
    chroot /var/lib/haproxy
    user haproxy
    group haproxy
    daemon

defaults
    log     global
    mode    http
    option  httplog
    option  dontlognull
    timeout connect 5000ms
    timeout client  50000ms
    timeout server  50000ms
    option  forwardfor    # Add X-Forwarded-For header
    option  http-server-close

frontend web_frontend
    bind *:80
    bind *:443 ssl crt /etc/ssl/certs/example.com.pem
    redirect scheme https if !{ ssl_fc }    # Force HTTPS
    default_backend web_servers

backend web_servers
    balance roundrobin
    option httpchk GET /health     # Health check endpoint
    http-check expect status 200

    server web-01 192.168.1.10:8080 check inter 5s fall 2 rise 3
    server web-02 192.168.1.11:8080 check inter 5s fall 2 rise 3
    server web-03 192.168.1.12:8080 check inter 5s fall 2 rise 3

listen stats
    bind *:8404
    stats enable
    stats uri /stats
    stats auth admin:StrongPassword42!

# Validate config before applying:
haproxy -c -f /etc/haproxy/haproxy.cfg

# Apply changes:
sudo systemctl reload haproxy

# Zero-downtime config reload:
sudo haproxy -f /etc/haproxy/haproxy.cfg -p /var/run/haproxy.pid -st $(cat /var/run/haproxy.pid)

HAProxy statistics and health checks

# Access stats page (if enabled in config):
# http://YOUR_LB_IP:8404/stats
# Shows: connections, bytes, response times, server status

# Check HAProxy status via CLI:
echo "show stat" | sudo socat stdio /var/lib/haproxy/stats | cut -d',' -f1,2,18,19 | column -t -s ','

# View backend server status:
echo "show servers state" | sudo socat stdio /var/lib/haproxy/stats

show servers state output

1 web_servers 1 web-01 2 0 0 0 0 ...  # 2 = UP
1 web_servers 2 web-02 2 0 0 0 0 ...  # 2 = UP
1 web_servers 3 web-03 0 0 0 0 0 ...  # 0 = DOWN (failed health check)

Keepalived for HA load balancers

# Keepalived provides a floating VIP between two HAProxy nodes:
sudo apt install -y keepalived

sudo nano /etc/keepalived/keepalived.conf

/etc/keepalived/keepalived.conf (on primary LB)

vrrp_script chk_haproxy {
    script "systemctl is-active haproxy"
    interval 2
    fall 3
}

vrrp_instance VI_1 {
    state MASTER
    interface eth0
    virtual_router_id 51
    priority 200           # Secondary uses 100
    authentication {
        auth_type PASS
        auth_pass mypassword42
    }
    virtual_ipaddress {
        10.0.0.100/24     # Floating VIP
    }
    track_script {
        chk_haproxy        # Drop to BACKUP if haproxy fails
    }
}

Conclusion

Configure health checks with fall 2 rise 3: a server is marked DOWN after 2 failed checks (fast failure detection) but only marked UP again after 3 consecutive passing checks (prevents flapping). The stats page at :8404/stats is indispensable for monitoring — it shows real-time connection counts, error rates, and server status in a browser-accessible format. Pair HAProxy with Keepalived for a fully redundant load balancer layer that handles HAProxy itself failing.