How to Monitor Your HAProxy Load Balancer with Vigilmon

HAProxy is the industry standard for high-performance TCP and HTTP load balancing. It handles traffic distribution for some of the world's most demanding production systems. But HAProxy failures are catastrophic: when your load balancer goes down, all backend services become unreachable simultaneously.

This guide shows you how to monitor HAProxy with Vigilmon — from the built-in stats page to backend server health.

---

HAProxy's Built-in Health Features

HAProxy has two native monitoring mechanisms:

1. Stats Page

HAProxy can expose a web-based stats page:

# In your haproxy.cfg
listen stats
    bind *:8404
    stats enable
    stats uri /stats
    stats refresh 30s
    stats auth admin:strongpassword

This gives you real-time data on:

• Active/down backend servers
• Current connections and queue depth
• Request rates and response times
• Health check status per backend

2. Health Check on Frontend

You can expose a simple health endpoint for external monitoring:

frontend health-check
    bind *:8405
    default_backend health-backend

backend health-backend
    server health-server 127.0.0.1:9000 check
    errorfile 503 /etc/haproxy/errors/503-health.http

---

Step 1: Enable HAProxy Stats API

Enable the CSV stats endpoint:

listen stats
    bind *:8404
    stats enable
    stats uri /stats
    stats auth monitoring:$STATS_PASSWORD
    stats hide-version
    stats refresh 10s

Now curl http://localhost:8404/stats;csv returns machine-readable stats.

---

Step 2: Create a Health Proxy Application

Node.js example:

const express = require('express');
const axios = require('axios');
const csvParser = require('csv-parse/sync');

const app = express();

app.get('/health/haproxy', async (req, res) => {
  try {
    const response = await axios.get(
      `http://localhost:8404/stats;csv`,
      {
        auth: { username: 'monitoring', password: process.env.STATS_PASSWORD },
        timeout: 5000
      }
    );

    // Parse CSV
    const rows = csvParser.parse(response.data, { columns: true, skip_empty_lines: true });

    // Find backend servers with DOWN status
    const backends = rows.filter(r => r.svname !== 'FRONTEND' && r.svname !== 'BACKEND');
    const downServers = backends.filter(r => r.status === 'DOWN');

    if (downServers.length > 0) {
      return res.status(503).json({
        status: 'degraded',
        down_servers: downServers.map(s => ({
          backend: s['# pxname'],
          server: s.svname,
          last_check: s.lastchk
        }))
      });
    }

    // Check if all backends are UP
    const allDown = backends.every(r => r.status === 'DOWN');
    if (allDown && backends.length > 0) {
      return res.status(503).json({ status: 'all_backends_down' });
    }

    res.json({
      status: 'ok',
      total_servers: backends.length,
      active_servers: backends.filter(r => r.status === 'UP').length
    });
  } catch (err) {
    res.status(503).json({ status: 'error', message: err.message });
  }
});

app.listen(3000);

Python example:

from fastapi import FastAPI, Response
import httpx
import csv
import io
import json

app = FastAPI()

@app.get("/health/haproxy")
async def health_haproxy():
    try:
        async with httpx.AsyncClient(timeout=5.0) as client:
            r = await client.get(
                "http://localhost:8404/stats;csv",
                auth=("monitoring", "password")
            )
            r.raise_for_status()

        reader = csv.DictReader(io.StringIO(r.text))
        servers = [row for row in reader 
                   if row['svname'] not in ('FRONTEND', 'BACKEND')]
        down = [s for s in servers if s['status'] == 'DOWN']

        if down:
            return Response(
                content=json.dumps({"status": "degraded", "down": len(down)}),
                status_code=503
            )

        return {"status": "ok", "servers": len(servers)}
    except Exception as e:
        return Response(content=str(e), status_code=503)

---

Step 3: Add Vigilmon HTTP Monitors

1. Log in to vigilmon.online → Add Monitor
2. Monitor 1 — HAProxy health:
   • URL: https://your-domain.com/health/haproxy
   • Alert if: Status != 200
3. Monitor 2 — Public frontend check:
   • URL: https://your-domain.com (the HAProxy frontend)
   • Alert if: Status != 200 or > 2000ms
4. Interval: 60 seconds

Vigilmon checks from multiple global regions — so you'll know if HAProxy is failing in one geographic area but not another.

---

Step 4: SSL Certificate Monitoring

If HAProxy terminates SSL (common setup), add a Vigilmon SSL certificate monitor:

1. Add Monitor → Type: SSL Certificate
2. Hostname: your-domain.com
3. Alert: If expires in < 14 days

---

HAProxy Monitoring Coverage Table

Monitor	Endpoint	Alerts On
HAProxy health	/health/haproxy	Backend servers DOWN
Public frontend	https://your-domain.com	HAProxy itself down
SSL certificate	your-domain.com	Cert expiry < 14 days

---

Common HAProxy Failure Patterns

Backend server goes offline: One of three web servers in a backend pool crashed. HAProxy's own health checks marked it DOWN, but the traffic continued to the remaining 2 servers. Vigilmon's backend monitor immediately alerted on the degraded state — allowing the team to investigate before the remaining 2 servers were overloaded.

HAProxy process crash: HAProxy's master process died after a misconfigured reload. All traffic stopped. The public frontend monitor caught it within 60 seconds.

SSL cert expires on HAProxy: A wildcard SSL cert wasn't renewed before expiry. HAProxy terminated connections with an expired cert. Vigilmon's SSL monitor would have alerted 14 days earlier.

---

Conclusion

HAProxy is a critical single point of failure for your infrastructure. Vigilmon's multi-region HTTP monitoring ensures you know the moment HAProxy or any backend server fails — not when your users start reporting problems.

Start monitoring your HAProxy setup free at vigilmon.online