Website Uptime Monitoring

Website Uptime Monitoring

Uptime Monitoring — basic observability layer: URL availability check via HTTP requests at intervals. Detects unavailability fact, not cause. Must be supplemented with infrastructure monitoring and error logging.

Tools

SaaS (no infrastructure):

• UptimeRobot — free up to 50 monitors, 5 min interval. Paid — 1 min.
• Better Uptime — aggregates multiple locations, status page, on-call scheduling
• Pingdom — Enterprise-level, RUM, API monitoring
• StatusCake — good price/quality ratio

Self-hosted:

• Uptime Kuma — Docker, web interface, many check types
• Gatus — YAML configuration, Kubernetes-friendly
• Blackbox Exporter + Prometheus + Grafana — for those with existing Prometheus

Uptime Kuma via Docker

# docker-compose.yml
services:
  uptime-kuma:
    image: louislam/uptime-kuma:latest
    restart: unless-stopped
    ports:
      - "3001:3001"
    volumes:
      - ./uptime-kuma-data:/app/data

Gatus: declarative monitoring

# gatus/config.yaml
endpoints:
  - name: Main Site
    url: https://mysite.com
    interval: 1m
    conditions:
      - "[STATUS] == 200"
      - "[RESPONSE_TIME] < 2000"
      - "[CERTIFICATE_EXPIRATION] > 168h"  # > 7 days until expiry

  - name: API Health
    url: https://api.mysite.com/health
    interval: 30s
    conditions:
      - "[STATUS] == 200"
      - "[BODY].status == UP"
      - "[RESPONSE_TIME] < 500"

  - name: Checkout Flow
    url: https://mysite.com/cart
    interval: 5m
    conditions:
      - "[STATUS] == 200"
      - "[BODY] pat *Add to cart*"

alerting:
  telegram:
    token: $TELEGRAM_BOT_TOKEN
    id: $TELEGRAM_CHAT_ID
    default-alert:
      enabled: true
      failure-threshold: 2
      success-threshold: 1

Multi-region monitoring

Check from multiple locations: otherwise might miss regional failure. Better Uptime and Pingdom include this in base plan.

For self-hosted — multiple Gatus instances in different regions + central aggregator.

Telegram alerting

# Simple webhook handler for alerts
import requests
import os

def send_alert(message: str, is_recovery: bool = False):
    emoji = "✅" if is_recovery else "🚨"
    requests.post(
        f"https://api.telegram.org/bot{os.environ['BOT_TOKEN']}/sendMessage",
        json={
            "chat_id": os.environ['CHAT_ID'],
            "text": f"{emoji} {message}",
            "parse_mode": "HTML",
        }
    )

What to monitor

Setting up Uptime Kuma or Gatus for site (10–20 endpoints) — 2–4 hours.