Monitor Your Home Network Free: Uptime Kuma + n8n Alerts

This page may contain affiliate links.

Your broadband drops at 2am. The router reboots, the NAS goes offline, and you don't find out until you try to load a file the next morning. Consumer routers are useless at telling you this happened — they just quietly reconnect and pretend nothing occurred. The fix is a self-hosted monitoring stack, and the two best free tools for the job are Uptime Kuma (for the checking) and n8n (for the alerting logic). Together they'll watch every device on your LAN, ping you the second something falls over, and cost you nothing beyond the electricity to run them.

This guide assumes you've got Docker running on something — a Raspberry Pi 4/5, an old mini PC, a Synology, or a VM on your main machine. If you haven't, any of those will do; a Pi 4 with 4GB of RAM is plenty for both containers.

Why Uptime Kuma beats the paid alternatives

Uptime Kuma is an open-source, self-hosted status monitor. It does the same job as UptimeRobot or Pingdom's free tiers, but without the monitor limits, the 5-minute check intervals, or the nagging upgrade prompts. You get 20-second checks, HTTP keyword matching, TCP port checks, DNS queries, ping, and push heartbeats — all from a single container.

Spin it up with a compose file in your home lab directory:

services:
  uptime-kuma:
    image: louislam/uptime-kuma:1
    container_name: uptime-kuma
    restart: unless-stopped
    ports:
      - '3001:3001'
    volumes:
      - ./uptime-kuma-data:/app/data

Run docker compose up -d, then hit http://your-pi-ip:3001 and create an admin account. The first thing to add is your router — a simple ping monitor to 192.168.1.1 (or whatever your gateway is). Then add your ISP's DNS resolver, and then a public site like 1.1.1.1 via HTTP. That trio tells you something useful: if the router pings but the internet doesn't, it's your ISP. If neither pings, it's your LAN.

Add monitors for anything else that matters: your NAS on port 5000, Home Assistant on 8123, Pi-hole on port 80, your CCTV NVR, the printer. Uptime Kuma's dashboard gives you a clean uptime percentage and response-time graph for each one, which is genuinely useful when you're arguing with your ISP about intermittent dropouts.

Setting up notifications the sensible way

Uptime Kuma has built-in notification integrations for Telegram, Discord, Slack, ntfy, Gotify, Pushover, email and about 90 others. For a home lab, Telegram is the sweet spot — free, instant, and it works on your phone without any port forwarding.

Create a bot via @BotFather in Telegram, grab the token, send your bot a message, then visit https://api.telegram.org/bot<TOKEN>/getUpdates to find your chat ID. In Uptime Kuma, go to Settings → Notifications → Setup Notification → Telegram, paste both values, and hit Test. You should get a message within seconds.

One tip: set a heartbeat retry of 2–3 on each monitor. A single dropped ping on Wi-Fi is noise, not an outage. Uptime Kuma will only fire the alert after consecutive failures, which cuts the false positives dramatically.

Where n8n earns its keep

Uptime Kuma's native notifications are fine for a simple "it's down" ping. But the moment you want anything smarter — a daily summary, escalation if something stays down for more than ten minutes, a log written to a Google Sheet, or a message that only fires during waking hours — you want n8n in the middle.

The pattern is this: point Uptime Kuma's Webhook notification at an n8n webhook URL instead of (or as well as) Telegram directly. n8n receives the JSON payload, and you build whatever logic you like. A few workflows worth building:

• Escalation: if a monitor is still down after 10 minutes, send a second alert to a different channel — or ring your phone via a Twilio call.
• Quiet hours: an IF node that checks the current hour and suppresses non-critical alerts between 23:00 and 07:00, so a NAS reboot doesn't wake you.
• Daily digest: a Cron node at 08:00 that pulls Uptime Kuma's status page API (/api/status-page/<slug>) and posts a summary of yesterday's uptime to Discord.
• Auto-remediation: if the Pi-hole container is down, SSH in via n8n's SSH node and restart it. Yes, this is as satisfying as it sounds.

Deploy n8n with its own compose service on port 5678, mount a data volume, and set N8N_SECURE_COOKIE=false if you're accessing it over plain HTTP on your LAN. The community edition is free and unrestricted for self-hosted use.

If you'd rather skip the trial-and-error of building these workflows from scratch, the Home Lab & Automation Pack is a set of ready-made AI prompts for scripting, troubleshooting and documenting exactly this kind of setup — including prompt templates for generating n8n workflow JSON and debugging Docker networking. It's from £9 and saves you an evening of Stack Overflow archaeology.

Making it survive a reboot (and a power cut)

A monitor that dies when the thing it monitors dies is useless. Two things to sort:

Restart policy. Both containers need restart: unless-stopped in their compose files — that's already in the example above. Without it, a host reboot leaves you blind.

Power. If your Pi is plugged straight into a wall socket, a brief power cut takes your monitoring offline at exactly the moment you'd want it running. A small UPS is the single best upgrade for a home lab. Even a modest 600VA unit will keep a Pi and router alive for well over an hour, and it'll email you when the mains drops. Have a look at a small UPS for your Pi if you haven't got one — it's the difference between a monitoring stack and a toy.

Finally, back up the Uptime Kuma data directory. It's just SQLite in ./uptime-kuma-data, so a weekly rsync to your NAS is enough. Losing your monitor history the first time you rebuild the stack is a rite of passage you only need to go through once.

What you get for free

Once this is running you'll know within 20 seconds if your broadband drops, if a container crashes, or if a device on your LAN stops responding — and you'll have the historical data to prove it. The whole stack runs comfortably in under 500MB of RAM, costs nothing in subscriptions, and keeps your monitoring data on your own hardware rather than someone else's cloud.

Start with three monitors and one Telegram notification. Get that working, then add n8n and one escalation workflow. Within a week you'll wonder how you ever ran a home network without it — and your ISP's support team is about to have a much harder time fobbing you off.