Homelab from Zero #5: Uptime Kuma, know when things break

This is part 5 of Homelab from Zero. In part 4 you added Pi-hole and gave the whole household ad-free browsing. Today we make the lab self-aware: Uptime Kuma watches your services and pings you the moment something goes down.

Why you need monitoring before it hurts

Right now, if Pi-hole or Heimdall stops responding you won’t know until someone in the house complains, or until you notice yourself. As the lab grows, that becomes a real problem: a service can fail silently for hours.

Uptime Kuma is a self-hosted monitoring tool that checks your services on a schedule (every 60 seconds by default) and sends you an alert if one stops answering. It’s a single container, it has a clean web dashboard, and it logs a timeline so you can see exactly when each outage started and how long it lasted.

Step 1: Run Uptime Kuma

SSH in and start the container:

docker run -d \
  --name uptime-kuma \
  -p 3001:3001 \
  -v ~/uptime-kuma:/app/data \
  --restart unless-stopped \
  louislam/uptime-kuma:1

Piece	Plain English
`-p 3001:3001`	The dashboard port
`-v ~/uptime-kuma:/app/data`	Keeps all your monitors and history across reboots
`--restart unless-stopped`	Starts automatically when the server boots

Step 2: Create your account

On your normal computer, visit:

http://192.168.1.73:3001

The first visit prompts you to set a username and password, this is a local account, nothing leaves the network. After that you land on the main dashboard: nothing monitored yet.

Step 3: Add your first monitors

Click Add New Monitor. Start with the two services already running.

Heimdall, the dashboard from part 3:

Monitor type: HTTP(s)
Friendly name: Heimdall
URL: http://192.168.1.73:8080
Heartbeat interval: 60 seconds

Pi-hole, the ad blocker from part 4:

Monitor type: HTTP(s)
Friendly name: Pi-hole
URL: http://192.168.1.73:8081/admin
Heartbeat interval: 60 seconds

Uptime Kuma starts checking immediately. Within a minute both should show a green Up badge and begin building a response-time graph. From here you can add anything with an IP or URL: your router’s admin page, a ping monitor for the server itself (192.168.1.73 as a ping target), or any other service you add later.

Step 4: Set up an alert

A monitor that doesn’t notify you is just a dashboard you have to remember to check. Go to Settings → Notifications and add one. The options cover email, Telegram, Discord, Slack, and a dozen others, pick whatever you’ll actually notice.

Email (SMTP) works with most mail providers. Fill in your SMTP host, port, and credentials, send a test notification, and confirm it arrives. From then on, every outage sends a message automatically.

No SMTP handy? Telegram bot notifications are the quickest alternative: create a bot with BotFather, then paste the token and your chat ID into Uptime Kuma’s Telegram notification fields. Takes about two minutes.

The commands you’ll actually use

Command	What it does
`docker logs uptime-kuma`	What is Kuma saying? (start here if the page won’t load)
`docker restart uptime-kuma`	Turn it off and on again
`docker stop uptime-kuma`	Stop monitoring: other services keep running
`docker ps`	Confirm Uptime Kuma (and everything else) is still up

What you have now

A monitoring dashboard watching every service in the lab, with alerts that reach you the moment something falls over, no polling, no waiting for a complaint. You set it up once and it runs in the background.

In part 6 we tackle the thing most homelabs skip until it’s too late: backups. One automated, tested setup using the 3-2-1 rule, so a dead drive doesn’t mean lost data.

← Back to blog

Table of contents