Summary
I’ve built, broken, and rebuilt many labs over the years. But in all my builds, some tools have always made the cut, and one of the more important ones is the monitoring stack. I need to know if a Docker container has crashed, a disk has filled up, or a service has stopped responding the moment it happens, not when I try to use it. This week we take a look at some of the top Docker containers to monitor your home lab. Like the Docker containers I self-hosted on a recent weekend, you can self-host all of these, and they are free to use. Uptime Kuma The easiest way to know when something goes down Uptime Kuma was featured in my previous list of Docker containers worth self-hosting, and it’s no surprise that it has made it into this one too. It remains the simplest way to know if your services are up or down. Uptime Kuma runs as a single Docker container. You point it at a URL, a TCP port, a ping address, or a DNS record, and it checks that target on a schedule. If something stops responding, it sends you an alert through one of over 90 notification services, including Discord, Telegram, Slack, and email. More importantly, it also tells you when things come back up, and explains what failed and why. Beyond the basic up/down checks, it has an HTTP keyword monitor that loads a page and looks for specific text. So even if your app returns a 200 OK status, Uptime Kuma will mark it as down if the expected content is missing. It also tracks SSL certificate expiry dates. On top of all that, it builds a clean, public-style status page that groups your services and shows incident notes, all without a separate tool. Netdata Live system metrics with zero configuration Uptime Kuma tells you that a service went down, but it won’t tell you why. Netdata installs an agent on your machine, and you immediately get live graphs for CPU, RAM, disk I/O, network traffic, and per-container stats, all at per-second resolution and with no configuration needed. The Netdata agent is free and open-source, and you can install it on as many machines as you want. Each agent runs its own web UI on port 19999, so you can check any node directly. If you want to see everything in one place, the optional Netdata Cloud dashboard pulls all your nodes together. The free tier covers 5 nodes, and the Homelab plan gives you unlimited nodes for personal use for about $90 a year. What’s more, it can auto-detect what’s running on the host and adds relevant metrics. Install it on a Proxmox node, and you get Proxmox-specific graphs without touching a config file. The built-in alerts are also practical. On one of my nodes, Netdata flagged a disk starting to degrade before I noticed any slowdown. Uptime Kuma would never have caught that because the services on that machine were still responding. Dozzle A browser-based live view of your Docker logs Knowing that something went down and seeing the system metrics around it gets you most of the way, but sometimes you need the real error message. That’s where Dozzle comes in. It gives you a browser-based view of live Docker container logs, so you don’t have to SSH into a machine and run docker logs -f on each container. Dozzle is easy to set up. A single docker run command with the Docker socket mounted read-only, and Dozzle is live on port 8080. It doesn’t store any logs itself. It reads them straight from Docker in real time, so you don’t have a database to maintain and nothing extra eating up storage. Also, the new log lines appear as they are written. Another handy feature is split-screen view. When a problem involves two containers that depend on each other, like a reverse proxy and the app behind it, being able to watch both log streams side by side makes troubleshooting much faster. You can also search inside a container’s logs, filter by container name, and download logs from the dropdown. For what it does, Dozzle is surprisingly small at around 7 MB compressed. Pulse Proxmox monitoring built by a solo developer If you run Proxmox, you know it doesn’t come with great built-in alerting. Pulse is an excellent alternative to Netdata, but specifically built for Proxmox. It monitors Proxmox VE nodes, Proxmox Backup Server, and now Docker containers from a single dashboard, and it sends alerts to apps like Discord, Slack, Telegram, and email when something breaks. You can install Pulse as a Docker container on any Linux host, or use the official installer script that creates an optimized LXC container directly on your Proxmox node. Once it’s running on port 7655, you turn on auto-discovery, and Pulse finds your Proxmox hosts on the network. Connect with an API token, and you immediately see CPU, memory, disk usage, backup job status, and physical disk health for each node. The alert system offers more than basic on/off notifications. You can set per-VM thresholds, configure quiet hours so you’re not woken up in the middle of the night for a non-critical alert, group related alerts together, and set cooldown periods. Komodo Manage Docker containers across multiple servers from one screen Once your lab grows past a single machine, managing containers across multiple hosts becomes its own problem. Komodo is a self-hosted dashboard that lets you manage Docker containers, compose stacks, and deployments across every server you own from one browser tab. Think of it as a Portainer alternative with Git-based deployments and CI/CD automation built in. Komodo runs as a Core server and a lightweight Periphery agent on each Docker host. Adding a new server is a one-command install using an onboarding key you generate in the web UI. Once connected, you can see all containers on that host, manage stacks, view logs, and trigger deployments. The standout feature is that it version-controls your Docker Compose files through Git, so every change is tracked and deployable with a push. I currently manage containers on my NAS and a couple of mini PCs, and switching between SSH sessions to check on each one was getting tedious. Komodo can restart a container, check logs, or even redeploy a stack from a single screen. It’s a good fit if you’ve already got a few Docker containers running on an old desktop and want to stop jumping between terminal windows. Five tools to keep your Docker containers up and running You don’t need all five on day one. Start with whichever problem is costing you time right now, and add from there. The first monitoring container I set up was Uptime Kuma, and one by one, the others found their place.