keepcalculator

06 Numbers and Powers

Homelab Projects to Try This Weekend in January

Card No. 06·83591 · Updated · Standard formula, shown below

If you have a spare server, Raspberry Pi, or old PC, January is an ideal month to start a homelab project because the cold weather keeps you indoors and many online services offer free tiers that pair well with self-hosted tools. This article lists concrete projects you can complete in a weekend, from network-wide ad blocking to a personal cloud storage system, with setup steps and common pitfalls explained.

Why January Is a Good Month for Homelab Work

January brings shorter days and colder temperatures in the Northern Hemisphere, which naturally shifts attention to indoor hobbies. For homelab enthusiasts, this is also a time when many cloud providers run promotions, but you do not need paid services to build something useful. The quiet holiday period means fewer distractions, and if you received new hardware as a gift, the weekend is perfect for putting it to work.

Another reason January works well is that many self-hosted applications have stable releases after the December holiday rush. Developers often push major updates in early January, so you can start with a fresh version rather than debugging an old one. Finally, the new year is a natural moment to set goals—like reducing your reliance on big tech services or learning Docker—and a homelab project gives you a tangible outcome.

Project 1: Network-Wide Ad Blocking with Pi-hole

Pi-hole is a lightweight DNS sinkhole that blocks ads and trackers for every device on your home network. You install it on a Raspberry Pi or any always-on Linux machine, point your router’s DNS settings to that machine, and within an hour you will see fewer ads on phones, laptops, and smart TVs.

To set it up, you need a machine with a static IP address and a router that lets you change the DNS server. The installation is a single command: curl -sSL https://install.pi-hole.net | bash. After that, you access the web interface at http://<your-pi-ip>/admin and set a password. The default blocklist covers most ads, but you can add more lists from sites like Firebog or OISD.

Common issues include forgetting to set a static IP (the Pi gets a new address after reboot) and not disabling the router’s built-in DNS feature. Also, some devices like Chromecasts use hardcoded DNS servers, so they will bypass Pi-hole—this is not a bug but a limitation you should know about.

Project 2: Personal Cloud Storage with Nextcloud

Nextcloud gives you a self-hosted alternative to Google Drive or Dropbox, with file sync, calendar, contacts, and even video calls. You can install it on a Linux server using the official Snap package or via Docker. For a weekend project, the Snap install is the fastest: sudo snap install nextcloud. After that, you run sudo nextcloud.manual-install <admin-user> <password> and then set your trusted domains in the config file.

To access it from outside your home, you need to set up port forwarding on your router and ideally use a dynamic DNS service like DuckDNS if your IP changes. However, for a first weekend, you can keep it local and just sync files between your laptop and phone while on Wi-Fi. Nextcloud also has mobile apps for Android and iOS that work well.

A common mistake is not configuring the data directory correctly, which can fill up your system drive. Make sure you point Nextcloud to a large storage disk or a dedicated drive. Also, enable HTTPS using Let’s Encrypt if you expose it to the internet—the Snap package has a built-in command for that.

Project 3: Home Automation with Home Assistant

Home Assistant is an open-source platform that connects smart home devices from different brands into one interface. You can run it on a Raspberry Pi with the dedicated Home Assistant Operating System, or in Docker on an existing server. The installation is straightforward: download the image, flash it to an SD card, and boot. Then you access the web UI at http://<ip>:8123 and follow the onboarding.

Once running, you can add devices like smart plugs, lights, and thermostats. The integrations are mostly plug-and-play, but some devices require a Zigbee or Z-Wave USB stick. For a weekend, start with a few Wi-Fi devices that use the local API (e.g., Tasmota or ESPHome) to avoid cloud dependencies.

The biggest challenge is understanding the YAML configuration, but the new interface includes a visual editor for most automations. A good first automation is turning on a lamp at sunset or sending a notification when your phone connects to the home Wi-Fi. Home Assistant also has a companion app that lets you use your phone as a presence sensor.

Project 4: Media Streaming with Jellyfin

Jellyfin is a free, open-source media server that organizes your movies, TV shows, music, and photos, and streams them to any device. Unlike Plex, it has no paid subscription and no mandatory online account. You can install it on Linux, Windows, or in Docker. For a quick setup, use the official repository for your distribution—for example, on Debian/Ubuntu: sudo apt install jellyfin.

After installation, you add your media folders in the dashboard, and Jellyfin scans them and downloads metadata like posters and descriptions. You can then watch on your TV via the Jellyfin app, on your phone, or in a browser. The server transcodes video on the fly if the client does not support the codec, but for most modern devices, direct play works without transcoding.

A common issue is file naming—Jellyfin works best when your files are in folders like Movies/Movie Name (Year)/Movie Name (Year).mkv. If the metadata is wrong, check the naming guide on the Jellyfin website. Also, if you want remote access, you can set up a reverse proxy like Caddy or Nginx, but that is a more advanced weekend project.

Project 5: Monitoring Your Network with Grafana and Prometheus

If you want to see what your homelab is doing, set up a monitoring stack with Prometheus for data collection and Grafana for dashboards. This is more advanced but still doable in a weekend if you use Docker. Create a docker-compose.yml file with Prometheus, Grafana, and a node exporter that runs on each host you want to monitor.

A minimal example includes three services: prometheus, grafana, and node-exporter. The node exporter exposes system metrics like CPU, memory, and disk usage on port 9100. Prometheus scrapes that endpoint, and Grafana queries Prometheus to display graphs. You can also add exporters for Docker containers, network devices, or even weather sensors.

The learning curve is steep because you need to understand PromQL (Prometheus query language) and how to configure data sources in Grafana. However, there are many pre-built dashboards you can import from the Grafana website. A good first dashboard is “Node Exporter Full” which shows CPU, RAM, disk, and network traffic. This project teaches you about time-series databases and alerting, which are valuable skills for any IT career.

How to Choose the Right Project for You

Consider your experience level and what you actually want to achieve. If you are new to Linux, start with Pi-hole or Nextcloud because they have simple installers and lots of tutorials. If you already know Docker, try Jellyfin or the monitoring stack. Also, think about the hardware you have: a Raspberry Pi 4 can run Pi-hole and Home Assistant, but for Jellyfin transcoding you might need a more powerful x86 machine.

Another factor is the time you can dedicate. Pi-hole takes about 30 minutes, Nextcloud about an hour, Home Assistant can take a few hours if you add devices, and the monitoring stack might take a full day if you are new to it. Be realistic about your weekend—it is better to finish one project than to start three and leave them half-configured.

Finally, consider what problem you want to solve. If you are annoyed by ads, Pi-hole. If you want to replace Dropbox, Nextcloud. If you want to automate your lights, Home Assistant. If you want a Netflix-like experience for your own media, Jellyfin. If you want to learn about observability, Grafana. There is no wrong choice, but picking one that matches your motivation will keep you engaged.

Common Problems and How to Avoid Them

One frequent issue is not setting a static IP for your server. If the IP changes, all your services become unreachable. On most routers, you can reserve an IP address based on the device’s MAC address. Do this before you start any project.

Another problem is firewall rules. Many Linux distributions enable a firewall by default (like UFW on Ubuntu), which blocks incoming connections. If you cannot access your service from another device, check the firewall and allow the specific port. For example, sudo ufw allow 80/tcp for a web interface.

Backups are often overlooked. Before you start, make sure you have a way to back up the configuration of your homelab. For Docker, you can back up the docker-compose.yml and the volumes. For Nextcloud, use the built-in backup app or just copy the data directory. A simple script that tars your config files and copies them to another disk is enough for a weekend project.

Finally, do not expose your services to the internet without security. If you want remote access, use a VPN like WireGuard or Tailscale instead of opening ports. If you must open a port, use HTTPS and strong passwords. The majority of homelab attacks come from bots scanning for open ports, so a VPN is the safest option.

Frequently Asked Questions

Can I run these projects on an old laptop? Yes. An old laptop with 4GB of RAM can run Pi-hole, Nextcloud, and Home Assistant simultaneously. For Jellyfin, it might struggle with 4K transcoding, but direct play will work. Use the laptop’s battery as a UPS by enabling the “always on” power setting in the BIOS.

Do I need a static IP from my ISP? No. For local-only projects, you do not need a public IP. For remote access, you can use a dynamic DNS service like DuckDNS or a VPN that does not require inbound connections, such as Tailscale.

What is the difference between Docker and a virtual machine? Docker shares the host OS kernel and runs isolated containers, while a VM runs a full OS with its own kernel. Docker is lighter and faster to start, but VMs offer better isolation. For homelab, Docker is usually easier for running single services, while VMs are useful for testing different operating systems.

How much power will these projects consume? A Raspberry Pi 4 draws about 5 watts, while an old laptop might draw 20-30 watts. Running 24/7, a Pi costs a few dollars a year, whereas a laptop could add $20-30 to your annual electricity bill. You can reduce power by using SSDs instead of spinning disks and by disabling unused peripherals.

Is it safe to run a homelab on the same network as my main computer? Yes, but you should segment your network if you are concerned about security. Many routers support guest networks or VLANs. For a weekend project, simply keeping your homelab on the same network is fine as long as you do not expose services to the internet.

What if I break something? That is part of learning. Most homelab services can be reinstalled in minutes. Before making changes, take a snapshot of your OS (if using a VM) or back up the configuration files. For Docker, you can recreate the container from the compose file, so the worst case is losing some data, which is why backups are important.

Related

Method: the standard formula for this calculation, rounded as each result states. See how the calculators work or report an error.

More cards in this drawer

The whole Numbers and Powers drawer →