Home Lab Chaos and Learning

Home Lab Chaos and Learning | Generated by AI

Home 2026.02

Question: Suppose I have a Mac and 2 Ubuntu laptops at home. What programs or funny projects can I do? Just Kafka and ZooKeeper for learning purposes? What other cool things can I do?

Answer:

You have a nice heterogeneous setup (macOS + Linux) that’s perfect for a small home lab / distributed playground. Kafka + ZooKeeper (or the modern KRaft mode without ZooKeeper) is a solid start for learning streaming / event-driven systems, but there are tons of more fun, silly, or practical projects you can build with these three machines.

Here are some ideas ranked roughly from “quick & funny” to “more serious learning / useful”:

Funny / Silly / “Why not?” Projects

Distributed useless computing cluster
Turn the three machines into a mini “supercomputer” that wastes electricity together: run BOINC (SETI@home style), Folding@home across all three, or Mine some joke cryptocurrency (or just simulate mining with silly scripts). Bonus: write a Python script that pings between machines and shows fake “processing power” stats on the Mac screen.
Over-engineered home meme / notification system
One Ubuntu runs ntfy.sh or Gotify server. Write tiny scripts on all machines (including Mac) that send ridiculous notifications: “Dad’s Mac just opened Reddit again”, “Ubuntu-2 CPU at 0.1% – it’s napping”, or meme quotes every hour via cron. Add sound effects on the Mac using say or afplay.
Multi-room chaotic music / sound server
Run Mopidy or Jellyfin + Snapcast on one Ubuntu (server), control from Mac/iPhone, and sync audio playback across all three laptops (if they have speakers). Play the same song everywhere → instant annoying home surround system. Or make one machine the “DJ troll” that randomly changes volume/playlist.
Fake smart home with zero IoT
Use Home Assistant (on one Ubuntu) but control fake “devices”: scripts that pretend to be lights (change terminal color), coffee maker (print ASCII coffee), or robot vacuum (move mouse cursor randomly on Mac).

Cool & Actually Useful Learning Projects

Mini Kubernetes / container orchestration playground
Install k3s (lightweight Kubernetes) on all three machines → one as master (maybe Mac with multipass VMs or kind), two Ubuntu as workers. Deploy silly things like 404-game, Minecraft server pods that auto-scale, or a distributed “fortune” teller service.
Self-hosted personal cloud stack
- Ubuntu-1: Nextcloud or Immich (photos) + PostgreSQL/MariaDB
- Ubuntu-2: MinIO (S3-compatible object storage) or TrueNAS SCALE in a VM
- Mac: Use it as client + Time Machine backup target to one Ubuntu via SMB/NFS
  Sync files/photos across all devices privately.
Distributed task queue fun
Beyond Kafka/ZooKeeper:
- Set up Celery + Redis/RabbitMQ across machines
- Or use Dask.distributed: Mac as scheduler, two Ubuntu as workers → run embarrassingly parallel Python tasks (image processing, ML training chunks, password cracking demo, rendering Mandelbrot sets across nodes).
Monitoring & observability overkill
Prometheus + Grafana on one Ubuntu, node-exporter on all three → beautiful dashboard showing CPU, funny metrics like “number of open Safari tabs on Mac”, temperature if sensors work, uptime bragging rights.
GitLab / Gitea + CI/CD playground
Self-host Gitea or full GitLab on one Ubuntu. Set up GitLab Runner on all three machines (including Mac runner). Practice pipelines: build Docker images, deploy joke websites, or auto-backup your dotfiles.
Media & backup fortress
Jellyfin / Plex on Ubuntu-1 (media server), sync media folders with Syncthing or rsync from Mac. Use restic/BorgBackup on all machines → encrypted backups to each other (3-2-1 rule on a budget).
Learning modern data/streaming beyond basic Kafka
- Redpanda (Kafka-compatible, easier setup)
- Apache Flink or Spark cluster (tiny scale) for processing fake streaming data (e.g., Twitter-like feed simulator)
- Temporal.io or n8n for workflow orchestration across machines
Network security playground
One Ubuntu as pfSense/OPNsense router/firewall (needs second NIC), Mac and other Ubuntu behind it → experiment with VLANs, WireGuard VPN between machines, intrusion detection (Suricata), or Pi-hole for ad-blocking.

Start small: pick one project, use Docker/Podman everywhere for consistency (Mac supports both), Ansible from Mac to manage the two Ubuntu boxes. Most fun comes from over-engineering something stupid — that’s the spirit of home labs!

References:

Back Donate