Welcome to My Homelab: From Tinkering to Clustering

There’s something uniquely satisfying about a homelab. For some, it’s a weekend experiment gone wild. For others, it’s a sandbox for learning, tinkering, and sometimes—just sometimes—pulling your hair out over a misbehaving VM at 1 AM.

For me, it’s a bit of all that—and then some.

This post marks the start of documenting my homelab journey. I’ve learned (often the hard way) that no setup is ever truly finished, and that’s part of the appeal. So whether you’re here because you’re knee-deep in ZFS tuning or just curious about building your own lab, welcome. Let’s kick things off with how I got here and where I’m headed.

Where It All Started: The Rabbit Hole Beckons

My homelab journey didn’t start with a rack full of servers or blinking 10Gb switches. Like many of us, it began with a repurposed desktop, a spare SSD, and the thought: “I wonder if I can self-host that.”

Spoiler alert: I could. And then I wanted to host everything.

Docker containers were my gateway drug. Once I had Plex running and a few ARR stack apps automating my media life, I realized I needed something more robust than a single box. I’d caught the bug—and I was all in.

The Beast Grows: My Current Setup

Today, my homelab has evolved into something far more powerful (and way more complicated). At the heart of it is a 3-node Proxmox VE cluster, built from a mix of power-efficient desktops and small form factor machines. These nodes run a mix of VMs and LXC containers, powering everything from media servers to Home Assistant and InfluxDB-backed dashboards.

Each node is equipped with either 1Gbps or 2.5Gbps NICs, which gives me decent internal bandwidth—though I’ll admit I’m already eyeing 10GbE like a hungry sysadmin at a clearance sale.

Networking is handled by a Unifi-based setup with a sprinkling of QNAP and TP-Link switches. VLANs segment different workloads, and I’m slowly but surely turning this thing into a mini datacenter (minus the cooling bill and badge access).

The Good, the Bad, and the Unexpected

Let’s talk honesty for a minute. While things are mostly working well, not everything is sunshine and Docker.

What’s Going Smoothly:

• Proxmox clustering: High availability works surprisingly well for my needs. I can reboot a node and most services stay up with little interruption.
• Home Assistant: I’m using it to collect environmental and power data, and it’s been a game-changer for understanding my household energy patterns.
• Docker orchestration: Running my ARR stack, Plex, and even Nextcloud in containers has been mostly seamless.

What’s… not:

• VM Migration Woes: I recently ran into issues trying to live-migrate VMs between nodes. The culprit? Local-only storage. Turns out, not having shared storage kind of defeats the point of clustering. Who knew?
• Storage Bottlenecks: My current storage is a mix of local ZFS and NFS shares from my TrueNAS box. It works… until it doesn’t. I’m hitting I/O walls with certain workloads, especially during backups or heavy media scans.
• Trying to Be Too Efficient: I’m always chasing lower power draw without sacrificing performance. It’s a delicate balance between “green” and “grumpy.”

Looking Ahead: Big Plans, Bigger Lessons

I’m currently knee-deep in planning a fast shared storage solution—possibly Ceph, maybe ZFS over iSCSI, or even NVMe-over-TCP if I feel like punishing myself. The goal is simple: enable seamless VM migration and better fault tolerance across nodes without torching my power budget.

I’m also exploring:

• Grafana dashboards for real-time power and temp monitoring
• High availability for key services (think: reverse proxies, DNS, media)
• A proper Proxmox Backup Server build, likely using an NFS share from TrueNAS

And let’s not forget the rabbit holes I haven’t even discovered yet. AI-powered automations via Home Assistant? A 10Gb fiber backbone? Who knows what’s next?

Let’s Build This Together

This blog isn’t meant to be a how-to manual (though I’ll be sharing guides and configs). It’s more like a build diary—complete with wins, fails, and why-the-heck-is-that-container-using-12GB-of-RAM moments.

If you’re running your own homelab, thinking about starting one, or just like seeing how deep the nerd hole goes, stick around. I’ll be documenting everything—from upgrading switches to accidentally nuking a VM (yep, done that too).

What’s your homelab story? I’d love to hear it. Drop a comment, share a photo, or even just tell me which part of my setup you’d improve. This is as much your space as mine.

More posts coming soon—with fewer mishaps. Maybe.

Until next time, keep your ZFS pools scrubbed and your containers tidy.