📖 Chapter 01 — The Question
Where this all started. Why the way you ask matters more than what you know.
Learning Objectives
By the end of this chapter, you'll understand:
- Why the framing of your first question determines the quality of every answer that follows
- What TK was actually asking when he asked "what's a pool?"
- The mental shift from "follow instructions" to "understand the system"
Introduction
This chapter is short. It's also the most important one in the volume, because the entire handbook is an answer to the question it documents.
In July 2026, TK sent a message to ChatGPT. It wasn't particularly long, and it didn't use any technical vocabulary. But every sentence in it was the right sentence to write, in the right order. That's what made the response — and the project that grew from it — possible.
The original question
Verbatim, with no edits:
I need a TrueNAS and ZFS crash course. I need to understand the basic terms. Pool, vdev, dataset, users, permissions, etc. From my research, I think I want to use RAIDZ2. I have 6 hard drives of 12TB each. Me and my family will be the only ones using it for now. Maybe one day I'll save some space for friends so they can log in with their own credentials and use my NAS. But for now, I'm a bit lost with the terminology and am not able to decode what kind of setup I need. It's mostly for backing up pictures, videos and documents. The data will be accessed via SMB shares between other Windows and Mac computers on the local network, but the whole family can share the same folders. Or maybe I should create a separate dataset so only I can access for things like important documents. I also want to be able to log in from anywhere in the world and upload files via my phone.
What's actually being asked
Read it once and it sounds like a TrueNAS question. Read it twice and you see something else: TK isn't asking how do I install TrueNAS? He's asking what setup do I need?
That's a fundamentally different question. The first one is a procedure. The second is a design. The first has a single right answer. The second has tradeoffs, alternatives, and consequences.
Three details make the question extraordinary:
- It admits ignorance without embarrassment. "I'm a bit lost with the terminology." This is the foundation of every productive technical conversation. The alternative — pretending to understand, then designing a system on top of a misunderstanding — is how 90% of home lab disasters happen.
- It states a hypothesis. "From my research, I think I want to use RAIDZ2." TK had done the reading. He wasn't asking ChatGPT to do his research; he was asking ChatGPT to verify a hypothesis against someone else's experience. This is exactly how a senior engineer talks to a junior one.
- It names the actual goal. "Pictures, videos and documents. Accessed by Windows and Mac. From anywhere in the world via phone." The goal isn't TrueNAS. The goal is "a place my family can trust with the irreplaceable." The technology is the means, not the end.
Why this matters for the rest of the handbook
Every chapter in this volume is an answer to that question. Every tradeoff we discuss — RAIDZ2 vs RAIDZ3, separate datasets vs nested folders, SMB vs NFS, Tailscale vs port forwarding — comes back to the same three criteria TK named in the first message:
- Will the family actually use this?
- Will TK be able to maintain this in two years?
- Will it survive being reached for from an airport Wi-Fi?
When in doubt, the answer is the one that scores highest on those three.
Whenever you start a new project — a volume, a feature, a service — write the question first. Then design the answer. The question is the spec. If you can't write the question, you don't have a project yet.
Starting with the technology instead of the question. "I want to use Proxmox" is not a project. "I want to run my home lab without paying AWS for development environments" is. The first one makes you a hostage to Proxmox's roadmap. The second one lets you change tools as better ones appear.
Engineering Note
Most lab disasters start with a question shaped by the answer. "How do I set up a ZFS pool?" produces a ZFS pool. "What's the most resilient home storage I can build for under $X with my family's usage pattern?" produces a system the family will actually use for a decade. The question is the design. Get it right, and the rest is just execution.
Summary
TK's first question was the right question. It admitted ignorance, stated a hypothesis, and named a real goal. The rest of the handbook is the answer, refined across six volumes.
Checklist
- ⬜ For your own lab: write your first question, in full, the way TK wrote his. Save it. Refer to it whenever you're making a design decision.
- ⬜ Before you start any chapter in this volume, re-read TK's question. Remind yourself what the project is actually for.
Looking Ahead
Chapter 2 builds the mental model. By the end of it, you'll have a single picture of how TrueNAS works — drives, vdevs, pools, datasets, shares — that you'll reference mentally every time you click anything in the web UI.