rttf

rttf · the book · version 1.1.1

Self-Hosting Everything on a Three-Node k3s Cluster

A home lab that runs real apps, restores its own backups and tells you when it breaks. Twelve chapters, each one a thing that runs on three small VMs right now, with the mistakes left in.

  • PDF + EPUB
  • Every number from a real run
  • 3 chapters free
The book's cover next to three cluster nodes, all Ready
12chapters, each a thing that runs
98A5 pages, PDF and EPUB
3chapters free to read first
0made-up numbers

The problem

Most home-lab guides stop at "it runs".

Then the backup quietly stops, the monitoring eats the RAM, and an upgrade drops a few database columns. This book is about everything after "it runs", told from a lab where all of that actually happened.

The usual guide

  • Ends when the pods are Running
  • Backups mentioned, never restored
  • Upgrades left as an exercise
  • Mistakes edited out

This book

  • Starts where the guides stop
  • A real point-in-time restore, step by step
  • The majors night, with its way back
  • Mistakes left in, with what changed

What you get

Twelve chapters, one lab

Each chapter ends with a checklist you can run against your own setup.

Why Kubernetes at home

Four servers, three deploy rituals, one cluster, and when Docker Compose is the better answer.

Hardware and cluster

Why a NAS crashed the cluster, one VM template for everything, Flux as the only way in, and moving it all in 40 minutes.

Networking

Names instead of IPs, one tunnel out, zero open ports, and the IPv6 setting that made Chrome forget every name.

The data layer

Postgres, MariaDB and S3 in one shared namespace, and the plugin that broke pages without a single error.

Monitoring and backups

Prometheus on a diet (204 rules to 46), the backup that never ran, and restore tests that found bugs in themselves.

Upgrades, apps, losing a node, habits

The update checker that couldn't count past 1,000, the 200 that lied, any one of three nodes pulled while the cluster carries on, and a weekly review that fits in fifteen minutes.

Try before you buy

Three chapters are free to read

If you like these, the rest of the book reads the same.

Is it for you?

Written for small labs, on purpose

Yes, if you run (or want to run) k3s on one to a few small machines, and want it boring in the good way.

No, if you want a Kubernetes certification guide or HA across several physical machines: this is one admin, three VMs on one box, and honest trade-offs.

Questions

Before you buy

What format is it?

PDF (A5) and EPUB, about 19,800 words. Both download right after purchase.

Do I need the recipe packs?

No. The book is complete on its own. The recipes are the same setups as tested Flux manifests, if you'd rather start from working code.

Is it about my exact hardware?

No. The lab runs on one Proxmox host and a NAS, but every chapter explains the why, so it carries over to mini PCs, old laptops or a rented server.

Is it finished?

Yes, all twelve chapters. It is versioned (1.1 added the chapter on losing a node), and corrections and new chapters go into the same download.

Everything after "it runs".

Twelve chapters, three of them free to read first. PDF and EPUB.

Get the book €19