Table of Contents
| episode | airdate | recorded | status | short_title | full_title | description | summary | guest | guest_title | guest_links | headshot | youtube_url | youtube_id | fireside_player | slug | tags | topics_to_avoid | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 028 | 2026-09-01 | prep | What Is a Homelab? | What a Homelab Actually Is, with Linux Prepper's James Sundquist | The IT Guy Show 028 | James Sundquist of the Linux Prepper podcast joins me to compare labs: what homelab means to each of us, the services we run, and where the line between a homelab and production really sits. | Linux Prepper's James Sundquist and I compare labs: what homelab means to each of us, what we actually run, and how close it gets to production. | James Sundquist | Host of the Linux Prepper podcast |
|
host.png | itguyshow-ep28-what-is-a-homelab |
|
Abstract
Ask ten people what a homelab is and you will get ten answers, and most of them are really just a description of whatever that person happens to own. I have been running a homelab series on this show all year, mostly working through my own migration out of colocation and into an infrastructure as code workflow. James Sundquist has been running Linux Prepper, a show built around FOSS, self-hosting, and being able to keep your own stuff running when the internet is having a bad day.
We landed on the same hobby from different directions. I lean enterprise: Proxmox, containers, bootc, CI pipelines, infrastructure I can rebuild from a git push. James leans lean and resilient: minimally viable hardware, services running under systemd without a container in sight, and local networking that does not care whether your ISP is up. Put those two labs side by side and there is a lot to learn from each other.
So we compare notes. What each of us means by the word homelab, what we actually run and why, who Docker containers are really for, and the question that matters most once other people in your house depend on your gear: when does a homelab quietly become production, and how would you even know if it broke?
Hook
James Sundquist hosts Linux Prepper, a show about FOSS, self-hosting, and being able to keep your own stuff running when the internet is having a bad day. We built our homelabs for completely different reasons. Today we are putting them side by side to figure out what the word actually means, and where the hobby ends and production begins.
Agenda
Introductions
- Who are you?
- What do you do?
- What do you do for fun?
- Tell people about Linux Prepper and where the "prepper" angle comes from
1. Define Your Terms
- What does "homelab" actually mean to each of us?
- Is it the hardware, the services, or the practice of learning by breaking things?
- What is the minimally viable homelab? Can it be one old laptop?
- The fundamentals: what does someone actually need to understand before they buy anything?
- Where does "self-hosting" end and "homelab" begin, or are they the same thing?
2. Show Me Your Lab
- My side: Proxmox, 16 VMs, Forgejo with runners, Fedora bootc, and a slow crawl toward infrastructure as code
- James's side: the case for running services with
systemctl --userand no containers at all - Database-free services, and when SQLite is the right answer instead of Postgres
- Getting started with Postgres and SQLite without over-engineering day one
- What each of us would rip out and never run again
3. Who Are Containers Actually For?
- James's question, and it is a good one
- Where containers earn their keep: reproducibility, rollback, and moving hosts without a weekend of downtime
- Where a plain systemd unit and a config file is the simpler right answer
- Does the answer change when it is your lab versus someone else's paycheck?
4. When Does a Homelab Become Production?
- The moment your family notices it is down
- How do you know the services you are running are actually working? Monitoring, alerting, or vibes
- Backups, and the difference between having them and having tested them
- What we each borrow from our day jobs, and what we deliberately leave at work
5. Resilience and Independence
- Current 32-bit support and what still runs on hardware nobody else wants
- Local and offline networking stacks: what keeps working when the internet does not
- Migrating out of shared hosting
- Moving from Tailscale to Headscale, and why you would bother
Wrap Up
- Where to find James and Linux Prepper
- Final thoughts and key takeaways
Closing
- Like / Subscribe
- Next:
- Sign-off: on behalf of my guest, James Sundquist, and myself, Eric the IT Guy Hendricks
Keywords
homelab, self-hosting, Linux Prepper, James Sundquist, minimally viable hardware, 32-bit support, offline networking, local first, Docker containers, Podman, systemd user services, systemctl --user, SQLite, Postgres, database free services, Tailscale, Headscale, shared hosting migration, Proxmox, bootc, infrastructure as code, monitoring, homelab versus production
Links
- Linktree: https://linktr.ee/itguyeric
- Linux Prepper: https://podcast.james.network/@linuxprepper
- James's network: https://james.network/
Blog Post
Draft. Rewrite from the transcript after recording.
Ask ten people what a homelab is and you will get ten answers, and most of them are really just an inventory of whatever that person already owns. So I sat down with someone who built his lab for completely different reasons than I built mine.
{{< episode >}}
A few things worth carrying away from this one:
- We define it differently, and both definitions hold up. I think of a homelab as infrastructure practice. James comes at it from resilience and independence. Same hobby, two good reasons.
- Containers are not the automatic answer. James makes a real case for running services under systemd without containers at all, and it sharpened how I think about my own setup.
- Production is a feeling, not a tier. The moment someone else in your house notices the thing is down, you have a service, not a project. We got into monitoring, backups, and how you would even know.
Social Posts
Draft. Rewrite after recording.
Ask ten people what a homelab is and you will get ten answers. Most of them are just an inventory of whatever that person already owns.
So for episode 028 I brought on James Sundquist from the Linux Prepper podcast, because he built his lab for completely different reasons than I built mine. I come at this from the enterprise side: Proxmox, containers, bootc, everything rebuildable from a git push. James comes at it from resilience: minimally viable hardware, services running under systemd with no containers at all, and networking that keeps working when your ISP does not.
A few things you'll pick up:
- What "homelab" actually means, and whether it is the hardware or the practice
- Where containers earn their keep, and where a plain systemd unit is the better answer
- When a homelab quietly becomes production, and how you would know if it broke
- 32-bit support, offline networking, and moving from Tailscale to Headscale
[PASTE YOUTUBE URL]
#Homelab #SelfHosting #Linux #OpenSource #DevOps
Twitter/X
Draft. Rewrite after recording.
We built our homelabs for completely different reasons. For 028 I compared notes with James Sundquist of Linux Prepper: containers and plain systemd, minimally viable hardware, and the moment a homelab quietly becomes production. [PASTE YOUTUBE URL] #Homelab #SelfHosting #Linux
Bluesky / Mastodon
Draft. Rewrite after recording.
Ask ten people what a homelab is and you get ten answers. For 028 I sat down with James Sundquist from Linux Prepper to compare labs: containers and plain systemd units, minimally viable hardware, offline-first networking, and when a homelab stops being a hobby and starts being production.
[PASTE YOUTUBE URL]
#Homelab #SelfHosting #Linux #OpenSource
YouTube Ad
- Final URL: [PASTE YOUTUBE URL]
- Display path 1: IT-Guy-Show
- Display path 2: Homelab
- Short headline: The IT Guy Show
- Long headline: Two Homelabs, Built For Completely Different Reasons
- Description: Linux, homelabs, and open source for the people who actually run the infrastructure.
- CTA: Watch Now