Maybe my use case is abnormal, but I allocate the majority of my resources to a primary VM where I run everything, including containers, etc. but by running Proxmox now I can backup my entire server and even transfer it across the network. If I ever have some software to try out, I can do it in a new VM rather than on my main host. I can also ‘reboot’ my ‘server’ without actually rebooting the real computer, which meant less fan noises and interruption back when I used an actual rack mounted server at home.

Primarily, docker isn't isolation. Where isolation is important, VMs are just better.

Outside of that:

Docker & k8s are great for sharing resources, VMs allow you to explicitly not share resources.

VMs can be simpler to backup, restore, migrate.

Some software only runs in VMs.

Passing through displays, USB devices, PCI devices, network interfaces etc. often works better with a VM than with Docker.

For my setup, I have a handful of VMs and dozens of containers. I have a proxmox cluster with the VMs, and some of the VMs are Talos nodes, which is my K8s cluster, which has my containers. Separately I have a zimaboard with the pfsense & reverse proxy for my cluster, and another machine with pfsense as my home router.

My primary networking is done on dedicated boxes for isolation (not performance).

My VMs run: plex, home assistant, my backup orchestrator, and a few windows test hosts. This is because:

- The windows test hosts don't containerise well; I'd rather containerise them. - plex has a dedicated network port and storage device, which is simpler to set up this way. - Home Assistant uses a specific USB port & device, which is simpler to set up this way. - I don't want plex, home assistant, or the backup orchestrator to be affected by issues relating to my other services / k8s. These are the services where small transient or temporary issues would impact the whole household.

Also note, I don't use the proxmox container support (I use talos) for two reasons. 1 - I prefer k8s to manage services. 2 - the isolation boundary is better.

Proxmox is essentially a clustered hypervisor (a KVM wrapper, really). It has some overlap with K8s (via containers), but is simpler for what I do and has some very nice features, namely around backups, redundancy/HA, hardware passthrough, and the fact that it has a usable GUI.

I also use K8s at work, so this is a nice contrast to use something else for my home lab experiments. And tbh, I often find that if I want something done (or something breaks), muscle-memory-Linux-things come back to me a lot quicker than some obscure K8s incantation, but I suppose that's just my personal bias.

Several of my VMs (which are very different than containers, obviously - even though I believe VMs on K8s _can_ be a thing...) run (multiple) docker containers.

personally : proxmox /VM is great if You'd like to separate physical HW. In my case - virtualized TrueNAS means I can give it a whole SATA controller and keep this as an isolated storage machine.

Whatever uses that storage usually runs in a Docker inside an LXC container.

If I need something more isolated (think public facing cloudflare) - that's a separate docker in another network routed through another OPNSense VM.

Desktop - VM where I passed down a whole GPU and a USB hub.

Best part - it all runs on a fairly low power HW (<20W idle NAS plus whatever the harddrives take - generally ~5W / HDD).

it's so you can have a machine to run docker on, basically.

especially useful if you want multiple of those, and also helpful if you don't want one of them anymore.

Not all things run as containers.

If you want to go another, related rabbit hole, check out the DataHoarder subreddit. But don't blame me, if you’re buying terabytes of storage over the next few months :)

I have Proxmox running on top of a clean Debian install on my NUC, I wanted to allow Plex to use the hardware decoding and it got a bit funny trying to do that with Plex running in a VM, so it runs on the host and I use VMs for other stuff

> My most recent learning - DDR4 ECC UDIMMs are comically expensive. To the point where I considered just replacing the entire platform with something RDIMM rather than swapping to ECC sticks.

DDR4 anything is becoming very expensive right now because manufacturers have been switching over to DDR5.

>AGH is worth considering because it has built in DoH

Technitium has all the bells and whistles along with being cross platform.

https://technitium.com/dns/

It’s another league, but I don’t get the point of mixing enterprise rack-mounts with Raspberry Pis.

I always need lower power consumption. I'm in the UK and my power costs are $0.40 per kWh. Even running a raspberry pi 5 24/7 would cost me $25 per year

I just commented on this above, but I actually got for the Pi for free and it's a very capable device. I wouldn't buy one for this use case (nor do I really recommend it, but it _does_ work).

It's a neat chip but I couldn't bring myself to spend in excess of the price of the UPS ($439.00 for RMCARD at time of writing). I ended up hooking my NAS via USB to my existing home server via NUT and it's been working well.

Most “homelabs” are built by a developer LARPing as a sysadmin, with a user population of one (themselves) or zero for most of the features.

It’s the SUV that has off-road tires but never leaves the pavement, the beginner guitarist with an arena-ready amp, the occasional cook with a $5k knife. No judgment, everyone should do what they want, but the discussions get very serious even though the stakes are low.

Where’d we get this term? I hear “home lab” and I think of having equipment to accomplish something new, not… running ordinary server software in fairly ordinary ways. Like Tony Stark designing his suits has a “home lab”. People 3D printing Warhammer figures or with a couple little servers running PiHole and Wireguard and such… not so much?

I’ve had one or two machines running serving stuff at home for a couple decades [edit: oh god, closer to 2.5 decades…], including serving public web sites for a while, and at no point would I have thought the term “home lab” was a good label for what I was doing.

some people think it's not "homelabbing" unless you're doing things the way it's done at large scale. i think these people are aiming to enter IT as a career and consider a homelab to be a resume project.

but proxmox and kubernetes are overkill, imo, for most homelab setups. setting them up is a good learning experience but not necessarily an appropriate architecture for maintaining a few mini PCs in a closet long term.

you can ignore the gatekeeping.

Run whatever you like!

I personally enjoy the big machines (I've also always enjoyed meaninglessly large benchmark numbers on gaming hardware) and enterprise features, redundancy etc. (in other words, over-engineering).

I know others really enjoy playing with K8s, which is its own rabbit hole.

My main goal - apart from the truly useful core services - is to learn something new. Sometimes it's applicable to work (I am indeed an SWE larping as a sysadmin, as another commenter called out :-) ), sometimes it's not.

You can run whatever. You don't need specific software