Nextstep is Homepage with all your services on it.

I mean, I kinda do- for public services on my website. but I don’t why I would do a dedicated one- when almost all need services authentication.

but I do use homarr for as an internal “homepage”/dashboard. Great idea though.

Nice.

I spent an entire weekend doing this and then just sat there staring at it like the Wolverine looking at a picture meme.

You have each “docker” container in there own lxc running docker? That’s a silly way to do it if I’m understanding correctly. You’re adding overhead to containers for little to no benefit. You could put them all in the same lxc or Proxmox supports OCI containers natively now.

Proxmox supports OCI containers natively now

As of 9.1, and kind of. To be clear what its doing is not a drop-in replacement for docker, but using that image as a basis for an lxc - so you’re not just pulling a new image to update, more like re-deploying, so keep your stored data elsewhere, outside of the lxc. So what was handled by pulling an image or a change to a compose is done manually in proxmox now, but it does make things like backups/snapshots much more consistent and proxmox-native.

Yeah, but if you were doing lxc + docker per container you might as well use this oci as lxc method. I think both methods are silly, but that’s just my opinion.

If its a single image, I’d agree. If its a compose with multiple images though, better off combining into a single lxc or VM.

I think the misunderstanding is that I’m running every single service as Docker inside its own LXC. That’s not actually what I’m doing.

I split things by service boundary. For example, Pi-hole + Unbound has its own LXC, Lemmy has its own LXC, Immich has its own LXC, etc. Most of those run natively in LXC without Docker at all.

Docker is only there where the application actually expects/benefits from it. I’m not adding a Docker layer everywhere just for the sake of it.

The reason I prefer this over one giant LXC with everything docker- inside. is the management side.

Yes, there is some overhead. But LXC containers are much lighter than VMs. They share the host kernel and use Linux namespaces/cgroups for isolation, so you’re not paying the cost of a full virtualized OS. The extra memory/CPU overhead is minimal compared to the management benefits I get from separating services.

The Docker overhead only exists on the services that actually need Docker. A lot of my services run directly in their own LXC without Docker at all.

Could I put everything into one LXC and run Docker Compose? Sure. It would probably work fine. But for my homelab, ease of maintenance and recovery is just as important as raw resource efficiency.

How does that make it easier for maintenance?

My last gen Homeserver was like this too. It worked okay, but network configuration, especially port forwarding and ipv6 sucked.

My current generation is just debian stable again, it just works.

Ahhh…yes. The PronMaster 6000

Congrats

Nice specs, nice setup. You might want to add another rainbow sticker or two, just in case :p

Wait, we’re putting stickers in the case now?

where are they in the case??? hahaha (edit: i am still getting used to lemmy comments, so i thought you said I had stickers in my case- I didn’t see you replied to someone haha)

I believe by replying to my comment, per Lemmy rules, totally not mine. You’ve contractually agreed to putting at least 5, but no more than 100 rainbow stickers on the inside of your case.

Yes, you may call them “RGB”.

Yes, they make your NAS go faster.

Yes, they improve airflow, but only when oriented correctly.

No, they don’t make you less straight. They’re “closeted” officially.

Welcome to Lemmy :), you’ll fit right in lmao

Thank you!!

Low emission leds!

I see ’nett hier’ sticker, i vote up.

Dope little setup. Love the BIOS sticker.

@organicpu love the stickers ! All of them ! :)
Have fun with your beauty beast.

Thanks!!!

This is pretty cool, and I’m guessing the hardware doesn’t even break a sweat.

My entire setup runs on two AMD mini-PCs, one running a Ryzen 5700u with Proxmox for VM hosting, the other is a 5500u running bare metal Ubuntu.

Proxmox server hosts:

• Asset UPnP
• Jellyfin
• Audiobookshelf
• (formerly) Plex (I ditched it)
• Lyrion
• LRCLIB offline lyrics server (via LRCLIB DB dump)
• Home Assistant VM
• Navidrome instance

Probably a few other things I forgot about…

The bare metal Ubuntu server currently just runs Audiomuse-AI for music analysis, but when not scanning it’s generally idle. Only set this up because early versions of Audiomuse-AI didn’t play nice for me in a VM even when r running with CPU in host mode.

When the hardware gets cheaper I wouldn’t mind a full size server again with a dedicated GPU (I decommissioned my old one due to defective RAM), but for now the mini-PCs have been great. They’re quiet and absolutely sip power, so efficient and quiet.

Yea! It really doesn’t break a sweat. Most of the time it sits around 5-20W (legit jumps between 5W and 20W) while hosting everything, with around 5-15% CPU utilization. Though I do share it with other people, so it may not be as “idle” as I think it is. With all services stopped and nothing running, it obviously uses way less power.

I also have a Windows VM for someone else, which basically acts as a terminal server for development. They run Linux and need Windows for developing something, but they said “I don’t want Windows anywhere on my physical system”, so I gave them a VM to use instead.

I also like using my server as a playground to test things. It saves me having to install VM software on my PCs and keeps experiments separate.

Originally, I used GPU passthrough to a VM, but I stopped doing that because I wanted multiple services to be able to use the GPU while still keeping everything isolated. Unfortunately, my GPU isn’t nice enough for NVIDIA vGPU support.

So passing it directly into containers was the next best option - although definitely the more tedious route to set up. The efficiency gains are pretty nice though.

It also only pulls around 300W at max load (CPU + GPU), but the load is very variable and heavy usage isn’t that dramatic. Normal “heavy” use is closer to around 120W (CPU + GPU).

In the future, I’d love to move to an ARM-based server for even better efficiency, once the hardware becomes more readily available and practical for server use. The efficiency gains are really interesting, but the ecosystem and availability still need some time to catch up.

Also, your setup sounds really nice. Those mini-PC “servers” are honestly impressive for what they can do nowadays. The power efficiency and silence are hard to beat, and having two separate machines gives you a nice balance of reliability and flexibility. It’s cool seeing how far you can push small hardware with the right services.

Nice setup! Might check out some services I didn’t know about yet.

But I might add one thing, because I am/was in the same boat: Friends Don’t Let Friends Use Ollama

Woah, thanks for the article!

Great article, they always gave of that energy tbh. But I did not bother and also didn’t use it enough that it would matter. Good to have a write up about it. Thanks for sharing.

Commenting because I’m suuuuper interested but can’t read it now, but I never look at saved comments. Thank you!

I’ll summarize.

Shitty VC backed company. Just a wrapper for llamacpp. Just use llamacpp. If you want a front end, I recommend Anything LLM

Wait, what is this SnapOtter thing you mention? Apparently it’s based on some AI tool, but what has me really intrigued is its Whisper capabilities? There was another project I tried to use a few years ago called Willow, in order to have a VTT connection to home assistant. community.home-assistant.io/t/…/664916

The creator tragically died in a car accident soon after though and the project went kaput.

This looks like maybe it could do the same thing???

this is Snapottter snapotter.com

Yeah I was already viewing the website, but I’m interested if it can replace Willow

Why is this labeled ai?

Ollama probably

Ollama/Openwebui, Immich, Snapotter and cause degoog can use ollama.

Hey @organicpu@lemmy.organic-server.org did you use a LLM like Ollama to write your post?

No shade if you did, I just want to make sure we’re categorizing things correctly at our end.

Nah, I haven’t. I guess I’m just very verbose- I just want to make sure I’m clearly understood.

It’s less any verbosity, and more that the sentence and post structures sound very ai. Not that you’d be the only person that just writes like that, but especially the concluding paragraph sounds exactly like something an LLM would output.

I mean, I somewhat understand where you’re coming from. But I’ve always written in this kind of style, and I think the reason the post ended up sounding this way is simply because it’s meant to share information I personally find interesting or useful. There isn’t really much of a way to explain a setup like this without having sections, lists, and explanations.

Honestly it’s a bit difficult to write something like this and have it not sound “AI-generated”.

Doesn’t help, a lot of people seem to associate the dashes (“-”) with AI writing nowadays, but I’ve been using them for years, so unfortunately that probably doesn’t help my case- rip haha.

Regarding the:

“Dive into the Details. If you’re curious about the configs, the philosophy, or just want to see the messy reality behind the setup, head over to organic-server.org. You’ll find some more of my ramblings there. Or, if you have a specific question, I’m always happy to chat here.”

That part was mainly there because I’ve seen a lot of homelab posts where someone shares a cool setup, but then never responds to questions or expands on anything. I wanted to make it clear that I’m happy to discuss the details if anyone is interested.

I mean, you can write however you like. But it does kinda give the vibes of it being written like a click bait YouTube speaks.

Appreciate the post and all the details, I’m on the verge of setting up my own so I’ll probably come back when I do :)

I broke my immich during an update, i ponder waiting for the next update or reinstalling it

Might as well go for it. There will be more updates in the future.

I run docker within LXC too, but I have just a pair of docker instances with a collection of compose stacks on each. I really need stickers.

Same. When I first started down that path I kept finding posts about how it was impossible to set up and maintain but it’s been pretty painless for me. I basically had no clue what I was doing and I figured it out just fine.

@ndupont Stickers, yes. Colored sticks. I put little stickers all over the computers. Most stickers have mundane details (like host name, Subnet, etc.) but I also use a sticker for each major service. Everything else is documented in notebooks or spreadsheets. It’s probably overkill, but, between my memory and the home-grown gradual evolution of my #homelab, stickers really useful. I have 9 or so computers (rack-mounted), more controllers spread around (e.g., ESP32s running WLED, Pi Zeros), and nearly 250 individual devices for #homeautomation. I’m trying to learn about FOSS for configuration control but that’s slow going.

250! Wow, I thought I was already nuts with my 42 ZigBee devices + 10 reflashed sonoff over WiFi and my 8 cameras, doorbell etc… Let’s round up to 60; you are way ahead (oO)

I run everything on a single N100 mini NAS/PC with a remote slave backup NAS and the odd raspberry pi.

I’m more concerned about having to document in case I die than being able to keep all in mind, so far

@ndupont Friend, you won’t need the documentation in case you die. You’ll need it because you will forget (like me) what you did and why.

The day I hit a tree with my sports car or what have you, my family would certainly be less sad if they could keep the lights on, control the heating and so on.

It’s made so you could remove node red and still have basic functionality, but I really need to document a bit, share passwords, so a trusted friend could help them just in case

Acronyms, initialisms, abbreviations, contractions, and other phrases which expand to something larger, that I’ve seen in this thread:

[Thread #68 for this comm, first seen 31st Jul 2026, 18:30] [FAQ] [Full list] [Contact] [Source code]

Damn that case takes me back

I really like those big cases. Lots of room for expansion and cooling I’ve got an old CoolerMaster Cosmos II.

I took the bottom rails and retrofitted casters on it. It’s huge and hard to move around. Stands about 30" tall.

I love the Cosmos cases! I’m still using the Cosmos SE for my gaming pc. Computers should have handles!

This one? That’s a pretty nice look to it.

That’s the one! Thanks.

Love the front sticker

And yes - I’m running DSM (Arc Loader) in a VM with PCIe passthrough for the NAS. It’s a bit overkill, but I wanted proper storage virtualization with direct hardware access.

Whats this like, I’m running TrueNAS (the BSD version) with PCIe pass-through for my HBA and a couple of SATA ports - all in a VM on the latest proxmox

What’s your backup strategy?

Nice that’s similar to my setup! As for a website hub yours looks neat, mine is plain as hell unless you’re a family member, thanks to Authelia – but making public info is on my todo list – I tried custom-building a website many years back and adapted that. (If anyone discovers security flaws please lmk, but I’m preeetty sure I did good)

Gonna share mine :3

• i3 6100 (2 cores, 4 threads), iGPU passthrough to a LibreELEC VM so I could finally replace the Google CCwGTV 4K*)
• 24GB of RAM (16GB+8GB, mobo is ITX so that’s it)
• RTX 3060 12GB (handles GPU-accelerated processing with GPU passthrough to Emby and LLM LXC’s)
• 256GB Samsung 850 Evo SATA SSD for everything but media
• SATA HDDs for media:

1. Shucked Barracuda Pro 10TB
2. Ironwolf Pro 20TB
3. Ironwolf Pro 24TB

• Proxmox 9.2.3, kernel 7.0.6-2 (not quite sure why it’s not using .14)

I conspire that it was being updated to become weaker. Reasons: 1. Menu got laggier; 2. WiFi got real glitchy and wouldn’t connect on wake and I had to either: - d/c and r/c, - power cycle, - or forget the WiFi and reenter details in order to make it work And as the months went on reconnecting stopped working, power cycling stopped working and it stopped giving me a keyboard, instead prompting to “enter on your remote device” so my last remaining technique was to install the Google Home app, set everything up and use my phone keyboard to enter the WiFi creds. I conspired that as it was fine until recently and it went downhill so rapidly, that it must have been forced obsolescence and they’ll probably release a replacement device. Surprise – one month after I replaced it with Kodi, Google released the “Google TV Streamer 4K”.

Once I can upgrade my 2017 flagship to a 2027 midrange I’ll be transferring the i7-7700K and the other half of the 2x16 RAM, and I’m in dire need of more storage so a 30TB Ironwolf is in my sights (but fuck, it’s 2x its pre AI-pocalypse price (aka Sep 2025).

Getting some cameras and running frigate+ is a good way to utilise the GPU. My 2060 6gb runs 10-30% usage with 3 cameras. Can do cool things like send me a notification using ntfy if a person is seen between 2300-0500 something like that.

I did have ollama set up for local models but I’ve found anything that will run on 6gb isn’t worth using anyway.

Your homelab looks gay

Ah yes, an appimage packaged into a Flatpak packaged around Docker packaged into a proot+bubblewrap script which is packaged back into a Docker container (but a newer version) running under a Snap packaging of LXC under Flatpak Snap.

The app is running Java of course.

Um, OK… what do you DO with this? It just seems like a hermetic exercise in digital hoarding.

“Because it was there”

“HugeNerd”

“Ok so that was a fucking lie”

Why not?

How do you orchestrate all those lxc containers and vms?

Nice! I have a similar build! I recently discovered proxmox and now have my home assistant, Plex server, and a full arr stack up on it. I should look into automated backing up of photos 🤔. Turns out, hording all my old gaming PC hardware was the right call. With component prices these day it feels like a small fortune having extra hardware, drives, and ram laying around! (The wife finally agrees lol 🤣)

I’ve always heard that mixing ram kits is bad, I’m curious about the 16 + 8 channels. Does that give better performance?

Nice to have lightweight Proxmox and LXC, but this rtx3050 is draining power 24/7?

I really hope you remembered to give yourself a bit of leeway on your PSU. I forgot that the last time I upgraded my pc. ended up with a lot of power failures running a pc that needed 850 watts on a 750 watt PSU. I’ve since stopped being so stupid and gave myself extra leeway with a 1k watt PSU. (my pc needed around 850 watts and I hate not having a bit of leeway for upgrading).

Yess I do.

At some point, the old power supply started making strange noises, so I replaced it with a new one. I went from 500W to 1000W, mainly for the extra headroom. I know it’s overkill.

I’m sure in about 10-15 years 1000W PSUs will not be overkill…