I currently keep everything LAN-only because I haven’t figured out how to properly set up outside access yet.
(I would like to have Home Assistant available either over the Internet or via VPN so that automations keyed off people’s location outside the home would work.)
Two GPUs? Is that a thing? How does that work on a desktop?
GPUs these days aren’t like your old Voodoo, with its daisy-chained VGA port and one-way, fixed-function graphics pipeline. They can actually send the results of their calculations back to the CPU over the PCIe bus instead of only out to the monitor!
(In all seriousness though, you don’t actually need two GPUs.)
I really don’t need this and I don’t get their appeal [of RGB].
It costs manufacturers pennies to include but gives them an excuse to mark the price up by dollars, so every single one of them shoves it down all our throats.
Well, let’s see:
You no longer have to set jumpers to “master” or “slave” on your hard drives, both because we don’t put two drives on the same ribbon cable anymore and because the terminology is considered kinda offensive.
Speaking of jumpers, there’s a distinct lack of them on motherboards these days compared to the ones you’re familiar with: everything’s got to be configured in firmware instead.
There’s a thing called “plug 'n play” now, so you don’t have to worry about IRQ conflicts etc.
Make sure your power supply is “ATX”, not just “AT”. The computer has a soft on/off switch controlled through the motherboard now – the hard switch on the PSU itself can just normally stay on.
Cooling is a much bigger deal than it was last time you built a PC. CPUs require not just heat sinks now, but fans too! You’re even going to want some extra fans to cool the inside of the case instead of relying on the PSU fan to do it.
A lot more functionality is integrated onto motherboards these days, so you don’t need nearly as big a case or as many expansion slots as you used to. In fact, you could probably get by without any ISA slots at all!
(Side note: Make sure to follow good practices. Feel free to ask if you want more information)
Not OP, but I’d like some more information about following good practices, please, especially in terms of “the best way” to make services available outside my lan (forwarding ports vs. a reverse proxy vs. a tunnel vs. a vpn – assuming some of those terms aren’t the same thing and I’m too much of a noob to realize).
The other major problem I’ve ran into, is that HAOS assumes that you would have no need to run any other Docker services other than those that are add-ons or Home Assistant itself.
With the caveat that I can tell just from your post that I certainly know way less about this stuff than you do, HAOS’ assumption seems pretty reasonable to me. Isn’t the point of using HAOS (as opposed to installing HA some other way) that you’d be either (a) using it by itself on bare-metal hardware, or (b) using it in a VM? I’m running HAOS and Docker in two different VMs on Proxmox, and it’s working fine for me so far.
(The first complaint you mentioned, about reverse proxies and subpaths, sounds a lot more legitimate. In fact, that’s something I’d like to learn more about because I haven’t yet figured out how to make my HA install – or anything, for that matter – accessible outside my LAN and “Tailscale Funnel” sounds intriguing.)
https://alternativeto.net/software/duolingo/?license=opensource
Of the programs in that list, the only one I’ve heard of before is LibreLingo, and I’m not sure how good or bad it is. (It seems different enough from LinguaCafe that they might be complement each other more than compete.)
I think $700-800 for a server with SFP ports sounds like good value in terms of price relative to capability, but the absolute price and capability are probably overkill for a residential use-case (even a homelab one). It’s a no-brainer if you’re the Other Linus (the Tech Tips one) and have unlimited budget for all the latest electro-bling in your house, but if you’re any sort of normal person you don’t need 10 gig networking yet.
Does Minisforum make anything with 4 ethernet ports and a <100W TDP in the <$300 range? If so, get that instead.
are the plans to turn this a little bit into a P2P archive.org?
Now that would be cool!
I’m not sure what’s required for AI. I feel like it is similar to crypto mining, massive compute but relatively small amounts of data.
If you’re talking about training models, I think it requires both massive compute and massive amounts of data.
I’ve looked all over the Internet for <16" short-depth cases multiple times over the years, and I’ve learned the most important question is this: what do you want to put in it?
I’ll tell you right now that some of the things I’ve wanted, like…
A 2U with hot-swap caddies all the way across the front (like this or this, but <16" deep at the cost of fitting only an ITX motherboard), or
A 4- or 5U chassis that can fit an EATX / SSI EEB (12"x13") motherboard and an ATX PSU at the same time, or
A chassis of any size that has both the motherboard/PCI I/O and the drive bays on the front (front drive bays are normal and you can get front access I/O, but not both at the same time)
…simply do not exist, as far as I can tell. I’m pretty sure all of these things are geometrically possible (I did the math), but apparently I’m the only one who wants such weird stuff.
You can get a basic-bitch whatever-U case that supports a mATX motherboard and hard-mounting a couple of internal drives, with sharp metal edges, a shitty plastic door, and a price double (or more) what similarly low quality would cost in a desktop form-factor all day long, though.
This might be an X/Y problem. Why do you think you need HDMI output on a server?
Another question: Why do you have several debians-vm’s? You also could take one, right?
As I wrote in my other reply, you typically want a separate VM for each service so that the OS configurations don’t conflict, and also so that you can shut down the VM for one service (e.g. for installing updates or migrating to another cluster node) without causing downtime to other services.
a) What Linux OS do i take? Ubuntu Server?
Typically folks either pick what they like best or pick what’s recommended by the service they’re trying to run. (Remember, typically you run one service per VM, so everything about the VM can be tailored to that service. That’s pretty much the whole point of virtualization – so that you don’t have to get multiple services cooperating on the same machine.)
My default go-to would be Debian, but again, it’s really a matter of personal preference.
b) Should it be headless?
GUIs take up disk space, RAM and CPU cycles, so it’s more efficient not to have them (especially when you’re virtualizing and therefore running separate copies per VM). However, this is 2023, not 1993, so it’s not that big a deal.
would there be any advantages of installing an OS with a GUI?
The advantage would be that you could administer the VM and the service inside it using a GUI, if you’re into that sort of thing.
In general, most services are designed to be administered over SSH or via a web interface, so a GUI shouldn’t be necessary. Also, in general you ought to be scripting the administration of your VMs themselves using e.g. Ansible, so a GUI shouldn’t be necessary for that, either.
No, containers are basically sandboxed applications+dependencies running on top of the host’s kernel. VMs run their own separate kernel. If anything, a container is less “wrapped” than a VM.
Do you even need a reverse proxy if you’re using Tailscale? What advantage does it give you over setting up your DHCP correctly such that you can access your services by hostname?
What made you pick Mealie over other stuff like Nextcloud Cookbook or Grocy or whatnot?
I’m suddenly more motivated to do my French lessons on Duolingo now.
Data caps on home internet services should be illegal. They should also be much higher on mobile, but that’s a whole other topic.
I’m not convinced mobile deserves to have caps at all, either!
As far as I’m concerned, there’s no reason to limit the amount of data transfer except in times of congestion, and I also don’t see any reason the amount of data transferred during un-congested times should have any bearing on who gets throttled.
Why can’t you use the app to do locations?