Thanks for sharing but I still don’t know what this actually does? As it’s win only at the moment I can’t give it a test run but the readme is so high level that I’m not sure.

Like, does it interface with my ollama models? Is it integrating with remote models? What’s it locating actually?

The screenshot had has the criteria included though. Relevant part: either be for children or for everyone.

I have set up an lts kernel in addition to the zen I use by default. See:

https://wiki.archlinux.org/title/Kernel

Disclaimer: this only works when something with image creation goes wrong with an update. Which didn’t happen to me ever - unless I did a mistake or tested some kernel stuff. I only had bootloader errors when I screwed up pacman though. The fallback kernel in that case is on a USB stick…

Hehe true. And even that happened to me after a couple of tired “Syu enter”. But then again I learned something new with nearly every repair!

Oh agreed! That’s why I’m with OP actually that arch might not be the right distro to go for.

The person I replied to basically said “that’s what you deserve for not doing it properly” if I understood it correctly - that’s what I’m confused about as well.

What precaution would you expect OP to would’ve done though? A fallback kernel would be my guess - that’s something many casual oriented distro do out of the box basically. . I read your post as “you’re right, don’t use arch” - something btw which I tend to agree with although I wouldn’t say that’s because of the precautions.

I use arch because there’s no black box magic. For an end user who expects or wants that… Yes, arch might not be the right choice.

Just curious, what distro do you use that systemd is not the default? (I at least you didn’t change it after the fact if you don’t have any feelings (towards unit systems ;) ) )

The first link goes into amazing detail on that. In short: all your information concerning location as well as current IP and some other metadata gets send to a basically unknown company with no transparency on how that data is handled.

I highly recommend reading the first, linked post though!

Cups

linux printing server - if you want to share a printer over network or just use one locally on a linux machine.

(not OP but same boat) Doesn’t really matter to me because google knows my servers external IP which is a non-issue: I don’t expect google to try to attack me individually but crawl data about me. There is no automatic link between my server and my personal browsing habits.

In terms of attack vector vs ease of use , self hosting searxng is a nobrainer for me - but I do have an external server available for things like that anyway so no additional overhead needed.

A Dockerfile itself is the instruction set. There is a certain minimum requirement expected from a server admin that differs from end-user requirements.

The ease of docker obfuscates that quite a bit but if you want to go full bare metal (or full AWS or GCS, etc etc) then you need to manage the full admin part as well - including custom deployments.

No worries I phrased that quite weird I think.

A NAS is only more power efficient if the additional power of a full server is not used. If for some reason the server is still needed than the NAS will be additional power consumption and not save anything.

(for example I run some quite RAM and compute heavy things on my server which no stock NAS could handle I think).

That would replace the computer with the NAS though and is not true for a server that you’d want to extend, right?

Is that 370watt across all of them or per fat server? I ask because three m5 sound like a lot of power drain!.

And thanks for sharing!

I didn’t know that about the immich app, thanks for pointing it out!

Then you need a third application (e.g. syncthing) to replicate the auto upload functionality of Nextcloud.

Personally I don’t want to have same functionality in a different stack because of pipeline issues. Doesn’t solve OPs issue I just wanted to point out that your solution might have drawbacks OP didn’t see at first glance :)

Thank you! That’s really interesting, the performance with a pi 3 was way worse - even more than the pure spec difference would’ve lead me to believe.

The OCR devs have made a really awesome job!

You are running a specific module of a project locally - not the whole project. The web server is an integral part - leaving it out makes you do a bit of the leg work: you’d need to figure out how the websites get built and deployed and then reverse engineer that for your android environment.

Personally I’m fascinated by that attempt and it could be an awesome learning opportunity. To be honest I don’t have the motivation to follow your path down this rabbit hole though.

If you decide to follow up I’d appreciate you giving updates from time to time about your insights! ♥

Thanks for sharing! The only thing I’m surprised to see in your list is paperless - how long does OCR take on a pi?

“Being a router” is what they are good for! Even needed.

Edit to be more specific: two switches in each of the 10gbit and redundant uplink would be a setup I can see, depending on your line.

No overkill there :)