Yup, debian is where I was before Ubuntu, and where I went back to. Still what I run mostly, plus a few different flavors of it (proxmox for example).

Though I’m also running an arch desktop on one of my play machines, kind of reminds me of having to write my x conf out in the 90s! Not bad overall.

(Never giving up my deb stable servers though!)

Yes, and I don’t consider that an “easy to disable” option for regular users, but that’s just my opinion.

“Easy to disable” is also the wrong approach, IMO. It should have been “easy to enable” - stuff like this should always be opt-in, not opt-out. Opt-out, to me, demonstrates a company’s motivations more than anything else.

Ehh… not at first. That was a later release.

OK, so what you probably won’t get much out of would be load balancing knowledge, from your description the CPU far outpaces everything else you have running services today. To get a good handle on that sort of thing, its handy to have comparable hardware for each node.

But the CPU is more than enough for most general task services, so yeah that will do fine. In terms of the GPU, yes, that will work for AI tasks as far as I know, most of the hardware I’m using for that is work stuff I get my hands on, so I couldn’t tell you much about the performance of the 3070 specifically, and I doubt a 6000 Ada as a reference w9uld be helpful, so maybe others can chime in on that aspect.

Since its mostly for learning, yeah, go for it. If you want to run i5 24x7, I’d probably want to separate out some of that CPU from that PSU purely for power management/cost to run, but yes its more than adequate for most services you’d throw on there.

Most of the servers I’m running are using a CPU that came out about 5 years before that Ryzen, but they are also lower wattage systems. Since they dont need a ton of CPU at all times, this is more the ideal for continually running home services, but not the only way to do it.

So build away and enjoy

Work for… What exactly?

Server requirements vary, depending on what they are doing.

ETA: I run about a dozen servers at home. Of them, none are more than what you have there. And I run a lot of services on each. Its what they are doing that matters.

#3 is the route I’m going.

Bigscreen is still pretty rough though, I’m trying to see if I can resolve some open issues to submit back to resolve, but in the meantime I’m going to start playing with flex launcher - https://complexlogic.github.io/flex-launcher/

Its likely to be the way I go as of now.

Lutris to be a gaming interface (retro games and Roms), jellyfin for movies/shows/music, gcompris for some kids educational stuff, etc.

I want to figure out a remote that I like and get some CEC testing done, may look towards using my homeassistant to act as a control system if its a pain (and most CEC is implemented poorly IMO).

But I’m done with stuff like Chromecast, rokus, etc.

If you’re trying out mobile or bigscreen, sure, that’s an option.

If I’m trying to get bigscreen to work, and I need to make code changes to do so, a window is better.

Super useful for running plasma-mobile or bigscreen for testing too!

If its the only bit its doing or its critical for other services, yes, nofail is a good choice.

If you can map a network drive (very east fstab edit BTW), then yes, its a great way to go.

That’s what I do, I have two 5-bay NASs, both use all 4 uplinks (LAG) to my switch, and my media server is an LXC on an 8th gen intel, with GPU passthrough.

If you reboot your nas, you may need to reconnect from the server. If you reboot your server, you dont have to do anything since its connecting when it starts up. If you end up needing more space, you just mount that new NAS alongside it.

To me its the better approach.

That’s one core

I’d lean towards the pi being the problem, but you can test the network throughput with iperf, and would want to test the videos outside of Kodi on the pi, so you could also check top and see what the processing looks like.

If I remember my pi 4 hardware decoding specs correctly, I believe h.264, MPEG 2, and VC1, and some support for HEVC. If I had to guess, you may have some codecs that aren’t handled by hardware acceleration, and instead just CPU.

My best rec would be to use either a dedicated stream box (like a fire stick, Nvidia shield, etc) which has better codec support, or pick up like a little Intel n100 based system, which will handle a drastically wider set of codecs with full acceleration support.

Right now I’ve got a Roku and a Google TV Chromecast, and I’ve been trying with various environments on an old Lenovo m910q so I can find my favorite fit of UI/distro. The Roku and Chromecast never stutter, and I don’t do transcoding for inside the home. Works with 4K HDR HEVC no problem.

Edit: Autocorrect annoyances.

Tiny/mini/micro makes up my server environment (and two customs using old cases and replaced parts).

Storage is a 1520+ and the two customs, with the 1515+ for backups I don’t want to lose (syncs to two other locations).

Tiny/mini/micro is the majority of compute tasks, mostly proxmox, LXC’s, and a few VMs.

The little machines have plenty of processing power, usually nvme but I can add it on if needed. Combine it with network storage, and you don’t need anything else imo.

Bonus is they are small and cheap as off lease machines being auctioned off.

Plenty familiar with qubes, just not familiar with any current consoles.

That’s gaming services, so I guess it’s windows only then, you’re right. Like I said, no idea regarding anything about current consoles for me. Haven’t played on a console since the 360.

I’ll be totally candid, I have no idea what version of Xbox is current these days.

But didn’t they have a CVE a few months ago they previously claimed wasn’t an issue?

Edit: Found it. https://msrc.microsoft.com/update-guide/vulnerability/CVE-2024-28916

… Just watch it again.

(For that “just tape them up” energy)

There is nothing Microsoft I would consider “top tier” when it comes to security.

Defender does a great job for many AV tasks. Crowdstrike does more, and protection isn’t tied to windows updates.

This isn’t a situation where companies just chose not to use the free item, the free item has other costs (management overhead) and is missing some features.

The best answer, of course, is to not use windows for anything that needs to be secure.

Edit: For those who think I’m wrong, cool. I’m not but you are welcome to disagree.

There is a difference between the free defender and paid for defender. If you’re a home user, check out defenderui.com to get (many, not all) features that are normally limited to intune/gpo.

A full and proper deployed defender stack is very good, but in terms of management… The approach to different os’s is practically cobbled together, the webui is horrific, and it lacks some basic functionality. A problem to manage a system like this is a problem to deploy a system like this.

If you’re on the free Defender level, you are not getting anywhere near the same features as falcon, there is absolutely zero question about that.

Most mobile clients you’re going to get your search and browsing through OPDS - so a library and a search function, but no tag support. Just (afaik) author, title, publisher, year, etc.

So that kind of fuzzy sorting is, at best, limited to the web interface for servers that support it (like Kavita). Which means browsing in almost any context native to a reader device/app is not going to support tagging.

If that changes, then sure, it could be plenty useful as a single giant list with neatly browsable tags. As of what’s out there now and usable (again, afaik) it is not.

Kavita does that, which is what I use (though through the web interface, not using opds on an android app for example), but it would still make browsing just a giant single list.

And I agree, fuzzy has value, which is why I don’t want to separate major things like Science Fiction vs Fantasy. But there isn’t exactly going to be significant overlap between historical romance novels and an instructional book on erlang, so that book on erlang is just going to get lost in the library.

That’s why I separate, it’s just too much for a single large directory.