You might consider using something like Cloudflared or Tailscale’s Funnels to proxy the connections through to prevent DDOSing and apply ACLs. You can still use your domains with those.

Been using this for years, runs like a top.

Easydns.ca

OK, yah, that’s good point about swarms. I’ve generally not used any swarmed filesystem stuff where I needed persistence, just shared databases, so it hasn’t come up.

Well, I know you can define volumes for other filesystem drivers, but with bind mounts, you don’t need to define the bind mount as you do, you can just specify the path directly in the container volumes and it will bind mount it. I was just wondering if there was any actual benefit to defining the volume manually over the simple way.

Is there any advantage to bind mounting that way? I’ve only ever done it by specifying the path directly in the container, usually ./data:data or some such. Never had a problem with it.

Absolutely. Even if your ISP is firewalling, never trust they will maintain it, and some of these cheapshit routers they use are awful. Use your own router and put it on the ISP routers DMZ.

Then you’d have to share them back from the host to the guest using NFS or samba, and there’s no reason to have them on the host. No, you’re fine. That’s what I would consider the right way to do it for what you describe as your purpose. Putting more layers in there isn’t serving a useful purpose, which is why I say to not bother with a NAS to store them, if your primary use of them is via the Arr stack VM.

They seem to go in spurts on the cryptobullshit. Sometimes you go for months without hearing anything about it, then it’s part of the next 5 LUP shows. I can certainly do without it, and I think he gets emails shitting on him for it and he tones down for a while.

I did walk away from all his shows for a couple years there because he was on and on about how anyone that’s still mad at Microsoft is stupid, and I couldn’t take it. It was like he was on the payroll there for a while.

It’s unfortunate, I’d have an issue too, I have Canadian farmer levels of swearing I don’t even notice myself doing it. And some of the sayings are probably racist as hell but it’s not meant that way, it’s just carryover from the old days.

In the end I think it worked out well for him, but some understanding that not every culture is as uptight and puritanical as America would have been in order.

Ah, I did not know that, but that’s one of the shows I don’t listen to. I listen to the Changelog shows for much of my dev stuff.

You’re doing it fine. You can overprovision your zpool that way for other VMs with datasets on the pool, and it only uses what it needs for each set. And you can exclude the media pool from backups and repllications if you like. There’s no performance hit the way you’r doing it.

I sure wouldn’t bother with any sort of NAS intermediary, you can share out the media pool via the Jellyfin VM if you really need to, but I can’t see why you would if you’re using Arr stack components already.

Jupiter Broadcasting? Still alive and kicking. Not sure what you mean by digitial ocean, thry had merged with Linux Academy for some odd reason, then A Cloud Guru bought LA, then Joe Ressington got shitcanned for using a very british slang in a meeting and seemed like right after that Chris took it back independent and he kept on with LUP, Selfhosted and a couple others I don’t listen to.

Ressington has gone on to make what I think is a very successful Linux podcasting network, Late Night Linux, certainly enjoy their shows. I’ve been a patron of his stuff for several years now, and there’s a good bunch of people hosting those shows these days.

INtersting link to the LLM DDOSing against FOSS project sites/repos that’s been going on.

I love this line:

and robots.txt is the only thing they don’t read off the site

What a shitshow.

ITT: “Haha, yah AI makes shitty insecure code!”

<mad scrabbling in background to review all the code committed in the last year>

I’m using vscode/Roocode with Gosucoder shortprompt, with Requesty providing models. Generally I’ll use R1 to outline a project and Claude to implement. The shortprompt seems to reduce the context quite a bit and hence the cost. I’ve heard about Cursor but haven’t tried it yet.

When you’re using local models, which ones are you using? The ones I mention don’t seem to give me much I can use, but I’m also probably asking more of them because I see what Claude can do. It might also be a problem with how Roocode uses them, though when I just jump into a chat and ask it to spit out code, I don’t get much better.

Check the returns policy, but if you could buy a large external drive at something like BestBuy, do your copy then return it, that might be a lot safer than what you’re talking about. Just a thought.

That koboldcpp is pretty interesting. Looks like I can load a draft model for spec decode as well as a pile of other things.

What local models have you been using for coding? I’ve been disappointed with things like deepseek-coder and the qwen-coder, it’s not even a patch on Claude, but that damn cost for anthropic has been killing me.

Fair enough, but it’s damn handy and simple to use. And I don’t know how to do speculative decoding with ollama, which massively speeds up the models for me.

Windows Docker is so bad, I don’t even know why it’s a thing.

Some good planning might make the migration less painful. I would recommend a ZFS or other COW storage solution under the docker host so you can do snapshot backups and not have to worry about quiesing databases, etc.