I used proxmox to set up my ZFS pools and use bind mounts. It’s fine, I’m sure it’s a “grass is greener” thing.

Home labbing is a winter hobby, so in the summer months I hate the time spent updating all the machines when I could be outside.

If I had purely Docker set up, in winter I’d be complaining that “everything is too simple” and “I want more control” etc.

You can pry proxmox from my cold dead hands.

I do sometimes dream of running everything in Docker though for how easy it is to update. I’ve got the community scripts running and still it’s a bit of a maintenance job.

A TrueNAS + Docker machine is pretty tempting. If I were to migrate, that’s where I’d go.

Don’t start here. Get something tiny: some ewaste, a rPi3/4 or an n100.

Build a Pihole to block ads, malicious sites and trackers on your network

Risk free, tonnes of learning opportunities, huge utility, tonnes of documentation and guides to help.

Once you’ve built a couple Piholes (break and rebuild then) you’ll have an idea of what you might want to do next and what is achievable for you.

I don’t believe so. Maybe someone’s written a script on github, I haven’t looked.

A thing I like about lazylibrarian is that it just keeps rerolling until success. You probably miss good files just because LL couldn’t parse the folder structure or something, but it’s just set and forget.

Perhaps this could be modified to work. Like time is set to zero and file size to zero.

Not that I use LL, I just think it’s neat… From a purely onlooker POV.

Here’s my list, saved you a click. Ignore the *.iso, I added that for the seven seas peoples. Obviously, all I share are Linux ISOs

*.exe

*.sh

*.lnk

*.iso

*.zip

*.zipx

*.iz

*.izh

*.arj

*.scr

*.lnk

*.cmd

*.msi

*.bat

*.scf

Have your torrent client not download malicious file types like .exe, doesn’t solve the immediate problem but it helps in two ways.

First you don’t help spread the nonsense.

Second qbit will mark the torrent as complete and sonarr will then flag it for manual import with the reason no valid files to import. They’re easy and quick to spot and reroll that way.

Oooo this might be the path I take to finally get off IPv4. Cheers. I’ve already set up reverse proxies, but finally updating to 1999 technology seems like a good plan.

+1

I’m running a media/back up server for 4 households on a single n100 mini pc and a couple USB drives. It’s a “good enough”, low cost, high wife acceptance factor entry point into self-hosting. It’ll happily age into a firewall if I want to build a better box later on.

It’s revealing what I do/don’t need vs what I want. It’s teaching me what people use, what they don’t and where I might want to go in the future.

If I could go again I’d probably get a n100 2Bay Ugreen thing. Then it’d age into a local back up and I wouldn’t have to deal with USB drives.

I must have been having more basic problems than you. I found LLMs to present the most common solution, and generally the most common way of setting it up is the “right-way”, At least for a beginner. Then I’d quiz it on what docker compose environments do, what “ports: ####:####” meant, how I could route one container through another. All very basic stuff. Challenge: ask gpt

what does "ports:

-####:####" mean in a docker compose?

Then tell me it doesn’t spit out something a hobbiest could understand, immediately start applying, and is generally correct? Beginners, still verify what gpt spits out.

By the time I wanted to do non-standard stuff I was better equipped with the fundamentals of hobbiest deployment and how to coax an LLM into doing what I needed. It won’t write an Nginx config for you, or an ACL file, but with the documentation and an LLM you could teach yourself to write one.

Goes without saying I’d take the output of the LLM to Google for verification, then back to the LLM for a hobbiest’s explaination, back to Google for verification… Also, all details are place holders: don’t give it your email, api-keys, domains, nothing. Learn to scrub your input there and it’ll be a habit here is a bonus too.

Properly made software has great documentation and logs. If you know how to access those logs and read documentation (both skills in themselves)… Not to mention not all software is “properly made” some of it is bare bones and just works™. Works it do, absolutely not a criticisms for FOSS projects, I love your stuff keep making it, and I’ll keep finding ways to teach myself to use it.

For all its flaws. Low level tech support, rubber duck, command explainer is something LLMs do really well. Kept my early mistakes off the web and got me where I needed to be most times.

Thank you, saved. I’d still be a little but fucked, but not entirely.

I guess I should get on with learning/donating to this soon while the sun’s still shining with Tailscale. Slower, more gentle transition, fuck up in some dev environment instead of my actual server. Goddamn was tailscale so easy though.

Big words. I hope, though don’t trust, they can live up to them. But if tailscale goes, I’m just plain fucked. Thats certainly an indicator they’re worth some money to me, but there’s many a FOSS project before I get to paying a VC one.

As an aside, an interesting service would be a fund allocation type thing. You donate £x, tick which services you use and the funds get divvied up by what you use. Only able to donate £10 but use a lot of services? Each service gets very little, too little to donate as an individual, so little the individual doesn’t. But, on aggregate (with hundreds, or dozens of users) it would add up to a worthwhile donation. I thought of "round robin"ing my donations: pihole gets 10 this month, jellyfin the next, audiobookshelf the month after that… but yikes the admin.

Funds are donated when £x is accrued at the end of the month, and the service is maintained by earning interest on the funds held through the month. Idealistic, ripe for abuse, and out of my league to write and administrate. I promise I’d publish all the finances to keep me honest though.

Someone doesn’t know the folly of extending straight lines graphs into the future.

Vikunja feature request: once a day export “due today” tasks to printer, mark as done when printed. One day I’ll learn python and could script this myself… One day.

Vikunja is how my fiancée and I keep track of housework, we’re both neurospicy. But, like the author if I forget one day, the system completely breaks until I make a conscious effort to start again. By then the list is over run with "overdue"s it’s a little disheartening.

Hardware wise I’d go AIO. A mini and a pair of mirrored USB drives is my setup. I have an off-site backup running: another mini + USB. Finally, I have an inherited laptop as a redundant network box/local backup/immich compute. I have 5 households on my network, and aside from immich spiking in resources (hence the laptop), I have overhead to spare.

An n100 mini (or n150, n200, whatever) is cheap enough and powerful enough for you to jump in, decide if you want to spend more later. They’re small, quiet, reasonable Value for Money, easy Wife Acceptance Factor, and can age into a bunch of devices if you decided self hosting isn’t for you. I’d make a retro console out of any spare mini.

This way, when spending £x00s on a server, you’ll have some idea on what you actually need/want. The n100 can the age into a firewall/network box/local back up/etc if/when you upgrade.

All that said. An AIO storage-compute box is where I’m headed. I now know I need a dedicated graphics card for the immich demand. I now know I want a graphics card for generative AI hobby stuff. I know how much storage I need for everyone’s photos, and favorite entertainment, permanent stuff. I know how much storage I need for stuff being churned, temporary stuff. I now know I don’t care about high availability/clusters. I now know… Finally, the ‘Wife’ has grown used to having a server in the house: it’s a thing I have, and do, which she benefits from. So, a bigger, more expensive, and probably louder box, is an easier sell.

You’re right, I don’t take advantage of any of these features. I should.

Partly because of lack of know how on my part. So I don’t trust myself to successfully have it log into my email, get what it needs and leave everything else untouched. My main uploads, payslips and bank statements, are behind their own apps too.

Partly because paperless is isolated in it’s own little container (in my setup at least) so access to the consume folder is behind another step, I could syncthing it… I just haven’t.

And partly because I use the android app as my main interaction with Paperless. The app uses my phone as a good-enough scanner.

Paperless - Pay slips, Bank statements, MOT records, Insurance policies, User manuals, restaurant menus. All filed and searchable. Letters I get are photographed, uploaded and immediately disposed of, zero stress.