this is going to sound like gatekeeping, because it is.

yunohost allows anyone that couldn’t find the difference between /dev/null and their asshole the ability to deploy services on hardware. be it locally or remote VPC.

everything about that goes against the tenants that created the selfhost movement.

it also causes problems when those same ignorant “hosters” reach out to the community for help and again, can’t comprehend the technical nature of the issue at hand. they are akin to a plant that requires too much water and too much sun in the garden, and I generally pluck them out and toss them into the mulch pile.

that said, I’m sure there’s plenty of people who use yuno that have a technical background but lack the time/effort to DIY. my issue isn’t with them.

but you can fuck all the children you want as long as you’re in “the club”.

selfhosting

yunohost

pick one.

it’s pretty rare for dumbasses to point themselves out these days.

you’re doing gods work son. keep it up!

I would start with the NAS first. with a proper NAS and a solid network the storage requirements for the rest of your servers lessens.

for example, I’ve got an off-the-shelf NAS solution that hosts all my movies, music, etc that is mounted via docker in my Plex container on a different server. because it’s usually streaming to four different devices at any given moment, and the torrent’s running against the same NAS, I’m using a 10gbe copper line.

depending on your needs, a 1gbe should suffice.

having a NAS would also help with backups. I have a 5 bay NAS, one of them is dedicated for backups of both servers and cloud storage from the NAS (personal files, tax documents, etc).

also, when building your NAS whatever you think you need, double it. if you think 5tb is enough, you’ll want to get 10tb or even 15tb. I current run 15tb(8tb drives) in a raid 10 with a 20tb backup drive.

this kind of configuration allows me to run 1tb or even sub tb drives on my servers and reduces my overall costs to replace if anything goes wrong. with a raid 10 on my primary storage array I can easily replace bad drives.

the only time I’ll really hurt is if my backup drive fails. since it was so expensive due to the volume. but backup drives never fail, right? 😉

unfortunately because of AI all those prices have increased. if I were to build it today it would cost me around $1700. adjusted for current pricing my whole lab would probably cost around $10k (thanks ram!).

good luck and god speed

yes. all of the images I use are cached and stored in my locally hosted gitlab registry.

I think I’ve got around 120-140 images. a lot of what I have is just in case of an emergency.

I’ve always imagined I could build and run technological infrastructure after a social collapse or something, so I have a lot of images that could be a good basis to start with. Most OS images, popular DB images, etc. it would probably never work, but I’d rather have the option than not.

I’m currently running three hosts with a collection of around 40 containers.

one is the house host, one is the devops host, and one is the AI host.

I maintain images on the devops host and deploy them regularly. when one goes down or a container goes down, I am notified through mqtt on my phone. all hosts, services, ports, certs, etc are monitored.

no problems here. git gud I suppose?

• routine patching
• siem log aggregation
• proper alerting metrics and notifications
• routine virus scanning
• proper network segregation between your NATd network and your personal network
• firewall firewall firewall
• expose your applications to the internet through a WAF, never directly

if you can do all these things properly, then there shouldn’t be too much danger in selfhosting your apps publicly.

wekan and Jotty are very lightweight and flexible.

you know what would be really helpful? cross this with data from moderation. content removals, bans, permabans.

then, layer in mods/communities and you might find why users are leaving.

and here I thought it was because I’ve blocked close to 300 users.

I2P is a secure network protocol. to your ISP it’s just an encrypted stream.

it can work across any network connection. Bluetooth, Lora, ARRL, etc.

the way I interpreted your comment presented the possibility that the ISP would cut services based on the content being hosted, not as a wide area communications disruption.

https://geti2p.net/en/

your shits gay and you sound retarded

you’re that guy. congrats!

I2P

makes sense but wasn’t that already fixed with dockers rootless patches?

question, why would you leave docker for podman?

PostgreSQL

fuckin gross!