Bezerker03

Oh wow today I learned. I thought it was just containers still. My apologies. Looks like it’s been a thing since 5.0 lts.

I mean that’s fine if you dont want kernel space isolation. Lxd and proxmox are not the same.

I mean you could. Shell accounts did this back in the day but yes users could still abuse the system.

I have two nginx ingress running on my cluster. One of private one public. Public one is what’s exposed on 80 and 443 to the net.

The private is only available via VPN or lan. The public is for services I want internet exposed.

My family have a VPN network set up to my lan on their router and have access to most services but the public stuff is for the internet friends

It wasnt containerized sadly but remember in a container you still share (albeit split by cgroups) kernel space and the kernel. Only userland is isolated.

So kernel level sploits are still a concern. Wasn’t the case here but still.

Oh sweet. Good to know!

Hmm. Not sure. Sunshine is a reverse engineered version of Nvidia gamestream. Prolly have to rely on steam for remote streaming with amd I’m not sure sorry. :(

Sunshine and moonlight. Works amazing.

As others mentioned other things need to connect to your instance so valid ssl matters

Is imagine the rest of the fediverse will refuse to connect tho.

Ultimately this is the thing to worry about. Threads will get the largest communities and as a result the main amount of attention and when/if Meta decides to defederate, it will ruin things. Also, people will generally give zero shits about federation because 99.9% of content will be on meta’s instance.

Ironically, the main thing keeping fediverse from being more popular (the decentralized approach and “multiple places the same community can exist”) are going to be the thing that kills it if Meta gets involved and becomes the big boy.

Idunno what is arguably worse. The fediverse being restricted to more “technical” folks who give a shit, and thus a far more limited audience than a central platform, or being suddenly disconnected from the hivemind after taking all the content.

(Fwiw, I absolutely think that the Threads fediverse plan is to totally absorb all the content and become the main place for it then possibly pull the plug but honestly at that point they won’t even need to the usage stats will basically do the same thing for them.)

Yeah I noticed that. Dunno they fulfilled my orders so we shall see lol. Ify. Instance goes offline randomly now we know why. Hahaa

yeah its pretty nice. I am running this lemmy instance on the 48 dollar one, and have another fully upgraded for my own other services.

When you order the 3 vcpu 4.5gb one, you can upgrade cpu cores, number of ips, and ram. Maxed out minus the IPs and you get it for around 161.

Well the server to run lemmy must be at least from that repo debian based. (Ubuntu or debian). Per the requirements.

Anisvle prolly could run on windows. Never tested.

Had my entire home setup (all my arr services, nextcloud, home assistant, monitoring, etc) all running in my k3s setup on like 5 vms at home. Had velero backups of it etc.

Fast forward to i have no idea what happened and my masters just died. Nothing should sync anymore etc. Nobody in k3s community had an idea either. So lost my entire cluster and the backups weren’t too useful since the cluster itself was dead.

Rebuilt with Talos. But man that sucked.

Fwiw racknerd still has their 2022 black Friday deals active. I just got 8 core 12 gb ram 100 gb SSD 12tb bandwidth for 161 annual (little over 13 a month). Just search it on Google. It’s still live on their store site.

Just a follow-up thanks for this recommendation. Working great!

I have a house with a basement and a fiber connection I run my stuff in. I also have a pair of vps I use for things from racknerd that were black Friday deals (160 a year for 8 core 12 gb ram)

Awesome! Thanks!!!