Here’s a link to an instance of radicle for Yuzu. It’s a p2p git server implementation. I haven’t looked too much into it yet, but the tech seems interesting.

It was many moons ago, but I recall this is correct. You had to place locations in a specific pattern for some of the mission items and it was totally possible to do with a single cart, just you basically needed a guide to figure out all the combinations to create the right map. It was not a trivial thing, and I recall using gamefaqs a lot.

FFTA is one of two games I have ever 100%'d. Played that game so much back in the day. The reward for 100% is awesome, and I have very fond memories for that game.

Ritz from FFTA. Always loved her character, a protagonist gone rogue to pursue what they believe in, even against their own friends. Then, when confronted directly, backs down and recognizes our own determination to end the isekai. Really excellent character arc and growth throughout the game. Loved that she also breaks the rules of the game itself by being a human with viera job classes.

I love the QoL improvements of those games, but I do love the small intimate story of LA a bit more. Don’t get me wrong, I adore the Subrosians, adorable nerds, but the little seagull at the end of LA gets me every time. You’re not a traditional hero in LA.

If you run it in podman, podman can export into a kubernete file, but its been a long time since I’ve tried it though. podman kube generate $CONTAINERNAME

is podman-compose really dead? Their github page looks active at a glance. The tooling is so similar, I use podman for local testing, and deploy to docker, but I’ve also done the reverse. As long as your not using really exotic parameters its really just a drop in replacement, I’ve even used GPU passthrough for AI project no problem in both docker and podman. At the end of the day, they’re just slightly different frontends for the same backend.

As far as docker support, its often as simple as just providing a Dockerfile, which is basically the same thing as your build scripts. These days I’ve often used the Dockerfile INSTEAD of the readme to find help compiling some projects.

You primary use cases seem very GPU heavy, could you not save some budget by going for an F series CPU maybe even go to a 13th gen? Might even be able to save more by going for DDR4. In gaming DDR4/5 seem reasonably close, but I have no idea how RAM sensitive productivity stuff is. Then you could swing up to a 4070ti or better depending on sales.

Totally reasonable, something like LVM can at least get you to a raid1 setup, pretty easily.

Raid0 (combining both drives’ capacities) is not really tiered storage. You would want Raid1 (each drive is a copy of the other drive ), but doing this isn’t a backup. How will you be monitoring the drives so that you know if one of them actually fails?

I don’t think the RPi has a new enough kernel, but with bcachefs you can do tiered storage. By combining the storage of the ssd + hardrives, into a single block device, then make the ssd the read/write cache, and give the whole pool replicas=2, so that that if one drive dies you still have the failover of the other drive. Do be aware this setup is still not a backup however.

It does make sense. Thank you. I appreciate the link!

However, my cloud usage is purely as a proxy/load balancer, as none of my cloud providers hold any actual data. They’re just routing traffic, and all data/processing is on premises. What I’m interested in, is how to setup something like what you describe, but on premises also. From a design stand point, if I wanted to protect myself from a ransomware attack, obviously my cloud backups would be lost because they’re a mounted filesystem during a backup eventually. So I don’t know how to wrap my head around handling this, just storage design wise as specific tools I can figure out. How does one create a recovery point, and keep it safe from something like this? Just image the entire file system from a live booted offline environment? Feels like a chicken-egg problem to me.

I’ve thought about how I could handle disaster recovery for my homelab environment, but I haven’t come to any good solutions. For example, if my main concern was being hit by crypto. I can’t just recover from a regular backup, since I’m not sure how I can make a backup without that backup just being encrypted along side everything else. Since I mainly just backup everything to my file server, which is then synced to the cloud. In that setup, my cloud backups would be lost as well.

Would you have some starting points on how others handle disaster recovery? I’d like to avoid manually making an offline backup, because inevitably I’d forget to do it, which would make it useless anyway.

I believe they changed some of their licensing from the fallout of their IPO. Just worth noting for the selfhosting crowd. I know terraform is being forked entirely, but I’m unfamiliar with the specifics beyond that.

This is super cool! Do you know if other systems have open source flash cart projects like this?

No worries, I’ll link to some Arch Wiki stuff to help explain. Containers are a very cool system for isolating environments. Similar to how python uses VENV to contain all the dependencies for a python program, containers let you have a full environment beyond just the python stuff. I use podman to actually run the container on my computer. You use a Containerfile, to define what you want this environment to look like, and docker/podman does all the hard work for you, by making an image file that holds the whole thing in one place separate from our real OS.

This is my start script.

#!/usr/bin/env bash

podman run -it --rm --name stablediff2 -p 7860:7860 \
-e COMMANDLINE_ARGS="--api --listen --port 7860 --enable-insecure-extension-access --medvram-sdxl --cors-allow-origins *" \
        --device /dev/dri:dev/dri \
        --device /dev/kfd:/dev/kfd \
        -v ./models:/dockerx/stable-diffusion-webui/models:z \
        -v ./repos:/dockerx/stable-diffusion-webui/repositories:z \
        -v ./extensions:/dockerx/stable-diffusion-webui/extensions:z \
        -v ./embeddings:/dockerx/stable-diffusion-webui/embeddings:z \
        -v ./outputs:/dockerx/stable-diffusion-webui/outputs:z \
        -v ./inputfiles:/dockerx/stable-diffusion-webui/inputfiles:z
        localhost:stablediffusion:latest

This is just telling podman to start the container, give it an actual terminal to connect to, remove the container if it stops running, give it a name, and tell it what ports it can run on.

podman run -it --rm --name stablediff2 -p 7860:7860

These are the arguments passed to the webui start script itself, mostly for my own convenience. The medvram-sdxl is not required, since my card has enough vram, but then I can’t be doing anything else with it. So I sacrifice a bit of generation speed for more free memory for the rest of my computer. I’m running this locally, so insecure extension access also doesn’t matter since I’m the only one using this, just makes installing extensions from the webui directly.

-e COMMANDLINE_ARGS="--api --listen --port 7860 --enable-insecure-extension-access --medvram-sdxl --cors-allow-origins *" \

These are just the device files that correspond to my GPU, so that the container has access to it. Without this, the container would only have access to CPU based generation. Everything else is just the folders that holds my models, extensions etc. You have to give the container exactly what you want it to, because its isolated away from your normal files unless you tell it otherwise.

--device /dev/dri:dev/dri \
--device /dev/kfd:/dev/kfd \

This is iterations per second, I believe. It’s basically a measure of how fast stablediffusion is is running a particular generation of an image. It lets people compare performance across different software and hardware configurations.

10 it/s

NixOS is the name of the GNU/Linux operating system I’m using, similar to how MacOS is different than Windows, NixOS is another type of operating system. I’ve only been using it for a few months, but its extremely cool. Before that I mostly used Debian and Fedora, but the main difference between NixOS and them is that you can define you whole OS as a configuration files, and then the tools it’s designed around build your system for you. So instead of say, installing a program, opening it up and going into settings and changing everything to be how you like it. You can instead just make a file that lists everything the way you want it from the start, and Nix installs the program and sets it all up all in one go. It has a pretty big learning curve, and its features are numerous that I have yet to take full advantage of them. Probably not the best to start with if you are new to GNU/Linux systems, but once you see the benefits of why it does things differently, its awesome.

Hopefully that explains most of the words I used. Pardon my formatting, as I don’t know markdown very well and I think I separated everything okay. :)

by default, set up to run in a venv

It does, but since I’m running inside a container, I disable that behavior, and run it as a user package. Some extensions also require additional libraries, but they don’t pull the correct ROCm dependencies and I have to modify part of the install scripts to manually define the correct versions.

The main webui code is excellent, even if sometimes the documention is out of step because of how fast everything moves. Its the extensions that are not always to the same level of quality that make fiddling with python dependencies a bit of extra work.

I run it in a container on a NixOS host yes, eventually I’ll learn how to do it in a flake but my nix skills aren’t quite there yet. EDIT: I use a 6900xt, and some quick runs I did give me roughly 10 it/s. Which feels reasonably fast, only a couple seconds per image.

I use stable diffusion with automatic1111’s webui ran locally with an AMD GPU. I use the card for gaming and encoding too, so the cost for just AI is basically free. The webui is excellent, and I learn about new things it can do every time I use it. Setting it up took some time, but nothing beyond what I am familiar with. I do loathe that so much data science/AI stuff is python based, because python’s dependency management is an unruly beast, but oh well.

What’s your budget? That more than anything else will shape your build.

I don’t know if this is what you are looking for but I used :z with podman mounting and it Just Works*.

podman run -d -v /dir:/var/lib/dir:z image

From the documentation :z or :Z relabels volumes for host and container usage depending.