In my opinion trying to set up a highly available fault tolerant homelab adds a large amount of unnecessary complexity without an equivalent benefit. It’s good to have redundancy for essential services like DNS, but otherwise I think it’s better to focus on a robust backup and restore process so that if anything goes wrong you can just restore from a backup or start containers on another node.
I configure and deploy all my applications with Ansible roles. It can programmatically create config files, pass secrets, build or start containers, cycle containers automatically after config changes, basically everything you could need.
Sure it would be neat if services could fail over automatically but things only ever tend to break when I’m making changes anyway.
I watched a video from a guy who used machine learning to play Pokemon and he did a great analysis of the process. The most interesting part to me was how small changes to the reward system could produce such bizarre and unexpected behavior. He gave out rewards for exploring new areas by taking screenshots after every input and then comparing them against every previous one. Suddenly it became very fixated on a specific area of the game and he couldn’t figure out why. Turns out there was both flowers and water animating in that area so it triggered a lot of rewards without actually exploring. The AI literally got distracted looking at the beautiful landscape!
Anyway, that example helped me understand the challenges of this sort of software design. Super fascinating stuff.
Seamlessly syncing game saves between my Deck and my primary gaming PC is so nice. Before I travel I just make sure to wake up the deck long enough to get updates and sync saves.
For non steam games I use syncthing but that always requires just a little bit of work.
I jumped into Tumbleweed recently and have really been liking it. Last time I used Linux with a desktop environment I was using Gnome and KDE was a lot unglier. Things have definitely changed.
It’s like looking through a telescope. Everything within the lens is clear and detailed, but anything on the periphery might as well not exist. Very useful state of mind for certain coding tasks.
I use Ansible for all my deployments and just got a PXE boot set up with a preseed file to automate the install process and get the host ready to run playbooks.
I’ve been really pleased with this strategy overall. I think that Ansible works really well for programmatically generating config files which in turn makes moving applications between servers effortless. I control docker volume mounts with ansible variables and encrypt secrets with ansible vault so I can do everything all in one place.
Troubleshooting issues is a lot easier and recovering from a backup is faster and a requires less effort since I can just pull down the Ansible config from git and redeploy.
They love to blame flops like this on things like feminism, but generally it’s just because Disney has no idea what they are doing anymore after killing off the A-listers and then releasing nothing but wet farts for the last five years.
It’s pretty laborious to do this for casual browsing though. The websites I visit regularly where it’d be worth configuring this aren’t the ones with cookies I’m worried about.
Flatpaks are sandboxed to user space. I use Flatseal which allows you to grant flatpaks additional permissions. I used it to allow the flatpak version of syncthing to sync files that it otherwise lacked read/write permissions for.
That solution has worked really well for me and resolved my main frustration with flatpaks.
Understanding it or figuring out where it came from would be even worse than killing it. The characters knowing almost nothing except what they observe directly is what makes it so terrifying.
Obi Wan was the main character of the prequels but they kept thinking it was Anakin. Messes up the framing of a lot of events particularly in the first movie. By the third one I think they figured that out, but it has plenty of other problems even though the narrative focus was better.