I suspect this machine might be memory constrained and if so zfs might push it to its limits if it’s already close.

If it has <8G and doesn’t already have decent headroom I’d think twice about ZFS depending on how its gong to be used

I worry I could be risking data corruption or something swapping to this setup

I really hope this is just a turn of speech and you’re not actually planning to put swap on those HDDs

If you launch a process from CLI, you can redirect its standard output (stdout) and standard error (stderr) to files. Errors messages might be there.

foo >> /tmp/gamelog.out 2>> /tmp/gamelog.err  

Maybe the application is already saving logs or crashdumps to disk. If the game has just crashed, you can search for files modified during the last minute: find $HOME /var/log -mmin -1

That might turn up some clues on where to look.

If you have to dig deeper, look into strace (premium subscribers only).

Bah.

Mint’s fine and Kate runs all right on it and Cinnamon. All it takes is configuring your normal Qt theming to be pretty.

With your edit that yields a succinct proof that things are nuanced 😉

Nope but I guess a workaround would be to make a oneshot workaround-nvidia-gpu.service systemd unit file that runs the command and have the lxc autostart depend on it?

Might be something about PCI resets that running the command triggers 🤷‍♀️

Right, there’s the immutable root aspect. Guessing the other answer you got fills in the missing piece there and that Silverblue perhaps mounts the system flatpaks on a different r/w filesystem than the read-only /. Check output of mount to see.

At the end of the day it’s up to you if you prefer to keep the system clean and run flatpak unprivileged, or centralize updates under root.

The one catch I can think of with flatpak --user is that it obviously won’t work if /home is mounted with noexec, which is otherwise a good security measure (and IMO not doing that defeats a lot of the security wins of immutable distros). Unless you apply the same mounting strategy to the flatpak xdg user dirs, which is certainly an option but not something everyone will bother with. But then again maybe that’s exactly what you want anyway to make your Flatpak installations smoothly portable across distros.

You are not helping!

(But also not wrong)

BunDGiE LiNuxXx 🤪

There are some bug fixes that are relevant for installs and base images. For example, security update in GPG is probably not a big deal for you but might be for someone building and pushing things from these.
Kernel, firmware and microcode updates might only affect a small minority of users depending on hardware.

Internet should work if i disable the vpn app though, so idk what’s wrong there.

I don’t suppose you enabled the kill-switch feature?

https://mullvad.net/en/help/using-mullvad-vpn-app#temporarily-blocked-internet

Operating and securing Postgres is a steeper learning curve. MariaDB is more forgiving for best-effort shoestring setups without compensating scalability for it.

As a dev I’m agnostic, as an owner and computer scientiest I prefer Postgres, as a sysadmin or *Ops I will put my hand up for MariaDB any day if I’ll be on call or maintain deployments.

And thank you for the refining exchange!

I also recognize that both the rave scene and free software are enabled in part by people with cushy high-paying jobs and what Lemmy would call rich kids who don’t mind sinking some money (and sometimes employer goodwill) into their passion without expecting any returns.

You can replicate across more than one provider and do automated regular monitoring that backups are still accessible.

If one goes down you hopefully have time to figure out a replacment before the other(s) do.

Probably not worth it for a bunch of xvid dvdrips or historical archives of full system-level backups but for critical data it’s sensible.

I haven’t dug into them deeper but Fossify have what seems to be decent basic options for all of the base Android apps: Phone, SMS, keyboard, camera, etc.

Just replacing all the stock apps with the Fossify suite looks like it could be an easy privacy win for someone stuck on a device with locked bootloader and dodgy stock apps from vendor or Google.

I’m involved with people organizing free rave parties of all sizes and production grades and it’s something I hold dear so your analogy hits close to home!

They all have income streams from outside the scene, including the ones responsible for events with thousands of attendants. While there are countless stories of people making industry connections promoting their careers and getting work there, a DJ or producer expecting they will be able to sustain a professional career purely through scene exposure or free parties is delusional.

That a few have been fortunate and resourceful enough to do so for a while is great but it’s not an indictement of the scene if one of them makes a “The Scene Is Dead” post on Instagram that they’re tired of the freeloaders and only doing paid gigs from now on. If they then continue publicly theorizing on how one could successfully financially exploit this community, they shouldn’t be surprised if the people who have been volunteering (usually a better characterization than charity IMO) for years feel rubbed the wrong way.

it’s bizarre to me to see the “fuck AI in every way” place turn around and attack this guy

Agreed in the mobbing of the wider thread but I hope you don’t see that going on here?

No errors or output from the add?

I don’t see anything wrong in what you are doing assuming you have permissions but if it’s just for your user you can flatpak --user to install in your homedir instead of system-wide.

Also convenient for distro-hoppers as you can just share or copy the flatpak dirs between home directories so you don’t even have to redownload for every reinstall.

What you can do is segregate networks.

If the browser runs in, say, a VM with only access to the intranet and no internet access at all, this risk is greatly reduced.

Hex Launcher: https://f-droid.org/packages/com.mrmannwood.hexlauncher

Pie Launcher: https://f-droid.org/packages/de.markusfisch.android.pielauncher

Similar approach in both of these

Best coupled with frequent refactoring and breaking of APIs so any community efforts at documentation are eternally outdated.