Recent searches
Search options
#systemd
cool, i now get push notifications via #ntfy when any #systemd service fails and it will include the logs as well! this is a lot better than manually checking the status and logs every morning.
i thought this would be impossible in #nix due to infinite recursion and that i would have to manually set OnFailure= for every service, but i found a blog post with an easy workaround instead:
https://medium.com/red6-es/systemd-service-failure-notification-system-92761903c557
New Blog post after 3 years: "Using a simple python script in NixOS as a systemd Service without packaging it" https://euer.krebsco.de/using-a-simple-python-script-in-nixos-as-systemd-service-without-packaging-it.html #nixos #systemd #python
Working with a volunteer group who has a relatively simple Python app deployed via Docker Compose.
What did they get from this? "Automatic restarts."
Wrote them a #systemd service definition to run the app out of a plain virtualenv. All the same benefits, way fewer moving parts in production.
"ParticleOS: Systemd's Very Own Linux Distro in Making".
"In its current state, ParticleOS seems well-suited for developers who want full control over their system."
Wrong. People who want full control over their system don't use systemd. You should face it.
And I'm daring to make a bold statement: This seems to me that here #systemd actually enhances and helps with the application of the #UnixPhilosophy of using a modular design with multiple, small services and interconnecting them.
By creating reliable connections between them. Or if some connection breaks unrecoverably to properly propagate this to all affected users. And by using a simple, short, descriptive, non-programmatic language for service files.
I have to say, I'm getting more comfortable with #systemd services. And I find it fascinating how with its templates and powerful dependency feature I could achieve some wild, very dynamic stacking of @gstreamer, @pipewire, #mpd, #mpv and #SAP / #RFC2974 with only a few lines for various systemd services. So for instance a #multicast SAP announcement of @schenklradio will only appear if something plays on that pipewire device, which only happens if that radio is on air and reachable.
Proving Lennart Poettering wrong is what drives innovation.
ParticleOS – ein systemd-Betriebssystem
https://linuxnews.de/particleos-ein-systemd-betriebssystem/ #systemd #poettering #distributions #linux
I setup #systemd-boot on my computer, but there were a couple of annoyances:
The #kernel isn't signed for whatever reason. I wired in
sbctl signto the install script, but that wasn't completely straightforward.Since #Windows is installed in a separate #EFI (because Windows likes to fuck up the entire EFI partition sometimes) it couldn't "see" the Windows Boot Manager. I copied it to the other partition, but it will have to be manually updated whenever Microsoft changes it. Maybe that doesn't happen that often idk.
Side problem is that #Ubuntu's nvidia-lowlatency kernel isn't set up to reject unsigned modules, so it's a bit of a security hole, but also means that I don't have to figure out getting #DKMS to use the correct key. Right now I don't have a dependency on a DKMS-built driver. I used to use one for my dock and NVidia, but it seems that #NVidia doesn't need it now? and I don't use the video in my dock anymore because the #DisplayLink driver is annoying regardless of secure boot issues, because it doesn't work from power on. This isn't a problem with Linux, just DisplayLink in general.
I've been looking into hardening systemd and I'm curious why so many units are not hardened by default. Like why isn't it a standard practice to harden systemd units? Do the exposed parts shown in the output of `systemd analyze security <unit>` have that much of an impact on the overall security of the unit? I would think these exposure points would add up to a vulnerability that shouldn't be neglected.
https://blog.sergeantbiggs.net/posts/hardening-applications-with-systemd/
I think I'm nearly ready to release the first version of wpm - a process manager for #Windows inspired by my experiences using #systemd and #NixOS, written in #Rust
Setting up custom desktop environments and complex local development environments on Windows is about to get a whole lot easier - goodbye Docker!
@schenklradio @ffhl @videolan @tokudan entweder muss ich mir vll. noch was mit #systemd basteln, dass das nicht per #SAP (#RFC2974) announced wird, wenn gerade keine Sendung läuft. Weil so ist es außerhalb der Sendezeiten dann komplett still, wenn man's auswählt, was Leute verwirren könnte. Oder gibt es zufällig alternativ sonst noch einen Zwischeneinspieler? Hatte letzten Samstag zum ersten mal reingehört und das klang so, als ob es da am Ende was voraufgezeichnetes gab?
J'ai un problème avec systemd et mes recherches n'ont menées à rien. J'explique le problème, ce que j'ai fait et si vous avez une idée pour m'aider, n'hésitez pas !
Le problème :
Dans le status :
code=exited, status=226/NAMESPACE
Dans journalctl :
radicale.service: Failed to set up mount namespacing: /run/systemd/unit-root/var/cache/radicale: No such file or directory
radicale.service: Failed at step NAMESPACE spawning radicale: No such file or director
J'ai essayé de retirer une partie des options de sécurité de mon fichier unit :
# Optional security settings
PrivateTmp=true
ProtectSystem=strict
ProtectHome=true
PrivateDevices=true
ProtectKernelTunables=true
ProtectKernelModules=true
ProtectControlGroups=true
NoNewPrivileges=true
ReadWritePaths=/var/lib/radicale/ /var/cache/radicale/
J'ai essayé d'en passer une partie à false (j'ai vu ça sur Internet j'ai pas trop cherché à comprendre) et ça n'a pas fonctionné.
Si vous avez une idée pour m'aider, merci !
To download the latest web Remote Extension Host of #VSCodium
https://github.com/gbraad-devenv/fedora/issues/77#issuecomment-2781339941
and to run with #systemd:
https://github.com/gbraad-vscode/code-systemd/issues/2#issuecomment-2781345108
After this, you can open http://[ipaddress]:8000
Note: if you want to restrict this, you can modify the service to bind to only host 127.0.0.1 or something like your Tailscale IP address.
I just found this website about systemd: https://systemd-by-example.com/
It's a teaching website and playground to learn how to do dependency management with systemd units.
You can follow the instructions locally with a podman container or play with how the example systemd units interact on the website.
makadamia:~ # mount -o bind / /mnt
makadamia:~ # mount -o bind / /mnt
Broadcast message from systemd-journald@makadamia (Sat 2025-04-05 22:32:53 CEST):
#systemd[1]: Caught <ABRT>, from our own process.
Broadcast message from systemd-journald@makadamia (Sat 2025-04-05 22:32:53 CEST):
systemd[1]: Freezing execution.
@pid_eins I can imagine why it happens. But is it expected? Or should it be handled more gracefully?