In an effort to push even deeper into GNU/ #Linux Nerd-dom, I've been considering my options for going #Systemd-less so I can be just a little bit more snooty about my #OSS software choices. `/s`
- convert existing #Arch install to #Artix
- go real old school with #Slackware
- #void #voidlinux
As much as I have a soft spot for the nostalgia of slackware, I think pacman and the Arch repos have really spoiled me.
I may wind up on Artix, but I'm going to at least give void a try and see.
RE: https://mastodon.social/@muylinux/116325047906076086
sabia que #systemd no era fiable y tenia varios problemas, esto solo ayuda a cimentar mis resabios
Okay, I'm down to it here. I need to decide which non-#systemd OS I'm switching to (for the next five minutes, anyway; curse my DistroHopper nature...)
Here's the list I'm down to, although I am open to suggestions:
* **#Alpine**: PRO - Super fast and lightweight, I'm fairly familiar with the toolset; CON: Limited package set and support, no easy path for Noctalia/Dank install
* **#Artix**: PRO: Based on Arch with Arch tooling including Pacman and AUR access. Easy install. CON: Embracing #xLibre, which is run by a... fellow of questionable beliefs, let's be nice. Would have to do extra work to run Wayland instead with uncertain future
(continued, 1 of 2)
#SystemD Contributor Harassed Over Optional #AgeVerification Field, Suggests Installer-Level Disabling
#privacy #security #age
It's FOSS interviewed a software engineer whose long-running open source contributions include Python code for the Arch Linux installer and maintaining packages for NixOS. But "a recent change he made to systemd has pushed him into the spotlight" after he'd added the optional birthDate field for ...
[$] Objections to systemd age-attestation changes go overboard
In early March, Dylan M. Taylor submitted a pull request to add a field to store a user's birth date in systemd's JSON user records. This was done to allow applications to store th [...]
https://lwn.net/Articles/1064706/ #LWN #Linux #Debian #systemd
Running your own identity provider is all fun and games until you're debugging OIDC token flows at 2 AM.
If you want to deploy Keycloak 26 the right way - with proper network isolation, no plaintext passwords, and systemd-native declarative configs. I just published a new deep-dive.
We're ditching compose files and building a production-ready, daemonless stack using Podman Quadlets and systemd.
Read the full guide here: https://blog.hofstede.it/keycloak-26-on-podman-with-quadlets-identity-management-the-systemd-way/
#Linux #Podman #Keycloak #systemd #DevOps #Containers #SelfHosted #RHEL #Security
Tips buat yang jalanin #container terutama menggunakan #podman dan #quadlet.
Jika container tersebut butuh konsistensi data pada storage, misalnya database, biasanya untuk memberhentikan container itu butuh waktu yang cukup lama karena container tersebut akan melakukan "bersih-bersih" terlebih dahulu.
Timeout default saat stop container quadlet adalah 10 detik (yang merupakan default dari podman). Seringkali ini tidaklah cukup. Jika diabaikan, aplikasi di container tersebut akan diterminasi paksa sebelum dia selesai bersih-bersih.
Agar tidak di-terminasi paksa jika lebih dari 10 detik, naikkan batas timeout misal menjadi 120 detik (atau nilai lain yang dianggap wajar).
Caranya cukup edit file quadlet anda:
SP⟁CED GO⟁T
Ivan
Jiri Grönroos
Springy Owlbear 
PrivacyDigest
LWN.net
openSUSE Linux
Larvitz 
Sek Sakeudeung[Container]
...
...
...
PodmanArgs=--stop-timeout 120
...
...
...
[Service]
...
...
...
TimeoutStopSec=120
...
...
...
Catatan, [Container] dan [Service] tidak perlu anda ketik ulang, karena biasanya sudah ada di file quadlet tersebut. "..." juga hanya sebagai ilustrasi saja, tidak perlu anda ketik ulang.
Mengapa butuh dua baris tersebut, di [Container] dan juga [Service]?
Pada section --stop-timeout= pada section [Container] itu fungsinya untuk memberitahu podman bahwa batas timeout adalah 120 detik.
Sedangkan TimeoutStopSec= pada section [Service] untuk menaikkan batas timeout #systemd ketika menunggu podman selesai melakukan terminasi.
Jadi pada [Container] itu untuk timeout podman menunggu aplikasi selesai. Sedangkan pada [Service] untuk timeout systemd menunggu podman selesai.
I'm trying to figure out how I can support the scripting options of wg-quick within my tool. Obviously we can't do it within .network and .netdev files, but I also don't see a way to trigger a unit based on a specific interface becoming routable and configured. Perhaps there is a way using systemd.path magic?
ಚಿರಾಗ್ 🌹✊🏾Ⓥ🌱🇵🇸 (he/him)