nobody cares

>>108640297
fpbp
/thread

>>108640280
thanks, but I'll stick to windows 7

>>108640297
>>108640313
Yes, low IQ troglodytes dont care.

>>108640280
>[ *** ] A stop job is running for User Manager for UID 1000 (34s / 2min 2s)
kek, that's exactly the reason that made me switch from shitstemd 7 years ago
>inb4 it's fixed now
yeah I don't care, too late

>>108640280
The duty imparted on the linux community is to build a viable alternative to systemD that doesn't ask all of it's users to be sysadmins.

>>108640280
>>108640558
And everybody had the past 10 years to do it instead of building protest distros and complaining.

File: 1557817321509.png (226 KB, 440x495)

226 KB PNG

>>108640457
>>108640280

>>108640558
Who but sysadmins are dealing with the init?

>>108640580
Everyone who has to set up software and services on their machine.

>>108640580
>Void and artix are gonna get mass adoption, bro. It your moral responsibilty to install gentoo as if that will fix the software ecosystem that everyone neglected, bro.

>>108640558
>sudo pacman -S packagename-initname (on Artix)
Yeah, certainly a sysadmin thing

>>108640558
There's nothing stopping you from putting

service &

in your .profile like you used to. The rest of will enjoy having an actual service manager.

#define _XOPEN_SOURCE 700
#include <signal.h>
#include <unistd.h>

int main()
{
    sigset_t set;
    int status;

    if (getpid() != 1) return 1;

    sigfillset(&set);
    sigprocmask(SIG_BLOCK, &set, 0);

    if (fork()) for (;;) wait(&status);

    sigprocmask(SIG_UNBLOCK, &set, 0);

    setsid();
    setpgid(0, 0);
    return execve("/etc/rc", (char *[]){ "rc", 0 }, (char *[]){ 0 });
}

you don't need more

>>108640608
Don't need to. Alt inits won't stop the government from making your computer legally unable to access the internet.

my only experience with other innits was runit, its very archaic but simple and correct. honestly, no reason to use it over systemd.

File: Screenshot at 2026-04-10 (...).png (984 KB, 1366x768)

984 KB PNG

>>108640280
systemd is so stupid, all you ever need is sysvinit and a separate user services manager, it is such a simple thing even an idiot like me was able to build one.

File: Screenshot at 2026-04-10 (...).png (240 KB, 1366x768)

240 KB PNG

>>108640742
also worth mentioning, i built the dumb thing in posix shell and even hand wrote the manpage, i just can't grasp why this thing that feels like it should have been written 15 or even 20 years ago was not written and instead we got systemdicks, is really the foss space so shittastic?

ill use bsd if i want x (and i do)

>>108640585
Like???? If you are aware that services even exist and want to edit/create them you should also be capable of at least reading https://wiki.gentoo.org/wiki/OpenRC_to_systemd_Cheatsheet
But that's such a faggot line or argument, almost no one has to fuck with services on a daily basis if they're not sysadmins.
You really want to argue that lil ecker timmy playing csgo needs to deal with services? lmao
>>108640594
Who exactly are you quoting?

>>108640742
pretty sure you can have debian with sysv, isnt it slow and the reason systemd come along?

>>108640916
I call that line of argument the "Schrodinger's Retard", which is a magical person (often made out of straw) who's smart enough to be aware of and trying to deal with extremely nitpicky subjects but somehow is stupid enough to the point of being incapable of doing a simple web search or reading a manual.

>>108640929
1) yes you can have debian with sysvinit, but is a second class citizen experience, some packages will uninstall sysvinit and give you systemd, devuan is a first class citizen experience with sysvinit and nothing will ever install systemd as those packages are not in the repos.
2) fun enough nowadays sysvinit is faster than systemd if you set parallelization, just use insserv and set CONCURRENCY=makefile in /etc/default/rcS

>>108640916
>lil ecker timmy playing csgo
How do you convince them to choose one?

>>108640742
>awesome (X11)
But Wayland is the future.

>>108641123
the closest thing to awesome in wayland (some) is very experimental and yet not a 1:1 thing, not to mention that i know what every x11 problem is and how to solve it since back in 2012, i'd rather not learn new problems and spend years figuring out how to solve them, thanks.

>>108641145
I was being sarcastic, since you didn't notice.
>the closest thing to awesome in wayland[...]
Is mangowc, though yes, I will say it is very much still in development.

But artix won't have this https://pixelcluster.github.io/VRAM-Mgmt-fixed/

>>108640280
> [ *** ] A stop job is running for User Manager for UID 1000 (34s / 2min 2s)
You can get past that when you're rebooting by mashing ctrl+alt+delete really fast. You have to do it seven times in under two seconds though.

>>108640280
i don't use linux for freedom or politics or..
It literally works better for what i do than windows.
If windows was better for me than linux, i would use it instead, but it is not so i don't.

>>108641186
>ctrl+alt+delete
good to know. although this is rare, it happened to me the other day due to a defective flash drive, the moment i removed the drive it shutdown immediately. so yeah its working great.

>>108641232
It's not that useful on shutdown because it triggers a reboot, though I suppose it could allow you to shutdown once you're rebooted.

File: 1762021927657853.png (44 KB, 320x326)

44 KB PNG

>>108641184
>extra performance locked behind init system

File: 13624762472466.jpg (37 KB, 512x497)

37 KB JPG

>>108640579

>>108640614
okay, I'll bite. why is

sigprocmask

not inside of the fork?

systemd has a LOT of issues, but if you're having stop job issues more than once in a blue moon, you either fucked up your configurations or are using a distro maintained by retards who fucked up their configurations.

I also have yet to find something does systemd's core functionality as well as it does. I don't want a slightly better sysvinit.

>>108642316
kek, lmao even, here's the ackchully code https://gitlab.steamos.cloud/holo/dmemcg-booster/-/blob/main/src/main.rs?ref_type=heads
it ain't even 500 lines, the "hard systemd dependencies" are just dbus interfaces, actual work seems to be done mostly with "/sys/fs/cgroup", my guess is by the time the kernel patches get upstreamed someone will have figured out a python script to do the work of the dmemcg-booster in a way that works with sysvinit, doesn't seem so hard desu, simply not having a kernel with the capabilities rn for install is a hindrace to developing a non systemdicks dependant solution.

>>108642640
you're critiquing a linux init system to jobless losers on 4chan. anyone who has actually modified init scripts know that they're all shit.

>>108640280
>"systemd is le bad because some youtuber said so"

>>108642677
debian's init-d-script library fixes all initscripts criticisms, it just works.

>>108640280
Systemd makes the system boot faster.

>>108642677
>>108642640
OpenRC is nice. It does a lot of the same things Systemd does.
>Declarative init script format (but still support for rc scripts too)
>Hotplugging
>User services
>CGroups
>etc

Having said that though, Systemd has gotten to a point where you can largely ignore it now. There are still some parts that are a pile of garbage but they're not showstoppers.

>>108642754
can it still boot if a service fails?

>>108642754
> Systemd has gotten to a point where you can largely ignore it now
Good. Let’s get rid of it then.

File: 0feae77d8b6e087159a271513(...).jpg (117 KB, 421x480)

117 KB JPG

>>108640280
I mostly liked Artix when I tried it in various VMs (aside from that one time where Calamares completely shat itself, but that might be my fault for picking BTRFS) and it has some of the least retarded maintainers of any distro at the moment, but I can't bring myself to try it again purely because pacman sucks ass and I fucking hate the fact that updooter distros like Arch are gradually usurping everything else. If your OS doesn't reliably let me install local, offline software like Windows XP could 25 years ago, upgrade or downgrade those programs at will, or even install updates without having to run pacman -Syu every 5 seconds, then it's objectively useless dogshit that may as well be bricked if anything catastrophic happens to your internet connection or the project's repos/server infrastructure, systemd or no systemd
Debian is the only distro I'm aware of that comes even close to meeting any of these requirements, but it's dead, so that's cool I guess

>>108642825
Yes. It continues after a timeout like Systemd.

why did they suddenly introduce this? i dont remember it happening until a few weeks/months ago

why can't i just "reboot now" and have it work? why do I need to run some other program?

>>108641184
>But artix won't have this https://pixelcluster.github.io/VRAM-Mgmt-fixed/
>>108642316
>>108642671
Already have: https://codeberg.org/dsalt/gamevram
What else?

>>108640280
>don’t like systemd
>i can’t run mullvad without systemd
what do

>>108644614
Why wouldnt you be able to run it withour systemd?

>>108640280
Is it possible to migrate from cachyos to artix? Or better to just reinstall?

>>108644622
there's an Arch -> Artix migration guide on the Artix wiki, it'd probably work for Cachy too

>>108644614
There are mullvad service scripts for all Artix inits on the AUR.

>>108642731
my system boots (and shutsdown) way faster with dinit.

>>108642830
>let's get rid of this battle tested, stable code
No.

>>108640280
I'm considering migrating from to Arch to Artix. Could I expect significantly better performance with openrc, dinit, or runit? I've heard systemd is a resource hog.

>>108642900
>it's dead
its only getting better.
>>108644815
how long? i havent counted my time, but it boot in few seconds. shut down is instantaneous as well.

>>108640457
I'm diagnosed with autism and even i think you need to go touch some grass.

>>108644870
You heard wrong, I don't know why retarded nocoders here think that less code equals more performance but that's just simply not true at all, you get more performance by choosing the correct algorithm for the problem, utilizing cache correctly, running your shit in parallel, etc etc.

>>108640280
Not a hueg systemd hater, but the vast majority of my Linux experience is with Gentoo, so I'm orders of magnitude more familiar with openrc.

Gemmy 31B writing code
>introspective
>self correcting
>a lot of induction
Qwen 3.6 MoE writing code
>deterioration
>selects between 3 wrong solutions when you try to correct it and doesn't use the feedback
>vision encoder is ass

So obviously benchmaxxed.

I keep trying Artix but it feels janky I used the premade image and community versions because I don't know how to do the base version so maybe that's it.

The xlibre version was very janky the Wayland one was fine x11 default runs like ass on Nvidia so I don't use it.

Only other non systemd distro I have used is void and I honestly liked it a lot but again I don't know how to set it up from base on my Nvidia machine but in my AMD laptop it's perfect.

For the time being I'm just sticking with Arch.

>>108644382
>pyshit
Grim

>>108640280
>A stop job is running ... (34s / 2min 2s)
absolute state of shitstemd

Imagine thinking that convincing a bunch of teenagers to install your heckin' rebel OS so they can watch porn is going to help the legal battle that we all have to win to be private on computers. This is just round one, I hope you people realize that this ends in your firmware and you'll feed them a win if you're going to shill stuff to the rebel teens that will make their parents hate you.

>>108647424
by the time we have LGBT flags in our firmware, china will have developed risc-v far enough to make it usable

>>108640280
it is too difficult to ditch NixOS

>>108640280
>i want to be free of corporate interests
>I know! I'll use the linux kernel!
You already lost, kiddo.

>>108648249
saw something similar on the devuan forums, i was amazed that they really parrot this bullshit while leeching on everybody else.

>>108645987
better than the rustroon slop that hards depends on systemdicks, doesn't seem too complex that it cannot be rewritten in bash or even posix shell tho

don't care, i use systemd in gentoo because it works

File: 1598870436232.jpg (94 KB, 1440x1080)

94 KB JPG

>>108649905
>don't care, i use systemd in gentoo because it works

>>108642640
systemd brings the ethernet interface down on EVERY shutdown for 0 fucking gain

and when you're running from NFS root (I do), this gives you a hard freeze on every reboot/shutdown

It's such a basic fucking concept, you don't even have to check for it, just DON'T bring the ethernet interface down ffs, why even

sexless troons, that's why

>>108650421
>it brings down the ethernet for no reason
you can override this
>it freezes because it brings down ethernet before unmounting an nfs share
so configure the nfs mount to depend on the network unit

systemd isn't magic, if you don't configure your system properly it won't work properly

>>108640487
It's not lol, I get it everytime I shutdown/reboot.

>Come home.
>artix
try making easyeffects and a bunch of other stuff run on this piece of shit distro

>>108640929
>isnt it slow
no, systemd thrashing your hard drive and ssd is what's slow

>>108640280
>[ *** ] A stop job is running for User Manager for UID 1000 (34s / 2min 2s)
Absolute retard here who recently switched to Linux, what exactly is this?
I get it ~30% of the time...

>>108640297
/thread
>>108640457
imagine dying on this dumbass hill

>>108640280
That feeling when I've been using Fartix longer than all of you and before Luke Smith pushed an entire fleet towards the distro. Let me just do this gay ass captcha and release this brappy post all over you /g/tards.

>>108650682
it means a program you were running under your user account (UID 1000 is nearly always the first real/human user) isn't exiting properly when you go to shutdown/reboot.
if you get this often, you should find out what program is the issue and fix it. if you don't want systemd to wait as long before trying harder (SIGKILL rather than SIGTERM) or giving up and continuing anyway, you can configure the number of seconds it waits. while i prefer to fix the issue properly, if this does happen i have it set to wait only 20 seconds, as if something doesn't quit in under 20 seconds (only my fast cpu/ssd) then it's probably not going to

File: 1766310020496794.jpg (38 KB, 730x684)

38 KB JPG

>>108651178
Cheers Anon! For now I lowered my DefaultTimeoutStopSec to 10, but I will check my logs the next time this comes up to see whats being naughty.

>>108651404
np, for anyone reading try to avoid being tempted to set it too low, especially on slower hardware. there's a balance to be had between faster and safer, like from fastest/least safe to slowest/most safe;
- press reset button on the computer case
- alt+sysrq+rb (instant soft reboot)
- alt+sysrq+rsub (sync and unmount volumes then soft reboot, but don't even attempt to close any programs cleanly first)
(the above are for humour purposes, don't use them)
- alt+sysrq+reisub (tell programs to exit, then sync/unmount and soft reboot. this is the safest way to reboot if for whatever reason you are unable to reboot normally)
- hold ctl+alt+del until system does an emergency reboot
- reboot normally (switch to runlevel 6)

>>108651443
>hold ctl+alt+del until system
systemd* i meant to write, may need to do it from a non-graphical tty

alt+sysrq+reisub is a good one to know if you ever find yourself in a situation where you would otherwise have to hit the reset button, since outside of kernel panics it should always work and it way safer than hitting reset

how about you abandon linux and come to bsd already

File: 653995_700b-1498702847.jpg (56 KB, 569x412)

56 KB JPG

>>108651462

>>108651462
Bsd is a lot fussier about your hardware, if it wasn't I'd be running it on everything.

>>108651462
I've moved to Haiku. I don't regret it.

>>108640280
Artix is trash. Void is what the straight white man calls home

>>108640280
I tried to install gentoo on my laptop because I like the logo and features but I failed to properly install the kernel twice and I gave up>>108640280