In reference to: https://lemmy.world/post/23862757
I use Void btw
Image text:
Most people rejected his message.
“Systemd is Satan’s creation! Pure Evil!”
They hated Talking Pig because He told them the truth.
system d struggle sessions should be integrated into SystemD
Install what you want we’re in the land of the free (and open source software) here
Upvoted because it belongs in this community, and should not be silenced, even though it is the wrong opinion
I don’t really get the hate for systemd. At least for someone who started really using Linux after it was introduced, it always seemed easier to control and manage than the init.d stuff.
Obviously it’s a hassle to migrate if you have a ton of legacy services, but it’s pretty nice.
It’s because you now need to do
systemctl restart sshdinstead of/etc/init.d/sshd restart, I see no other reason than having to learn new syntax.Arguably, init.d scripts were easier to understand, and systemd is a bit of a black box, it somehow works, but who knows where it writes logs or saves the process pid (it’s all in the documentation somewhere), with init.d script you can just open the script itself and look.
Don’t minimize those strengths. Init.d scripts are something you can figure out just knowing a bit of shell script, or historical knowledge from before there was an internet. For something I rarely use, why do I need to learn something more complex to do the same thing - I either haven’t been sold on all the new functionality they piled in or do not need it. After all these years crowing about the Unix/linux way being many independent flexible tools that can work together, why do we now have this all-in-one monstrosity that might as well have come directly from Microsoft?
I prefer the ini files systemd uses to bash scripts
Systemd syntax is not that hard if you read the manual. I think every hardcode Linux user hates systemd because it automatically does the thinkering for you and you can control your processes with simple commands
Yeah, the manpages for systemd are large but also informative. Most of us only use a small subset of the features—much like we never explored everything possible with separate init programs.
Having used Linux on the desktop for some two decades and worked as a Linux sysadmin for a good while I don’t miss the init scripts. My impression is more that a certain cohort wants to pretend that service management is easy by ignoring large amounts of it. It’s easy to write a bad init script that breaks when you really need it, or be out of your depth with more complex cases.
Not to mention the whole conformity by convention thing. Systemd unit files are descriptive and predictable by their nature. So-called init scripts didn’t really have to be scripts, they just usually were, and their arguments and output and behaviour was also unenforced—there’s nothing really stopping you from writing a compiled program that self-daemonizes and place the binary with the init scripts rather than in /bin. Ultimately people who make programs also have to be good at writing init programs with that setup.
So we’d have people doing dumb shit themselves and getting angry at others doing dumb shit. PHP was also pretty popular and full of dumb shit. Lots of “worse is better” to go around.
Ultimately it’s more of the stuff covered in Bryan Cantrill’s Platform as a reflection of values. Some of us value predictability and correctness, others feel it’s a straitjacket. There’s no way of pleasing everyone with the same platform.
And currently the people who want to distribute their own riced-out init programs in bash, perl, php, node.js and so on are SOL. (They can still use them on their own machines.)
The more I actually learn about systemd, the cooler it is.
For example, you can automatically mount SMB shares using systemd, with automatic remount on failures. It’s far superior to using fstab once you know how.
https://anteru.net/blog/2019/automatic-mounts-using-systemd/
I think it’s okay to not 100% know every little detail of how a system works, as long as it’s possible to find out what you need when you need it.
This post was sponsord by the Backdoor Buddies
I have the following complaints about systemd:
- It was created basically by lennart because after RHEL 6 did pretty much the worst implementation ever of upstart he got NIH syndrome about it
- Red Hat played a lot of dirty politics early on to get systemd everywhere (my tinfoil hat theory is that Red Hat let Lennart’s NIH syndrome run away with it because they thought having more control over the init system would be beneficial)
- It’s subsuming everything, often with no real benefit over what it replaces.
The first two aren’t actually issues with systemd, but rather are political issues I have around the way Red Hat bullies the rest of the Linux ecosystem. I’m not going to let that become a stopping point for my using what is actually a fairly good piece of tech. The third is actually an ongoing issue, but it’s not enough for me to try throwing the baby out with the bathwater. It is, however, IMO a continuation of Red Hat’s sketchy political play.
What kind of dirty politics are we talking about here? I remember when Arch switched, the stated reasons from the devs was that their old init system was bad and nobody wanted to maintain it, for example.
There were several cases of shenanigans from other Red Hat controlled projects yanking upstart configs and sysvinit scripts from their projects and replacing them exclusively with systemd units even though those configs had active maintainers (often people who worked at Canonical or Google). This made packaging those supposedly community owned but de facto Red Hat controlled projects more difficult for any system that didn’t use systemd, since the packagers had to scramble to find or recreate those files and then maintain patch series for them. They also very quickly jumped on adding systemd-specific integrations where similar integrations to make the services work better with upstart had been rejected because services weren’t supposed to favour an init system.
Something not necessarily (or provably) from Red Hat - a whole lot of misinformation about upstart suddenly started appearing on mailing lists and message boards when Debian was considering whether to use upstart or systemd. While I think they made the right decision to go with systemd, that sudden influx of new accounts complaining about upstart likely influenced the decision in ways I’m really not comfortable with.
I don’t dislike systemd. I’m happy to use it and think it works quite well for many (though definitely not all) of the things it does. But I am concerned about how it’s yet another case of Red Hat having a large amount of control over the Linux ecosystem and Red Hat controlled projects and the supporters of Red Hat projects using dirty tricks to further that control. And with systemd consuming more and more of how a Linux system works, I am concerned about the influence that gives Red Hat. Are we going to see
systemd-packagedthat manages your packages, but somehow the patches to make it work with non-RPM packages keep getting rejected or just held up for years at a time? (We’ve already seen similar things with xdg portals, where portals Red Hat wants get approved and merged very quickly, but portals proposed by Canonical or SuSE spend years “in review” with more and more petty changes requested, sometimes to be rejected because a Red Hat backed portal that only implements part of the functionality suddenly appeared and was approved within a week or two.)
In practice, what makes it so bad?
It’s new and different, and the Linux boomers who are still stuck on ALSA and ext2 hate it.
what is it? systemD is new?
When you entered the scene before epoch 0 I guess it is.
Everything else aside, my biggest gripes are with service control. Instead of just “service” they had to invent a new name that was super close to an existing function (systemctl vs sysctl) and reverse the switch order. (service sshd stop vs systemctl stop sshd.service)
Besides that, I absolutely hate that all the service configs are not in a standard location. Well, you get things like sshd.conf which are still in etc, but the systemctl configs are who knows where.
There are more important things to hate on with systemd, but I went for the superficial this time and I absolutely hate service management with systemd now.
wrt conf file location, they’re only generally in /usr/lib/systemd, /etc/systemd, or /run/systemd. You can always find out what’s getting read with
systemctl cat <service-name>. Way easier to find stuff than with some other random programs imo, I’ve seen crap have default conf files in dumb places like/usr/share/<service-name>/lib/etc.
Bad usability, binary logs, crummy architecture.
I still can’t wrap my head around why SystemD has become the defacto standard & why aren’t devs trying out OTHER init-systems
- It was doing new things.
- It was easier to learn.
- The other init systems were (are) stagnant.
Imagine trying to get new, young developers to adopt C or Pascal when the likes of Rust and Python exist. You can make arguments for a thing’s superiority based on moral standards (which are always subjective), but morality is a poor metric. If everything was done based on that, the Linux ecosystem would be in the same state as the GNU Hurd kernel.
It’s pretty good at starting services. It just keeps adding bundled things people wouldn’t use otherwise, in a fairly microsoft fashion
Is systemD open source ?
There’s also no uppercase d in
systemd, the word is entirely lowercase (but I’ll still write it with an uppercase s at the start of sentences).
The choice of init system is up to the distro maintainers because init scripts are usually created and maintained by the packager of a given application. Debian for example chose its init system via a democratic vote. Distros that focus on different init systems exist, like the Debian fork Devuan.
There is nothing wrong about it, also it’s easy to use which is a plus to me
Hueheue i use artix btw
Now we just need to find a way to integrate systemd into wayland and watch people lose their mind.
It sort of is, the whole elogind thing.
You don’t have to use systemd. However, the rest of the world left you behind. Systemd isn’t controversial since everyone has adopted it. No one is making you use it but keep in mind you are a very small minority. The rest if the community moved on after systemd was release 10 years ago.
This is fine for the memes but outside of that it is silly.
Windows isn’t controversial since everyone has adopted it. No one is making you use it but keep in mind you are a very small minority.
Windows has about 80% market share (decreasing) in a very specific and shrinking niche (desktop PC’s).
All other computing devices used by most people daily on the client and server side are dominated by some form of Linux or BSD.Hexbear user spotted (or at least that’s what my first impression is with the weird image)
Windows isn’t controversial since everyone uses it. That’s a true fact.
Hexbear user spotted (or at least that’s what my first impression is with the weird image)
Heck no, that’s just an ancient meme to indicate it’s just banter/harmless trolling, not an attempt at serious discourse.
I guess we’re old now, if folks don’t know trollface anymore.
The more you know I guess. I wasn’t on the internet in 2008.
What’s weird about the image? It shows up as a trollface to me. Are you like, 14?
I’ve only seen that image used by Hexbear users
If I was 14 I would’ve been born in 2010. I’m young but not that young. (I’m in my 20’s)
Trollface was probably the most popular meme of the last decade, you must have been living under a rock
I wasn’t on any social media prior to about 2 or 3 years ago
You’re joking, it was on Twitter a decade ago
They’re usually a little more creative, like this:
Poettering worked for Red Hat from 2008 to 2022.[2][3] He then joined Microsoft.
In 2017, Poettering received the Pwnie Award for Lamest Vendor Response to vulnerabilities reported in systemd.
This Mastodon stream from Lennart Poettering describes a sudo replacement — called run0 — that will be part of the upcoming systemd 256 release. It takes a rather different approach to the execution of privileged commands, avoiding the use of setuid (which he calls “SUID”) permissions entirely.
Basically Microsoft bloat confirmed, everyone switch back to OpenRC lol
Hard disagree, I have been using Linux for over two decades and find sysd superior to sysv
I’m about 2 decades in too, really not here to argue since everything has already been said multiple times. I do see systemd in a somewhat similar light as Pulseaudio. Yes, some good ideas there and it’s a useful tool, but it wasn’t the be-all end-all solution.
Syswire when?
Iirc we can use emoji now, so Sys🎶 is a valid option.
Maybe, but SysV Init was just fucking awful. It’s a shame.
It’s not evil. It’s merely
- the wrong tool
- built wrong
- on wrong principles
- by a bad team
- who has poor coding and interaction
and now RedHat’s wunderkinder has moved onto Microsoft where he’s a better fit. Ideally, we can go back to Linux again.
Simple.
As someone who ran security for an enterprise OS company, I can’t see why there’s any debate on this. Are we used to choosing comfy things despite the safety concerns, now, or just when Lennart shits them out?
Are we used to choosing comfy things despite […] concerns
People have been choosing convenience above everything else for a while, personally I find that doing so is even glorified at times.
I’ve never really had issues with systemd, but I must say when I was setting up void I did really enjoy the runit init system 🤷♂️
