SystemD is blamed for long boot times and being heavy and bloated on resources. I tried OpenRC and Runit on real hardware (Ryzen 5000-series laptop) for week each and saw only 1 second faster boot time.
I’m old enough to remember plymouth.service (graphical image) being the most slowest service on boot in Ubuntu 16.04 and 18.04. But I don’t see that as an issue anymore. I don’t have a graphical systemD boot on my Arch but I installed Fedora Sericea and it actually boots faster than my Arch despite the plymouth (or whatever they call it nowadays).
My 2 questions:
- Is the current SystemD rant derived from years ago (while they’ve improved a lot)?
- Should Linux community rant about bigger problems such as Wayland related things not ready for current needs of normies?
Nah, it’s fine. Boot times are considerably faster than sys.v in most cases, and it has a huge amount of functionality. Most people I work with have adopted it and much prefer it to the old init.d and sys.v systems.
People’s problem with systemd (and there are fewer people strongly against it than before) seem to break down into two groups:
-
They were happy with sys.v and didn’t like change. Some were unhappy with how distros adopted it. (The debian wars in particular were really quite vicious)
-
It does too much. systemd is modular, but even so does break one of the core linux tenets - “do one thing well”. Despite the modularity, it’s easy to see it as monolithic.
But regardless of feelings, systemd has achieved what it set out to do and is the defacto choice for the vast majority of distros, and they adopted it because it’s better. Nobody really cares if a user tries to make a point by not using it any more, they’re just isolating themselves. The battle was fought and systemd won it.
One of my biggest problems with critics of systemd is that a lot of the same people who make that second point also argue against wayland adoption when xorg does the exact same thing as systemd. It makes me feel like they’re just grumpy stubborn old Linux nerds from the 90s who just hate anything that’s not what they learned Linux with.
Which is sad, because honestly I think it’s kind of not great that an unnecessarily massive project has gained such an overwhelming share of users when the vast majority of those users don’t need or use most of what it does. Yeah, the init systems from before systemd sucked, but modern alternatives like runit or openrc work really well. Unfortunately they get poorly supported because everyone just assumes you have systemd. I don’t like the lack of diversity. I think it’s a problem that any init system “won”.
Unfortunately they get poorly supported because everyone just assumes you have systemd.
No, they get poorly supported because they were a pain to support even before systemd ever showed up. I for one was extremely tired of writing the same shit over and over again in every init script and then going through the tedious process of porting the script to every platform for minor idiosyncrasies of the various distros (start-stop-daemon available or not was one I remember, the general bash/GNU vs. BSD stuff you get with any script was another) from 10 year old RHEL to modern ones.
Maybe
systemdgets grouped withwaylandandxorgwith other init systems simply because of usability?I mean, I got used to the thought that what I prefer is less usable, because some pretentious UX designers say so, and we Unix nerds use inconvenient things because we are all perverts.
But when I read about industrial design and ergonomics, it seems that my preferences are consistent with what I read, and all those UX designers and managers should just be fired for incompetence and malice.
Back to wayland/xorg and runit/systemd (for example), same reason FreeBSD may seem easier to set up and use than an “advanced” Linux distribution - there’s less confusion.
Xorg, or X11, “used to” do the “minimum necessary” for a remote display system… in the 80s. Graphics tech has changed A LOT in the last 40 years, with most of the stuff getting offloaded to GPUs, so the whole X11 protocol became more and more bloated as it kept getting new and optional features without dropping backwards compatibility.
The point against Wayland, was dropping support for remote displays, while kind of having an existential crysis for several years during which it didn’t know what it wanted to become. Hopefully that’s clear now.
OpenRC and runit are indeed working alternatives, but OpenRC is kind of a hack over init.rd, while runit relies a bit too much on storing all its status in the filesystem. Systemd has a cleaner approach and a more flexible service configuration.
I like it too. Very easy to work with and set up services as needed.
“do one thing well”
Arguably, Systemd does exactly that: orchestrate the parallel starting of services, and do it well.
The problem with init.d and sys.v is they were not designed for multi-core systems where multiple services can start at once, and had no concept of which service depended on which, other than a lineal “this before that”. Over the years, they got extended with very dirty hacks and tons of support functions that were not consistent between distributions, and still barely functional.
Systemd cleaned all of that up, added parallel starting taking into account service dependencies, which meant adding an enhanced journaling system to pull status responses from multiple services at once, same for pulling device updates, and security and isolation configs.
It’s really the minimum that can be done (well) for a parallel start system.
I just hate the syntax, systemctl start apache2 feels like dumb manager speak over service apache2 start.
But other then that I love how systemd has been for me.
Y’know, I felt that way to begin with and it certainly took a long time for my fingers to adjust, but I’ve grown to adjust to that.
And it’s better - you can do: “systemctl restart Service1 Service2 Service3” Before, with “system Service1 restart” you could only action on service at a time.
Plus, it’s linux, so you can set up aliases to change the order into anything you like, even carry on using the old muscle memory formats. (Although I don’t encourage this if you intend working on multiple servers!)
How so? I like the systemctl syntax more, since it allows for starting/stopping many units at once. It also supports a much richer set of commands than service ever did.
it just feels like a manager decided the command should read like english, made the decision then went back to never entering a command again in the terminal again. every day, i get to decide, should i enter “systemctl restart problem_service” all again or hit up on the keyboard and and hold back, then rewrite over the previous status command. bit less work if the status/stop/start/restart bit was on the end like it used to be.
In BASH ALT+T will swap the last white spaced separated strings… It’s still annoying but makes “systemctl problem_server start/status/restart” a bit easier. CTRL+W will clear the current string to whitespace, so up arrow, ALT+T, CTRL+W, status, ALT+T, Enter.
The bit was on the end because it was an argument to the script specific to that program. Instead, the control is now at the start because it is an argument to systemctl itself. This removes the ability to define custom controls, but enables you to control many things at once.
Yeah,
command subcommand args.... Theserviceformat makes more sense when you’re seeing it as “run this script to control this service”. Thesystemctlformat makes more sense as a frontend subcommand to control systemd itself.
Thanks a lot. I truly hope this is the big picture and SystemD whiners are just a fringe minority lol
systemd is modular, but even so does break one of the core linux tenets - “do one thing well”.
Linux itself (i.e. the kernel) breaks the hell out of that so-called core tenet. Have you looked at
make menuconfigat any point? There’s everything but the kitchen sink in there.Apples to oranges, and you can have a minimal kernel tailored to your needs.
Is there somewhere I can read about the Debian wars? I am curious about that 🤓
Prepare for a rabbit hole… but this ought to get you started…
They’re the main reasons, yes, but there are plenty others .
-
systemd is a godsend when you need service control while getting actual work done, at scale.
there are legitimate things to criticize but in general the rants are incompetent preaching to the uninformed.
systemd is a godsend when you need service control while getting actual work done, at scale.
there are legitimate things to criticize but in general the rants are incompetent preaching to the uninformed.
Service control was systemd’s main benefit and what it most excelled at. Having shell scripts for everything was a legitimate pain. It was all the other pieces of the ecosystem that it was wanting to subsume that got people upset (logging, cron, time, hostname, login, etc). Journald/binary logs was the main sticking point that I recall, though I figured it was a trade-off that was worth it, especially since you could have journald keep dumping to text anyway.
Ok, so I have a very unique background in systemd. I worked at Red Hat supporting it basically as the primary support and I’ve worked with the developers of systemd at Red Hat directly. I no longer work there.
So first off, it’s “systemd” all lower case. I don’t care, but for some reason Lennart Pottering (creator) does.
systemd was a MASSIVE change. And Red Hat did a TERRIBLE job relaying it. To the point where I’m still trying to get my company to understand that it can NOT be treated like the old init systems. You can NOT just drop an init script in place and walk away and hope it works. Because a LOT of times it doesn’t. Due to forks, switch users, etc.
systemd is NOT an init system. RHEL 5 and older had sysvinit as it’s init systemd. RHEL 6 had UpStart as it’s init system and looked exactly like sysvinit that no one even noticed. systemd again is NOT an init system. Init system is 1 part of systemd. systemd does a lot of cool things. It bundles applications together, it manages those applications and can restart them or kill children, it can do resource constraints, it separates out users from the system, and lots more.
Because it is not an init system there is a LOT LOT LOT of bad recommendations out on the internet where someone has X problem and person suggests Y and IT WORKS! … except it doesn’t REALLY work as far as systemd is concerned and you’ll hit other issues or your application takes longer to start or stop and people just blame systemd.
It is systemd’s fault that it has done an ATROCIOUS job of helping people adapt. It’s a great example of RTFM. systemd’s man pages are INCREDIBLE and extensive, but when you drop so much knowledge it becomes more difficult to find what you want/need. systemd.index and systemd.directives are your best bet.
So systemd does a lot of amazing things that sysvinit never attempted to do. It’s never attempted to explain anything it expects everyone just learn magically. it’s INCREDIBLY complex, but once you understand it’s basics you can more easily get an application running, but as soon as there’s a problem it’ll just break your brain.
To give you an example, sshd’s old init script is like 250 lines of bash. systemd’s unit file comparative is like 12. Because systemd handles a LOT of what you manually had to handle before. BUT to get to that 12 you literally have to learn EVERYTHING new.
There is no “is it good or bad” here really imo. It’s a completely different fundamental design. Red Hat made it for themselves. Other distros picked it up. It can be argued that lots of folks followed Debian and Debian had a few Red Hat board members that were pushing it. Whether they pushed it of their own accord or because they were with Red Hat I don’t have a clue.
What I can say is at my current company they’re suffering from a LOT of systemd issues and they don’t even realize it. I’ve been working with Red Hat to try to get Insights to alert people to the failures and we’re making progress.
To see if you have issues just to start run the two following commands:
# systemctl list-units --failed # systemd-cglsIf you have any units that are failed, investigate those. If you don’t need them, disable them. As for the systemd-cgls this shows HOW systemd is grouping things. ANY application that runs as a service (or daemon or application or runs in the background or however you wanna say it) should be under system.slice. ONLY humans logging into the system (meat bags NOT applications switching to users) should be in user.slice. A LOT of times what happens is an old init script is dropped in place, they start it, it has a switch user and systemd assumes it’s a user and puts it into user.slice. systemd does NOT treat anything in user.slice the same as in system.slice and this WILL eventually cause problems.
So again, is it good or bad? Eh. It does a lot of cool things, but they did a MASSIVE disservice to ALL of us by just expecting to relearn absolutely EVERYTHING.
sshd’s init script under OpenRC is 87 lines, of which around half are blanks, comments, closing braces, and other boilerplate. Granted, that still makes the real code maybe three times the size of your systemd unit file, but the difference isn’t as impressive as you’re making out.
95% of people shouldn’t need to poke around in their init scripts or unit files anyway. If you actually need to do that, your use case is already somewhat unusual.
As an end user, unless you’re running a server, then no you shouldn’t have to mess with any of it.
If you’re running a server or a sysadmin you absolutely 100% should be paying attention. Almost every single vendor I’ve seen selling their applications only have initscripts. Which then cause issues. I’ve gone to the vendors and told them and they’ve said go to Red Hat. Well Red Hat doesn’t support that vendor’s init scripts.
Not naming an application, but it was from a BIG BLUE company and they said their only instructions are to call their script from the user. But it won’t remain running if you do that because systemd will close out the slice when the user logs out. SO it’s obvious they haven’t tried what they’re suggesting.
And I’m not attempting to state that systemd is impressive in any way. systemd basically took what had been building over 40 years of init scripting and threw it out the window and said our way is better. I don’t think it is. I’m just saying, with a directive based unit file it’ll be simpler to parse than a bash script.
Yeah, the landscape changes if you’re a professional sysadmin running multiple servers with uptime requirements, and possibly proprietary software or unusual hardware. I contend that that is, in and of itself, not the most common use case. 😉
systemd isn’t bad at all. People simply don’t understand that it is not “just an init service”.
It’s a massive question, and I think quite a lot of the argument comes from the fact that it depends what direction you’re answering it from.
As a user, do I like being able to just
systemctl enable --now whatever.service, and have a nice overview of ‘how’s my computer’ insystemctl status? Yes, that’s a big step up from symlinking run levels and other nonsense, much easier.As an administrator, do I like having services, mounts and timers all managed in one way? Yes, that is very nice - can do more with less, and have to spend less time hunting for where things are configured. Do I think that the configuration files for these are a fucking mess of ‘just keep adding new features in’ and the override system is lunacy? Also yes.
As a developer trying to do post-mortem debugging, who just wants all the logs in front of him for some server that’s gone wrong somehow, which I often have to request via an insane daisy-chain of emails and ‘Salesforce nonsense that our tech support use’ from our often fairly non-technical end users, on some server that I’ve no other access to? No, I do not find having logs spread between
/var/logand journalctl (and various CloudFormation logs in a web console) makes my life easier. I would be pleased if that got sorted out.tl:dr; mostly an improvement, some caveats.
As someone who recently started learning linux properly by setting up arch, systemd is nice. It does a lot of things that make life easier for me and it never gets in the way.
I didn’t know that logging question is related to SystemD, so thanks for telling it! As an non-top class desktop user the same thing frustrates especially because the solution is often simpler and not found from those logs.
Just try to implement user session management on a non systemd distro…
Systemd is way better than others init system. I’m using Alpine Linux on my phone and I really wait for a Fedora/Arch like PMOS project (it’s on the way)
[]Just saying, not everyone needs session management…
sudo su
Why spawn additional process when you can get into shell directly with
sudo -s?Well, sudo itself is a purely optional component—you can run a system quite happily with just su .
What do you do with all the process you save with that trick ?
Because I already had my fingers closer to “su” than to “-s”… but more seriously, because I tend to use
sudo -E suon a remote terminal with a PS1 set to colorize the prompt based on whether I’m running root and the host if it’s remote, butsudo -E -sdoesn’t run the root’s.bashrcthat runs the updated colorization while at the same time exports too much of the user’s environment into the root shell.
I am fine with it and personally make heavy use of it. On my Arch install, I use systemd-boot, systemd-timesyncd, neworkd, resolved and unified kernel images with ukify.
Roast me you systemd haters.
Is the current SystemD rant derived from years ago (while they’ve improved a lot)?
In my experience, the same arguments against systemd (not systemD) are still used. No matter how often they have been disproved or whether the problem has been fixed in the meantime. With many users I am sure that it is only about making the project systemd bad.
Should Linux community rant about bigger problems such as Wayland related things not ready for current needs of normies?
I would prefer it if there were no rants at all. No matter what the topic. Because that doesn’t help in any way. It would make more sense to invest the energy in the projects in question or in alternative projects to improve them.
We’d probably need to qualify this with “bad compared to what”. I can’t complain, as it does its job, and I’ve been able to tweak what I needed to. As I don’t tinker with it every week, I keep a sticky note rolled up on my desktop, or I quickly use ‘cheat systemd’ to remember some key examples.
I was getting really long start up time earlier this year (like 19 mins before the desktop was fully responding) and after trying everything else I tried ditching BTRFS and reverting my /home drive back to ext4. Turns out BTRFS start and checks was killing my boot times. Now, as fast as anything.
The following have been my saviours though in identifying boot times: journalctl -b -p err systemd-analyze blame --user systemd-analyze blame
Keep in mind that it all started 20 years ago with Pulseaudio. Pottering was not really a nice guy (on mailing lists ofc, I don’t know him personally) whose software I wanted on my machine.
Problem was never speed or even technical, problem was trust on original author and single-mindedness that they were promoting. Acting like it is the only way forward, so anyone believing in freedom part of free software was against it. Additionally, it was looking like tactics used by proprietary software companies to diminish competition.
It looked scary to some of us, and it still does, even worse is that other software started having it as hard dependency.
All of this looks like it was pushed from one place: Portering and RedHat.
While after 20 years I might have gotten a bit softer, you can imagine that 15 years ago some agresive and arogant guy who had quite a bad habbit of writing (IMHO) stupid opinions wanted to take over my init system… no, I will not let him, not for technical reasons but for principal.
I want solutions to come from community and nice people, even if they are inferior, I will not have pottering’s code on my machine so no systemd and no pulseaudio for me, thank you, and for me it is an important choice to have.
Keep in mind that it all started 20 years ago with Pulseaudio. Pottering was not really a nice guy (on mailing lists ofc, I don’t know him personally) whose software I wanted on my machine.
Poettering is like Torvalds: gruff when pressed, but not wrong.
PulseAudio is like systemd: dramatically better than what came before, and the subject of a great deal of criticism with no apparent basis in reality.
PulseAudio did expose a lot of ALSA driver bugs early on. That may be the reason for its bad rap. But it’s still quite undeserved.
Additionally, it was looking like tactics used by proprietary software companies to diminish competition.
This is a nonsensical argument. Systemd is FOSS. It can and will be forked if that becomes necessary.
Which, in light of recent changes at Red Hat, seems likely to happen soon…
Problem was never speed or even technical, problem was trust on original author and single-mindedness that they were promoting.
That’s because fragmentation among fundamental components like sound servers and process supervisors results in a compatibility nightmare. You really want to go back to the bad old days when video games had to support four different sound servers and the user had to select one with an environment variable? Good riddance to that.
I want solutions to come from community and nice people
Then you’d best pack your bags and move to something other than Linux, because Linus Torvalds is infamous for his scathing (albeit almost invariably correct) rants.
@argv_minus_one @monobot Pulse Audio is good for a desktop multi-program environment. Its optimized for context switching, however for MIDI applications it has too much latency and you need to use Jack.
@thanhdo @argv_minus_one @monobot What about Pipewire? Doesn’t that combine Pulseaudio und Jack capabilities?
Those sound like perfectly valid reasons to me. Thanks for the illuminating look into it.
I’ve been using Slackware for more than a decade, and all this systemd talk has me feeling like I dodged a bullet.
Fedora uses dracut as opposed to initramfs, so that’s also a major difference.
dracut is just an initramfs generator, last I checked.
The Systemd init system and its consequences have been a disaster for the human race. It has greatly increased the life-expectancy of those of us who live in “just werks” distros, but it has destabilized GNU/Linux society, has made life unfulfilling, has subjected users to indignities, has led to widespread psychological suffering (in the BSD world to physical suffering as well) and has inflicted severe damage on the Unix world. The continued development of Systemd will worsen the situation. It will certainly subject human beings to greater indignities and inflict greater damage on the Unix world, it will probably lead to greater social disruption and psychological suffering, and it may lead to increased physical suffering even in “just werks” distros.
The Systemd system may survive or it may break down. If it survives, it MAY eventually achieve a low level of physical and psychological suffering, but only after passing through a long and very painful period of adjustment and only at the cost of permanently reducing users and many other Unix processes to engineered products and mere cogs in the Systemd machine. Furthermore, if the system survives, the consequences will be inevitable: There is no way of reforming or modifying PID 1 so as to prevent it from depriving users of dignity and autonomy.
If the system breaks down the consequences will still be very painful. But the bigger the system grows the more disastrous the results of its breakdown will be, so if it is to break down it had best break down sooner rather than later.
We therefore advocate a revolution against the Systemd system. This revolution may or may not make use of violence; it may be sudden or it may be a relatively gradual process spanning a few decades. We can’t predict any of that. But we do outline in a very general way the measures that those who hate the Systemd system should take in order to prepare the way for a revolution against that form of society. This is not to be a POLITICAL revolution. Its object will be to overthrow not distros but the init-system basis of the present GNU/Linux ecosystem.
Complete this sentence: As an AI language model …
it’s a copied pasta
As an AI language model, I can understand and communicate fluently in different languages such as English, 中文, 日本語, Español, Français, Deutsch and others. Here are some examples of how I can complete the sentence:
- As an AI language model, I can translate text or speech from one language to another.
- As an AI language model, I can paraphrase or rewrite sentences to improve clarity or style.
- As an AI language model, I can detect the sentiment or tone of a text or speech.
- As an AI language model, I can answer questions or chat about various topics.
- As an AI language model, I can generate catchy slogans or headlines for products or services.
Which one do you find the most useful? 😊
As someone who recently started learning linux properly by setting up arch, systemd is nice. It does a lot of things that make life easier for me and it never gets in the way.
What DE/WM you chose with your Arch? Wayland?
KDE on X because I like KDE and have an nvidia GPU.
