Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Linux distros without systemd

ungleich.ch

Post reply on HN
Print view View original

131–140 of 449 posts

Re: Linux distros without systemd

#131

Post by otterley »

nnq wrote: post #123

What's the recommended way for someone to learn systemd properly? Is there a readable manual somewhere that documents everything? That's the only thing I find scary about such big systems, the lack of comprehensive docs and a "starting point for learning them" from 0 to hero. Like, with Git there's tons of nice tutorials like "learn kit bottom up by reinventing it in way X" etc. plus a good "default book". What's the…

Good question! A great introduction that was written while the author was still building systemd can be found here:

http://0pointer.de/blog/projects/systemd.html

Also, the "systemd for administrators" series:

https://gist.github.com/bcremer/8cdf6900c35dda65f387

Re: Linux distros without systemd

#132

Post by heinrichhartman »

nnq wrote: post #123

What's the recommended way for someone to learn systemd properly? Is there a readable manual somewhere that documents everything? That's the only thing I find scary about such big systems, the lack of comprehensive docs and a "starting point for learning them" from 0 to hero. Like, with Git there's tons of nice tutorials like "learn kit bottom up by reinventing it in way X" etc. plus a good "default book". What's the…

I found the man pages to be a pretty good read.

Use `man -t systemctl` to produce a pdf version.

There are a couple of files you want to read, e.g. systemd.service, systemd.unit, systemd, deamon etc.

But those are linked at the `SEE ALSO` section.

Re: Linux distros without systemd

#133

Post by creeble »

marcosdumay wrote: post #97
meddlepal wrote: post #13

Earlier quoted context omitted.

It seems the systemd hate mostly comes from a philosophical position these days than anything grounded in real user impact. You either agree or disagree with what systemd does, or do not care because you are not aware of it (which I would argue is a good thing). Personally I think it is great. Sysvinit scripts were terrible to write and frequently had bugs because the scripts were usually afterthoughts "Oh yea... I g…

Well, just excuse my philosophical stance of not liking my computers breaking by a new random cause every couple of months.

This is my experience as well.

Suddenly logins take 30s. Why? Some new systemd problem with reverse DNS.

Suddenly reboots don't happen. Why? Systemd is spinning on a task that didn't used to spin. Yes, probably fixable, but not discovered until trying to reboot.

Can't login at coffee shop WiFi suddenly. Why? Some random systemd DNS issue.

Is my system more secure? Maybe, but possibly because it breaks so often that low uptime keeps it safe!

For my servers, I seriously have to cross my fingers and have the colo phone number handy whenever I reboot, fearful that they won't restart because of some random systemd interaction that happened on some previous upgrade. It's gotten that bad.

Re: Linux distros without systemd

#134

Post by dual_basis »

api wrote: post #78
AsyncAwait wrote: post #12

I get that not everybody likes systemd, but a lot of the criticisms seem misguided and looking at sysvinit with role-colored glasses. For example, people are complaining about logind being included, but forget that consolekit was unmaintained before that. Yes, there's parts to systemd that not everyone will use, like container support, but in that case you can pretty safely ignore that use case. For me, systemd has p…

My gripe is bad UX. The whole system is needlessly obtuse and esoteric. Git is another popular thing with this problem, but in the case of git it provides so much value it's enough to get me to overcome its obtuseness. For systemd it's not "enough better" to get me to forgive it for being designed with no thought whatsoever to simplicity or ease of use. Another thing that makes me forgive git more is that its underly…

I strongly agree that git has a UX issue, and the widespread adoption of it over mercurial is a perfect example of inertia leading us to get stuck in a local minima.

Re: Linux distros without systemd

#135

Post by russellbeattie »

sjwright wrote: post #10

One of the better videos on this subject: https://www.youtube.com/watch?v=6AeWu1fZ7bY BSDCan 2018—Benno Rice: The Tragedy of systemd "systemd is, to put it mildly, controversial. Depending on who you ask it's either a complete violation of the UNIX philosophy, a bloated pile of bugs, a complete violation of the elegant simplicity it replaced or, it most cases, some or all of the above. So why have so many Linux distr…

Basically, people don't like systemd because they don't like change, and groups like shitting on other groups of people?

I'm not sure the presenter is condescending enough. Maybe he should have given an abbreviated history of Unix as well, oh wait, he did that.

Re: Linux distros without systemd

#136

Post by eropple »

Crinus wrote: post #118
Piezoid wrote: post #91

Earlier quoted context omitted.

> They pushed for Gnome to have a hard dependency on it so that everyone would be _forced_ to use it instead of just letting it be accepted by merits. We should standardize on the interfaces, avoiding hard dependencies. There is dbus for that.

dbus is also a dependency though. If you are on the bus (heh) that says systemd is a hard dependency that should be avoided then dbus is also the same thing.

I agree with your logical consistency, but I'm curious: is D-Bus sufficiently specified that one could write an alternative to it? That is, is it a hard dependency or is it itself an interface? I'm not nearly sufficiently experienced with desktop Linux application development to know. ;)

Re: Linux distros without systemd

#137

Post by darkpuma »

sjwright wrote: post #117
darkpuma wrote: post #105

Earlier quoted context omitted.

You are presuming that everyone gives a damn about open source as an end unto itself, not merely a means to an end. Ironically, the people who do consider it an end unto itself are probably the people most likely to despise the term "open source" and prefer "Free Software". Ask RMS what he thinks about SystemD and I bet you he'll shrug and say it's fine because it's Free, and that's all he really cares about. Needles…

Trust me, I don't give a damn about open source as an end in itself—to me RMS means root mean square first, Roads and Maritime Services second, and occasionally I might be reminded of an unwashed beard.

> "Trust me, I don't give a damn about open source as a means to an end"

See, I think that's where you differ from most people. To most people, I believe open source is a means to an end, not an end unto itself. What they want first and foremost is something that works, and it happens to be the case that having the source code freely available facilitates having something that works and gives them a sense of comfort, an assurance, that it will continue to work in the foreseeable future. So appealing to "it's opensource, so you should like it" totally misses the mark. The argument should be "it works, therefore you should like it."

Had SystemD never given anybody any trouble, I don't think it would be hated. I think SystemD is hated because (particularly during early adoption years) there were a lot of things about it that didn't work for a lot of people. This made them angry, because it violated what they actually cared about. Not permissive licensing, but rather having something that works. The principle of open source is not axiomatic to most people; it only has value when it is part of having something that works.

Somewhere else I see this mentality: Android. There are a few among the 'community' who would have me pop open the champagne because "Linux won" on account of Android being an incredible commercial success. But why should I celebrate that? Why is "number of CPUs running Linux" the metric I should be glued to? Because Linux is open source? So what? Honestly, who gives a damn?

Re: Linux distros without systemd

#138

Post by JohnFen »

dmitryminkovsky wrote: post #113
marcosdumay wrote: post #95

Earlier quoted context omitted.

When there was a problem with a sysvinit procedure, one could simply fix it. As a result, mostly nobody ever had a problem with it. When there is a bug with systemd, you have to fix systemd. And your version of systemd is completely different from the head, because head was rewritten last month (in a process that persists bugs) and yours is a year old. You can't just use head either, because every low level system on…

> As a result, mostly nobody ever had a problem with it. The number of broken init scripts I have encountered throughout the years, that went unfixed for months if not years, stand in stark contrast to this statement. And I’m talking packages like uWSGI on Ubuntu 14.04! This is exactly why I gravitated immediately to systemd. The idea that init scripts actually worked and when they didn’t were easily fixable is not t…

> The people who prefer initv are the people who enjoy digging in other people’s (untestable) Bash scripts

This is demonstrably untrue. I strongly prefer SysV (which is not to say I think SysV is wonderful or anything) over systemD for a number of architectural and functional reasons.

But I do not enjoy digging into other people's init scripts. Fortunately, I rarely have to do that. I think the last time I had a problem with someone's init script was over 7 years ago.

Re: Linux distros without systemd

#139

Post by acdha »

marcosdumay wrote: post #95
AsyncAwait wrote: post #12

I get that not everybody likes systemd, but a lot of the criticisms seem misguided and looking at sysvinit with role-colored glasses. For example, people are complaining about logind being included, but forget that consolekit was unmaintained before that. Yes, there's parts to systemd that not everyone will use, like container support, but in that case you can pretty safely ignore that use case. For me, systemd has p…

When there was a problem with a sysvinit procedure, one could simply fix it. As a result, mostly nobody ever had a problem with it. When there is a bug with systemd, you have to fix systemd. And your version of systemd is completely different from the head, because head was rewritten last month (in a process that persists bugs) and yours is a year old. You can't just use head either, because every low level system on…

> When there was a problem with a sysvinit procedure, one could simply fix it. As a result, mostly nobody ever had a problem with it.

To the extent that this isn't flat-out wrong, it's only because SysV init had so few features and so the actual patch had to go to one of many different upstreams. Having had to fix the same classes of problem in hundreds of different places like that has given me an appreciation for how long it's been since I've needed to do anything with systemd other than use it.

One important thing to remember is that the ability to have a standard way to start a process, as a non-privileged user, keep it running, with logging, etc. is something which has been standardized in the Windows world since the early 90s. It's really nice not to keep having to spend time working around the lack of a check-box competition feature for a quarter of a century.

Re: Linux distros without systemd

#140

Post by eropple »

tyingq wrote: post #77
AsyncAwait wrote: post #12

I get that not everybody likes systemd, but a lot of the criticisms seem misguided and looking at sysvinit with role-colored glasses. For example, people are complaining about logind being included, but forget that consolekit was unmaintained before that. Yes, there's parts to systemd that not everyone will use, like container support, but in that case you can pretty safely ignore that use case. For me, systemd has p…

It's not the init replacement part that bugs me. It's the expansion into audio, kernel logging, resolvers, containers, and so forth that is the issue. It's a monolith of sorts that breaks the loosely coupled toolbox concept of Unix.

Systemd is a bunch of loosely coupled tools, though; it's not a monolith at all. PulseAudio was written by Lennart but isn't part of systemd. You don't have to use systemd's resolver. systemd-nspawn is a completely separate process and isn't required for systemd as a whole.

It's worth watching Benno Rice (a BSD developer, actually) discuss systemd and some of the misconceptions, some of which you're expressing, therein: https://www.youtube.com/watch?v=o_AIw9bGogo

Post reply on HN