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

Why pro-systemd and anti-systemd people will never get along

uselessd.darknedgy.net

Post reply on HN
Print view View original

51–60 of 113 posts

Re: Why pro-systemd and anti-systemd people will never get along

#51

Post by vertex-four »

exelius wrote: post #43
hello_there wrote: post #37

What I really dislike about these debates about systemd is that usually nobody mentions the alternatives. From reading the debate I get the impression that both sysvinit and systemd have major problems, but I'm never presented with a solution. So if I'm going to be against sysvinit as well as systemd, then what should i be supporting?

The main alternatives seem to be OpenRC and Upstart. One of the biggest criticisms of systemd by its detractors is that it's unproven and not stable; and neither of these address that concern. Upstart was developed by Ubuntu, which has stopped supporting it and is adopting systemd because they recognized the need to standardize on a single system. After the Debian committee voted to go with systemd, Ubuntu (being a D…

> The main alternatives seem to be OpenRC and Upstart.

There's also outliers like nosh[0], which implements at least some of the good bits of systemd (socket activation is in there), and the ideas and general architecture are solid enough that it could be extended to implement more (device/path dependencies through udev and inotify are two that I hear people wanting). Then separate pieces of software can implement the remainder of what systemd implements (its many dbus APIs).

[0] http://homepage.ntlworld.com/jonathan.deboynepollard/Softwar...

Re: Why pro-systemd and anti-systemd people will never get along

#52

Post by jamescun »

nanoscopic wrote: post #49

That was a very long and detailed explanation of the sides of the argument. It misses one huge thing: Consistency. The elephant in the room is that most people who dislike systemd simply dislike it because they learned and used init.d scripts, runlevels, and chkconfig, and that is the extent of their understanding. The main hatred for systemd is coming from people simply being resistant to change. If it ain't broke,…

That is a rather broad estimation of all those who dislike systemd.

While there are people who most likely dislike systemd because they are resistant to change, they don't constitute everyone with objections. I for one dislike systemd, not because I am resistant to change, but on a purely technical basis (which I won't discuss as it has already been reiterated ad nauseam).

I must also add that I dislike sysv and like that the community is discussing alternatives; but the race to systemd, with some serious technical failings which have not been addressed, scares me.

Re: Why pro-systemd and anti-systemd people will never get along

#53

Post by fst01 »

justincormack wrote: post #26
fidotron wrote: post #11

It will be very interesting to see how this all plays out in the long run, though the short term is utterly depressing. My hunch is Lennart will be vindicated, on the basis that systemd as it exists right now is not right, but it can be evolved into something much better, and while that may not have occurred to other projects here the stakes are enormous. There is a definite strand to the resistance that any change t…

Yes, I have a side project to experiment with scripting the OS with Lua [1] - it has enough functionality to bring up a system (networking, mounts, namespaces etc) without any other code. [1] https://github.com/justincormack/ljsyscall

And using "Scriptable operating systems with Lua[1]" there is no userspace needed?

[1] http://dl.acm.org/authorize?N80603 [2] http://2014.splashcon.org/track/dls14-papers#Proceedings

Re: Why pro-systemd and anti-systemd people will never get along

#54

Post by uselessdguy »

nanoscopic wrote: post #49

That was a very long and detailed explanation of the sides of the argument. It misses one huge thing: Consistency. The elephant in the room is that most people who dislike systemd simply dislike it because they learned and used init.d scripts, runlevels, and chkconfig, and that is the extent of their understanding. The main hatred for systemd is coming from people simply being resistant to change. If it ain't broke,…

No, I specifically addressed this:

--------------------------------

"The rather huge scope and opinionated nature of systemd leads to people yearning for the days of sysvinit. A lot of this is ignorance about good design principles, but a good part may also be motivated from an inability to properly convey desires of simple and transparent systems. In this way, proponents and opponents get caught in feedback loops of incessantly going nowhere with flame wars over one initd implementation (that happened to be dominant), completely ignoring all the previous research on improving init, as it all gets left to bite the dust. Even further, most people fail to differentiate init from rc scripts, and sort of hold sysvinit to be equivalent to the shoddy initscripts that distros have written, and all the hacks they bolted on top like LSB headers and startpar(2). This is a huge misunderstanding that leads to a lot of wasted energy."

-------------------------------

This isn't about people "hating change". It looks like it, because a lot of people who defend sysvinit aren't really doing that as much as they are defending minimal and transparent systems. In fact, there's way too many people who don't understand "init". Init is the first userspace process that is started. That's it. Init doesn't mean "manages services", "manages processes" or anything like that. Those are separate concepts. The sooner we realize this, the sooner we can have some more innovative architectures for managing services, as we're still trapped in this mental cage.

Moreover, it's not just systemd haters who are resistant to change. A lot of systemd lovers are, as well. In fact, the reason we didn't fix the problem earlier and stuck with SysV for so long was precisely because people didn't care about init, and didn't want to change their flawed ways. Well, at least in the Linux communities. Many of the people who resisted change when presented with non-SysV approaches back in the day are the same who now support systemd and lament on how much "systemd haters don't like change".

systemd, of course, went significantly beyond service management, and thus had a much bigger impact than previous designs which were rather focused on one problem domain. Thus, systemd simply became far more prominent (and controversial) than anything else because of its huge ambitions.

Re: Why pro-systemd and anti-systemd people will never get along

#55

Post by chc »

jimktrains2 wrote: post #39
zokier wrote: post #33

Earlier quoted context omitted.

> The more I realize all the untapped potential lying around, the more I realize how so many Linux users are living in their monoculture. Unfortunately there is no one there to amplify all the good efforts and hidden gems scattered all over the place, so you have people just reading Phoronix and LWN articles and standing in their bubble. Meanwhile, all the non-Linux Unices and the "toy project" builders are doing gre…

> This is one of the main reasons I approve of systemd. It is something completely new fighting against the old entrenched cruft. Most people aren't anti-systemd because it's new, but because it doesn't conform to their philosophy of how their computer should run. There could be newer init systems that conform to their ideals, but systemd is not one of them.

Could there be, or are there? Because it is entirely possible that their philosophy is too rigid to permit a system that is substantially different from what exists now, in which case "They're against it because it's new" seems like a fair assessment.

Re: Why pro-systemd and anti-systemd people will never get along

#56

Post by lmm »

nanoscopic wrote: post #49

That was a very long and detailed explanation of the sides of the argument. It misses one huge thing: Consistency. The elephant in the room is that most people who dislike systemd simply dislike it because they learned and used init.d scripts, runlevels, and chkconfig, and that is the extent of their understanding. The main hatred for systemd is coming from people simply being resistant to change. If it ain't broke,…

I don't hate change. I just want to be able to carry on using FreeBSD (for reasons that, believe it or not, are primarily technical), and Pottering seems to think that means I shouldn't be allowed use Gnome.

Re: Why pro-systemd and anti-systemd people will never get along

#57

Post by mercurial »

vidarh wrote: post #23
ck2 wrote: post #19

I don't want anything bad to happen to Lennart, I just want him to take his football and go home. Then suddenly init.d will look very open and friendly again. Maybe the same thing can happen to grub2 and I wouldn't mind going back to good old grub.conf

> Then suddenly init.d will look very open and friendly again. Having used systemd on a few machines by now, I won't ever go back. It's utterly naive to think that systemd would go away if Lennart were to.

Not to mention that quite a few people contribute to systemd [1], even if Lennart is still leading the pack.

1: https://github.com/systemd/systemd/graphs/contributors?from=...

Re: Why pro-systemd and anti-systemd people will never get along

#58

Post by nova »

api wrote: post #14
fidotron wrote: post #11

It will be very interesting to see how this all plays out in the long run, though the short term is utterly depressing. My hunch is Lennart will be vindicated, on the basis that systemd as it exists right now is not right, but it can be evolved into something much better, and while that may not have occurred to other projects here the stakes are enormous. There is a definite strand to the resistance that any change t…

I agree. So far I've seen it as a classic conservatives vs. liberals debate. The conservatives have no interest at all in changing the Unix paradigm under Linux. It works perfectly well, thank you. The liberals see it as archaic and unsuited to a wide variety of new use cases including rapid-deployment VMs, containers, mobile and laptop computers, and the desktop in general. I side with the liberals here, but with a…

“The whole modern world has divided itself into Conservatives and Progressives. The business of Progressives is to go on making mistakes. The business of Conservatives is to prevent mistakes from being corrected. Even when the revolutionist might himself repent of his revolution, the traditionalist is already defending it as part of his tradition. Thus we have two great types -- the advanced person who rushes us into ruin, and the retrospective person who admires the ruins. He admires them especially by moonlight, not to say moonshine. Each new blunder of the progressive or prig becomes instantly a legend of immemorial antiquity for the snob. This is called the balance, or mutual check, in our Constitution.”

― G.K. Chesterton

Re: Why pro-systemd and anti-systemd people will never get along

#59

Post by e7620 »

hello_there wrote: post #37

What I really dislike about these debates about systemd is that usually nobody mentions the alternatives. From reading the debate I get the impression that both sysvinit and systemd have major problems, but I'm never presented with a solution. So if I'm going to be against sysvinit as well as systemd, then what should i be supporting?

I happen to like modern daemontools-like init systems a lot, (runit, s6) but I use various init systems depending on what I want, and that's great IMO.

The problem is that systemd is that is being forcefully pushed to all distros. For instance, by merging udev with systemd. We want to keep all the alternatives open.

Re: Why pro-systemd and anti-systemd people will never get along

#60

Post by justincormack »

fst01 wrote: post #53
justincormack wrote: post #26

Earlier quoted context omitted.

Yes, I have a side project to experiment with scripting the OS with Lua [1] - it has enough functionality to bring up a system (networking, mounts, namespaces etc) without any other code. [1] https://github.com/justincormack/ljsyscall

And using "Scriptable operating systems with Lua[1]" there is no userspace needed? [1] http://dl.acm.org/authorize?N80603 [2] http://2014.splashcon.org/track/dls14-papers#Proceedings

Well, you could use it for that in theory yes, though no one has tried yet as far as I know...
Post reply on HN