Hacker News Reader

Comment by smhenderson | Hacker News Reader

I know this has been discussed all over the place to the point of exhaustion but I am still somewhat shocked that Debian adopted systemd so quickly. For a distribution that is known for it's stability and conservative approach to new software I just think this happened way too fast.

I've been a happy Debian user for many years but lately I've fallen back to using *BSD (particularly OpenBSD). I guess that puts me in the "old and afraid of change" camp but for me it's nice to be back using something that looks and, more importantly, _works_ like Unix.

Debian will remain my goto distribution when I want/need Linux but it's no longer my first choice...

View on HN

What are you talking about? Debian had systemd on the discussion list for years and it was the default init system for other distros for just as long. What exactly about years of discussion doesn't align with the conservative choices of their other software defaults?

Here's an article where debian is talking about systemd in 2011 [0]. Forgive me for beign a bit shocked when people say things like 'it happened so fast!', but it's almost like people just sincerely want to hate systemd.

[0]http://lwn.net/Articles/452865/

edit: I really wanted to ignore this, but it's eating away at me.

>and, more importantly, _works_ like Unix.

Maybe I'm just cynical, but I get the feeling people who pretend their interpretation of the unix philosophy regarding systemd is the superior interpretation, probably didn't even know what the tenets of unix philosophy were before the drama.

Maybe it's relative to systemd reach in the OS and the fact that it's not a no brainer (this topic as proof). How many deep changes Debian did before (I think there was a glibc/eglibc swap at one point) ?

So how many years of discussion have to happen before you would consider it thoroughly discussed? How many years of top distro's using it by default before you would consider it vetted?

From my interpretation of smhenderson's post, one of Debian values was a very slow pace of change and old but well known pieces. In this light, systemd still seems too controversial. Fedora adopted it because it was an in-house project, archlinux tries to use cutting edge as much as possible, so it's irrelevant for that interpretation.

I only read part of the Debian voting process, it's wasn't years but I admit not knowing how long systemd was discussed in Debian's ML before that. On archlinux board's, adoption was quickly reached (without too much user troubles).

Well in the typical Arch way it was a case of dictatorial "we are going systemd, deal with it".

Also one of the busybodies within systemd development is directly involved with Arch maintenance...

Btw, the big pedia claims that OpenSuse and Mageia(?) also use systemd. At least one of them is a fork of what used to be a RHEL fork with a more desktop orientation.

Then again, the adoption may well have happened before systemd started absorbing/reimplementing so many other functions.

Look, I come to hacker news to agree to disagree. If you like systemd than great but as far as I'm concerned it's still a moving target, does way more than replace the init system, uses binary where traditional Unix uses text and in general inserts itself into way too many things for me to be comfortable with it.

Smarter people than me have done a great job of debating both sides of the issue and it looks like systemd has won. That doesn't mean I have to like it and thankfully I can fall back to using OpenBSD since I want a Unix that looks and feels like the Unix I've been using my whole career.

I don't care if you disagree with me but please don't assume that I "don't even know the tenets of unix philosphy" becuase I happen to view systemd in a way that you apparently do not. If all the major distributions are going with systemd than OK, but the beauty of free software is that I have choices and I've made mine.

I think Debian has taken a conservative approach. For example, Arch adopted systemd as its default years ago, which means it was already stable enough for daily use years ago (although I think Arch helped pave the way for its adoption by the larger distros). systemd is not really "new" software anymore; it's been around the block a few times. Even RHEL7 has adopted systemd.

I suppose one might say that this is a core component of the system and even three or so years of testing is not enough... But, I think at that point, we're off in the realm of opinion.

Fedora was using it as default for a year before Arch. Redhat definitely paved the way. Many were slow to adopt it because it was such a radical change and most of the changes were seen as boons only for the virtualization stack that Redhat was working with.

I love Arch too, but damn! I remember the "upgrade" process to systemd that wrecked my system (not unrecoverably, but before they released the fix path, I already reinstalled.)

Huh. I didn't know that about Fedora. I admit I don't keep up with the Red Hat world as much as I should. So, apparently systemd has been in production distros for, what, four or five years now?

I don't remember the systemd transition in Arch being too bad, though it was sort of rough around the edges. I guess that roughness is why (mentally) I hadn't expected anyone to have adopted systemd before Arch.

It wasn't even bad on Fedora, it was in rawhide for over a year being worked on before entering Fedora 15, albeit they transitioned from upstart to systemd, which I'd think is a lot less significant a transition than from sysv to systemd.

Also, systemd has really "grown" over the years. Adopting systemd when it was just about handing services was one thing. But now it does networking (firewall management included), job scheduling, and more.

It is almost as if you can see the frog slowly boiling...

It was devastatingly bad for desktop consistency to have networkmanager not be default on so many distros for so long. It made writing GUI network managers a tremendous PITA.

And nobody is even shipping a distro with networkd as the only network daemon. They are all still using networkmanager. I doubt anyone will switch any time soon either, since the faculties of networkd are a lot more limited.

And I'm wondering how a project that is meant to encompass a common base userspace OS is boiling anyone by adopting things like network management into it. That seems pretty base OS nowadays to me. I'll worry if once everyone is on systemd we start getting horror stories of the upstream systemd developers blocking contributions and fixes because they conflict with what they want in the project (a la Gnome, not like KDE).

1. I wonder how many "disable networkmanager" comments i have seen around the net in relation to fixing some networking issue. Seems to be up there with Pulseaudio...

2. It would really surprise me if networkmanager finds itself being anything more than a frontend for networkd within a year or two.

At this point in time i am starting to think that going back to pre-freedesktop Linux may be better for my ulcer than trying to keep up with the desktop (developers) uber alles threadmill.

> For a distribution that is known for it's stability and conservative approach to new software

I imagine many people might trust systemd more than the overlapping set of init/cron/at/supevisord/monit/forever/user generated SysV init shell scripts of varying quality.

So quickly ? This non-issue has been pending for months now, mainly because of non-technical, ridiculous (ie. religious war with sysv) reasons, and we will at last go forward with it. Yes, it is far from being perfect (I personally never really liked systemd), but this is a huge step forward anyway. It's 2015 and we are still using crappy init shell scripts to start demons, and crappy text files to configure every piece of network. We need to move forward.

Can you tell me exactly when text files have started being "crappy"?

"Crappy" in the sense that every piece of network has its own file format, which is parsed by different tools.

But the worst part is probably init scripts.

$ wc -l /etc/init.d/smartd 653 /etc/init.d/smartd

653 lines of code to start a demon, because of the differences between all distros.

Guys, seriously ?

It's actually because of, I'm sorry to put it so bluntly, distribution maintainer incompetence in the form of not modularizing common script functions, bolting on hacks like start-stop-daemon, startpar and insserv to achieve primitive asynchronicity and service dependencies, and to an extent the failings of sysvinit itself with providing little tools to contain the environment and enforcing messy symlink farms.

What you're doing is the equivalent of looking at some old Mach performance statistics and concluding that microkernels cannot possibly work... never mind there's one alongside every smartphone.

Around 1998 or so. They always were an incompatible mess of differing configuration formats, "standards" and kludges, but it was ok for the first 2-3 decades of UNIX...

It's 2015 and people are still making false dichotomies with sysvinit.

>I know this has been discussed all over the place to the point of exhaustion but I am still somewhat shocked that Debian adopted systemd so quickly. For a distribution that is known for it's stability and conservative approach to new software I just think this happened way too fast.

Yeah, they should give it a few decades of thought...

And maybe check out if this ANSI C thing is worth supporting over plain ole K&R C.

"...one fool says `this is old, and therefore good`; the other says `this is new, and therefore better`..."

Only neither is new. One is several decades old and broken, the second is newer but still almost a decade old, and we know it does work and what it fixes...

I don't really understand the force that makes so many distros adopt systemd either.

Who/what is driving this despite all the controverse?

Personal opinions from the user perspective aside, it's significantly easier to maintain from a distribution / package maintainer perspective, so it's unsurprising they'd want to switch.

I think that force is just the developers and maintainers of distros and upstream projects.

Code contributions, even if controversial, pack a much stronger punch than emails, forum posts, blog posts, etc.

The fact that it did many things easily and correctly that init/cron/monit only did after navigating a minefield of gotchas, reinventing the wheel, and employing obnoxious hacks to do things that ought to be simple (parallel init). Init scripts are easier to tweak, no doubt about it, but I'm equally sure that if I was actually throwing together a distro I would want a modern init system.

Also, remember that launchd on OSX has been around for a decade. People from the linux world who encountered it tended to think "hey, this is kinda nice, I wish we had something like it." Perhaps launchd itself had a predecessor somewhere, I dunno. But the point is that people had seen these systems in the wild, so when someone showed up who was serious about seeing the project through there were a bunch of maintainers whose reaction was "about damn time" rather than "what is this?"

I've found service files much easier to tweak then init scripts. When 90% of all init scripts become a single line exec command, there's not a lot to go wrong.

Just this periodic strange urge to adopt superior technology that fixes several broken things at once, instead of keep going with what you're familiar with and sorta-works for decades...

One major thing that we get from systemd is cgroup management, that gives resource and namespace management at a per process level. This has actually been in the kernel since 2008 but to use it you had to have some really bad ass programmers to give you an access interface to the features. It's now trivial to run a hosting server where each site has a set amount of cpu and ram without the virtualization overhead. Before that the option was openvz and before the option was virtualization.

There is a very loud minority in the linux community that doesn't want things to change. There are a ton of linux users who's lives this improves, and they are not as noisy.

Debian is the larges distro and when the maintainers took a vote systemd won with like a 60+% majority. Almost all other distros that are going to, have switched over the last 3 years.

1/ Maintaining init scripts is a significant part of the work to be done in a distribution, and systemd makes this task easier.

2/ Gnome started requiring systemd (via the dependency of GDM on logind). So, if they want to keep supporting Gnome, they have a choice between switching to systemd or developping some shims. And it's probably safe to assume other major pieces of software (KDE?) will adopt similar requirements.

I have been looking at OpenBSD as well for an upcoming Debian related transition. I have nothing against systemd as such, this is pure selfishness. I an happy to let others do the debugging...

You're moving to openbsd so you'll have a more hands off system? Good luck with that.

One of my favourite things about FreeBSD is that it acknowledges that updating will always have edge cases you don't think of. So there's a text file that you read as part of the update workflow, and when specific steps are necessary, you follow them. I've never had an upgrade go wrong with FreeBSD, which is more than I can say for any flavour of Linux.

For what it's worth, Gentoo has a similar system. Packages can show a message when they are installed (including updates) and for system-wide changes the package manager will tell you to to read 'eselect news'.

I loved Gentoo's package management system. I managed to break my system more with Gentoo than any other distro, but that wasn't the package manager's fault—mostly mine and/or package maintainers. I wish it'd gone as mainstream as dpkg and rpm—I prefer widely-used distros these days because I like being able to google any problem and find an answer (I'm well past the days where I found managing my OS to be anything resembling fun) but I miss Portage.

I miss OpenRC, for that matter. Using that is the only time I've felt something like contentment while managing init.

It's more that when you do need to get hands on, the code's pretty minimalist, the manuals are very good quality, there are fewer moving parts so less to go wrong, there's a good focus on code quality etc etc.

Couldn't agree more. OpenBSD is a great bsd for those that want to understand their system inside and out and feel confident that the code quality is there, but it's just not designed to be an ease of use, low barrier to entry, bsd.

Well so far so good. It looks like a series of dead simple scripts and config files. Oh, and the pf stuff appears to be awesome...

Systemd has been in use for a little over 4 years. There may be some issues with distro specific implementations but that should be flushed out in the normal debian testing which had been going on with systemd for over a year?

Well it may not happen quickly, depending on how long it takes to get it to stable.

> but it's no longer my first choice

What is out of interest?

I've recently replaced three lenny servers (two LAMPs and a name server) and rather than go with Wheezy I migrated the three of them to OpenBSD. On one I added apache from ports, the other I was able to use OpenBSD's new httpd and I found I really like OpenBSDs built in named.

Other than that I've been playing around with FreeBSD and have revisited Slackware which was the first Linux I was introduced to via a book/cdrom back in the mid 90's.

Having had some success with OpenBSD on servers I'm running it on a laptop as one of two workstations at work. The other has Wheezy on it. I'm able to do most of my day to day work on OpenBSD but do find myself remoting into the Linux box for some things (e.g. mysql-workbench).

I am curious what you'll do (ie. what platform / distro you'll move to) do if / when *bsd adopts launchd?

OpenBSD will never have launchd (licensing policy)!

launchd use the Apache License how can they have a license policy against that?

I guess you know *BSD might eventually adopt something similar.

This is the great part, because the will probably have API compatibility for a lot of the little parts.(like logind) That means we will not have a mono culture going forward.

Where did you get that idea? The BSD people (just FreeBSD, mostly iXsystems employees) are simply working on a port of OS X's launchd, which has no similar session/seat management component as logind, from what I know.

At least OpenBSD had a GSoC to investigate a logind replacement: http://www.openbsdfoundation.org/gsoc2014.html#systemd

Reply on news.ycombinator.com