5 ms·

Systemd Invasion into Linux Server Space

29 points by lelf 8y ago

twblalock 8y ago
This is a slanted, conspiratorial rant full of personal attacks.
- msla 8y ago
  > This is a slanted, conspiratorial rant full of personal attacks. Par for the course as regards reportage on systemd, in other words. It's weird: People are so het up on hating systemd it's like they forgot to give any truly coherent reasons why we should hate systemd. They just assume everyone does, and launch into the rant on that basis. It reminds me of people thinking disco is terrible and so much worse than any other kind of dance music, or that Nickleback is utterly inexcusable as opposed to being a somewhat mediocre rock band, or the endless stream of hatred for Modern architecture and Modern art. Once you remove the constant blaring hatred, there's just nothing there.
  - fao_ 8y ago
    Have... have you actually read https://suckless.org/sucks/systemd/ https://suckless.org/sucks/systemd/ ? There are solid reasons for disliking systemd.
    - icedchai 8y ago
      I've been using Linux since 1993, FreeBSD since 1995. systemd has a weird, "non-unixy" feel to it. It does too much. Yeah, the fast bootup times are nice.
      - majewsky 8y ago
        > systemd has a weird, "non-unixy" feel to it. It just occurred to me that this is just the Tanenbaum-Torvalds debate all over again. Context: https://en.wikipedia.org/wiki/Tanenbaum%E2%80%93Torvalds_debate https://en.wikipedia.org/wiki/Tanenbaum%E2%80%93Torvalds_deb...
        yellowapple 8y ago
        Question is: is systemd the Linux or the MINIX in this particular debate?
      - fao_ 8y ago
        > Yeah, the fast bootup times are nice. Every machine I've put SystemD on takes an extra 20 seconds to boot. This includes thinkpads, and other machines. I've tried a mix of distributions, all of the ones with SystemD take longer. I've even tried altering things to take less time. So I don't see how 'faster boot' can be considered part of systemd's bag of tricks.
        icedchai 8y ago
        Maybe it is all a lie. I mostly run it on servers.
- yellowapple 8y ago
  It's definitely a rant, but most of it (that I've read at least) is well-reasoned and not at all ad hominem in nature (aside perhaps from an irrational hatred of Red Hat in general).
xyzzy_plugh 8y ago
systemd is such a colossal step in the right direction, this is nothing but FUD. There's certainly a case to be made about running the project in a more community-friendly manner, but systemd is mature and pretty much everywhere now, so it's hard to argue what the benefits might've been.
zzzcpan 8y ago
Systemd doesn't matter much in the server space, since the server space is moving away from using host's init system for anything. Where it is actually problematic is linux desktop. I wasted so much time dealing with it on desktop, it's not even funny.
fao_ 8y ago
The Server Space is arguably the main place SystemD was designed for, and probably the only place it makes sense to deploy it. There are a subset of use-cases that it does make sense to run SystemD, and the server-space is the place where those cases lie. In my opinion, for the security and stability of most systems, initd is superior. Lennart Poettering has a long history of ignoring blatant security problems, and SystemD itself has a long history of containing system-breaking bugs, and in general being over-engineered for most use-cases. See: https://suckless.org/sucks/systemd/ https://suckless.org/sucks/systemd/ and http://without-systemd.org/wiki/index.php/Arguments_against_systemd http://without-systemd.org/wiki/index.php/Arguments_against_... and http://without-systemd.org/wiki/index.php/List_of_articles_critical_of_systemd http://without-systemd.org/wiki/index.php/List_of_articles_c... EDIT: There's really nothing here I can say that hasn't been talked about elsewhere. What I will say is that, elsewhere on hacker news, people seem to be preoccupied with "Language Safety", and Rust, and the adjacent languages that provide that. The reason for this is to reduce bugs and potential exploits. You know what else causes less bugs? LESS CODE, AND LESS FEATURES. The less code there is running on a system, the less code there is to exploit, and the less bugs there are likely to be. PID 1 is a SACRED, HOLY, Process Identifier. Code that runs with this PID has control over the ENTIRE SYSTEM. An exploit in PID 1 is LITERALLY game over. Let us not forget: https://www.theregister.co.uk/2017/07/28/black_hat_pwnie_awards/ https://www.theregister.co.uk/2017/07/28/black_hat_pwnie_awa... (Links to bug reports are halfway down the page)
- pulisse 8y ago
  > You know what else causes less bugs? LESS CODE, AND LESS FEATURES. The less code there is running on a system, the less code there is to exploit, and the less bugs there are likely to be. PID 1 is a SACRED, HOLY, Process Identifier. Code that runs with this PID has control over the ENTIRE SYSTEM. An exploit in PID 1 is LITERALLY game over. That's an odd point to make in defense of systemd, since one of the primary objections to its use on servers is that it has vastly more surface area than is required for process supervision.
  - kohtatsu 8y ago
    That point wasn't in defence of systemd.
chomp 8y ago
This article is so biased, I had to flag it, I'm sorry. I can appreciate a good systemd HCI argument. Maybe a Unix philosophy argument, but those are pretty debunked these days. I can appreciate an attack surface/system responsibility argument. But this article delves into obnoxious tropes, biases, rants against systemd, and doesn't present anything that's thought provoking.
- zzzcpan 8y ago
  The article is very long, I haven't even read most of it but already saw some good arguments. You are dismissing it for no reason.
  - chomp 8y ago
    Claiming that boot time doesn't matter to servers is a good argument? Quoting Karl Rove is a good argument? Lennart Poettering is a trojan horse? How can this sway anyone?
    - yellowapple 8y ago
      > Claiming that boot time doesn't matter to servers is a good argument? I'd say so, yeah. Unless you think boot time does matter on a server, in which case I'd ask why.
majewsky 8y ago
I will always love systemd, if only for the major achievement of bluntly paving over a ton of distributions' stupid idiosyncrasies. Units look identical on all distributions. Network configuration looks identical on all distributions. /etc/os-release looks identical on all distributions. As a developer, and as an admin dealing with different distros all the time, this is a godsend.

29 points by lelf 8y ago

twblalock 8y ago

This is a slanted, conspiratorial rant full of personal attacks.

msla 8y ago
> This is a slanted, conspiratorial rant full of personal attacks. Par for the course as regards reportage on systemd, in other words. It's weird: People are so het up on hating systemd it's like they forgot to give any truly coherent reasons why we should hate systemd. They just assume everyone does, and launch into the rant on that basis. It reminds me of people thinking disco is terrible and so much worse than any other kind of dance music, or that Nickleback is utterly inexcusable as opposed to being a somewhat mediocre rock band, or the endless stream of hatred for Modern architecture and Modern art. Once you remove the constant blaring hatred, there's just nothing there.
- fao_ 8y ago
  Have... have you actually read https://suckless.org/sucks/systemd/ https://suckless.org/sucks/systemd/ ? There are solid reasons for disliking systemd.
  - icedchai 8y ago
    I've been using Linux since 1993, FreeBSD since 1995. systemd has a weird, "non-unixy" feel to it. It does too much. Yeah, the fast bootup times are nice.
    - majewsky 8y ago
      > systemd has a weird, "non-unixy" feel to it. It just occurred to me that this is just the Tanenbaum-Torvalds debate all over again. Context: https://en.wikipedia.org/wiki/Tanenbaum%E2%80%93Torvalds_debate https://en.wikipedia.org/wiki/Tanenbaum%E2%80%93Torvalds_deb...
      - yellowapple 8y ago
        Question is: is systemd the Linux or the MINIX in this particular debate?
    - fao_ 8y ago
      > Yeah, the fast bootup times are nice. Every machine I've put SystemD on takes an extra 20 seconds to boot. This includes thinkpads, and other machines. I've tried a mix of distributions, all of the ones with SystemD take longer. I've even tried altering things to take less time. So I don't see how 'faster boot' can be considered part of systemd's bag of tricks.
      - icedchai 8y ago
        Maybe it is all a lie. I mostly run it on servers.
yellowapple 8y ago
It's definitely a rant, but most of it (that I've read at least) is well-reasoned and not at all ad hominem in nature (aside perhaps from an irrational hatred of Red Hat in general).

xyzzy_plugh 8y ago

systemd is such a colossal step in the right direction, this is nothing but FUD. There's certainly a case to be made about running the project in a more community-friendly manner, but systemd is mature and pretty much everywhere now, so it's hard to argue what the benefits might've been.

zzzcpan 8y ago

Systemd doesn't matter much in the server space, since the server space is moving away from using host's init system for anything. Where it is actually problematic is linux desktop. I wasted so much time dealing with it on desktop, it's not even funny.

fao_ 8y ago

The Server Space is arguably the main place SystemD was designed for, and probably the only place it makes sense to deploy it. There are a subset of use-cases that it does make sense to run SystemD, and the server-space is the place where those cases lie. In my opinion, for the security and stability of most systems, initd is superior. Lennart Poettering has a long history of ignoring blatant security problems, and SystemD itself has a long history of containing system-breaking bugs, and in general being over-engineered for most use-cases. See: https://suckless.org/sucks/systemd/ https://suckless.org/sucks/systemd/ and http://without-systemd.org/wiki/index.php/Arguments_against_systemd http://without-systemd.org/wiki/index.php/Arguments_against_... and http://without-systemd.org/wiki/index.php/List_of_articles_critical_of_systemd http://without-systemd.org/wiki/index.php/List_of_articles_c... EDIT: There's really nothing here I can say that hasn't been talked about elsewhere. What I will say is that, elsewhere on hacker news, people seem to be preoccupied with "Language Safety", and Rust, and the adjacent languages that provide that. The reason for this is to reduce bugs and potential exploits. You know what else causes less bugs? LESS CODE, AND LESS FEATURES. The less code there is running on a system, the less code there is to exploit, and the less bugs there are likely to be. PID 1 is a SACRED, HOLY, Process Identifier. Code that runs with this PID has control over the ENTIRE SYSTEM. An exploit in PID 1 is LITERALLY game over. Let us not forget: https://www.theregister.co.uk/2017/07/28/black_hat_pwnie_awards/ https://www.theregister.co.uk/2017/07/28/black_hat_pwnie_awa... (Links to bug reports are halfway down the page)

pulisse 8y ago
> You know what else causes less bugs? LESS CODE, AND LESS FEATURES. The less code there is running on a system, the less code there is to exploit, and the less bugs there are likely to be. PID 1 is a SACRED, HOLY, Process Identifier. Code that runs with this PID has control over the ENTIRE SYSTEM. An exploit in PID 1 is LITERALLY game over. That's an odd point to make in defense of systemd, since one of the primary objections to its use on servers is that it has vastly more surface area than is required for process supervision.
- kohtatsu 8y ago
  That point wasn't in defence of systemd.

chomp 8y ago

This article is so biased, I had to flag it, I'm sorry. I can appreciate a good systemd HCI argument. Maybe a Unix philosophy argument, but those are pretty debunked these days. I can appreciate an attack surface/system responsibility argument. But this article delves into obnoxious tropes, biases, rants against systemd, and doesn't present anything that's thought provoking.

zzzcpan 8y ago
The article is very long, I haven't even read most of it but already saw some good arguments. You are dismissing it for no reason.
- chomp 8y ago
  Claiming that boot time doesn't matter to servers is a good argument? Quoting Karl Rove is a good argument? Lennart Poettering is a trojan horse? How can this sway anyone?
  - yellowapple 8y ago
    > Claiming that boot time doesn't matter to servers is a good argument? I'd say so, yeah. Unless you think boot time does matter on a server, in which case I'd ask why.

majewsky 8y ago

I will always love systemd, if only for the major achievement of bluntly paving over a ton of distributions' stupid idiosyncrasies. Units look identical on all distributions. Network configuration looks identical on all distributions. /etc/os-release looks identical on all distributions. As a developer, and as an admin dealing with different distros all the time, this is a godsend.