• floquant@lemmy.dbzer0.com
    link
    fedilink
    arrow-up
    22
    ·
    edit-2
    6 days ago

    Systemd is fine.

    Journald is fine.

    But someone pass me a mace I can beat systemd-resolved and systemd-logind to death with

    EDIT: Oh come on

  • ronflex@lemmy.world
    link
    fedilink
    arrow-up
    10
    ·
    6 days ago

    Systemd has simplified my life on a few occasions, and it seems to be reliable from what I can tell. At the end of the day if I can get the OS to do what I want in a relatively simple matter, that’s all I care about.

    • Lettuce eat lettuce@lemmy.ml
      link
      fedilink
      arrow-up
      1
      ·
      2 days ago

      In all seriousness, I’ve yet to encounter a situation where Systemd made any meaningful negative difference in my Linux experience.

      I’ve never had problems with any init system, Systemd or otherwise.

  • whoisearth@lemmy.ca
    link
    fedilink
    arrow-up
    19
    arrow-down
    1
    ·
    7 days ago

    My biggest complaint with systemd…

    Service xxx stop/start/restart is so much easier than

    Systemctl stop/start/restart xxx

    It fucking annoys me

    • Wildmimic@anarchist.nexus
      link
      fedilink
      English
      arrow-up
      44
      arrow-down
      1
      ·
      8 days ago

      I fully agree. I am a user with a bit of technical background, but not a lot of detailled knowledge about the inner workings of an operating system (i know boolean logic and basic programming structures - in Pascal lol - from the 90’s, what a transistor does and stuff, how to build my own PCs and handle filesystems and troubleshooting).

      With init scripts, i hit a wall pretty fast.

      With Systemd i know how to start, stop and configure services, and the suite built around it uses the same conventions everywhere, making the everyday life with Linux for someone like me so much easier and more transparent than ever before.

    • michaelmrose@lemmy.world
      link
      fedilink
      English
      arrow-up
      8
      arrow-down
      2
      ·
      7 days ago

      Have you considered that just “reaping old process IDs” wasn’t enough responsibility for an init daemon on a secure, robust system? That maybe it should be protecting other parts of the system and tracking the liveness of a desired service?

      What is the benefit of specifically doing that in init?

      If I see an argument like this then I can only assume the interlocutor doesn’t do software engineering.

      Its more likely that the user simply has simple needs like running stuff at startup which any init system can do and doesn’t see as much benefits as poster.

      Also who loves systemd-resolved?

      • azertyfun@sh.itjust.works
        link
        fedilink
        arrow-up
        12
        arrow-down
        1
        ·
        7 days ago

        Being able to assign a nameserver per interface with a domain wildcard is a fucking godsend. I use it every day with a hook script because my job uses some private domains but I don’t want to send my entire DNS history through the VPN. Now ~job.com goes to tun0 and that’s the end of it.

        systemd-resolved is not perfect but with libnss’s overly rigid nature the only alternative for my use-case would be to recreate similar functionality to resolved with dnsmasq – which is just objectively worse especially when you want to use DHCP sometimes but not always. Why reinvent the wheel? resolved does its job and does it well. I had some issues with it a few years ago but have been using it for the past couple years without complaint.

      • anyhow2503@lemmy.world
        link
        fedilink
        arrow-up
        12
        arrow-down
        2
        ·
        7 days ago

        What is the benefit of specifically doing that in init?

        What’s the alternative?

        Also who loves systemd-resolved?

        I don’t think I will ever love anything DNS-related, but it’s still the best solution I’ve used for name resolution on a system with many interfaces.

      • Rooty@lemmy.world
        link
        fedilink
        arrow-up
        18
        ·
        8 days ago

        I still remember the bad old days of stale repositories and compiling from scratch. Never again.

        • corsicanguppy@lemmy.ca
          link
          fedilink
          English
          arrow-up
          11
          ·
          8 days ago

          There was 25 years between c;m;mi and lennart’s cancer, filled with excellent choices better than either.

      • Cenzorrll@lemmy.world
        link
        fedilink
        arrow-up
        3
        ·
        edit-2
        7 days ago

        I just had an issue with the vscodium flatpak, been using it for two months with no issue in an online course, got to learning GUIs, import module, doesn’t exist. I couldn’t figure out why it wasn’t there, installed three different python versions of it three different ways, still nothing. Couldn’t even get vscodium to point to a different interpreter that I knew was there (yet it doesn’t say it’s not there, just that some things won’t work). Still nothing. Three hours later, after trying everything I could think of, I realized that it was because I installed the flatpak version when it clicked that it worked in Geany and I didn’t have python 3.13 in my repos, yet that was the only one I could see in vscodium.

    • Trainguyrom@reddthat.com
      link
      fedilink
      English
      arrow-up
      43
      arrow-down
      1
      ·
      8 days ago

      Since I started actually doing system administration and actually interacting directly with SystemD all of the hate for it I’d soaked up from enthusiast forums melted away. I’ve never used any of the other init systems so maybe I’m missing out, but I do appreciate SystemD for what it does