News for hackers. Stuff that matters.
Re:Systemd 248 RC3: systemd-oomd is now considered fully sup... by cmeacham98 on Saturday March 13, 2021 @07:14PM (#26451491)
This is a common mistake. systemd-oomd is not "in" systemd the init system. Similar to systemd-resolved, systemd-networkd, systemd-boot, and others it just has the systemd name because it is developed by the systemd team. A minimal system using sytemd for the init system and other software for everything else only needs sytemd, dbus, and journald.

I use several of these projects because I like the systemd design philosophy (ex the way the CLI interface and config files are) and trust the developers of their software. However, there are popular altenatives available for all of these components, should you choose to use them:

systemd-resolved: bind9, dnsmasq, powerdns

systemd-networkd: NetworkManager, ConnMan

systemd-boot: GRUB2, Refind, EFISTUB

systemd-oomd: earlyoom, nohang, oomd (the facebook one, not systemd)

Re:Systemd 248 RC3: systemd-oomd is now considered fully sup... by dathinab on Saturday March 13, 2021 @07:31PM (#26451570)
I agree but I wouldn't call systemd a init system.

It's a service manager for your Linux OS which manage all services down to initial startup related services.

Just viewing it as a init system is misleading.

As a side note I think a service manager directly above the kernel is a must have for any modern server or desktop operating system ;=)

Re:Systemd 248 RC3: systemd-oomd is now considered fully sup... by viraptor on Saturday March 13, 2021 @08:50PM (#26452042)
> I agree but I wouldn't call systemd a init system.

Systemd is the name of the init system. Systemd is also the name of the project developing multiple systemd-* services. Maybe they could help the situation a bit by renaming the service itself systemd-init, but we've had enough time for people to learn to deal with a tiny bit of ambiguity...

[dead]
Re:Systemd 248 RC3: systemd-oomd is now considered fully sup... by IgorPartola on Saturday March 13, 2021 @08:03PM (#26451758)
You know I feel that way about the Linux kernel. Everything is done by the Linux kernel developers and it’s shipped as one monolithic thing. Wouldn’t it be better to get drivers for networks cards from other places? Or have to pick from a list of memory management packages that get downloaded from GitHub on every reboot? And why do they include things like spin lock implementations? Should that be an NPM package? /s
Re:Systemd 248 RC3: systemd-oomd is now considered fully sup... by badsectoracula on Monday March 15, 2021 @01:18AM (#26462849)
You're joking but having drivers separate from the rest of the Linux kernel and with a stable ABI would certainly help other OSes as well (e.g. ReactOS, Haiku).
Re:Systemd 248 RC3: systemd-oomd is now considered fully sup... by viraptor on Saturday March 13, 2021 @08:52PM (#26452052)
> I suddenly have to use systemd-* over them.

You really don't. You have the choice to go with a distro which doesn't use them, or to mix&match yourself.

Re:Systemd 248 RC3: systemd-oomd is now considered fully sup... by zbentley on Sunday March 14, 2021 @10:56AM (#26455923)
> Like, hell, what is systemd-boot and when and why it replaced grub2?

I've been really happy with systemd-boot; in my limited (used it on probably a dozen machines across several distros at this point) experience, it's better at autodetecting nonstandard images/locations than grub, and is vastly friendlier to weird partitioning schemes. It also has a much more minimal config surface than grub, which I like (since, across distros, I no longer have to learn a distro-specific set of nested/included grub config files and conventions, nor do I have to worry about what will be auto-regenerated when a distro does a kernel update versus what configs I need to edit by hand). However, that limited config surface may be a drawback to people who need to customize very esoteric boot cases.

I also aesthetically prefer systemd-boot's minimalism and focus on newer (UEFI) boot strategies compared to grub; the latter is huge in terms of functionality. A lot of that size is needed for rare cases and older boot configurations (and if you fall into either group, grub is what you should use), but it complicates the "happy path" on post-2010 computers.

https://en.wikipedia.org/wiki/Gummiboot_(software)

Re:Systemd 248 RC3: systemd-oomd is now considered fully sup... by candiddevmike on Saturday March 13, 2021 @11:17PM (#26452768)
I personally prefer systemd-boot over grub2, it's dead simple to configure and works really well for UEFI systems (and dual booting windows).