Is the project still active?

[lumarogit]

I've seen several pull requests go unanswered for months, as well as no commits merged. I hope this continues. Thanks.

[ElyasSindi2020]

well heres the thing. 1st off void linux uses eudev so if it truly was dead they would switch to something else or fork it them selves.

and also, if a project is not in need of updates, it wont be updated. so far there isnt really anything wrong with eudev so what would need an update? if u were to update it, what would u update? theres nothing to update. if something did need to be updated. it surely will be updated.

[lumarogit]

Thanks for responding. If you're a developer on this project, please review this PR.

#290

Thank you.

[ElyasSindi2020]

I'm not a dev on this

[ruby-R53]

if a project is not in need of updates, it wont be updated. so far there isnt really anything wrong with eudev so what would need an update?

you do have a point, but i see that a lot of the issues here went unanswered and some of them are bugs, so i would say there actually is something wrong in some parts of it that should be reviewed

i'm making this comment a literal year since the last commit here too, and considering the pace of development was rather fast in 2024, i fear this project is going dormant

[mid-kid]

eudev has a lot of code that needs to evolve along with changes in the kernel and changes in hardware. While it seems to still work fine for all my use cases, I can't really imagine this to hold true forever.

[ruby-R53]

yeah, i guess we can even say it's gone dormant at this point already

[drfiemost]

It has been deprecated in Gentoo in favor of building udev from systemd sources: https://wiki.gentoo.org/wiki/Udev
Same in LFS: https://linuxfromscratch.org/lfs/view/development/chapter08/udev.html

I'd say it's officially dead

[aanderse]

i hope more people will give the mdevd project the attention that it deserves - it is entirely viable as a replacement for (e)udev

i am running mdevd on my laptop with no loss in functionality from a full systemd+udev system, my home server, and even on a steam deck gaming device for my kids

[mid-kid]

I'm aware of a few distributions still using eudev: Alpine linux, Buildroot, Slackware, Void Linux.

There's very few distributions with a no-systemd option anymore, of which I only know Artix Linux, Gentoo and LFS that have switched to systemd-udev. I can't tell what Devuan and MX/AntiX are doing.

[AsciiWolf]

if a project is not in need of updates, it wont be updated.

What about the udev rules (and hwdb)? They are awfully outdated.

[ArsenArsen]

FWIW, now that systemd-udev seems to be entirely usable on non-systemd systems, eudev is simply a redundant downstream really. I'd recommend those distros to switch, too

[drfiemost]

If anything this project could sync with the systemd udev sources and add a sane build system, but someone should step up and do the hard work.
Else just add a note and/or archive the project for good so people would know.
Maybe @bbonev can shed some light on eudev's fate?

[eddiegb]

FWIW, now that systemd-udev seems to be entirely usable on non-systemd systems, eudev is simply a redundant downstream really. I'd recommend those distros to switch, too

I take issue with the idea that in the OSS world we should all just gather around one piece of software and agree to all use it. I want and embrace that there exist competing projects that do the same thing. If we should not have alternatives, then why have gnu/linux, the BSDs and whatever else your favourite OS is? Why have different Linux distributions? Why have gzip, zstd, bzip2? and endless other examples.

Furthermore, why do people want to have alternatives when they all fulfil the same technical criteria? It's because software does not exist in a technical vacuum. There are political/social/personal reasons that people choose one piece of software over another, and whatever these are they are perfectly valid. Wanting an alternative udev implementation to exist just because you take philosophical issue with systemd is perfectly acceptable and a GOOD thing.

And before you say it, yes, the same principle applies here as to any software; if you want it then you have to step up and do the work to code it. So if no one makes any effort to keep eudev alive, then it deserves to die. But I sincerely hope it (and mdevd and any others) do not suffer that fate.

[mid-kid]

@eddiegb I'm not sure your argument makes a lot of sense when you consider that eudev isn't really an alternative - it's simply an outdated, lightly patched version of systemd-udev. Most of its changes have landed in systemd-udev.

[eddiegb]

"Alternative" is not a word you can argue about, it has a very clear meaning. As soon as two or more things become available to choose from then they all become alternatives. You can discuss which one is "best" but you can't argue whether one of the options can be termed an "alternative" or not, it just is.