Arch Linux

mpan

Member

Registered: 2012-08-01

Posts: 1,386

Website

Re: The future of `reflector` [was: Spring cleanup '25]

Note: this is a reply to a arch-dev-public thread, which can’t be posted the normal ways due to the current unavailability of the latter.
I ask everybody to refrain from pointless discussion, “me-tooing,” or anything that would not normally happen on the MLs. I intentionally neither post it in nor link from the main reflector thread, to not bring unwanted attention. Thanks.

Time to clean up old cruft again. Please head to https://archlinux.org/devel/reports/unneeded-orphans/ and adopt packages you want to keep in the repos.
(...)
I will start dropping packages to AUR in ~ 10 days.
(…)
reflector
(…)

Re: Spring cleanup '25 [1]: Aside from being quite popular among our users (probably for historical reasons and because it's currently the only solution offered in the official repo as far as I can tell), `reflector` is included and used in our ISO [2].
As such, it should *not* be dropped in the current state of things.

Actually I don't think that's a good reason to keep it. If anything it should be removed from ISO as well in my opinion.
I don't think we should be including anything like that because then we should consider also anything from https://wiki.archlinux.org/title/Mirror … de_ranking
So let's keep it simple.

Removal from the ISO may be a reasonable choice. It’s not an essential tool and, as an automatically started service, it was giving people trouble.

Personally I also believe benchmarking tools, that use others’ resources, shouldn’t be run without person’s explicit need. Privacy concerns may be raised by some users too, given running the ISO unconditionally contacts a large number of network services.

But I don’t see a reason to remove reflector from the repos, unless no new maintainer appears in due time.⁽¹⁾ The package is present on 58% pkgstats-ed machines:⁽²⁾ more popular than chromium, zsh, or ruby. Upon removal the only remaining options will be using Xyne’s repo or AUR. Not a huge deal for seasoned Archers, but I don’t think fresh users should be told to configure 3rd party repos or jump into AUR, if that’s not really necessary.
____
⁽¹⁾ Due time: when reflector’s PKGBUILD needs to be updated.
⁽²⁾ Webapp at https://pkgstats.archlinux.de/packages/reflector

Last edited by mpan (2025-03-28 10:30:48)

---

Sometimes I seem a bit harsh — don’t get offended too easily!

Lone_Wolf

Administrator

From: Netherlands, Europe

Registered: 2005-10-04

Posts: 13,473

Re: Re: The future of `reflector` [was: Spring cleanup '25]

Unfortunately Xyne has been MIA for almost a year and no one has been able to reach them.

Reflector should probably be considered unmaintained.

---

Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.

clean chroot building not flexible enough ?
Try clean chroot manager by graysky

twelveeighty

Member

Registered: 2011-09-04

Posts: 1,298

Re: Re: The future of `reflector` [was: Spring cleanup '25]

I'll argue the counter point for the record: Reflector isn't a vital part of the distro because if you remove it from [extra] no machine is going to break. Manually managing mirrors isn't that hard since it can be described in a single paragraph on the Wiki[1]. Additionally, pacman-contrib is actively maintained, with its upstream code also in Arch's GitLab [2] and provides 90% of the Reflector functionality. The "fresh users" vs. AUR is a straw man, because that doesn't fit Arch's User-centric philosophy. A "fresh user" should be considered someone who's willing to read documentation and solve their own problems.

To be clear: I'm not against Reflector. I just think the distro will be fine without it until such time that its maintainer resurfaces.

[1] https://wiki.archlinux.org/title/Mirror … ng_mirrors
[2] https://gitlab.archlinux.org/pacman/pacman-contrib