July 2025 - Arch-general - lists.archlinux.org

folder in /var/cache/pacman/pkg
by Frank 01 Aug '25

01 Aug '25

Good afternoon, I just discovered this folder /var/cache/pacman/pkg/download-LyI4Km/ with the following content: -rw-r--r-- 1 alpm alpm 34091008 5. Apr 07:49 calibre-8.0.1-3- x86_64.pkg.tar.zst.part -rw-r--r-- 1 alpm alpm 62468096 5. Apr 07:49 electron34-34.5.0-2- x86_64.pkg.tar.zst.part -rw-r--r-- 1 alpm alpm 19755008 2. Apr 21:00 mesa-1:25.0.3-1- x86_64.pkg.tar.zst -rw-r--r-- 1 alpm alpm 60813312 5. Apr 07:49 qt6-webengine-6.9.0-1- x86_64.pkg.tar.zst.part -rw-r--r-- 1 alpm alpm 37257216 5. Apr 07:49 telegram-desktop-5.13.1- 2-x86_64.pkg.tar.zst.part -rw-r--r-- 1 alpm alpm 106354016 5. Apr 01:17 zen-browser-bin-1.11b-2- x86_64.pkg.tar.zst -rw-r--r-- 1 alpm alpm 566 5. Apr 01:17 zen-browser-bin-1.11b-2- x86_64.pkg.tar.zst.sig Don't know where this is coming from but am wondering whether it's safe to delete it? THX Frank

5 5

Fwd: Possible malicous PKGBUILD
by David C Rankin 31 Jul '25

31 Jul '25

Dear Arch Officials, Will you please review the perpetual "moderation" that is attached to my AUR e-mail. This is going on a year and is quite ridiculous. This is my third request for review of this unfortunate matter. -------- Forwarded Message -------- Subject: Re: Possible malicous PKGBUILD Date: Wed, 30 Jul 2025 09:22:24 +0000 From: aur-general-owner(a)lists.archlinux.org To: drankinatty(a)gmail.com Your message to the Aur-general mailing-list was rejected for the following reasons: The message comes from a moderated member The original message as received by Mailman is attached. -- David C. Rankin, J.D.,P.E.

2 1

Re: Licensing changes in package sources: does one link mentioned here 404?
by u34＠net9.cf 24 Jul '25

24 Jul '25

Sven-Hendrik Haase <svenstaro(a)archlinux.org> wrote: > Hey all, > > With [RFC40][0] and [RFC52][1] merged, we're now going full steam to > make sure the changes outlined in those RFCs are soon implemented. > If you are a packager and only care about how this concerns you without > the nitty gritty details, here's what you need to know in short: > > 1. PKGBUILDs and auxiliary files (e.g. config files, etc.) that Arch > Linux contributors provided will be licensed under 0BSD. Full current > list [here][3]. > 2. Licensing information will be documented in a `REUSE.toml` file per > repository and will be maintained by package maintainers going forward. > 3. We will generate a `REUSE.toml` (using `pkgctl license setup`) for > most packages. > 4. You may have to make changes to your packages (details below)! > 5. When moving package source repositories from the AUR, either contact > past contributors and ask for consent to license under 0BSD, or rewrite > the package sources. > > The changes introduced in [RFC40][0] and [RFC52][1] have a general > impact on packaging work. > > To clearly document which licenses apply to which files, we are > introducing REUSE [2], which provides a structured configuration file > format (`REUSE.toml`) for mapping files to concrete licenses. > In the near future, we will add these files to package source repos in > an automated fashion using `pkgctl license setup`. > While doing so, we will exclude some packages added after 2024-11-01, > see [this issue][4] for further details. > [4]: > https://gitlab.archlinux.org/archlinux/packaging/licensing-package-sources/… I get a 404. -- u34 > > In some cases (around 750 out of 12462 repos), packager maintainers will > need to manually perform some changes, as adding a `REUSE.toml` will not > work out of the box. There will be a todo list on Arch web for those > packages where manual action is required. > The most common reasons for this are: > > 1. The package's license isn't specified in an SPDX conforming fashion > (e.g. `GPL2` instead of `GPL-2.0-or-later`). There are open MRs on > GitLab to make some packages SPDX conforming [5]. We will go through > these and we'll try to merge all of these. > 2. Multiple license identifiers are specified in the PKGBUILD and > SRCINFO files and patch files are present in the package source repository. > In this case, we can't automatically pick a suitable upstream license. > > By default, all files in package source repos must be licensed under 0BSD. > 0BSD is very permissive and files licensed under its terms can be > contributed upstream under their desired license. > > For some files, 0BSD might not be suitable, and contributors need to > choose a different license depending on the situation. Here are some > common cases: > > - Included files originating from contributors that have not agreed to > the 0BSD license. > If the files are available under a FOSS license, mark the files under > that particular license in the `REUSE.toml`. > If no license is provided, the files need to be rewritten or removed. > - Patches taken from upstream and committed to the repository need to be > annotated with the correct upstream license. > > `pkgctl license` will be available from devtools v1.4.0 onwards. After > this release, you will encounter a warning when running `pkgctl release` > if your package source repository isn't compliant. > > If there are any open questions with existing packages and the new reuse > setup, reach out over the mailing list or in the packaging channel. > > Cheers, > Rafael and Sven > > [0]: https://rfc.archlinux.page/0040-license-package-sources/ > [1]: https://gitlab.archlinux.org/archlinux/rfcs/-/merge_requests/52 > [2]: https://reuse.software/ > [3]: > https://gitlab.archlinux.org/archlinux/devtools/-/blob/b8c475b3f402809fc65b… > [4]: > https://gitlab.archlinux.org/archlinux/packaging/licensing-package-sources/… > [5]: > https://gitlab.archlinux.org/search?scope=merge_requests&search=spdx&state=…

2 1

Good PR from bad AUR FF bin package situation courtesy of The Register
by David C Rankin 23 Jul '25

23 Jul '25

All, Despite the unfortunate AUR RAT issue, Arch received a good review of its handling of the matter: Arch Linux users told to purge Firefox forks after AUR malware scare https://go.reg.cx/tdml/dfd67/68a9047f/70c7d2e3/4hwN The article spoke well of Arch throughout. That's a good reflection on Arch governance and reaffirmation of the the way the distro has been managed. Well done. (I'm not affiliated with The Register, but worth subscribing to the excellent daily newsletter) -- David C. Rankin, J.D.,P.E.

1 0

About the 2 patches and GitLab requests
by c4a＠monocles.eu 23 Jul '25

23 Jul '25

Within the last 24 hours, I have sent 2 patches. Since the mailing list clearly isn't the right place to post them, I decided to request a GitLab account. However, this decision of mine has changed. As a preemptive action to prevent any drama, I have sent a message in the support ticket to disregard the request. None of this is done maliciously to harm Arch Linux or flood people's mailboxes. I really want this to be the last message about this. I mainly just was curious about the bring up process for packages. And I wanted to bring up a couple packages for the Arch Linux team since they are out of date.

2 1

[PATCH] upgpkg: 7zip 25.00
by c4a＠monocles.eu 22 Jul '25

22 Jul '25

From: Charles <c4a(a)monocles.eu> Unfortunately I had to add a prepare function to patch out the -Werror flag using sed. Apart from that the package appears to work just fine. --- PKGBUILD | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/PKGBUILD b/PKGBUILD index ab0524b..14c55f6 100644 --- a/PKGBUILD +++ b/PKGBUILD @@ -1,8 +1,8 @@ # Maintainer: Evangelos Foutras <foutrelis(a)archlinux.org> pkgname=7zip -pkgver=24.09 -pkgrel=3 +pkgver=25.00 +pkgrel=1 pkgdesc="File archiver for extremely high compression" arch=('x86_64') url="https://www.7-zip.org" @@ -13,7 +13,12 @@ provides=('p7zip') conflicts=('p7zip') replaces=('p7zip') source=(https://7-zip.org/a/7z${pkgver//./}-src.tar.xz) -sha256sums=('49c05169f49572c1128453579af1632a952409ced028259381dac30726b6133a') +sha256sums=('bff9e69b6ca73a5b8715d7623870a39dc90ad6ce1f4d1070685843987af1af9b') + +prepare() { + cd "$srcdir" + sed -i 's/-Werror//g' CPP/7zip/7zip_gcc.mak +} build() { local _platform_flags=() -- 2.50.1

3 2

[PATCH] upgpkg: less 679
by c4a＠monocles.eu 22 Jul '25

22 Jul '25

From: Charles <c4a(a)monocles.eu> --- PKGBUILD | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/PKGBUILD b/PKGBUILD index 95a9671..b279e80 100644 --- a/PKGBUILD +++ b/PKGBUILD @@ -3,7 +3,7 @@ # Contributor: judd <jvinet(a)zeroflux.org> pkgname=less -pkgver=668 +pkgver=679 pkgrel=1 epoch=1 pkgdesc='A terminal based program for viewing text files' @@ -14,7 +14,7 @@ depends=('glibc' 'ncurses' 'pcre2') validpgpkeys=('AE27252BD6846E7D6EAE1DD6F153A7C833235259') # Mark Nudelman source=("https://www.greenwoodsoftware.com/$pkgname/$pkgname-$pkgver.tar.gz" "$pkgname-$pkgver.tar.gz.sig::https://www.greenwoodsoftware.com/$pkgname/$pkgname-$pkgver.sig") -sha256sums=('2819f55564d86d542abbecafd82ff61e819a3eec967faa36cd3e68f1596a44b8' +sha256sums=('9b68820c34fa8a0af6b0e01b74f0298bcdd40a0489c61649b47058908a153d78' 'SKIP') prepare() { -- 2.50.1

2 1

Bind in systemd-nspawn Container
by Pocket 18 Jul '25

18 Jul '25

Is it possible to run bind inside a nspawn container? Created a container with pacstrap -k base into /var/lib/machines/Bind/ set root password in container Using host networking machinectl start Bind machinectl login Bind Login as root root@Bind named]# pacman -Syy&&pacman -S bind vim Configured bind as resolver per archlinux wiki BIND root@Bind named]# systemctl start named&&systemctl status named ● named.service - Internet domain name server Loaded: loaded (/usr/lib/systemd/system /named.service; enabled; preset: disabled) Active: active (running) since Thu 2025-07-17 19:45:23 EDT; 17ms ago Invocation: f5eddb5e40fb4c548d8eae7671c563fd Main PID: 1109 ((named)) Tasks: 1 (limit: 19108) CPU: 9ms CGroup: /system.slice/named.service └─1109 "(named)" Jul 17 19:45:23 Bind systemd[1]: Started Internet domain name server. [root@Bind named]# dig @127.0.0.1 redhat.com ;; communications error to 127.0.0.1#53: connection refused ;; communications error to 127.0.0.1#53: connection refused ;; communications error to 127.0.0.1#53: connection refused ; <<>> DiG 9.20.11 <<>> @127.0.0.1 redhat.com ; (1 server found) ;; global options: +cmd ;; no servers could be reached [root@Bind named]# dig @localhost. redhat.com ;; communications error to ::1#53: connection refused ;; communications error to ::1#53: connection refused ;; communications error to ::1#53: connection refused ;; communications error to 127.0.0.1#53: connection refused ; <<>> DiG 9.20.11 <<>> @localhost. redhat.com ; (2 servers found) ;; global options: +cmd ;; no servers could be reached root@Bind named]# ps aux|grep named named 1109 0.0 0.1 223180 22328 ? Ssl 19:45 0:00 /usr/bin/named -f -u named root 1135 0.0 0.0 3796 2064 pts/1 S+ 19:47 0:00 grep named [root@pocket ~]# ps aux|grep named root 22107 0.0 0.0 16348 10928 pts/1 Sl+ 19:44 0:00 vim etc/named.conf vu-Bind+ 22145 0.0 0.1 223180 22328 ? Ssl 19:45 0:00 /usr/bin/named -f -u named root 22228 0.0 0.0 6468 2048 pts/0 S+ 19:47 0:00 grep named From journalctl: Jul 17 20:00:34 Bind named[1306]: couldn't add command channel 127.0.0.1#953: permission denied Jul 17 20:00:34 Bind named[1306]: configuring command channel from '/etc/rndc.key' Jul 17 20:00:34 Bind named[1306]: couldn't add command channel ::1#953: permission denied Jul 17 20:00:34 Bind named[1306]: listening on IPv4 interface lo, 127.0.0.1#53 Jul 17 20:00:34 Bind named[1306]: creating IPv4 interface lo failed; interface ignored Jul 17 20:00:34 Bind named[1306]: listening on IPv4 interface end0, 192.168.50.5#53 Jul 17 20:00:34 Bind named[1306]: creating IPv4 interface end0 failed; interface ignored Jul 17 20:00:34 Bind named[1306]: IPv6 socket API is incomplete; explicitly binding to each IPv6 address separately Jul 17 20:00:34 Bind named[1306]: listening on IPv6 interface lo, ::1#53 Jul 17 20:00:34 Bind named[1306]: creating IPv6 interface lo failed; interface ignored Jul 17 20:00:34 Bind named[1306]: listening on IPv6 interface end0, 2603:6010:6c00:3f2e:2ecf:67ff:fed8:6900#53 Jul 17 20:00:34 Bind named[1306]: creating IPv6 interface end0 failed; interface ignored Jul 17 20:00:34 Bind named[1306]: listening on IPv6 interface end0, fe80::2ecf:67ff:fed8:6900%2#53 Jul 17 20:00:34 Bind named[1306]: creating IPv6 interface end0 failed; interface ignored Looks like it is running in the container but can not bind to any interfaces Looks like I am missing something or it just doesn't work. Anyone have any magic pixie dust? -- Hindi madali ang maging ako

2 2

Systemd resolved
by Pocket 17 Jul '25

17 Jul '25

I have a router that has DHCP and assigns IP and DNS servers The DNS server assigned are 192.168.50.4 192.168.50.2 fe80::dea6:32ff:fe25:41d2 192.168.50.4 pi-hole 192.168.50.2 Bind DNS recursive fe80::dea6:32ff:fe25:41d2 pi-hole The pi-hole server uses 192.168.50.2 as its DNS name resolution resolvectl returns this Global Protocols: -LLMNR -mDNS -DNSOverTLS DNSSEC=allow-downgrade/supported resolv.conf mode: stub Fallback DNS Servers: 1.1.1.1#cloudflare-dns.com 9.9.9.9#dns.quad9.net 8.8.8.8#dns.google 2606:4700:4700::1111#cloudflare-dns.com 2620:fe::9#dns.quad9.net 2001:4860:4860::8888#dns.google Link 2 (end0) Current Scopes: DNS Protocols: +DefaultRoute -LLMNR -mDNS -DNSOverTLS DNSSEC=allow-downgrade/supported Current DNS Server: fe80::dea6:32ff:fe25:41d2 DNS Servers: 192.168.50.4 192.168.50.2 fe80::dea6:32ff:fe25:41d2 Default Route: yes Link 3 (wlan0) Current Scopes: none Protocols: -DefaultRoute -LLMNR -mDNS -DNSOverTLS DNSSEC=allow-downgrade/supported Default Route: no The following for end0 wired connection Current DNS Server: fe80::dea6:32ff:fe25:41d2 changes or cycles through the three DNS servers, changing about every 5 minutes or upon browsing the web with firefox ( firefox is set to use my DNS servers) Is that how resolved really works? I was under the impression that once the "Current DNS Server" was set it would stay that way unless the "Current DNS Server" failed. I there a setting other than /etc/systemd/resolved.conf ? [Resolve] DNS=192.168.50.4 fe80::dea6:32ff:fe25:41d2 Domains=~. I would really like to have it assigned by the router instead of at each client. -- Hindi madali ang maging ako

2 2

Inquiry About Creating a Modified Distribution Based on Arch Linux
by 王奕宁 14 Jul '25

14 Jul '25

Dear Arch Linux Development Team, I am working on a customized Linux distribution named LinuxToGo, which would be explicitly based on Arch Linux (not a fork). Our goal is to create a specialized Linux distribution derived from Arch Linux, focused exclusively on providing a robust, portable Linux experience bootable directly from external drives (e.g., USB sticks, SSDs, etc.). All releases will prominently state: "LinuxToGo is NOT an official Arch Linux product. It is a modified version of Arch Linux. Official resources can be found at archlinux.org." The project website/documentation will clearly credit Arch Linux as the foundation and link to the official project. Technical Approach The distribution would utilize the Arch Linux package repositories ([core], [extra], [community]) directly where possible. Customizations (e.g., installer scripts, default configurations, or curated software sets) will be developed/maintained separately in our repositories. We will strictly adhere to all relevant Arch Linux trademarks and branding guidelines. I would like to confirm whether this approach aligns with Arch Linux’s policies, particularly regarding: The use of the Arch Linux name/brand in the context of "based on Arch Linux." Distributing pre-configured ISOs that utilize official repositories. Any specific compliance requirements we should follow. We deeply respect the Arch Linux community and philosophy, and wish to ensure full alignment with your guidelines before proceeding. Thank you for your time and consideration. Wtada233 wtada233(a)outlook.com

6 5