systemd (259~rc2-1)
[PTS] [DDPO]
COMMITS: VCS has seen 5 commits since the debian/259_rc2-1 tag
- Git: https://salsa.debian.org/systemd-team/systemd.git
-
- Branch: debian/master
- Path: debian/changelog
- Repo size: 107462656
- Browser: https://salsa.debian.org/systemd-team/systemd
- Last scan: 2025-12-05 22:35:17+00
- Next scan: 2025-12-11 23:17:00+00
- Merge requests: 3
- CI pipeline status: success
- Debian changelog in Git:
systemd (259~rc2-1) unstable; urgency=medium
* d/t/control: ensure unit-tests autopkgtest pulls in dlopened libraries
for test
* Update upstream source from tag 'upstream/259_rc2' Update to upstream
version '259~rc2' with Debian dir
cf601b8ede8f71b078792ce228102fe98830a5fc
* Install new varlinkctl bash completion script
-- Luca Boccassi <bluca@debian.org> Wed, 26 Nov 2025 12:41:32 +0000
- This branch is 5 commits ahead of tag debian/259_rc2-1
- Git log:
commit 89509d96921e1e40457d430e32e0aed56a768252
Author: Nick Rosbrook <enr0n@ubuntu.com>
Date: Fri Dec 5 14:04:56 2025 -0500
d/t/tests-in-lxd: re-construct --pin-packages arguments for autopkgtest
When --pin-packages is passed to autopkgtest, and --disable-apt-fallback
is NOT passed (default), fallback logic will be invoked if an upgrade
calculation cannot be calculated during testbed setup. This can happen
e.g. testing with a package trigger from experimental.
Unfortunately, in tests-in-lxd, even though we copy in the exact
/etc/apt configs, the fallback logic is not triggered because we didn't
use --pin-packages. This means that it's possible for tests-in-lxd to
fail due to the upgrade issue, even though the same tests in the normal
testbed would succeed after the fallback.
Add a hack to re-construct the --pin-packages arguments based on the
content of /etc/apt/preferences.d/autopkgtest-*.pref.
Gbp-Dch: Short
commit 6b77249c71d5758585a577bcb67273bbe18583e4
Author: Nick Rosbrook <enr0n@ubuntu.com>
Date: Thu Dec 4 12:37:07 2025 -0500
d/extra/dbus-1: rename systemd-localed-read-only.conf
It is unclear why exactly, but dbus-daemon and dbus-broker appear to
handle configuration precedence, parsing, etc. differently. With
dbus-daemon, the current filename results in this configuration being
ignored, whereas dbus-broker applies it correctly.
Renaming the file to org.freedesktop.locale1.read-only.conf, however,
appears to work with both implementations.
LP: #2133889
Gbp-Dch: Short
commit 819831c19af2a29ea775d02973e82923c06c8289
Author: Helmut Grohne <helmut@subdivi.de>
Date: Sat Nov 29 12:29:24 2025 +0100
Update architecture match for 50-pid-max.conf
In theory we would like to match any 64bit CPU, but architecture
patterns cannot. Use CPU matches and add all CPUs known to dpkg.
commit 0ddff89e9d531ccae284fa9c20b6c24ca27f78aa
Author: Helmut Grohne <helmut@subdivi.de>
Date: Sat Nov 29 12:38:05 2025 +0100
Mirror dmi_arches from meson.build into debian/udev.install
The relevant files are installed for any architecture matched by by
the dmi_arches variable in the upstream meson.build. It does not care
about the libc or kernel, so we start using any patterns here. Notably,
mips was the only architecture missing in the udev.rules. x32 is implied
by any-amd64 and loong64 was duplicated. armel and armhf are both
matches by any-arm as the suffix is part of the libc ABI earlier in the
tuple.
commit 398e8791db99f9bbada83af23662ac0a5c7ae913
Author: Luca Boccassi <luca.boccassi@gmail.com>
Date: Thu Nov 27 17:49:39 2025 +0000
d/t/control: pull in optional libs for boot-and-services too