]> git.ipfire.org Git - thirdparty/systemd.git/log
thirdparty/systemd.git
6 weeks agoresolved: always progress DS queries
Ronan Pigott [Mon, 29 Apr 2024 09:17:23 +0000 (02:17 -0700)] 
resolved: always progress DS queries

If we request a DS and the resolver offers an unsigned SOA, a new
auxiliary transaction for the DS will be rejected as a loop, and we
might not make any progress toward finding the DS we need. Let's ensure
that we at least always check the parent in this case.

Fixes: 47690634f157 ("resolved: don't request the SOA for every dns label")
6 weeks agomeson: copy prefix mapping CFLAGS when building BPF objects
Luca Boccassi [Sat, 27 Apr 2024 16:55:27 +0000 (17:55 +0100)] 
meson: copy prefix mapping CFLAGS when building BPF objects

Otherwise the filenames will contain variable paths and break reproducibility

6 weeks agomeson: define 's390' for 's390x' when building BPF objects
Luca Boccassi [Sat, 27 Apr 2024 17:06:42 +0000 (18:06 +0100)] 
meson: define 's390' for 's390x' when building BPF objects

The kernel headers match on __s390__ so the build fails

../src/nsresourced/bpf/userns_restrict/userns-restrict.bpf.c:159:6: error: Must specify a BPF target arch via __TARGET_ARCH_xxx
void BPF_KPROBE(userns_restrict_free_user_ns, struct work_struct *work) {
     ^
/usr/include/bpf/bpf_tracing.h:817:20: note: expanded from macro 'BPF_KPROBE'
        return ____##name(___bpf_kprobe_args(args));                        \
                          ^
/usr/include/bpf/bpf_tracing.h:797:41: note: expanded from macro '___bpf_kprobe_args'
                                        ^
/usr/include/bpf/bpf_helpers.h:195:29: note: expanded from macro '___bpf_apply'
                            ^
note: (skipping 2 expansions in backtrace; use -fmacro-backtrace-limit=0 to see all)
/usr/include/bpf/bpf_tracing.h:789:72: note: expanded from macro '___bpf_kprobe_args1'
                                                                       ^
/usr/include/bpf/bpf_tracing.h:563:29: note: expanded from macro 'PT_REGS_PARM1'
                            ^
<scratch space>:125:6: note: expanded from here
 GCC error "Must specify a BPF target arch via __TARGET_ARCH_xxx"

6 weeks agoMerge pull request #32547 from YHNdnzj/minor-cleanup
Mike Yuan [Mon, 29 Apr 2024 13:08:06 +0000 (21:08 +0800)] 
Merge pull request #32547 from YHNdnzj/minor-cleanup

Some cleanups prompted during review

6 weeks agoboot: fix argument name mismatch in two functions
Antonio Alvarez Feijoo [Mon, 29 Apr 2024 10:22:49 +0000 (12:22 +0200)] 
boot: fix argument name mismatch in two functions

6 weeks agocore: Add systemd.crash_action= kernel command line argument
Daan De Meyer [Mon, 29 Apr 2024 06:46:58 +0000 (08:46 +0200)] 
core: Add systemd.crash_action= kernel command line argument

Required for integration tests to power off on PID 1 crashes. We
deprecate systemd.crash_reboot and related options by removing them
from the documentation but still parsing them.

6 weeks agojournalctl: also check arg_file_stdin with other journal location
Mike Yuan [Mon, 29 Apr 2024 08:47:11 +0000 (16:47 +0800)] 
journalctl: also check arg_file_stdin with other journal location
options

Prompted by #32491

6 weeks agodocs/DEBUGGING.md: use an underscore in the kernel command line option
nl6720 [Mon, 29 Apr 2024 11:39:14 +0000 (14:39 +0300)] 
docs/DEBUGGING.md: use an underscore in the kernel command line option

systemd-debug-generator(8) lists it as `systemd.debug_shell`.

According to 1d84ad944520fc3e062ef518c4db4e1d3a1866af, kernel command
line options should use an underscore instead of a dash.

6 weeks agocore: Pass NULL error in dump_impl()
Daan De Meyer [Sun, 28 Apr 2024 17:42:51 +0000 (19:42 +0200)] 
core: Pass NULL error in dump_impl()

If mac_selinux_access_check() or bus_verify_bypass_dump_ratelimit_async()
fail, we goto "ratelimited" where we set a custom D-BUS error. In
"ratelimited", we call sd_bus_error_setf() which eventually hits an
assert_return(!bus_error_is_dirty()). Avoid hitting this assertion by
passing NULL as the error to mac_selinux_access_check() and
bus_verify_bypass_dump_ratelimit_async() since we will override the error
immediately anyway if either fails.

We modify both functions as well to allow passing a NULL error and fix
the argument name as well while we're at it.

6 weeks agoMerge pull request #32556 from YHNdnzj/development-freeze
Frantisek Sumsal [Mon, 29 Apr 2024 11:46:41 +0000 (13:46 +0200)] 
Merge pull request #32556 from YHNdnzj/development-freeze

development-freeze: suppress warning for some labels

6 weeks agobuild(deps): bump pkg/debian from `4b1f868` to `640ff73`
dependabot[bot] [Mon, 29 Apr 2024 10:00:20 +0000 (10:00 +0000)] 
build(deps): bump pkg/debian from `4b1f868` to `640ff73`

Bumps pkg/debian from `4b1f868` to `640ff73`.

---
updated-dependencies:
- dependency-name: pkg/debian
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
6 weeks agodevelopment-freeze: suppress warning for some labels 32556/head
Mike Yuan [Mon, 29 Apr 2024 10:59:02 +0000 (18:59 +0800)] 
development-freeze: suppress warning for some labels

6 weeks agolabeler: add l10n 🌍
Mike Yuan [Mon, 29 Apr 2024 10:56:54 +0000 (18:56 +0800)] 
labeler: add l10n 🌍

6 weeks agopam: Setup logging to syslog
Daan De Meyer [Sun, 28 Apr 2024 19:27:23 +0000 (21:27 +0200)] 
pam: Setup logging to syslog

We already log to syslog using pam_syslog() for logs generated directly
within our pam plugins. However, any logs generated by our generic logging
macros that are invoked within a pam plugin will log to the console. Let's
make sure our generic logging macros are set up to log to syslog as well.

6 weeks agopo: Translated using Weblate (French)
Pierre GRASSER [Mon, 29 Apr 2024 08:36:18 +0000 (10:36 +0200)] 
po: Translated using Weblate (French)

Currently translated at 100.0% (233 of 233 strings)

Co-authored-by: Pierre GRASSER <pierre.grasser@proton.me>
Translate-URL: https://translate.fedoraproject.org/projects/systemd/main/fr/
Translation: systemd/main

6 weeks agoNEWS: fix minimum dracut version required for systemd v256
Antonio Alvarez Feijoo [Mon, 29 Apr 2024 06:55:36 +0000 (08:55 +0200)] 
NEWS: fix minimum dracut version required for systemd v256

Follow-up for d4e9be97a3

6 weeks agofs-util: modernize touch_file a bit 32547/head
Mike Yuan [Mon, 29 Apr 2024 08:23:00 +0000 (16:23 +0800)] 
fs-util: modernize touch_file a bit

6 weeks agofs-util: try AT_EMPTY_PATH first for futimens_opath
Mike Yuan [Mon, 29 Apr 2024 08:22:37 +0000 (16:22 +0800)] 
fs-util: try AT_EMPTY_PATH first for futimens_opath

6 weeks agoefivars: drop unneeded UTIME_NOW
Mike Yuan [Mon, 29 Apr 2024 07:45:14 +0000 (15:45 +0800)] 
efivars: drop unneeded UTIME_NOW

futimens(2) defaults to current time if 'times' is NULL.

6 weeks agopam_systemd_loadkey: add missing PAM_EXTERN
Mike Yuan [Mon, 29 Apr 2024 07:46:32 +0000 (15:46 +0800)] 
pam_systemd_loadkey: add missing PAM_EXTERN

6 weeks agomkosi: Ignore version from versioned dependencies in .SRCINFO
Daan De Meyer [Sun, 28 Apr 2024 13:21:15 +0000 (15:21 +0200)] 
mkosi: Ignore version from versioned dependencies in .SRCINFO

Dependencies in .SRCINFO can be versioned. Let's make sure we ignore
any specified versions when grepping it for dependencies. Also update
the arch submodule to the latest to make sure the change works.

6 weeks agovarious: drop unnecessary DISABLE_WARNING_FORMAT_NONLITERAL
Mike Yuan [Sun, 28 Apr 2024 07:59:36 +0000 (15:59 +0800)] 
various: drop unnecessary DISABLE_WARNING_FORMAT_NONLITERAL

We use _printf_ for these functions, so there shouldn't
be a warning in the first place.

6 weeks agohwdb: Add a common Logitech M185/M225 mouse variant
Maciej S. Szmigiero [Sun, 28 Apr 2024 08:51:57 +0000 (10:51 +0200)] 
hwdb: Add a common Logitech M185/M225 mouse variant

Measured/tested on a Logitech N1901 mouse that apparently identifies as
one.

6 weeks agodocs: update link for SUSE whitepaper
spdfnet [Sun, 28 Apr 2024 09:10:46 +0000 (10:10 +0100)] 
docs: update link for SUSE whitepaper

Signed-off-by: spdfnet <32593931+spdfnet@users.noreply.github.com>
6 weeks agonetwork: DHCP version logging typos
Sebastian Pucilowski [Sun, 28 Apr 2024 00:34:02 +0000 (10:34 +1000)] 
network:  DHCP version logging typos

Some DHCP client log messages report "DHCP4" or "DHCP6" instead of
"DHCPv4" or "DHCPv6" as used within the rest of the codebase. Typos
fixed.

6 weeks agoNEWS: add warnings about read-only fs and libkmod being dlopen'ed
Zbigniew Jędrzejewski-Szmek [Sat, 27 Apr 2024 18:44:15 +0000 (20:44 +0200)] 
NEWS: add warnings about read-only fs and libkmod being dlopen'ed

Closes https://github.com/systemd/systemd/issues/32511.

6 weeks agoMerge pull request #32520 from YHNdnzj/sd-daemon-followup
Luca Boccassi [Sat, 27 Apr 2024 13:22:35 +0000 (15:22 +0200)] 
Merge pull request #32520 from YHNdnzj/sd-daemon-followup

sd-daemon: minor cleanup and follow-up

6 weeks agoMerge pull request #32516 from YHNdnzj/core-cleanup
Mike Yuan [Sat, 27 Apr 2024 11:43:27 +0000 (19:43 +0800)] 
Merge pull request #32516 from YHNdnzj/core-cleanup

core: several cleanups

6 weeks agosd-daemon: downgrade log level for library code, use correct errno 32520/head
Mike Yuan [Sat, 27 Apr 2024 11:34:45 +0000 (19:34 +0800)] 
sd-daemon: downgrade log level for library code, use correct errno

Follow-up for 13b67b61b3b4a5356f5d1b29b51137b8e336aa55

6 weeks agosd-daemon: minor modernization, use assert_return
Mike Yuan [Sat, 27 Apr 2024 11:26:49 +0000 (19:26 +0800)] 
sd-daemon: minor modernization, use assert_return

6 weeks agonetworkd: Correct documentation for LinkLocalAddressing
Mathias Lang [Thu, 25 Apr 2024 10:29:53 +0000 (12:29 +0200)] 
networkd: Correct documentation for LinkLocalAddressing

LinkLocalAddressing accepts a boolean. This can be seen by looking at
`link_local_address_family_from_strong(cont char *s)` in
`src/network/netword-util.c#L102-108` which falls back to
`address_family_from_string`, defined two lines above (L100)
using `DEFINE_STRING_TABLE_LOOKUP_WITH_BOOLEAN`.

6 weeks agopo: Translated using Weblate (Polish)
Piotr Drąg [Sat, 27 Apr 2024 09:23:10 +0000 (11:23 +0200)] 
po: Translated using Weblate (Polish)

Currently translated at 100.0% (233 of 233 strings)

Co-authored-by: Piotr Drąg <piotrdrag@gmail.com>
Translate-URL: https://translate.fedoraproject.org/projects/systemd/main/pl/
Translation: systemd/main

6 weeks agotreewide: fix a few typos in NEWS, docs, comments, and log messages
Dmitry V. Levin [Sat, 27 Apr 2024 08:00:00 +0000 (08:00 +0000)] 
treewide: fix a few typos in NEWS, docs, comments, and log messages

6 weeks agocore/namespace: modernize mount_private_dev 32516/head
Mike Yuan [Sat, 27 Apr 2024 06:25:36 +0000 (14:25 +0800)] 
core/namespace: modernize mount_private_dev

6 weeks agocore/namespace: modernize clone_device_node
Mike Yuan [Sat, 27 Apr 2024 08:46:14 +0000 (16:46 +0800)] 
core/namespace: modernize clone_device_node

6 weeks agoFix spelling errors found by Lintian
Luca Boccassi [Fri, 26 Apr 2024 15:58:00 +0000 (16:58 +0100)] 
Fix spelling errors found by Lintian

6 weeks agocore/namespace: use FOREACH_ARRAY at one more place
Mike Yuan [Fri, 26 Apr 2024 03:07:59 +0000 (11:07 +0800)] 
core/namespace: use FOREACH_ARRAY at one more place

6 weeks agocore/execute: switch mount_apivfs to tristate
Mike Yuan [Thu, 25 Apr 2024 17:44:15 +0000 (01:44 +0800)] 
core/execute: switch mount_apivfs to tristate

No functional change, just refactoring.

6 weeks agocore/manager-serialize: deserialize handoff-timestamp fds
Mike Yuan [Sat, 27 Apr 2024 06:08:10 +0000 (14:08 +0800)] 
core/manager-serialize: deserialize handoff-timestamp fds

Follow-up for 817062e6211e7559ff91bcb7bb5dc4e472fc1646

Fixes #32512

6 weeks agocore: use close_and_replace more when deserializing
Mike Yuan [Sat, 27 Apr 2024 06:12:53 +0000 (14:12 +0800)] 
core: use close_and_replace more when deserializing

Addresses https://github.com/systemd/systemd/pull/32441#discussion_r1579149873

6 weeks agoman/systemd.exec: mount_switch_root uses pivot_root rather than chroot
Mike Yuan [Fri, 26 Apr 2024 09:48:53 +0000 (17:48 +0800)] 
man/systemd.exec: mount_switch_root uses pivot_root rather than chroot

6 weeks agocore/load-fragment: add one missing assertion
Mike Yuan [Fri, 26 Apr 2024 09:44:32 +0000 (17:44 +0800)] 
core/load-fragment: add one missing assertion

6 weeks agounits: add Before=shutdown.target to systemd-networkd-persistent-storage.service
Luca Boccassi [Fri, 26 Apr 2024 17:43:01 +0000 (18:43 +0100)] 
units: add Before=shutdown.target to systemd-networkd-persistent-storage.service

It's ordered with networkd, but just in case. Lintian complains:

W: systemd: systemd-service-file-shutdown-problems [usr/lib/systemd/system/systemd-networkd-persistent-storage.service]

Follow-up for 91676b645886382328c261efd5e83490d16e75cf

6 weeks agoMerge pull request #32510 from arthurzam/bash-v256
Luca Boccassi [Fri, 26 Apr 2024 20:16:09 +0000 (22:16 +0200)] 
Merge pull request #32510 from arthurzam/bash-v256

shell-completions: add bash completion for run0 & systemd-vpick

6 weeks agoMerge pull request #32439 from CodethinkLabs/simple-mkosi-integration-tests
Daan De Meyer [Fri, 26 Apr 2024 18:48:00 +0000 (20:48 +0200)] 
Merge pull request #32439 from CodethinkLabs/simple-mkosi-integration-tests

Enable some simple mkosi integration tests

6 weeks agoMerge pull request #32504 from bluca/test_failure
Luca Boccassi [Fri, 26 Apr 2024 17:55:24 +0000 (19:55 +0200)] 
Merge pull request #32504 from bluca/test_failure

test: do not fail if mknod() fails in a build system

6 weeks agoMerge pull request #32505 from mrc0mmand/fix-build-without-importd
Luca Boccassi [Fri, 26 Apr 2024 17:35:35 +0000 (19:35 +0200)] 
Merge pull request #32505 from mrc0mmand/fix-build-without-importd

import: skip the whole subdir when building with -Dimportd=false

6 weeks agotest/meson.build: Add some runnable integration tests 32439/head
Richard Maw [Thu, 18 Apr 2024 09:39:00 +0000 (10:39 +0100)] 
test/meson.build: Add some runnable integration tests

6 weeks agoTEST-81-GENERATORS: Clean /proc/cmdline of unusual mounts
Richard Maw [Sat, 24 Feb 2024 18:13:59 +0000 (18:13 +0000)] 
TEST-81-GENERATORS: Clean /proc/cmdline of unusual mounts

6 weeks agoTEST-36-NUMAPOLICY: Bump PID1 strace time to 3 seconds
Richard Maw [Fri, 26 Apr 2024 11:54:34 +0000 (12:54 +0100)] 
TEST-36-NUMAPOLICY: Bump PID1 strace time to 3 seconds

CI can be slow and only waiting 1 second makes it flaky there.

6 weeks agoTEST-26-SYSTEMCTL: Create /etc/init.d if needed
Richard Maw [Wed, 24 Apr 2024 13:44:27 +0000 (14:44 +0100)] 
TEST-26-SYSTEMCTL: Create /etc/init.d if needed

OpenSUSE doesn't disable sysv compat but also may not have anything in
/etc/init.d.

6 weeks agoTEST-25-IMPORT: Skip if importctl not installed
Richard Maw [Wed, 24 Apr 2024 11:16:37 +0000 (12:16 +0100)] 
TEST-25-IMPORT: Skip if importctl not installed

machinectl import-* was split out into importctl
which is a separately configurable component to machinectl
and might not always be available.

6 weeks agoTEST-19-CGROUP: Skip on opensuse
Richard Maw [Thu, 25 Apr 2024 11:18:57 +0000 (12:18 +0100)] 
TEST-19-CGROUP: Skip on opensuse

OpenSUSE images seem to be unhappy with either how they are built
or what they are being asked to do.

The listed device-mapper failure is just one of the strange errors,
I have also seen it fail to propagate cgroup properties into new cgroups
that were previously guaranteed to exist.

6 weeks agoTEST-17-UDEV: Skip credentials test if service not loaded
Richard Maw [Tue, 23 Apr 2024 21:30:24 +0000 (22:30 +0100)] 
TEST-17-UDEV: Skip credentials test if service not loaded

It does not appear to exist on OpenSUSE yet.

6 weeks agotest: modernize test-dirent-util 32504/head
Luca Boccassi [Fri, 26 Apr 2024 16:09:04 +0000 (17:09 +0100)] 
test: modernize test-dirent-util

Follow-up for 6a57d86bf9d7f8d6f5d339f57665dfb2e1d191c5

6 weeks agotest: do not fail if mknod() fails in a build system
Luca Boccassi [Fri, 26 Apr 2024 12:30:31 +0000 (13:30 +0100)] 
test: do not fail if mknod() fails in a build system

mknod fails in a Salsa container when building Debian packages, skip the check

Follow-up for 6a57d86bf9d7f8d6f5d339f57665dfb2e1d191c5

6 weeks agoshell-completion: add bash-completion for systemd-vpick command 32510/head
Arthur Zamarin [Fri, 26 Apr 2024 10:29:47 +0000 (13:29 +0300)] 
shell-completion: add bash-completion for systemd-vpick command

Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
6 weeks agoshell-completion: add bash-completion for run0 command
Arthur Zamarin [Fri, 26 Apr 2024 10:15:55 +0000 (13:15 +0300)] 
shell-completion: add bash-completion for run0 command

Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
6 weeks agoshell-completions: install new completions which were forgotten
Arthur Zamarin [Fri, 26 Apr 2024 15:43:38 +0000 (18:43 +0300)] 
shell-completions: install new completions which were forgotten

Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
6 weeks agoimport: skip the whole subdir when building with -Dimportd=false 32505/head
Frantisek Sumsal [Fri, 26 Apr 2024 15:07:39 +0000 (17:07 +0200)] 
import: skip the whole subdir when building with -Dimportd=false

Resolves: #32501

6 weeks agomountfsd: drop explicit dependencies for systemd-mountwork
Frantisek Sumsal [Fri, 26 Apr 2024 15:06:06 +0000 (17:06 +0200)] 
mountfsd: drop explicit dependencies for systemd-mountwork

As the executables are automagically linked against libshared.

6 weeks agotest: mountnfsd -> mountfsd
Frantisek Sumsal [Fri, 26 Apr 2024 14:53:43 +0000 (16:53 +0200)] 
test: mountnfsd -> mountfsd

I keep accidentally writing mountnfsd instead of mountfsd, and a couple
of instances managed to get into the repo.

6 weeks agomkosi: use generic ubuntu kernel for udev and nvme tests
Richard Maw [Tue, 23 Apr 2024 15:36:15 +0000 (16:36 +0100)] 
mkosi: use generic ubuntu kernel for udev and nvme tests

TEST-17-UDEV expects to find scsi_debug and TEST-84-STORAGETM expects to
find nvmet-tcp.

This isn't ideal as it adds firmware, microcode and other drivers to the
initramfs, but there's no linux-modules-extra virtual package
to just include the extra modules.

6 weeks agomkosi: Use default opensuse kernel for tests
Richard Maw [Tue, 23 Apr 2024 20:39:04 +0000 (21:39 +0100)] 
mkosi: Use default opensuse kernel for tests

The kvmsmall kernel lacks important modules like the dummy network
device which TEST-17-UDEV requires very early on.

6 weeks agotest: Disable network by default in mkosi tests
Richard Maw [Thu, 25 Apr 2024 09:00:15 +0000 (10:00 +0100)] 
test: Disable network by default in mkosi tests

6 weeks agoMerge pull request #32499 from mrc0mmand/docs-fix-method-name
Luca Boccassi [Fri, 26 Apr 2024 14:47:25 +0000 (16:47 +0200)] 
Merge pull request #32499 from mrc0mmand/docs-fix-method-name

docs: drop invalid links & fix method name

6 weeks agoMerge pull request #32498 from DaanDeMeyer/mkosi
Luca Boccassi [Fri, 26 Apr 2024 14:44:19 +0000 (16:44 +0200)] 
Merge pull request #32498 from DaanDeMeyer/mkosi

mkosi: Set CleanPackageMetadata=yes for minimal-base

6 weeks agosd-daemon: Replace SO_LINGER with shutdown() + recv() 32498/head
Daan De Meyer [Fri, 26 Apr 2024 13:02:56 +0000 (15:02 +0200)] 
sd-daemon: Replace SO_LINGER with shutdown() + recv()

Let's shutdown the write end and wait for EOF from the other side
before continuing to make sure that the receiver has received all
data we sent on the socket.

6 weeks agodocs: PrepareForSuspend() is not a valid signal name 32499/head
Frantisek Sumsal [Fri, 26 Apr 2024 12:27:03 +0000 (14:27 +0200)] 
docs: PrepareForSuspend() is not a valid signal name

6 weeks agodocs: drop invalid links
Frantisek Sumsal [Fri, 26 Apr 2024 12:45:52 +0000 (14:45 +0200)] 
docs: drop invalid links

6 weeks agoTEST-50-DISSECT: Don't log image contents to console
Daan De Meyer [Fri, 26 Apr 2024 11:51:49 +0000 (13:51 +0200)] 
TEST-50-DISSECT: Don't log image contents to console

These can contain quite a few files, so let's not write every single
file and directory straight to the console to speed things up.

6 weeks agomkosi: Set CleanPackageMetadata=yes for minimal-base
Daan De Meyer [Fri, 26 Apr 2024 11:46:45 +0000 (13:46 +0200)] 
mkosi: Set CleanPackageMetadata=yes for minimal-base

We don't want to install any more packages in derived images so let's
not store repository metadata in the image to make it smaller.

6 weeks agomkosi: Update to latest
Daan De Meyer [Fri, 26 Apr 2024 11:11:37 +0000 (13:11 +0200)] 
mkosi: Update to latest

6 weeks agoNEWS: add back deprecation notices that have not been implemented yet
Luca Boccassi [Fri, 26 Apr 2024 09:43:46 +0000 (10:43 +0100)] 
NEWS: add back deprecation notices that have not been implemented yet

6 weeks agoMerge pull request #32494 from arthurzam/bash-importctl
Luca Boccassi [Fri, 26 Apr 2024 09:33:37 +0000 (11:33 +0200)] 
Merge pull request #32494 from arthurzam/bash-importctl

shell completion: add bash completion for importctl & fix machinectl

6 weeks agoshell completion: fix machinectl import-{tar,raw} 32494/head
Arthur Zamarin [Fri, 26 Apr 2024 09:10:26 +0000 (12:10 +0300)] 
shell completion: fix machinectl import-{tar,raw}

Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
6 weeks agoshell completion: add bash completion for importctl
Arthur Zamarin [Fri, 26 Apr 2024 08:59:58 +0000 (11:59 +0300)] 
shell completion: add bash completion for importctl

Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
6 weeks agopo: Translated using Weblate (Korean)
김인수 [Fri, 26 Apr 2024 08:36:03 +0000 (10:36 +0200)] 
po: Translated using Weblate (Korean)

Currently translated at 100.0% (233 of 233 strings)

Co-authored-by: 김인수 <simmon@nplob.com>
Translate-URL: https://translate.fedoraproject.org/projects/systemd/main/ko/
Translation: systemd/main

6 weeks agoMerge pull request #32490 from YHNdnzj/namespace-cleanup
Daan De Meyer [Fri, 26 Apr 2024 06:29:32 +0000 (08:29 +0200)] 
Merge pull request #32490 from YHNdnzj/namespace-cleanup

core: trivial cleanups for setup_namespace

7 weeks agocore/exec-invoke: use FOREACH_ARRAY more, drop unnecessary initialization 32490/head
Mike Yuan [Sun, 21 Apr 2024 16:03:53 +0000 (00:03 +0800)] 
core/exec-invoke: use FOREACH_ARRAY more, drop unnecessary initialization
for some fields

7 weeks agocore/namespace: check error first, add missing assertion
Mike Yuan [Sun, 21 Apr 2024 15:32:01 +0000 (23:32 +0800)] 
core/namespace: check error first, add missing assertion

7 weeks agomount-setup: fix typo
Mike Yuan [Sat, 20 Apr 2024 19:09:22 +0000 (03:09 +0800)] 
mount-setup: fix typo

7 weeks agoNEWS: finalize for v256~rc1 v256-rc1
Luca Boccassi [Thu, 25 Apr 2024 21:30:35 +0000 (22:30 +0100)] 
NEWS: finalize for v256~rc1

7 weeks agoMerge pull request #32488 from bluca/chores
Luca Boccassi [Thu, 25 Apr 2024 23:59:36 +0000 (01:59 +0200)] 
Merge pull request #32488 from bluca/chores

Chores for 256~rc1

7 weeks agosd-radv: remove unnecessary check
Luca Boccassi [Thu, 25 Apr 2024 15:18:08 +0000 (16:18 +0100)] 
sd-radv: remove unnecessary check

in6_addr_prefix_intersect() returns a bool, so the check < 0 is
dead code. Flagged by codeql.

Follow-up for c7c56724a5d7cc2b6b0d9d5f95b1055811014843

7 weeks agomeson: update version numbers for 256~rc1 32488/head
Luca Boccassi [Thu, 25 Apr 2024 21:26:08 +0000 (22:26 +0100)] 
meson: update version numbers for 256~rc1

7 weeks agoUpdate hwdb
Luca Boccassi [Thu, 25 Apr 2024 21:23:52 +0000 (22:23 +0100)] 
Update hwdb

7 weeks agoNEWS: update contributors list
Luca Boccassi [Thu, 25 Apr 2024 21:21:51 +0000 (22:21 +0100)] 
NEWS: update contributors list

7 weeks agovarlink: make errors returned by verify_unix_socket() systematic
Lennart Poettering [Thu, 25 Apr 2024 15:23:24 +0000 (17:23 +0200)] 
varlink: make errors returned by verify_unix_socket() systematic

Previously, if we encountered a non-socket fd we'd return ENOTSOCK the
first time, but the subsequent times we'd return ENOMEDIUM, due to
caching. Let's make sure we return the same errors all the the time.

7 weeks agoMerge pull request #32445 from DaanDeMeyer/mkosi-images
Daan De Meyer [Thu, 25 Apr 2024 19:59:23 +0000 (21:59 +0200)] 
Merge pull request #32445 from DaanDeMeyer/mkosi-images

mkosi: Build minimal images and enable related integration tests

7 weeks agosd-daemon: Set SO_LINGER on AF_VSOCK notify socket fds
Daan De Meyer [Thu, 25 Apr 2024 17:35:07 +0000 (19:35 +0200)] 
sd-daemon: Set SO_LINGER on AF_VSOCK notify socket fds

We're still seeing problems in mkosi CI where we lose notify messages
sent over vsock from virtual machines because the virtual machine shuts
down before the socket buffers are flushed.

Let's try setting SO_LINGER on vsock sockets so that the kernel waits
until the buffers are flushed when we close the file descriptor.

To make sure it works, let's get rid of our previous workaround that
sends EXIT_STATUS earlier to reduce the chance of it not arriving.
This way, if it doesn't work, we'll notice it immediately due to CI
failures.

7 weeks agoMerge pull request #32484 from DaanDeMeyer/mkosi-ci
Daan De Meyer [Thu, 25 Apr 2024 19:13:09 +0000 (21:13 +0200)] 
Merge pull request #32484 from DaanDeMeyer/mkosi-ci

Two mkosi CI improvements

7 weeks agomkosi: Build minimal images and enable related integration tests 32445/head
Daan De Meyer [Wed, 24 Apr 2024 19:21:34 +0000 (21:21 +0200)] 
mkosi: Build minimal images and enable related integration tests

This commit adds definitions to build the minimal_0 and minimal_1
images with mkosi and includes them into the system image. We also
move the building of the various app-xxx and similar images that are
extremely minimal into the tests itself by moving the related logic
from install_verity_minimal() into a new function
install_extension_images() in util.sh. Because the mkosi /usr is
read-only, we now place the extension images in /tmp instead of
/usr/share.

Co-authored-by: Richard Maw <richard.maw@codethink.co.uk>
Co-authored-by: sam-leonard-ct <sam.leonard@codethink.co.uk>
7 weeks agomkosi: Only keep failed test journals and other logs for 7 days 32484/head
Daan De Meyer [Thu, 25 Apr 2024 17:52:10 +0000 (19:52 +0200)] 
mkosi: Only keep failed test journals and other logs for 7 days

The default retention of 90 days seems a bit long, so drop it to 7
days.

7 weeks agomkosi: Only archive outputs in systemd and systemd-stable repositories
Daan De Meyer [Thu, 25 Apr 2024 17:50:38 +0000 (19:50 +0200)] 
mkosi: Only archive outputs in systemd and systemd-stable repositories

Private forks would very quickly reach their quota or spend lots of
money trying to archive all these artifacts, so let's make sure it
only happens on our own repositories.

7 weeks agoTEST-50-DISSECT: Make sure logging sockets are mounted into images
Daan De Meyer [Thu, 25 Apr 2024 14:58:44 +0000 (16:58 +0200)] 
TEST-50-DISSECT: Make sure logging sockets are mounted into images

Otherwise we lose valuable logging from systemd-executor when things
go wrong since it can only log to the journal and not to the console
in these cases.

7 weeks agoMerge pull request #32475 from bluca/skip
Luca Boccassi [Thu, 25 Apr 2024 16:48:57 +0000 (18:48 +0200)] 
Merge pull request #32475 from bluca/skip

test: skip testsuite-50.mountnfsd if kernel/polkit are too old

7 weeks agoFix _do_log conditions in log_exec_* macros
Łukasz Stelmach [Thu, 25 Apr 2024 14:10:49 +0000 (16:10 +0200)] 
Fix _do_log conditions in log_exec_* macros

The expression for calculation of the _do_log values in the log_exec_*
macros need to be the same as the unit_log_level_test() function, used
to calculate _do_log in unit.h. The only difference between execute.h
and unit.h is the lack of the Unit structure.

Fixes: b646fc324a ("core: ensure execute/spawn functions can work without Unit object")
Fixes: 210ca71cb5 ("core/execute: clean up log_exec_full_errno and friends")
Signed-off-by: Łukasz Stelmach <l.stelmach@samsung.com>
7 weeks agoMerge pull request #32477 from DaanDeMeyer/fixes
Daan De Meyer [Thu, 25 Apr 2024 16:24:09 +0000 (18:24 +0200)] 
Merge pull request #32477 from DaanDeMeyer/fixes

Various fixes

7 weeks agoMerge pull request #32474 from poettering/varlink-no-pidfd
Luca Boccassi [Thu, 25 Apr 2024 15:27:18 +0000 (17:27 +0200)] 
Merge pull request #32474 from poettering/varlink-no-pidfd

varlink: tweak fallback for SO_PEERPIDFD on old kernels