]> git.ipfire.org Git - thirdparty/systemd.git/log
thirdparty/systemd.git
18 months agomachine-id-setup: acquire machine ID from /run/machine-id if possible 32915/head
Yu Watanabe [Sun, 19 May 2024 00:14:27 +0000 (09:14 +0900)] 
machine-id-setup: acquire machine ID from /run/machine-id if possible

If machine ID is previously stored at /run/machine-id, then let's reuse
it. This is important on switching root and /etc/machine-id was previously
a mount point.

Fixes #32908.

18 months agomachine-id-setup: use empty_or_root()
Yu Watanabe [Sat, 18 May 2024 23:46:43 +0000 (08:46 +0900)] 
machine-id-setup: use empty_or_root()

18 months agoMerge pull request #32917 from YHNdnzj/ptyfwd-cleanup
Luca Boccassi [Sun, 19 May 2024 12:44:54 +0000 (14:44 +0200)] 
Merge pull request #32917 from YHNdnzj/ptyfwd-cleanup

ptyfwd: a few fixes/cleanups

18 months agotest: sync journal before read
Yu Watanabe [Sat, 18 May 2024 22:12:48 +0000 (07:12 +0900)] 
test: sync journal before read

Workaround for #32834 and #32890.

18 months agorun: pass the pty slave fd to transient service 32917/head
Mike Yuan [Sun, 19 May 2024 01:07:21 +0000 (09:07 +0800)] 
run: pass the pty slave fd to transient service

The rationale is similar to 40e1f4ea7458a0a80eaf1ef356e52bfe0835412e.

Currently, we only pass TTYPath=/dev/pts/... to
the transient service spawned by systemd-run.
This is a bit problematic though, when ExecStartPre=
or ExecStopPost= is used. Since when these control
processes get to run, the main process is not yet
started/has already exited, hence the slave suffers
from the same vhangup problem as the mentioned commit.

By passing the slave fd in, the service manager will
hold the fd open as long as the service is alive.

Fixes #32916

18 months agorun: do not log "Error on PTY forwarding logic" when disconnected due to user operation
Mike Yuan [Sun, 19 May 2024 00:53:07 +0000 (08:53 +0800)] 
run: do not log "Error on PTY forwarding logic" when disconnected due to user operation

18 months agovmspawn: use openpt_allocate where appropriate
Mike Yuan [Sat, 18 May 2024 23:59:24 +0000 (07:59 +0800)] 
vmspawn: use openpt_allocate where appropriate

18 months agoptyfwd: trivial coding style and log cleanups
Mike Yuan [Sun, 19 May 2024 01:16:56 +0000 (09:16 +0800)] 
ptyfwd: trivial coding style and log cleanups

18 months agoMerge pull request #32914 from yuwata/test-64-storage
Luca Boccassi [Sun, 19 May 2024 00:50:36 +0000 (02:50 +0200)] 
Merge pull request #32914 from yuwata/test-64-storage

test: several follow-ups for TEST-64-STORAGE

18 months agotest: wipe filesystem before moving to the next test case 32914/head
Yu Watanabe [Sat, 18 May 2024 20:10:28 +0000 (05:10 +0900)] 
test: wipe filesystem before moving to the next test case

18 months agotest: install modinfo to test image
Yu Watanabe [Sat, 18 May 2024 19:54:25 +0000 (04:54 +0900)] 
test: install modinfo to test image

Follow-up for 6c2d47d6d3ad25ffd7527c7f4de31457ee1b25d8.

Fixes the following unexpected skip:
```
[    6.163670] TEST-64-UDEV-STORAGE.sh[596]: + modinfo btrfs
[    6.164102] TEST-64-UDEV-STORAGE.sh[726]: /usr/lib/systemd/tests/testdata/units/TEST-64-UDEV-STORAGE.sh: line 726: modinfo: command not found
[    6.164683] TEST-64-UDEV-STORAGE.sh[727]: + echo 'This test requires the btrfs kernel module but it is not installed, skipping the test'
[    6.165069] TEST-64-UDEV-STORAGE.sh[728]: + tee --append /skipped
[    6.166801] TEST-64-UDEV-STORAGE.sh[728]: This test requires the btrfs kernel module but it is not installed, skipping the test
[    6.167177] TEST-64-UDEV-STORAGE.sh[596]: + exit 77
```

18 months agoMerge pull request #32902 from yuwata/emergency-action-fixes
Yu Watanabe [Sat, 18 May 2024 07:32:36 +0000 (16:32 +0900)] 
Merge pull request #32902 from yuwata/emergency-action-fixes

several fixes for emergency actions and document update

18 months agoMerge pull request #32909 from arthurzam/bash-v256
Yu Watanabe [Sat, 18 May 2024 07:30:03 +0000 (16:30 +0900)] 
Merge pull request #32909 from arthurzam/bash-v256

shell-completion: add more missing args for bash

18 months agoshell-completion: add systemd-vmspawn to meson.build 32909/head
Arthur Zamarin [Sat, 18 May 2024 06:38:53 +0000 (09:38 +0300)] 
shell-completion: add systemd-vmspawn to meson.build

Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
18 months agoshell-completion: add missing args to bash systemd-nspawn
Arthur Zamarin [Sat, 18 May 2024 06:34:38 +0000 (09:34 +0300)] 
shell-completion: add missing args to bash systemd-nspawn

Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
18 months agoshell-completion: add missing args to bash systemd-run
Arthur Zamarin [Sat, 18 May 2024 06:34:38 +0000 (09:34 +0300)] 
shell-completion: add missing args to bash systemd-run

Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
18 months agoinstall-file: make fs_make_very_read_only() static
Yu Watanabe [Fri, 17 May 2024 18:31:37 +0000 (03:31 +0900)] 
install-file: make fs_make_very_read_only() static

18 months agoman: refer FailureAction= and SuccessAction= for explaining allowed values in JobTime... 32902/head
Yu Watanabe [Fri, 17 May 2024 17:11:34 +0000 (02:11 +0900)] 
man: refer FailureAction= and SuccessAction= for explaining allowed values in JobTimeoutAction=

The allowed values are explained in FailureAction= and SuccessAction=,
rather than StartLimitAction=.

18 months agocore: refuse invalid emergency actions for SuccessAction= and friends in user service...
Yu Watanabe [Fri, 17 May 2024 17:07:34 +0000 (02:07 +0900)] 
core: refuse invalid emergency actions for SuccessAction= and friends in user service manager

Especially, soft-reboot is not supported by user service manager.

Fixes a bug in 13ffc60749df0ca7c76cfcac317b41a05679b364 and
3cf848f6cd2151ccff89073bffb102f966d64ced.

18 months agoman: mention soft-reboot in bootup(7)
Yu Watanabe [Fri, 17 May 2024 17:33:13 +0000 (02:33 +0900)] 
man: mention soft-reboot in bootup(7)

18 months agounits: add JobTimeoutAction= to exit.target and friends
Yu Watanabe [Fri, 17 May 2024 16:26:14 +0000 (01:26 +0900)] 
units: add JobTimeoutAction= to exit.target and friends

For consistency with other targets, e.g. poweroff.target or
reboot.target.

18 months agoman/soft-reboot: order surviving services before shutdown.target
Mike Yuan [Fri, 17 May 2024 13:07:17 +0000 (21:07 +0800)] 
man/soft-reboot: order surviving services before shutdown.target

Prompted by #32895

Rather than ordering with each power operation targets,
ordering against shutdown.target which is a valid
synchronization point. This has no effect if soft-reboot
is being performed.

18 months agoman: add note about selinux to rc-local
Zbigniew Jędrzejewski-Szmek [Fri, 17 May 2024 12:31:16 +0000 (14:31 +0200)] 
man: add note about selinux to rc-local

This feature is deprecated, but if users use it, they are likely to be
tripped up by the wrong selinux context, so add a note.

https://discussion.fedoraproject.org/t/systemd-rc-local-service-doesnt-work-on-fedora-40-until-selinux-contexts-are-set-on-rc-local-script

18 months agocore/execute: do not use format str for log_object_internal
Mike Yuan [Fri, 17 May 2024 09:35:43 +0000 (17:35 +0800)] 
core/execute: do not use format str for log_object_internal

This reverts part of 210ca71cb5d9f41df3b23eef28d6742bd90a0362.

Another fallout from that commit...

Fixes #32877

18 months agomountpoint-util: Deal with kernel API breakage in "norecovery" mount option
Daan De Meyer [Fri, 17 May 2024 08:46:12 +0000 (10:46 +0200)] 
mountpoint-util: Deal with kernel API breakage in "norecovery" mount option

"norecovery" was deprecated for btrfs in
https://github.com/torvalds/linux/commit/74ef00185eb864252156022ff129b01549504175
and removed in
https://github.com/torvalds/linux/commit/a1912f712188291f9d7d434fba155461f1ebef66.

Let's drop our assumption that btrfs supports "norecovery" and first query for the
new name of the option followed by querying for the old name.

18 months agotest: wait for underlying .device unit being active before invoking systemd-mount
Yu Watanabe [Fri, 17 May 2024 06:04:31 +0000 (15:04 +0900)] 
test: wait for underlying .device unit being active before invoking systemd-mount

Fixes following failure:
===
May 17 04:12:04 TEST-74-AUX-UTILS.sh[2684]: + systemd-mount --owner=testuser /dev/loop0 /tmp/tmp.DVQdo2ou53/mnt
(snip)
May 17 04:15:04 systemd[1]: dev-loop0.device: Job dev-loop0.device/start timed out.
May 17 04:15:04 systemd[1]: dev-loop0.device: Job 5812 dev-loop0.device/start finished, result=timeout
May 17 04:15:04 systemd[1]: Timed out waiting for device dev-loop0.device - /dev/loop0.
May 17 04:15:04 systemd[1]: tmp-tmp.DVQdo2ou53-mnt.mount: Job 5804 tmp-tmp.DVQdo2ou53-mnt.mount/start finished, result=dependency
May 17 04:15:04 systemd[1]: Dependency failed for tmp-tmp.DVQdo2ou53-mnt.mount - /tmp/tmp.DVQdo2ou53/mnt.
May 17 04:15:04 systemd[1]: tmp-tmp.DVQdo2ou53-mnt.mount: Job tmp-tmp.DVQdo2ou53-mnt.mount/start failed with result 'dependency'.
May 17 04:15:04 systemd[1]: systemd-fsck@dev-loop0.service: Job 5805 systemd-fsck@dev-loop0.service/start finished, result=dependency
May 17 04:15:04 systemd[1]: Dependency failed for systemd-fsck@dev-loop0.service - File System Check on /dev/loop0.
May 17 04:15:04 systemd[1]: systemd-fsck@dev-loop0.service: Job systemd-fsck@dev-loop0.service/start failed with result 'dependency'.
May 17 04:15:04 systemd[1]: dev-loop0.device: Job dev-loop0.device/start failed with result 'timeout'.
(snip)
May 17 04:15:04 systemd-mount[2856]: A dependency job for tmp-tmp.DVQdo2ou53-mnt.mount failed. See 'journalctl -xe' for details.

18 months agotest: dynamically generate list of test cases
Yu Watanabe [Fri, 17 May 2024 01:04:37 +0000 (10:04 +0900)] 
test: dynamically generate list of test cases

Follow-up for #32666.

18 months agotest: wait for sessions being closed
Yu Watanabe [Fri, 17 May 2024 05:00:12 +0000 (14:00 +0900)] 
test: wait for sessions being closed

If a session in closing state, the user state will be in online.

Fixes #32698.

18 months agoMerge pull request #32875 from yuwata/network-route-wireguard
Daan De Meyer [Fri, 17 May 2024 07:10:33 +0000 (09:10 +0200)] 
Merge pull request #32875 from yuwata/network-route-wireguard

network/route: fix unexpected removal of routes for wireguard

18 months agotest: wait for partition device being processed by udevd
Yu Watanabe [Fri, 17 May 2024 05:21:44 +0000 (14:21 +0900)] 
test: wait for partition device being processed by udevd

Fixes #32697.

18 months agotest: wait for slice unit being (de)activated
Yu Watanabe [Fri, 17 May 2024 04:25:25 +0000 (13:25 +0900)] 
test: wait for slice unit being (de)activated

Fixes #32731.

18 months agoMerge pull request #32885 from yuwata/test-do-not-fill-journal
Daan De Meyer [Fri, 17 May 2024 06:54:24 +0000 (08:54 +0200)] 
Merge pull request #32885 from yuwata/test-do-not-fill-journal

test: do not fill journal with garbage

18 months agoMerge pull request #32880 from yuwata/unit-soft-reboot
Mike Yuan [Fri, 17 May 2024 06:06:44 +0000 (14:06 +0800)] 
Merge pull request #32880 from yuwata/unit-soft-reboot

units: do not soft-reboot before soft-reboot.target reached

18 months agotest: sync journal before reading journal
Yu Watanabe [Fri, 17 May 2024 04:03:44 +0000 (13:03 +0900)] 
test: sync journal before reading journal

Otherwise, expected lines may not be processed or not sync()ed to disk.

Fixes #32712.

18 months agotest: wait for partition processed by udevd
Yu Watanabe [Fri, 17 May 2024 01:12:25 +0000 (10:12 +0900)] 
test: wait for partition processed by udevd

Fixes #32695.

18 months agohwdb: Enable JP-IK LEAP W502's touchpad toggle key
Jian-Hong Pan [Fri, 17 May 2024 03:32:30 +0000 (11:32 +0800)] 
hwdb: Enable JP-IK LEAP W502's touchpad toggle key

The laptop JP-IK LEAP W502 has touchpad toggle key (Fn+F9), but it does
not work. Because, the scancode maps to a wrong key code:

Event: time 1715846095.224900, type 4 (EV_MSC), code 4 (MSC_SCAN), value 9d
Event: time 1715846095.224900, type 1 (EV_KEY), code 97 (KEY_RIGHTCTRL), value 1
Event: time 1715846095.224900, -------------- SYN_REPORT ------------
Event: time 1715846095.230985, type 4 (EV_MSC), code 4 (MSC_SCAN), value db
Event: time 1715846095.230985, type 1 (EV_KEY), code 125 (KEY_LEFTMETA), value 1
Event: time 1715846095.230985, -------------- SYN_REPORT ------------
Event: time 1715846095.232903, type 4 (EV_MSC), code 4 (MSC_SCAN), value 76
Event: time 1715846095.232903, type 1 (EV_KEY), code 85 (KEY_ZENKAKUHANKAKU), value 1
Event: time 1715846095.232903, -------------- SYN_REPORT ------------

Map the scancode 76 to KEY_F21 to enable the touchpad toggle key.

18 months agotest: do not fill journal with diff 32885/head
Yu Watanabe [Fri, 17 May 2024 05:15:51 +0000 (14:15 +0900)] 
test: do not fill journal with diff

18 months agotest: do not fill journal with "wait"
Yu Watanabe [Fri, 17 May 2024 05:14:00 +0000 (14:14 +0900)] 
test: do not fill journal with "wait"

18 months agounits: do not soft-reboot before soft-reboot.target reached 32880/head
Yu Watanabe [Fri, 17 May 2024 02:38:16 +0000 (11:38 +0900)] 
units: do not soft-reboot before soft-reboot.target reached

Otherwise, at the time systemd-soft-reboot.service succeeds,
services which has Conflicts= and Before=soft-reboot.target may
not be stopped yet, and may be SIGKILLed.

Especially, systemd-journald.service has the dependencies, thus
journal may be corrupted. See #32223.

Follow-up for 13ffc60749df0ca7c76cfcac317b41a05679b364.

Fixes #32834.

18 months agounits: drop dependencies of soft-reboot.target from systemd-journald@.service
Yu Watanabe [Fri, 17 May 2024 02:50:44 +0000 (11:50 +0900)] 
units: drop dependencies of soft-reboot.target from systemd-journald@.service

The service deos not have DefaultDependencies=no. Hence it has dependencies
of shutdown.target, and dependencies of soft-reboot.target are not
necessary.

Follow-up for f89985ca494b79b2beed47e1f10d46ef2d59ce3e.

18 months agotest-network: use different destination from gateway 32875/head
Yu Watanabe [Fri, 17 May 2024 00:28:46 +0000 (09:28 +0900)] 
test-network: use different destination from gateway

Previously, one of the test route has the same address in destination
and gateway. Even it is a test case, that's super spurious. Let's use a
different address.

18 months agonetwork/route: fix unexpected removal of routes for wireguard
Yu Watanabe [Thu, 16 May 2024 21:40:40 +0000 (06:40 +0900)] 
network/route: fix unexpected removal of routes for wireguard

Fixes a bug introduced by 8d01e44c1f0e00b414d36bd1b46ecff548242208.

If a .netdev file for a wireguard interface requests to configure
routes for the interface, the routes were removed during configuring
another interface.

Fixes #32859.

18 months agorepart: remove duplicate word in --help
Zbigniew Jędrzejewski-Szmek [Sat, 11 May 2024 11:10:14 +0000 (13:10 +0200)] 
repart: remove duplicate word in --help

My bad; follow-up for 4955d2e36706bba9ba6595b0c5babb8f17f13465.

18 months agotest: Enable TEST-69-SHUTDOWN for mkosi
Daan De Meyer [Thu, 16 May 2024 10:28:51 +0000 (12:28 +0200)] 
test: Enable TEST-69-SHUTDOWN for mkosi

In mkosi, we run the test inside the VM instead of outside. To simplify
the implementation we drop the reboot part and only verify that we can
schedule and cancel shutdowns and that the wall messages are sent as
expected.

18 months agoRevert "CI: add manual workflow to publish pages to fix submodule issue"
Luca Boccassi [Thu, 16 May 2024 14:31:52 +0000 (15:31 +0100)] 
Revert "CI: add manual workflow to publish pages to fix submodule issue"

Nope, it's broken, never mind

This reverts commit b0f9c3c6483eadd630b0a6a4de23db761e8158c8.

18 months agoCI: add manual workflow to publish pages to fix submodule issue
Luca Boccassi [Thu, 16 May 2024 14:19:32 +0000 (15:19 +0100)] 
CI: add manual workflow to publish pages to fix submodule issue

Cloning the fedora/centos submodules fails because --depth 1 is used. Fork the actions workflow and use fetch-depth: 0 to disable it.

18 months agoMerge pull request #32863 from YHNdnzj/switch-root-followup
Luca Boccassi [Thu, 16 May 2024 13:53:18 +0000 (15:53 +0200)] 
Merge pull request #32863 from YHNdnzj/switch-root-followup

Two minor followups

18 months agoMerge pull request #32842 from DaanDeMeyer/cryptsetup
Daan De Meyer [Thu, 16 May 2024 13:50:55 +0000 (15:50 +0200)] 
Merge pull request #32842 from DaanDeMeyer/cryptsetup

test: Enable TEST-24-CRYPTSETUP for mkosi

18 months agotest: Enable TEST-24-CRYPTSETUP for mkosi 32842/head
Daan De Meyer [Wed, 15 May 2024 11:19:19 +0000 (13:19 +0200)] 
test: Enable TEST-24-CRYPTSETUP for mkosi

Encrypted /var is skipped because meson's limitations make per test
images not really feasible and we can't encrypt /var by default because
it slows down the image build too much.

Co-authored-by: Richard Maw <richard.maw@codethink.co.uk>
18 months agonetwork: configure a tun host0 interface in a container
Helmut Grohne [Wed, 15 May 2024 05:21:53 +0000 (07:21 +0200)] 
network: configure a tun host0 interface in a container

While containers often have their host0 network provided by veth when
the container runtime is privileged, unprivileged containers tend to
have their network provided via slirp4netns or pasta. These tools use a
tun interface rather than a veth interface and systemd should still set
configure such networks.

We should have different .network files for the veth and tun use cases
as there may arise a need to configure them differently.  We should not
rename 80-container-host0.network as that would cause existing drop-ins
to no longer apply.

Closes: #32095
Fixes: f139393dd20a ("network: use Kind= instead of Driver=")
Signed-off-by: Helmut Grohne <helmut@subdivi.de>
18 months agotest: add coverate for Compress=yes config option
Luca Boccassi [Thu, 16 May 2024 10:03:42 +0000 (11:03 +0100)] 
test: add coverate for Compress=yes config option

Avoid regressions like https://github.com/systemd/systemd/issues/32856

Follow-up for 2ef7cdc4255883d1c50860661248c9db69a86aa1

18 months agomkosi: Update centos to latest
Daan De Meyer [Thu, 16 May 2024 10:14:17 +0000 (12:14 +0200)] 
mkosi: Update centos to latest

18 months agojournal-importer: Consider ECONNRESET as EOF
Daan De Meyer [Wed, 15 May 2024 19:40:12 +0000 (21:40 +0200)] 
journal-importer: Consider ECONNRESET as EOF

Otherwise we log a noisy error when we get ECONNRESET.

18 months agoswitch-root: update comment regarding dropped mounts 32863/head
Mike Yuan [Thu, 16 May 2024 11:15:11 +0000 (19:15 +0800)] 
switch-root: update comment regarding dropped mounts

Follow-up for 07c5c2ab8b80f75a1c89fc405db1ce1dfd08534b

18 months agofd-util: check and log error properly for fd_reopen_propagate_append_and_position
Mike Yuan [Thu, 16 May 2024 11:12:38 +0000 (19:12 +0800)] 
fd-util: check and log error properly for fd_reopen_propagate_append_and_position

Follow-up for b8e25bff386548f679902c6b797a5a8ff1542c8b

18 months agope-binary: .initrd section is optional for UKI
Yu Watanabe [Wed, 15 May 2024 21:08:27 +0000 (06:08 +0900)] 
pe-binary: .initrd section is optional for UKI

.osrel is also optional, but sd-boot and bootctl requires it.
So, let's keep .osrel section at least now.

Fixes #32774.

18 months agoMerge pull request #32800 from YHNdnzj/preserve-cred-mounts
Zbigniew Jędrzejewski-Szmek [Thu, 16 May 2024 10:45:40 +0000 (12:45 +0200)] 
Merge pull request #32800 from YHNdnzj/preserve-cred-mounts

switch-root: preserve the whole cred mount tree (/run/credentials/)

18 months agoTEST-24-CRYPTSETUP: Store tokens in /usr
Daan De Meyer [Wed, 15 May 2024 12:00:51 +0000 (14:00 +0200)] 
TEST-24-CRYPTSETUP: Store tokens in /usr

We want to be able to boot with empty /var.

18 months agorepart: Improve error message
Daan De Meyer [Wed, 15 May 2024 14:21:45 +0000 (16:21 +0200)] 
repart: Improve error message

18 months agoMerge pull request #32857 from dtardon/drop-assert
Luca Boccassi [Thu, 16 May 2024 10:04:51 +0000 (12:04 +0200)] 
Merge pull request #32857 from dtardon/drop-assert

Drop invalid asserts

18 months agocore: Skip private /tmp for generators in manager test runs
Daan De Meyer [Wed, 15 May 2024 18:54:33 +0000 (20:54 +0200)] 
core: Skip private /tmp for generators in manager test runs

For manager test runs, the generator output paths are located in
/tmp, which means that if we mount a private /tmp for generators,
we lose all the generated units (actually the generators will just
fail because the directories don't exist, but if they did exist,
we'd still lose all the units).

Let's avoid the problem by skipping the private /tmp for manager
test runs. This also avoids any possible privilege issues with
mounting a private /tmp that might happen in this scenario.

18 months agojournald-server: drop another invalid assert() 32857/head
David Tardon [Thu, 16 May 2024 07:03:40 +0000 (09:03 +0200)] 
journald-server: drop another invalid assert()

config_parse_forward_to_socket() is not used for parsing unit files
either.

Follow-up-for: f31cff849d3db074d0d8157e4117d9b8372c03e8

18 months agojournald-server: drop invalid assert()
David Tardon [Thu, 16 May 2024 06:59:50 +0000 (08:59 +0200)] 
journald-server: drop invalid assert()

config_parse_compress() is not used for parsing unit files, hence unit
is always NULL.

Fixes #32856

Follow-up-for: 6813be2eba24a2f864d5d47326437dbff38cd675

18 months agoMerge pull request #32784 from YHNdnzj/release-version
Luca Boccassi [Wed, 15 May 2024 18:25:59 +0000 (20:25 +0200)] 
Merge pull request #32784 from YHNdnzj/release-version

docs/RELEASE: use meson.version for tags

18 months agohwdb.d/60-keyboard.hwdb: enable Clevo quirk for model V5x0TU
Michał Kopeć [Wed, 15 May 2024 15:33:39 +0000 (17:33 +0200)] 
hwdb.d/60-keyboard.hwdb: enable Clevo quirk for model V5x0TU

Enable the display switch function key quirk for a new model.

Signed-off-by: Michał Kopeć <michal.kopec@3mdeb.com>
18 months agosystemctl: fix "applying zero offset to null pointer" UBSan error
Yu Watanabe [Wed, 15 May 2024 15:45:06 +0000 (00:45 +0900)] 
systemctl: fix "applying zero offset to null pointer" UBSan error

Fixes #32837.

18 months agoTEST-82-SOFTREBOOT: test that creds for surviving units are carried over 32800/head
Mike Yuan [Tue, 14 May 2024 11:34:11 +0000 (19:34 +0800)] 
TEST-82-SOFTREBOOT: test that creds for surviving units are carried over

18 months agoswitch-root: preserve the whole cred mount tree (/run/credentials/)
Mike Yuan [Tue, 14 May 2024 09:03:00 +0000 (17:03 +0800)] 
switch-root: preserve the whole cred mount tree (/run/credentials/)

Currently, during soft-reboot, some services may survive,
but their associated credential mounts are dropped.
Let's instead preserve them, as discussed.

18 months agotest/test-rpm-macros.sh: add build directory to pkg-config search path
Radoslav Kolev [Tue, 14 May 2024 09:19:22 +0000 (12:19 +0300)] 
test/test-rpm-macros.sh: add build directory to pkg-config search path

If tests are run during build time, without an already installed
systemd they fail to resolve the sysusersdir and tpmfilesdir pkg-config variables.

18 months agoMerge pull request #32844 from YHNdnzj/mount-defaultdep-followup
Luca Boccassi [Wed, 15 May 2024 16:23:26 +0000 (18:23 +0200)] 
Merge pull request #32844 from YHNdnzj/mount-defaultdep-followup

core/mount: correctly set DefaultDependencies=no for mounts from mountinfo

18 months agoMerge pull request #32818 from keszybz/libsystemd-network-size-check
Luca Boccassi [Wed, 15 May 2024 15:33:22 +0000 (17:33 +0200)] 
Merge pull request #32818 from keszybz/libsystemd-network-size-check

Check packet size in libsystemd-network

18 months agocore/mount: correctly set DefaultDependencies=no for mounts from mountinfo 32844/head
Mike Yuan [Wed, 15 May 2024 13:49:00 +0000 (21:49 +0800)] 
core/mount: correctly set DefaultDependencies=no for mounts from mountinfo

Follow-up for 3c0a1b1e70c1fe6e0b6707fab7dc86b1500d0f82

Before this commit, DefaultDependencies=no is set in
mount_add_extras(). However, when generating mount units
from /proc/self/mountinfo, we don't have a unit in memory
yet, and mount_setup_new_unit() doesn't call into
mount_add_extras().

Fixes #32838

18 months agocore/swap: rename variable 'new' -> 'new_unit'
Mike Yuan [Wed, 15 May 2024 13:52:47 +0000 (21:52 +0800)] 
core/swap: rename variable 'new' -> 'new_unit'

Follow-up for ba31a5018f99864c22dd4e0f10712456c7abc934

'new' is already used as a macro for memory allocation.
Let's avoid the confusion here.

Addresses https://github.com/systemd/systemd/pull/31902#discussion_r1557672858

18 months agoman: fix typo 'ot' -> 'or'
Luca Boccassi [Wed, 15 May 2024 10:57:06 +0000 (11:57 +0100)] 
man: fix typo 'ot' -> 'or'

18 months agocore: Fix assertion in parse_smbios_strings()
Daan De Meyer [Wed, 15 May 2024 10:26:33 +0000 (12:26 +0200)] 
core: Fix assertion in parse_smbios_strings()

18 months agoTEST-08-INITRD: Fix result check
Daan De Meyer [Wed, 15 May 2024 10:32:57 +0000 (12:32 +0200)] 
TEST-08-INITRD: Fix result check

18 months agoMerge pull request #32835 from DaanDeMeyer/test
Daan De Meyer [Wed, 15 May 2024 09:58:22 +0000 (11:58 +0200)] 
Merge pull request #32835 from DaanDeMeyer/test

test: Enable TEST-08-INITRD on mkosi

18 months agotest: Enable TEST-08-INITRD on mkosi 32835/head
Daan De Meyer [Wed, 15 May 2024 08:09:53 +0000 (10:09 +0200)] 
test: Enable TEST-08-INITRD on mkosi

18 months agotest: Rename "shutdown initrd" to "exitrd"
Daan De Meyer [Wed, 15 May 2024 06:48:46 +0000 (08:48 +0200)] 
test: Rename "shutdown initrd" to "exitrd"

18 months agoTEST-64-UDEV-STORAGE: Give higher priority
Daan De Meyer [Wed, 15 May 2024 05:27:00 +0000 (07:27 +0200)] 
TEST-64-UDEV-STORAGE: Give higher priority

Some of the udev storage tests can take quite a while, so let's make
sure they start early enough.

18 months agoTEST-85-NETWORK: Give higher priority
Daan De Meyer [Wed, 15 May 2024 05:26:15 +0000 (07:26 +0200)] 
TEST-85-NETWORK: Give higher priority

Some of the networkd tests can take quite a while, so let's make sure
they start early enough.

18 months agoTEST-21-DFUZZER: Fix priority
Daan De Meyer [Wed, 15 May 2024 06:26:34 +0000 (08:26 +0200)] 
TEST-21-DFUZZER: Fix priority

Tests with higher priority, not lower priority are started first.

18 months agoMerge pull request #32666 from DaanDeMeyer/mkosi-network
Daan De Meyer [Wed, 15 May 2024 05:30:04 +0000 (07:30 +0200)] 
Merge pull request #32666 from DaanDeMeyer/mkosi-network

test: Add TEST-85-NETWORK to run systemd-networkd-tests.py

18 months agoFinalize for v256~rc2 v256-rc2
Luca Boccassi [Tue, 14 May 2024 23:29:01 +0000 (00:29 +0100)] 
Finalize for v256~rc2

18 months agoMerge pull request #32830 from yuwata/home-skel
Luca Boccassi [Tue, 14 May 2024 23:26:15 +0000 (01:26 +0200)] 
Merge pull request #32830 from yuwata/home-skel

home: fix ownership of files copied from skelton directory

18 months agotest: add tests for "homectl --skel" 32830/head
Yu Watanabe [Tue, 14 May 2024 21:34:43 +0000 (06:34 +0900)] 
test: add tests for "homectl --skel"

18 months agoTEST-21-DFUZZER: Fix script name
Daan De Meyer [Tue, 14 May 2024 18:52:28 +0000 (20:52 +0200)] 
TEST-21-DFUZZER: Fix script name

18 months agohome: fix ownership of files copied from skelton directory
Yu Watanabe [Tue, 14 May 2024 21:19:42 +0000 (06:19 +0900)] 
home: fix ownership of files copied from skelton directory

Fixes #32808.

18 months agocore,vconsole-setup: treat locking failure as non-fatal
Zbigniew Jędrzejewski-Szmek [Tue, 14 May 2024 18:00:48 +0000 (20:00 +0200)] 
core,vconsole-setup: treat locking failure as non-fatal

Locking of the tty device and then /dev/console was added to synchronize
vconsole-setup with other writers to the console. But it turns out that often
the locking doesn't work and we carved out various cases where we ignore
failure:
- lack of permissions (in the user manager)
- missing device node

It turns out that there's at least one more failure mode: we get -EIO when the
console is (mis-)configured to point to an invalid device. E.g. in
rhbug#2273069 the reporter has a VM in Proxmox without a virtual console
configured and has 'console=tty console=ttyS0' on the kernel cmdline. I
couldn't reproduce this under libvirt, but failure with EIO has been reported
by at least four users in #30501.

Note that in systemd-vconsole-setup we report this is a hard failure, while
in the manager, we only do a debug line. So it's possible that the failure
also occured there, causing the rest of the setup of the tty to be skipped
without further notice.

Ignore the locking failure, since there's just too many ways it can fail. If we
proceed without a lock, we're back to the situation before we started locking,
which wasn't too bad. OTOH, skipping setup of the console is problematic for
users, and it seems better to try to do the setup without locking.

Fixes https://github.com/systemd/systemd/issues/30501,
https://bugzilla.redhat.com/show_bug.cgi?id=2273069.

18 months agotest: Enable TEST-54-CREDS on mkosi
Daan De Meyer [Tue, 14 May 2024 20:14:27 +0000 (22:14 +0200)] 
test: Enable TEST-54-CREDS on mkosi

18 months agotest: Add TEST-85-NETWORK to run systemd-networkd-tests.py 32666/head
Daan De Meyer [Mon, 6 May 2024 13:23:24 +0000 (15:23 +0200)] 
test: Add TEST-85-NETWORK to run systemd-networkd-tests.py

This adds a testsuite unit to run systemd-networkd-tests.py. This is
mkosi only for now as python is not available in the images set up
by the bash framework. We give the test a lower priority as it takes
a while to run so we want to start it as soon as possible.

18 months agotest: Allow using TEST_MATCH_TESTCASE with systemd-networkd-tests.py
Daan De Meyer [Tue, 14 May 2024 13:22:06 +0000 (15:22 +0200)] 
test: Allow using TEST_MATCH_TESTCASE with systemd-networkd-tests.py

18 months agomkosi: Disable journald rate-limiting
Daan De Meyer [Mon, 13 May 2024 19:19:14 +0000 (21:19 +0200)] 
mkosi: Disable journald rate-limiting

18 months agoMerge pull request #32817 from YHNdnzj/core-varlink
Luca Boccassi [Tue, 14 May 2024 20:40:01 +0000 (22:40 +0200)] 
Merge pull request #32817 from YHNdnzj/core-varlink

core/core-varlink: trivial cleanups

18 months agopo: Update translation files
Weblate [Tue, 14 May 2024 20:06:31 +0000 (22:06 +0200)] 
po: Update translation files

Updated by "Update PO files to match POT (msgmerge)" hook in Weblate.

Co-authored-by: Weblate <noreply@weblate.org>
Translate-URL: https://translate.fedoraproject.org/projects/systemd/main/
Translation: systemd/main

18 months agoMerge pull request #32816 from bluca/chores
Luca Boccassi [Tue, 14 May 2024 20:07:40 +0000 (22:07 +0200)] 
Merge pull request #32816 from bluca/chores

Chores for RC2

18 months agotest: Enable TEST-82-SOFTREBOOT on mkosi
Daan De Meyer [Tue, 14 May 2024 18:08:00 +0000 (20:08 +0200)] 
test: Enable TEST-82-SOFTREBOOT on mkosi

18 months agomkosi: Restore job for F40
Zbigniew Jędrzejewski-Szmek [Tue, 14 May 2024 19:02:40 +0000 (21:02 +0200)] 
mkosi: Restore job for F40

This partially reverts ecf8468dd4abbf006cb607d023018edee51c851d.
dist-git was made compatible again with F40.

18 months agoMerge pull request #32812 from DaanDeMeyer/test
Daan De Meyer [Tue, 14 May 2024 18:50:38 +0000 (20:50 +0200)] 
Merge pull request #32812 from DaanDeMeyer/test

test: Only add tpm2.target dependency to TEST-70-TPM2.service

18 months agosysusers: simplify meson config
Zbigniew Jędrzejewski-Szmek [Mon, 13 May 2024 16:53:51 +0000 (18:53 +0200)] 
sysusers: simplify meson config

There were two parts, for templated and non-templated files, and
they were more different than they should be.