systemd/.git
4 years agostat-util: don't try to open path on path_is_temporary_fs()
Lennart Poettering [Wed, 6 Jan 2021 17:05:56 +0000 (18:05 +0100)]
stat-util: don't try to open path on path_is_temporary_fs()

I mean, the old code at least used O_PATH, but still, we shouldn't
allocate/close an fd if we don't have to.

(cherry picked from commit 15308e5083391f6a1b9ce25c5b7323f37544eab8)

4 years agosystemctl: have is-enabled return success for aliases when calling into pid1 too
Luca Boccassi [Tue, 5 Jan 2021 13:44:26 +0000 (13:44 +0000)]
systemctl: have is-enabled return success for aliases when calling into pid1 too

commit 15d7ab87c4e5917f5788f1f8dce327a1e272bea3 introduced the
change to add an 'alias' state, but it was wired to systemctl
only when running in 'client-side' mode. Return success as
expected and documented also when running in 'server-mode'.

Fixes https://github.com/systemd/systemd/issues/18134

(cherry picked from commit bf3b428f73eb51f4abb64a80ea5be346b9eeb2fe)

4 years agoman: fix path reference to unit file
Gaël PORTAY [Thu, 5 Mar 2020 17:38:31 +0000 (12:38 -0500)]
man: fix path reference to unit file

The unit files are located at path /usr/lib/systemd/system.

This fixes the path reference to the unit file by adding the missing
path component system.

(cherry picked from commit 0215f04a7ebd1ab1da4b4279d5057953ae1ebcaa)

4 years agodocs: fix the link to boot loader specification
Gaël PORTAY [Wed, 30 Dec 2020 14:29:36 +0000 (09:29 -0500)]
docs: fix the link to boot loader specification

The boot loader specification link points to the boot loader interface
documentation.

This fixes the link to point to BOOT_LOADER_SPECIFICATION instead of
BOOTLOADER_INTERFACE which is itself.

(cherry picked from commit 7745379ea8c1d08ecde9ccd15af1a76cdac91cc5)

4 years agonetwork: fix possible memory leak
Yu Watanabe [Tue, 29 Dec 2020 17:11:32 +0000 (02:11 +0900)]
network: fix possible memory leak

When set_put() returns 0, then already stored rule will be unref()ed
from Manager.

(cherry picked from commit fdce9324c7132a9a6d3b2fe9f8449dcfc8309114)

4 years agoresolve: field size in dns resource record may be zero
Yu Watanabe [Tue, 29 Dec 2020 16:48:35 +0000 (01:48 +0900)]
resolve: field size in dns resource record may be zero

(cherry picked from commit b652cccab9ed4091ec347f6e924be131105279e9)

4 years agosiphash: introduce siphash24_compress_safe()
Yu Watanabe [Tue, 29 Dec 2020 16:44:04 +0000 (01:44 +0900)]
siphash: introduce siphash24_compress_safe()

(cherry picked from commit 0b71a7e01d8d04a3668c349e2320c42c786f2382)

4 years agofuzzers: set maximum length for several fuzzers
Yu Watanabe [Tue, 29 Dec 2020 08:26:43 +0000 (17:26 +0900)]
fuzzers: set maximum length for several fuzzers

Hopefully fixes oss-fuzz#19081, oss-fuzz#21731, oss-fuzz#21755.
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=19081
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=21731
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=21755

(cherry picked from commit 57dda0de3896e57f6d3f0625b2d5e13bbdb48d79)

4 years agoshared/dns: fix dlopen_idn return code check
Luca Boccassi [Mon, 28 Dec 2020 00:23:14 +0000 (00:23 +0000)]
shared/dns: fix dlopen_idn return code check

Fixes https://github.com/systemd/systemd/issues/18078

(cherry picked from commit 5def1f11f84fd4000a0f7c9ed1294b04f0cea703)

4 years agoman/systemd-nspawn: document hashing machine name for uid base
Dmitry Borodaenko [Wed, 23 Dec 2020 01:54:04 +0000 (17:54 -0800)]
man/systemd-nspawn: document hashing machine name for uid base

Explicitly document the behavior introduced in #7437: when picking a new
UID shift base with "-U", a hash of the machine name will be tried
before falling back to fully random UID base candidates.

(cherry picked from commit 68709a636c838e0754b49caa6ff2d4168e3c99c8)

4 years agoudev: fix memleak
Yu Watanabe [Mon, 21 Dec 2020 06:28:19 +0000 (15:28 +0900)]
udev: fix memleak

Fixes #18039.

(cherry picked from commit 9df1e11e31588c5df39aa7efed05414273f93f69)

4 years agojournal-importer: ignore invalid field at one more place
Yu Watanabe [Mon, 21 Dec 2020 07:33:20 +0000 (16:33 +0900)]
journal-importer: ignore invalid field at one more place

Fixes oss-fuzz#28817.
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=28817

(cherry picked from commit 8786d4bbe43b5f6493982bcb5211e010f99deb57)

4 years agoman/localtime: document default timezone
Luca BRUNO [Mon, 21 Dec 2020 09:55:04 +0000 (09:55 +0000)]
man/localtime: document default timezone

This explicitly documents the default `UTC` timezone which is used
by systemd and (other softwares) when `/etc/localtime` is not present.

Ref: https://github.com/systemd/systemd/blob/v247/src/basic/time-util.c#L1460-L1469
(cherry picked from commit 3d909037e3bb86bb14ceef98325b2759b938a4b3)

4 years agoman/systemd.netdev: clarify the wireguard AllowedIPs= setting
Florian Klink [Sun, 20 Dec 2020 17:24:05 +0000 (18:24 +0100)]
man/systemd.netdev: clarify the wireguard AllowedIPs= setting

`AllowedIPs=` only affects "routing inside the network interface
itself", as in, which wireguard peer packets with a specific destination
address are sent to, and what source addresses are accepted from which
peer.

To cause packets to be sent via wireguard in first place, a route via
that interface needs to be added - either in the `[Routes]` section on
the `.network` matching the wireguard interface, or outside of networkd.

This is a common cause of misunderstanding, because tools like wg-quick
also add routes to the interface. However, those tools are meant as a
"extremely simple script for easily bringing up a WireGuard interface,
suitable for a few common use cases (from their manpage).

Networkd also should support other usecases - like setting AllowedIPs to
0.0.0.0/0 and ::/0 and having a dynamic routing protocol setting more
specific routes (or the user manually setting them).

Reported-In: https://github.com/systemd/systemd/issues/14176
(cherry picked from commit c6b90e5c5e54e98b6aed38677f77d8491f2e49c8)

4 years agologs-show: refuse data which contain invalid fields
Yu Watanabe [Tue, 15 Dec 2020 19:50:39 +0000 (04:50 +0900)]
logs-show: refuse data which contain invalid fields

(cherry picked from commit 805d67c565d57e0915162164f7e5e3026a29a2c5)

4 years agojournal: refuse data which contain invalid fields
Yu Watanabe [Tue, 15 Dec 2020 19:44:31 +0000 (04:44 +0900)]
journal: refuse data which contain invalid fields

Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=25353.

(cherry picked from commit f2bd032044ca3cd4b454dd0ba86719effcf34dc0)

4 years agojournal: move journal_field_valid() to journal_file.c
Yu Watanabe [Tue, 15 Dec 2020 19:36:14 +0000 (04:36 +0900)]
journal: move journal_field_valid() to journal_file.c

(cherry picked from commit adce225a104d0b7503aa7322db15d1c6dd8b8093)

4 years agotest: use modern qemu numa arguments
Christian Ehrhardt [Tue, 15 Dec 2020 11:05:14 +0000 (12:05 +0100)]
test: use modern qemu numa arguments

Upgrading to qemu 5.2 breaks TEST-36-NUMAPOLICY like:
  qemu-system-x86_64: total memory for NUMA nodes (0x0) should
  equal RAM size (0x20000000)

Use the new (as in >=2014) form of memdev in test 36:
 -object memory-backend-ram,id=mem0,size=512M -numa node,memdev=mem0,nodeid=0

Since some target systems are as old as qemu 1.5.3 (CentOS7) but the new
kind to specify was added in qemu 2.1 this needs to add version parsing and
add the argument only when qemu is >=5.2.

Fixes #17986.

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>
(cherry picked from commit 43b49470d1f2808555c07f64cd0a1529b7ddd559)

4 years agobus-util: improve logging when we can't connect to the bus
Lennart Poettering [Mon, 14 Dec 2020 15:36:00 +0000 (16:36 +0100)]
bus-util: improve logging when we can't connect to the bus

Previously, we'd already have explicit logging for the case where
$XDG_RUNTIME_DIR is not set. Let's also add some explicit logging for
the EPERM/ACCESS case. Let's also in both cases suggest the
--machine=<user>@.host syntax.

And while we are at it, let's remove side-effects from the macro.

By checking for both the EPERM/EACCES case and the $XDG_RUNTIME_DIR case
we will now catch both the cases where people use "su" to issue a
"systemctl --user" operation, and those where they (more correctly, but
still not good enough) call "su -".

Fixes: #17901
(cherry picked from commit 1ecb46724cae151606bc825f0e39f14d4dfe1a0e)

4 years agosd-bus: make credential acquisition more graceful
Lennart Poettering [Mon, 14 Dec 2020 12:16:39 +0000 (13:16 +0100)]
sd-bus: make credential acquisition more graceful

So far when asked for augmented bus credentials and the process was
already gone we'd fail fatally. Let's make this graceful instead, and
never allow augmenting fail due to PID having vanished — unless the
augmenting is the explicit and only purpose of the requested operation.

This should be safe as clients have to explicitly query the acquired
creds anyway and handle if they couldn't be acquired. Moreover we
already handle permission problems gracefully, thus clients must be
ready to deal with missing creds.

This is useful to make selinux authorization work for short-lived client
proceses. PReviously we'd augment creds to have more info to log about
(the selinux decision would not be based on augmented data however,
because that'd be unsafe), and would fail if we couldn't get it. Now,
we'll try to acquire the data, but if we cannot acquire it, we'll still
do the selinux check, except that logging will be more limited.

(cherry picked from commit f8ecc2c00df7bd810557f3056ec12f6a0730812d)

4 years agosd-bus: 'ret' parameter to sd_bus_query_sender_creds() is not optional, check for it
Lennart Poettering [Mon, 14 Dec 2020 12:20:28 +0000 (13:20 +0100)]
sd-bus: 'ret' parameter to sd_bus_query_sender_creds() is not optional, check for it

(cherry picked from commit 1ca37419b13b836d7fb2b9815d5efb6dccc62134)

4 years agonetwork: fix IPv6PrivacyExtensions=kernel handling
Florian Klink [Wed, 23 Dec 2020 11:57:47 +0000 (12:57 +0100)]
network: fix IPv6PrivacyExtensions=kernel handling

When set to "kernel", systemd is not supposed to touch that sysctl.

5e0534f1c13cd50ec2b143a8b18156cd37e502f7, part of
https://github.com/systemd/systemd/pull/17240 forgot to handle that
case.

Fixes https://github.com/systemd/systemd/issues/18003

(cherry picked from commit d3ccb1b4ba71b7bdf263aab8b1b17d1274d27a47)

4 years agonetwork: fix typo
Florian Klink [Wed, 23 Dec 2020 11:56:45 +0000 (12:56 +0100)]
network: fix typo

IPv6 privacy extensions are plural, not singular.

(cherry picked from commit 7eeaf72b88e43cb100763cec1db1b750928f2a1b)

4 years agodissect: fix root hash signature autodiscovery
Luca Boccassi [Sun, 10 Jan 2021 15:06:04 +0000 (15:06 +0000)]
dissect: fix root hash signature autodiscovery

The root hash signature is auto discovered only if the root hash was specified
manually. Ensure that an auto discovered root hash is also enough.

(cherry picked from commit 90f989861e1f7fd4465a8dddd1721b54ecb3f273)

4 years agocryptsetup: add support for workqueue options
Jonathan G. Underwood [Tue, 22 Dec 2020 20:04:52 +0000 (20:04 +0000)]
cryptsetup: add support for workqueue options

This commit adds support for disabling the read and write
workqueues with the new crypttab options no-read-workqueue
and no-write-workqueue. These correspond to the cryptsetup
options --perf-no_read_workqueue and --perf-no_write_workqueue
respectively.

(cherry picked from commit 227acf0009bde2cd7f8bc371615b05e84137847d)

4 years agotest-login: skip consistency checks when logind is not active
Zbigniew Jędrzejewski-Szmek [Wed, 16 Dec 2020 14:56:44 +0000 (15:56 +0100)]
test-login: skip consistency checks when logind is not active

There are two ways in swich sd_login_* functions acquire data:
some are derived from the cgroup path, but others use the data serialized
by logind.

When the tests are executed under Fedora's mock, without systemd-spawn
but instead in a traditional chroot, test-login gets confused:
the "outside" cgroup path is visible, so sd_pid_get_unit() and
sd_pid_get_session() work, but sd_session_is_active() and other functions
that need logind data fail.

Such a buildroot setup is fairly bad, but it can be encountered in the wild, so
let's just skip the tests in that case.

/* Information printed is from the live system */
sd_pid_get_unit(0, …) → "session-237.scope"
sd_pid_get_user_unit(0, …) → "n/a"
sd_pid_get_slice(0, …) → "user-1000.slice"
sd_pid_get_session(0, …) → "237"
sd_pid_get_owner_uid(0, …) → 1000
sd_pid_get_cgroup(0, …) → "/user.slice/user-1000.slice/session-237.scope"
sd_uid_get_display(1000, …) → "(null)"
sd_uid_get_sessions(1000, …) → [0] ""
sd_uid_get_seats(1000, …) → [0] ""
Assertion 'r >= 0' failed at src/libsystemd/sd-login/test-login.c:104, function test_login(). Aborting.

(cherry picked from commit ac5644635dba54ce5eb0ff394fc0bc772a984849)

4 years agojournalctl: don't skip the entries that have the same seqnum v247.2
shenyangyang4 [Thu, 10 Dec 2020 11:44:31 +0000 (06:44 -0500)]
journalctl: don't skip the entries that have the same seqnum

These two judgement can't judge that two entries are repeating fully.
So i think seqnum is needed to make full judgement.

(cherry picked from commit b17f651a17cd6ec0ceac7835f2f8607fbd9ddb95)

4 years agosd-bus: use SOCK_CLOEXEC on one more socket
Lennart Poettering [Fri, 11 Dec 2020 11:04:11 +0000 (12:04 +0100)]
sd-bus: use SOCK_CLOEXEC on one more socket

(cherry picked from commit 68a3d9153883b90c99ea2aec20075146ce58beaa)

4 years agoresolved: create stub-resolv.conf symlink with correct security label
Ondrej Mosnacek [Mon, 14 Dec 2020 15:36:27 +0000 (16:36 +0100)]
resolved: create stub-resolv.conf symlink with correct security label

Use symlink_atomic_label() instead of symlink_atomic() as the symlink
may need a different label than the parent directory.

Signed-off-by: Ondrej Mosnacek <omosnace@redhat.com>
(cherry picked from commit 7b87bece5ded3e8f068df8402901198c069ab5cf)

4 years agoefi: Only use arm flags if supported
Andrew Balmos [Fri, 11 Dec 2020 03:15:24 +0000 (22:15 -0500)]
efi: Only use arm flags if supported

Support gcc 8 on arm

(cherry picked from commit 361f41645cdf920d431e2d68dcfa3f98088c2e03)

4 years agocore: detect_container() may return negative errno
Yu Watanabe [Mon, 14 Dec 2020 16:13:32 +0000 (01:13 +0900)]
core: detect_container() may return negative errno

(cherry picked from commit bcdb3b7d5076cf6ad17cb70df8db22d876880ada)

4 years agomeson: Fix reallocarray check
Khem Raj [Sun, 13 Dec 2020 00:15:57 +0000 (16:15 -0800)]
meson: Fix reallocarray check

reallocarray() is defined in stdlib.h, so that would be right header to
check for its presense.

Signed-off-by: Khem Raj <raj.khem@gmail.com>
(cherry picked from commit 5bb20fd3d33f7e866a0845f15c1ab5b595147f1e)

4 years agonetwork: do not assume address ready callback is always set to static addresses
Yu Watanabe [Fri, 11 Dec 2020 05:22:35 +0000 (14:22 +0900)]
network: do not assume address ready callback is always set to static addresses

(cherry picked from commit b788a429d19acda8acb1c4b543da1adc2e49f4a9)

4 years agonetwork: drop assertions to check link state in netlink callback handlers
Yu Watanabe [Thu, 10 Dec 2020 05:16:22 +0000 (14:16 +0900)]
network: drop assertions to check link state in netlink callback handlers

As, the link may be dropped while configuring addresses or routes.

Fixes #17920.

(cherry picked from commit b2219cc4b6a2832c29eb82e90625d9c6ac61e73c)

4 years agonetwork: do not reconfigure interface when the link gains carrier but udev not initia...
Yu Watanabe [Fri, 11 Dec 2020 03:15:45 +0000 (12:15 +0900)]
network: do not reconfigure interface when the link gains carrier but udev not initialized it yet

When an interface gains carrier but udev have not initialized the
interface or link_initialized_handler() has not been called yet,
then link_configure will be called twice. Thus LLDP client will be
configured twice, and triggers assertion.

Fixes #17929.

(cherry picked from commit 8bceafa7771800c8bdfc1b74ef39824103f5c129)

4 years agoveritysetup: also place udev socket dep
Lennart Poettering [Thu, 10 Dec 2020 10:48:37 +0000 (11:48 +0100)]
veritysetup: also place udev socket dep

In light of #17848, also place udev socket dep in veritysetup, it's the
same issue after all.

(cherry picked from commit 8374bf4fa2663b19f329801abfd8b29e66897380)

4 years agocryptsetup: Fix crypto device missing issue after bootup
Jinyuan Si [Fri, 4 Dec 2020 02:38:28 +0000 (10:38 +0800)]
cryptsetup: Fix crypto device missing issue after bootup

Normally, the udev rules operate on "change" events. But when
coldplugging, there's an "add" event present. The udev rules have to
recognize this and do some actions in this particular situation, too.
Also, we don't want the nodes to be created prematurely on "add"
events while not coldplugging. The udev rules will check
DM_UDEV_PRIMARY_SOURCE_FLAG to see if the device was activated
correctly before and if not, it ignore the "add" event totally.
This way the udev rules can support udev triggers generating "add"
events (e.g. "udevadm trigger --action=add" or
"echo add > /sys/block/<dm_device>/uevent").

In this case, the udevd service is started after
systemd-cryptsetup@config.service, is started, which will cause udevd
service to miss the "change" uevent with DM_UDEV_PRIMARY_SOURCE_FLAG
flag generated by systemd-cryptsetup@config.service. To solve this
issue, we let the cryptsetup service be started after the udevd
service.

(cherry picked from commit 239952e890fd4b53859f84bbc43d910e68a8f6bc)

4 years agonetwork: fix SIGABRT related to unreachable route with DHCP6
Yu Watanabe [Fri, 4 Dec 2020 11:50:34 +0000 (20:50 +0900)]
network: fix SIGABRT related to unreachable route with DHCP6

After #17834, unreachable routes generated through DHCP6 are managed by
Manager. But they are referrenced by the DHCP6 uplink. So, the routes
managed by Manager must be freed after all Link objects are freed.

Follow-up for 575f14eef010101c60e9d4d970e542c815be1994.

Fixes SIGABRT reproted in #17831.

(cherry picked from commit 450fa34bd0315f2ca4728050b37c3a3d9e0f1b78)

4 years agonetwork: revert previous changes to address_compare_func()
Yu Watanabe [Fri, 4 Dec 2020 07:41:08 +0000 (16:41 +0900)]
network: revert previous changes to address_compare_func()

This partially reverts fe841414ef157f7f01d339c5d5730126e7b5fe0a and
2a236f9fc0ff8fb2152032551436fde74da7217a.

For IPv4, kernel compares the local address, prefix, and prefixlen.
For IPv6, kernel compares only the local address.
Let's follow the kernel's comparison way.

Fixes #17831.

(cherry picked from commit 1d30fc5cb64ecba2f03fe42aa0d8c65c3decad82)

4 years agoudev: Fix sound.target dependency
Takashi Iwai [Wed, 9 Dec 2020 09:56:51 +0000 (10:56 +0100)]
udev: Fix sound.target dependency

The recent bug report indicated a race at device creation and the
sound.target dependencies, and the cause turned out to be the condition
of the sound.target trigger.  Currently it's set for "card*", but this
is actually the parent object; i.e. the sound.target is triggered before
the sound devices are created.

For assuring the whole sound device creations beforehand, we need to use
"controlC*" instead of "card*"; as already described in
78-sound-card.rules, this is guaranteed to be the last device, and can
be used as a synchronization point.

BugLink: https://bugzilla.suse.com/show_bug.cgi?id=1179363
Signed-off-by: Takashi Iwai <tiwai@suse.de>
(cherry picked from commit 5926e592fa5e6290b9f4588939945869adb5c55f)

4 years agomeson: specify correct libqrencode version in meson dep
Luca Boccassi [Wed, 9 Dec 2020 12:33:54 +0000 (13:33 +0100)]
meson: specify correct libqrencode version in meson dep

(cherry picked from commit a6c7811f0d3888e2fa545cd80d7815049b5cb084)

4 years agoudev: link_update() should fail if the entry in symlink dir couldn't have been created
Franck Bui [Wed, 9 Dec 2020 08:46:08 +0000 (09:46 +0100)]
udev: link_update() should fail if the entry in symlink dir couldn't have been created

Follow-up for 30f6dce62cb3a738b20253f2192270607c31b55b

(cherry picked from commit 4e1db59274c4b31ba5369270a489420245616eb4)

4 years agoman: document that automount units are privileged
Lennart Poettering [Wed, 9 Dec 2020 13:04:21 +0000 (14:04 +0100)]
man: document that automount units are privileged

Fixes: #17886
(cherry picked from commit eef34a1d8fe5ca44bdab189b682642096c53e661)

4 years agologind: fix closing of button input devices
Lennart Poettering [Mon, 7 Dec 2020 09:51:15 +0000 (10:51 +0100)]
logind: fix closing of button input devices

This is a fix of #17751. Specifically:

1. Sort #include headers again

2. Remove tabs, as per coding style

3. Don't install fds in half-initialized objects

4. Use asynchronous_close() everywhere

That all said:

Quit frankly, I am not convinced we should do all this at all. If
close()ing of these input devices is really that slow, then this should
probably be fixed in the kernel, not worked around in userspace like
this.

(cherry picked from commit c74d5fe25d53263c143f0a9c2698d8bb483e398c)

4 years agoUpdate logind-button.c
nihilix-melix [Fri, 27 Nov 2020 10:21:22 +0000 (11:21 +0100)]
Update logind-button.c

(cherry picked from commit eee582e7951fa8e328d1fcdfcff940254070ccba)

4 years agoasync: add trivial cleanup wrapper for asynchronous_close()
Lennart Poettering [Wed, 9 Dec 2020 09:07:12 +0000 (10:07 +0100)]
async: add trivial cleanup wrapper for asynchronous_close()

(cherry picked from commit 1d9aa4d572b661fd6500c55ab524141332f76230)

4 years agoSilence cgroups v1 read-only filesystem warning
Daan De Meyer [Mon, 7 Dec 2020 22:18:28 +0000 (22:18 +0000)]
Silence cgroups v1 read-only filesystem warning

Avoid warning messages when booting systemd-nspawn containers and using
hybrid or legacy cgroups. systemd-nspawn mounts the cgroups v1 controller
tree as read-only so these errors are expected and not problematic.
Partially fixes #17862.

Test plan:

- Before: `mkosi --default .mkosi/mkosi.fedora boot`

```
‣ Processing default...
Spawning container image on /home/daan/projects/systemd/image.raw.
Press ^] three times within 1s to kill container.
systemd 247 running in system mode. (+PAM +AUDIT +SELINUX -APPARMOR +IMA +SMACK +SECCOMP +GCRYPT +GNUTLS +OPENSSL +ACL +BLKID +CURL +ELFUTILS +FIDO2 +IDN2 -IDN +IPTC +KMOD +LIBCRYPTSETUP +LIBFDISK +PCRE2 +PWQUALITY +P11KIT +QRENCODE +BZIP2 +LZ4 +XZ +ZLIB +ZSTD +XKBCOMMON +UTMP +SYSVINIT default-hierarchy=unified)
Detected virtualization systemd-nspawn.
Detected architecture x86-64.

Welcome to Fedora 33 (Thirty Three)!

Queued start job for default target Graphical Interface.
-.slice: Failed to migrate controller cgroups from , ignoring: Read-only file system
system.slice: Failed to delete controller cgroups /system.slice, ignoring: Read-only file system
[  OK  ] Created slice system-getty.slice.
[  OK  ] Created slice system-modprobe.slice.
user.slice: Failed to delete controller cgroups /user.slice, ignoring: Read-only file system
[  OK  ] Created slice User and Session Slice.
[  OK  ] Started Dispatch Password Requests to Console Directory Watch.
[  OK  ] Started Forward Password Requests to Wall Directory Watch.
[  OK  ] Reached target Local Encrypted Volumes.
[  OK  ] Reached target Paths.
[  OK  ] Reached target Remote File Systems.
[  OK  ] Reached target Slices.
[  OK  ] Reached target Swap.
[  OK  ] Listening on Process Core Dump Socket.
[  OK  ] Listening on initctl Compatibility Named Pipe.
[  OK  ] Listening on Journal Socket (/dev/log).
[  OK  ] Listening on Journal Socket.
[  OK  ] Listening on User Database Manager Socket.
dev-hugepages.mount: Failed to delete controller cgroups /dev-hugepages.mount, ignoring: Read-only file system
         Mounting Huge Pages File System...
sys-fs-fuse-connections.mount: Failed to delete controller cgroups /sys-fs-fuse-connections.mount, ignoring: Read-only file system
         Mounting FUSE Control File System...
         Starting Journal Service...
         Starting Remount Root and Kernel File Systems...
system.slice: Failed to delete controller cgroups /system.slice, ignoring: Read-only file system
```

After: `mkosi --default .mkosi/mkosi.fedora boot`

```
‣ Processing default...
Spawning container image on /home/daan/projects/systemd/mkosi.output/image.raw.
Press ^] three times within 1s to kill container.
systemd 247 running in system mode. (+PAM +AUDIT +SELINUX -APPARMOR +IMA +SMACK +SECCOMP +GCRYPT +GNUTLS +OPENSSL +ACL +BLKID +CURL +ELFUTILS +FIDO2 +IDN2 -IDN +IPTC +KMOD +LIBCRYPTSETUP +LIBFDISK +PCRE2 +PWQUALITY +P11KIT +QRENCODE +BZIP2 +LZ4 +XZ +ZLIB +ZSTD +XKBCOMMON +UTMP +SYSVINIT default-hierarchy=unified)
Detected virtualization systemd-nspawn.
Detected architecture x86-64.

Welcome to Fedora 33 (Thirty Three)!

Queued start job for default target Graphical Interface.
[  OK  ] Created slice system-getty.slice.
[  OK  ] Created slice system-modprobe.slice.
[  OK  ] Created slice User and Session Slice.
[  OK  ] Started Dispatch Password Requests to Console Directory Watch.
[  OK  ] Started Forward Password Requests to Wall Directory Watch.
[  OK  ] Reached target Local Encrypted Volumes.
[  OK  ] Reached target Paths.
[  OK  ] Reached target Remote File Systems.
[  OK  ] Reached target Slices.
[  OK  ] Reached target Swap.
[  OK  ] Listening on Process Core Dump Socket.
[  OK  ] Listening on initctl Compatibility Named Pipe.
[  OK  ] Listening on Journal Socket (/dev/log).
[  OK  ] Listening on Journal Socket.
[  OK  ] Listening on User Database Manager Socket.
         Mounting Huge Pages File System...
         Mounting FUSE Control File System...
         Starting Journal Service...
         Starting Remount Root and Kernel File Systems...
[  OK  ] Mounted Huge Pages File System.
[  OK  ] Mounted FUSE Control File System.
[  OK  ] Finished Remount Root and Kernel File Systems.
         Starting Create Static Device Nodes in /dev...
[  OK  ] Finished Create Static Device Nodes in /dev.
[  OK  ] Reached target Local File Systems (Pre).
[  OK  ] Reached target Local File Systems.
         Starting Restore /run/initramfs on shutdown...
[  OK  ] Finished Restore /run/initramfs on shutdown.
[  OK  ] Started Journal Service.
         Starting Flush Journal to Persistent Storage...
[  OK  ] Finished Flush Journal to Persistent Storage.
         Starting Create Volatile Files and Directories...
[  OK  ] Finished Create Volatile Files and Directories.
         Starting Network Name Resolution...
         Starting Update UTMP about System Boot/Shutdown...
[  OK  ] Finished Update UTMP about System Boot/Shutdown.
[  OK  ] Reached target System Initialization.
[  OK  ] Started Daily Cleanup of Temporary Directories.
[  OK  ] Reached target Timers.
[  OK  ] Listening on D-Bus System Message Bus Socket.
[  OK  ] Reached target Sockets.
[  OK  ] Reached target Basic System.
         Starting Home Area Manager...
         Starting User Login Management...
         Starting Permit User Sessions...
[  OK  ] Finished Permit User Sessions.
[  OK  ] Started Console Getty.
[  OK  ] Reached target Login Prompts.
         Starting D-Bus System Message Bus...
[  OK  ] Started D-Bus System Message Bus.
[  OK  ] Started Home Area Manager.
[  OK  ] Started User Login Management.
[  OK  ] Reached target Multi-User System.
[  OK  ] Reached target Graphical Interface.
         Starting Update UTMP about System Runlevel Changes...
[  OK  ] Finished Update UTMP about System Runlevel Changes.
[  OK  ] Started Network Name Resolution.
[  OK  ] Reached target Host and Network Name Lookups.

Fedora 33 (Thirty Three) (built from systemd tree)
Kernel 5.9.11-arch2-1 on an x86_64 (console)
```

(cherry picked from commit ecd3c86b60655d892f5ec68549226cc5fa4be818)

4 years agomanager: Fix HW watchdog when systemd starts before driver loaded
Michael Marley [Tue, 8 Dec 2020 02:27:38 +0000 (21:27 -0500)]
manager: Fix HW watchdog when systemd starts before driver loaded

When manager_{set|override}_watchdog is called, set the watchdog timeout
regardless of whether the hardware watchdog was successfully initialized.  If
the watchdog was requested but could not be initialized, then instead of
pinging it, attempt to initialize it again.  This ensures that the hardware
watchdog is initialized even if the kernel module for it isn't loaded when
systemd starts (which is quite likely, unless it is compiled in).

This builds on work by @danc86 in https://github.com/systemd/systemd/pull/17460,
but fixes the issue of not updating the watchdog timeout with the actual value
from the hardware.

Fixes https://github.com/systemd/systemd/issues/17838

Co-authored-by: Dan Callaghan <djc@djc.id.au>
Co-authored-by: Michael Marley <michael@michaelmarley.com>
(cherry picked from commit 61927b9f116bf45bfdbf19dc2981d4a4f527ae5f)

4 years agocgroup: Also set blkio.bfq.weight
Pavel Hrdina [Wed, 25 Nov 2020 08:05:36 +0000 (09:05 +0100)]
cgroup: Also set blkio.bfq.weight

Commit [1] added a workaround when unified cgroups are used but missed
legacy cgroups where there is the same issue.

[1] <https://github.com/systemd/systemd/commit/2dbc45aea747f25cc1c3848fded2ec0062f96bcf>

Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
(cherry picked from commit 35e7a62ca32a30169a94693b831e53c832251984)

4 years agonss-resolve: varlink_call() set error_id only when r >= 0
Yu Watanabe [Sun, 6 Dec 2020 13:41:35 +0000 (22:41 +0900)]
nss-resolve: varlink_call() set error_id only when r >= 0

Fixes #17870.

(cherry picked from commit 3c229a9afb670025a3c1792bf5ac233b5a566ec5)

4 years agomissing: Define several syscall numbers for Alpha arch
Matt Turner [Sun, 6 Dec 2020 23:45:49 +0000 (18:45 -0500)]
missing: Define several syscall numbers for Alpha arch

(cherry picked from commit d6c0fdf4923420bd808094d02df9f05d739f4b17)

4 years agoDon't assume /run/systemd exists when creating unit-root
Daan De Meyer [Thu, 3 Dec 2020 23:08:21 +0000 (23:08 +0000)]
Don't assume /run/systemd exists when creating unit-root

When running tests in a mkosi container, /run/systemd might not
exist yet in the container which causes test-execute to fail.

Fixes #17842.

(cherry picked from commit 77f16dbd6d93f2b4a96984254cca25cab03479af)

4 years agomkosi: Add findutils to Fedora config
Daan De Meyer [Fri, 4 Dec 2020 22:23:11 +0000 (22:23 +0000)]
mkosi: Add findutils to Fedora config

find is needed by test_exec_dynamicuser.

(cherry picked from commit 8f2c4c5e93718ac6d6f790de61fced5ed143de5d)

4 years agomkosi: Add rpm to Fedora BuildPackages as it's needed by pkg-config
Daan De Meyer [Fri, 4 Dec 2020 20:05:19 +0000 (20:05 +0000)]
mkosi: Add rpm to Fedora BuildPackages as it's needed by pkg-config

(cherry picked from commit 8d036c446adc4dd8f9f1919916462752209c29a4)

4 years agomkosi: Replace iptables-dev with libiptc-dev in debian config
Daan De Meyer [Mon, 30 Nov 2020 23:03:52 +0000 (23:03 +0000)]
mkosi: Replace iptables-dev with libiptc-dev in debian config

Latest debian unstable doesn't have iptables-dev anymore. Instead,
let's use libiptc-dev.

(cherry picked from commit aef61dc888056454c543535e1410c3b299d623a5)

4 years agodissect: don't declare unused variables on archs that have no GPT discovery
Lennart Poettering [Thu, 3 Dec 2020 19:58:00 +0000 (20:58 +0100)]
dissect: don't declare unused variables on archs that have no GPT discovery

Fixes: #17839
(cherry picked from commit 62ea0ed08d0fba0bcf0a7fbde2d2ec1d317b66c7)

4 years agoresolved: synthesize NODATA instead of NXDOMAIN if gateway exists, but of other protocol
Lennart Poettering [Fri, 6 Nov 2020 09:26:44 +0000 (10:26 +0100)]
resolved: synthesize NODATA instead of NXDOMAIN if gateway exists, but of other protocol

Fixes: #11192
(cherry picked from commit 877884fc0da76170dce5abe7b3883eee334156b1)

4 years agolocal-addresses: make returning accumulated list optional
Lennart Poettering [Fri, 6 Nov 2020 09:26:40 +0000 (10:26 +0100)]
local-addresses: make returning accumulated list optional

(cherry picked from commit c3a8c6aa42092d87b73a3653d96258bbfe32fd28)

4 years agoresolved: improve log message when we use TCP a bit
Lennart Poettering [Tue, 10 Nov 2020 17:45:59 +0000 (18:45 +0100)]
resolved: improve log message when we use TCP a bit

DNS-over-TLS being in use isn't precisely the same as "UDP not
supported". Let's make this clearer.

(cherry picked from commit 6c0bacc146f1eea16f3b177a02fadf28901f9302)

4 years agonetwork: ignore broadcast address for /31 or /32 addresses
Yu Watanabe [Thu, 3 Dec 2020 01:19:35 +0000 (10:19 +0900)]
network: ignore broadcast address for /31 or /32 addresses

As they do not have broadcast address.
See https://tools.ietf.org/html/rfc3021

(cherry picked from commit 2a236f9fc0ff8fb2152032551436fde74da7217a)

4 years agonetwork: fix verification for broadcast address
Yu Watanabe [Wed, 2 Dec 2020 10:26:41 +0000 (19:26 +0900)]
network: fix verification for broadcast address

Fixes a bug caused by fe841414ef157f7f01d339c5d5730126e7b5fe0a.

(cherry picked from commit 05a7023d242b9012216c661f253df1c9c3d45b39)

4 years agonetwork: do not set broadcast if prefixlen is 31 or 32
Yu Watanabe [Wed, 2 Dec 2020 10:19:06 +0000 (19:19 +0900)]
network: do not set broadcast if prefixlen is 31 or 32

After fe841414ef157f7f01d339c5d5730126e7b5fe0a, broadcast address is
also compared with existing one to determine whether the address is
foregin or not. So, the address object should not contain unnecessary
information.

Fixes #17803.

(cherry picked from commit df8aa086420e6ce61e32898b2bb7c88b87fd1c43)

4 years agostub: don't ever respond to datagrams coming in on non-localhost addreses, on the...
Lennart Poettering [Thu, 5 Nov 2020 14:51:12 +0000 (15:51 +0100)]
stub: don't ever respond to datagrams coming in on non-localhost addreses, on the stub

(cherry picked from commit 565147b7bb2156277a2efaccd920d3ca9c21f8a8)

4 years agoresolved: never allow _gateway lookups to go to the network
Lennart Poettering [Wed, 11 Nov 2020 16:38:21 +0000 (17:38 +0100)]
resolved: never allow _gateway lookups to go to the network

Make them rather fail than go to the network.

Previously we'd filter them on LLMNR (explicitly) and MDNS (implicitly,
because it doesn't have .local suffix), but not on DNS.

In order to make _gateway truly reliable, let's not allow it to go to
DNS either, and keep it local.

This is particular relevant, as clients can now request lookups without
local RR synthesis, where we'd rather have NXDOMAIN returned for
_gateway than have it hit the network.

(cherry picked from commit fbbc72189f7844df8500bb10a58988f70bf90c99)

4 years agoresolved: lower SERVFAIL cache timeout from 30s to 10s
Lennart Poettering [Thu, 12 Nov 2020 16:52:09 +0000 (17:52 +0100)]
resolved: lower SERVFAIL cache timeout from 30s to 10s

Apparently 30s is a bit too long for some cases, see #5552. But not
caching SERVFAIL at all also breaks stuff, see explanation in
201d99584ed7af8078bb243ce2587e5455074713.

Let's try to find some middle ground, by lowering the cache timeout to
10s. This should be ample for the problem
201d99584ed7af8078bb243ce2587e5455074713 attackes, but not as long as
half a miute, as #5552 complains.

Fixes: #5552
(cherry picked from commit 19bcef9dc3fde342f138394333ab04d7e44b7da2)

4 years agodns-domain: try IDN2003 rules if IDN2008 doesn't work
Lennart Poettering [Mon, 9 Nov 2020 22:10:43 +0000 (23:10 +0100)]
dns-domain: try IDN2003 rules if IDN2008 doesn't work

This follows more closely what web browsers do, and makes sure emojis in
domains work.

Fixes: #14483
(cherry picked from commit d80e72ec602c2af2983842ad87e4443fce89d423)

4 years agovirt: Properly detect nested UML inside another hypervisor
Christopher Obbard [Wed, 2 Dec 2020 14:20:39 +0000 (14:20 +0000)]
virt: Properly detect nested UML inside another hypervisor

UML runs as a user-process so it can quite easily be ran inside of
another hypervisor, for instance inside a KVM instance. UML passes
through the CPUID from the host machine so in this case detect_vm
incorrectly identifies as running under KVM. So check we are running
a UML kernel first, before we check any other hypervisors.

Resolves: #17754

Signed-off-by: Christopher Obbard <chris.obbard@collabora.com>
(cherry picked from commit c8037dbf05da586b6a210ac04f145d99f424971f)

4 years agoresolved: properly check per-link NTA list
Lennart Poettering [Tue, 17 Nov 2020 17:31:53 +0000 (18:31 +0100)]
resolved: properly check per-link NTA list

We need to check for parent domains too. We did this correctly for the
system-wide NTA list, but not for the per-link one. Let's fix that.

(cherry picked from commit 7e8a93b77c3c4d4df1e8c3177dc9553c94fac759)

4 years agomeson: use '_' as separator in fuzz test names
Yu Watanabe [Wed, 2 Dec 2020 09:05:06 +0000 (18:05 +0900)]
meson: use '_' as separator in fuzz test names

Follow-up for d448888924c1d4815cb97bcd5d94419812c053b9 and ca121e20c42219e3bc4e5cb63dcc96cc5eae2879.

Fixes #17568.

(cherry picked from commit 0f82a2ab5c8d812791aca9686bdcc45f39c62431)

4 years agoman: mention that --key= is about *secret* keys
Lennart Poettering [Wed, 4 Nov 2020 15:23:11 +0000 (16:23 +0100)]
man: mention that --key= is about *secret* keys

(cherry picked from commit ce82de671f63105e9839978b432e2b015783ffc0)

4 years agomeson: check that cxx variable is set before using it
Luca Boccassi [Thu, 3 Dec 2020 16:17:43 +0000 (16:17 +0000)]
meson: check that cxx variable is set before using it

In some cases it is not defined. Eg in a yocto build:

src/systemd/meson.build:61:15: ERROR: Unknown variable cxx.

(cherry picked from commit 442bc2afee6c5f731c7b3e76ccab7301703a45a7)

5 years agoscope: on unified, make sure to unwatch all PIDs once they've been moved to the cgrou... v247.1
Franck Bui [Mon, 30 Nov 2020 14:26:15 +0000 (15:26 +0100)]
scope: on unified, make sure to unwatch all PIDs once they've been moved to the cgroup scope

Commit 428a9f6f1d0396b9eacde2b38d667cbe3f15eb55 freed u->pids which is
problematic since the references to this unit in m->watch_pids were no more
removed when the unit was freed.

This patch makes sure to clean all this refs up before freeing u->pids by
calling unit_unwatch_all_pids().

(cherry picked from commit e9eec8b5d2c106c5dd51382a155e6045c7c17c1a)

5 years agosd-event: fix delays assert brain-o (#17790)
Vito Caputo [Tue, 1 Dec 2020 08:26:54 +0000 (00:26 -0800)]
sd-event: fix delays assert brain-o (#17790)

s/sizeof/ELEMENTSOF/

Bug introduced in 34b87517749caa4142b19eb3c63bdf349fafbc49.

(cherry picked from commit cb9d621ebbfa30bbd620c17e143daeb0d78c12f0)

5 years agopo: add units/systemd-journald.service.in to POTFILES.skip
Piotr Drąg [Sat, 28 Nov 2020 12:43:52 +0000 (13:43 +0100)]
po: add units/systemd-journald.service.in to POTFILES.skip

(cherry picked from commit a75471ced1b4527870e21e96fbb81017e76fbcc8)

5 years agotest-networkd-conf: add missing assert_se
Zbigniew Jędrzejewski-Szmek [Sat, 28 Nov 2020 10:27:31 +0000 (11:27 +0100)]
test-networkd-conf: add missing assert_se

Coverity CID#1437351. Fixup for f706340979a.

(cherry picked from commit 78cafcbc7ec7eb19d40a07dfa75ad562f7ded643)

5 years agoNEWS: really fix kernel version number reference
Zbigniew Jędrzejewski-Szmek [Fri, 27 Nov 2020 11:51:48 +0000 (12:51 +0100)]
NEWS: really fix kernel version number reference

Fixes #17736.

(cherry picked from commit 832eedd192deb4e3520deb6a62c9acae23541b67)

5 years agooom: fix oomd.conf install location
Christian Hesse [Thu, 26 Nov 2020 20:29:30 +0000 (21:29 +0100)]
oom: fix oomd.conf install location

The file is read in pkgsysconfdir, so install it there.

5 years agohome: fix homed.conf install location
Christian Hesse [Thu, 26 Nov 2020 20:22:41 +0000 (21:22 +0100)]
home: fix homed.conf install location

The install location changed in d7aa78c32f076c305ceeb183fc06eb1de0960deb,
I think this was not intentional. Keep the condition but revert the path.

5 years agoMerge pull request #17738 from keszybz/hwdb-news-update-v247 v247
Zbigniew Jędrzejewski-Szmek [Thu, 26 Nov 2020 18:00:50 +0000 (19:00 +0100)]
Merge pull request #17738 from keszybz/hwdb-news-update-v247

Update hwdb and news for v247

5 years agoshared/seccomp-util: address family filtering is broken on ppc
Zbigniew Jędrzejewski-Szmek [Thu, 26 Nov 2020 10:23:54 +0000 (11:23 +0100)]
shared/seccomp-util: address family filtering is broken on ppc

This reverts the gist of da1921a5c396547261c8c7fcd94173346eb3b718 and
0d9fca76bb69e162265b2d25cb79f1890c0da31b (for ppc).

Quoting #17559:
> libseccomp 2.5 added socket syscall multiplexing on ppc64(el):
> https://github.com/seccomp/libseccomp/pull/229
>
> Like with i386, s390 and s390x this breaks socket argument filtering, so
> RestrictAddressFamilies doesn't work.
>
> This causes the unit test to fail:
> /* test_restrict_address_families */
> Operating on architecture: ppc
> Failed to install socket family rules for architecture ppc, skipping: Operation canceled
> Operating on architecture: ppc64
> Failed to add socket() rule for architecture ppc64, skipping: Invalid argument
> Operating on architecture: ppc64-le
> Failed to add socket() rule for architecture ppc64-le, skipping: Invalid argument
> Assertion 'fd < 0' failed at src/test/test-seccomp.c:424, function test_restrict_address_families(). Aborting.
>
> The socket filters can't be added so `socket(AF_UNIX, SOCK_DGRAM, 0);` still
> works, triggering the assertion.

Fixes #17559.

5 years agoNEWS: fix kernel version number reference
Zbigniew Jędrzejewski-Szmek [Thu, 26 Nov 2020 13:01:09 +0000 (14:01 +0100)]
NEWS: fix kernel version number reference

Fixes #17736.

5 years agoNEWS: version 247
Zbigniew Jędrzejewski-Szmek [Thu, 26 Nov 2020 12:45:06 +0000 (13:45 +0100)]
NEWS: version 247

5 years agohwdb: update chromiumos autosuspend rules
Zbigniew Jędrzejewski-Szmek [Thu, 26 Nov 2020 12:54:08 +0000 (13:54 +0100)]
hwdb: update chromiumos autosuspend rules

5 years agohwdb: update for v247
Zbigniew Jędrzejewski-Szmek [Thu, 26 Nov 2020 12:53:03 +0000 (13:53 +0100)]
hwdb: update for v247

Only some small changes, because we updated recently. As usual, it seems that there are mostly
additions with a smaller amount of corrections, no big removals.

5 years agodocs/RELEASE: clarify which steps are done when
Zbigniew Jędrzejewski-Szmek [Thu, 26 Nov 2020 12:48:35 +0000 (13:48 +0100)]
docs/RELEASE: clarify which steps are done when

5 years agotest: use cap_last_cap() for max supported cap number, not capability_list_length()
Dan Streetman [Wed, 25 Nov 2020 20:22:24 +0000 (15:22 -0500)]
test: use cap_last_cap() for max supported cap number, not capability_list_length()

This test assumes capability_list_length() is an invalid cap number,
but that isn't true if the running kernel supports more caps than we were
compiled with, which results in the test failing.

Instead use cap_last_cap() + 1.

If cap_last_cap() is 63, there are no more 'invalid' cap numbers to test with,
so the invalid cap number test part is skipped.

5 years agoTranslated using Weblate (Chinese (Simplified))
Whired Planck [Thu, 26 Nov 2020 11:35:55 +0000 (12:35 +0100)]
Translated using Weblate (Chinese (Simplified))

Currently translated at 83.9% (157 of 187 strings)

Co-authored-by: Whired Planck <fungdaat31@outlook.com>
Translate-URL: https://translate.fedoraproject.org/projects/systemd/master/zh_CN/
Translation: systemd/master

5 years agoMerge pull request #17622 from yuwata/udev-database-set-udev-version
Zbigniew Jędrzejewski-Szmek [Thu, 26 Nov 2020 11:48:05 +0000 (12:48 +0100)]
Merge pull request #17622 from yuwata/udev-database-set-udev-version

sd-device: make sd_device_has_current_tag() compatible with udev database generated by older udevd

5 years agoadd touchpad-edge-detect output for the Lenovo Thinkpad Extreme to evdev.hwdb
Lee Whalen [Thu, 26 Nov 2020 04:31:47 +0000 (20:31 -0800)]
add touchpad-edge-detect output for the Lenovo Thinkpad Extreme to evdev.hwdb

5 years agoMerge pull request #17709 from yuwata/test-seccomp-skip
Yu Watanabe [Thu, 26 Nov 2020 00:31:10 +0000 (09:31 +0900)]
Merge pull request #17709 from yuwata/test-seccomp-skip

test: skip several tests in test-seccomp when running under valgrind or ASAN

5 years agosd-device: make sd_device_has_current_tag() and friends compatible with database...
Yu Watanabe [Mon, 16 Nov 2020 10:47:42 +0000 (19:47 +0900)]
sd-device: make sd_device_has_current_tag() and friends compatible with database version 0

5 years agosd-device: introduce database version and save it in udev database V field
Yu Watanabe [Wed, 25 Nov 2020 21:23:14 +0000 (06:23 +0900)]
sd-device: introduce database version and save it in udev database V field

5 years agosd-device: make device_add_property_internal() inline
Yu Watanabe [Mon, 16 Nov 2020 10:28:40 +0000 (19:28 +0900)]
sd-device: make device_add_property_internal() inline

5 years agoman: slightly update the man page of sd_bus_message_read_basic()
Yu Watanabe [Wed, 25 Nov 2020 13:20:48 +0000 (22:20 +0900)]
man: slightly update the man page of sd_bus_message_read_basic()

Follow-up for 73a1d7d2433edd1872ec53db3e804009298ebb1d.

5 years agoman: Fixed an incomplete sentence
igo95862 [Wed, 25 Nov 2020 13:15:33 +0000 (16:15 +0300)]
man: Fixed an incomplete sentence

5 years agoman: Specify that sd_bus_message_read_basic returns 0 if end of array had been reached.
igo95862 [Wed, 25 Nov 2020 12:53:52 +0000 (15:53 +0300)]
man: Specify that sd_bus_message_read_basic returns 0 if end of array had been reached.

5 years agodocs: document what VPNs should do to systemd-resolved.service
Lennart Poettering [Fri, 20 Nov 2020 11:33:16 +0000 (12:33 +0100)]
docs: document what VPNs should do to systemd-resolved.service

Fixes: #17588 #17512

Prompted-by: #17529
(Also relevant: #6076)

5 years agocore/mount: mount command may fail after adding the corresponding proc mountinfo...
Yu Watanabe [Tue, 17 Nov 2020 00:13:59 +0000 (09:13 +0900)]
core/mount: mount command may fail after adding the corresponding proc mountinfo entry

Hopefully fixes #17617.

5 years agoMerge pull request #17715 from yuwata/specifier-follow-ups
Zbigniew Jędrzejewski-Szmek [Wed, 25 Nov 2020 12:22:56 +0000 (13:22 +0100)]
Merge pull request #17715 from yuwata/specifier-follow-ups

Sort specifiers

5 years agocryptsetup: use log_warning_errno() where we can, instead of log_warning()
Lennart Poettering [Tue, 24 Nov 2020 12:56:48 +0000 (13:56 +0100)]
cryptsetup: use log_warning_errno() where we can, instead of log_warning()