]> git.ipfire.org Git - thirdparty/systemd.git/log
thirdparty/systemd.git
3 years agoTranslated using Weblate (Turkish) 17102/head
Oğuz Ersen [Sat, 19 Sep 2020 12:22:13 +0000 (14:22 +0200)] 
Translated using Weblate (Turkish)

Currently translated at 100.0% (133 of 133 strings)

Co-authored-by: Oğuz Ersen <oguzersen@protonmail.com>
Translate-URL: https://translate.fedoraproject.org/projects/systemd/master/tr/
Translation: systemd/master

3 years agoTranslated using Weblate (Danish)
scootergrisen [Sat, 19 Sep 2020 12:22:13 +0000 (14:22 +0200)] 
Translated using Weblate (Danish)

Currently translated at 100.0% (187 of 187 strings)

Translated using Weblate (Danish)

Currently translated at 100.0% (187 of 187 strings)

Translated using Weblate (Danish)

Currently translated at 95.7% (179 of 187 strings)

Translated using Weblate (Danish)

Currently translated at 78.6% (147 of 187 strings)

Co-authored-by: scootergrisen <scootergrisen@gmail.com>
Translate-URL: https://translate.fedoraproject.org/projects/systemd/master/da/
Translation: systemd/master

3 years agoMerge pull request #17082 from poettering/nspawn-ctty-tweaks
Anita Zhang [Fri, 18 Sep 2020 21:26:14 +0000 (14:26 -0700)] 
Merge pull request #17082 from poettering/nspawn-ctty-tweaks

nspawn controlling tty tweaks

3 years agoupdate TODO
Lennart Poettering [Fri, 18 Sep 2020 20:26:50 +0000 (22:26 +0200)] 
update TODO

3 years agoMerge pull request #16981 from keszybz/use-crypt_ra
Lennart Poettering [Fri, 18 Sep 2020 17:46:08 +0000 (19:46 +0200)] 
Merge pull request #16981 from keszybz/use-crypt_ra

Use crypt_ra to allocate scratch area for password hashing

3 years agoUpdate 60-sensor.hwdb
Mitsuha_QuQ [Fri, 18 Sep 2020 01:33:31 +0000 (09:33 +0800)] 
Update 60-sensor.hwdb

Removed the blank

3 years agoUpdate 60-sensor.hwdb
Mitsuha_QuQ [Thu, 17 Sep 2020 14:56:00 +0000 (22:56 +0800)] 
Update 60-sensor.hwdb

Add Cube KNote 5 ACCEL matrix

3 years agotest: add test cases for RuntimeDirectoryPreserve=yes
Yu Watanabe [Tue, 1 Oct 2019 15:22:58 +0000 (00:22 +0900)] 
test: add test cases for RuntimeDirectoryPreserve=yes

3 years agoMerge pull request #17085 from yuwata/network-configure-mdb-entries-on-bridge-master
Zbigniew Jędrzejewski-Szmek [Fri, 18 Sep 2020 11:06:44 +0000 (13:06 +0200)] 
Merge pull request #17085 from yuwata/network-configure-mdb-entries-on-bridge-master

network: allow to configure mdb entries on bridge master

3 years agosocket: unified documentation for Accept to use 'yes/no' everywhere
Renaud Métrich [Fri, 18 Sep 2020 06:58:52 +0000 (08:58 +0200)] 
socket: unified documentation for Accept to use 'yes/no' everywhere

3 years agoMerge pull request #17092 from keszybz/libtool-excorcism
Lennart Poettering [Fri, 18 Sep 2020 08:14:09 +0000 (10:14 +0200)] 
Merge pull request #17092 from keszybz/libtool-excorcism

Libtool excorcism

3 years agoMerge pull request #17093 from poettering/dissect-tweaks2
Lennart Poettering [Fri, 18 Sep 2020 08:13:31 +0000 (10:13 +0200)] 
Merge pull request #17093 from poettering/dissect-tweaks2

dissect: some more tweaks

3 years agoMerge pull request #15050 from ssahani/fq-pie
Yu Watanabe [Thu, 17 Sep 2020 18:50:02 +0000 (03:50 +0900)] 
Merge pull request #15050 from ssahani/fq-pie

network: tc: add support for FQ-PIE packet scheduler

3 years agodissect: show partition output even if os-release is missing 17093/head
Lennart Poettering [Thu, 17 Sep 2020 15:52:00 +0000 (17:52 +0200)] 
dissect: show partition output even if os-release is missing

It's just too useful to immediately see with "systemd-dissect" what
"systemd-repart" generated for us without having to populate it with
/etc/os-release. Hence let's log a message if /etc/os-release is
missing, but proceed otherwise and show the partition table.

3 years agodissect: use structured initialization, it's prettier
Lennart Poettering [Thu, 17 Sep 2020 14:56:09 +0000 (16:56 +0200)] 
dissect: use structured initialization, it's prettier

3 years agodissect: is_loop_device() returns negative on error, don't mistake that is true
Lennart Poettering [Thu, 17 Sep 2020 15:24:33 +0000 (17:24 +0200)] 
dissect: is_loop_device() returns negative on error, don't mistake that is true

3 years agodissect: always invalidate secondary arch partitions if we found primary arch
Lennart Poettering [Thu, 17 Sep 2020 15:32:34 +0000 (17:32 +0200)] 
dissect: always invalidate secondary arch partitions if we found primary arch

Let's suppress the secondary arch data, since we never ever want to
mount it if we found the primary arch.

Previously we only suppressed in the Verity case, but there's little
reason to entertain the idea of a secondary arch in non-Verity
environments either, we are not going to use them, and should not do
decryption or anything like that.

3 years agodissect: three trivial fixes
Lennart Poettering [Thu, 17 Sep 2020 15:27:48 +0000 (17:27 +0200)] 
dissect: three trivial fixes

Uppercase first char of log message, and indicate correct program name.

Reindent comment table at one place.

Use correct, specific, enum type at one more place.

3 years agomkosi: update rules in mkosi.default.networkd-test boot 17092/head
Zbigniew Jędrzejewski-Szmek [Thu, 17 Sep 2020 13:51:34 +0000 (15:51 +0200)] 
mkosi: update rules in mkosi.default.networkd-test boot

We need to include ninja-build in the packages list because meson doesn't
depend on it (because it supports other backends too).

Also drop xz-devel, it's not crucial for the test.

3 years agohwdb: Chuwi Hi12 (#17042)
ErrantSpore [Thu, 17 Sep 2020 14:52:52 +0000 (08:52 -0600)] 
hwdb: Chuwi Hi12 (#17042)

Different bios version includes different pn. Submitted change fixed my inverted screen after reboot.
I've stepped back to the bios version 5.11 28.04.2016 in an attempt to troubleshoot Windows Hello locking the sign
in on reboot. The screen again rotated, but 90 degrees to the right. I created a new hwdb line to resolve that issue. The bios version changes the dmi string including the svn and pn.

3 years agonspawn: add --console=autopipe mode 17082/head
Lennart Poettering [Thu, 17 Sep 2020 14:26:14 +0000 (16:26 +0200)] 
nspawn: add --console=autopipe mode

By default we'll run a container in --console=interactive and
--console=read-only mode depending if we are invoked on a tty or not so
that the container always gets a /dev/console allocated, i.e is always
suitable to run a full init system /as those typically expect a
/dev/console to exist).

With the new --console=autopipe mode we do something similar, but
slightly different: when not invoked on a tty we'll use --console=pipe.
This means, if you invoke some tool in a container with this you'll get
full inetractivity if you invoke it on a tty but things will also be
very nicely pipeable. OTOH you cannot invoke a full init system like
this, because you might or might not become a /dev/console this way...

Prompted-by: #17070
(I named this "autopipe" rather than "auto" or so, since the default
mode probably should be named "auto" one day if we add a name for it,
and this is so similar to "auto" except that it uses pipes in the
non-tty case).

3 years agonspawn: don't become TTY controller just to undo it later again
Lennart Poettering [Wed, 16 Sep 2020 20:34:43 +0000 (22:34 +0200)] 
nspawn: don't become TTY controller just to undo it later again

Instead of first becoming a controlling process of the payload pty
as side effect of opening it (without O_NOCTTY), and then possibly
dropping it again, let's do it cleanly an reverse the logic: let's open
the pty without becoming its controller first. Only after everything
went the way we wanted it to go become the controller explicitly.

This has the benefit that the PID 1 stub process we run (as effect of
--as-pid2) doesn't have to lose the tty explicitly, but can just
continue running with things. And we explicitly make the tty controlling
right before invoking actual payload.

In order to make sure everything works as expected validate that the
stub PID 1 in the container really has no conrolling tty by issuing the
TIOCNOTTY tty and expecting ENOTTY, and log about it.

This shouldn't change behaviour much, it just makes thins a bit cleaner,
in particular as we'll not trigger SIGHUP on ourselves (since we are
controller and session leader) due to TIOCNOTTY which we then have to
explicitly ignore.

3 years agonspawn: fix fd leak on failure path
Lennart Poettering [Wed, 16 Sep 2020 20:16:10 +0000 (22:16 +0200)] 
nspawn: fix fd leak on failure path

3 years agonspawn: print log notice when we are invoked from a tty but in "pipe" mode
Lennart Poettering [Wed, 16 Sep 2020 20:12:29 +0000 (22:12 +0200)] 
nspawn: print log notice when we are invoked from a tty but in "pipe" mode

If people do this then things are weird, and they should probably use
--console=interactive (i.e. the default) instead.

Prompted-by: #17070
3 years agonspawn: check return of setsid()
Lennart Poettering [Wed, 16 Sep 2020 20:11:48 +0000 (22:11 +0200)] 
nspawn: check return of setsid()

Let's verify that everything works the way we expect it to work, hence
check setsid() return code.

3 years agoTranslated using Weblate (Italian)
Milo Casagrande [Tue, 15 Sep 2020 08:29:26 +0000 (10:29 +0200)] 
Translated using Weblate (Italian)

Currently translated at 100.0% (187 of 187 strings)

Co-authored-by: Milo Casagrande <milo@milo.name>
Translate-URL: https://translate.fedoraproject.org/projects/systemd/master/it/
Translation: systemd/master

3 years agopo: update Japanese translation
Yu Watanabe [Thu, 17 Sep 2020 11:51:15 +0000 (20:51 +0900)] 
po: update Japanese translation

3 years agomkosi: stop pulling in libtool
Zbigniew Jędrzejewski-Szmek [Thu, 17 Sep 2020 13:50:59 +0000 (15:50 +0200)] 
mkosi: stop pulling in libtool

Not useful since 72cdb3e783.

3 years agodissect: wrap verity settings in new VeritySettings structure
Lennart Poettering [Tue, 15 Sep 2020 20:09:08 +0000 (22:09 +0200)] 
dissect: wrap verity settings in new VeritySettings structure

Just some refactoring: let's place the various verity related parameters
in a common structure, and pass that around instead of the individual
parameters.

Also, let's load the PKCS#7 signature data when finding metadata
right-away, instead of delaying this until we need it. In all cases we
call this there's not much time difference between the metdata finding
and the loading, hence this simplifies things and makes sure root hash
data and its signature is now always acquired together.

3 years agoMerge pull request #16929 from ssahani/network-bare-udp
Yu Watanabe [Thu, 17 Sep 2020 11:34:15 +0000 (20:34 +0900)] 
Merge pull request #16929 from ssahani/network-bare-udp

network: introduce Bare UDP Tunnelling

3 years agoMerge pull request #17073 from ssahani/vxlan
Yu Watanabe [Thu, 17 Sep 2020 11:33:14 +0000 (20:33 +0900)] 
Merge pull request #17073 from ssahani/vxlan

network: Allow vxlan to be created without .network file

3 years agotest-network: add test for bridge MDB entries on bridge master 17085/head
Yu Watanabe [Thu, 17 Sep 2020 05:32:03 +0000 (14:32 +0900)] 
test-network: add test for bridge MDB entries on bridge master

3 years agonetwork: old kernel may not support to configure bridge MDB entries on bridge master
Yu Watanabe [Thu, 17 Sep 2020 09:02:09 +0000 (18:02 +0900)] 
network: old kernel may not support to configure bridge MDB entries on bridge master

3 years agonetwork: allow to configure bridge MDB entries on bridge master
Yu Watanabe [Thu, 17 Sep 2020 05:09:17 +0000 (14:09 +0900)] 
network: allow to configure bridge MDB entries on bridge master

3 years agotest-network: add tests for FQ-PIE 15050/head
Yu Watanabe [Thu, 17 Sep 2020 08:18:48 +0000 (17:18 +0900)] 
test-network: add tests for FQ-PIE

3 years agonetwork: tc: refuse to set 0 for FlowQueuePIE.PacketLimit=
Yu Watanabe [Thu, 17 Sep 2020 08:16:49 +0000 (17:16 +0900)] 
network: tc: refuse to set 0 for FlowQueuePIE.PacketLimit=

3 years agoudev: if a tablet has BTN_0, label it as ID_INPUT_TABLET_PAD
Peter Hutterer [Thu, 17 Sep 2020 03:06:19 +0000 (13:06 +1000)] 
udev: if a tablet has BTN_0, label it as ID_INPUT_TABLET_PAD

Graphics tablet devices comprise multiple event nodes, usually a Pen, Finger
and Pad node (that's how the kernel postfixes them). Pen and Pad are labeled
as ID_INPUT_TABLET but the pad doesn't actually send stylus events - it
doesn't usually have BTN_TOOL_PEN, merely BTN_STYLUS.

For the last several years, libwacom has set ID_INPUT_TABLET_PAD for all pad
devices known to it based on vid/pid and a "* Pad" name match. That does not
cover devices not in libwacom. libinput relies on ID_INPUT_TABLET_PAD to
initialize the pad backend.

We can't drop ID_INPUT_TABLET without breaking userspace, but we can add
ID_INPUT_TABLET_PAD ourselves - where a device has BTN_0 in addition to
BTN_STYLUS, let's add it as a pad.

There are some devices (notably: bamboos) that use BTN_LEFT instead of BTN_0
but they are relatively rare and there's a risk of mislabeling those devices,
so let's just stick with BTN_0 only.

3 years agonetwork: TC - introduce fq pie
Susant Sahani [Mon, 9 Mar 2020 13:03:49 +0000 (14:03 +0100)] 
network: TC - introduce fq pie

see https://patchwork.ozlabs.org/patch/1233253/

3 years agosd-netlink: add fq pie options
Susant Sahani [Mon, 9 Mar 2020 13:03:08 +0000 (14:03 +0100)] 
sd-netlink: add fq pie options

3 years agotest-network: add tests for BareUDP netdev 16929/head
Yu Watanabe [Thu, 17 Sep 2020 06:31:54 +0000 (15:31 +0900)] 
test-network: add tests for BareUDP netdev

3 years agoFix compilation without libcryptsetup (#17071)
masmullin2000 [Thu, 17 Sep 2020 07:07:03 +0000 (03:07 -0400)] 
Fix compilation without libcryptsetup (#17071)

3 years agotest-network: add test for Independent= for vxlan 17073/head
Yu Watanabe [Thu, 17 Sep 2020 06:58:41 +0000 (15:58 +0900)] 
test-network: add test for Independent= for vxlan

3 years agonetwork: Allow vxlan to be created without .network file
Susant Sahani [Wed, 16 Sep 2020 01:03:30 +0000 (01:03 +0000)] 
network: Allow vxlan to be created without .network file

3 years agonetwork: refuse to configure IPv4LL for bareudp netdev
Yu Watanabe [Thu, 17 Sep 2020 06:18:18 +0000 (15:18 +0900)] 
network: refuse to configure IPv4LL for bareudp netdev

3 years agonetwork: Introduce bare UDP
Susant Sahani [Mon, 14 Sep 2020 03:43:55 +0000 (03:43 +0000)] 
network: Introduce bare UDP

3 years agosd-netlink: Add support for bare UDP
Susant Sahani [Mon, 14 Sep 2020 03:42:35 +0000 (03:42 +0000)] 
sd-netlink: Add support for bare UDP

3 years agoutil: update kernel headers (v5.9-rc5)
Yu Watanabe [Thu, 17 Sep 2020 05:51:09 +0000 (14:51 +0900)] 
util: update kernel headers (v5.9-rc5)

3 years agoutil: wireguard is merged into upstream kernel
Yu Watanabe [Thu, 17 Sep 2020 05:49:47 +0000 (14:49 +0900)] 
util: wireguard is merged into upstream kernel

3 years agoMerge pull request #17059 from yuwata/network-mdb-follow-ups
Yu Watanabe [Thu, 17 Sep 2020 04:52:44 +0000 (13:52 +0900)] 
Merge pull request #17059 from yuwata/network-mdb-follow-ups

network: follow ups for bridge MDB support

3 years agoMerge pull request #17080 from poettering/repart-umask-fix
Anita Zhang [Wed, 16 Sep 2020 23:58:19 +0000 (16:58 -0700)] 
Merge pull request #17080 from poettering/repart-umask-fix

repart: don't mark image files generated with --empty=create executable

3 years agoMerge pull request #17067 from keszybz/rc-local-path
Zbigniew Jędrzejewski-Szmek [Wed, 16 Sep 2020 20:59:53 +0000 (22:59 +0200)] 
Merge pull request #17067 from keszybz/rc-local-path

Expose the path to rc.local in the man page

3 years agorepart: don't mark image files generated with --empty=create executable 17080/head
Lennart Poettering [Wed, 16 Sep 2020 15:51:25 +0000 (17:51 +0200)] 
repart: don't mark image files generated with --empty=create executable

3 years agoMerge pull request #17076 from poettering/dissect-cleanup
Lennart Poettering [Wed, 16 Sep 2020 16:42:12 +0000 (18:42 +0200)] 
Merge pull request #17076 from poettering/dissect-cleanup

minor cleanups to the dissector code

3 years agotest-execute: kill Waldo and Quux
Topi Miettinen [Wed, 16 Sep 2020 11:09:55 +0000 (14:09 +0300)] 
test-execute: kill Waldo and Quux

Retire /var/lib{,/private}/{quux,waldo} after they have served their purpose.

3 years agotest-network: add test for [BridgeMDB] section 17059/head
Yu Watanabe [Tue, 15 Sep 2020 07:28:09 +0000 (16:28 +0900)] 
test-network: add test for [BridgeMDB] section

3 years agonetwork: add debug message for configuring MDB entries
Yu Watanabe [Tue, 15 Sep 2020 07:27:24 +0000 (16:27 +0900)] 
network: add debug message for configuring MDB entries

3 years agonetwork: configure bridge MDB entries after bridge has carrier
Yu Watanabe [Tue, 15 Sep 2020 07:20:36 +0000 (16:20 +0900)] 
network: configure bridge MDB entries after bridge has carrier

3 years agonetwork: drop [BridgeMDB] entries if Bridge= is not set
Yu Watanabe [Tue, 15 Sep 2020 07:07:45 +0000 (16:07 +0900)] 
network: drop [BridgeMDB] entries if Bridge= is not set

3 years agonetwork: check MulticastGroupAddress= is neither a local multicast address nor all...
Yu Watanabe [Tue, 15 Sep 2020 06:37:25 +0000 (15:37 +0900)] 
network: check MulticastGroupAddress= is neither a local multicast address nor all nodes address

3 years agoin-addr-util: introduce in4_addr_is_local_multicast()
Yu Watanabe [Tue, 15 Sep 2020 06:34:22 +0000 (15:34 +0900)] 
in-addr-util: introduce in4_addr_is_local_multicast()

3 years agoin-addr-util: introduce in6_addr_is_link_local_all_nodes()
Yu Watanabe [Tue, 15 Sep 2020 06:33:56 +0000 (15:33 +0900)] 
in-addr-util: introduce in6_addr_is_link_local_all_nodes()

3 years agonetwork: drop unnecessary headers
Yu Watanabe [Tue, 15 Sep 2020 07:40:34 +0000 (16:40 +0900)] 
network: drop unnecessary headers

3 years agonetwork: move functions in networkd-mdb.c
Yu Watanabe [Tue, 15 Sep 2020 07:34:23 +0000 (16:34 +0900)] 
network: move functions in networkd-mdb.c

3 years agoMerge pull request #17066 from keszybz/allow-loopback-addresses
Yu Watanabe [Wed, 16 Sep 2020 15:02:08 +0000 (00:02 +0900)] 
Merge pull request #17066 from keszybz/allow-loopback-addresses

Allow loopback addresses

3 years agoMerge pull request #16675 from topimiettinen/exec-syscall-error-action
Lennart Poettering [Wed, 16 Sep 2020 15:01:44 +0000 (17:01 +0200)] 
Merge pull request #16675 from topimiettinen/exec-syscall-error-action

SystemCallFilter, SystemCallErrorAction error actions kill and log

3 years agoman: reorder autogenerated dbus api lists
Zbigniew Jędrzejewski-Szmek [Wed, 16 Sep 2020 08:01:13 +0000 (10:01 +0200)] 
man: reorder autogenerated dbus api lists

3e5f04bf6468fcb79c080f02b0eab08f258bff0c was trying to do the right thing, but
the resulting list does not match the autogenerated order (which is the same as
the order in vtable definition). I assume the addition was done manually. Fix
the order so that dbus-docs-fresh test is not unhappy.

3 years agogpt: move GPT_LINUX_GENERIC definition next to the user GPT partition types 17076/head
Lennart Poettering [Wed, 16 Sep 2020 14:31:00 +0000 (16:31 +0200)] 
gpt: move GPT_LINUX_GENERIC definition next to the user GPT partition types

No code changes, just some reordering.

3 years agodissect: introduce PartitionDesignator as real type
Lennart Poettering [Tue, 15 Sep 2020 16:02:28 +0000 (18:02 +0200)] 
dissect: introduce PartitionDesignator as real type

3 years agodissect: rename mount_options_from_part() → mount_options_from_designator()
Lennart Poettering [Tue, 15 Sep 2020 16:02:28 +0000 (18:02 +0200)] 
dissect: rename mount_options_from_part() → mount_options_from_designator()

After all, it actually takes a designator argument, not a partition
index or so.

3 years agoman: substitute path to rc.local in the man page 17067/head
Zbigniew Jędrzejewski-Szmek [Tue, 15 Sep 2020 15:56:32 +0000 (17:56 +0200)] 
man: substitute path to rc.local in the man page

Different systems use different paths for it and users are confused when the
man page has a path different than the one on the local system.

https://bugzilla.redhat.com/show_bug.cgi?id=1876905

3 years agotest: remove test-dissect-image
Lennart Poettering [Tue, 15 Sep 2020 15:53:32 +0000 (17:53 +0200)] 
test: remove test-dissect-image

It does pretty much exactly what systemd-dissect does and is a manual
test, hence let's remove it as redundant code.

3 years agomeson: RC_LOCAL_SCRIPT_PATH_START to RC_LOCAL_PATH
Zbigniew Jędrzejewski-Szmek [Tue, 15 Sep 2020 15:44:26 +0000 (17:44 +0200)] 
meson: RC_LOCAL_SCRIPT_PATH_START to RC_LOCAL_PATH

RC_LOCAL_SCRIPT_PATH_START and RC_LOCAL_SCRIPT_PATH_STOP were was originally
added in the conversion to meson based on the autotools name. In
44508946534eee032927c263b79464832656dd6e RC_LOCAL_SCRIPT_PATH_STOP was dropped.
We don't need to use such a long name.

3 years agosocket-proxy: allow localhost addresses 17066/head
Zbigniew Jędrzejewski-Szmek [Tue, 15 Sep 2020 12:52:47 +0000 (14:52 +0200)] 
socket-proxy: allow localhost addresses

With this change, only manager_connect() in timesync.d uses AI_ADDRCONFIG. There
we are connecting to a remove server, so the flag is appropriate.

3 years agosd-bus: allow localhost addresses
Zbigniew Jędrzejewski-Szmek [Tue, 15 Sep 2020 12:49:36 +0000 (14:49 +0200)] 
sd-bus: allow localhost addresses

By settings AI_ADDRCONFIG in hints we cannot for example resolve "localhost"
when the local machine only has a loopback interface. This seems like an
unnecessary restriction, drop it.

Inspired by https://bugzilla.redhat.com/show_bug.cgi?id=1839007.

3 years agofstab-generator: add 'nofail' when NFS 'bg' option is used
Franck Bui [Tue, 15 Sep 2020 09:49:09 +0000 (11:49 +0200)] 
fstab-generator: add 'nofail' when  NFS 'bg' option is used

Otherwise a 'Before=remote-fs.target' dependency is added to the mount unit
which defeats the purpose of 'bg' option.

3 years agotests: try to allocate a delegated scope unit for our cgroup tests
Lennart Poettering [Mon, 14 Sep 2020 16:18:27 +0000 (18:18 +0200)] 
tests: try to allocate a delegated scope unit for our cgroup tests

Prompted-by: #16632
3 years agoexec: SystemCallLog= directive 16675/head
Topi Miettinen [Sun, 30 Aug 2020 18:56:13 +0000 (21:56 +0300)] 
exec: SystemCallLog= directive

With new directive SystemCallLog= it's possible to list system calls to be
logged. This can be used for auditing or temporarily when constructing system
call filters.

---
v5: drop intermediary, update HASHMAP_FOREACH_KEY() use
v4: skip useless debug messages, actually parse directive
v3: don't declare unused variables with old libseccomp
v2: fix build without seccomp or old libseccomp

3 years agoexec: Add kill action to system call filters
Topi Miettinen [Wed, 5 Aug 2020 13:31:26 +0000 (16:31 +0300)] 
exec: Add kill action to system call filters

Define explicit action "kill" for SystemCallErrorNumber=.

In addition to errno code, allow specifying "kill" as action for
SystemCallFilter=.

---
v7: seccomp_parse_errno_or_action() returns -EINVAL if !HAVE_SECCOMP
v6: use streq_ptr(), let errno_to_name() handle bad values, kill processes,
 init syscall_errno
v5: actually use seccomp_errno_or_action_to_string(), don't fail bus unit
parsing without seccomp
v4: fix build without seccomp
v3: drop log action
v2: action -> number

3 years agotest-libcrypt-util: before doing anything check what methods are available 16981/head
Zbigniew Jędrzejewski-Szmek [Tue, 15 Sep 2020 07:30:39 +0000 (09:30 +0200)] 
test-libcrypt-util: before doing anything check what methods are available

On centos7 ci:

--- test-libcrypt-util begin ---
Found container virtualization none.
/* test_hash_password */
ew3bU1.hoKk4o: yes
$1$gc5rWpTB$wK1aul1PyBn9AX1z93stk1: no
$2b$12$BlqcGkB/7BFvNMXKGxDea.5/8D6FTny.cbNcHW/tqcrcyo6ZJd8u2: no
$5$lGhDrcrao9zb5oIK$05KlOVG3ocknx/ThreqXE/gk.XzFFBMTksc4t2CPDUD: no
$6$c7wB/3GiRk0VHf7e$zXJ7hN0aLZapE.iO4mn/oHu6.prsXTUG/5k1AxpgR85ELolyAcaIGRgzfwJs3isTChMDBjnthZyaMCfCNxo9I.: no
$y$j9T$$9cKOWsAm4m97WiYk61lPPibZpy3oaGPIbsL4koRe/XD: no

3 years agotest-libcrypt-util: skip test on ppc64 with no xcrypt
Zbigniew Jędrzejewski-Szmek [Tue, 15 Sep 2020 07:18:31 +0000 (09:18 +0200)] 
test-libcrypt-util: skip test on ppc64 with no xcrypt

I'm tired of trying to figure this out.

3 years agoshared/libcrypt-util: do not refuse passwords if some other hash is unsupported
Zbigniew Jędrzejewski-Szmek [Fri, 11 Sep 2020 06:27:43 +0000 (08:27 +0200)] 
shared/libcrypt-util: do not refuse passwords if some other hash is unsupported

3 years agoshared/libcrypt-util: add fallback for crypt_ra()
Zbigniew Jędrzejewski-Szmek [Thu, 10 Sep 2020 16:30:22 +0000 (18:30 +0200)] 
shared/libcrypt-util: add fallback for crypt_ra()

Following the style in missing_syscall.h, we use a non-conflicting name
for the function and use a macro to map to the real name to the replacement.

3 years agoCheck for crypt_gensalt_ra() instead of relying on libxcrypt presence
Zbigniew Jędrzejewski-Szmek [Thu, 10 Sep 2020 16:21:41 +0000 (18:21 +0200)] 
Check for crypt_gensalt_ra() instead of relying on libxcrypt presence

Since the loop to check various xcrypt functions is already in place,
adding one more is cheap. And it is nicer to check for the function
directly. People like to backport things, so we might get lucky even
without having libxcrypt.

3 years agomeson: test if we have libcrypt_ra
Zbigniew Jędrzejewski-Szmek [Tue, 8 Sep 2020 18:02:31 +0000 (20:02 +0200)] 
meson: test if we have libcrypt_ra

We always seem to have either libcrypt_r and not the other two, or all
three. So the fallback for libcrypt_ra needs to be based on libcrypt_r.

3 years agoshared/libcrypt-util: include fewer headers
Zbigniew Jędrzejewski-Szmek [Tue, 8 Sep 2020 16:26:24 +0000 (18:26 +0200)] 
shared/libcrypt-util: include fewer headers

Now that we wrap crypt_r/ra uses, we can include the header only in libcrypt-util.c.

3 years agoMake test_password_{one,many} also use crypt_ra()
Zbigniew Jędrzejewski-Szmek [Tue, 8 Sep 2020 14:27:47 +0000 (16:27 +0200)] 
Make test_password_{one,many} also use crypt_ra()

3 years agoAdd reciprocal test for password hashing and checking
Zbigniew Jędrzejewski-Szmek [Tue, 8 Sep 2020 13:28:21 +0000 (15:28 +0200)] 
Add reciprocal test for password hashing and checking

3 years agoMove test_password_{one,many} to libcrypt-util.c
Zbigniew Jędrzejewski-Szmek [Tue, 8 Sep 2020 13:21:21 +0000 (15:21 +0200)] 
Move test_password_{one,many} to libcrypt-util.c

They are only used under src/home/, but I want to add tests in test-libcrypt-util.c.
And the functions are almost trivial, so I think it is OK to move them to shared.

3 years agobusctl: add missing shortopt -l
Juergen Hoetzel [Tue, 15 Sep 2020 06:09:51 +0000 (08:09 +0200)] 
busctl: add missing shortopt -l

3 years agocore: use strv_free_and_replace() at one more place
Yu Watanabe [Sat, 15 Aug 2020 09:26:23 +0000 (18:26 +0900)] 
core: use strv_free_and_replace() at one more place

3 years agoshared/libcrypt-util: use libcrypt_ra()
Zbigniew Jędrzejewski-Szmek [Tue, 8 Sep 2020 13:13:44 +0000 (15:13 +0200)] 
shared/libcrypt-util: use libcrypt_ra()

This lets the libc/xcrypt allocate as much storage area as it needs.
Should fix #16965:

testsuite-46.sh[74]: ==74==ERROR: AddressSanitizer: stack-buffer-overflow on address 0x7f3e972e1080 at pc 0x7f3e9be8deed bp 0x7ffce4f28530 sp 0x7ffce4f27ce0
testsuite-46.sh[74]: WRITE of size 131232 at 0x7f3e972e1080 thread T0
testsuite-46.sh[74]:     #0 0x7f3e9be8deec  (/usr/lib/clang/10.0.1/lib/linux/libclang_rt.asan-x86_64.so+0x9feec)
testsuite-46.sh[74]:     #1 0x559cd05a6412 in user_record_make_hashed_password /systemd-meson-build/../build/src/home/user-record-util.c:818:21
testsuite-46.sh[74]:     #2 0x559cd058fb03 in create_home /systemd-meson-build/../build/src/home/homectl.c:1112:29
testsuite-46.sh[74]:     #3 0x7f3e9b5b3058 in dispatch_verb /systemd-meson-build/../build/src/shared/verbs.c:103:24
testsuite-46.sh[74]:     #4 0x559cd058c101 in run /systemd-meson-build/../build/src/home/homectl.c:3325:16
testsuite-46.sh[74]:     #5 0x559cd058c00a in main /systemd-meson-build/../build/src/home/homectl.c:3328:1
testsuite-46.sh[74]:     #6 0x7f3e9a88b151 in __libc_start_main (/usr/lib/libc.so.6+0x28151)
testsuite-46.sh[74]:     #7 0x559cd0583e7d in _start (/usr/bin/homectl+0x24e7d)
testsuite-46.sh[74]: Address 0x7f3e972e1080 is located in stack of thread T0 at offset 32896 in frame
testsuite-46.sh[74]:     #0 0x559cd05a60df in user_record_make_hashed_password /systemd-meson-build/../build/src/home/user-record-util.c:789
testsuite-46.sh[74]:   This frame has 6 object(s):
testsuite-46.sh[74]:     [32, 40) 'priv' (line 790)
testsuite-46.sh[74]:     [64, 72) 'np' (line 791)
testsuite-46.sh[74]:     [96, 104) 'salt' (line 809)
testsuite-46.sh[74]:     [128, 32896) 'cd' (line 810)
testsuite-46.sh[74]:     [33152, 33168) '.compoundliteral' <== Memory access at offset 32896 partially underflows this variable
testsuite-46.sh[74]:     [33184, 33192) 'new_array' (line 832) <== Memory access at offset 32896 partially underflows this variable
testsuite-46.sh[74]: HINT: this may be a false positive if your program uses some custom stack unwind mechanism, swapcontext or vfork
testsuite-46.sh[74]:       (longjmp and C++ exceptions *are* supported)
testsuite-46.sh[74]: SUMMARY: AddressSanitizer: stack-buffer-overflow (/usr/lib/clang/10.0.1/lib/linux/libclang_rt.asan-x86_64.so+0x9feec)

It seems 'struct crypt_data' is 32896 bytes, but libclang_rt wants more, at least 33168?

3 years agoAdd a helper function that does make_salt+crypt_r
Zbigniew Jędrzejewski-Szmek [Tue, 8 Sep 2020 12:52:13 +0000 (14:52 +0200)] 
Add a helper function that does make_salt+crypt_r

No functional change.

3 years agoMerge pull request #16998 from zonque/networkd/mdb
Yu Watanabe [Tue, 15 Sep 2020 07:30:39 +0000 (16:30 +0900)] 
Merge pull request #16998 from zonque/networkd/mdb

network: add support for setting MDB entries

3 years agosocket-util: reset length argument for second getsockopt call in fd_[gs]et_rcvbuf
Christian Göttsche [Mon, 14 Sep 2020 17:40:42 +0000 (19:40 +0200)] 
socket-util: reset length argument for second getsockopt call in fd_[gs]et_rcvbuf

In case the first getsockopt() call changes the value.

3 years agoMerge pull request #16976 from keszybz/systemctl-service-log-levels
Yu Watanabe [Tue, 15 Sep 2020 00:34:07 +0000 (09:34 +0900)] 
Merge pull request #16976 from keszybz/systemctl-service-log-levels

systemctl: add service-log-{level,target} verbs

3 years agoMerge pull request #17055 from keszybz/two-coverity-fixes
Yu Watanabe [Tue, 15 Sep 2020 00:22:39 +0000 (09:22 +0900)] 
Merge pull request #17055 from keszybz/two-coverity-fixes

Two coverity fixes

3 years agoMerge pull request #17049 from mrc0mmand/code-and-spell-check
Lennart Poettering [Mon, 14 Sep 2020 21:00:02 +0000 (23:00 +0200)] 
Merge pull request #17049 from mrc0mmand/code-and-spell-check

tree-wide: assorted cleanups/fixes

3 years agobasic/copy: vodify unlinkat() call 17055/head
Zbigniew Jędrzejewski-Szmek [Mon, 14 Sep 2020 20:34:43 +0000 (22:34 +0200)] 
basic/copy: vodify unlinkat() call

Not much we can do at this point.

Coverity CID#1432670.

3 years agotest: use pclose() for popen()
Zbigniew Jędrzejewski-Szmek [Mon, 14 Sep 2020 20:32:52 +0000 (22:32 +0200)] 
test: use pclose() for popen()

Coverity CID#1432701.

3 years agosystemctl: give a nice hint about org.freedesktop.LogControl1 when applicable 16976/head
Zbigniew Jędrzejewski-Szmek [Fri, 11 Sep 2020 14:00:22 +0000 (16:00 +0200)] 
systemctl: give a nice hint about org.freedesktop.LogControl1 when applicable