platform/upstream/systemd.git
5 years agoMerge pull request #13344 from ddstreet/test-functions
Yu Watanabe [Sun, 18 Aug 2019 15:34:35 +0000 (00:34 +0900)]
Merge pull request #13344 from ddstreet/test-functions

Minor/cosmetic changes to test/test-functions, some test.sh

5 years agotest: increase qemu timeout for TEST-18 and TEST-19
Dan Streetman [Thu, 15 Aug 2019 20:27:05 +0000 (16:27 -0400)]
test: increase qemu timeout for TEST-18 and TEST-19

These tests runs under qemu, and on some testbeds, without acceleration.
On those systems, the current 180 second overall test timeout is too
short to run the test.

Increasing the timeout to 600s should be enough, even for slow
non-accelerated qemu testbeds.

5 years agoman: add missing comma
Theo Ouzhinski [Sat, 17 Aug 2019 17:21:50 +0000 (13:21 -0400)]
man: add missing comma

5 years agotrivial spelling fixes (#13339)
Mr-Foo [Sat, 17 Aug 2019 16:30:57 +0000 (16:30 +0000)]
trivial spelling fixes (#13339)

5 years agotest/test-functions: add mkdir to import_initdir
Dan Streetman [Sat, 17 Aug 2019 16:24:00 +0000 (12:24 -0400)]
test/test-functions: add mkdir to import_initdir

This dir is created by create_empty_image_rootdir, as well as indirectly
by some other functions, but it should be created by import_initdir so
the newly-exported $initdir exists and can be used immediately without
relying on other functions to create it.

5 years agotest: replace $TESTDIR/root with $initdir
Dan Streetman [Thu, 15 Aug 2019 13:39:31 +0000 (09:39 -0400)]
test: replace $TESTDIR/root with $initdir

The $initdir var is already set to $TESTDIR/root, it should be used
instead of direct use of $TESTDIR/root.

5 years agotest: TEST-33 and TEST-36 should use create_empty_image_rootdir
Dan Streetman [Sat, 17 Aug 2019 16:20:56 +0000 (12:20 -0400)]
test: TEST-33 and TEST-36 should use create_empty_image_rootdir

This recently added function handles mkdir and mounting of the empty image,
as well as umounting after test_setup.

5 years agotest/test-functions: avoid stderr noise, only umount on cleanup if mountpoint
Dan Streetman [Tue, 13 Aug 2019 11:50:59 +0000 (07:50 -0400)]
test/test-functions: avoid stderr noise, only umount on cleanup if mountpoint

Only umount it during cleanup if the $TESTDIR/root dir is a mountpoint.
This avoids adding noise to the stderr log such as:

mountpoint: /var/tmp/systemd-test.waLOFT/root: No such file or directory

5 years agoMerge pull request #13317 from ddstreet/werror
Zbigniew Jędrzejewski-Szmek [Fri, 16 Aug 2019 08:19:09 +0000 (10:19 +0200)]
Merge pull request #13317 from ddstreet/werror

Fix build warnings, so Ubuntu CI can pass --werror to meson

5 years ago po: update Ukrainian translation (#13329)
Yuri Chornoivan [Fri, 16 Aug 2019 07:55:19 +0000 (10:55 +0300)]
 po: update Ukrainian translation (#13329)

* Update Ukrainian translation

* po: update Ukrainian translation

5 years agounit-file: downgrade accidentaly high-prio debug log message
Lennart Poettering [Thu, 15 Aug 2019 10:34:31 +0000 (12:34 +0200)]
unit-file: downgrade accidentaly high-prio debug log message

5 years agonetworkctl: avoid outputting '(null)' for LLDP ports without description
Lennart Poettering [Thu, 15 Aug 2019 17:28:06 +0000 (19:28 +0200)]
networkctl: avoid outputting '(null)' for LLDP ports without description

5 years agosrc/boot/efi/meson.build: if meson --werror is true, set gcc -Werror
Dan Streetman [Tue, 13 Aug 2019 00:34:43 +0000 (20:34 -0400)]
src/boot/efi/meson.build: if meson --werror is true, set gcc -Werror

This part of the build does not use the normal meson parameters, so
we need to explicitly check for the meson --werror parameter, and if
it's true, set the gcc -Werror parameter for this subdir's build.

5 years agosrc/boot/efi/linux: elide __attribute__((regparm(0))) on non-i386
Dan Streetman [Tue, 13 Aug 2019 11:02:33 +0000 (07:02 -0400)]
src/boot/efi/linux: elide __attribute__((regparm(0))) on non-i386

This attribute is x86_32-only, so when building on non-intel archs it
generates a compiler warning.  When building with -Werror this turns
into an error, so only include the attribute on i386 arch builds.

5 years agosrc/boot/efi/shim: elide __attribute__((sysv_abi)) on non-intel archs
Dan Streetman [Tue, 13 Aug 2019 10:45:04 +0000 (06:45 -0400)]
src/boot/efi/shim: elide __attribute__((sysv_abi)) on non-intel archs

This attribute is x86-only, so when building on non-intel archs it
generates a compiler warning.  When building with -Werror this turns
into an error, so only include the attribute on intel archs.

5 years agosrc/basic/missing_syscall: add comment lines for PR 13319 changes
Dan Streetman [Thu, 15 Aug 2019 01:08:36 +0000 (21:08 -0400)]
src/basic/missing_syscall: add comment lines for PR 13319 changes

Add a comment line explaining that the syscall defines might be
defined to invalid negative numbers, as libseccomp redefines them
to negative numbers if not defined by the kernel headers, which is
not obvious just from reading the code checking for defined && > 0

5 years agoREADME: add an OSS-Fuzz badge
Evgeny Vereshchagin [Thu, 15 Aug 2019 07:23:55 +0000 (07:23 +0000)]
README: add an OSS-Fuzz badge

Since bug reports, backtraces, coverage reports and build logs are scattered
across at least four different places and there is no publicly available dashboards
the badge can point to, let's just point it to the build logs, which hopefully are going to be
a little bit more usable once https://github.com/google/oss-fuzz/issues/2690 is
addressed.

5 years agoupdate-utmp: fix assertion failure if rescue.target, multi-user.target and graphical...
Tommi Rantala [Mon, 5 Aug 2019 11:01:58 +0000 (14:01 +0300)]
update-utmp: fix assertion failure if rescue.target, multi-user.target and graphical.target are all inactive

If rescue.target, multi-user.target and graphical.target are all
inactive, get_current_runlevel() is not able to determine current
runlevel, and returns with zero. This zero runlevel value results to
assertion failure in utmp_put_runlevel().

 # systemctl stop rescue.target multi-user.target graphical.target
 # systemctl start systemd-update-utmp-runlevel.service

 systemd[1]: Stopped target Graphical Interface.
 systemd[1]: Stopped target Multi-User System.
 systemd[1]: Starting Update UTMP about System Runlevel Changes...
 systemd-update-utmp[67]: Assertion 'runlevel > 0' failed at src/shared/utmp-wtmp.c:275, function utmp_put_runlevel(). Aborting.
 systemd[1]: systemd-update-utmp-runlevel.service: Main process exited, code=dumped, status=6/ABRT
 systemd[1]: systemd-update-utmp-runlevel.service: Failed with result 'core-dump'.
 systemd[1]: Failed to start Update UTMP about System Runlevel Changes.

Let's just print a warning in this case and skip the utmp update, to
avoid systemd-update-utmp-runlevel.service failures.

5 years agoMerge pull request #13273 from RPigott/zshcomp
Zbigniew Jędrzejewski-Szmek [Wed, 14 Aug 2019 16:21:12 +0000 (18:21 +0200)]
Merge pull request #13273 from RPigott/zshcomp

Enable resolvectl zsh completions

5 years agosysusers: properly mark generated accounts as locked
Lennart Poettering [Wed, 7 Aug 2019 10:35:29 +0000 (12:35 +0200)]
sysusers: properly mark generated accounts as locked

Previously, we'd only set the shell to /usr/bin/nologin and lock the
password for system users. Let's go one step further and also lock the
whole account.

This is a paranoid safety precaution, since neither disabling the shell
like this nor disabling the password is sufficient to lock an account,
since remote shell tools generally allow passing different shells, and
logins into ftp or similar protocols don't know the shell concept anyway.
Moreover, in times of ssh authentication by password is just one
option of authentication among many.

Takes inspiration from the recommendations in usermod(8)'s -L switch:

    "Note: if you wish to lock the account (not only access with a
    password), you should also set the EXPIRE_DATE to 1."

5 years agoMerge pull request #13298 from RPigott/busctl
Zbigniew Jędrzejewski-Szmek [Wed, 14 Aug 2019 16:18:55 +0000 (18:18 +0200)]
Merge pull request #13298 from RPigott/busctl

zsh: add busctl completions

5 years agohwdb: acceleration mount matrix for a Medion Akoya E3221 (#13310)
EtherGraf [Wed, 14 Aug 2019 16:15:50 +0000 (18:15 +0200)]
hwdb: acceleration mount matrix for a Medion Akoya E3221 (#13310)

5 years agosrc/basic/missing_syscall: change #ifndef to #if ! (defined && > 0)
Dan Streetman [Thu, 25 Jul 2019 11:57:30 +0000 (07:57 -0400)]
src/basic/missing_syscall: change #ifndef to #if ! (defined && > 0)

The #ifndef check used to work for missing __NR_* syscall defines, but
unfortunately libseccomp now redefines missing syscall number to negative
numbers, in their public header file, e.g.:
https://github.com/seccomp/libseccomp/blob/master/include/seccomp.h.in#L801

When systemd is built, since it includes <seccomp.h>, it pulls in the
incorrect negative value for any __NR_* syscall define that's included in
the seccomp.h header (for those syscalls that the kernel headers don't
yet define, e.g. when built with older/stable-distro kernels).  This leads
to bugs like:
https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1821625

This changes the check so that it can override the negative number that
libseccomp defines, instead of trying to use the negative syscall number.
To avoid gcc warnings (which are failures with meson --werror), this checks
without generating a redefinition gcc warning.

I have no idea why libseccomp decided to define missing syscalls
to negative numbers inside their *public* header file, causing
problems like this.

5 years agoshared/unit-file: fix systemctl cat user@.service
Zbigniew Jędrzejewski-Szmek [Tue, 6 Aug 2019 19:20:07 +0000 (21:20 +0200)]
shared/unit-file: fix systemctl cat user@.service

I assumed that unit_name_to_instnace() returns NULL if there is no instance.
In fact it returns "", so the check for instance was wrong.

Also avoid unnecessary call to unit_name_is_valid().

5 years agoMerge pull request #13302 from yuwata/network-set-put-fixes
Lennart Poettering [Wed, 14 Aug 2019 14:18:06 +0000 (16:18 +0200)]
Merge pull request #13302 from yuwata/network-set-put-fixes

network: fixes related to set_put()

5 years agoMerge pull request #13318 from ddstreet/s390_seccomp
Lennart Poettering [Wed, 14 Aug 2019 14:13:58 +0000 (16:13 +0200)]
Merge pull request #13318 from ddstreet/s390_seccomp

add s390 pkey_mprotect syscall number, and s390 mmap syscall definitions

5 years agocore: Avoid race when starting dbus services
Mattias Jernberg [Thu, 11 Jul 2019 16:13:46 +0000 (18:13 +0200)]
core: Avoid race when starting dbus services

In high load scenarios it is possible for services to be started
before the NameOwnerChanged signal is properly installed.

Emulate a callback by also queuing a GetNameOwner when the match is
installed.

Fixes: #12956

5 years agotest: ppc64* qemu is qemu-system-ppc64
Daniel Black [Wed, 14 Aug 2019 01:48:26 +0000 (11:48 +1000)]
test: ppc64* qemu is qemu-system-ppc64

5 years agohwdb: Fix airplane mode spam on HP Spectre x360 Convertible
Francesco Pennica [Tue, 13 Aug 2019 10:10:56 +0000 (12:10 +0200)]
hwdb: Fix airplane mode spam on HP Spectre x360 Convertible

Added additional rule matching my HP Spectre x360 Convertible from 2015 (skylake)

5 years agoshell-completion: complete --match argument for busctl
Ronan Pigott [Wed, 14 Aug 2019 03:27:56 +0000 (20:27 -0700)]
shell-completion: complete --match argument for busctl

5 years agosrc/shared/seccomp-util.c: Add mmap definitions for s390
Dan Streetman [Thu, 25 Jul 2019 00:48:49 +0000 (20:48 -0400)]
src/shared/seccomp-util.c: Add mmap definitions for s390

5 years agosrc/basic/missing_syscall: add s390 syscall number for __NR_pkey_mprotect
Dan Streetman [Wed, 24 Jul 2019 23:58:48 +0000 (19:58 -0400)]
src/basic/missing_syscall: add s390 syscall number for __NR_pkey_mprotect

The syscall number for s390 was added to the kernel at:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=b41c51c8e194c0bdfb4b1778a137aea8246c86cd

5 years agoTODO: add more, and drop implemented stuff
Lennart Poettering [Tue, 13 Aug 2019 12:40:30 +0000 (14:40 +0200)]
TODO: add more, and drop implemented stuff

5 years agotest: store coredumps in journal
Frantisek Sumsal [Mon, 12 Aug 2019 22:14:54 +0000 (00:14 +0200)]
test: store coredumps in journal

To make debugging much easier, especially for crashes in tests under
QEMU, let's store the entire coredump bundle in the systemd journal,
which is usually kept around by various CIs. Right now, we usually end
up with a journal, but without the coredump itself, which is pretty
useless.

5 years agohwdb: Add axis overrides for HP Envy x360 (#13304)
William Wold [Mon, 12 Aug 2019 22:03:55 +0000 (18:03 -0400)]
hwdb: Add axis overrides for HP Envy x360 (#13304)

5 years agoMerge pull request #13303 from yuwata/bootctl-segfaults
Frantisek Sumsal [Mon, 12 Aug 2019 19:16:51 +0000 (19:16 +0000)]
Merge pull request #13303 from yuwata/bootctl-segfaults

bootctl: fix segfaults

5 years agonetworkd: Keep track of static neighbors
William A. Kennington III [Fri, 19 Apr 2019 07:53:34 +0000 (00:53 -0700)]
networkd: Keep track of static neighbors

We need to keep track of the static neighbors that are configured on the
interface so that we can delete stale entries that were removed.

5 years agobootctl: arg_dolloar_boot_path() may return NULL
Yu Watanabe [Sun, 11 Aug 2019 15:05:28 +0000 (00:05 +0900)]
bootctl: arg_dolloar_boot_path() may return NULL

5 years agobootctl: clear arg_xbootldr_path when acquire_xbootldr() succeeds
Yu Watanabe [Sun, 11 Aug 2019 15:04:44 +0000 (00:04 +0900)]
bootctl: clear arg_xbootldr_path when acquire_xbootldr() succeeds

5 years agonetwork: add missing link_ref()
Yu Watanabe [Sun, 11 Aug 2019 00:11:20 +0000 (09:11 +0900)]
network: add missing link_ref()

5 years agonetwork: fix potential memleaks related to set_put()
Yu Watanabe [Sat, 10 Aug 2019 23:56:37 +0000 (08:56 +0900)]
network: fix potential memleaks related to set_put()

5 years agoman: add example for IPv6 Prefix Delegation
frederik [Fri, 9 Aug 2019 10:09:26 +0000 (12:09 +0200)]
man: add example for IPv6 Prefix Delegation

5 years agofirstboot: drop duplicate trailing whitespace from root pw question
Lennart Poettering [Fri, 9 Aug 2019 14:35:36 +0000 (16:35 +0200)]
firstboot: drop duplicate trailing whitespace from root pw question

Since ask_password() (and related calls) already append one char, we
ended up appending two. That's not pretty. Let's fix this, and do it
like in all other cases ask_password() (or an equivalent function) is
called.

5 years agodocs: fix env var name in random seed markdown documentation
Lennart Poettering [Thu, 8 Aug 2019 11:31:40 +0000 (13:31 +0200)]
docs: fix env var name in random seed markdown documentation

5 years agoFix typo in `analyze-security.c`.
Johannes Christ [Sat, 10 Aug 2019 12:41:52 +0000 (14:41 +0200)]
Fix typo in `analyze-security.c`.

5 years agozsh: add busctl completions
Ronan Pigott [Sat, 10 Aug 2019 08:21:35 +0000 (01:21 -0700)]
zsh: add busctl completions

5 years agofuzzit: get MSan to track origins
Evgeny Vereshchagin [Thu, 8 Aug 2019 18:30:44 +0000 (18:30 +0000)]
fuzzit: get MSan to track origins

It's just a follow-up to https://github.com/systemd/systemd/pull/13281
that should make it a little bit easier to make sense of
MSan reports.

https://clang.llvm.org/docs/MemorySanitizer.html#origin-tracking

5 years agoMerge pull request #13281 from evverx/unleash-msan
Evgeny Vereshchagin [Thu, 8 Aug 2019 15:38:21 +0000 (18:38 +0300)]
Merge pull request #13281 from evverx/unleash-msan

fuzzit: an attempt to unleash MSan on all the fuzzers

5 years agofuzzit: collapse a series of commands
Evgeny Vereshchagin [Thu, 8 Aug 2019 13:23:53 +0000 (13:23 +0000)]
fuzzit: collapse a series of commands

Now that the ids are gone, we can generate commands simply
using names as we go

5 years agofuzzit: switch to a new organization
Evgeny Vereshchagin [Thu, 8 Aug 2019 11:56:23 +0000 (11:56 +0000)]
fuzzit: switch to a new organization

5 years agoman: Fix grammar in systemd.kill.xml
Matthew Leeds [Wed, 7 Aug 2019 20:10:50 +0000 (13:10 -0700)]
man: Fix grammar in systemd.kill.xml

5 years agofuzzit: unleash MSan on all the fuzzers
Evgeny Vereshchagin [Wed, 7 Aug 2019 21:45:19 +0000 (21:45 +0000)]
fuzzit: unleash MSan on all the fuzzers

5 years agologind: set description for inhibitor event source
Lennart Poettering [Mon, 15 Jul 2019 18:29:21 +0000 (20:29 +0200)]
logind: set description for inhibitor event source

5 years agonetworkd: Routes should take the gateway into account
William A. Kennington III [Fri, 19 Apr 2019 00:52:28 +0000 (17:52 -0700)]
networkd: Routes should take the gateway into account

Otherwise, changing the default gateway doesn't purge old gateway routes
left on the system during daemon restart. This also fixes removing other
foreign gateway routes that don't match the expected configuration.

Tested:
    Changed gateway addresses prior to the patch and they lingered on
    the system during each reconfiguration. Applied this patch and
    reconfigured gateways and other routes multiple times and it removed
    the foreign routes that had gateways that didn't match.

Signed-off-by: William A. Kennington III <william@wkennington.com>
5 years agoshell-completion: add resolvectl commands to zsh completion
Ronan Pigott [Wed, 7 Aug 2019 05:50:06 +0000 (22:50 -0700)]
shell-completion: add resolvectl commands to zsh completion

5 years agoshell-completion: enable resolvectl zsh completion
Ronan Pigott [Wed, 7 Aug 2019 04:00:36 +0000 (21:00 -0700)]
shell-completion: enable resolvectl zsh completion

5 years agoMerge pull request #13265 from keszybz/timedated-ntp-logging
Yu Watanabe [Tue, 6 Aug 2019 23:07:57 +0000 (08:07 +0900)]
Merge pull request #13265 from keszybz/timedated-ntp-logging

Timedated logging and service priority

5 years agotimedated: always enable&start the service with highest priority
Zbigniew Jędrzejewski-Szmek [Mon, 5 Aug 2019 10:50:11 +0000 (12:50 +0200)]
timedated: always enable&start the service with highest priority

This removes a special case that was implemented before: if some service
was already enabled, we'd treat it as having higher priority.

From https://bugzilla.redhat.com/show_bug.cgi?id=1735584#c4:
> Setting ntp off and on should give the same result as just setting it
> on. There should be no stickiness (hidden state). It should behave like
> running an ansible role.
>
> The other service might have been enabled because no other was installed at
> the time. If I install a new NTP service with a higher priority, setting ntp
> on should enable and start the new service, and disable all other. Also, if
> for some reason multiple services are enabled, after setting ntp on there
> should be only one enabled to avoid systemd selecting between them randomly
> on the next boot.

5 years agosystemctl: Add unit file states to state help message
Ronan Pigott [Tue, 6 Aug 2019 02:07:31 +0000 (19:07 -0700)]
systemctl: Add unit file states to state help message

5 years agoman: fix typo in resolvectl(1) man page
Ronan Pigott [Mon, 5 Aug 2019 22:41:16 +0000 (15:41 -0700)]
man: fix typo in resolvectl(1) man page

5 years agotimedated: log about unit enablement actions and enhance logs in general
Zbigniew Jędrzejewski-Szmek [Sun, 4 Aug 2019 09:06:59 +0000 (11:06 +0200)]
timedated: log about unit enablement actions and enhance logs in general

The general idea is that for things which can occur repeatedly, like SetNTP
being called with the same argument, we only log at debug level. For things
which change state, like when we realize that a unit that wasn't enabled
before needs to be enabled, we log at info level.

Also, don't return success if there are no units loaded.

5 years agounit: drop Before=sysinit.target from systemd-random-seed.service
Yu Watanabe [Sun, 4 Aug 2019 23:05:54 +0000 (08:05 +0900)]
unit: drop Before=sysinit.target from systemd-random-seed.service

Follow-up for 26ded55709947d936634f1de0f43dcf88f594621.

The commit says,

> Note that with this change sysinit.target (and thus early boot) is NOT
systematically delayed until the entropy pool is initialized,

But the dependency was not dropped.

This was found by David Seifert (@SoapGentoo).

5 years agomeson: create (empty) /etc/systemd/system during installation
Zbigniew Jędrzejewski-Szmek [Mon, 5 Aug 2019 13:02:13 +0000 (15:02 +0200)]
meson: create (empty) /etc/systemd/system during installation

We explicitly create /etc/systemd/user and other parts of the basic directory
tree. I think we should create /etc/systemd/system too. (The alternative would
be to not create those other directories too, but I think it's nice to have
the basic directory structure in place after installation.)

https://bugzilla.redhat.com/show_bug.cgi?id=1737362

5 years agoMerge pull request #12884 from mrc0mmand/numapolicy-test
Frantisek Sumsal [Mon, 5 Aug 2019 18:02:40 +0000 (18:02 +0000)]
Merge pull request #12884 from mrc0mmand/numapolicy-test

test: introduce TEST-36-NUMAPOLICY

5 years agoshared/exit-status: fix off-by-one in comment
Zbigniew Jędrzejewski-Szmek [Mon, 5 Aug 2019 14:37:53 +0000 (16:37 +0200)]
shared/exit-status: fix off-by-one in comment

5 years agoshared/exit-status: fix lookup
Zbigniew Jędrzejewski-Szmek [Mon, 5 Aug 2019 14:36:45 +0000 (16:36 +0200)]
shared/exit-status: fix lookup

FLAGS_SET() is the wrong operator here, because we want to see if
*any* bits are set. Add test.

https://github.com/systemd/systemd/pull/12884#issuecomment-518238410

5 years agoMerge pull request #13258 from yuwata/login-revert-handle-ignore-assertion
Zbigniew Jędrzejewski-Szmek [Mon, 5 Aug 2019 13:43:23 +0000 (15:43 +0200)]
Merge pull request #13258 from yuwata/login-revert-handle-ignore-assertion

Revert "logind: remove unused check"

5 years agotest: add a simple sanity check for systems without NUMA support
Frantisek Sumsal [Mon, 5 Aug 2019 12:38:45 +0000 (14:38 +0200)]
test: add a simple sanity check for systems without NUMA support

5 years agotest: give strace some time to initialize
Frantisek Sumsal [Mon, 22 Jul 2019 22:56:04 +0000 (00:56 +0200)]
test: give strace some time to initialize

The `coproc` implementation seems to be a little bit different in older
bash versions, so the `strace` is sometimes started AFTER `systemctl
daemon-reload`, which causes unexpected fails. Let's help it a little by
sleeping for a bit.

5 years agotest: skip the test on systems without NUMA support
Frantisek Sumsal [Tue, 2 Jul 2019 07:52:45 +0000 (09:52 +0200)]
test: skip the test on systems without NUMA support

5 years agotest: make sure the strace process is indeed dead
Frantisek Sumsal [Mon, 1 Jul 2019 17:53:45 +0000 (19:53 +0200)]
test: make sure the strace process is indeed dead

It may take a few moments for the strace process to properly terminate
and write all logs to the backing storage

5 years agotest: support MPOL_LOCAL matching in unpatched strace versions
Frantisek Sumsal [Mon, 1 Jul 2019 11:08:26 +0000 (13:08 +0200)]
test: support MPOL_LOCAL matching in unpatched strace versions

The MPOL_LOCAL constant is not recognized in current strace versions.
Let's match at least the numerical value of this constant until the
strace patch is approved & merged.

5 years agotest: replace `tail -f` with journal cursor which should be...
Frantisek Sumsal [Mon, 1 Jul 2019 07:27:59 +0000 (09:27 +0200)]
test: replace `tail -f` with journal cursor which should be...

more reliable

5 years agotest: introduce TEST-36-NUMAPOLICY
Frantisek Sumsal [Tue, 25 Jun 2019 21:01:40 +0000 (23:01 +0200)]
test: introduce TEST-36-NUMAPOLICY

5 years agotest: allow overriding QEMU_OPTIONS from the outside
Frantisek Sumsal [Tue, 25 Jun 2019 20:45:15 +0000 (22:45 +0200)]
test: allow overriding QEMU_OPTIONS from the outside

5 years agoudev: do not try to import properties on commented out lines
Yu Watanabe [Sat, 3 Aug 2019 21:08:06 +0000 (06:08 +0900)]
udev: do not try to import properties on commented out lines

Fixes #13257.

5 years agotimedated: minor code simplifications
Zbigniew Jędrzejewski-Szmek [Sun, 4 Aug 2019 09:06:33 +0000 (11:06 +0200)]
timedated: minor code simplifications

5 years agoMerge pull request #13256 from keszybz/minor-sd-network-cleanup
Yu Watanabe [Sun, 4 Aug 2019 08:07:03 +0000 (17:07 +0900)]
Merge pull request #13256 from keszybz/minor-sd-network-cleanup

Minor sd-network cleanup and other coverity-inspired changes

5 years agoRevert "logind: remove unused check"
Yu Watanabe [Sat, 3 Aug 2019 20:43:34 +0000 (05:43 +0900)]
Revert "logind: remove unused check"

This reverts commit f2330acda408a34451d5e15380fcdd225a672473.

Fixes #13255.

5 years agologin: drop space in empty line
Yu Watanabe [Sat, 3 Aug 2019 20:42:53 +0000 (05:42 +0900)]
login: drop space in empty line

5 years agopid1: replace asprintf() with strjoin()
Zbigniew Jędrzejewski-Szmek [Sat, 3 Aug 2019 15:46:56 +0000 (17:46 +0200)]
pid1: replace asprintf() with strjoin()

It's nicer. And coverity doesn't need to complain about unchecked return
value (CID#1401780).

5 years agosd-hwdb: use return value from ordered_hashmap_iterate()
Zbigniew Jędrzejewski-Szmek [Sat, 3 Aug 2019 15:43:51 +0000 (17:43 +0200)]
sd-hwdb: use return value from ordered_hashmap_iterate()

Why not? Coverity CID#1402329.

5 years agolibsystemd-network: make option_append() atomic and make the code a bit clearer
Zbigniew Jędrzejewski-Szmek [Sat, 3 Aug 2019 14:49:39 +0000 (16:49 +0200)]
libsystemd-network: make option_append() atomic and make the code a bit clearer

Comparisons are done in the normal order (if (need > available), not if (available < need)),
variables have reduced scope and are renamed for clarity.

The only functional change is that if we return -ENAMETOOLONG, we do that
without modifying the options[] array.

I also added an explanatory comment. The use of one offset to point into three
buffers is not obvious.

Coverity (in CID#1402354) says that sname might be accessed at bad offset, but
I cannot see this happening. We check for available space before writing anything.

5 years agounit: make logind can access ESP
Yu Watanabe [Fri, 2 Aug 2019 20:05:43 +0000 (05:05 +0900)]
unit: make logind can access ESP

Fixes the following error:
```
logind[601]: Failed to open file system "/dev/block/259:1": Operation not permitted
```

5 years agoMerge pull request #13253 from yuwata/udev-update-logs
Zbigniew Jędrzejewski-Szmek [Sat, 3 Aug 2019 12:15:52 +0000 (14:15 +0200)]
Merge pull request #13253 from yuwata/udev-update-logs

udev: update log messages

5 years agoudev-node: fix misleading log messages
Yu Watanabe [Fri, 2 Aug 2019 21:04:34 +0000 (06:04 +0900)]
udev-node: fix misleading log messages

This fixes the following logs:
```
drm_dp_aux2: Handling device node '/dev/drm_dp_aux2', devnum=c238:2, mode=037777777777, uid=4294967295, gid=4294967295
drm_dp_aux2: Preserve permissions of /dev/drm_dp_aux2, 037777777777, uid=4294967295, gid=4294967295
```

5 years agoudev-event: log device name on spawning commands
Yu Watanabe [Fri, 2 Aug 2019 20:38:13 +0000 (05:38 +0900)]
udev-event: log device name on spawning commands

5 years agotime-util: improve detection of synchronized clock
Miroslav Lichvar [Thu, 1 Aug 2019 09:47:18 +0000 (11:47 +0200)]
time-util: improve detection of synchronized clock

Instead of checking for the STA_UNSYNC flag in the timex status, check
the maximum error. It is updated by the kernel, increasing at a rate of
500 ppm. The maximum value is 16 seconds, which triggers the STA_UNSYNC
flag.

This follows timedatex and allows timedated to correctly detect a clock
synchronized by chronyd when configured to not synchronize the RTC.

5 years agoMerge pull request #13243 from keszybz/two-cryptsetup-quickfixes
Zbigniew Jędrzejewski-Szmek [Thu, 1 Aug 2019 15:31:12 +0000 (17:31 +0200)]
Merge pull request #13243 from keszybz/two-cryptsetup-quickfixes

Two cryptsetup quickfixes

5 years agocryptsetup: don't assert on variable which is optional
Zbigniew Jędrzejewski-Szmek [Thu, 1 Aug 2019 06:15:43 +0000 (08:15 +0200)]
cryptsetup: don't assert on variable which is optional

https://github.com/systemd/systemd/commit/50d2eba27b9bfc77ef6b40e5721713846815418b#commitcomment-34519739

In add_crypttab_devices() split_keyspec is called on the keyfile argument,
which may be NULL.

5 years agocryptsetup: use unabbrieviated variable names
Zbigniew Jędrzejewski-Szmek [Thu, 1 Aug 2019 06:13:13 +0000 (08:13 +0200)]
cryptsetup: use unabbrieviated variable names

Now that "ret_" has been added to the output variables, we can name
the internal variables without artificial abbrevs.

5 years agoMerge pull request #13225 from mrc0mmand/networkd-test-replace-adduser-with-useradd
Frantisek Sumsal [Wed, 31 Jul 2019 18:59:03 +0000 (18:59 +0000)]
Merge pull request #13225 from mrc0mmand/networkd-test-replace-adduser-with-useradd

test: use `useradd` instead of `adduser`

5 years agotest: print an error when networkctl returns an unexpected EC
Frantisek Sumsal [Tue, 30 Jul 2019 18:27:34 +0000 (20:27 +0200)]
test: print an error when networkctl returns an unexpected EC

If networkctl crashes, like recently with SIGABRT, it returns absolutely
no output, which may be confusing during debugging. Help it a little
with a short informative message.

5 years agoanalyze-security: move assert above first use of the variable (#13238)
Jan Synacek [Wed, 31 Jul 2019 14:13:35 +0000 (16:13 +0200)]
analyze-security: move assert above first use of the variable (#13238)

5 years agoMerge pull request #13239 from poettering/coverity-fixes
Zbigniew Jędrzejewski-Szmek [Wed, 31 Jul 2019 14:08:52 +0000 (16:08 +0200)]
Merge pull request #13239 from poettering/coverity-fixes

four coverity fixes

5 years agoman: reword DeviceAllow= documentation
Lennart Poettering [Wed, 31 Jul 2019 08:20:16 +0000 (10:20 +0200)]
man: reword DeviceAllow= documentation

Don't claim we'd use cgroup.deny much. It's just a way to remove stuff
from device lists, which is nothing we allow users to explicitly
configure.

Also, extend documentation when wildcards may be used, and when not.

5 years agojob: fix coverity issue
Lennart Poettering [Wed, 31 Jul 2019 07:44:28 +0000 (09:44 +0200)]
job: fix coverity issue

Fixes coverity issue 1403550

5 years agotest-chase-symlinks: fix coverity issue
Lennart Poettering [Wed, 31 Jul 2019 07:42:24 +0000 (09:42 +0200)]
test-chase-symlinks: fix coverity issue

Fixes coverity issue 1403708

5 years agounit-file: fix coverity issue
Lennart Poettering [Wed, 31 Jul 2019 07:39:12 +0000 (09:39 +0200)]
unit-file: fix coverity issue

Fixes coverity issue 1403820

5 years agocryptsetup-generator: fix coverity issue
Lennart Poettering [Wed, 31 Jul 2019 07:38:15 +0000 (09:38 +0200)]
cryptsetup-generator: fix coverity issue

Fixes coverity issue 1403772