Daniel Machon [Mon, 5 Oct 2015 09:15:16 +0000 (11:15 +0200)]
Update NEWS
Fixed spelling error: seriban -> serbian
Daniel Mack [Mon, 5 Oct 2015 07:47:58 +0000 (09:47 +0200)]
NEWS update
Kay Sievers [Sun, 4 Oct 2015 18:25:44 +0000 (20:25 +0200)]
Merge pull request #1457 from martinpitt/master
keymap: Add Thinkpad Yoga 12 (2015)
Lennart Poettering [Sun, 4 Oct 2015 15:34:44 +0000 (18:34 +0300)]
Merge pull request #1454 from kostich/master
po, catalog: added Serbian translation.
Martin Pitt [Sun, 4 Oct 2015 12:31:00 +0000 (14:31 +0200)]
keymap: Add Thinkpad Yoga 12 (2015)
Fixes #1440
Марко М. Костић (Marko M. Kostić) [Sat, 3 Oct 2015 20:07:29 +0000 (22:07 +0200)]
po, catalog: added Serbian translation.
Lennart Poettering [Sat, 3 Oct 2015 10:04:47 +0000 (12:04 +0200)]
Merge pull request #1450 from evverx/man-journal-fields
man: systemd.journal-fields: add info about _TRANSPORT=audit
Evgeny Vereshchagin [Sat, 3 Oct 2015 02:38:06 +0000 (05:38 +0300)]
man: systemd.journal-fields: add info about _TRANSPORT=audit
Daniel Mack [Thu, 1 Oct 2015 20:58:02 +0000 (22:58 +0200)]
Merge pull request #1439 from poettering/CODING_STYLE
Coding style
Lennart Poettering [Thu, 1 Oct 2015 19:11:12 +0000 (21:11 +0200)]
logind: remove warn_melody() logic
The internal speaker is usually not available on modern latops that
support suspend, and even if it is available in the hardware, most
distributions turned support for it off in the kernel. And even if it is
enabled, it's probably still a bad idea to make use of it for the
suspend-failures. If anything a proper sound should be played.
Long story short, let's remove support of this anachronism.
Lennart Poettering [Thu, 1 Oct 2015 19:10:55 +0000 (21:10 +0200)]
update CODING_STYLE with various additions
Kay Sievers [Thu, 1 Oct 2015 17:10:06 +0000 (19:10 +0200)]
Merge pull request #1428 from franciozzy/tagenhance
Enhance matching of TAG keys
Daniel Mack [Thu, 1 Oct 2015 14:35:43 +0000 (16:35 +0200)]
Merge pull request #1438 from poettering/rfkill-rework
An rfkill rework among other things
Lennart Poettering [Thu, 1 Oct 2015 14:29:13 +0000 (16:29 +0200)]
Merge pull request #1430 from evverx/driver-syslog
journald: add syslog fields for driver messages
Lennart Poettering [Thu, 1 Oct 2015 14:24:04 +0000 (16:24 +0200)]
update NEWS
Lennart Poettering [Thu, 1 Oct 2015 12:32:48 +0000 (14:32 +0200)]
rfkill: rework and make it listen on /dev/rfkill
With this rework we introduce systemd-rfkill.service as singleton that
is activated via systemd-rfkill.socket that listens on /dev/rfkill. That
way, we get notified each time a new rfkill device shows up or changes
state, in which case we restore and save its current setting to disk.
This is nicer than the previous logic, as this means we save/restore
state even of rfkill devices that are around only intermittently, and
save/restore the state even if the system is shutdown abruptly instead
of cleanly.
This implements what I suggested in #1019 and obsoletes it.
Lennart Poettering [Thu, 1 Oct 2015 12:28:13 +0000 (14:28 +0200)]
core: add new setting Writable= to ListenSpecial= socket units
Writable= is a new boolean setting. If ture, then ListenSpecial= will
open the specified path in O_RDWR mode, rather than just O_RDONLY.
This is useful for implementing services like rfkill, where /dev/rfkill
is more useful when opened in write mode, if we want to not only save
but also restore its state.
Lennart Poettering [Thu, 1 Oct 2015 12:27:20 +0000 (14:27 +0200)]
systemctl: fix how we decode types and states
There was a bad memory access among other smaller issues.
Lennart Poettering [Thu, 1 Oct 2015 12:26:42 +0000 (14:26 +0200)]
man: minor wording improvements to USB FunctionFS documentation
Daniel Mack [Thu, 1 Oct 2015 11:28:09 +0000 (13:28 +0200)]
fix typo in NEWS
Daniel Mack [Thu, 1 Oct 2015 10:16:55 +0000 (12:16 +0200)]
Merge pull request #1434 from reverendhomer/patch-1
systemctl: fix memory leak in systemctl_parse_argv
Daniel Mack [Thu, 1 Oct 2015 10:16:16 +0000 (12:16 +0200)]
Merge pull request #1433 from michich/log-errno
tree-wide: add missing errno arguments to log_*_errno()
Michal Schmidt [Thu, 1 Oct 2015 09:28:40 +0000 (11:28 +0200)]
udev: fix format string after conversion to log_*_errno()
Use %m where previously %s was used together with strerrno().
Fixes:
e53fc357a9b "tree-wide: remove a number of invocations of
strerror() and replace by %m"
Michal Schmidt [Thu, 1 Oct 2015 09:18:05 +0000 (11:18 +0200)]
sd-device: fix format strings after conversion to log_*_errno()
Use %m where previously %s was used together with strerrno().
Fixes:
e53fc357a9b "tree-wide: remove a number of invocations of
strerror() and replace by %m"
reverendhomer [Thu, 1 Oct 2015 09:09:00 +0000 (12:09 +0300)]
systemctl: fix memory leak in systemctl_parse_argv
This commit fixes Coverity #1325228
Michal Schmidt [Thu, 1 Oct 2015 09:04:08 +0000 (11:04 +0200)]
tree-wide: add missing errno arguments to log_*_errno()
A few of the recent conversions to log_*_errno() were missing the errno
value arguments.
Fixes:
e53fc357a9b "tree-wide: remove a number of invocations of
strerror() and replace by %m"
Daniel Mack [Thu, 1 Oct 2015 07:53:15 +0000 (09:53 +0200)]
Merge pull request #1429 from poettering/dhcp-hostname
DHCP and journal remoting fix, as well as NEWS update
Daniel Mack [Thu, 1 Oct 2015 07:46:40 +0000 (09:46 +0200)]
Merge pull request #1427 from dbuch/master
sd-pppoe: use strna() instead of strempty()
Daniel Mack [Thu, 1 Oct 2015 07:46:09 +0000 (09:46 +0200)]
Merge pull request #1426 from poettering/log-syntax
logging fixes and more
Evgeny Vereshchagin [Thu, 1 Oct 2015 00:36:11 +0000 (03:36 +0300)]
journald: add syslog fields for driver messages
Lennart Poettering [Wed, 30 Sep 2015 21:49:26 +0000 (23:49 +0200)]
journal: make sure to set MHD_USE_PIPE_FOR_SHUTDOWN for libmicrohttpd servers
This makes sure libmicrohttpd won't call shutdown() on our listening
sockets, which make sure socket activation and re-activation will work
cleanly.
See:
https://github.com/systemd/systemd/pull/1286
https://lists.gnu.org/archive/html/libmicrohttpd/2015-09/msg00014.html
Fixes #1286
Lennart Poettering [Wed, 30 Sep 2015 21:35:18 +0000 (23:35 +0200)]
dhcp: make sure we can deal with a single trailing NUL byte in the hostname
Fixes #1337
Lennart Poettering [Wed, 30 Sep 2015 21:35:10 +0000 (23:35 +0200)]
update NEWS
Daniel Buch [Wed, 30 Sep 2015 21:16:13 +0000 (23:16 +0200)]
sd-pppoe: use strna() instead of strempty()
In cases where we use for instance. log_debug() with a statement + string
it is better to output (Service-name: N/A) instead of (Service-name: )
Felipe Franciosi [Wed, 30 Sep 2015 21:07:16 +0000 (22:07 +0100)]
Support OP_NOMATCH for TAG key
The TAG key can be used in rules for event matching. At the moment, it
does not support inequality tests. This patch enhances the key test to
validate the rule if it does not contain a given TAG (by TAG!="value").
Signed-off-by: Felipe Franciosi <felipe@paradoxo.org>
Lennart Poettering [Wed, 30 Sep 2015 20:35:58 +0000 (22:35 +0200)]
Merge pull request #1425 from kaysievers/wip
gpt-auto-generator: check fstab for /boot entries
Lennart Poettering [Wed, 30 Sep 2015 20:33:50 +0000 (22:33 +0200)]
Merge pull request #1422 from keszybz/sd-daemon-mq-badf
sd-daemon: fix return value for sd_is_mq
Lennart Poettering [Wed, 30 Sep 2015 20:32:09 +0000 (22:32 +0200)]
Merge pull request #1419 from keszybz/shell-completion
Shell completion tweaks
Lennart Poettering [Wed, 30 Sep 2015 20:22:03 +0000 (22:22 +0200)]
core: make sure we expose the empty string when we detect no virtualization
On the bus, stick to exposing the empty string as "no virtualization",
instead of none.
Fixes #1423
Lennart Poettering [Wed, 30 Sep 2015 20:16:17 +0000 (22:16 +0200)]
tree-wide: remove a number of invocations of strerror() and replace by %m
Let's clean up our tree a bit, and reduce invocations of the
thread-unsafe strerror() by replacing it with printf()'s %m specifier.
Lennart Poettering [Wed, 30 Sep 2015 19:50:22 +0000 (21:50 +0200)]
basic: split out cpu set specific APIs into cpu-set-util.[ch]
Lennart Poettering [Wed, 30 Sep 2015 18:28:51 +0000 (20:28 +0200)]
conf-parser: make use of DEFINE_PARSER macro to parse mode_t
Let's unify more code.
Also, rework the macro to accept a trailing semicolon, to make the code
prettier.
Lennart Poettering [Wed, 30 Sep 2015 18:19:57 +0000 (20:19 +0200)]
util: there cannot be trailing garbage when parsing cpu sets
extract_first() already skips trailing whitespace, hence no reason to
explicitly check for it.
Lennart Poettering [Wed, 30 Sep 2015 18:16:51 +0000 (20:16 +0200)]
util: rename parse_cpu_set() to parse_cpu_set_and_warn()
It's pretty untypical for our parsing functions to log on their own.
Clarify in the name that this one does.
Lennart Poettering [Wed, 30 Sep 2015 18:07:24 +0000 (20:07 +0200)]
log: move log_invalid_utf8() to log.h
Also, make sure it follows the same scheme as log_syntax() does in its
behaviour.
Lennart Poettering [Wed, 30 Sep 2015 16:22:42 +0000 (18:22 +0200)]
tree-wide: clean up log_syntax() usage
- Rely everywhere that we use abs() on the error code passed in anyway,
thus don't need to explicitly negate what we pass in
- Never attach synthetic error number information to log messages. Only
log about errors we *receive* with the error number we got there,
don't log any synthetic error, that don#t even propagate, but just eat
up.
- Be more careful with attaching exactly the error we get, instead of
errno or unrelated errors randomly.
- Fix one occasion where the error number and line number got swapped.
- Make sure we never tape over OOM issues, or inability to resolve
specifiers
Lennart Poettering [Wed, 30 Sep 2015 20:25:36 +0000 (22:25 +0200)]
Merge pull request #1424 from evverx/journalctl-rotate
journalctl: add --rotate option
Kay Sievers [Wed, 30 Sep 2015 20:24:52 +0000 (22:24 +0200)]
gpt-auto-generator: check fstab for /boot entries
We need to prevent the creation of the gpt automount unit, which will not
get overridden by the fstab mount unit.
https://github.com/systemd/systemd/issues/1378
Evgeny Vereshchagin [Wed, 30 Sep 2015 20:02:58 +0000 (23:02 +0300)]
man: journalctl: add info about --rotate
Evgeny Vereshchagin [Wed, 30 Sep 2015 19:58:05 +0000 (22:58 +0300)]
shell-completion: journalctl: add --rotate
Evgeny Vereshchagin [Wed, 30 Sep 2015 19:54:58 +0000 (22:54 +0300)]
journalctl: add --rotate option
shortcut for `systemctl kill --kill-who main --signal SIGUSR2 systemd-journald`
Zbigniew Jędrzejewski-Szmek [Sat, 5 Sep 2015 13:20:15 +0000 (15:20 +0200)]
sd-daemon: fix sd_is_mq for non-mq fds
mq_getattr returns -1/EBADF for file descriptors which are not mq.
But we should return 0 in this case.
We first check that fd is a valid fd, so we can assume that if
mq_getattr returns EBADF, it is simply a non-mq fd. There is a slight
race, but there doesn't seem to be a nice way to fix it.
Daniel Mack [Wed, 30 Sep 2015 16:40:57 +0000 (18:40 +0200)]
Merge pull request #1421 from keszybz/formatting-fixes
logind: avoid a dot before parenthesis
Daniel Mack [Wed, 30 Sep 2015 16:37:18 +0000 (18:37 +0200)]
Merge pull request #1420 from keszybz/doc-updates
Doc updates
Zbigniew Jędrzejewski-Szmek [Wed, 30 Sep 2015 16:33:31 +0000 (12:33 -0400)]
systemctl: do not set variable to be immediately overwritten
Zbigniew Jędrzejewski-Szmek [Wed, 30 Sep 2015 15:48:33 +0000 (11:48 -0400)]
NEWS: typos
Zbigniew Jędrzejewski-Szmek [Wed, 30 Sep 2015 15:14:32 +0000 (11:14 -0400)]
man: update description of exit.target
exit.target is now used for both system and user sessions,
so remove "on user service manager exit". Also reword that
paragraph: services will be killed before the manager exits,
even if they do not conflict with shutdown target, but we
recommend that they conflict with shutdown target so that
systemd schedules them to be stopped immediately when starting
to exit.
In the first paragraph, containers should be mentioned last,
and the more general systems first.
Zbigniew Jędrzejewski-Szmek [Mon, 28 Sep 2015 15:44:28 +0000 (11:44 -0400)]
logind: avoid a dot before parenthesis
systemd-logind[27]: System is rebooting. (Applied kernel updates.)
is changed to
systemd-logind[27]: System is rebooting (Applied kernel updates).
Users should not add a dot in the sentence in --message, i.e. the correct usage is now:
$ systemctl reboot --message "Applied kernel updates"
Zbigniew Jędrzejewski-Szmek [Wed, 30 Sep 2015 14:41:41 +0000 (10:41 -0400)]
man: add link to kernel docs for net_cls
Zbigniew Jędrzejewski-Szmek [Mon, 28 Sep 2015 15:42:30 +0000 (11:42 -0400)]
man: use "=" when referring to configuration file settings
This convention is almost universal in systemd man pages, and makes
it easier to visually parse the docs.
Also fix some markup along the way.
Zbigniew Jędrzejewski-Szmek [Mon, 28 Sep 2015 15:40:25 +0000 (11:40 -0400)]
man: add link to functionfs docs
Also reword some ungainly grammar constructs.
Daniel Mack [Wed, 30 Sep 2015 14:35:49 +0000 (16:35 +0200)]
Merge pull request #1415 from aakoskin/bootchart-v2
[v2] bootchart: don't fail if "model name" is missing from cpuinfo
Daniel Mack [Wed, 30 Sep 2015 14:01:42 +0000 (16:01 +0200)]
Merge pull request #1418 from poettering/base-fs
base-fs: consider inability to create /root non-fatal
Daniel Mack [Wed, 30 Sep 2015 13:55:26 +0000 (15:55 +0200)]
Merge pull request #1417 from poettering/nspawn-and-more
Various fixes, primarily for nspawn, but other stuff too
Lennart Poettering [Wed, 30 Sep 2015 13:50:42 +0000 (15:50 +0200)]
base-fs: consider inability to create /root non-fatal
Fixes #1414
Lennart Poettering [Wed, 30 Sep 2015 13:39:29 +0000 (15:39 +0200)]
Merge pull request #1394 from steelman/fsck-force-reboot
fsck: start reboot.target irreversibly
Lennart Poettering [Wed, 30 Sep 2015 13:01:01 +0000 (15:01 +0200)]
analyze: add new set-log-target subcommand
We already have the property writable, hence let's add a command to set
it.
Lennart Poettering [Wed, 30 Sep 2015 12:16:40 +0000 (14:16 +0200)]
log: properly return -EINVAL from log_set_max_level_from_string()
If we just return the value we got from log_level_from_string() on
failure we'll return -1, which is not a proper error code.
log_set_target_from_string() did get this right already, hence let's fix
this here too.
Lennart Poettering [Wed, 30 Sep 2015 11:47:28 +0000 (13:47 +0200)]
nspawn: mount /sys as tmpfs, and then mount only select subdirs of the real sysfs below it
This way we can hide things like /sys/firmware or /sys/hypervisor from
the container, while keeping the device tree around.
While this is a security benefit in itself it also allows us to fix
issue #1277.
Previously we'd mount /sys before creating the user namespace, in order
to be able to mount /sys/fs/cgroup/* beneath it (which resides in it),
which we can only mount outside of the user namespace. To ensure that
the user namespace owns the network namespace we'd set up the network
namespace at the same time as the user namespace. Thus, we'd still see
the /sys/class/net/ from the originating network namespace, even though
we are in our own network namespace now. With this patch, /sys is
mounted before transitioning into the user namespace as tmpfs, so that
we can also mount /sys/fs/cgroup/* into it this early. The directories
such as /sys/class/ are then later added in from the real sysfs from
inside the network and user namespace so that they actually show whatis
available in it.
Fixes #1277
Aaro Koskinen [Wed, 30 Sep 2015 12:57:55 +0000 (15:57 +0300)]
bootchart: use get_proc_field and PROC_CPUINFO_MODEL
Use get_proc_field and PROC_CPUINFO_MODEL to get the CPU type. Also don't
fail if it missing, instead restore the old behaviour and print "Unknown".
Fixes the following error seen with v225 on powerpc:
Unable to read module name from cpuinfo.
Error generating svg file: No such file or directory
Aaro Koskinen [Wed, 30 Sep 2015 12:57:55 +0000 (15:57 +0300)]
architecture.h: define PROC_CPUINFO_MODEL
Define PROC_CPUINFO_MODEL to indicate the CPU model field in /proc/cpuinfo.
Aaro Koskinen [Wed, 30 Sep 2015 12:57:55 +0000 (15:57 +0300)]
fileio: make get_status_field() more generic
All users of get_status_field() expect the field pattern to occur in
the beginning of a line, and the delimiter is ':'.
Hardcode this into the function, and also skip any whitespace before ':'
to support fields in files like /proc/cpuinfo. Add support for returning
the full field value (currently stops on first whitespace).
Rename the function so it's easier to ensure all callers switch to new
semantics.
Daniel Mack [Wed, 30 Sep 2015 12:03:14 +0000 (14:03 +0200)]
fix typos in NEWS
Lennart Poettering [Wed, 30 Sep 2015 10:48:17 +0000 (12:48 +0200)]
nspawn: fix user namespace support
We didn#t actually pass ownership of /run to the UID in the container
since some releases, let's fix that.
Daniel Mack [Wed, 30 Sep 2015 10:46:00 +0000 (12:46 +0200)]
update NEWS
Daniel Mack [Wed, 30 Sep 2015 10:40:44 +0000 (12:40 +0200)]
update NEWS
Daniel Mack [Wed, 30 Sep 2015 10:23:57 +0000 (12:23 +0200)]
Add Coccinelle patch for strempty() usage
Lennart Poettering [Wed, 30 Sep 2015 10:24:06 +0000 (12:24 +0200)]
nspawn: order includes
Lennart Poettering [Wed, 30 Sep 2015 10:20:26 +0000 (12:20 +0200)]
nspawn: make sure mount_legacy_cgroup_hierarchy() can deal with NULL root directories
Lennart Poettering [Wed, 30 Sep 2015 08:56:07 +0000 (10:56 +0200)]
update TODO
Lennart Poettering [Tue, 29 Sep 2015 20:40:05 +0000 (22:40 +0200)]
systemctl: when reading legacy -t argument for shutdown, don't drop following parameter
We currently completely ignore the following parameter, but we really
should not, as that is actually the time to shut down on.
Lennart Poettering [Tue, 29 Sep 2015 20:39:49 +0000 (22:39 +0200)]
systemctl: various modernizations
Lennart Poettering [Wed, 30 Sep 2015 10:00:46 +0000 (12:00 +0200)]
Merge pull request #1412 from zonque/strempty
tree-wide: use strempty() where possible
Daniel Mack [Fri, 25 Sep 2015 11:36:54 +0000 (13:36 +0200)]
tree-wide: use strempty() where possible
Also add a Coccinell patch to detect such locations in the future.
Daniel Mack [Tue, 29 Sep 2015 20:10:40 +0000 (22:10 +0200)]
Merge pull request #1408 from poettering/systemctl-and-more
Systemctl and more
Lennart Poettering [Tue, 29 Sep 2015 13:09:10 +0000 (15:09 +0200)]
systemctl: when a shutdown is scheduled, always go via logind
Lennart Poettering [Tue, 29 Sep 2015 13:08:33 +0000 (15:08 +0200)]
systemctl: minor modernizations
Lennart Poettering [Tue, 29 Sep 2015 13:00:07 +0000 (15:00 +0200)]
systemctl: port verb dispatching to generic verbs.[ch] code
Let's parse the command line the same way in our various tools.
Lennart Poettering [Tue, 29 Sep 2015 12:00:21 +0000 (14:00 +0200)]
systemctl: prefer shutting down through logind even as root
Otherwise, wall messages will not be generated for root.
Lennart Poettering [Tue, 29 Sep 2015 12:00:12 +0000 (14:00 +0200)]
systemctl: split out code invoking SetExitCode() into function of its own
Lennart Poettering [Tue, 29 Sep 2015 11:59:32 +0000 (13:59 +0200)]
pager: minor modernizations
Lennart Poettering [Tue, 29 Sep 2015 11:38:36 +0000 (13:38 +0200)]
systemctl: don't assert on the arguments array unnecessarily
In most verbs it's fine to treat the arguments array being NULL is empty
array, hence do so.
Lennart Poettering [Tue, 29 Sep 2015 11:36:07 +0000 (13:36 +0200)]
systemctl: make use of strv_skip() where appropriate
Lennart Poettering [Tue, 29 Sep 2015 11:35:27 +0000 (13:35 +0200)]
systemctl: make use of log_error_errno() where we can
Also adds a couple of "return" statements, where they are missing.
Lennart Poettering [Tue, 29 Sep 2015 11:13:50 +0000 (13:13 +0200)]
systemctl: rework how we connect to busses
Sometimes we have to connect to the system manager directly (early boot,
initrd, late boot, ...), sometimes through the system bus (unprivileged,
remote, logind, ...). Instead of guessing in advance, which kind of
connection we require (and sometimes guessing incorrectly), let's make
sure each time we need bus connection we request the right bus
explicitly.
This way, we set up exactly the bus connections require, never guess
incorrectly, and do so only immediately when necessary.
As effect this reworks avoid_bus() into install_client_side(), since
that's all it determines now: whether to install unit files client-side
or server-side (i.e. in PID 1).
Lennart Poettering [Tue, 29 Sep 2015 11:06:28 +0000 (13:06 +0200)]
core: add a "Requires=" dependency between units and the slices they are located in
We place the processes we fork off in the cgroup anyway, and we probably
shouldn't be able to get that far if we couldn't set up the slice due to
resource problems or unmet conditions. Hence upgrade the dependency
between units and the slices they are located in from Wants= to
Requires=.
Lennart Poettering [Thu, 24 Sep 2015 11:34:08 +0000 (13:34 +0200)]
update-utmp: flush and close the bus used for connections after use
Lennart Poettering [Thu, 24 Sep 2015 11:33:55 +0000 (13:33 +0200)]
update-utmp: minor modernizations
Lennart Poettering [Thu, 24 Sep 2015 11:30:10 +0000 (13:30 +0200)]
bus-util: rename bus_open_transport() to bus_connect_transport()
In sd-bus, the sd_bus_open_xyz() family of calls allocates a new bus,
while sd_bus_default_xyz() family tries to reuse the thread's default
bus. bus_open_transport() sometimes internally uses the former,
sometimes the latter family, but suggests it only calls the former via
its name. Hence, let's avoid this confusion, and generically rename the
call to bus_connect_transport().
Similar for all related calls.
And while we are at it, also change cgls + cgtop to do direct systemd
connections where possible, since all they do is talk to systemd itself.
Lennart Poettering [Thu, 24 Sep 2015 10:42:30 +0000 (12:42 +0200)]
pager: minor simplification