If for some reason we can't query the firmware state, don't propagate
that to clients, but instead log about it, and claim that
reboot-to-firmware is not available (which is the right answer, since it
is not working).
Let's log about this though, as this is certainly relevant to know, even
though not for the client.
So far I avoided adding license headers to meson files, but they are pretty
big and important and should carry license headers like everything else.
I added my own copyright, even though other people modified those files too.
But this is mostly symbolic, so I hope that's OK.
SetLinger is authorized by the PolicyKit action "set-self-linger", if it is
not passed an explicit UID.
According to comments we were determining the default UID from the client's
session. However, user processes e.g. which are run from a terminal
emulator do not necessarily belong to a session scope unit. They may
equally be started from the systemd user manager [1][2]. Actually the
comment was wrong, and it would also have worked for processes
started from the systemd user manager.
Nevertheless it seems to involve fetching "augmented credentials" i.e.
it's using a racy method, so we shouldn't have been authenticating based
on it.
We could change the default UID, but that raises issues especially for
consistency between the methods. Instead we can just use the clients
effective UID for authorization.
This commit also fixes `loginctl enable-linger $USER` to match the docs
that say it was equivalent to `loginctl enable-linger` (given that $USER
matches the callers user and owner_uid). Previously, the former would not
have suceeded for unpriviliged users in the default configuration.
[1] It seems the main meaning of per-session scopes is tracking the PAM
login process. Killing that provokes logind to revoke device access. Less
circularly, killing it provokes getty to hangup the TTY.
[2] User units may be started with an environment which includes
XDG_SESSION_ID (presuambly GNOME does this?). Or not.
To maintain consistency with `loginctl user-status`, drop the fallback to
XDG_SESSION_ID for `loginctl enable-linger`. The fallback was unnecessary
and also incorrect: it passed the numeric value of the session identifier
as a UID value.
The manpages tell that such calls have quite limited meaning. logind has
a few in the implementation of what remains of the session concept.
At the same time, logind basically exposes sd_pid_get_session() as public
API. This is absolutely required, to retain compatability e.g. with Xorg.
But client code will work in more situations if it avoids assuming that it
runs in a session itself.
Its use inside the login session could be replaced with $XDG_SESSION_ID
(which pam_systemd sets). I don't know whether it would be useful to
change Xorg at this point or not. But if you were building something new,
you would think about whether you want to support running it in a systemd
service.
Comment these logind API features, acknowledging the reason they exist is
based in history. I.e. help readers avoid drawing implications from their
existence which apply to history, but not the current general case.
Finally, searching these revealed a call to sd_pid_get_session() in
implementing some types of logind inhibitors. So these inhibitors don't
work as intended when taken from inside a systemd user service :(. Comment
this as well, deferring it as ticket #6852.
If you try to run `loginctl user-status` on a non-logged in user to see
whether "Linger" is enabled, it doesn't work.
If you're already an expert in logind, the fact that the user is considered
unknown actually tells you the user is not lingering. So, probably they
they do not have lingering enabled. I think we can point towards this
without being misleading.
I also reword it because I thought it was slightly confusing to run
`loginctl user-status root` and get an error back about "User 0". Try to
be more specific, that it is "User ID 0".
It's confusing that the bus API has aliases like "session/self" that return
an error based on ENXIO, when it also has methods that return e.g.
NO_SESSION_FOR_PID for the same problem. The latter kind of error includes
more specifically helpful messages.
"user/self" is the odd one out; it returns a generic UnknownObject error
when it is not applicable to the caller. It's not clear whether this was
intentional, but at first I thought it was more correct. More
specifically, user_object_find() was returning 0 for "user/self", in the
same situations (more or less) where user_node_enumerator() was omitting
"user/self". I thought that was a good idea, because returning e.g. -ENXIO instead
suggested that there _is_ something specific on that path. And it could be
confused with errors of the method being called.
Therefore I suggested changing the enumerator, always admitting that there
is a handler for the path "foo/self", but returning a specific error when
queried. However this interacts poorly with tools like D-Feet or `busctl`.
In either tool, looking at logind would show an error message, and then go
on to omit "user/self" in the normal listing. These tools are very useful,
so we don't want to interfere with them.
I think we can change the error codes without causing problems. The self
objects were not listed in the documentation. They have been suggested to
other projects - but without reference to error reporting. "seat/self" is
used by various Wayland compositors for VT switching, but they don't appear
to reference specific errors.
We _could_ insist on the link between enumeration and UnknownObject, and
standardize on that as the error for the aliases. But I'm not aware of any
practical complaints, that we returned an error from an object that didn't
exist.
Instead, let's unify the codepaths for "user/self" vs GetUserByPid(0) etc.
We will return the most helpful error message we can think of, if the
object does not exist. E.g. for "session/self", we might return an error
that the caller does not belong to a session. If one of the compositors is
ever simplified to use "session/self" in initialization, users would be
able to trigger such errors (e.g. run `gnome-shell` inside gnome-terminal).
The message text will most likely be logged. The user might not know what
the "session" is, but at least we'll be pointing towards the right
questions. I think it should also be clearer for development / debugging.
Unifying the code paths is also slightly helpful for auditing / marking
calls to sd_bus_creds_get_session() in subsequent commits.
GetSessionByPID(0) can fail with NO_SESSION_FOR_PID. More obscurely, if
the session is abandoned, it can return NO_SUCH_SESSION. It is not clear
that the latter was intended. The message associated with the former,
hints that this was overlooked.
We don't have a document enumerating the errors. Any specific
error-handling in client code, e.g. translated messages, would also be
liable to overlook the more obscure error code.
I can't see any equivalent condition for GetUserByPID(0). On the other
hand, the code did not return NO_USER_FOR_PID where it probably should.
The relevant code is right next to that for GetSessionByPID(0), so it will
be simpler to understand if both follow the same pattern.
This didn't work during the initial conversion to meson, but should now.
A sufficiently new polkit is also required, for the .its rules files.
Note that https://github.com/mesonbuild/meson/blob/master/docs/markdown/i18n-module.md
says that 'install' argument was added in meson 0.43.0. If this is accurate,
warnigs might be generated with older mesons. Fedora has 0.43.0 across the
board, but other distros probably don't, but I guess that a warning is
prefereable to having to update do latest meson.
The advantages are:
- one less dependency (intltool)
- using the generic implementation instead of our open-coded calls
- we don't need to use the fake "_" prefixes in XML
Replaces #1609, fixes#7300.
Let's hook up the sysfs tree output with the output flags logic,
already used when dumping log lines or process trees. This way we get
very similar output handling for line breaking/ellipsation in all three
outputs of structured data.
Fixes: #7095
loginctl, machinectl, systemctl all have very similar implementations of
a get_output_flags() functions. Simplify it by merging two lines that
set the same flag.
- Remove the uaccess tag from /dev/dri/renderD*.
- Change the owning group from video to render.
- Change default mode to 0666.
- Add an option to allow users to set the access mode for these devices at
compile time.
In https://bugzilla.redhat.com/show_bug.cgi?id=1486859 error messages appera:
Sep 06 19:09:07 ld92.e.math.uh.edu audit[21482]: AVC avc: denied { read } for pid=21482 comm="systemd-logind" name="dbus-1" dev="tmpfs" ino=5548194 scontext=system_u:system_r:systemd_logind_t:s0 tcontext=unconfined_u:object_r:session_dbusd_tmp_t:s0 tclass=dir permissive=0
Sep 06 19:09:07 ld92.e.math.uh.edu systemd-logind[21482]: Failed to remove runtime directory /run/user/8664: Permission denied
But it's not clear which of the two rm_rf's is the source. Let's make
them different.
This adds new method calls Halt() and CanHalt() to the logind bus APIs.
They aren't overly useful (as the whole concept of halting isn't really
too useful), however they clean up one major asymmetry: currently, using
the "shutdown" legacy commands it is possibly to enqueue a "halt"
operation through logind, while logind officially doesn't actually
support this. Moreover, the path through "shutdown" currently ultimately
fails, since the referenced "halt" action isn't actually defined in
PolicyKit.
Finally, the current logic results in an unexpected asymmetry in
systemctl: "systemctl poweroff", "systemctl reboot" are currently
asynchronous (due to the logind involvement) while "systemctl halt"
isnt. Let's clean this up, and make all three APIs implemented by
logind natively, and all three hence asynchronous in "systemctl".
Moreover, let's add the missing PK action.
Fixes: #6957
These two sites _do_ match the definition of pid_is_valid(); they don't
provide any special handling for the invalid PID value 0. (They're used
by dbus methods, so the PID value 0 is handled with reference to the dbus
client creds, outside of these functions).
The advantage is that is the name is mispellt, cpp will warn us.
$ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/"
$ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;'
$ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g'
$ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g'
+ manual changes to meson.build
squash! build-sys: use #if Y instead of #ifdef Y everywhere
v2:
- fix incorrect setting of HAVE_LIBIDN2
This reverts commit ee043777be.
It broke almost everywhere it touched. The places that
handn't been converted, were mostly followed by special
handling for the invalid PID `0`. That explains why they
tested for `pid < 0` instead of `pid <= 0`.
I think that one was the first commit I reviewed, heh.
Following commit b498d6ea, I belated realized we should tighten the
assertions as well, to make sure that we're setting `m->action_what` to
represent an action in progress. (The check for an action in progress is
to compare `m->action_what` to zero)
This fixed https://bugzilla.redhat.com/show_bug.cgi?id=1476313
as much as I was able to reproduce it in a VM, at least.
E.g. this signal might wake the screen back up, providing a more visible
indicator of suspend failure. In my VM testing, it was also required in
order to unblock keyboard input in gnome-shell after the failed suspend.
At the same time, fix the error handling for scheduled shutdowns. This now
mirrors the behaviour of when you use `shutdown -k` - it sends all the
scary messages about shutting down, "but you'll have to do it [shut down
the system] yourself". It also avoids the risk of locking out the admin
(nologin file), in case they logged out for some reason (and they use
`sudo` instead of root).
Not that I have any idea why you'd want to use `shutdown -k`, but the code
is easier to analyze if it rolls back on error (in the absence of any code
comment as to why that's not wanted).
There is no justification not to wait an extra (default) 5 seconds, for
a more graceful shutdown of user programs. Again, you don't get to ignore
delay inhibitors for unscheduled shutdowns, short of
`systemctl poweroff -f`.
It is simplest if we move the test for `m->shutdown_dry_run` into
manager_scheduled_shutdown_handler().
However we need to not add such delays during a "dry run". Otherwise, we
would still have to be considered "in progress" for some seconds after our
admin has seen the final wall message. If they go to `poweroff`, we would
have blocked them with a misleading error message. Note this `poweroff`
will still process delay inhibitors as needed. If the admin planned to
use a more forceful method... eh. It's their responsibility to assess
whether that's safe.
There is an argument that the alternative behaviour could be used (racily!)
to kludge around them not being able to shutdown to "single user mode". If
we cared about that case, we would have easily preserved non-racy support
for it in `shutdown`.
Additionally, though I think this code does read more easily by reducing
inconsistencies, we didn't come up with any use case for delay inhibitors
v.s. shutdown.[1] The SIGTERM v.s. SIGKILL delay is more general, and we
allow a whole 90 seconds for it, not just 5. So I don't think keeping this
approach bears a risk of significant damage.
[1] https://www.freedesktop.org/wiki/Software/systemd/inhibit/
> We don't want to shutdown while a suspend is running, and vice versa.
> This would be confusing and could lead to data loss in the worst case.
https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1441253/comments/4
According to the above comment, if the conflicting operation is hung,
we don't want to force things when the admin has not passed a force option.
Similarly if you're not an admin, you probably shouldn't get to sneak
around this check by using a scheduled shutdown instead of an unscheduled
one. (And no-one so far thought it necessary to add such a permission in
PolKit).
Note that if the conflicting operation was _not_ hung, and we lost the
race with suspend, the system might not have shut down at the scheduled
time anyway. Which is no good if you were scheduling a power outage.
And scheduling a shutdown for an arbitrary time when the system is resumed,
does not seem a very useful semantic. More likely, scheduled shutdowns are
useful on systems which do not use suspend, such as multi-user servers.
(In which case even PolKit defaults likely don't let the users trigger
suspend).
Apparently, PAM documents that the PAM_TTY should come with a /dev
prefix, but we don't expect it so far, except that Wayland ends up
setting it after all, the way the docs suggest. Hence, let's simply drop
the /dev prefix if it is there.
Fixes: #6516
Checking for validity of a PID is relatively easy, but let's add a
helper cal for this too, in order to make things more readable and more
similar to uid_is_valid(), gid_is_valid() and friends.
Let's add a proper validation function, since validation isn't entirely
trivial. Make use of it where applicable. Also make use of
AUDIT_SESSION_INVALID where we need a marker for an invalid audit
session.
As a follow-up for db3f45e2d2 let's do the
same for all other cases where we create a FILE* with local scope and
know that no other threads hence can have access to it.
For most cases this shouldn't change much really, but this should speed
dbus introspection and calender time formatting up a bit.
This moves pretty much all uses of getpid() over to getpid_raw(). I
didn't specifically check whether the optimization is worth it for each
replacement, but in order to keep things simple and systematic I
switched over everything at once.
Now that we have support for key/switch masking in logind, we can relax
the rules by which logind picks the devices to watch a bit, after all we
won't wake up anymore for every single event, but instead only the
events we actually care about.
This should make power/suspend keys on normal usb/atkbd keyboards just
work.
This way logind will get woken up only when an actual event took place,
and not for every key press on the system.
The ioctl EVIOCSMASK was added by @dvdhrm already in October 2015, for
the use in logind, among others, hence let's actually make use of it
now.
While we are at it, also fix usage of the EVIOCGSW ioctl, where we
assumed a byte array, even though a unsigned long native endian array is
returned.
This patch ensures that session devices are saved for each session.
In order to make the revokation logic work when logind is restarted, the
session devices are now saved in the session state files and their respective
file descriptors sent to PID1's fdstore in order to keep them open accross
restart.
This is mandatory in order to keep the revokation logic working. Indeed in case
of input-devices, the same file descriptors must be shared by logind and a
given session controller in order EVIOCREVOKE to work otherwise multiple
sessions can have device access in parallel.
This should be the only remaining and missing piece for making logind fully
restartable.
Fixes: #1163
When assigning a new session controller to a session, the VT is prepared so the
controller can expect the VT to be in a good default state.
However when logind is restarted and a session controller already took control
of a session, there's no need to prepare th VT otherwise logind may screw up
the VT state set by the controller.
This patch prevents the preparation of the VT in this case.
Instead of always letting logind guess what the caller's session is, let's
give it the value from $XDG_SESSION_ID when it is present in the caller's
environment.
Nowadays terminal emulators are often running as services under systemd --user,
and not as part of an actual session, so all loginctl calls which depend on
logind guessing the session will fail. I don't see a reason not to honour
$XDG_SESSION_ID.
This applies to LockSession, UnlockSession, TerminateSession, ActivateSession,
SetUserLinger.
Fixes#6032.
Using conf.set() with a boolean argument does the right thing:
either #ifdef or #undef. This means that conf.set can be used unconditionally.
Previously I used '1' as the placeholder value, and that needs to be changed to
'true' for consistency (under meson 1 cannot be used in boolean context). All
checks need to be adjusted.
The indentation for emacs'es meson-mode is added .dir-locals.
All files are reindented automatically, using the lasest meson-mode from git.
Indentation should now be fairly consistent.
This is pretty ugly, because I don't know how to use a single
definition for two purposes:
- --version-script needs a path relative to the build root
- link_depends needs a path relative to source root
Also, link_depends does not accept files() output
[https://github.com/mesonbuild/meson/issues/1172], and I don't see a way to go
from files() output to a string path that can be used to craft the -Wl arg.
Ideally, a single files() result could be used in both places.
I'm leaving this as a separate commit for now.
It's crucial that we can build systemd using VS2010!
... er, wait, no, that's not the official reason. We need to shed old systems
by requring python 3! Oh, no, it's something else. Maybe we need to throw out
345 years of knowlege accumulated in autotools? Whatever, this new thing is
cool and shiny, let's use it.
This is not complete, I'm throwing it out here for your amusement and critique.
- rules for sd-boot are missing. Those might be quite complicated.
- rules for tests are missing too. Those are probably quite simple and
repetitive, but there's lots of them.
- it's likely that I didn't get all the conditions right, I only tested "full"
compilation where most deps are provided and nothing is disabled.
- busname.target and all .busname units are skipped on purpose.
Otherwise, installation into $DESTDIR has the same list of files and the
autoconf install, except for .la files.
It'd be great if people had a careful look at all the library linking options.
I added stuff until things compiled, and in the end there's much less linking
then in the old system. But it seems that there's still a lot of unnecessary
deps.
meson has a `shared_module` statement, which sounds like something appropriate
for our nss and pam modules. Unfortunately, I couldn't get it to work. For the
nss modules, we need an .so version of '2', but `shared_module` disallows the
version argument. For the pam module, it also didn't work, I forgot the reason.
The handling of .m4 and .in and .m4.in files is rather awkward. It's likely
that this could be simplified. If make support is ever dropped, I think it'd
make sense to switch to a different templating system so that two different
languages and not required, which would make everything simpler yet.
v2:
- use get_pkgconfig_variable
- use sh not bash
- use add_project_arguments
v3:
- drop required:true and fix progs/prog typo
v4:
- use find_library('bz2')
- add TTY_GID definition
- define __SANE_USERSPACE_TYPES__
- use join_paths(prefix, ...) is used on all paths to make them all absolute
v5:
- replace all declare_dependency's with []
- add more conf.get guards around optional components
v6:
- drop -pipe, -Wall which are the default in meson
- use compiler.has_function() and compiler.has_header_symbol instead of the
hand-rolled checks.
- fix duplication in 'liblibsystemd' library name
- use the right .sym file for pam_systemd
- rename 'compiler' to 'cc': shorter, and more idiomatic.
v7:
- use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D
- rename prefix to prefixdir, rootprefix to rootprefixdir
("prefix" is too common of a name and too easy to overwrite by mistake)
- wrap more stuff with conf.get('ENABLE...') == 1
- use rootprefix=='/' and rootbindir as install_dir, to fix paths under
split-usr==true.
v8:
- use .split() also for src/coredump. Now everything is consistent ;)
- add rootlibdir option and use it on the libraries that require it
v9:
- indentation
v10:
- fix check for qrencode and libaudit
v11:
- unify handling of executable paths, provide options for all progs
This makes the meson build behave slightly differently than the
autoconf-based one, because we always first try to find the executable in the
filesystem, and fall back to the default. I think different handling of
loadkeys, setfont, and telinit was just a historical accident.
In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs.
In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin),
but in Debian, those directories are not included in the path.
C.f. https://github.com/mesonbuild/meson/issues/1576.
- call all the options 'xxx-path' for clarity.
- sort man/rules/meson.build properly so it's stable
Embedding sd_id128_t's in constant strings was rather cumbersome. We had
SD_ID128_CONST_STR which returned a const char[], but it had two problems:
- it wasn't possible to statically concatanate this array with a normal string
- gcc wasn't really able to optimize this, and generated code to perform the
"conversion" at runtime.
Because of this, even our own code in coredumpctl wasn't using
SD_ID128_CONST_STR.
Add a new macro to generate a constant string: SD_ID128_MAKE_STR.
It is not as elegant as SD_ID128_CONST_STR, because it requires a repetition
of the numbers, but in practice it is more convenient to use, and allows gcc
to generate smarter code:
$ size .libs/systemd{,-logind,-journald}{.old,}
text data bss dec hex filename
1265204 149564 4808 1419576 15a938 .libs/systemd.old
1260268 149564 4808 1414640 1595f0 .libs/systemd
246805 13852 209 260866 3fb02 .libs/systemd-logind.old
240973 13852 209 255034 3e43a .libs/systemd-logind
146839 4984 34 151857 25131 .libs/systemd-journald.old
146391 4984 34 151409 24f71 .libs/systemd-journald
It is also much easier to check if a certain binary uses a certain MESSAGE_ID:
$ strings .libs/systemd.old|grep MESSAGE_ID
MESSAGE_ID=%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x
MESSAGE_ID=%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x
MESSAGE_ID=%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x
MESSAGE_ID=%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x
$ strings .libs/systemd|grep MESSAGE_ID
MESSAGE_ID=c7a787079b354eaaa9e77b371893cd27
MESSAGE_ID=b07a249cd024414a82dd00cd181378ff
MESSAGE_ID=641257651c1b4ec9a8624d7a40a9e1e7
MESSAGE_ID=de5b426a63be47a7b6ac3eaac82e2f6f
MESSAGE_ID=d34d037fff1847e6ae669a370e694725
MESSAGE_ID=7d4958e842da4a758f6c1cdc7b36dcc5
MESSAGE_ID=1dee0369c7fc4736b7099b38ecb46ee7
MESSAGE_ID=39f53479d3a045ac8e11786248231fbf
MESSAGE_ID=be02cf6855d2428ba40df7e9d022f03d
MESSAGE_ID=7b05ebc668384222baa8881179cfda54
MESSAGE_ID=9d1aaa27d60140bd96365438aad20286
And then show it, to make things a bit friendlier to the user if we fail
acquiring some props.
In fact, this fixes a number of actual bugs, where we used an error
structure for output that we actually never got an error in.
The 'Sessions' property for both org.freedesktop.login1.User and
org.freedesktop.login1.Seat is marked as EmitsChangedSignal(false).
Trying to emit a change signal that includes the 'Sessions' property
leads to the signal not being sent at all.
Fixes#5210.
gperf-3.1 generates lookup functions that take a size_t length
parameter instead of unsigned int. Test for this at configure time.
Fixes: https://github.com/systemd/systemd/issues/5039
We want that systemd --user gets its own keyring as usual, even if the
barebones PAM snippet we ship upstream is used. If we don't do this we get the
basic keyring systemd --system sets up for us.
Let's use chase_symlinks() everywhere, and stop using GNU
canonicalize_file_name() everywhere. For most cases this should not change
behaviour, however increase exposure of our function to get better tested. Most
importantly in a few cases (most notably nspawn) it can take the correct root
directory into account when chasing symlinks.
We don't have plural in the name of any other -util files and this
inconsistency trips me up every time I try to type this file name
from memory. "formats-util" is even hard to pronounce.
It is impossible to ship a fully generic PAM configuration upstream.
Therefore, ship a minimal configuration with the systemd --user requirements,
and add a note to DISTRO_PORTING documenting this.
Fixes#4284
The parsing functions for [User]TasksMax were inconsistent. Empty string and
"infinity" were interpreted as no limit for TasksMax but not accepted for
UserTasksMax. Update them so that they're consistent with other knobs.
* Empty string indicates the default value.
* "infinity" indicates no limit.
While at it, replace opencoded (uint64_t) -1 with CGROUP_LIMIT_MAX in TasksMax
handling.
v2: Update empty string to indicate the default value as suggested by Zbigniew
Jędrzejewski-Szmek.
v3: Fixed empty UserTasksMax handling.
This adds the boolean RemoveIPC= setting to service, socket, mount and swap
units (i.e. all unit types that may invoke processes). if turned on, and the
unit's user/group is not root, all IPC objects of the user/group are removed
when the service is shut down. The life-cycle of the IPC objects is hence bound
to the unit life-cycle.
This is particularly relevant for units with dynamic users, as it is essential
that no objects owned by the dynamic users survive the service exiting. In
fact, this patch adds code to imply RemoveIPC= if DynamicUser= is set.
In order to communicate the UID/GID of an executed process back to PID 1 this
adds a new "user lookup" socket pair, that is inherited into the forked
processes, and closed before the exec(). This is needed since we cannot do NSS
from PID 1 due to deadlock risks, However need to know the used UID/GID in
order to clean up IPC owned by it if the unit shuts down.
This reverts commit 8121f4d209.
The special 'key handling' inhibitors should always work regardless of
any *IgnoreInhibited settings – otherwise they're nearly useless.
Reverts: #3470Fixes: #3897
config_parse_user_tasks_max() was incorrectly accepting percentage value
between 1 and 99. Update it to accept 0% and 100%. This brings it in line
with TasksMax handling in systemd.
Let's change from a fixed value of 12288 tasks per user to a relative value of
33%, which with the kernel's default of 32768 translates to 10813. This is a
slight decrease of the limit, for no other reason than "33%" sounding like a nice
round number that is close enough to 12288 (which would translate to 37.5%).
(Well, it also has the nice effect of still leaving a bit of room in the PID
space if there are 3 cooperating evil users that try to consume all PIDs...
Also, I like my bikesheds blue).
Since the new value is taken relative, and machined's TasksMax= setting
defaults to 16384, 33% inside of containers is usually equivalent to 5406,
which should still be ample space.
To summarize:
| on the host | in the container
old default | 12288 | 12288
new default | 10813 | 5406
This way systemd is informed that we consider everything inside the scope as
"left-over", and systemd can log about killing it.
With this change systemd will log about all processes killed due to the session
clean-up on KillUserProcesses=yes.
sd-bus generally exposes bools as "int" instead of "bool" in the public API.
This is relevant when unmarshaling booleans, as the relevant functions expect
an int* pointer and no bool* pointer. Since sizeof(bool) is not necessarily the
same as sizeof(int) this is problematic and might result in memory corruption.
Let's fix this, and make sure bus_map_all_properties() handles booleans as
ints, as the rest of sd-bus, and make all users of it expect the right thing.
Delete the dbus1 generator and some critical wiring. This prevents
kdbus from being loaded or detected. As such, it will never be used,
even if the user still has a useful kdbus module loaded on their system.
Sort of fixes#3480. Not really, but it's better than the current state.
If "systemctl -H" is used, let's make sure we first terminate the bus
connection, and only then close the pager. If done in this order ssh will get
an EOF on stdin (as we speak D-Bus through ssh's stdin/stdout), and then
terminate. This makes sure the standard error we were invoked on is released by
ssh, and only that makes sure we don't deadlock on the pager which waits for
all clients closing its input pipe.
(Similar fixes for the various other xyzctl tools that support both pagers and
-H)
Fixes: #3543
We need to explicitly define authorizations for allow_inactive and
allow_active. Otherwise one is getting "Access denied" when run from a
local console:
$ loginctl enable-linger
Could not enable linger: Access denied
This reverts commit 483d8bbb4c.
In [1] Michel Dänzer and Daniel Vetter wrote:
>> The scenario you describe isn't possible if the Wayland compositor
>> directly uses the KMS API of /dev/dri/card*, but it may be possible if
>> the Wayland compositor uses the fbdev API of /dev/fb* instead (e.g. if
>> weston uses its fbdev backend).
>
> Yeah, if both weston and your screen grabber uses native fbdev API you can
> now screenshot your desktop. And since fbdev has no concept of "current
> owner of the display hw" like the drm master, I think this is not fixable.
> At least not just in userspace. Also even with native KMS compositors
> fbdev still doesn't have the concept of ownership, which is why it doesn't
> bother clearing it's buffer before KMS takes over. I agree that this
> should be reverted or at least hidden better.
TBH, I think that privilege separation between processes running under the same
UID is tenuous. Even with drm, in common setups any user process can ptrace the
"current owner of the display" and call DROP_MASTER or do whatever. It *is*
possible to prevent that, e.g. by disabling ptrace using yama.ptrace_scope, or
selinux, and so on, but afaik this is not commonly done. E.g. all Fedora
systems pull in elfutils-default-yama-scope.rpm through dependencies which sets
yama.ptrace_scope=0. And even assuming that ptrace was disabled, it is trivial
to modify files on disk, communicate through dbus, etc; there is just to many
ways for a non-sandboxed process to interact maliciously with the display shell
to close them all off. To achieve real protection, some sort of sandboxing
must be implemented, and in that case there is no need to rely on access mode
on the device files, since much more stringent measures have to be implemented
anyway.
The situation is similar for framebuffer devices. It is common to add
framebuffer users to video group to allow them unlimited access to /dev/fb*.
Using uaccess would be better solution in that case. Also, since there is no
"current owner" limitation like in DRM, processes running under the same UID
should be able to access /proc/<pid-of-display-server>/fd/* and gain access to
the devices. Nevertheless, weston implements a suid wrapper to access the
devices and then drop privileges, and this patch would make this daemon
pointless. So if the weston developers feel that this change reduces security,
I prefer to revert it.
[1] https://lists.freedesktop.org/archives/wayland-devel/2016-May/029017.html
That function doesn't draw anything on it's own, just returns a string, which
sometimes is more than one character. Also remove "DRAW_" prefix from character
names, TREE_* and ARROW and BLACK_CIRCLE are unambigous on their own, don't
draw anything, and are always used as an argument to special_glyph().
Rename "DASH" to "MDASH", as there's more than one type of dash.
For similar reasons as the recent addition of a limit on sessions.
Note that we don't enforce a limit on inhibitors per-user currently, but
there's an implicit one, since each inhibitor takes up one fd, and fds are
limited via RLIMIT_NOFILE, and the limit on the number of processes per user.
If we have a lot of simultaneous sessions we really shouldn't send the full
list of active sessions with each PropertyChanged message for user and seat
objects, as that can become quite substantial data, we probably shouldn't dump
on the bus on each login and logout.
Note that the global list of sessions doesn't send out changes like this
either, it only supports requesting the session list with ListSessions().
If cients want to get notified about sessions coming and going they should
subscribe to SessionNew and SessionRemoved signals, and clients generally do
that already.
This is kind of an API break, but then again the fact that this was included
was never documented.
Let's make sure we process session and inhibitor pipe fds (that signal
sessions/inhibtors going away) at a higher priority
than new bus calls that might create new sessions or inhibitors. This helps
ensuring that the number of open sessions stays minimal.
We really should put limits on all resources we manage, hence add one to the
number of concurrent sessions, too. This was previously unbounded, hence set a
relatively high limit of 8K by default.
Note that most PAM setups will actually invoke pam_systemd prefixed with "-",
so that the return code of pam_systemd is ignored, and the login attempt
succeeds anyway. On systems like this the session will be created but is not
tracked by systemd.
The macro determines the right length of a AF_UNIX "struct sockaddr_un" to pass to
connect() or bind(). It automatically figures out if the socket refers to an
abstract namespace socket, or a socket in the file system, and properly handles
the full length of the path field.
This macro is not only safer, but also simpler to use, than the usual
offsetof() + strlen() logic.
This way the user service will have a loginuid, and it will be inherited by
child services. This shouldn't change anything as far as systemd itself is
concerned, but is nice for various services spawned from by systemd --user
that expect a loginuid.
pam_loginuid(8) says that it should be enabled for "..., crond and atd".
user@.service should behave similarly to those two as far as audit is
concerned.
https://bugzilla.redhat.com/show_bug.cgi?id=1328947#c28
This ports over machinectl and loginctl to also use the new GetProcesses() bus
call to show the process tree of a container or login session. This is similar
to how systemctl already has been ported over in a previous commit.
We enable lingering for anyone who wants this. It is still disabled by
default to avoid keeping long-running processes accidentally.
Admins might want to customize this policy on multi-user sites.
Instead of KillOnlyUsers being a filter for KillUserProcesses, it can now be
used to specify users to kill, independently of the KillUserProcesses
setting. Having the settings orthogonal seems to make more sense. It also
makes KillOnlyUsers symmetrical to KillExcludeUsers.
This ensures that users sessions are properly cleaned up after.
The admin can still enable or disable linger for specific users to allow
them to run processes after they log out. Doing that through the user
session is much cleaner and provides better control.
dbus daemon can now be run in the user session (with --enable-user-session,
added in 1.10.2), and most distributions opted to pick this configuration.
In the normal case it makes a lot of sense to kill remaining processes.
The exception is stuff like screen and tmux. But it's easy enough to
work around, a simple example was added to the man page in previous
commit. In the long run those services should integrate with the systemd
users session on their own.
https://bugs.freedesktop.org/show_bug.cgi?id=94508https://github.com/systemd/systemd/issues/2900
With this option, systemctl will only print the rhs in show:
$ systemctl show -p Wants,After systemd-journald --value
systemd-journald.socket ...
systemd-journald-dev-log.socket ...
This is useful in scripts, because the need to call awk or similar
is removed.
It's possible that sd_bus_creds_get_tty() fails and thus
scheduled_shutdown_tty is NULL in method_schedule_shutdown().
Fix logind_wall_tty_filter() to get along with that, by showing the message on
all TTYs, instead of crashing in strcmp().
https://launchpad.net/bugs/1553040
Many subsystems define own pager_open_if_enabled() function which
checks '--no-pager' command line argument and open pager depends
on its value. All implementations of pager_open_if_enabled() are
the same. Let's merger this function with pager_open() from the
shared/pager.c and remove pager_open_if_enabled() from all subsytems
to prevent code duplication.
The deserialize_timestamp_value() is renamed timestamp_deserialize() to be more
consistent with dual_timestamp_deserialize()
And add the NULL check back on realtime and monotonic
systemd-logind uses mkdir_label and label_fix functions without calling
first mac_selinux_init. This makes /run/user/$UID/ directories not
labelled correctly on an Arch Linux system using SELinux.
Fix this by calling mac_selinux_init("/run") early in systemd-logind.
This makes files created in /etc/udev/rules.d and /var/lib/systemd to be
labelled through transitions in the SELinux policy instead of using
setfscreatecon (with mac_selinux_create_file_prepare).
Issue #2388 suggests the current TasksMax= setting for user processes is to low. Bump it to 12K. Also, bump the
container TasksMax= from 8K to 16K, so that it remains higher than the one for user processes.
(Compare: the kernel default limit for processes system-wide is 32K).
Fixes#2388
... to determine if color output should be enabled. If the variable is not set,
fall back to using on_tty(). Also, rewrite existing code to use
colors_enabled() where appropriate.
manager_{start,stop}_{slice,scope,unit} functions had an optional job
output parameter. But all callers specified job, so make the parameter
mandatory, add asserts. Also extract common job variable handling to
a helper function to avoid duplication.
Avoids gcc warning about job being unitialized.
gcc is confused by the common idiom of
return errno ? -errno : -ESOMETHING
and thinks a positive value may be returned. Replace this condition
with errno > 0 to help gcc and avoid many spurious warnings. I filed
a gcc rfe a long time ago, but it hard to say if it will ever be
implemented [1].
Both conventions were used in the codebase, this change makes things
more consistent. This is a follow up to bcb161b023.
[1] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61846
method_schedule_shutdown referenced org.freedesktop.login1.poweroff*
which is never registered in polkit.
Now refers to org.freedesktop.login1.power-off*
Signed-off-by: Joost Bremmer <toost.b@gmail.com>
GLIB has recently started to officially support the gcc cleanup
attribute in its public API, hence let's do the same for our APIs.
With this patch we'll define an xyz_unrefp() call for each public
xyz_unref() call, to make it easy to use inside a
__attribute__((cleanup())) expression. Then, all code is ported over to
make use of this.
The new calls are also documented in the man pages, with examples how to
use them (well, I only added docs where the _unref() call itself already
had docs, and the examples, only cover sd_bus_unrefp() and
sd_event_unrefp()).
This also renames sd_lldp_free() to sd_lldp_unref(), since that's how we
tend to call our destructors these days.
Note that this defines no public macro that wraps gcc's attribute and
makes it easier to use. While I think it's our duty in the library to
make our stuff easy to use, I figure it's not our duty to make gcc's own
features easy to use on its own. Most likely, client code which wants to
make use of this should define its own:
#define _cleanup_(function) __attribute__((cleanup(function)))
Or similar, to make the gcc feature easier to use.
Making this logic public has the benefit that we can remove three header
files whose only purpose was to define these functions internally.
See #2008.
If we requeue jobs, we are no longer interested in old jobs. Hence, we
better ignore any JobRemoved signals for old jobs and concentrate on our
replacements.
When queuing unit jobs, we should rather replace existing units than
fail. This is especially important when we queued a user-shutdown and a
new login is encountered. In this case, we better raplce the shutdown
jobs. systemd takes care of everything else.
If the last reference to a user is released, we queue stop-jobs for the
user-service and slice. Only once those are finished, we drop the
user-object. However, if a new session is opened before the user object is
fully dropped, we currently incorrectly re-use the object. This has the
effect, that we get stale sessions without a valid "systemd --user"
instance.
Fix this by properly allowing user_start() to be called, even if
user->stopping is true.
Just like user->slice, there is no reason to store the unit name in /run,
nor should we allocate it dynamically on job instantiation/removal. Just
keep it statically around at all times and rely on user->started ||
user->stopping to figure out whether the unit exists or not.
Few changes to user_new() and user_free():
- Use _cleanup_(user_freep) in constructor
- return 'int' from user_new()
- make user_free() deal with partially initialized objects
- keep reverse-order in user_free() compared to user_new()
- make user_free() return NULL
- make user_free() accept NULL as no-op
Currently, we allocate user->slice when starting a slice, but we never
release it. This is incompatible if we want to re-use a user object once
it was stopped. Hence, make sure user->slice is allocated statically on
the user object and use "u->started || u->stopping" as an indication
whether the slice is actually available on pid1 or not.
Lets not pretend we support changing XDG_RUNTIME_DIR via logind state
files. There is no reason to ever write the string into /run, as we
allocate it statically based on the UID, anyway. Lets stop that and just
allocate the runtime_path in "struct User" at all times.
We keep writing it into the /run state to make sure pam_systemd of
previous installs can still read it. However, pam_systemd is now fixed to
allocate it statically as well, so we can safely remove that some time in
the future.
Last but not least: If software depends on systemd, they're more than free
to assume /run/user/$uid is their runtime dir. Lets not require sane
applications to query the environment to get their runtime dir. As long as
applications know their login-UID, they should be safe to deduce the
runtime dir.
This new setting configures the TasksMax= field for the slice objects we
create for each user.
This alters logind to create the slice unit as transient unit explicitly
instead of relying on implicit generation of slice units by simply
starting them. This also enables us to set a friendly description for
slice units that way.
The macro is generically useful for putting together search paths, hence
let's make it truly generic, by dropping the implicit ".d" appending it
does, and leave that to the caller. Also rename it from
CONF_DIRS_NULSTR() to CONF_PATHS_NULSTR(), since it's not strictly about
dirs that way, but any kind of file system path.
Also, mark CONF_DIR_SPLIT_USR() as internal macro by renaming it to
_CONF_PATHS_SPLIT_USR() so that the leading underscore indicates that
it's internal.
The files are named too generically, so that they might conflict with
the upstream project headers. Hence, let's add a "-util" suffix, to
clarify that this are just our utility headers and not any official
upstream headers.
So far we had two pretty much identical calls in user-util.[ch]:
lookup_uid() and uid_to_name(). Get rid of the former, in favour of the
latter, and while we are at it, rewrite it, to use getpwuid_r()
correctly, inside an allocation loop, as POSIX intended.
There are more than enough calls doing string manipulations to deserve
its own files, hence do something about it.
This patch also sorts the #include blocks of all files that needed to be
updated, according to the sorting suggestions from CODING_STYLE. Since
pretty much every file needs our string manipulation functions this
effectively means that most files have sorted #include blocks now.
Also touches a few unrelated include files.
When the Suspend method is called, the only log message we write
(unless debugging is enabled) is "Operation finished.". This is
not very helpful when trying to figure out what is going on, so
add what operation we are talking about to the message:
"Operation 'sleep' finished.".
Hat tip to Daniel Aleksandersen for pointing this out.
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.
- 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
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"
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.
This also allows us to drop build.h from a ton of files, hence do so.
Since we touched the #includes of those files, let's order them properly
according to CODING_STYLE.
Adding additional keys prevents this gpio-keys powerswitch from working,
e.g. this wouldn't poweroff:
button@23 {
label = "power-switch";
linux,code = <116>;
gpios = <&gpio 23 1>;
};
button@25 {
label = "KEY_A";
linux,code = <30>;
gpios = <&gpio 25 1>;
};
Changing ATTRS{keys}=="116" to ATTRS{keys}=="*116*" makes the
power-switch and the A key both work properly.
(David: rephrase and merge-commits)
off_t is a really weird type as it is usually 64bit these days (at least
in sane programs), but could theoretically be 32bit. We don't support
off_t as 32bit builds though, but still constantly deal with safely
converting from off_t to other types and back for no point.
Hence, never use the type anymore. Always use uint64_t instead. This has
various benefits, including that we can expose these values directly as
D-Bus properties, and also that the values parse the same in all cases.
Allow passing a "dry-" prefix to the action parameter passed to
.ScheduleShutdown(). When strings with this prefix are passed,
the scheduled action will not take place. Instead, an info message
is logged.
The previous coccinelle semantic patch that improved usage of
log_error_errno()'s return value, only looked for log_error_errno()
invocations with a single parameter after the error parameter. Update
the patch to handle arbitrary numbers of additional arguments.
Turns this:
r = -errno;
log_error_errno(errno, "foo");
into this:
r = log_error_errno(errno, "foo");
and this:
r = log_error_errno(errno, "foo");
return r;
into this:
return log_error_errno(errno, "foo");
Extra details for an action can be supplied when calling polkit's
CheckAuthorization method. Details are a list of key/value string pairs.
Custom policy can use these details when making authorization decisions.
We treat an empty wall-message equal to a NULL wall-message since:
commit 5744f59a3e
Author: Lennart Poettering <lennart@poettering.net>
Date: Fri Sep 4 10:34:47 2015 +0200
logind: treat an empty wall message like a NULL one
Fix the shutdown scheduler to not deref a NULL pointer, but properly
check for an empty wall-message.
Fixes: #1120
In all cases where the function (or cg_is_empty_recursive()) ignoring
the calling process is actually wrong, as a process keeps a cgroup busy
regardless if its the current one or another. Hence, let's simplify
things and drop the "ignore_self" parameter.
On Dell and HP laptops the dock state/events (SW_DOCK) come from the "{Dell,HP}
WMI hotkeys" input devices. Tag them as power-switch so that login actually
considers them. Use a general match in case this affects other vendors, too.
Thanks to Andreas Schultz for debugging this!
https://launchpad.net/bugs/1450009
dbus-1.10 was just released, including systemd units to run
`dbus-daemon --session` as systemd user unit. This allows using a
user-bus with dbus1, just like we do per default with kdbus.
All the dbus libraries have already been fixed long ago to use the
user-bus as default. Hence, there's no need to set
DBUS_SESSION_BUS_ADDRESS= if we use the user-bus. However, gdm and
friends continue to spawn a session bus if this variable is not set
(instead of checking for the existence of the user-bus). Hence, we force
the user-bus, if it is available, in pam_systemd. Once gdm and friends
are fixed, we can continue to drop this again. However, that might take
a while.
With this in place, all that is needed to make the user-bus work is:
`systemctl --global enable dbus.socket`
If dbus.socket is not enabled, the legacy session-bus is still used.
Based on a patch by: Jan Alexander Steffens <jan.steffens@gmail.com>
Enable unprivileged users to set wall message on a shutdown
operation. When the message is set via the --message option,
it is logged together with the default shutdown message.
$ systemctl reboot --message "Applied kernel updates."
$ journalctl -b -1
...
systemd-logind[27]: System is rebooting. (Applied kernel updates.)
...
When the controlling process exits, any existing file descriptors
for that FD will be marked as hung-up and ioctls on them will
file with EIO. To work around this, open a new file descriptor
for the VT we want to clean up.
Thanks to Ray Strode for help in sorting out the problem and
coming up with a fix!
https://github.com/systemd/systemd/issues/989
The open_terminal() function adds retries in case a terminal
is in the process of being closed when we open it, and should
generally be used to open a terminal. We especially need it
for code that a subsequent commit adds that reopens the terminal
at session shut-down time; such races would be more likely in
that case.
Found by Ray Strode.
The following functions return immediately if a null pointer was passed.
* calendar_spec_free
* link_address_free
* manager_free
* sd_bus_unref
* sd_journal_close
* udev_monitor_unref
* udev_unref
It is therefore not needed that a function caller repeats a corresponding check.
This issue was fixed by using the software Coccinelle 1.0.1.
Since dacd6cee76 the two OOM's are
ignored as the value of r will be overwritten and we only log in
the fail section anyway.
This patch jumps to fail on OOM.
Note that this is different behavior compared to both the current
code and previous to dacd6cee76. Before
that commit we would log that saving the inhibit data failed, but
still write the file, though without the WHO/WHY section.
CID# 1313545
Use mfree() where we can.
Drop unnecessary {}.
Drop unnecessary variable declarations.
Cast syscall invocations where explicitly don't care for the return
value to (void).
Reword a comment.
If we get a weird signal, then we should log about it, but not return an
error, since sd-bus will not call us again then anymore, but for these
signals we match here we actually do want to be called on the next
invocation.
Let logind use the sd_bus_track helper object to track the controllers of
sessions. This does not only remove quite some code but also kills the
unconditional matches for all NameOwnerChanged signals.
The latter is something we should never ever do, as it wakes up the daemon
every time a client connects, which doesn't scale.