In systemctl-show we only show current swap if ever swapped or non-zero. This
reduces the noise on swapless systems, that would otherwise always show a swap
value that never has the chance to become non-zero. It further reduces the
noise for services that never swapped.
Linux's Control Group v2 interfaces exposes memory.peak, which contains the
"max memory usage recorded for the cgroup and its descendants since the
creation of the cgroup."
This commit adds a new property "MemoryPeak" for units and makes "systemctl
show" display this value if it is available.
Fixes#29878.
Signed-off-by: Florian Schmaus <flo@geekplace.eu>
This makes sure unit_watch_pid() and unit_unwatch_pid() will track
processes by pidfd if supported. Also ports over some related code.
Should not really change behaviour.
Note that this does *not* add support waiting for POLLIN on the pidfds
as additional exit notification. This is left for a later commit (this
commit is already large enough), in particular as that would add new
logic and not just convert existing logic.
Add a new boolean for units, SurviveFinalKillSignal=yes/no. Units that
set it will not have their process receive the final sigterm/sigkill in
the shutdown phase.
This is implemented by checking if a process is part of a cgroup marked
with a user.survive_final_kill_signal xattr (or a trusted xattr if we
can't set a user one, which were added only in kernel v5.7 and are not
supported in CentOS 8).
This ports over unit_main_pid() + unit_control_pid() to return PidRef*
pointers (which also means the underlying UnitVTable function pointers
are changed accordingly).
This then uses te functions to simplify the unit_kill() call, by
avoiding the kill() vtable indirection and instead just suing
unit_main_pid() and unit_control_pid() directly.
bus_unit_queue_job_one has two callers:
- bus_unit_queue_job which would do the appropriate transormations
to turn JOB_TRY_RESTART into JOB_TRY_RELOAD,
- and method_enqueue_marked_jobs which did not.
In effect, method_enqueue_marked_jobs() would queue restart jobs for
units which has Markers= needs-reload or needs-restart.
When the chunk of code which does the transformations is moved from
bus_unit_queue_job to bus_unit_queue_job_one, there is no change for
bus_unit_queue_job, and method_enqueue_marked_jobs is fixed.
The additional checks that are done seem reasonable to do from
method_enqueue_marked_jobs: we shouldn't be restarting units which are
configured to not allow that, or force unwanted start of dbus-broker.
As it says on the tin, configures the unit to survive a soft reboot.
Currently all the following options have to be set by hand:
Conflicts=reboot.target kexec.target poweroff.target halt.target
Before=reboot.target kexec.target poweroff.target halt.target
After=sysinit.target basic.target
DefaultDependencies=no
IgnoreOnIsolate=yes
This is not very user friendly. If new default dependencies are added,
or new shutdown/reboot types, they also have to be added manually.
The new option is much simpler, easy to find, and does the right thing
by default.
Let's make sure that D-Bus messages are always sent out when pending,
before we might GC a unit/job.
This is kinda a follow-up for 8db998981a,
and a similar logic really applies: GC should only be done if we
processed everything else, generated evertyhing else and really don't
need it anymore.
Follow-up for #27579
In #27579 we refused all StartUnit requests for Type=dbus units
if dbus is not running, which means if dbus is manually stopped,
user can't use systemctl to start Type=dbus units again, which
is incorrect.
The only culprit that leads to the cancellation of the whole
transaction mentioned in #26799 is job type conflict on dbus.
So let's relax the restriction and only refuse job enqueuing
if dbus has a stop job.
To summarize, the case we want to avoid is:
1. dbus has a stop job installed
2. StartUnit/ActivationRequest is received
3. Type=dbus service gets started, which has Requires=dbus.socket
4. dbus is pulled in again, resulting in job type conflict
What we can support is:
1. dbus is already stopped
2. StartUnit is received (possibly through systemctl, i.e. on private bus)
3. Type=dbus service gets started, which will wait for dbus to start
4. dbus is started again, thus the job for Type=dbus service
Replaces #27590Fixes#27588
dbus-broker issues StartUnit directly for activation requests,
so let's add a check on bus state in bus_unit_queue_job to refuse
that if dbus is not running.
Replaces #27570Closes#26799
The ignore directive specifies to not do anything with the given
unit and leave existing configuration intact. This allows distributions
to gradually adopt preset files by shipping a ignore * preset file.
Now that we have a potentially pinned fdstore let's add a concept for
cleaning it explicitly on user requested. Let's expose this via
"systemctl clean", i.e. the same way as user directories are cleaned.
I was changing how some properties are appended to the StartTransientUnit call
and messed up the message contents. When something is wrong with how the
message is structed, we would return a very generic
"Failed to start transient service unit: No such device or address".
Mention that it was property setting that failed, and translate ENXIO to a
different message. bus_unit_set_properties() or any of the children it calls
may also return other errors, in particular EBADMSG or ENOMEM, but the error
message that is generated for those is understandable, so we don't need to
"translate" them explicitly.
bus_unit_set_properties() is called from two places, so it seems nicer to
generate the message internally, rather than ask the caller to do that. Also,
now bus_unit_set_properties() always sets <error>, which is nicer for the
callers.
In various tools and services we have a per-system and per-user concept.
So far we sometimes used a boolean indicating whether we are in system
mode, or a reversed boolean indicating whether we are in user mode, or
the LookupScope enum used by the lookup path logic.
Let's address that, in introduce a common enum for this, we can use all
across the board.
This is mostly just search/replace, no actual code changes.
This augments the existing KillUnit() + Kill() methods with
QueueSignalUnit() + QueueSignal(), which are what sigqueue() is to
kill().
This is useful for sending our new SIGRTMIN+18 control signals to system
services.
Sometimes a freeze operation can hang due to the presence of kernel
threads inside the unit cgroup (e.g. QEMU-KVM). This ensures that the
ThawUnit operation invoked by systemd-sleep at wakeup always thaws the
unit.
Not wired in by any unit type yet, just the basic to allocate,
ref, deref and plug in to other unit types.
Includes recording the trigger unit name and passing it to the
triggered unit as TRIGGER_UNIT= env var.
E.g. Documentation or Markers could apply to any unit type. This already worked
partially, because a direct dbus call could be made:
After rebuild with the patch, but before the manager has been restarted:
$ build/systemctl --user set-property dev-zram0.swap Markers=+needs-restart
$ build/systemctl --user show -p Markers dev-zram0.swap
Markers=needs-restart
I noticed that that the rpm unit restart helper was throwing errors for target
units. We should just let the Markers be set for those too, even if it doesn't
do anything in the end. This way we don't need to special-case by unit type.
Do not go back to disk on each selinux access, but instead cache the
label off the inode we are actually reading. That way unit file contents
and unit file label we use for access checks are always in sync.
Based on discussions here:
https://github.com/systemd/systemd/pull/10023#issuecomment-1179835586
Replaces:
https://github.com/systemd/systemd/pull/23910
This changes behaviour a bit, because we'll reach and cache the label at
the moment of loading the unit (i.e. usually on boot and reload), but
not after relabelling. Thus, users must refresh the cache explicitly via
a "systemctl daemon-reload" if they relabelled things.
This makes the SELinux story a bit more debuggable, as it adds an
AccessSELinuxContext bus property to units that will report the label we are
using for a unit (or the empty string if not known).
This also drops using the "source" path of a unit as label source. if
there's value in it, then generators should manually copy the selinux
label from the source files onto the generated unit files, so that the
rule that "access labels are read when we read the definition files" is
upheld. But I am not convinced this is really a necessary, good idea.
Backward incompatible change to avoid returning 'skipped' if a condition causes
a job activation to be skipped when using StartUnitWithFlags().
Job results are broadcasted, so it is theoretically possible that existing
software could get confused if they see this result.
Replaces https://github.com/systemd/systemd/pull/22369
When an activation job is skipped because of a Condition*= setting failing,
currently the JobRemoved signal lists 'done' as the result, just as with
a successful job.
This is a problem when doing dbus activation: dbus-broker will receive a
signal that says the job was successful, so then it moves into a state where
it waits for the requested name to appear on the bus, but that never happens
because the job was actually skipped.
Add a new StartUnitWithFlags that changes the behaviour of the JobRemoved
signal to list 'done' or 'skipped'.
Fixes#21520
The function name `method_reload` is used both in dbus-unit.c and
dbus-manager.c for static functions.
With the previous addition of adding the function name to the audit
information on SELinux denials, rename the one (and its relatives) in
dbus-unit.c as most of the functions in src/core/dbus-unit.c are already
prefixed with `bus_unit_`.
Defines a "UNIT_DEPENDENCY_SLICE_PROPERTY" UnitDependencyMask type that
is used when adding slices to the dependencies hashmap. This type is
used to remove slice dependencies when they get overridden by new ones.
Fixes#20182
It's not "const", it can change any time if people change the fs, and we
don#t send out notifications for it. Hence don't claim it was const.
(Otherwise clients might cache it, but they should not)
Prompted-by: #20792
Try to infer the unused memory that a unit can claim before the
memory.max limit is reached, including any limit set on any parent
slice above the unit itself.
Mike Yuan
Florian Schmaus
Yu Watanabe
Daan De Meyer
Lennart Poettering
Luca Boccassi
Zbigniew Jędrzejewski-Szmek
msizanoen1
David Tardon
nl6720
amarjargal
наб
Luca Boccassi
Christian Göttsche
Anita Zhang
Vladimir Panteleev