man: add a new page with a general description of common syntax
We have a common parser, but for the user it might be
completely unobvious that the same general rules apply
to all those files. Let's add a page about the basic syntax
so that the more specific pages don't have to repeat those
details.
dropin: when looking for dropins for a unit, also look within "-" prefix unit dirs
This extends the logic by which we look for drop-ins for unit files when
loading them. Previously for a unit "foo-quux-bar.service" we'd look in
a directory "foo-quux-bar.service.d" accompanying it for extension
dropins. With this change we'll additionally look in:
"foo-quux-.service.d" and "foo-.service.d", i.e. we'll truncate the unit
name after every dash.
This is an alternative to templating for many services, as it permits
configuring defaults for sets of units that all use the same prefix in
the unit name. This is particularly useful in slice, mount and
automount units which reflect a hierarchy of concepts, as it permits
setting defaults for specific subsets of the tree. For example, in order
to provide every user with a memory of 1G it's now possible to do:
copy: drop _unlikely_() that isn't obviously the case
If a tool only invokes copy_bytes() a single time the _unlikely_() will always be
wrong, and is hence not useful. Let's drop it and let the compiler
figure our what to do, instead of misleading it.
copy: hide in copy_bytes() the strange way splice() handles O_NONBLOCK
splice() ignores O_NONBLOCK on pipes but not on other fds. Let's handle
that properly, and query O_ONBLOCK manually in that case, ensuring
systematic behaviour in either case.
Optionally, when we copy between fds with simple read/write, let's
return any remaining data we already read into the buffer if write
fails. This is useful to allow callers to use the read data otherwise,
perhaps implementing a different fallback for copying.
logind: refuse operations if the target unit is masked or unavailable
If hibernate.target is masked, and systemctl hibernate is invoked, havoc ensues.
logind starts the hibernation operation, but then doesn't go through with it;
gnome-shell segfaults. Let's be nice to the user and refuse doing anything in
that case.
$ sudo systemctl mask hibernate.target
$ busctl call org.freedesktop.login1 /org/freedesktop/login1 org.freedesktop.login1.Manager CanHibernate
s "no"
$ sudo systemctl hibernate
Failed to hibernate system via logind: Access denied
Failed to start hibernate.target: Unit hibernate.target is masked.
shared/sleep-fix: fix check if s-then-h is possible
can_sleep() returns 0 if the operation is impossible, but
the code assumed that negative is returned in that case,
in effect reporting s2h was possible even if hibernation or
suspend were not possible.
Devices in Hyper-V/Azure exist on vmbus and are identified by
UUID value. This patch adds a hardware table so that udevadm
can report properties. I chose names are based on the values
reported in Window Device Manager (for consistency).
The table includes several devices that are not used by Linux
but are present and ignored.
sleep: Add support for setting a disk offset when hibernating
The Linux kernel is adding support for configuring the offset
into a disk. This allows swapfiles to be more usable as users
will no longer need to set the offset on their kernel command
line.
Use this API in systemd when hibernating as well.
Signed-off-by: Mario Limonciello <mario.limonciello@dell.com>
analyze: give a hint what is it means that boot is still active
$ build/systemd-analyze time
Bootup is not yet finished (org.freedesktop.systemd1.Manager.FinishTimestampMonotonic=0).
Please try again later.
Hint: Use 'systemctl list-jobs' to see active jobs
analyze: Do no require a full d-bus bus for the plot command (#8539)
The plot command requires a full d-bus bus to fetch the host
information, which seems rather optional, and having a running dbus
daemon is not always desirable. So instead, we try to acquire a full
bus, and if that fails we acquire the systemd bus, in which case we
omit the host information from the output.
We refactor acquire_bus() into two new functions which in addition
makes the call sites clearer.
Martin Wilck [Sat, 7 Apr 2018 15:33:48 +0000 (17:33 +0200)]
systemd-udevd: limit children-max by available memory (#8668)
Udev workers consume typically 50-100MiB virtual memory.
On systems with lots of CPUs and relatively low memory, that may
easily cause workers to be OOM-killed.
This patch limits the number of workers to 8 per GiB memory.
But don't let the limit drop below the smallest value we had
without this patch (8 + 1 * 2 = 10); on small systems, udev's
memory footprint is likely lower.
Clarify checker/helper in systemd-fsck@.service manpage (#8674)
Clarify the helper/checker terminology in the systemd-fsck@.service manpage to
make the description more clear about what is responsible for deciding if a filesystem
needs checking.
Files which are installed as-is (any .service and other unit files, .conf
files, .policy files, etc), are left as is. My assumption is that SPDX
identifiers are not yet that well known, so it's better to retain the
extended header to avoid any doubt.
I also kept any copyright lines. We can probably remove them, but it'd nice to
obtain explicit acks from all involved authors before doing that.
Philip Sequeira [Thu, 5 Apr 2018 14:04:27 +0000 (14:04 +0000)]
nspawn: wait for network namespace creation before interface setup (#8633)
Otherwise, network interfaces can be "moved" into the container's
namespace while it's still the same as the host namespace, in which case
e.g. host0 for a veth ends up on the host side instead of inside the
container.
dissect: when pulling metadata from an image, don't bother with /home or ESP
When we try to read meta-data from an image, don't bother with mounting
/home or the ESP, as that's not where the metadata is. This not only
speeds things up a bit, but also has the benefit that setups where an
unencrypted root is mixed with an encrypted /home (which I have on one
of my own systems) won't result in errors that the crypto key is needed.
1. We'll now explicitly check that the child devices of a block device
we are interested in (i.e. the partitions) are block devices themselves.
On newer kernels the mmc rpmb stuff is actually exposed as char rather
than block device as before, and they probably should have been that in
the first place. By adding this check we'll hence filter out these weird
devices through a second rule too, that hopefully makes things a bit
more future-proof, should more devices like this be added eventually,
or other subsystems do a similar thing.
2. When counting partitions we'll now also check the devnum of the
device being non-null, which we already do when matching up the devices
in the second iteration. This should make things more robust, and
prevent other kinds of miscounting, which after all was the main
issue #8609 fixed.
If an rfkill device disappears between the time we get notified about
the existance and we fully opened it we might get ENXIO or ENODEV (i.e.
the two kinds of "device not found" errors, which are typically
generated when for example a device node has no actual backing device
behind it). let's handle that the same way as ENOENT, and downgrade the
log message to LOG_DEBUG.
tmpfiles: ignore "operational" errors during setup
We still get the errors logged, but we don't fail the service. This
is better for users because rerunning tmpfiles-setup.service a second
time is dangerous (c.f. cd9f5b68ce08375eb1d68a4ddaa7a24a5092d7ba).
Note that this only touches sd-tmpfiles-setup.service and
sd-tmpfiles-setup-dev.service. sd-tmpfiles-clean.service is as before.
tmpfiles: add a new return code for "operational failure" when processing
Things can fail, and we have no control over it:
- file system issues (immutable bits, file system errors, MAC refusals, etc)
- kernel refusing certain arguments when writing to /proc/sys or /sys
Let's add a new code for the case where we parsed configuration but failed
to execute it because of external errors.
units: use `systemctl exit` to kill the user manager (#8648)
Use `systemctl --user --force exit` to implement the systemd-exit
user service.
This removes our dependence on an external `kill` binary and the
concerns about whether they recognize SIGRTMIN+n by name or what their
interpretation of SIGRTMIN is.
Tested: `systemctl --user start systemd-exit.service` kills the
`systemd --user` instance for my user.
oss-fuzz: Fallback to `ninja-build` when available (#8641)
The ninja binary is deployed as `ninja-build` in older distros such as
RHEL 7/CentOS 7. Detect that and use `ninja-build` instead of `ninja`
when it's available.
dissect: Don't count RPMB and boot partitions (#8609)
Filter-out RPMB partitions and boot partitions from MMC devices when
counting partitions enumerated by the kernel. Also factor out the now
duplicated code into a separate function.
This complement the previous fixes to the problem reported in
https://github.com/systemd/systemd/issues/5806
Accept definitions to other AF_ constants, not just PF_ ones,
such as:
#define AF_LINUX AF_LOCAL
It may not be necessary to impose any restriction on the
definitions of the macros extracted, but for now
keep most of that requirement but match AF_* as well.
projects / thirdparty / systemd.git / log
