Let's add a way to get the type-specific D-Bus interface of a unit from
either its type or name to src/basic/unit-name.[ch]. That way we can
share it with the client side, where it is useful in tools like cgls or
machinectl.
Also ports over machinectl to make use of this.
It seems that systemd still uses legacy -n option. The option has been
originally designed to avoid write to /etc/mtab during boot when root
FS is not ready or read-only.
This is not necessary for long time, because /etc/mtab is not a real
file (it's symlink) and write to the file is impossible. All utils
should be able to detect the symlink and ignore mtab. This concept is
supported for very long time before systemd.
The userspase mount options are currently maintained by libmount
(mount(8) and mount.nfs) in /run/mount) which is tmpfs initialized
during early boot.
Return 1 from *_reload() methods to signify "we did something", just
like in *_start(). This causes "Reloading foo..." messages to be logged.
"Reloaded foo." messages are already logged.
Skip /proc/mountinfo entries for which libmount returns a NULL pointer
for 'source' or 'target'. This happened on Semaphore CI's build servers
when the test suite is run.
It's primarily just a property of the Manager object after all, and we
try to refer to PID 1 as "manager" instead of "systemd", hence let's to
stick to this here too.
The call is only used by the mount and automount unit types, but that's
already enough to consider it generic unit functionality, hence move it
out of mount.c and into unit.c.
This changes log_unit_info() (and friends) to take a real Unit* object
insted of just a unit name as parameter. The call will now prefix all
logged messages with the unit name, thus allowing the unit name to be
dropped from the various passed romat strings, simplifying invocations
drastically, and unifying log output across messages. Also, UNIT= vs.
USER_UNIT= is now derived from the Manager object attached to the Unit
object, instead of getpid(). This has the benefit of correcting the
field for --test runs.
Also contains a couple of other logging improvements:
- Drops a couple of strerror() invocations in favour of using %m.
- Not only .mount units now warn if a symlinks exist for the mount
point already, .automount units do that too, now.
- A few invocations of log_struct() that didn't actually pass any
additional structured data have been replaced by simpler invocations
of log_unit_info() and friends.
- For structured data a new LOG_UNIT_MESSAGE() macro has been added,
that works like LOG_MESSAGE() but prefixes the message with the unit
name. Similar, there's now LOG_LINK_MESSAGE() and
LOG_NETDEV_MESSAGE().
- For structured data new LOG_UNIT_ID(), LOG_LINK_INTERFACE(),
LOG_NETDEV_INTERFACE() macros have been added that generate the
necessary per object fields. The old log_unit_struct() call has been
removed in favour of these new macros used in raw log_struct()
invocations. In addition to removing one more function call this
allows generated structured log messages that contain two object
fields, as necessary for example for network interfaces that are
joined into another network interface, and whose messages shall be
indexed by both.
- The LOG_ERRNO() macro has been removed, in favour of
log_struct_errno(). The latter has the benefit of ensuring that %m in
format strings is properly resolved to the specified error number.
- A number of logging messages have been converted to use
log_unit_info() instead of log_info()
- The client code in sysv-generator no longer #includes core code from
src/core/.
- log_unit_full_errno() has been removed, log_unit_full() instead takes
an errno now, too.
- log_unit_info(), log_link_info(), log_netdev_info() and friends, now
avoid double evaluation of their parameters
A variety of changes:
- Make sure all our calls distuingish OOM from other errors if OOM is
not the only error possible.
- Be much stricter when parsing escaped paths, do not accept trailing or
leading escaped slashes.
- Change unit validation to take a bit mask for allowing plain names,
instance names or template names or an combination thereof.
- Refuse manipulating invalid unit name
Change cunescape() to return a normal error code, so that we can
distuingish OOM errors from parse errors.
This also adds a flags parameter to control whether "relaxed" or normal
parsing shall be done. If set no parse failures are generated, and the
only reason why cunescape() can fail is OOM.
If you have for example ext4 on iscsi devices it is possible to setup
qoutas there. Unfortunately, because such fstab entry contains _netdev,
systemd will not add dependency to quotaon.service.
Because the order of coldplugging is not defined, we can reference a
not-yet-coldplugged unit and read its state while it has not yet been
set to a meaningful value.
This way, already active units may get started again.
We fix this by deferring such actions until all units have been at
least somehow coldplugged.
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=88401
This change introduces a new state "tentative" for device units. Device
units are considered "plugged" when udev announced them, "dead" when
they are not available in the kernel, and "tentative" when they are
referenced in /proc/self/mountinfo or /proc/swaps but not (yet)
announced via udev.
This should fix a race when device nodes (like loop devices) are created
and immediately mounted. Previously, systemd might end up seeing the
mount unit before the device, and would thus pull down the mount because
its BindTo dependency on the device would not be fulfilled.
This patch removes includes that are not used. The removals were found with
include-what-you-use which checks if any of the symbols from a header is
in use.
After all it is now much more like strjoin() than strappend(). At the
same time, add support for NULL sentinels, even if they are normally not
necessary.
Add unit dependencies for dynamic (i. e. not from fstab) mounts. With that,
mount units properly bind to their underlying device, and thus get
automatically stopped/unmounted when the underlying device goes away.
This cleans up stale mounts from unplugged devices.
Thanks to Lennart Poettering for pointing out the fix!
Unit _start() and _stop() implementations can fail with -EAGAIN to delay
execution temporarily. Thus, we should not output status messages before
invoking these calls, but after, and only when we know that the
invocation actually made a change.
After all, mounts below these directories are pretty much guaranteed to
be virtual, and it's hence unnecessary to unmount them during shutdown.
Moreover, in less-priviliged containers we might lack the rights to
unmount them, hence don't even try.
http://lists.freedesktop.org/archives/systemd-devel/2015-January/027113.html
deb6120920 'man: there's actually no "fail" fstab option, but only
"nofail" removed it from our documentation, which I missed.
fstab(5) only mentions "auto", "noauto", and "nofail". Stick to
those three.
strempty() will return an empty string in case the input parameter is
a NULL pointer. The correct test to check for an empty string is
isempty(), so use that instead.
This fixes a regression from commit 17a1c59 ("core/mount: filter out
noauto,auto,nofail,fail options").
We passed the full option string from fstab to /bin/mount. It would in
turn pass the full option string to its helper, if it needed to invoke
one. Some helpers would ignore things like "nofail", but others would
be confused. We could try to get all helpers to ignore those
"meta-options", but it seems better to simply filter them out.
In our model, /bin/mount simply has no business in knowing whether the
mount was configured as fail or nofail, auto or noauto, in the
fstab. If systemd tells invokes a command to mount something, and it
fails, it should always return an error. It seems cleaner to filter
out the option, since then there's no doubt how the command should
behave.
https://bugzilla.redhat.com/show_bug.cgi?id=1177823
This fixes parsing of options in shared/generator.c. Existing code
had some issues:
- it would treate whitespace and semicolons as seperators. fstab(5)
is pretty clear that only commas matter. And the syntax does
not allow for spaces to be inserted in the field in fstab.
Whitespace might be escaped, but then it should not seperate
options. Treat whitespace and semicolons as any other character.
- it assumed that x-systemd.device-timeout would always be followed
by "=". But this is not guaranteed, hasmntopt will return this
option even if there's no value. Uninitialized memory could be read.
- some error paths would log, and inconsistently, some would just
return an error code.
Filtering is split out to a separate function and tests are added.
Similar code paths in other places are adjusted to use the new function.
-n is only allowed for root. /etc/mtab is nowadays almost always a link to /proc/,
so in practice this does not really matter too much, but should allow .mount units
to work in --user mode.
https://bugs.freedesktop.org/show_bug.cgi?id=87602
Containers do not really support .device, .automount or .swap units;
Systems compiled without support for swap do not support .swap units;
Systems without kdbus do not support .busname units.
With this change attempts to start a unsupported unit types will result
in an immediate "unsupported" job result, which is a lot more
descriptive then before. Also, attempts to start device units in
containers will now immediately fail instead of causing jobs to be
enqueued that never go away.
When creating a new mount unit after an event on /proc/self/mountinfo,
check the mount options as well as the fstype to determine if this is a
remote mount that requires network access.
