Let's remove a number of synchronization points from our service
startups: let's drop synchronous match installation, and let's opt for
asynchronous instead.
Also, let's use sd_bus_match_signal() instead of sd_bus_add_match()
where we can.
Let's always leave logging to the call that actually added the fields to
the bus message. This way we don't get duplicate logging whenver
bus_append_unit_property_assignment() ends up being called, which does
all its logging on its own (and probably should do, as it can output
much more precise errors).
This option is likely to be very useful for systemd-run invocations,
hence let's add a shortcut for it.
With this new concepts it's now very easy to put together systemd-run
invocations that leave zero artifacts in the system, including when they
fail.
In this mode, we'll directly connect stdin/stdout/stderr of the invoked
service with whatever systemd-run itself is invoked on. This allows
inclusion of "systemd-run" commands in shell pipelines, as unlike
"--pty" this means EOF of stdin/stdout/stderr are propagated
independently.
If --pty and --pipe are combined systemd-run will automatically pick the
right choice for the context it is invoked in, i.e. --pty when invoked
on a TTY, and --pipe otherwise.
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.
Since 95f1d6bfec we'll subscribe to unit
signals to figure out when to disconnect the pty. But that can only work
correctly if we actually subscribe to the unit's signals. Hence,
explicitly pin (and thus subscribe to) the unit we just created not only
in --wait mode but also in --pty mode.
Or to say this differently: we need to pin the unit in the same cases as
we install the signal match. 95f1d6bfec
forgot to do that.
This is relevant to make sure systemd-run works correctly in --user
mode, and correctly exits when the spawned service dies. To test:
systemd-run --user -t /bin/bash
And then press ^D. This will hang before this change, but exit cleanly
after it.
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.
This reworks systemd-run so that in --pty mode we watch the unit state
the way we do it in --wait mode. Whenever we notice that the service is
in failed or inactive state finish right-away, but first write all
unwritten characters we can read from the master TTY device.
This makes sure that when the TTY service fails before it opens the
slave PTY device we properly notice that and exit early, so that borked
start parameters result in immediate systemd-run failure. Previously,
we'd not notice this at all, as a PTY slave that never was opened won't
result in POLLHUP events, and we'd hence simply keep reading from it
forever.
In essence, --pty now enables the same unit watching logic that --wait
enables. However, unless --wait is specified we won#t show the final
summary, hence the effective difference should be pretty minimal.
Fixes: #3915
If the PTY forwarder is still around our TTY will have borked settings,
regarding newlines, hence explicitly close it before showing the
summary, so that it looks pretty.
D-Bus is inherently racy when a function returns an object path for a
newly allocated object the client shall watch: as the object already
exists before the client can subscribe to it, it might lose messages
from it.
Let's fix this, by explicitly querying unit properties right after
subscribing to its property changes.
Fixes: #4920
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.
This adds a new call get_user_creds_clean(), which is just like
get_user_creds() but returns NULL in the home/shell parameters if they contain
no useful information. This code previously lived in execute.c, but by
generalizing this we can reuse it in run.c.
This adds parse_nice() that parses a nice level and ensures it is in the right
range, via a new nice_is_valid() helper. It then ports over a number of users
to this.
No functional changes.
We'd call sd_bus_message_unref and then proceed to use
variables pointing into the reply buffer (fd and char*).
dup the fd and copy the string before destorying the reply.
This makes systemd-run run again for me.
https://bugzilla.redhat.com/show_bug.cgi?id=1337636
Previously we'd have generally useful sd-bus utilities in bust-util.h,
intermixed with code that is specifically for writing clients for PID 1,
wrapping job and unit handling. Let's split the latter out and move it into
bus-unit-util.c, to make the sources a bit short and easier to grok.
In 110ceee58e we removed the period after
printing the started units. This makes copying the unit name easier but
results in improper English.
This adds a colon before printing the units, which makes the output
look better.
Make sure we can properly process resource limit properties. Specifically, allow transient configuration of both the
soft and hard limit, the same way from the unit files. Previously, only the the hard rlimits could be configured but
they'd implicitly spill into the soft hard rlimits.
This also updates the client-side code to be able to parse hard/soft resource limit specifications. Since we need to
serialize two properties in bus_append_unit_property_assignment() now, the marshalling of the container around it is
now moved into the function itself. This has the benefit of shortening the calling code.
As a side effect this now beefs up the rlimit parser of "systemctl set-property" to understand time and disk sizes
where that's appropriate.
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.
Previously we used the process ID to generate transient unit names.
However, that is problematic as PIDs get reused easily, and applying
them to remote systems makes little sense.
Fortunately, each bus peer gets a unique, non-reusable ID assigned when
attaching to a bus, hence let's use that, if we can. In some cases we
cannot however, because we connect directly to PID's private socket, and
thus are not a proper bus peer with a unique ID. In that case generate a
random UUID to name the unit after.
Modernize the code a bit:
- Get rid of FOREACH_WORD_SEPARATOR() loop in favour of a
extract_first_word() loop.
- Remove find_binary()'s "local" flag. It's not reasonably possible to
look for binaries on remote systems, we hence should not pretend we
could.
- When we cannot find a suitable binary, return the last error returned
from access() rather than ENOENT unconditionally.
- Rework fsck_exists() and mkfs_exists() to return 1 on success, 0 if
the implementation is missing and negative on real errors. This is
more like we do it in other functions.
- Make sure we also detect direct fsck symlinks to "true", rather than
just absolute ones to /bin/true.
Previously, we'd allocate the TTY, spawn a service on it, but
immediately start processing the TTY and forwarding it to whatever the
commnd was started on. This is however problematic, as the TTY might get
actually opened only much later by the service. We'll hence first get
EIOs on the master as the other side is still closed, and hence
considered it hung up and terminated the session.
With this change we add a flag to the pty forwarding logic:
PTY_FORWARD_IGNORE_INITIAL_VHANGUP. If set, we'll ignore all hangups
(i.e. EIOs) on the master PTY until the first byte is successfully read.
From that point on we consider a hangup/EIO a regular connection termination. This
way, we handle the race: when we get EIO initially we'll ignore it,
until the connection is properly set up, at which time we start
honouring it.
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.
sd_bus_flush_close_unref() is a call that simply combines sd_bus_flush()
(which writes all unwritten messages out) + sd_bus_close() (which
terminates the connection, releasing all unread messages) +
sd_bus_unref() (which frees the connection).
The combination of this call is used pretty frequently in systemd tools
right before exiting, and should also be relevant for most external
clients, and is hence useful to cover in a call of its own.
Previously the combination of the three calls was already done in the
_cleanup_bus_close_unref_ macro, but this was only available internally.
Also see #327
This ports a lot of manual code over to sigprocmask_many() and friends.
Also, we now consistly check for sigprocmask() failures with
assert_se(), since the call cannot realistically fail unless there's a
programming error.
Also encloses a few sd_event_add_signal() calls with (void) when we
ignore the return values for it knowingly.
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
Otherwise it might happen that by the time PID 1 adds our process to the
scope unit the process might already have died, if the process is
short-running (such as an invocation to /bin/true).
https://bugs.freedesktop.org/show_bug.cgi?id=86520
Previously we always invoked the container PID 1 on /dev/console of the
container. With this change we do so only if nspawn was invoked
interactively (i.e. its stdin/stdout was connected to a TTY). In all other
cases we directly pass through the fds unmodified.
This has the benefit that nspawn can be added into shell pipelines.
https://bugs.freedesktop.org/show_bug.cgi?id=87732
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.
Pretty much everywhere else we use the generic term "machine" when
referring to containers in API, so let's do though in sd-bus too. In
particular, since the concept of a "container" exists in sd-bus too, but
as part of the marshalling system.
Support timer options --on-active=, --on-boot=, --on-startup=,
--on-unit-active=, --on-unit-inactive=, --on-calendar=. Each options
corresponding with OnActiveSec=, OnBootSec=, OnStartupSec=,
OnUnitActiveSec=, OnUnitInactiveSec=, OnCalendar= of timer
respectively. And OnCalendar= and WakeSystem= supported by
--timer-property= option like --property= of systemd-run.
And if --unit= option and timer options are specified the command can
be omitted. In this case, systemd-run assumes the target service is
already loaded. And just try to generate transient timer unit only.
If the format string contains %m, clearly errno must have a meaningful
value, so we might as well use log_*_errno to have ERRNO= logged.
Using:
find . -name '*.[ch]' | xargs sed -r -i -e \
's/log_(debug|info|notice|warning|error|emergency)\((".*%m.*")/log_\1_errno(errno, \2/'
Plus some whitespace, linewrap, and indent adjustments.
Using:
find . -name '*.[ch]' | while read f; do perl -i.mmm -e \
'local $/;
local $_=<>;
s/(if\s*\([^\n]+\))\s*{\n(\s*)(log_[a-z_]*_errno\(\s*([->a-zA-Z_]+)\s*,[^;]+);\s*return\s+\g4;\s+}/\1\n\2return \3;/msg;
print;'
$f
done
And a couple of manual whitespace fixups.
As a followup to 086891e5c1 "log: add an "error" parameter to all
low-level logging calls and intrdouce log_error_errno() as log calls
that take error numbers", use sed to convert the simple cases to use
the new macros:
find . -name '*.[ch]' | xargs sed -r -i -e \
's/log_(debug|info|notice|warning|error|emergency)\("(.*)%s"(.*), strerror\(-([a-zA-Z_]+)\)\);/log_\1_errno(-\4, "\2%m"\3);/'
Multi-line log_*() invocations are not covered.
And we also should add log_unit_*_errno().
systemd-run would fail when run with -M or -H and an absolute path,
if this path did not exists locally. Allow it to continue, since we
don't have a nice way of checking if the binary exists remotely.
The case where -M or -H is used and a local path is unchanged, and we
still iterate over $PATH to find the binary. We need to convert to an
absolute path, and we don't have a nice mechanism to check remotely,
so we assume that the binary will be located in the same place locally
and remotely.
http://lists.freedesktop.org/archives/systemd-devel/2014-November/025418.html
Since b5eca3a205 we don't attempt to GC
busses anymore when unsent messages remain that keep their reference,
when they otherwise are not referenced anymore. This means that if we
explicitly want connections to go away, we need to close them.
With this change we will no do so explicitly wherver we connect to the
bus from a main program (and thus know when the bus connection should go
away), or when we create a private bus connection, that really should go
away after our use.
This fixes connection leaks in the NSS and PAM modules.
getopt is usually good at printing out a nice error message when
commandline options are invalid. It distinguishes between an unknown
option and a known option with a missing arg. It is better to let it
do its job and not use opterr=0 unless we actually want to suppress
messages. So remove opterr=0 in the few places where it wasn't really
useful.
When an error in options is encountered, we should not print a lengthy
help() and overwhelm the user, when we know precisely what is wrong
with the commandline. In addition, since help() prints to stdout, it
should not be used except when requested with -h or --help.
Also, simplify things here and there.
There is a small number of the places in sources where we don't check
asprintf() return code and assume that after error the function
returns NULL pointer via the first argument. That's wrong, after
error the content of pointer is undefined.
Both systemd-analyze and systemd-run only access org.freedesktop.systemd1
on the bus. This patch allows using systemd-run --user and systemd-analyze
--user even if the user session's bus is not properly integrated with the
systemd user unit.
https://bugs.freedesktop.org/show_bug.cgi?id=79252 and other reports...
The code for parsing these properties is shared with "systemctl
set-property", which means all the resource control settings are
immediately available.
Previously the returned object of constructor functions where sometimes
returned as last, sometimes as first and sometimes as second parameter.
Let's clean this up a bit. Here are the new rules:
1. The object the new object is derived from is put first, if there is any
2. The object we are creating will be returned in the next arguments
3. This is followed by any additional arguments
Rationale:
For functions that operate on an object we always put that object first.
Constructors should probably not be too different in this regard. Also,
if the additional parameters might want to use varargs which suggests to
put them last.
Note that this new scheme only applies to constructor functions, not to
all other functions. We do give a lot of freedom for those.
Note that this commit only changes the order of the new functions we
added, for old ones we accept the wrong order and leave it like that.
By specifiy a Controller property when creating the scope a client can
specify a bus name that will be notified with a RequestStop bus signal
when the scope has been asked to shut down, instead of sending SIGTERM
to the scope processes themselves.
https://bugzilla.redhat.com/show_bug.cgi?id=1032695