Daniel Mack [Fri, 7 Mar 2014 16:29:01 +0000 (17:29 +0100)]
sd-bus: add support for policy upload on activator connections
Activator connections may upload policy when registering to the bus.
This patch contains code to translate between BusNamePolicy objects and
the kdbus specific items.
Daniel Mack [Fri, 7 Mar 2014 16:14:52 +0000 (17:14 +0100)]
sd-bus: sync kdbus.h (ABI break)
Bring kdbus.h in sync with recent policy related changes. Most notably,
KDBUS_MAKE_POLICY_OPEN is now gone. The rest doesn't matter for systemd
at this point, as the policy functionality was not yet used.
* 'see': The user/group/world is allowed to see a name on the bus
* 'talk': The user/group/world is allowed to talk to a name
* 'own': The user/group/world is allowed to own a name
Daniel Mack [Fri, 7 Mar 2014 13:43:59 +0000 (14:43 +0100)]
manager: flush memory stream before using the buffer
When the manager receives a SIGUSR2 signal, it opens a memory stream
with open_memstream(), uses the returned file handle for logging, and
dumps the logged content with log_dump().
However, the char* buffer is only safe to use after the file handle has
been flushed with fflush, as the man pages states:
When the stream is closed (fclose(3)) or flushed (fflush(3)), the
locations pointed to by ptr and sizeloc are updated to contain,
respectively, a pointer to the buffer and the current size of the
buffer.
These values remain valid only as long as the caller performs no
further output on the stream. If further output is performed, then the
stream must again be flushed before trying to access these variables.
Without that call, dump remains NULL and the daemon crashes in
log_dump().
Daniel Mack [Fri, 7 Mar 2014 10:41:18 +0000 (11:41 +0100)]
core/busname: add lookup string for BUSNAME_FAILURE_SERVICE_FAILED_PERMANENT
When a busname unit enters BUSNAME_FAILURE_SERVICE_FAILED_PERMANENT, the
serialization will not be able to look up the result as string via
busname_result_to_string(). This leads to an assertion trap during
daemon-reexec.
Armin K [Tue, 4 Mar 2014 15:23:41 +0000 (16:23 +0100)]
build-sys: Don't distribute generated udev rule
It contains hardcoded path to systemd-sysctl executable which
is /usr/lib/systemd/systemd-sysctl on latest stable release and
as such it will complain at runtime if rootprefix != prefix
gpt-auto-generator: automatically find the root disk of the system
When run in an initrd and no root= argument is set (or is set to
root=gpt-auto) we will automatically look for the root partition on the
same disk the EFI ESP is located on.
Since we look for swap, /home and /srv on the disk the root partition is
located on, we hence have a fully discoverable chain:
Firmware discovers the EFI ESP partition → the initrd discovers the
root partition → the host OS discovers swap, /home, and /srv.
Note that this requires an EFI boot loader that sets the
LoaderDevicePartUUID EFI variable, such as Gummiboot.
Susant Sahani [Wed, 5 Mar 2014 15:13:25 +0000 (20:43 +0530)]
sd-rtnl:introduce table-based lookup and typesafe read() functions
This patch introduces new netlink attribute parsing logic
which is table based lookup and sd_rtnl_message_read_*
methods for reading attributes. By doing this user does not
have to loop for the attribute values . Only providing the
attribute type it gets the attribute values which is optimized
and sd_rtnl_message_read_* methods are simplified.
Tom Gundersen [Wed, 5 Mar 2014 10:53:26 +0000 (11:53 +0100)]
networkd: listen to changes to the MAC address
Bridges will change their MAC address when other devices are enslaved. We need
the correct MAC address to acquire a DHCP lease, so take note of it whenever
it changes.
If we encounter an inconsistency in a file, let's just
ignore it. Otherwise, after previous patch, we would try,
and fail, to use this file in every invocation of sd_journal_next
or sd_journal_previous that happens afterwards.
journal: assume that next entry is after previous entry
With a corrupted file, we can get in a situation where two entries
in the entry array point to the same object. Then journal_file_next_entry
will find the first one using generic_arrray_bisect, and try to move to
the second one, but since the address is the same, generic_array_get will
return the first one. journal_file_next_entry ends up in an infinite loop.
systemd-run: add new --property= switch that can set arbitrary properties for the unit that is created
The code for parsing these properties is shared with "systemctl
set-property", which means all the resource control settings are
immediately available.
As it appears "ln -s --relative" in conjunction with "-f" is broken,
let's work around that by explicitly remove the destination of the
symlink before we create it.
Implements IPv4LL with respect to RFC 3927
(http://tools.ietf.org/rfc/rfc3927.txt) and integrates it
with networkd. Majority of the IPv4LL state machine is
taken from avahi (http://avahi.org/) project's autoip.
IPv4LL can be enabled by IPv4LL=yes under [Network]
section of .network file.
IPv4LL works independent of DHCP but if DHCP lease is
aquired, then LL address will be dropped.
[tomegun: removed a trailing newline and a compiler warning]
conf-parser: drop special casing in config_parse_path()
The code checked for two lvalues that aren't even using
config_parse_path(), so let's drop these checks and make the function
completely generic again.
logind: ignore lid switch events for 30s after each suspend and 3min after startup
This is needed to give USB docking stations and suchlike time to settle,
so that a display connected to an USB docking station can actually act
as a lid swith inhibitor correctly.
With this change we should have somewhat reliable docking station
support in place.
logind: ignore lid switch if more than 1 display is connected
Previously we expected the desktop environment to take an inhibitor
lock, but this opened a race on boot-up where logind might already be
running but no DE is active.
Hence, let's move checking for additional displays into logind. This
also opens up this logic for other DEs, given that only GNOME
implemented the inhibitor logic so far.