Announcing D-Bus 1.11.10
Simon McVittie
simon.mcvittie at collabora.co.uk
Thu Feb 16 22:28:34 UTC 2017
The “purple hair gives you telekinesis?” release.
1.11.x is a development branch for the adventurous, which will
eventually lead to a 1.12.x stable branch.
http://dbus.freedesktop.org/releases/dbus/dbus-1.11.10.tar.gz
http://dbus.freedesktop.org/releases/dbus/dbus-1.11.10.tar.gz.asc
git tag: dbus-1.11.10
git branch: master
Dependencies:
• AppArmor support requires at least libapparmor 2.8.95, reduced
from 2.10 in previous versions. One test requires 2.10 and is
skipped if building with an older version.
Enhancements:
• Do the Travis-CI build in Docker containers for Ubuntu LTS, Debian
stable and Debian testing in addition to the older Ubuntu that is
the default (fd.o #98889, Simon McVittie)
• Avoid some deprecated CMake functions (fd.o #99586, Ralf Habacker)
• Silence many -Wswitch-enum and -Wswitch-default warnings
(fd.o #98191; Thomas Zimmermann, Simon McVittie)
• Install a sysusers.d snippet so `dbus-daemon --system` can be used
with an unpopulated /etc (fd.o #99162, Lennart Poettering)
• Install pkg-config metadata on Unix even if building with CMake
(fd.o #99752, Ralf Habacker)
• Exclude auth mechanisms from REJECTED message if they are supported
in the code but but configured to be disallowed (fd.o #99621,
Ralf Habacker)
Fixes:
• Prevent symlink attacks in the nonce-tcp transport on Unix that could
allow an attacker to overwrite a file named "nonce", in a directory
that the user running dbus-daemon can write, with a random value
known only to the user running dbus-daemon. This is unlikely to be
exploitable in practice, particularly since the nonce-tcp transport
is really only useful on Windows.
On Unix systems we strongly recommend using only the unix: and systemd:
transports, together with EXTERNAL authentication. These are the only
transports and authentication mechanisms enabled by default.
(fd.o #99828, Simon McVittie)
• Avoid symlink attacks in the "embedded tests", which are not enabled
by default and should never be enabled in production builds of dbus.
(fd.o #99828, Simon McVittie)
• Fix the implementation of re-enabling a timeout so that its
countdown is restarted as intended, instead of continually
decreasing. (fd.o #95619; Michal Koutný, Simon McVittie)
• When receiving a message with file descriptors, do not start reading
the beginning of the next message, so that only one such message
is processed at a time. In conjunction with the fix for #95619
this means that processes sending many file descriptors, such as
systemd-logind on a system that receives very rapid ssh connections,
are not treated as abusive and kicked off the bus. Revert the previous
workaround that special-cased uid 0.
(fd.o #95263, LP#1591411; Simon McVittie)
• Do not require TMPDIR, TEMP or TMP to be set when cross-compiling
for Windows with CMake (fd.o #99586, Ralf Habacker)
• Do not set Unix-specific variables when targeting Windows
(fd.o #99586, Ralf Habacker)
• Install Unix executables to ${CMAKE_INSTALL_PREFIX}/bin as intended,
not ${CMAKE_INSTALL_PREFIX}/lib (fd.o #99752, Ralf Habacker)
• Use relative install locations in CMake on Unix to respect DESTDIR,
and use GNU-style install layout (fd.o #99721, #99752; Ralf Habacker)
• Install dbus-arch-deps.h correctly when using CMake
(fd.o #99586, #99721; Ralf Habacker)
• Improve argument validation for `dbus-test-tool spam`
(ffd.o #99693, Coverity #54759; Philip Withnall)
• Don't shift by a negative integer if a hash table becomes monstrously
large (fd.o #99641, Coverity #54682; Philip Withnall)
• Don't leak LSM label if dbus-daemon runs out of memory when dealing with
a new connection (fd.o #99612, Coverity #141058; Philip Withnall)
• Remove an unnecessary NULL check
(fd.o #99642, Coverity #141062; Philip Withnall)
• Improve error handling in unit tests and dbus-send
(fd.o #99643, #99694, #99712, #99722, #99723, #99724, #99758,
#99759, #99793, Coverity #54688, #54692, #54693, #54697, #54701,
#54710, #54711, #54714, #54715, #54718, #54721, #54724, #54726,
#54730, #54740, #54822, #54823, #54824, #54825; Philip Withnall)
• Do not print verbose messages' timestamps to stderr if the actual message
has been redirected to the Windows debug port (fd.o #99749, Ralf Habacker)
--
Simon McVittie, Collabora Ltd.
More information about the dbus
mailing list