platform/upstream/dbus.git
9 years agoProvide appropriate DBUS_USER and DBUS_TEST_USER under CMake
Ralf Habacker [Wed, 4 Feb 2015 12:50:03 +0000 (13:50 +0100)]
Provide appropriate DBUS_USER and DBUS_TEST_USER under CMake

[separated out from a larger commit -smcv]
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88964
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agoLink tests to test-utils-glib.c under CMake too
Ralf Habacker [Wed, 4 Feb 2015 12:23:34 +0000 (13:23 +0100)]
Link tests to test-utils-glib.c under CMake too

[Separated out from a larger commit -smcv]
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88964
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agodbus-monitor: use common code from dbus-test-tool
Simon McVittie [Wed, 4 Feb 2015 14:18:39 +0000 (14:18 +0000)]
dbus-monitor: use common code from dbus-test-tool

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=46787
Reviewed-by: Philip Withnall
[also sync up the cmake build system -smcv]

9 years ago1.9.9
Simon McVittie [Tue, 3 Feb 2015 23:13:51 +0000 (23:13 +0000)]
1.9.9

9 years ago1.9.8 development release
Simon McVittie [Tue, 3 Feb 2015 17:26:47 +0000 (17:26 +0000)]
1.9.8 development release

9 years agoTreat root as a valid candidate for TEST_USER_ME
Simon McVittie [Tue, 3 Feb 2015 19:35:39 +0000 (19:35 +0000)]
Treat root as a valid candidate for TEST_USER_ME

If spawn_dbus_daemon() can fail for TEST_USER_ME, then we'd have to
go through all the tests adding the ability to skip tests after
it fails, which is a fairly extensive change.

The tests have historically all run as whatever uid is supplied, and
if the tests are being run as root for some reason - perhaps in a CI
framework for an embedded platform that doesn't have non-root users,
or in an environment where you can be root or non-root but not both -
there is no particular reason to skip them.

9 years agoNEWS
Simon McVittie [Tue, 3 Feb 2015 16:25:56 +0000 (16:25 +0000)]
NEWS

9 years agoAdd a regression test for systemd activation
Simon McVittie [Mon, 26 Jan 2015 20:10:39 +0000 (20:10 +0000)]
Add a regression test for systemd activation

4.5 years after it was implemented, here is the regression test.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=57952
Reviewed-by: Philip Withnall
9 years agobus_context_log_and_set_error: add and use
Simon McVittie [Mon, 2 Feb 2015 18:15:27 +0000 (18:15 +0000)]
bus_context_log_and_set_error: add and use

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88810
Reviewed-by: Philip Withnall
9 years ago_dbus_set_error_valist: add
Simon McVittie [Mon, 2 Feb 2015 18:15:02 +0000 (18:15 +0000)]
_dbus_set_error_valist: add

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88810
Reviewed-by: Philip Withnall
9 years agobus_context_log_literal: add simplified version of bus_context_log
Simon McVittie [Mon, 2 Feb 2015 17:57:30 +0000 (17:57 +0000)]
bus_context_log_literal: add simplified version of bus_context_log

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88810
Reviewed-by: Philip Withnall
9 years agobus: put the printf attribute in the header where it will do more good
Simon McVittie [Mon, 2 Feb 2015 17:29:43 +0000 (17:29 +0000)]
bus: put the printf attribute in the header where it will do more good

Now we can actually notice incorrect format strings in other
translation units.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88810
Reviewed-by: Philip Withnall
9 years agobus driver: factor out bus_driver_check_caller_is_privileged, and allow root
Simon McVittie [Mon, 26 Jan 2015 19:12:01 +0000 (19:12 +0000)]
bus driver: factor out bus_driver_check_caller_is_privileged, and allow root

Unlike the initial mitigation for CVE-2014-8148, we now allow
uid 0 to call UpdateActivationEnvironment. There's no point in root
doing that, but there's also no reason why it's particularly bad -
if an attacker is uid 0 we've already lost - and it simplifies
use of this function for future things that do want to be callable
by root, like BecomeMonitor for #46787.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88810
Reviewed-by: Philip Withnall
9 years agoAdd a test for uid-controlled permissions
Simon McVittie [Mon, 26 Jan 2015 19:10:11 +0000 (19:10 +0000)]
Add a test for uid-controlled permissions

This is technical debt from mitigating CVE-2014-8148, which should
really have had a regression test at the time.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88810
Reviewed-by: Philip Withnall
9 years agoAdd infrastructure to run bits of tests under an alternative uid
Simon McVittie [Mon, 26 Jan 2015 15:47:59 +0000 (15:47 +0000)]
Add infrastructure to run bits of tests under an alternative uid

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88810
Reviewed-by: Philip Withnall
9 years agotest: implement GLib-style "installed tests"
Simon McVittie [Mon, 26 Jan 2015 15:47:22 +0000 (15:47 +0000)]
test: implement GLib-style "installed tests"

We run each test twice:

* once with the system's session.conf, as an integration test
  (test-cases that need a special configuration are automatically
  skipped)
* once with our special test configuration files, which provide better
  coverage

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88810
Reviewed-by: Philip Withnall
9 years agoGenerate test configuration files via build-time sed, not configure
Simon McVittie [Mon, 26 Jan 2015 18:55:08 +0000 (18:55 +0000)]
Generate test configuration files via build-time sed, not configure

This means we can generate a version that works when installed,
from the same source files.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88810
Reviewed-by: Philip Withnall
9 years agoFactor out some utility functions from test/dbus-daemon*
Simon McVittie [Mon, 26 Jan 2015 19:02:59 +0000 (19:02 +0000)]
Factor out some utility functions from test/dbus-daemon*

In the process, make test_kill_pid() safer: do not try to terminate
more than one pid, or the NULL handle.

Also stop leaking the address_fd in spawn_dbus_daemon, a pre-existing
bug that was spotted by Philip Withnall during review.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88810
Reviewed-by: Philip Withnall
9 years agoBump required GLib version to 2.36
Simon McVittie [Mon, 2 Feb 2015 16:04:52 +0000 (16:04 +0000)]
Bump required GLib version to 2.36

This is for g_close(), which the next commit will use. It also lets us
rely on g_type_init() being a no-op (since 2.32 the type system is
always initialized by a global constructor).

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88810
Reviewed-by: Philip Withnall
9 years agoUpdate .gitignore files
Lukasz Skalski [Tue, 3 Feb 2015 12:59:00 +0000 (13:59 +0100)]
Update .gitignore files

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88943
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agoNEWS
Simon McVittie [Mon, 2 Feb 2015 18:52:08 +0000 (18:52 +0000)]
NEWS

9 years agolcov: use builddir, not srcdir
Simon McVittie [Fri, 23 Jan 2015 19:08:32 +0000 (19:08 +0000)]
lcov: use builddir, not srcdir

It seems lcov (or gcc?) has changed its paths since last time this
worked.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88808
Reviewed-by: Philip Withnall
9 years agoEnable subdir-objects Automake option
Simon McVittie [Mon, 26 Jan 2015 13:49:12 +0000 (13:49 +0000)]
Enable subdir-objects Automake option

It has been supported since at least 1.10, and its absence is
deprecated since 1.14.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88808
Reviewed-by: Philip Withnall
9 years agobus: exit on fatal errors even if not syslogging
Simon McVittie [Thu, 22 Jan 2015 15:46:27 +0000 (15:46 +0000)]
bus: exit on fatal errors even if not syslogging

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88808
Reviewed-by: Philip Withnall
9 years agoFix 'dbus-daemon on Windows: Failed to bind socket "localhost:xxxx": Address already...
Ralf Habacker [Fri, 30 Jan 2015 19:19:01 +0000 (20:19 +0100)]
Fix 'dbus-daemon on Windows: Failed to bind socket "localhost:xxxx": Address already in use'.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=87999
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agoIn _dbus_listen_tcp_socket() set socket family to ipv4 in unspecified case for now.
Ralf Habacker [Tue, 27 Jan 2015 22:42:25 +0000 (23:42 +0100)]
In _dbus_listen_tcp_socket() set socket family to ipv4 in unspecified case for now.

This fix is required to support at least ipv4 tcp support calling
_dbus_listen_tcp_socket() without family.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=87999
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agoAdd manual tcp test case.
Ralf Habacker [Wed, 28 Jan 2015 21:44:16 +0000 (22:44 +0100)]
Add manual tcp test case.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=87999
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years ago_dbus_string_append_printf_valist: avoid leaking copy of varargs
Simon McVittie [Tue, 6 Jan 2015 22:12:16 +0000 (22:12 +0000)]
_dbus_string_append_printf_valist: avoid leaking copy of varargs

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88087
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agoUse pygi instead of pygobject 2
Simon McVittie [Thu, 6 Nov 2014 14:24:14 +0000 (14:24 +0000)]
Use pygi instead of pygobject 2

pygobject 2 is obsolete and unmaintained, and anyway this is for
optional functionality (full regression test coverage) rather than
anything that will be needed in production builds.

Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=85969

9 years agoFix build with --enable-maintainer-mode --disable-embedded-tests
Thiago Macieira [Mon, 29 Dec 2014 18:51:28 +0000 (16:51 -0200)]
Fix build with --enable-maintainer-mode --disable-embedded-tests

Otherwise, we get:
dbus/dbus-object-tree.c:323:1: error: 'find_subtree' defined but not used [-Werror=unused-function]

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=87837
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agoWhen generating random messages, exercise empty signature as intended
Simon McVittie [Tue, 6 Jan 2015 22:15:16 +0000 (22:15 +0000)]
When generating random messages, exercise empty signature as intended

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88089

9 years ago_dbus_marshal_validate_test: test both /foo/ and foo/
Simon McVittie [Tue, 6 Jan 2015 22:14:08 +0000 (22:14 +0000)]
_dbus_marshal_validate_test: test both /foo/ and foo/

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88089

9 years agoselinux: avoid leaking on systems where va_start allocates memory
Simon McVittie [Tue, 6 Jan 2015 22:11:48 +0000 (22:11 +0000)]
selinux: avoid leaking on systems where va_start allocates memory

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88087

9 years agoFix assertion in regression test to be equality test, not assignment
Simon McVittie [Tue, 6 Jan 2015 22:08:47 +0000 (22:08 +0000)]
Fix assertion in regression test to be equality test, not assignment

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88088

9 years agoMerge branch 'dbus-1.8'
Simon McVittie [Mon, 5 Jan 2015 15:00:45 +0000 (15:00 +0000)]
Merge branch 'dbus-1.8'

Conflicts:
NEWS
configure.ac

9 years agoMerge tag 'dbus-1.9.6'
Simon McVittie [Mon, 5 Jan 2015 14:58:48 +0000 (14:58 +0000)]
Merge tag 'dbus-1.9.6'

dbus-1.9.6

Conflicts:
NEWS
configure.ac

9 years ago1.8.15
Simon McVittie [Mon, 5 Jan 2015 14:57:26 +0000 (14:57 +0000)]
1.8.15

9 years agoMerge tag 'dbus-1.8.14' into dbus-1.8
Simon McVittie [Mon, 5 Jan 2015 14:56:44 +0000 (14:56 +0000)]
Merge tag 'dbus-1.8.14' into dbus-1.8

dbus-1.8.14

9 years agoFix usage of undefined variable DBUS_WIN32 on cmake build system
Ralf Habacker [Sun, 4 Jan 2015 07:12:06 +0000 (08:12 +0100)]
Fix usage of undefined variable DBUS_WIN32 on cmake build system

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88010
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agoMerge branch 'dbus-1.8'
Ralf Habacker [Mon, 5 Jan 2015 10:49:13 +0000 (11:49 +0100)]
Merge branch 'dbus-1.8'

Conflicts:
NEWS
configure.ac

9 years agoWindows cmake cross compile fix
Ralf Habacker [Sat, 3 Jan 2015 19:36:31 +0000 (20:36 +0100)]
Windows cmake cross compile fix

We need to include 'test' subdir in any case not only when using glib.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=88009
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agoFix of 'dbus-daemon can only handle 64 simultaneous connections on Windows'.
Ralf Habacker [Mon, 27 Jan 2014 10:58:37 +0000 (11:58 +0100)]
Fix of 'dbus-daemon can only handle 64 simultaneous connections on Windows'.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=71297
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
9 years agoMerge branch 'dbus-1.8' and prepare 1.9.6
Simon McVittie [Thu, 1 Jan 2015 23:48:13 +0000 (23:48 +0000)]
Merge branch 'dbus-1.8' and prepare 1.9.6

Conflicts:
NEWS
configure.ac
test/dbus-daemon.c

9 years agoPrepare release for Monday
Simon McVittie [Thu, 1 Jan 2015 23:42:41 +0000 (23:42 +0000)]
Prepare release for Monday

9 years agoHardening: only accept Stats function calls at the canonical object path
Simon McVittie [Fri, 19 Dec 2014 18:51:04 +0000 (18:51 +0000)]
Hardening: only accept Stats function calls at the canonical object path

These function calls are not a privilege escalation risk like
UpdateActivationEnvironment, but they might provide sensitive
information or be enhanced to provide sensitive information
in future, so the default system.conf locks them down to root-only.
Apply the same canonical-object-path hardening as for
UpdateActivationEnvironment.

We do not apply the uid check here because they are less dangerous
than UpdateActivationEnvironment, and because the ability to unlock
these function calls for specific uids is a documented configuration
for developers.

Reviewed-by: Thiago Macieira <thiago@kde.org>
[added missing #include; extended commit message -smcv]

9 years agoAdd a regression test for path-based UpdateActivationEnvironment hardening
Simon McVittie [Fri, 19 Dec 2014 19:17:14 +0000 (19:17 +0000)]
Add a regression test for path-based UpdateActivationEnvironment hardening

Reviewed-by: Thiago Macieira <thiago@kde.org>
9 years agoHardening: only allow the uid of the dbus-daemon to call UpdateActivationEnvironment
Simon McVittie [Fri, 19 Dec 2014 19:19:00 +0000 (19:19 +0000)]
Hardening: only allow the uid of the dbus-daemon to call UpdateActivationEnvironment

As with the previous commit, this is probably not actually privilege
escalation due to the use of an activation helper that cleans up its
environment, but let's be extra-careful here.

Reviewed-by: Thiago Macieira <thiago@kde.org>
[adjusted commit message -smcv]

9 years agoHardening: reject UpdateActivationEnvironment on non-canonical path
Simon McVittie [Fri, 19 Dec 2014 18:49:33 +0000 (18:49 +0000)]
Hardening: reject UpdateActivationEnvironment on non-canonical path

UpdateActivationEnvironment is the one dbus-daemon API call that is
obviously dangerous (it is intended for the session bus),
so the default system.conf does not allow anyone to call it.

It has recently come to the D-Bus maintainers' attention that some
system services incorrectly install D-Bus policy rules that allow
arbitrary method calls to any destination as long as they have a
"safe" object path. This is not actually safe: some system services
that use low-level D-Bus bindings like libdbus, including dbus-daemon
itself, provide the same API on all object paths.

Unauthorized calls to UpdateActivationEnvironment are probably just
resource consumption rather than privilege escalation, because on
the system bus, the modified environment is only used to execute
a setuid wrapper that avoids LD_PRELOAD etc. via normal setuid
handling, and sanitizes its own environment before executing
the real service. However, it's safest to assume the worst and
treat it as a potential privilege escalation.

Accordingly, as a hardening measure to avoid privilege escalation on
systems with these faulty services, stop allowing calls to
("/com/example/Whatever",
"org.freedesktop.DBus.UpdateActivationEnvironment")
and only allow ("/org/freedesktop/DBus",
"org.freedesktop.DBus.UpdateActivationEnvironment").

We deliberately continue to provide read-only APIs like
GetConnectionUnixUser at all object paths, for backwards compatibility.

Reviewed-by: Thiago Macieira <thiago@kde.org>
[adjusted commit message to note that this is probably only DoS -smcv]

9 years agoFix GetExtendedTcpTable crash on vista sp0.
Илья А. Ткаченко [Tue, 23 Dec 2014 08:45:24 +0000 (11:45 +0300)]
Fix GetExtendedTcpTable crash on vista sp0.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=77008
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
9 years ago1.8.13
Simon McVittie [Mon, 24 Nov 2014 22:52:22 +0000 (22:52 +0000)]
1.8.13

9 years ago1.9.5
Simon McVittie [Mon, 24 Nov 2014 22:51:44 +0000 (22:51 +0000)]
1.9.5

9 years agoStart spec 0.26
Simon McVittie [Mon, 10 Nov 2014 17:08:56 +0000 (17:08 +0000)]
Start spec 0.26

9 years ago1.9.4
Simon McVittie [Mon, 24 Nov 2014 13:08:06 +0000 (13:08 +0000)]
1.9.4

9 years agoRevert "Start spec 0.26"
Simon McVittie [Mon, 24 Nov 2014 13:05:37 +0000 (13:05 +0000)]
Revert "Start spec 0.26"

This reverts commit cdd9e9a65c40ac4c21dcd36e9bd0fd6c746ad753.

9 years agoUpgrade auth_timeout violation to a warning, since this branch has those
Simon McVittie [Tue, 18 Nov 2014 18:58:57 +0000 (18:58 +0000)]
Upgrade auth_timeout violation to a warning, since this branch has those

9 years agoMerge branch 'dbus-1.8'
Simon McVittie [Mon, 24 Nov 2014 13:05:09 +0000 (13:05 +0000)]
Merge branch 'dbus-1.8'

Conflicts:
NEWS
configure.ac

9 years ago1.8.12
Simon McVittie [Mon, 24 Nov 2014 13:01:40 +0000 (13:01 +0000)]
1.8.12

9 years agoRevert "config: change default auth_timeout to 5 seconds"
Simon McVittie [Tue, 18 Nov 2014 17:34:11 +0000 (17:34 +0000)]
Revert "config: change default auth_timeout to 5 seconds"

This reverts commit 54d26df52b6a394bea175651d1d7ad2ab3f87dea.

It appears this change may cause intermittent slow or failed boot,
more commonly on slower/older machines, in at least Mageia and
possibly also Debian. This would indicate that while the system
is under load, system services are not completing authentication
within 5 seconds.

This change was not the main part of fixing CVE-2014-3639, but does
help to mitigate that attack. As such, increasing this timeout makes
the denial of service attack described by CVE-2014-3639 somewhat
more effective: a local user connecting to the system bus repeatedly
from many parallel processes can cause other users' attempts to
connect to take longer.

If your machine boots reliably with the shorter timeout, and
resilience against local denial of service attacks is important
to you, putting this in /etc/dbus-1/system-local.conf
or a file matching /etc/dbus-1/system.d/*.conf can restore
the lower limit:

    <busconfig>
      <limit name="auth_timeout">5000</limit>
    </busconfig>

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=86431

9 years agoLog to syslog when auth_timeout drops an incomplete connection
Simon McVittie [Tue, 18 Nov 2014 12:01:44 +0000 (12:01 +0000)]
Log to syslog when auth_timeout drops an incomplete connection

This is a symptom of either a denial of service attack, or a
serious performance problem. Either way, sysadmins should know.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=86431

9 years agoUse DBUS_SYSTEM_LOG_WARNING for failure to alter fd-limits
Simon McVittie [Tue, 4 Nov 2014 14:51:22 +0000 (14:51 +0000)]
Use DBUS_SYSTEM_LOG_WARNING for failure to alter fd-limits

Reviewed-by: Alban Crequy <alban.crequy@collabora.co.uk>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=85105

9 years agoDBusSystemLogSeverity: add DBUS_SYSTEM_LOG_WARNING
Alban Crequy [Mon, 3 Nov 2014 16:04:45 +0000 (16:04 +0000)]
DBusSystemLogSeverity: add DBUS_SYSTEM_LOG_WARNING

Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=85105

9 years agoMerge branch 'dbus-1.8'
Simon McVittie [Fri, 14 Nov 2014 19:14:22 +0000 (19:14 +0000)]
Merge branch 'dbus-1.8'

9 years agoREADME, HACKING: add some brief notes on reporting security vulnerabilities
Simon McVittie [Fri, 14 Nov 2014 19:14:13 +0000 (19:14 +0000)]
README, HACKING: add some brief notes on reporting security vulnerabilities

We now have a private mailing list that can be the security contact.

9 years agoMerge branch 'dbus-1.8'
Simon McVittie [Fri, 14 Nov 2014 18:58:52 +0000 (18:58 +0000)]
Merge branch 'dbus-1.8'

9 years agoNEWS
Simon McVittie [Fri, 14 Nov 2014 18:58:44 +0000 (18:58 +0000)]
NEWS

9 years agoSet error when message delivery is denied due to receive rule
Jacek Bukarewicz [Fri, 14 Nov 2014 18:39:38 +0000 (18:39 +0000)]
Set error when message delivery is denied due to receive rule

This makes bus_context_check_security_policy follow convention of
setting errors if function indicates failure and has error parameter.
Notable implication is that AccessDenied error will be sent if sending message
to addressed recipient is denied due to receive rule. Previously, message
was silently dropped.

This also fixes assertion failure when message is denied at addressed recipient
while sending pending auto activation messages.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=86194

9 years agoMerge branch 'dbus-1.8'
Simon McVittie [Mon, 10 Nov 2014 17:10:09 +0000 (17:10 +0000)]
Merge branch 'dbus-1.8'

Conflicts:
NEWS

9 years agonext version number will be 1.8.12
Simon McVittie [Mon, 10 Nov 2014 17:09:52 +0000 (17:09 +0000)]
next version number will be 1.8.12

9 years agoStart spec 0.26
Simon McVittie [Mon, 10 Nov 2014 17:08:56 +0000 (17:08 +0000)]
Start spec 0.26

9 years ago1.9.3
Simon McVittie [Mon, 10 Nov 2014 17:07:31 +0000 (17:07 +0000)]
1.9.3

9 years ago1.8.11
Simon McVittie [Mon, 10 Nov 2014 17:06:44 +0000 (17:06 +0000)]
1.8.11

9 years ago1.9.2, spec 0.25
Simon McVittie [Thu, 6 Nov 2014 21:02:27 +0000 (21:02 +0000)]
1.9.2, spec 0.25

9 years agoMerge branch 'dbus-1.8'
Simon McVittie [Thu, 6 Nov 2014 20:59:57 +0000 (20:59 +0000)]
Merge branch 'dbus-1.8'

Conflicts:
NEWS
configure.ac

9 years agoEmbargoed security release for Monday
Simon McVittie [Thu, 6 Nov 2014 15:39:51 +0000 (15:39 +0000)]
Embargoed security release for Monday

9 years agoCVE-2014-7824: set fd rlimit to 64k for the system dbus-daemon
Simon McVittie [Tue, 4 Nov 2014 14:41:54 +0000 (14:41 +0000)]
CVE-2014-7824: set fd rlimit to 64k for the system dbus-daemon

This ensures that our rlimit is actually high enough to avoid the
denial of service described in CVE-2014-3636 part A.
CVE-2014-7824 has been allocated for this incomplete fix.

Restore the original rlimit for activated services, to avoid
them getting undesired higher limits.

(Thanks to Alban Crequy for various adjustments which have been
included in this commit.)

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=85105
Reviewed-by: Alban Crequy <alban.crequy@collabora.co.uk>
9 years agoNEWS
Simon McVittie [Thu, 6 Nov 2014 14:32:50 +0000 (14:32 +0000)]
NEWS

9 years agoAdd NetBSD to the list of platforms where credentials-passing a pid should work
Simon McVittie [Mon, 15 Sep 2014 17:37:59 +0000 (18:37 +0100)]
Add NetBSD to the list of platforms where credentials-passing a pid should work

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=69702
Reviewed-by: Patrick Welche <prlw1@cam.ac.uk>
Reviewed-by: Alban Crequy <alban.crequy@collabora.co.uk>
Tested-by: Patrick Welche <prlw1@cam.ac.uk>
9 years agotest_processid: only assert that it works if we expect it to work
Simon McVittie [Mon, 15 Sep 2014 17:38:49 +0000 (18:38 +0100)]
test_processid: only assert that it works if we expect it to work

Otherwise, this would fail on, for instance, QNX.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=69702
Reviewed-by: Patrick Welche <prlw1@cam.ac.uk>
Reviewed-by: Alban Crequy <alban.crequy@collabora.co.uk>
Tested-by: Patrick Welche <prlw1@cam.ac.uk>
9 years agowhitespace/comment fixes
Patrick Welche [Thu, 30 Oct 2014 12:39:26 +0000 (12:39 +0000)]
whitespace/comment fixes

[originally part of the previous commit -smcv]

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=69702
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Alban Crequy <alban.crequy@collabora.co.uk>
9 years agoImplement NetBSD credentials-passing with LOCAL_PEEREID
Patrick Welche [Tue, 29 Jul 2014 13:08:20 +0000 (14:08 +0100)]
Implement NetBSD credentials-passing with LOCAL_PEEREID

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=69702
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Alban Crequy <alban.crequy@collabora.co.uk>
Tested-by: Patrick Welche <prlw1@cam.ac.uk>
9 years agodbus-daemon test: don't assert we pass uid/pid on unknown Unix platforms
Simon McVittie [Mon, 15 Sep 2014 17:04:36 +0000 (18:04 +0100)]
dbus-daemon test: don't assert we pass uid/pid on unknown Unix platforms

We know that Linux, FreeBSD and OpenBSD are "first class citizens"
for credentials-passing, with NetBSD not far behind: people have
turned up on the bug tracking system and told us that tests passed.
On other Unixes, we can't really assert that it works, until someone
who runs them tells us that it worked for them.

Additions to these lists are welcome.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=69702
Reviewed-by: Patrick Welche <prlw1@cam.ac.uk>
Reviewed-by: Alban Crequy <alban.crequy@collabora.co.uk>
Tested-by: Patrick Welche <prlw1@cam.ac.uk>
9 years agoNEWS
Simon McVittie [Wed, 29 Oct 2014 14:21:19 +0000 (14:21 +0000)]
NEWS

9 years agodbus-spawn: do not forget the exec() errno when the grandchild exits
Simon McVittie [Thu, 11 Sep 2014 09:59:32 +0000 (10:59 +0100)]
dbus-spawn: do not forget the exec() errno when the grandchild exits

As is already noted in a comment in
_dbus_babysitter_set_child_exit_error(), if the grandchild fails
to exec() the desired process, we get both CHILD_EXEC_FAILED (with
an errno) and CHILD_EXITED (with a status), and we want to report
the former, since it is more informative. However, clearing
sitter->errnum meant we lose the errno value.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=24821
Reviewed-by: Ross Lagerwall
9 years agodbus-tutorial: replace the entire GLib section with "use GDBus"
Simon McVittie [Wed, 10 Sep 2014 12:53:39 +0000 (13:53 +0100)]
dbus-tutorial: replace the entire GLib section with "use GDBus"

Also provide links to relevant GLib and Qt documentation.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=25140

9 years agoStop asserting that we're not using the dummy lock implementation
Simon McVittie [Wed, 8 May 2013 15:58:08 +0000 (16:58 +0100)]
Stop asserting that we're not using the dummy lock implementation

That implementation no longer exists, so neither 0xABCDEF nor 0xABCDEF2
has any special meaning any more.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=54972
Signed-off-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: Alban Crequy <alban.crequy@collabora.co.uk>
9 years agoUse a better NoReply message for disconnection with reply pending
Simon McVittie [Wed, 29 Oct 2014 14:10:48 +0000 (14:10 +0000)]
Use a better NoReply message for disconnection with reply pending

As an implementation detail, dbus-daemon handles this situation by
artificially triggering a timeout (even if its configured timeout for
method calls is in fact infinite). However, using the same debug message
for both is misleading, and can lead people who are debugging a service
crash to blame dbus-daemon instead, wasting their time.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=76112

9 years agodbus-sysdeps-unix: document the assumption that makes our use of credentials secure
Simon McVittie [Thu, 11 Sep 2014 12:08:21 +0000 (13:08 +0100)]
dbus-sysdeps-unix: document the assumption that makes our use of credentials secure

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=83499
Reviewed-by: Alban Crequy <alban.crequy@collabora.co.uk>
9 years agoConsistently save and restore errno
Simon McVittie [Mon, 8 Sep 2014 19:18:22 +0000 (20:18 +0100)]
Consistently save and restore errno

Some functions in dbus-transport-socket.c make a (wrapped)
socket syscall, then call other APIs, then test the result and
errno of the socket syscall.

This would break horribly if those "other APIs" overwrote errno with
their own value (... and this is part of why errno is an awful API).

Notably, if running under DBUS_VERBOSE, _dbus_verbose() is basically
fprintf(), which sets errno; and our Unix fd-passing support
makes calls of the form _dbus_verbose ("Read/wrote %i unix fds\n", n)
between the syscall and the result processing.

Maybe one day we'll convert all of dbus' syscall wrappers to either
raise a DBusError, or use the "negative errno" convention that systemd
borrowed from the Linux kernel, and in particular, we would need to
do that if we ever ported it to a platform where socket error reporting
was not basically errno. However, in practice everyone uses something
derived from BSD sockets, so "this sets errno, you know what errno is"
is a good enough internal API if we make sure to use it correctly.

Nothing calls _dbus_get_is_errno_nonzero(), so I just removed it instead
of converting it to the new calling convention.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=83625

9 years agoNEWS
Simon McVittie [Wed, 29 Oct 2014 13:59:24 +0000 (13:59 +0000)]
NEWS

9 years agoImplement getter, setter for ALLOW_INTERACTIVE_AUTHORIZATION flag
Simon McVittie [Thu, 25 Sep 2014 12:29:50 +0000 (13:29 +0100)]
Implement getter, setter for ALLOW_INTERACTIVE_AUTHORIZATION flag

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=83449
Reviewed-by: Lennart Poettering
9 years agobuild: include dbus-sysdeps-unix.h for _dbus_fd_set_close_on_exec
Patrick Welche [Tue, 28 Oct 2014 15:31:58 +0000 (15:31 +0000)]
build: include dbus-sysdeps-unix.h for _dbus_fd_set_close_on_exec

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=85563
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agodbus-spec, dbus-protocol: add ALLOW_INTERACTIVE_AUTHORIZATION flag
Simon McVittie [Thu, 25 Sep 2014 12:19:24 +0000 (13:19 +0100)]
dbus-spec, dbus-protocol: add ALLOW_INTERACTIVE_AUTHORIZATION flag

Heavily based on a patch from Lennart Poettering.

This is useful for authentication frameworks such as polkit, but this
flag is supposed to be generic, and not be bound to any implementation
of such a framework.

The dbus specification already clarifies that unknown flags must be
ignored, the reference implementation and the other implementations we
checked indeed ignore any new flags, hence we should be fine with
compatibility here.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=83449

9 years agoInclude test-dbus and test-bus in cmake 'make check' target.
Ralf Habacker [Tue, 14 Jan 2014 17:59:18 +0000 (18:59 +0100)]
Include test-dbus and test-bus in cmake 'make check' target.

Because test-dbus and test-bus lives in subdirectory dbus/bus, we need
to define make 'check' in top level source directory.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=73689
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agotest-bus, test-dbus: close any inherited fds from caller
Simon McVittie [Fri, 24 Oct 2014 12:41:13 +0000 (13:41 +0100)]
test-bus, test-dbus: close any inherited fds from caller

It is probably a bug for them to pass us any fds without close-on-exec;
but apparently CMake has this bug, and so does at least some NetBSD GUI
environment. Cope.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=73689
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=83899
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
9 years agocmake: only copy session.conf and system.conf into test data dir
Simon McVittie [Fri, 24 Oct 2014 12:42:57 +0000 (13:42 +0100)]
cmake: only copy session.conf and system.conf into test data dir

Historically, CMake used the glob *.conf.in whereas Autotools listed
the files explicitly. This used to be equivalent, but broke down
when we added example-*.conf.in which are just snippets rather than
complete configuration files (they're intended to go in session.d
or system.d, or otherwise get included by the main config file).

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=73689
Reviewed-by: Ralf Habacker <ralf.habacker@freenet.de>
9 years agoAdd configure checks for accept4, dirfd, inotify_init1 and unix_fd_passing to cmake...
Ralf Habacker [Fri, 17 Oct 2014 13:42:06 +0000 (15:42 +0200)]
Add configure checks for accept4, dirfd, inotify_init1 and unix_fd_passing to cmake build system.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=73689
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agoAdd configure check for pipe2 to cmake build system.
Ralf Habacker [Tue, 14 Oct 2014 14:18:54 +0000 (16:18 +0200)]
Add configure check for pipe2 to cmake build system.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=73689
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agoNEWS
Simon McVittie [Tue, 14 Oct 2014 13:12:14 +0000 (14:12 +0100)]
NEWS

9 years agodbus-test-tool: add black-hole mode
Alban Crequy [Wed, 2 Jul 2014 13:18:26 +0000 (14:18 +0100)]
dbus-test-tool: add black-hole mode

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=34140
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>
9 years agodbus-test-tool spam: add --messages-per-conn=N
Alban Crequy [Wed, 2 Jul 2014 15:36:44 +0000 (16:36 +0100)]
dbus-test-tool spam: add --messages-per-conn=N

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=34140
Reviewed-by: Simon McVittie <simon.mcvittie@collabora.co.uk>