Can't just substitute sysconfdir into a header file -- it's meant to be
expanded in a Makefile. Otherwise, unexpanded ${prefix} will end up in a
header file.
We do that for NMCONFDIR already, let's use it here too.
Fixes: 2144457fab
When you want to run valgrind for a test, you either had to
invoke valgrind manually, or doing it via `make check` (provided
you configured --with-valgrind).
Make it more convenient to run valgrind by passing the test
to run to the "run-test-valgrind.sh" wrapper.
This also allows to pass -p/-s to the test, which is not possible
during `make check` because selecting tests conflicts with "--tap".
The following invocations are largely equivalent and work as
expected:
$ ./tools/run-test-valgrind.sh ./src/platform/tests/test-link-linux -p /link/software/detect/vlan
$ NMTST_DEBUG=no-debug,p=/link/software/detect/vlan ./tools/run-test-valgrind.sh ./src/platform/tests/test-link-linux
Instead of using libnl-route-3 library to serialize netlink messages,
construct the netlink messages ourselves.
This has several advantages:
- Creating the netlink message ourself is actually more straight
forward then having an intermediate layer between NM and the kernel.
Now it is immediately clear, how a platform request translates to
a netlink/kernel request.
You can look at the kernel sources how a certain netlink attribute
behaves, and then it's immediately clear how to set that (and vice
versa).
- Older libnl versions might have bugs or missing features for which
we needed to workaround (often by offering a reduced/broken/untested
functionality). Now we can get rid or workaround like _nl_has_capability(),
check_support_libnl_extended_ifa_flags(), HAVE_LIBNL_INET6_TOKEN.
Another example is a libnl bug when setting vlan ingress map which
isn't even yet fixed in libnl upstream.
- We no longer need libnl-route-3 at all and can drop that runtime
requirement, saving some 400k.
Constructing the messages ourselves also gives better performance
because we don't have to create the intermediate libnl object.
- In the future we will add more link-type support which is easier
to support by basing directly on the plain kernel/netlink API,
instead of requiring also libnl3 to expose this functionality.
E.g. adding macvtap support: we already parsed macvtap properties
ourselves because of missing libnl support. To *add* macvtap
support, we also would have to do it ourself (or extend libnl).
Constructing the libnl3 object only to parse the message
is wasteful. It involves several memory allocations, thread
synchronization and parsing fields that we don't care about.
But moreover, older libnl version might not support all the
fields we are interested in, hence we have workarounds like
_nl_link_parse_info_data(). Certain features might not fully
work unless libnl supports it too (although kernel would).
As we already parse the data ourselves sometimes, just go
all they way and don't construct the intermediate libnl object.
This will allow us to drop the _nl_link_parse_info_data() workarounds
in next commits. That is important, because _nl_link_parse_info_data()
sidesteps our platform cache and is not in sync with the cache (not to
mention the extra work to explicitly refetch the data on every lookup).
Also, it gets us 60% on the way to no longer needing 'libnl-route-3.so'
at all and eventually drop requiring the library.
Allows to enable more-asserts more granularly.
Unfortunately, the old check was "${enable_more_asserts} == "yes", thus
we cannot extend "--enable-more-assert=level" because that would mean
that the same build script cannot set the option on both old and new
NetworkManager.
Thus, add a new option --with-more-asserts=level. If you put the
following in your build script, it will work as expected whether
you build a new or an old version of NetworkManager.
./configure --enable-more-asserts --with-more-asserts=5
For libnm library, "nm-dbus-interface.h" contains defines like the D-Bus
paths of NetworkManager. It is desirable to have this header usable without
having a dependency on "glib.h", for example for a QT application. For that,
commit c0852964a8 removed that dependancy.
For libnm-glib library, the analog to "nm-dbus-interface.h" is
"NetworkManager.h", and the same applies there. Commit
159e827a72 removed that include.
However, that broke build on PackageKit [1] which expected to get the
version macros by including "NetworkManager.h". So at least for libnm-glib,
we need to preserve old behavior so that a user including
"NetworkManager.h" gets the version macros, but not "glib.h".
Extract the version macros to a new header file "nm-version-macros.h".
This header doesn't include "glib.h" and can be included from
"NetworkManager.h". This gives as previous behavior and a glib-free
include.
For libnm we still don't include "nm-version-macros.h" to "nm-dbus-interface.h".
Very few users will actually need the version macros, but not using
libnm.
Users that use libnm, should just include (libnm's) "NetworkManager.h" to
get all headers.
As a special case, a user who doesn't want to use glib/libnm, but still
needs both "nm-dbus-interface.h" and "nm-version-macros.h", can include
them both separately.
[1] https://github.com/hughsie/PackageKit/issues/85
Fixes: 4545a7fe96
NetworkManager set wpa_supplicant's fragment_size option to 1300. But if MTU
was lower, wpa_supplicant failed with "l2_packet_send - sendto: Message too
long" due to fragmentation of EAP-TLS or EAP-PEAP packets.
Actually, MTU has to be 14 bytes bigger than the "fragment_size" parameter.
Ideally, wpa_supplicant would take MTU in the account and adjust the
fragmentation limit accordingly. See discussion in
http://lists.shmoo.com/pipermail/hostap/2015-August/033546.htmlhttps://bugzilla.gnome.org/show_bug.cgi?id=755145
GNOME release tooling repacks the bz2 to a xz anyway. This makes it easier for
packagers to use tarballs created with "make dist" in place of distribution
tarballs.
Add --without-libnm-glib, for people who don't want to build the
legacy client libraries. When building with this option, dbus-glib and
libdbus are not required.
Port remaining bits to gdbus and remove stray dbus-glib references
Drop the dbus-glib version check from configure, since nothing depends
on new dbus-glib any more.
Move nm-dbus-glib-types.h and nm-gvaluearray-compat.h from include/ to
libnm-util/ since they are now only used by libnm-util and libnm-glib.
Previously, when compiling NetworkManager with libaudit support, it
was disabled by default and only used after setting logging.audit=true.
Turn that around. If we compile NetworkManager with audit support, we also
enable it by default. The user can then explicitly disable it by
configuring logging.audit in NetworkManager.conf.
But also, add a configure option 'yes-disabled-by-default' to compile
with audit support, but have it disabled by default. This would be the
previous behavior, but it must be enabled explicitly.
Fixes: be49a59fb6
Introduce some primitives to deliver messages about relevant
configuration changes to the Linux audit subsystem through libaudit
(if enabled at build time) and to the logging system.
The PKG_CHECK_MODULES macro shipped with modern versions (at least
0.24) of pkg-config already calls AC_SUBST to generate ${PKG}_CFLAGS
and ${PKG}_LIBS variables in Makefiles.
Remove the unneeded occurrences of AC_SUBST after PKG_CHECK_MODULES in
configure.ac; this should be safe because we are already assuming that
pkg-config version is recent enough in some other places.
Rather than randomly including one or more of <glib.h>,
<glib-object.h>, and <gio/gio.h> everywhere (and forgetting to include
"nm-glib-compat.h" most of the time), rename nm-glib-compat.h to
nm-glib.h, include <gio/gio.h> from there, and then change all .c
files in NM to include "nm-glib.h" rather than including the glib
headers directly.
(Public headers files still have to include the real glib headers,
since nm-glib.h isn't installed...)
Also, remove glib includes from header files that are already
including a base object header file (which must itself already include
the glib headers).
Originally, ibft settings were handled by "ifcfg-rh" plugin. Later, we added
a separate "ibft" plugin and moved the functionality there.
The problem was that users quite possibly had a configuration like
[main]
plugins=ifcfg-rh
in their "NetworkManager.conf". That meant, after upgrade users would
no longer have ibft support.
We fixed that by installing "/etc/NetworkManager/conf.d/10-ibft-plugin.conf"
which was read after the main file and contained:
[main]
plugins+=ibft
We no longer want to install configuration snippets with our core packages to
/etc. Avoid the regression by changing the meaning of "ifcfg-rh". By enabling
"ifcfg-rh" you now implicitly enable "ibft" plugin as well. This can be
turned off via "no-ibft". And you can continue to enable "ibft" plugin
alone.
The 'example' settings plugin is (obviously) unused, but it is also
badly maintained and no longer best-practice in several ways:
- it directly reads "NetworkManager.conf" instead of using NMConfig.
- it parses device specs itself, instead of using
nm_match_spec_split().
- read_connections() doesn't ensure that loading a file
does not replace a previously loaded one (due to conflicting
UUID). In general the example doesn't show/handle the complexity
of potential UUID conflicts.
Instead of fixing these issues it is better to ensure our main plugin
('keyfile') corresponds to current best practices. Should we ever add a
new pluginww, 'keyfile' should be the example.
After the hostname functionality has been moved from plugins to core,
the ifcfg-suse plugin contains only boilerplate code with no actual
functionality.
Remove the plugin, mark it as deprecated in manual page and print a
warning when it is selected in configuration file.
How to write and read the machine hostname is something that has been
handled until now by plugins; this is questionable since the method
using for storing the hostname should depend only on the distro used
and not on which plugins are enabled.
This commit moves all hostname-related functions from plugins to the
core and allows to specify the method used to load and store the
hostname at build time with the
--with-hostname-persist=default|suse|gentoo
configure option.
'default' method stores the hostname to /etc/hostname and monitors it
to detect runtime changes.
When the selected method is 'suse', the hostname gets read from and
written to /etc/HOSTNAME; the file /etc/sysconfig/network/dhcp is also
read to detect if the hostname is dynamic and thus invalid. Both files
are monitored for changes.
'gentoo' method relies on /etc/conf.d/hostname for storing the
hostname.
This reverts commit 7daf63461de4195b1626ca15f835fc7cbc56e847.
Turns out the removal of the second set of [] in configure.ac causes the command
to be wrong in 'configure' and the test to be incorrect.
We switched to user-mode PPPoE client to work around a bug in kernel
PPPoE code that caused pppd not to be notified when the connection was
terminated with a PADT before the LCP Termination Request.
The kernel bug has now been fixed upstream with commit 287f3a943fef
"pppoe: Use workqueue to die properly when a PADT is received", queued
for v4.1.
Since the issue affected only very particular scenarios and could
somehow be solved with the right configuration (see bugzilla entry),
we can safely revert the patch and restore the use of kernel mode
PPPoE.
This reverts commit 7955806a02.
https://bugzilla.gnome.org/show_bug.cgi?id=742939
Add a new 'rc-manager' configuration parameter that allows to select
the strategy used to write resolv.conf; currently supported values
are: none|resolvconf|netconfig, 'none' meaning that NM directly writes
the file.
The default value of the parameter is 'none'; however if a
RESOLVCONF_PATH (or NETCONFIG_PATH) is specified at build time, the
default value will be 'resolvconf' (or 'netconfig').
Even Fedora is no longer shipping the WiMAX SDK, so it's likely we'll
eventually accidentally break some of the code in src/devices/wimax/
(if we haven't already). Discussion on the list showed a consensus for
dropping support for WiMAX.
So, remove the SDK checks from configure.ac, remove the WiMAX device
plugin and associated manager support, and deprecate all the APIs.
For compatibility reasons, it is still possible to create and save
WiMAX connections, to toggle the software WiMAX rfkill state, and to
change the "WIMAX" log level, although none of these have any effect,
since no NMDeviceWimax will ever be created.
nmcli was only compiling in support for most WiMAX operations when NM
as a whole was built with WiMAX support, so that code has been removed
now as well. (It is still possible to use nmcli to create and edit
WiMAX connections, but those connections will never be activatable.)
NM core uses nm-logging which is entirely configurable at runtime.
Other components use glib-logging, which can also be partly configured
via G_MESSAGES_DEBUG.
It makes sense to have a compile time option to enable some
logging statements that are only useful for heavy debugging.
For glib-logging, this is a way to enable/disable extra logging.
For nm-logging, we could alternatively configure a least log-level
that is enabled at compile time (that way, we could configure to prune all
LOGL_TRACE logging). While that might be useful (too), this gives
an alternative way to disable/enable logging.
Add a configure option --enable-more-logging and a NM_MORE_LOGGING define
for that.
If we don't find this useful after a while, we can simply remove it,
because our logging statements are not part of a "stable" behavior.
NM already has two kinds of assertions:
- g_assert*(), conditionally compiled via #ifndef G_DISABLE_ASSERT
- g_return*(), conditionally compiled via #ifndef G_DISABLE_CHECKS
In theory, one should be able to disable both asserts and NM should
still work correctly (and possibly more efficient). In practice,
hardly anybody is testing such a configuration and it might be broken.
Especially, we don't disable asserts for production builds, both because
of less test coverage and because it might reduce our ability to debug.
Add a new configure option --enable-more-asserts, which defines
NM_MORE_ASSERTS and nm_assert(). This is for expensive asserts,
that -- contrary to the asserts above -- are disabled by default.
This is useful for extended debugging.
