result_cb invokes a function pointer provided by the user. Nothing prevents
the user from destroying the NMRemoteConnection in the callback, which leads
to a crash. Take an additional ref of NMRemoteConnection to keep it
alive.
This probably caused a crash for nm-applet:
https://bugzilla.redhat.com/show_bug.cgi?id=1030403
Signed-off-by: Thomas Haller <thaller@redhat.com>
If you called nm_remote_connection_delete() on a connection whose only
ref was held by the NMRemoteSettings, then the callback would never
get called, because NMRemoteSettings would drop its ref before then
(when the connection emitted the 'removed' signal), so the callback
would get cancelled.
Fix this by taking an extra ref on the connection around the D-Bus
call in this case.
https://bugzilla.redhat.com/show_bug.cgi?id=997568https://bugzilla.gnome.org/show_bug.cgi?id=706141
Use the D-Bus connection helper whenever we need a connection to
NM, which by default tries to use a private connection instead of
the shared bus connection whenever the user is root. Doing this
by default will not change the behavior of libnm-glib, and allows
tools like nmcli and libnm-glib-using clients to work in minimal
environments (those without a bus daemon) by default.
When using a private connection, we need to use dbus_g_proxy_new_for_peer()
because the bus isn't involved. Since many parts of libnm-glib create a
proxy for their corresponding remote object, consolidate the proxy creation
logic.
A later patch will add logic to use a private connection versus a bus-based
one.
24cda2bc broke the ability to set NMConnection:path back to NULL after
it had been set, as part of a hack to try to make
NMRemoteConnection:dbus-path work. Fix that by moving the hack
entirely into NMRemoteConnection.
https://bugzilla.gnome.org/show_bug.cgi?id=693829
In order to resolve NMRemoteConnection-valued properties, NMObject
needs to be able to create NMRemoteConnections. But NMObject assumes
that all the objects it will be creating have "dbus-connection" and
"dbus-path" properties. So add those properties to NMRemoteConnection,
aliasing the existing "bus" and "path" properties (and ensure that
whichever version gets set, we keep that value, rather than letting it
get overwritten by the NULL default value of the other one).
https://bugzilla.gnome.org/show_bug.cgi?id=693669
If a class implements init_async, it should implement init_finish too,
rather than assuming the default implementation will do the right
thing (which it briefly didn't in glib 2.33).
In some situations, objects might get used after being disposed, so
clear out their various priv fields so we don't try to access unreffed
objects, freed strings, etc.
https://bugzilla.gnome.org/show_bug.cgi?id=674473
Implement GInitable and GAsyncInitable in NMObject, with
implementations that synchronously or asynchonously load all
properties, and change _nm_object_ensure_inited() to run
g_initable_init().
Update the object/object-array property handling to initialize the
objects after creating them (synchronously or asynchronously,
according to the situation), so that they will have all of their
properties preloaded before they are ever visible to the caller.
Move the non-blocking/non-failable parts of various objects'
constructor() methods to constructed(), and move the blocking/failable
parts to init(), and implement init_async() methods with non-blocking
versions of the blocking methods.
Make nm_device_new() and nm_client_new() call
_nm_object_ensure_inited(), to preserve the behaviour formerly
enforced by their construct() methods, that properties are guaranteed
to be initialized before any signals involving them are emitted.
Most of the code was using dbus_g_proxy_call() directly, but there
were some leftover uses of the generated bindings. Make things more
consistent by using dbus_g_proxy_call() everywhere, and stop building
the -bindings.h files.
Don't delete them if we don't have permission for them, since
we may get permission for them later via Update. But to listen
for Update we need the connection around.
When the connection becomes invisible to a user (ie, the permissions
of the connection no longer allow that user to view the connection)
then we have to hid the connection from clients. But we can't
just dispose of it, because visibility changes are signaled with
Update signals on the connection itself, and thus we need to keep
the connection around just in case it becomes visible to the user
again. But if it's invisible, make sure we clear out the settings
since they may have changed.
Simplifies code internally, and makes it easier for clients as well in
some cases where they want to control what ends up in the resulting
hash and what does not.
The Settings.Connection interface is now only provided by NetworkManager
itself since there is only one settings service. NM can validate
requests for secrets internally and thus there's no need to lock down
GetSecrets using a separate D-Bus interface, since PolicyKit provides
that functionality on systems where this is desirable (ie multi-user).
Single-user systems that do not have PolicyKit will inherently trust
the user already, or if not D-Bus auth is flexible enough to lock
down the GetSecrets method individually even if it's not on a separate
D-Bus interface.
Second, since only clients like connection editors or applets will be
calling the GetSecrets method, there's no need for 'hints' or
'request_new' arguments here since this method should never trigger
an interactive secrets request. Only NM should send those requests
when it knows it needs to ask the user, either during connection or
after validating the incoming GetSecrets request. A connection editor
type application should never be able to trigger the normal
"What's your passphrase" dialog thats provided by the secret agent
for that user's session.
Add the necessary annotations (the mininum required, that is those
on return values. NULL parameters or container types may require
more), and the Autotools stuff to get a NetworkManager GIR for
libnm-util and a NMClient for libnm-glib.
Just for consistency, make settings related stuff live under the
org.freedesktop.NetworkManager namespace, rather than its own
org.freedesktop.NetworkManagerSettings namespace. Renames are done for
DBus interface names, DBus object paths, and polkit actions.
Remove the org.freedesktop.NetworkManagerSystemSettings bus name and
have everybody talk to org.freedesktop.NetworkManager. Now that we have
a single settings service that's embedded in the main daemon, we don't
need separate names anymore.
NMSettingsConnectionInterface was created to allow the daemon and NM
clients to have common code that handled both system and user
connections. It's no longer needed now that user settings services are
gone.
This concludes the flattening of libnm-glib.
In continuation of the theme, the removal of user settings services
means that the distinction between NMSysconfigConnection and
NMExportedConnection is no longer needed. Merge NMExportedConnection
into NMSysconfigConnection.
Remove code related to "connection scope" and such. Later, we will also
do lots of code flattening and simplification that's possible now that
user settings are gone.
The old NMExportedConnection was used for both client and server-side classes,
which was a mistake and made the code very complicated to follow. Additionally,
all PolicyKit operations were synchronous, and PK operations can block for a
long time (ie for user input) before returning, so they need to be async. But
NMExportedConnection and NMSysconfigConnection didn't allow for async PK ops
at all.
Use this opportunity to clean up the mess and create GInterfaces that both
server and client objects implement, so that the connection editor and applet
can operate on generic objects like they did before (using the interfaces) but
can perform specific operations (like async PK verification of callers) depending
on whether they are local or remote or whatever.
