colin/NetworkManager
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
14,110 Commits 4 Branches 0 Tags
14c16a8b22ed53c40db2c14cb0e76da2c844917c
Commit Graph

14110 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Dan Williams
701c05ad69 build: remove spurious comma from configure.ac 2016-01-20 12:21:40 -06:00
Dan Williams
3b3f108dfd docs, trivial: fix some documentation issues 2016-01-20 11:27:22 -06:00
Dan Williams
0a12a1711c libnm-core,core,cli: fix spelling of NM_IP_TUNNEL_MODE_UNKNOWN 
Fixes: b614a5ec61
Fixes: ae8c7a8967
Fixes: 570fdce93f
 2016-01-20 11:27:22 -06:00
Dan Williams
b610d2dc0c docs: ignore some private libnm & libnm-core headers 2016-01-20 11:27:21 -06:00
Dan Williams
d59c1d4c8a libnm,vpn: restore export of deprecated NMVpnPluginOld symbols 
Deprecated of course, but shouldn't have been removed from the ABI.

Fixes: 867227dd4a
 2016-01-20 11:26:49 -06:00
Lubomir Rintel
0f199943de Revert "Set the default powersave value to enabled" 
libnm-core & more discussion were missing; backing out for now.

This reverts commit 15ab4bda5b.
 2016-01-20 13:53:40 +01:00
Beniamino Galvani
80e8b9cca9 merge: branch 'bg/ipv4-dad-rh1259063' 
Add support for detecting IPv4 duplicate addresses.

https://bugzilla.redhat.com/show_bug.cgi?id=1259063
 2016-01-20 11:57:13 +01:00
Beniamino Galvani
75068a085a device: detect duplicate IPv4 addresses when method=auto 2016-01-20 11:53:47 +01:00
Beniamino Galvani
28f6e8b4d2 device: detect duplicate IPv4 addresses when method=manual 2016-01-20 11:53:47 +01:00
Beniamino Galvani
83b712aa42 core: add test for NMArpingManager 2016-01-20 11:53:47 +01:00
Beniamino Galvani
7a0fe9b21e core: add NMArpingManager 
Add a new object which implements the logic for announcing IP
addresses and detecting duplicates using arping.

Based-on-patch-by: Jiří Klimeš <jklimes@redhat.com>
 2016-01-20 11:53:47 +01:00
Jiří Klimeš
df6b0fa99e cli: add support for new ipv4.dad-timeout property 2016-01-20 11:53:47 +01:00
Jiří Klimeš
adbbf3aa5c ifcfg-rh: read/write ipv4.dad-timeout using ARPING_WAIT 
ARPING_WAIT is used for DAD by Red Hat initscrips (ifup-eth).
 2016-01-20 11:53:47 +01:00
Jiří Klimeš
31ea5a99cb libnm: add NMSettingIPConfig 'dad-timeout' property 
The property is used to control duplicate address detection:
 * -1 means default value
 * 0 means no DAD is performed
 * > 0 means timeout (in milliseconds) for arping responses

[bgalvani: moved setting from NMSettingIP4Config]
 2016-01-20 11:53:47 +01:00
Beniamino Galvani
f607a16994 device/macvlan: implement dispose() 
When the object gets disposed, parent must be unreferenced and signals
disconnected.

Fixes: 4de8851eca

https://bugzilla.redhat.com/show_bug.cgi?id=1299803
 2016-01-20 11:41:59 +01:00
Mathieu Trudel-Lapierre
15ab4bda5b Set the default powersave value to enabled 
This is an Ubuntu-specific patch to enable wireless powersave by default.

Should it not work for you for some reason, you can safely add:

powersave=0

to the section:
[802-11-wireless]

In the configuration file for your connection under
/etc/NetworkManager/system-connections/.
 2016-01-20 11:27:47 +01:00
Thomas Haller
26721b8a88 core: use guint type for glib's source ids 2016-01-20 10:27:36 +01:00
Thomas Haller
a99a0d194d dhcp: clear watch_id in child-watch callback 2016-01-20 10:27:36 +01:00
Beniamino Galvani
701d98acc5 device: fix check on device state in carrier_changed() 
The function can now be called on unrealized devices before the
initial unmanaged flags are set and for those devices
nm_device_get_managed() will return TRUE. Since we only accept
states > UNMANAGED, return early when the condition is not met.

Fixes the following failed assertion:

carrier_changed: assertion 'priv->state >= NM_DEVICE_STATE_UNAVAILABLE' failed

https://bugzilla.gnome.org/show_bug.cgi?id=760844
 2016-01-20 09:53:10 +01:00
Thomas Haller
7231626d31 core: use NM_GOBJECT_PROPERTIES_DEFINE() 2016-01-19 21:20:36 +01:00
Thomas Haller
f427909bf8 shared: add NM_GOBJECT_PROPERTIES_DEFINE() 
Helper macros to define GObject property enum,
the obj_properties array, and a _notify() function.

We should move away from invoking property changed notifications by
string, as we already do for signals. This macro simplifies the
implementation of establishes a pattern for naming and usage.
 2016-01-19 21:20:36 +01:00
Lubomir Rintel
ecd729706e relese: fix NEWS formatting 
The double spacing was probably a mistake. Also, there was an extra line break.
 2016-01-19 15:10:35 +01:00
Lubomir Rintel
83d043b112 release: bump version to 1.1.90 (1.2-beta1) 2016-01-19 14:45:36 +01:00
Thomas Haller
76f90812f4 device: clear queued_state callback in dispose 
dispose() calls _cleanup_generic_pre() which in turn already calls
nm_device_queued_state_clear(). So we would expect that at the end
of dispose() no queued-state is pending.

However, there there are crashes reports in queued_set_state() with the
device instance already destructed (rh#1270247). Add this check trying
to avoid the crash and closing in on the cause.

Related: https://bugzilla.redhat.com/show_bug.cgi?id=1180827
Related: https://bugzilla.redhat.com/show_bug.cgi?id=1270247
Related: https://bugzilla.redhat.com/show_bug.cgi?id=1298009
 2016-01-19 13:00:20 +01:00
Christian Kirbach
9e0ca0058b po: update German (de) translation (bgo #760806) 
https://bugzilla.gnome.org/show_bug.cgi?id=760806
 2016-01-19 12:04:28 +01:00
Beniamino Galvani
9bc5600aa4 man: document the effects of supported signals 2016-01-19 11:48:49 +01:00
Thomas Haller
1e90ef2043 device: remove unrealized device without connections on idle handler 
Emitting the NM_DEVICE_REMOVED signal can cause the destruction of the
device, thus it should be emitted on an idle handler.
 2016-01-19 11:38:29 +01:00
Thomas Haller
b0fc51841a config: implement nm_config_change_flags_to_string() via NM_UTILS_FLAGS2STR_DEFINE() 2016-01-19 11:14:51 +01:00
Thomas Haller
a841d40039 callouts: always print the environment even if no scripts are present 
Commit ac6522bbf1 moved logging of the environment
after the check for @error_message to avoid a crash. However, in case of no
scripts (request->scripts->len == 0) we still want to log the environment,
so move it back to where it was, but avoid the crash.
 2016-01-19 11:01:33 +01:00
Lubomir Rintel
f7c76b372c device: don't remove a devices on connection availability checks 
The availability checks are called from places that don't assume the device
will be removed mid-air. Call the removal routine only when we're the very
last thing that's being done.
 2016-01-18 21:09:28 +01:00
Lubomir Rintel
ac6522bbf1 dispatcher: only dump environment when there's no error 
(gdb) bt
  #0  0x00007f206bd1ec58 in handle_action (dbus_dispatcher=<optimized out>, context=0x7f2058003190, str_action=<optimized out>, connection_dict=0x7f205800bc40, connection_props=0x7f205800b900,
      device_props=0x7f205800b150, device_ip4_props=0x7f205800b120, device_ip6_props=0x7f205800b0f0, device_dhcp4_props=0x7f205800b0c0, device_dhcp6_props=0x7f205800b090, vpn_ip_iface=0x7f206c257990 "",
      vpn_ip4_props=0x7f205800b000, vpn_ip6_props=0x7f205800af90, request_debug=1, user_data=0x7f206c242980) at nm-dispatcher.c:722
  #1  0x00007f2068313dac in ffi_call_unix64 () at ../src/x86/unix64.S:76
  #2  0x00007f20683136d5 in ffi_call (cif=cif@entry=0x7ffdfde6d2f0, fn=<optimized out>, rvalue=0x7ffdfde6d260, avalue=avalue@entry=0x7ffdfde6d140) at ../src/x86/ffi64.c:522
  #3  0x00007f2069ba4678 in g_cclosure_marshal_generic (closure=0x7f206c23a9e0, return_gvalue=0x7ffdfde6d440, n_param_values=<optimized out>, param_values=<optimized out>, invocation_hint=<optimized out>,
      marshal_data=0x0) at gclosure.c:1454
  #4  0x00007f2069ba3e38 in g_closure_invoke (closure=0x7f206c23a9e0, return_value=return_value@entry=0x7ffdfde6d440, n_param_values=14, param_values=param_values@entry=0x7f206c256d10,
      invocation_hint=invocation_hint@entry=0x7ffdfde6d490) at gclosure.c:768
  #5  0x00007f2069bb575d in signal_emit_unlocked_R (node=node@entry=0x7f206c2505a0, detail=detail@entry=0, instance=instance@entry=0x7f206c2442a0, emission_return=emission_return@entry=0x7ffdfde6d5d0,
      instance_and_params=instance_and_params@entry=0x7f206c256d10) at gsignal.c:3553
  #6  0x00007f2069bbc641 in g_signal_emitv (instance_and_params=instance_and_params@entry=0x7f206c256d10, signal_id=signal_id@entry=6, detail=detail@entry=0, return_value=return_value@entry=0x7ffdfde6d5d0)
      at gsignal.c:3048
  #7  0x00007f206bd21aa1 in _nmdbus_dispatcher_skeleton_handle_method_call (connection=<optimized out>, sender=<optimized out>, object_path=<optimized out>,
      interface_name=0x7f2058010b20 "org.freedesktop.nm_dispatcher", method_name=0x7f205800ab00 "Action", parameters=<optimized out>, invocation=0x7f2058003190, user_data=0x7f206c2442a0)
      at nmdbus-dispatcher.c:1066
  #8  0x00007f2069ec8ba6 in g_dbus_interface_method_dispatch_helper (invocation=0x7f2058003190, method_call_func=0x7f206bd218d0 <_nmdbus_dispatcher_skeleton_handle_method_call>, interface=0x7f206c2442a0)
      at gdbusinterfaceskeleton.c:609
  #9  skeleton_intercept_handle_method_call (connection=<optimized out>, sender=sender@entry=0x7f205800ee50 ":1.1007", object_path=object_path@entry=0x7f20580108d0 "/org/freedesktop/nm_dispatcher",
      interface_name=interface_name@entry=0x7f2058010b20 "org.freedesktop.nm_dispatcher", method_name=method_name@entry=0x7f205800ab00 "Action", parameters=parameters@entry=0x7f205800af00,
      invocation=invocation@entry=0x7f2058003190, user_data=user_data@entry=0x7f206c2442a0) at gdbusinterfaceskeleton.c:650
  #10 0x00007f2069eb0bc1 in call_in_idle_cb (user_data=0x7f2058003190) at gdbusconnection.c:4873
  #11 0x00007f20696a279a in g_main_dispatch (context=0x7f206c23f9a0) at gmain.c:3109
  #12 g_main_context_dispatch (context=context@entry=0x7f206c23f9a0) at gmain.c:3708
  #13 0x00007f20696a2ae8 in g_main_context_iterate (context=0x7f206c23f9a0, block=block@entry=1, dispatch=dispatch@entry=1, self=<optimized out>) at gmain.c:3779
  #14 0x00007f20696a2dba in g_main_loop_run (loop=0x7f206c23fa80) at gmain.c:3973
  #15 0x00007f206bd1dcd4 in main (argc=1, argv=0x7ffdfde6da88) at nm-dispatcher.c:948
  (gdb) print error_message
  $4 = 0x7f206bd22980 "Missing or invalid required value path!"
  (gdb)
 2016-01-18 19:12:17 +01:00
Thomas Haller
988ad91976 manager: refactor construction of manager singleton instance 
nm_manager_setup() should only call g_object_new() and leave
most of the initialization to the GObject constructor.
 2016-01-18 18:12:15 +01:00
Thomas Haller
be9ff2c51a manager: initilize NETWORKING_ENABLED as constructor paramter 2016-01-18 14:22:50 +01:00
Thomas Haller
c8e7957c63 core: add NMLOG macros to nm-manager.c 2016-01-18 14:04:56 +01:00
Thomas Haller
4b917123ef platform: use NM_UTILS_ENUM2STR_DEFINE() to implement to-string functions 2016-01-18 13:10:18 +01:00
Thomas Haller
1d15d8c3ef core: add nm_utils_enum2str() 2016-01-18 13:10:18 +01:00
Thomas Haller
94fabafc1a platform: use RT_SCOPE_* names in nm_platform_route_scope2str() 2016-01-18 13:10:18 +01:00
Thomas Haller
262ba62708 core/trivial: add code comment explaining route-handling and gateway 2016-01-18 13:10:18 +01:00
Lubomir Rintel
636380d19d release: update NEWS 2016-01-18 13:06:30 +01:00
Thomas Haller
320a3dee14 device: merge branch 'lr/default-unmanaged-bgo746566' (part 4) 
Merge an early part of 'lr/default-unmanaged-bgo746566' to fix
issues related to realizing devices.

https://bugzilla.gnome.org/show_bug.cgi?id=746566
 2016-01-18 11:56:26 +01:00
Thomas Haller
3d463e2f69 device: only set NMDevice as "real" at the end of nm_device_realize_finish() 
Only set the device as "real" at the end of nm_device_realize_finish(),
not already during nm_device_realize_start()/realize_start_setup().
 2016-01-17 22:40:09 +01:00
Thomas Haller
9c2266741b device: cleanup unrealized device without connection 
During nm_device_unrealize(), the connection might still
have some connections that only get removed later. We must
garbage collect unrealized devices when they loose their
available-connections, not during unrealize.

Fixes: 436ec5b8e3
 2016-01-17 22:40:09 +01:00
Thomas Haller
eab9cca87c device: refactor cp_connection* functions to use @self pointer 2016-01-17 22:40:09 +01:00
Thomas Haller
2292cb63bf core: always unrealize software devices when platform link disappears 
We possibly need the unrealized device for connections that need
this. nm_device_unrealize() will check if there are still any
available-connections and possibly emit DEVICE_REMOVED signal.

Fixes: 7e5f27a21c
 2016-01-17 22:40:04 +01:00
Thomas Haller
7830b79d50 device: minor logging changes to device creation 2016-01-17 22:38:28 +01:00
Thomas Haller
ae7e468630 trivial: fix whitespace 2016-01-17 22:38:28 +01:00
Thomas Haller
7a2a96f8ef platform/tests: add test for handling IPv4 zero gateways 
Adding a route via iproute2 with explicit gateway 0.0.0.0
sets the route scope to GLOBAL, contrary to a on-link device-route
which has scope LINK.

Add a test adding two such routes.
 2016-01-16 21:27:12 +01:00
Thomas Haller
4c79b605bc platform: implement nmp_object_is_visible() based on nmp_object_is_alive() 
An object should only be visible if it is also alive.
Hence, refactor nmp_object_is_visible() to be reject
object that fail nmp_object_is_alive().
 2016-01-16 13:43:23 +01:00
Thomas Haller
3294e06c28 platform: add "link" scope to nm_platform_route_scope2str() 2016-01-16 13:43:23 +01:00
Lubomir Rintel
c4bd0cc5cb device: always process hop limit and MTU from the RA 
Apply MTU and hop limit when managed or other flag is set as well. This wasn't
the case for the first RA and the previous commit made it slightly worse by
always ignoring it.

Fixes: ad2584c375
 2016-01-15 12:08:24 +01:00