Daniel Wagner [Mon, 13 Aug 2012 08:40:30 +0000 (10:40 +0200)]
session: Remove RoamingPolicy documetation
Remove this setting from the D-Bus API and move to the configuration
plugin.
Daniel Wagner [Mon, 13 Aug 2012 08:40:29 +0000 (10:40 +0200)]
session: Move EmergencyCall to configuration plugin
Instead fallback using the configuration plugin for selecting
the emergency application.
Now we don't have to update all other sessions which makes
the whole code logic more straight forward. If a session is
marked as an emergency call and this application calls
Connect()/Disconnect() it will succedd and block all other
sessions when they try to call Connect() or Disconnect().
Daniel Wagner [Mon, 13 Aug 2012 08:40:28 +0000 (10:40 +0200)]
session: Remove EmergencyCall documentation
The EmergencyCall setting which should be set by the platform integrator
and not by the application developer is moved the configuration plugin.
Daniel Wagner [Mon, 13 Aug 2012 08:40:27 +0000 (10:40 +0200)]
session: Remove StayConnected implementation
Daniel Wagner [Mon, 13 Aug 2012 08:40:26 +0000 (10:40 +0200)]
session: Remove StayConnected documentation
This setting is implicitly already available through AutoConnect.
Having yet another way to express this doesn't make it work any better.
Daniel Wagner [Mon, 13 Aug 2012 08:40:25 +0000 (10:40 +0200)]
session: Move Priority to configuration plugin
The Priority setting which should be set by the platform integrator
and not by the application developer is moved the configuration plugin.
Daniel Wagner [Mon, 13 Aug 2012 08:40:24 +0000 (10:40 +0200)]
session: Remove Priority documentation
Remove this setting from the D-Bus API and move it to the
configuration plugin.
Daniel Wagner [Mon, 13 Aug 2012 08:40:23 +0000 (10:40 +0200)]
session: Add configuration plugin
Daniel Wagner [Mon, 13 Aug 2012 08:40:22 +0000 (10:40 +0200)]
session: Remove SessionMarker implementation
Daniel Wagner [Mon, 13 Aug 2012 08:40:21 +0000 (10:40 +0200)]
session: Remove SessionMarker documentation
SO_MARK can only be set if the application has the CAP_NET_ADMIN
capability. We don't want to give that to all application for
good reasons and thus this approach won't work. Also touching all
applications to get this feature working is not a good idea. We
are going to use a cgroup controller to set SO_MARK and therefore
we don't need it in the Session API.
Daniel Wagner [Mon, 13 Aug 2012 08:40:20 +0000 (10:40 +0200)]
session: Remove PeriodicConnect implementation
Daniel Wagner [Mon, 13 Aug 2012 08:40:19 +0000 (10:40 +0200)]
session: Remove PeriodicConnect documentation
The idea was to let ConnMan know and optimize the connection
scheduling. Doing this per Session is a bit over-engineered.
The only use cases that have come up so far was for things like an
email client checking for new message.
Past experience showed that this kind of interface lead easily to
live locks. If there is need for such a feature, doing it on a per
technology would make more sense. For the time beeing it will be
removed. Later it could still be added to the Technology API.
If an application still wants to have a connection on regular
basis it can call the Session Connnect() method. ConnMan is still
able to defer any connection request.
Daniel Wagner [Mon, 13 Aug 2012 08:40:18 +0000 (10:40 +0200)]
session: Update ConnectionType documentation
The ConnectionType configuration option will be moved to
the config plugin when we know how the application can be
uniquely identified in a persistent way.
Daniel Wagner [Mon, 13 Aug 2012 08:40:17 +0000 (10:40 +0200)]
session: Remove IdleTimeout implementation
Daniel Wagner [Mon, 13 Aug 2012 08:40:16 +0000 (10:40 +0200)]
session: Move IdleTimeout to Technology API
Having an IdleTimeout per Session is not really reasonable. It is
hard to define the right behavoir if the timeouts differ per Session.
Doing this per technology is easier to understand and also makes
sense when no Session is used.
IdleTimeout is only documented here, it's not yet implemented.
Daniel Wagner [Mon, 13 Aug 2012 08:40:15 +0000 (10:40 +0200)]
session: Update overview doc
Update to the new D-Bus API and the simplified connection
algorithms (state).
Alok Barsode [Wed, 29 Aug 2012 14:07:15 +0000 (17:07 +0300)]
service: Do not report error twice
In request_input_cb(), __connman_service_indicate_error() unconditionally
reports an error via agent after determing the service state. The error
reporting and user connectable state will be properly handled by
service_indicate_state().
Patrik Flykt [Tue, 28 Aug 2012 07:51:30 +0000 (10:51 +0300)]
inet: Properly remove IPv4 address and set interface down
The interface was not properly set down and thus the services
were not removed.
Fixes BMC#25659
Jaganath Kanakkassery [Thu, 16 Aug 2012 12:51:19 +0000 (18:21 +0530)]
gdbus: Fix compilation error if macro "error" is defined
The variable "signature" used in error is not defined and "args" is now
a struct instead of a string.
Patrik Flykt [Fri, 24 Aug 2012 07:05:04 +0000 (10:05 +0300)]
pptp: Fix PPTP option typo
Fixes BMC#25660
Jukka Rissanen [Thu, 23 Aug 2012 11:19:37 +0000 (14:19 +0300)]
pptp: Get VPN gateway resolved IP address
We should not use the Host variable from provider as it can
contain a non-resolved address.
Fixes BMC#25658
Jukka Rissanen [Thu, 23 Aug 2012 11:19:36 +0000 (14:19 +0300)]
l2tp: Get VPN gateway resolved IP address
We should not use the Host variable from provider as it can
contain a non-resolved address.
Jukka Rissanen [Thu, 23 Aug 2012 11:19:35 +0000 (14:19 +0300)]
provider: Return VPN gateway IP address
If 'provider->host_ip[0]' is set, it contains the
resolved IP address. If not set, the 'Host' property
already contains an IP address.
Jukka Rissanen [Wed, 22 Aug 2012 08:28:01 +0000 (11:28 +0300)]
vpn: Accept also "true" or "1" as a boolean value
Currently user is able to set the boolean VPN config value
only by setting the variable to "yes". This is not very
intuitive so accept also the value of "true" or "1".
Jukka Rissanen [Tue, 21 Aug 2012 10:38:04 +0000 (13:38 +0300)]
device: Refactor device scan function
In wifi plugin this means that we combine three wifi scan
functions (normal, fast, hidden) into one scan function.
Marcel Holtmann [Wed, 22 Aug 2012 08:22:36 +0000 (10:22 +0200)]
Release 1.6
Jukka Rissanen [Tue, 21 Aug 2012 08:14:45 +0000 (11:14 +0300)]
doc: Manual page for connman configuration file
Jukka Rissanen [Tue, 21 Aug 2012 08:14:44 +0000 (11:14 +0300)]
doc: Manual page for connman
Jukka Rissanen [Tue, 21 Aug 2012 08:14:43 +0000 (11:14 +0300)]
git: Do not ignore connman manual pages
Patrik Flykt [Mon, 20 Aug 2012 12:43:52 +0000 (15:43 +0300)]
service: Always zero the timeout id when timer triggers
The timer id was not cleared in case the signal creation failed.
Patrik Flykt [Mon, 20 Aug 2012 12:43:51 +0000 (15:43 +0300)]
service: Send updated properties after ServicesChanged
Updated service properties are not sent before the new service
has been announced in a 'ServicesChanged' signal.
Fixes BMC#21285
Jukka Rissanen [Tue, 21 Aug 2012 10:55:09 +0000 (13:55 +0300)]
wifi: Ignore duplicate hidden APs
We must not add duplicate entries to list of hidden AP.
There can be more that one service file for each AP if
we have multiple wifi cards in the system.
Fixes BMC#24779
Jukka Rissanen [Tue, 21 Aug 2012 10:55:08 +0000 (13:55 +0300)]
wifi: Always Scan hidden APs
We should scan hidden AP even if the AP is not marked
as autoconnectable so that the user is able to connect
to it manually.
Tomasz Bursztyka [Tue, 21 Aug 2012 13:45:11 +0000 (16:45 +0300)]
wispr: Fix refcounting issue
Patrik Flykt [Mon, 20 Aug 2012 09:29:08 +0000 (12:29 +0300)]
service: Restore error clearing on Agent reply
If the invalid key error is not cleared on Agent reply, the
service cannot be reconnected.
Jukka Rissanen [Mon, 9 Jul 2012 14:17:08 +0000 (17:17 +0300)]
service: Allow connection if there are interfaces available
Allow user to initiate connection if there are multiple devices
of the same technology type available (like having multiple wifi
cards).
Patrik Flykt [Thu, 16 Aug 2012 09:21:34 +0000 (12:21 +0300)]
inet: Clear IP addresses when interface is set down
Clear interface IPv4 address by setting it to 0.0.0.0. IPv6
addresses are cleared automatically when the IFF_DYNAMIC flag is
set at the time the interface is brought down.
By removing the IP address and netmask, netlink properly reports
new addresses assigned by DHCPv4.
Patrik Flykt [Thu, 16 Aug 2012 09:20:45 +0000 (12:20 +0300)]
service: Clear service error on successfull connection attempts
Move clearing of the error to __connman_service_connect() so it is
cleared for all service connect code paths.
Patrik Flykt [Thu, 16 Aug 2012 09:20:26 +0000 (12:20 +0300)]
ipconfig: Use correct ipconfig ipaddress for FIXED method
Use ipconfig->address when reporting IP addresses for services
with FIXED method. AUTO method is used only by IPv6, don't report
any addresses should it erroneously appear for IPv4.
Fixes BMC#25084
Tomasz Bursztyka [Fri, 17 Aug 2012 07:16:42 +0000 (10:16 +0300)]
technology: Move SetProperty to async method call
Handle properly -EINPROGRESS error when enabling/disabling technology.
This fixes the existing timeout-based implementation.
Tomasz Bursztyka [Fri, 17 Aug 2012 07:16:41 +0000 (10:16 +0300)]
technology: Refactor enable/disable technology
- Cleanup the code (remove useless vars, remove duplicate code)
- Separate dbus logic to feature logic
- Fixes issue:
"dbus.exceptions.DBusException: net.connman.Error.Failed: Success"
Daniel Wagner [Fri, 17 Aug 2012 07:47:07 +0000 (09:47 +0200)]
dundee: Fix error code handling in add_device()
If no error happens then we should just return and not run
into the cleanup code. And when we cleanup we should not double
free pointers.
Tomasz Bursztyka [Thu, 16 Aug 2012 13:36:37 +0000 (16:36 +0300)]
bluetooth: Add the device into hash table when relevant
Fixes BMC#25215
Patrik Flykt [Thu, 16 Aug 2012 11:54:31 +0000 (14:54 +0300)]
timezone: Add comment
Andrei Gherzan [Wed, 15 Aug 2012 19:38:51 +0000 (22:38 +0300)]
timezone: If there is no d_type support use fstatat()
This is useful for filesystems where d_type is always DT_UNKNOWN,
for example reiserfs, jffs2 and ubifs. In this case use the
fstatat() function.
Marcel Holtmann [Wed, 15 Aug 2012 11:05:17 +0000 (13:05 +0200)]
Release 1.5
Jukka Rissanen [Tue, 14 Aug 2012 14:05:35 +0000 (17:05 +0300)]
provider: Return EINPROGRESS to the caller when connect is pending
Jukka Rissanen [Tue, 14 Aug 2012 14:05:34 +0000 (17:05 +0300)]
service: Remember provider pending dbus message
The provider connect dbus reply message is stored in service
until provider is connected.
Jukka Rissanen [Tue, 14 Aug 2012 14:05:33 +0000 (17:05 +0300)]
service: Allow setting path when sending pending dbus reply
Jukka Rissanen [Tue, 14 Aug 2012 14:01:03 +0000 (17:01 +0300)]
connection: Disconnect VPN when underlaying service disconnects
Fixes BMC#25128
Jukka Rissanen [Tue, 14 Aug 2012 14:01:02 +0000 (17:01 +0300)]
inet: Add function to return route to a given host
The netlink message handling is modified from iproute.c in iproute2 package.
Julien Massot [Tue, 14 Aug 2012 14:21:46 +0000 (16:21 +0200)]
wifi: Try only twice when connecting to a non-favorite service
Julien Massot [Tue, 14 Aug 2012 14:21:45 +0000 (16:21 +0200)]
service: Add getter for favorite flag
Tomasz Bursztyka [Mon, 13 Aug 2012 10:28:52 +0000 (13:28 +0300)]
iptables: Load table at the right places
Using -j/-m options without -t one, will segfault due to table not loaded
before hand.
Tomasz Bursztyka [Mon, 13 Aug 2012 10:28:51 +0000 (13:28 +0300)]
iptables: Refactor default return code usage
Tomasz Bursztyka [Mon, 13 Aug 2012 10:28:50 +0000 (13:28 +0300)]
tools: Refactor where table should be loaded
Fixes segfault issues when using a non-builtin target without providing table name.
like: -A OUTPUT -j IDLETIMER (...)
will crash when: -t filter -A OUTPUT -j IDLETIMER (...) will success
Jukka Rissanen [Mon, 13 Aug 2012 09:51:31 +0000 (12:51 +0300)]
service: Disconnect provider after connection timeout
The provider was not disconnected after a connection timeout.
This lead to weird issues in VPN code; like following provider
connect request would disconnect the earlier VPN connection and
report EALREADY to the user.
Lucas De Marchi [Wed, 8 Aug 2012 16:23:59 +0000 (13:23 -0300)]
doc: Improve description of service state property
Add missing "disconnect" state and point to doc/overview-api.txt for
more information about state transitions.
Daniel Wagner [Fri, 10 Aug 2012 12:23:18 +0000 (14:23 +0200)]
dundee Create ipconfig before setting index
Since connman_network_set_index() does not create magically the
ipconfig for us anymore, we need to create it inside set_connected().
Daniel Wagner [Thu, 9 Aug 2012 16:27:53 +0000 (18:27 +0200)]
bluetooth: Defer add_networks() if adapter is not powered yet
In the case we add a new adapter and the adapter is powered off
we would ignore the reported devices. Later when the adapter was
powered on it would just not add those networks.
Instead silently ignoring the devices, store the D-Bus paths in
a hash table and process them as soon the adapter is powered.
Fixes BMC#25322
Daniel Wagner [Thu, 9 Aug 2012 16:27:52 +0000 (18:27 +0200)]
bluetooth: Remove unused argument in check_networks()
Daniel Wagner [Thu, 9 Aug 2012 16:27:51 +0000 (18:27 +0200)]
bluetooth: Remove unused argument in add_network()
Daniel Wagner [Fri, 10 Aug 2012 09:59:04 +0000 (11:59 +0200)]
ofono: Set IPv6 nameservers
Elena Tebesoi [Fri, 10 Aug 2012 09:59:03 +0000 (11:59 +0200)]
network: Enable IPv6 on network interface
This patch enables IPv6 usage for network interfaces
with static/manual configuration.
Elena Tebesoi [Fri, 10 Aug 2012 09:59:02 +0000 (11:59 +0200)]
ipconfig: Allow NULL gateway for IPv6
This patch makes possible configuring a GPRS connection with
IPV6 settings and without gateway.
Daniel Wagner [Fri, 10 Aug 2012 09:59:01 +0000 (11:59 +0200)]
ofono: Create ipconfig before setting index
Since connman_network_set_index() does not create magically the
ipconfig for us anymore, we need to create it inside set_connected().
Daniel Wagner [Fri, 10 Aug 2012 09:59:00 +0000 (11:59 +0200)]
network: Only set index in connman_network_set_index()
Service ipconfig structures are no longer created in this
function.
Daniel Wagner [Fri, 10 Aug 2012 09:58:59 +0000 (11:58 +0200)]
service: Allow plugins to lookup services by networks
Daniel Wagner [Fri, 10 Aug 2012 09:58:58 +0000 (11:58 +0200)]
service: Allow plugins access to connman_service_create_ip4/6config
Julien Massot [Tue, 7 Aug 2012 16:57:06 +0000 (18:57 +0200)]
technology: Save tethering identifier and passphrase
Julien Massot [Tue, 7 Aug 2012 16:57:05 +0000 (18:57 +0200)]
technology: Rename save_state function by technology_save
Gustavo Padovan [Thu, 9 Aug 2012 21:57:26 +0000 (18:57 -0300)]
dundee: Clean up device in case of setup failure
If the creation and registration of Bluetooth DUN device fails we need to
clean up before return, otherwise we leak memory.
Gustavo Padovan [Thu, 9 Aug 2012 21:57:25 +0000 (18:57 -0300)]
bluetooth: Add device to hash before registration
During the connman_device_register() procedure a lookup to the
bluetooth_devices hash table happens, however the device is not on the
hash at this point and the look out fails.
If the registration fails, technology_disable() returns the Failed
message on D-Bus with the error status zero. That happens because we
don't have any device registered.
This patch moves the insertion of the device to before the device
registration.
Jukka Rissanen [Fri, 10 Aug 2012 08:21:26 +0000 (11:21 +0300)]
dhcp: Apply received hostname option according to conf file setting
We do not use the received DHCP hostname option (12) if system
config file setting says so.
Fixes BMC#25597
Jukka Rissanen [Fri, 10 Aug 2012 08:21:25 +0000 (11:21 +0300)]
main: Introduce conf file option to disable hostname updates
The AllowHostnameUpdates in system config file can be used
to ignore the hostname update request that for example
can be received via DHCP hostname option.
Jukka Rissanen [Wed, 8 Aug 2012 10:22:48 +0000 (13:22 +0300)]
provider: Discard routes to VPN server via VPN tunnel
These obsolete routes must be discarded as we cannot
contact VPN server via VPN tunnel. If the VPN server
address is not yet resolved, we do that while starting
the VPN client process.
Example of non working routes:
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
0.0.0.0 0.0.0.0 0.0.0.0 UG 0 0 0 vpn0
10.10.15.0 0.0.0.0 255.255.255.0 U 0 0 0 wlan0
10.10.15.1 0.0.0.0 255.255.255.255 UH 0 0 0 wlan0
10.252.112.0 0.0.0.0 255.255.240.0 U 0 0 0 vpn0
193.229.0.40 10.10.15.1 255.255.255.255 UGH 0 0 0 wlan0
193.229.0.42 10.10.15.1 255.255.255.255 UGH 0 0 0 wlan0
113.108.211.91 0.0.0.0 255.255.255.255 UH 0 0 0 vpn0
113.108.211.91 10.10.15.1 255.255.255.255 UGH 0 0 0 wlan0
The correct routes should be like this:
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
0.0.0.0 0.0.0.0 0.0.0.0 U 0 0 0 vpn0
10.10.15.0 0.0.0.0 255.255.255.0 U 0 0 0 wlan0
10.10.15.1 0.0.0.0 255.255.255.255 UH 0 0 0 wlan0
10.252.112.0 0.0.0.0 255.255.240.0 U 0 0 0 vpn0
193.229.0.40 10.10.15.1 255.255.255.255 UGH 0 0 0 wlan0
193.229.0.42 10.10.15.1 255.255.255.255 UGH 0 0 0 wlan0
113.108.211.91 10.10.15.1 255.255.255.255 UGH 0 0 0 wlan0
Jukka Rissanen [Wed, 8 Aug 2012 10:22:47 +0000 (13:22 +0300)]
provider: Allow VPN connection if not in connected state
Allow user to connect to VPN via dbus and supplying all
VPN parameters if the provider already exists but the VPN
is not in connected state.
This means that the user can call the Manager.ConnectProvider interface
multiple times to connect to VPN service. This patch is needed so that
OpenConnect can work correctly. Without this patch the OpenConnect cookie
cannot be passed to openconnect client program.
Jukka Rissanen [Wed, 8 Aug 2012 10:22:46 +0000 (13:22 +0300)]
connection: Set VPN default route to the VPN interface
Do not bother setting individual routes for default gateway so
for VPN we set the whole interface as a default gateway route.
Jukka Rissanen [Wed, 8 Aug 2012 10:22:45 +0000 (13:22 +0300)]
connection: Add more debugging when setting VPN routes
Grant Erickson [Wed, 8 Aug 2012 16:59:27 +0000 (09:59 -0700)]
service: Signal error property changes
In order to allow clients to update the property in a manner similar to
other service properties, signal changes in the service "Error" property
key/value pair.
The value "" is used when no error is asserted because D-Bus property
changed convention has no semantics for indicating a key/value pair was
removed.
Jukka Rissanen [Tue, 7 Aug 2012 14:08:41 +0000 (17:08 +0300)]
service: Initiate autoconnect after changing ipconfig method
Jukka Rissanen [Tue, 7 Aug 2012 14:08:40 +0000 (17:08 +0300)]
service: Do not set the service state when changing IPv4 method
The service state will be set to proper state if we are
actually doing the connection so do not set the state
when we are just setting the IPv4 method.
Jukka Rissanen [Tue, 7 Aug 2012 14:08:39 +0000 (17:08 +0300)]
service: Do not try to connect to service that cannot create connection
Check if the service IPv4/6 ipconfig is turned OFF, in this
case do not try to connect to that service because it will
not succeed.
Fixes BMC#25587
Jukka Rissanen [Tue, 7 Aug 2012 14:08:38 +0000 (17:08 +0300)]
ipconfig: Add function to check if the ipconfig is OFF
Grant Erickson [Tue, 7 Aug 2012 15:12:35 +0000 (08:12 -0700)]
main: Add support for specifying a non-default configuration file.
This patch adds support for specifying a main configuration file
different from the default value of /etc/connman/main.conf. This
is valuable during debug to specify alternate debug-only
configurations.
Daniel Wagner [Mon, 6 Aug 2012 16:34:06 +0000 (18:34 +0200)]
main: Remove unused compat argument
Andrei Gherzan [Mon, 6 Aug 2012 20:54:33 +0000 (23:54 +0300)]
storage.c: Search for settings file even if d_type is DT_UNKNOWN.
This is useful for filesystems where d_type is always DT_UNKNOWN,
for example in reiserfs.
Patrik Flykt [Mon, 6 Aug 2012 06:05:19 +0000 (09:05 +0300)]
TODO: Add WiMAX removal task
Daniel Wagner [Fri, 3 Aug 2012 11:30:14 +0000 (13:30 +0200)]
device: Remove unused phyindex
Marcel Holtmann [Tue, 24 Jul 2012 21:10:50 +0000 (23:10 +0200)]
Release 1.4
Tomasz Bursztyka [Tue, 24 Jul 2012 13:28:02 +0000 (16:28 +0300)]
wifi: Return -EBUSY when tethering
Fixes BMC#25561
Tomasz Bursztyka [Tue, 17 Jul 2012 13:32:06 +0000 (16:32 +0300)]
wispr: Add a reference counter mechanism on contexts
Basically, wispr provides a pointer (a wispr/portal context) to third parties
(agent, gweb...) which are answering asynchronously. However, if the technology
related to the service owning such context is disabled: this context is not
valid anymore, leading to a crash when third part answers after such action.
In order to avoid such issue, wispr implement an internal refcount system.
Fixes BMC#25479
Daniel Wagner [Wed, 18 Jul 2012 10:56:10 +0000 (12:56 +0200)]
AUTHORS: Mention Ross' contributions
Samuel Ortiz [Wed, 18 Jul 2012 10:23:49 +0000 (12:23 +0200)]
doc: Add coding-style.txt
This is a copy of the oFono coding-style document, with oFono instances
replaced with ConnMan ones.
It's also in sync with the neard one.
Ross Burton [Wed, 18 Jul 2012 09:37:38 +0000 (10:37 +0100)]
storage: check that the string isn't empty before splitting
If the string was non-NULL but empty (str="\0"), the following \0 assignment
would write to str[-1] and thus cause memory corruption.
On PPC and MIPS, this was causing crashes in glibc.
Tomasz Bursztyka [Wed, 11 Jul 2012 06:59:40 +0000 (09:59 +0300)]
device: Do not set regdom when device is not powered
Fixes crash bug:
connmand[6761]: plugins/ofono.c:get_properties_reply() /huawei_0 path /huawei_0 org.ofono.NetworkRegistration
connmand[6761]: plugins/ofono.c:netreg_properties_reply() /huawei_0
connmand[6761]: plugins/ofono.c:netreg_update_regdom() /huawei_0 MobileContryCode 460
connmand[6761]: src/device.c:connman_device_ref_debug() 0x8686ac0 ref 3 by plugins/wifi.c:991:wifi_set_regdom()
connmand[6761]: Aborting (signal 11) [src/connmand]
connmand[6761]: ++++++++ backtrace ++++++++
connmand[6761]: #0 0xb7749400 in
connmand[6761]: #1 0x806aa0b in g_supplicant_interface_set_country() at nat.c:0
connmand[6761]: #2 0x8062297 in wifi_set_regdom() at wifi.c:0
connmand[6761]: #3 0x80a607d in connman_technology_set_regdom() at ??:0
connmand[6761]: #4 0x806fefb in netreg_update_regdom() at ofono.c:0
connmand[6761]: #5 0x8071f3b in netreg_properties_reply() at ofono.c:0
connmand[6761]: #6 0x806f747 in get_properties_reply() at ofono.c:0
connmand[6761]: #7 0x4a374b62 in /usr/lib/libdbus-1.so.3
connmand[6761]: #8 0x4a3602b7 in /usr/lib/libdbus-1.so.3
connmand[6761]: #9 0x4a3639f5 in /usr/lib/libdbus-1.so.3
connmand[6761]: #10 0x8054250 in message_dispatch() at mainloop.c:0
connmand[6761]: #11 0x49f3e9d7 in /lib/libglib-2.0.so.0
connmand[6761]: #12 0x49f3de38 in /lib/libglib-2.0.so.0
connmand[6761]: #13 0x49f3e13e in /lib/libglib-2.0.so.0
connmand[6761]: #14 0x49f3e5a6 in /lib/libglib-2.0.so.0
connmand[6761]: #15 0x8053ab6 in main() at nat.c:0
connmand[6761]: #16 0x49cfc4c4 in /lib/libc.so.6
connmand[6761]: +++++++++++++++++++++++++++
Forest Bond [Mon, 16 Jul 2012 13:09:01 +0000 (09:09 -0400)]
gdhcp: Set client state to INIT_SELECTING on restart
With the previous behavior, we would end up sending a discover message
and then ignoring the reply due to the client state mismatch, so a
restart would never result in successful acquisition of a lease.
RFC 2131 specifies that the client should return to INIT state in
various restart scenarios. However, also note that it would not make
sense in any case to send a discover message and ignore the reply.
Forest Bond [Wed, 20 Jun 2012 21:06:28 +0000 (17:06 -0400)]
rtnl: Support PPP connections
The ARP header type for PPP connections is ARPHRD_PPP (512).
Grant Erickson [Mon, 16 Jul 2012 16:48:10 +0000 (09:48 -0700)]
gresolv: Use predefined mnemonics rather than magic numbers
Use predefined mnemonics from arpa/nameserv.h for the return value from
ns_msg_getflag rather than magic numbers.
Grant Erickson [Mon, 16 Jul 2012 16:48:09 +0000 (09:48 -0700)]
gresolv: Do not update successful status with unsuccessful one
When performing a resolver lookup from timeserver or wpad, both perform
queries with an unspecified address family. This means that both A and
AAAA record queries are issued. In cases where a valid, successful A
response comes back but where the AAAA query results in a timeout, do
not smash the successful A status with the time out AAAA status;
otherwise, the timeserver or wpad will appear to fail to them when, in
fact, the A query was successful and more than satisfies its unspecified
address family requirement.
Partial fix for BMC#25486.