Previously the tests required that a NMDHCPClient object was
created, but all that's really being tested is the DHCP option
parsing and that doesn't need a client object now that the
option parsing has been split out.
Also, the options test wasn't in src/dhcp-manager/, so put it there.
Lastly, the test ran for both dhclient and dhcpcd, but the code
executed for both cases was the same, so there's no reason to keep
running the test for both clients.
Instead of creating most routes with metric 0 and then fixing them
just before applying them, create the routes with the correct metric
in the first place (so that NMIP4Config and NMIP6Config don't have to
try to guess whether "metric 0" means "unset" or "actually metric 0").
It's only used to keep the DHCPManager up-to-date with hostname changes,
and that can be accomplished in much less code by just having NMManager
set a hostname on the DHCPManager itself.
At a later point, we will have to make a copy of @dhcp_anycast_addr to start
the client asynchronously. Although the length of the guint8 array *should*
always be 6 byte (being a MAC address), it's nicer to just pass on the
GByteArray instance instead, which knows how many byte are actually
set.
Signed-off-by: Thomas Haller <thaller@redhat.com>
Add hidden command line option --run-from-build-dir; with that, helpers
like nm-avahi-autoipd.action and nm-dhcp-helper will be called from the
build tree instead of libexecdir, which allows testing without having to
install first.
Helper paths are now stored in global variables instead of macros, and
get modified with that new option.
https://bugzilla.gnome.org/show_bug.cgi?id=698752
Rather than passing specific bits of data to NMDHCPManager and
NMDnsManager, just let them call nm_config_get() and then get the data
themselves.
Also, remove the GError argument from nm_dhcp_manager_new(), since the
function never returned NULL. This in turn means there is no longer
any need for a distinction between nm_dhcp_manager_new() and
nm_dhcp_manager_get(), so remove the former.
DHCP lease file names are built by NetworkManager and contain connection UUID
which makes them NM-specific. Their new location belongs to NetworkManager and
doesn't have to be guessed. With no guessing, we don't need distribution-specific
conditionals.
Note: This change may require modifications to the selinux policy. But after all
these files actually belong to NetworkManager as well as the instance of dhclient
that uses them.
When NM was registering all of its enum types by hand, it was using
NamesLikeThis rather than the default names-like-this for the "nick"
values. When we switched to using glib-mkenums, this resulted in
dbus-glib using different strings for the D-Bus error names, causing
compatibility problems.
Fix this by using glib-mkenums annotations to manually fix all the
enum values back to what they were before. (This can't be done in a
more automated way, because the old names aren't 100% consistent. Eg,
"UNKNOWN" frequently becomes "UnknownError" rather than just
"Unknown".)
Rather than generating enum classes by hand (and complaining in each
file that "this should really be standard"), use glib-mkenums.
Unfortunately, we need a very new version of glib-mkenums in order to
deal with NM's naming conventions and to fix a few other bugs, so just
import that into the source tree temporarily.
Also, to simplify the use of glib-mkenums, import Makefile.glib from
https://bugzilla.gnome.org/654395.
To avoid having to run glib-mkenums for every subdirectory of src/,
add a new "generated" directory, and put the generated enums files
there.
Finally, use Makefile.glib for marshallers too, and generate separate
ones for libnm-glib and NetworkManager.
If your distributor forgot to set up the client path for you, this
makes NM look in the right places and allows you to use either client
if it's installed, no matter what NM was configured with.
Since the same interface could be used for both DHCPv4 and DHCPv6 we
can't just use 'iface' for tracking DHCP client lease changes. Instead
use a generated client ID, and track DHCP events based on the client's
PID instead of interface name.
Based on patches by Johan Bilien <jobi@via.ecp.fr>,
nick loeve <trickie@gmail.com>, and Roy Marples <roy@marples.name>
with significant changes for dhclient formatting and test cases.
Note that dhclient needs help before it can actually parse
classless static routes by adding the following to the
dhclient.conf file:
option rfc3442-classless-static-routes code 121 = array of unsigned integer 8;
Ensure zombie children get cleaned up. To get notifications when children
die abnormally, g_spawn_async() requires G_SPAWN_DO_NOT_REAP_CHILD, but
that requires calling waitpid() yourself if you've removed the child watch
handler before the process has actually died, which NM needs to do in a few
places. So ensure that everything uses G_SPAWN_DO_NOT_REAP_CHILD and also
cleans up after the child when required. Should fix problems trying to
activate mobile broadband connections after a previous failure.
* src/dhcp-manager/nm-dhcp-dhclient.c
src/dhcp-manager/nm-dhcp-dhcpcd.c
- Use G_SPAWN_DO_NOT_REAP_CHILD
* src/dhcp-manager/nm-dhcp-manager.c
- (nm_dhcp_device_destroy): ensure child is cleaned up
- (nm_dhcp_client_stop, nm_dhcp_manager_cancel_transaction_real): always
block on child quitting, since the non-blocking functionality was
never actually used
* src/dnsmasq-manager/nm-dnsmasq-manager.c
- (dm_watch_cb): child is already reaped here
- (ensure_killed, nm_dnsmasq_manager_stop): block until child is dead
* src/nm-device.c
- (aipd_cleanup): block until child is dead
* src/named-manager/nm-named-manager.c
- (run_netconfig): don't use G_SPAWN_DO_NOT_REAP_CHILD if we aren't
event bothering to watch the child
* src/ppp-manager/nm-ppp-manager.c
- (ppp_watch_cb): child is already reaped here
- (ensure_killed, nm_ppp_manager_stop): block until child is dead
* src/vpn-manager/nm-vpn-service.c
- (vpn_service_watch_cb): child is already reaped here
- (nm_vpn_service_daemon_exec): use G_SPAWN_DO_NOT_REAP_CHILD so that
status of the child is actually tracked
- (ensure_killed, finalize): block until child is dead
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@4020 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* introspection/Makefile.am
introspection/nm-device.xml
introspection/nm-dhcp4-config.xml
- Add bits for the DHCP4Config property of the device, and the DHCP4Config
itself
* src/nm-device-interface.c
src/nm-device-interface.h
- Add the DHCP4Config property
* src/nm-device.c
- Keep track of DHCP4 options via a new DHCP4Config property and notify
D-Bus clients when it changes
* src/nm-dhcp4-config.c
src/nm-dhcp4-config.h
- Simple object to store DHCP4 options, export them over D-Bus, and
notify when they change
* src/dhcp-manager/nm-dhcp-manager.c
src/dhcp-manager/nm-dhcp-manager.h
- (nm_dhcp_manager_set_dhcp4_config, copy_dhcp4_config_option): copy and
filter server-returned DHCP options into an NMDHCP4Config object
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@3829 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Patch from Roy Marples <roy@marples.name>
* configure.in
- Add --with-dhcp-client option
* src/dhcp-manager/Makefile.am
- pass DHCP_CLIENT_PATH on compile line
* src/dhcp-manager/nm-dhcp-manager.c
src/dhcp-manager/nm-dhcp-manager.h
- Genericize for both dhcpcd and dhclient
* src/dhcp-manager/nm-dhcp-dhclient.c
- Move dhclient stuff out to it's own file from nm-dhcp-manager.c
* src/dhcp-manager/nm-dhcp-dhcpcd.c
- Implement support for dhcpcd too
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@3823 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* libnm-util/nm-setting-ip4-config.c
libnm-util/nm-setting-ip4-config.h
- Add properties for DHCP Client Identifier and DHCP Hostname
* src/dhcp-manager/nm-dhcp-manager.c
src/dhcp-manager/nm-dhcp-manager.h
- (nm_dhcp_manager_begin_transaction): take the connection's ip4-config
setting as an argument to pass on to the dhclient config file
creation function
- (nm_dhcp_manager_cancel_transaction_real): remove dhclient config when
DHCP is torn down
- (dhclient_run): punt config file handling to create_dhclient_config()
- (create_dhclient_config): create an interface-specific dhclient
config file since there may need to be interface-specific options
passed to dhclient
- (merge_dhclient_config): merge normal distro dhclient config file and
add options from the connection
- (nm_dhcp_device_new): generate the interface specific dhclient
config file path once
- (nm_dhcp_device_destroy): handle partially initialized objects; free
dhclient config file path
* src/nm-device.c
- (real_act_stage3_ip_config_start): pass ip4-config, if any, to the
DHCP manager when starting DHCP
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@3805 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Kill dhcdbd until it's dead, dead, dead. Based on a patch from
Robert Frank <rfrank@redhat.com>
* src/dhcp-manager/nm-dhcp-manager.c
src/dhcp-manager/nm-dhcp-manager.c
src/nm-device.c
- Spawn and communicate with dhclient directly, through means of a
custom dhclient callout script. Process callout D-Bus signals
with dbus-glib instead of hand-rolled dbus. DHCP timeouts are now
sent via gobject signals rather than being driven by the dhcp manager
directly.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2666 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* src/dhcp-manager/nm-dhcp-marshal-main.c: Add.
* src/dhcp-manager/nm-dhcp-marshal.list: Add.
* src/dhcp-manager/nm-dhcp-manager.c:
* src/dhcp-manager/nm-dhcp-manager.h:
- Convert it to GObject since we need to signal state changes.
- Remove all references to other NM classes, this class is one
of the lowest classes in our hierarchy.
- One less class to use NMActRequest.
- Make it singleton, one less user of NMData.
- Remove a couple of sleep() calls.
- Convert a bunch of low-level dbus API calls to dbus-glib calls.
One less class to use the NM's custom tailored signal handlig.
* Makefile.am: Generate marshallers, add them to build.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2283 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Threading removal related cleanups:
- Use the glib default main context. Remove the device main context
member from NMDevice, and the main_context member from NMData. Change
all the idle and timeout scheduler functions to use plain
g_idle_add() and g_timeout_add().
- As a side-effect of the first change, nm_dbus_manager_get() no longer
takes an argument; fix that up too.
- Remove all locking, which is useless since we no longer use threads. For
example, nm_get_device_by_iface_locked() has been removed. The global
device list lock, the AP List lock, and all static locks in
NetworkManagerPolicy.c have been removed. The locking utility functions
in NetworkManagerUtils.c have also been removed.
- Other cleanups in spacing and code style
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2205 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* Huge DBus refactor:
- Create a "DBus Manager" object which manages the connection and
sends signals on NameOwnerChanged and connection/disconnection events,
handles reconnection to the bus if NM gets kicked off, and abstracts
signal handling
- Remove DBusConnection members from places where they are no
longer needed due to the refactor, like the dbus-connection
property of the named manager, and from NMData
- Reformats a bunch of the code to gnome style
(8-space tabs, braces on same line as statement, 80-col width).
Consider it open season to reformat any bits to gnome style.
style that aren't already.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2061 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* src/dhcp-manager/nm-dhcp-manager.c: Use preprocessor defines and not
open-coded integer constants. Add state_to_string() to map a given
state to a textual description, and provide that when notifying of
state change.
* src/dhcp-manager/nm-dhcp-manager.h: Provide defines for the dhcdbd
states, copied and cleaned up from dhcdbd.d. Ideally, we would use
this header directly, but it is currently not installed on most
systems.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1559 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* src/nm-activation-request.c
- Change dhcp_state member of the NMActRequest structure
from guint8 to guint32
* src/dhcp-manager/nm-dhcp-manager.[ch]
- (nm_dhcp_manager_get_state_for_device): return guint32 rather
than guint8 to match the dbus argument. Turns out we were
overwriting memory since we were passing in only a guint8
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1245 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* refactor NMDevice into a GObject-based framework with separate
objects for wired and wireless. The following files are no
longer used but should stick around for a bit so we don't
loose code through the cracks:
NetworkManagerDevice.c
NetworkManagerDevice.h
NetworkManagerWireless.c
NetworkManagerWireless.h
The intent here is to allow each device type to manage its own
connection & activation life-cycle, ie to allow wireless devices
to interface with wpa_supplicant, etc. There's a fair bit of
encapsulation breakage right now that should gradually get pulled
back into each device, along with things like periodic property
updates and link probing.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1244 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* Kill dhcpcd. We now use "dhcdbd", a dbus daemon that controls dhclient.
This means that NetworkManager shouldn't have DHCP issues anymore. It also
means you need dhcdbd, which you can get here (get the latest one):
http://people.redhat.com/jvdias/dhcdbd/
Technically NetworkManager can use any DHCP daemon that uses the same DBUS
interface as dhcdbd.
* Rewrite device activation to facilitate the new DHCP infrastructure and
future improvements. Its now "activation request" based, ie there is a single
activation request composed of the device, access point, and other info which
follows the entire activation process. There are 5 stages of the activation
process which correspond to:
1) Device preparation
2) Device configuration (bring it up, set ESSID/Key/etc)
3) IP Config Start (fire off DHCP if we're using it)
4) IP Config Get (grab config from DHCP or static config files)
5) IP Config Commit (set device's IP address, DNS, etc)
Note that there is no longer a "scanning" step, since the access point must
be known _before_ activation starts. If the access point drops out or does
not exist for some reason, the entire activation process fails and must be
restarted for a different access point or device.
Patch from Bill Moss:
* gnome/applet/applet.c
- Fix type of vpn_failure dialog -> vpn_banner dialog
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@597 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
