Where we can do so, let's use ifindex since that's actually unique
and doesn't change when the interface name changes. We already use
ifindex in a bunch of places, and netlink *only* uses ifindex, so
this will make it easier later when we move over to ifindexes fully.
The non-blocking connection is really only good for listening for
events. It doesn't work for request/response operations (like
refilling link and address caches) because the message receive loop
in libnl will break out from the EAGAIN before it gets the response
it needs to update the cache with.
This is most evident with link cache refills when requesting the
interface index from the name, or vice-versa; the refill request
exits early with EAGAIN (due to the non-blocking nature of the
connection's socket) and the cache isn't refilled, and the index
lookup fails. We need to use blocking netlink operations in quite
a few places besides index lookups, from address/route operations
to getting the initial device carrier state.
So, split the montior's netlink connection into a non-blocking
event listener connection, and a synchronous connection which gets
used for immediate operations. This also has the effect of
validation the synchronous operations for security, which wasn't
done before in nm-netlink.c (though it wasn't really a problem).
Got the flags wrong in the previous commit; MULTI isn't about
multicast/unicast, but about multi-part packets. Instead we need
to check the netlink sockaddr structure for the group mask.
The RA flags aren't in the link flags, they are in the special
PROTINFO flags that the IPv6 stack sends. To get these, because
libnl doesn't have native support for them, we get to parse the
netlink messages directly. Furthermore, the PROTINFO message
isn't sent unless it's explicitly requested with a
RTM_GETLINK/AF_INET6 message, meaning we get to poll for it
periodically.
So switch over to the netlink monitor object (killing a lot of
duplicate code) and start requesting the PROTINFO bits from
netlink.
We need the IFLA_PROTINFO messages, and these are apparently sent
(still by the kernel) but with our own PID. So the current checks
that limit the netlink PID to zero block these messages out. Instead
(like udev) we should be checking the actual sender of the message
usign unix socket credentials.
Second, at least at this point only privileged processes can send
netlink multicast messages, so as long as the:
* the other end of the socket is UID 0 AND
* the netlink PID is 0 OR
* the message is multicast and sent to our netlink PID
the we accept the message and process it as normal. For another
example of this, see 'netlink.c' from the dhcp6s program; do a
Google search for "IFLA_PROTINFO" to find it.
Sometimes cause the cache not to refill with all interfaces,
meaning NM sometimes got the wrong carrier state from the
kernel which prevented NM from taking over existing connections.
Automatic IPv6 configuration is handled by the kernel, but to
integrate it properly with NetworkManager, we need to watch what the
kernel does to see whether or not it was successful (so that we can
let the user know if there is no IPv6 router present, for example).
NMIP6Manager takes care of this.
Found by Ricardo Salveti de Araujo <ricardo.salveti@openbossa.org>
The link cache was updated immediately, but the carrier state signals
were emitted a lot later, when the cache data was already stale. So
just update the cache at the same time we emit the signals. The
carrier-state-request stuff wasn't originally converted to deferred
for any netlink-specific reason, just to smooth the initial device
creation process in NM.
We really have wanted to use IFF_LOWER_UP since 2.6.17 for carrier
status, since that's tied to netif_carrier_ok()/netif_carrier_on() in
the kernel. See kernel commit b00055aacdb172c05067612278ba27265fcd05ce
for the introduction of IFF_LOWER_UP.
Patch from Tambet Ingo <tambet@gmail.com>
Fix mishandling of netlink error floods (rh #459205, novell #443429, lp #284507)
* src/nm-netlink-monitor.c
- Remove bits for using a non-default GMainContext, which weren't used
- (nm_netlink_monitor_error_handler): don't leak the GError, and report
the actual error code
* src/NetworkManager.c
- (nm_error_monitoring_device_link_state): disconnect error handler when
an error flood occurs to avoid pegging the CPU
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@4334 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Fix Linus' bug in rh #134886
* src/nm-device-802-3-ethernet.c
- (constructor): request initial carrier state
* src/nm-netlink-monitor.c
- (nm_netlink_monitor_request_status): schedule emission of carrier
signals after refilling the link cache. Because the refill is a
synchronous operation, the normal message hander won't get called
since libnl has already consumed the messages.
- (deferred_emit_carrier_state): emit carrier states from an idle handler
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@3669 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Base the NMNetlinkMonitor class on libnl instead of hand-rolled netlink.
* src/nm-netlink-monitor.c
src/nm-netlink-monitor.h
- Remove handrolled netlink, use libnl instead
* src/nm-device-802-3-ethernet.c
- (constructor, nm_device_802_3_ethernet_carrier_off,
nm_device_802_3_ethernet_carrier_on): use new names
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@3181 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Convert to using interface indexes as the primary method of identifying
devices inside NetworkManager. Indexes are (?) stable, but devices can
be renamed at any time. Device object paths now refer to the device
index rather than the name, and you can map those two manually if you like
by looking in the /sys/class/net/<name>/ifindex file. Also moves most
netlink-related code to nm-netlink.c, and cleans up nm-netlink-monitor.c
to use interface indexes rather than names.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2731 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* src/nm-netlink-monitor.c
- (nm_netlink_monitor_class_init, nm_netlink_monitor_event_handler):
don't need the 'wireless-event' signal anymore since that's all
handled by wpa_supplicant
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2727 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* src/vpn-manager/nm-vpn-manager.c: Handle the DBUS state changes itself.
Handle device state changes and disconnect VPN if it's device deactivates.
* src/nm-dbus-nm.c:
* src/nm-dbus-nm.h:
* src/nm-dbus-device.c:
* src/nm-dbus-device.c:
* src/nm-dbus-net.c:
* src/nm-dbus-net.h: Remove. All of it is implemented byt the new dbus API.
* src/NetworkManagerMain.h: Get rid of all but 3 properties of NMData.
* src/nm-device.c (nm_device_get_by_udi):
(nm_device_get_by_iface): Remove. This doesn't belong here and is already
implemented in the correct location (NMManager).
Rip out all the test_device stuff.
* src/NetworkManagerPolicy.c: Remove the leftover activation success and
failure handlers, it's all done by NMDevice already.
* src/NetworkManager.c: Move the signal handling here from nm-logging.c
Remove the iochannel hack to route the unix signals to the main thread since
we're not threaded anymore.
* src/NetworkManagerAP.c: Implement HWAddress property.
* src/NetworkManagerDbus.c: Remove the dbus signal sending code, it happens
automatically with dbus-glib.
* src/nm-netlink-monitor.c:
* src/nm-netlink-monitor.h:
- Move it low in the class hierarchy, don't reference any NM types.
- Remove private data from the header.
- Use type safe checks in public API methods.
- Make it a singleton so we don't have to pass the single reference around.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2339 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
Patch by Jürg Billeter <j@bitron.ch>:
* src/nm-logging.c: Add printf modifier to fix warning on 64-bit
systems.
* src/nm-netlink-monitor.c: Include <net/if.h> instead of <linux/if.h>
as we prefer glibc over kernel headers, if possible.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1641 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Process netlink messages in device subclasses rather than in
NetworkManager.c. Also add support for recognizing Wireless Events.
* configure.in
- Find GLIB_GENMARSHAL
* src/Makefile.am
- Since we're marshalling custom types for wireless event signals,
we get to create our own marshallers using GLIB_GENMARSHAL
* src/NetworkManager.c
- (nm_monitor_wired_link_state): renamed to nm_monitor_setup
- (nm_monitor_setup): renamed from nm_monitor_wired_link_state, and
cut down somewhat. We no longer process signals here.
- (nm_data_new): create the netlink monitor here, and remove a
useless call to nm_policy_schedule_device_change_check()
- (nm_data_free): get rid of the netlink monitor here
- (nm_device_link_activated, nm_device_link_deactivated): removed
- (main): don't create the netlink monitor here, let nm_data_new
do that. Call nm_policy_schedule_device_change_check() right
before we jump to the mainloop to figure out which device
to use first
* src/NetworkManagerSystem.[ch]
- (nm_system_get_rtnl_index_from_iface, nm_system_get_iface_from_rtnl_index):
convert back and forth from interface names to interface
indexes
* src/nm-device-802-11-wireless.c
- (real_init): connect to wireless-event signals from the netlink
monitor object
- (nm_device_802_11_wireless_event): new function, schedule handler
for wireless event signals from the netlink monitor object. We
want the handler to run in the device's context
- (wireless_event_helper): handle wireless-event signals from netlink
- (nm_device_802_11_wireless_dispose): disconnect wireless-event
signal handler
* src/nm-device-802-11-wireless.h
- remove unused prototype for nm_device_802_11_wireless_new
* src/nm-device-802-3-ethernet.c
- (real_init): new function; set up signal handlers for link events
- (nm_device_802_3_ethernet_link_activated): new function, schedule
handler for netlink link activated events on device's main loop
- (link_activated_helper): when we get a link activated event, set
the device's link to be active
- (nm_device_802_3_ethernet_link_deactivated): new function; schedule
handler for netlink link deactivated events on device's main loop
- (link_deactivated_helper): when we get a link deactivated event, set
the device's link to be inactive
- (nm_device_802_3_ethernet_dispose): disconnect signal handler on
dispose
* src/nm-device-802-3-ethernet.h
- remove unused prototype for nm_device_802_3_ethernet_new
* src/nm-device.[ch]
- (nm_get_device_by_iface_locked): variant of nm_get_device_by_iface
but locks the device list
- (nm_device_set_active_link): a little bit of cleanup and de-indenting
* src/nm-netlink-monitor.[ch]
- (nm_netlink_monitor_class_install_signals): New signal
"wireless-event"
- (nm_netlink_monitor_new): keep reference to NMData so we can get
at the device list
- (nm_netlink_monitor_event_handler): expand for wireless events too
* src/nm-marshal-main.c
- Include generated nm-marshal.c and nm-marshal.h
* src/nm-marshal.list
- List of custom marshal functions
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1555 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
Patch from Bill Moss <bmoss@clemson.edu>
* Consolidate writes of access point information updates to the info daemon
so that we only do it when the connection to the access point was
successful. Also consolidates updates to GConf in the Gnome applet.
* src/nm-netlink-monitor.c
- Silence compile warning when calling g_object_new()
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@830 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
* src/nm-netlink-monitor.c (nm_netlink_monitor_new):
remove unneeded NULL arg from g_object_new(). Any
warnings caused by not having the extra NULL are just a
result of a bug in glib 2.7.0 - 2.7.2.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@809 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
