summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* platform/core: add support for IPv4 PtP/peer addressesdcbw/ptp-addressDan Williams2013-10-159-19/+56
| | | | | | | | | | | | | | OpenVPN's 'p2p' topology does send a different peer address than the local address for tunX, plus the server may also push routes that use the peer address as the next hop. Unless NetworkManager configures the tunX interface with the peer address, the kernel refuses to add the routes with the peer nexthop, because the kernel has no idea how to talk to the peer, because the peers address is not reachable over any routes (none have yet been added) and the peer's address is not assigned to any interface yet. OR: could we just add a host route through the tunX pointing to the peer address, and leave the PtP peer address unset?
* core: print ifindex when logging UDEV eventThomas Haller2013-10-152-4/+11
| | | | Signed-off-by: Thomas Haller <thaller@redhat.com>
* core: don't have IP4 and IP6 configs on slavesDan Winship2013-10-145-97/+126
| | | | | | | | | | | | | | | Although it's convenient in some places to have IP configs on all connections, it makes more sense in other places to not have IP configs on slaves. (eg, it's confusing for nmcli, etc, to report a full NMSettingIP4Config on a slave device). So revert parts of the earlier patch. However, it's still safe to assume that s_ip4 != NULL if method != DISABLED, so some of the earlier simplifications can stay. Also, add nm_utils_get_ip_config_method(), which returns the correct IP config method for a connection, whether the connection has IP4 and IP6 settings objects or not, and use that to keep some more of the simplifications from the earlier patch.
* trivial: whitespace fixThomas Haller2013-10-141-17/+17
| | | | Signed-off-by: Thomas Haller <thaller@redhat.com>
* nm-online: check return value of nm_client_new()Jiří Klimeš2013-10-141-1/+6
|
* libnm-glib: fix a crash in nm_client_new() (rh #1010288)Jiří Klimeš2013-10-141-5/+30
| | | | | | | | We have to check if 'client' is valid when calling _nm_object_ensure_inited(). Creation of NMClient object can fail, because its parent NMObject's constructor() returns NULL for D-Bus errors. https://bugzilla.redhat.com/show_bug.cgi?id=1010288
* core: fix crash for bridge-slave with missing NMSettingBridgePort settingThomas Haller2013-10-111-11/+18
| | | | Signed-off-by: Thomas Haller <thaller@redhat.com>
* ifcfg-rh: fix handling of minimal ifcfg filesDan Winship2013-10-112-71/+50
| | | | | | | | | ifcfg-rh had the rule that if an ifcfg file had no BOOTPROTO and no IPv4 addresses, then it should be treated as method=auto for compatibility. But in fact, current ifup treats it as method=disabled, so we should too. https://bugzilla.gnome.org/show_bug.cgi?id=708875
* settings: make connections always have s_ip4 and s_ip6Dan Winship2013-10-1112-189/+302
| | | | | | | | | | | | | | | | Make sure that all connections returned from NMSettings or created via AddAndActivateConnection have an NMSettingIP4Config and an NMSettingIP6Config, with non-NULL methods, and get rid of now-unnecessary checks for those. Also move the slaves-can't-have-IP-config checks into the platform-independent code as well. This also gets rid of spurious "ignoring IP4/IP6 configuration" warnings in ifcfg-rh when reading a slave ifcfg file. Partly based on a patch from Pavel. https://bugzilla.gnome.org/show_bug.cgi?id=708875
* libnm-util: remove reference to non-existent "Posix" namespaceDan Winship2013-10-111-4/+1
| | | | | There's no way in gobject-introspection to annotate something as returning an arbitrary C struct type, so don't try.
* libnm-util, libnm-glib: fix up some gtk-doc commentsDan Winship2013-10-1110-46/+49
| | | | | | | | | gtk-doc recognizes that #NMFoos is the plural of #NMFoo now, so you don't need to put an empty comment between the type name and the "s" to make it work. (Unfortunately, it's not smart enough to realize that "NMIP4Addresses" is the plural of "NMIP4Address".) Also, add some missing "#"s noticed along the way.
* core: fix PropertiesChanged signals for IP-related propertiesDan Williams2013-10-091-16/+30
| | | | | | | | | | | | | | | | | | To present a consistent API to clients, the IP-related properties are only valid when the device has finished IP configuration. But they are set before that happens, and their change notifications were emitted before the IP configuration was considered valid. Re-emit the change notifications when the device enters the IP_CHECK state (and thus has IP configuration) and also when the device deactivates to enusre clients have up-to-date IP-related property information. For the changes to has_ip_config(), the priv->ipX_state checks are not necessary since the device will have valid IP configuration when it enters the IP_CHECK state. The other checks can be consolidated into a single statement. Acked-by: Dan Winship
* trivial: remove unnecessary warningDan Williams2013-10-091-1/+0
|
* core: avoid use-after-free in libnm-util/bond.verify()Thomas Haller2013-10-081-2/+2
| | | | Signed-off-by: Thomas Haller <thaller@redhat.com>
* po: updated Brazilian Portuguese (pt_BR) translation (bgo #709675)Enrico Nicoletto2013-10-081-355/+505
| | | | | | https://bugzilla.gnome.org/show_bug.cgi?id=709675 Signed-off-by: Thomas Haller <thaller@redhat.com>
* core: allow IPv4 to proceed if IPv6 is globally disabled but set to "auto" ↵Dan Williams2013-10-071-2/+2
| | | | | | | | | | (rh #1012151) If the user disabled IPv6 support in the kernel with "ipv6.disable=1" on the kernel boot line, then any attempts to open IPv6 sockets (which libndp does) will fail. This failed the entire connection, even if IPv6's "may-fail" property was TRUE. Instead, just fail IPv6 and allow IPv4 to proceed. If IPv4 fails or is disabled, then other logic will fail the entire connection.
* ifcfg-rh: ignore default routes in route6 file (rh #991807)Jiří Klimeš2013-10-072-4/+13
| | | | | | | Base on patch from Francesco Prelz <Francesco Prelz mi infn it>: https://mail.gnome.org/archives/networkmanager-list/2013-January/msg00095.html https://bugzilla.redhat.com/show_bug.cgi?id=991807
* ifcfg-rh: fix ignoring updates that don't change anythingDan Williams2013-10-041-8/+3
| | | | | | | | | | connection_from_file() requires the 'error' parameter. Not passing a valid 'error' parameter causes the function to fail and return NULL, which mean that commit_changes() would always re-write the connection instead of ignoring commits where nothing has actually changed. connection_from_file() no longer requires the unmanaged, keyfile, or routefile parameters, so remove them.
* test: launch dbus for test-remote-settings-clientThomas Haller2013-10-041-1/+5
| | | | | | | | | | | | | | | | Running `make check` on systems without running dbus failed in test-remote-settings-client.c:383 make[4]: Entering directory `/tmp/NetworkManager/libnm-glib/tests' /tmp/NetworkManager/libnm-glib/tests/test-remote-settings-client /tmp/NetworkManager/libnm-glib/tests test-remote-settings-service.py ** (/tmp/NetworkManager/libnm-glib/tests/.libs/lt-test-remote-settings-client:26983): WARNING **: Error connecting to D-Bus: Unable to autolaunch a dbus-daemon without a $DISPLAY for X11 make[4]: *** [check-local] Trace/breakpoint trap (core dumped) Modify the Makefile to start the dbus-daemon, if it is not yet running. Signed-off-by: Thomas Haller <thaller@redhat.com>
* bluez: fix leak of NMBluezDevice in bluez_connect_cbThomas Haller2013-10-031-1/+3
| | | | Signed-off-by: Thomas Haller <thaller@redhat.com>
* core: fix nm_ip6_config_replace() nameserver address comparisonDan Williams2013-10-031-2/+2
|
* trivial: print route prefix when printing RA discovered routesDan Williams2013-10-031-1/+1
|
* rdisc: mask host bits off RA prefix (rh #1008104) (bgo #709230)Dan Williams2013-10-031-2/+21
| | | | | | | | | | | | | | | | | | | | | | Some RA implementations (like radvd) dump whatever the user configures onto the wire, accepting a prefix of "2001:db8:1:0::1/64" without masking the host bits off. This causes NetworkManager to send that route down to the kernel, which *does* mask the host bits off. This causes a mismatch between the route NetworkManager expects the kernel to create, and what the kernel actually created, when searching for the kernel object in the platform's refresh_object() function: cache = choose_cache (platform, object); cached_object = nl_cache_search (choose_cache (platform, object), object); kernel_object = get_kernel_object (priv->nlh, object); kernel_object is NULL since 'object' (a route which came from the RA prefix) is not the same as the object the kernel actually did create. Ensure we match kernel behavior by fixing up prefixes for dumb router advertisement services.
* cli: fix a possible crash when casting NULL using NM_CONNECTION (rh #1011942)Jiří Klimeš2013-10-031-2/+7
| | | | https://bugzilla.redhat.com/show_bug.cgi?id=1011942
* libnm: more valgrinding fixesDan Winship2013-10-036-7/+20
|
* libnm-glib: fix the leak that breaks the shell network status iconDan Winship2013-10-031-7/+5
| | | | | | | | | | | | | All NMObjects created in response to property changes were getting leaked, which in particular included all NMAccessPoint objects, which meant that after disconnecting and reconnecting a wifi interface some number of times (depending on how many access points were in the area), gnome-shell would be watching so many D-Bus AccessPoint objects (most of which didn't exist any more) that it would hit dbus-daemon's limit on the number of match rules you can register, which meant that NMActiveConnections created after that point wouldn't be able to register for PropertiesChanged notifications, which meant that the network status icon would get out of sync.
* cli: enhance TAB-completion in nmcli editorJiří Klimeš2013-10-031-64/+150
| | | | | | | | | | | - completion now works for command shortcuts too nmcli> g ethe<TAB> - completion now allows leading spaces nmcli> desc wifi.ss<TAB> - fix completion of settings when property is already there nmcli> set con<TAB>.id - fix completion of properties in setting.property nmcli> des con.inter<TAB> bla
* cli: fix bash completion to show general optionsThomas Haller2013-10-031-2/+2
| | | | | | | | | | This fixes an error in the following example: $ nmcli con add type bridge con-name test-bridge <TAB> Before, general options such as 'autoconnect' and 'ifname' were wrongly not suggested. Signed-off-by: Thomas Haller <thaller@redhat.com>
* bluez: fix creation of NAP connection for bluez deviceThomas Haller2013-10-021-1/+4
| | | | Signed-off-by: Thomas Haller <thaller@redhat.com>
* team: do not g_source_remove after g_child_watch_add signal was emittedThomas Haller2013-10-021-0/+3
| | | | | | | | g_child_watch_add only emits one signal and gets automatically removed from the source. No need for double removal, so unset teamd_process_watch before calling teamd_cleanup. Signed-off-by: Thomas Haller <thaller@redhat.com>
* team: fix missing teamd_cleanup in certain casesThomas Haller2013-10-021-17/+14
| | | | | | Also ensure that teamd_timeout_cb does a state transition of the device. Signed-off-by: Thomas Haller <thaller@redhat.com>
* team: fix startup of team by ignoring teamd_dbus_vanished at firstThomas Haller2013-10-021-0/+12
| | | | | | | | g_bus_watch_name immediatly emits a vanished signal, if the name does not exist at the begining. So the first signal, does not indicate an actual change of state and must be ignored. Signed-off-by: Thomas Haller <thaller@redhat.com>
* team: fix a crash on team devices (rh #1013593)Jiří Klimeš2013-10-021-1/+1
| | | | | | | | | | | teamd was being watched using g_bus_watch_name(). But when the NM team device was destroyed NM tried to remove the watcher with g_source_remove() instead of g_bus_unwatch_name(). Thus the watcher was not removed and teamd_dbus_appeared() was called on disposed device. https://bugzilla.redhat.com/show_bug.cgi?id=1013593#c28 Signed-off-by: Thomas Haller <thaller@redhat.com>
* team: ensure that teamd_start does not get called when already startingThomas Haller2013-10-021-8/+20
| | | | Signed-off-by: Thomas Haller <thaller@redhat.com>
* trivial: whitespace fixDan Williams2013-10-021-0/+1
|
* cli: add primary bonding option in bash completionThomas Haller2013-10-021-1/+2
| | | | Signed-off-by: Thomas Haller <thaller@redhat.com>
* trivial: remove stray ')' from nmcli prompts in questionsJiří Klimeš2013-10-021-13/+13
|
* cli: accept 'primary' bonding optionJiří Klimeš2013-10-022-2/+33
|
* cli: enhance questionnaire for bond connections (rh #1007355)Jiří Klimeš2013-10-021-49/+66
| | | | | | | | | | | | | | | | | | The session now looks like this: There are optional arguments for 'bond' connection type. Do you want to provide them? (yes/no) [yes] Bonding mode [balance-rr]: Bonding monitoring mode (miimon or arp) [miimon]: miimon Bonding miimon [100]: Bonding downdelay [0]: Bonding updelay [0]: ...when answered arp... Bonding arp-interval [0]: Bonding arp-ip-target [none]: https://bugzilla.redhat.com/show_bug.cgi?id=1007355
* cli: fix memory leaks on questionnairesJiří Klimeš2013-10-021-63/+158
|
* dhcp: fix IPv6 address prefixes (rh #1013583)Dan Williams2013-10-011-1/+1
| | | | | | 53e55aab365a263df5d2e13efe28a5dc5a7b8432 mistakenly moved the address prefix without moving the memset() to initialize the address. Make sure the address is initialized before trying to do anything with it.
* libnm-glib: make properties-changed debugging available at runtimeDan Williams2013-10-011-14/+22
| | | | | | Use an environment variable LIBNM_GLIB_DEBUG=properties-changed to indicate that properties-changed debugging messages should be printed. Also cleans up the debug output formatting.
* libnm-util: only remove ARP when MIIMON is being set to non-zero and vice versaJiří Klimeš2013-10-011-2/+4
| | | | | miimon or arp_interval == 0 means 'disable'. So we should not remove options of one mode when the other mode is actually being disabled (set to zero).
* team: remove variable teamd_on_dbus that was always FALSEThomas Haller2013-10-011-5/+1
| | | | | | | The variable teamd_on_dbus in nm-device-team was never set to TRUE. Remove it. Also, before teamd_dbus_vanished did never cleanup anything. Signed-off-by: Thomas Haller <thaller@redhat.com>
* team: fix wrong assertion in device-teamThomas Haller2013-10-011-3/+7
| | | | | | | | | | | | The assertion that priv->teamd_pid is set does not hold. For example, if no teamd binary can be found, "Stage 1 of 5 (Device Prepare)" fails with "Activation (nm-team) to start teamd: not found". This causes the device state to become "failed" and later "disconnected", which in turn calls teamd_stop without a valid teamd_pid set. Signed-off-by: Thomas Haller <thaller@redhat.com>
* team: chain up parent dispose() in NMDeviceTeam dispose() (rh #1013593)Jiří Klimeš2013-10-011-0/+2
| | | | | | | | | NMDeviceTeam dispose() did not chain up to parent dispose(). This prevented parent dispose() to be called (only finalize()) and thus link_changed_cb() and device_ip_changed() handlers were not disconnected. Later these handlers were called on invalid device resulting in a crash. https://bugzilla.redhat.com/show_bug.cgi?id=1013593
* cli: fix missing type option 'team' in mcli completionThomas Haller2013-09-301-1/+1
| | | | Signed-off-by: Thomas Haller <thaller@redhat.com>
* cli: use a generic function asking for both IPv4/IPv6 in questionnaireJiří Klimeš2013-09-301-39/+33
|
* cli: be more verbose when adding IP addresses in questionnaire (rh #1006450)Jiří Klimeš2013-09-301-8/+22
| | | | | | | | | | | | | | | | | Now the session looks like this: Do you want to add IP addresses? (yes/no) [yes] Press <Enter> to finish adding addresses. IPv4 address (IP[/plen] [gateway]) [none]: 10.0.0.22/24 10.0.0.1 Address successfully added: 10.0.0.22/24 10.0.0.1 IPv4 address (IP[/plen] [gateway]) [none]: 10.0.0.22/24 10.0.0.1 Warning: address already present: 10.0.0.22/24 10.0.0.1 IPv4 address (IP[/plen] [gateway]) [none]: 192.168.1.1 abc Error: invalid gateway 'abc' IPv4 address (IP[/plen] [gateway]) [none]: IPv6 address (IP[/plen] [gateway]) [none]: abcd::1234 ::2 blabla Address successfully added: abcd::1234 ::2 Warning: ignoring garbage at the end: 'blabla' IPv6 address (IP[/plen] [gateway]) [none]:
* rdisc: do not crash on NDP init failures (rh #1012151)Jiří Klimeš2013-09-302-7/+16
| | | | | | | | ndp_open() can fail, do not assume it cannot. When 'ipv6.disable=1' kernel parameter is used, the whole IPv6 stack is disabled and the attempt to create PF_INET6 socket fails. https://bugzilla.redhat.com/show_bug.cgi?id=1012151
View Lorry Controller Status