Commit graph

2431 commits

Author SHA1 Message Date
Dan Williams
d815cb9f33 logging: fix format string/argument disagreement
Now that the logging fixes make format checking actually work, fix
the issues it brings up.
2011-03-19 12:44:14 -05:00
Dan Williams
e7be98260e logging: clean up logging macros and helper
There were two specific problems with the logging macros:

1) the existing varargs usage didn't allow for format string checking,
which is bad, since it could make logging segfault if the arguments
don't match the format string

2) it didn't allow logging usage without wrapping {}, ie this didn't work:

if (foo)
   nm_log_dbg (...)

blah blah

Fix all that by using the varargs stuff correctly.
2011-03-19 12:42:29 -05:00
Dan Williams
b2036e264c wimax: log a message when connecting to an NSP 2011-03-17 23:58:50 -05:00
Dan Williams
1701df4b75 core: add active connection state DEACTIVATING
Not used yet, but will be when device deactivating state gets
used.  Should be 100% backwards compatible with users that don't
know about it for now.
2011-03-17 14:23:21 -05:00
Dan Williams
23943e7ce2 core: handle new device states
The device states aren't used yet, but handle them anyway.
2011-03-17 13:42:22 -05:00
Dan Williams
8ae173635d trivial: fix comment 2011-03-17 13:42:21 -05:00
Dan Williams
b339074fd9 core: fix up active connection state for newer device states 2011-03-17 13:42:21 -05:00
Dan Williams
11ca4c7a91 supplicant: add 'authenticating' state
Caused a harmless warning, but lets add it anyway.
2011-03-17 13:42:21 -05:00
Dan Williams
05f70e34a5 vpn: cancel the right secrets request
Code still called nm_act_request_cancel_secrets() but the VPN
connection no longers uses the underlying activation request for
secrets.  Fix that.
2011-03-17 13:42:21 -05:00
Mu Qiao
8d60cafd18 ifnet: fix tests and distcheck 2011-03-17 11:05:46 -05:00
Dan Williams
c3dc2a0608 agents: only commit connection changes if there were system secrets
It's pointless to save the connection if there weren't any system
secrets being changed.
2011-03-16 20:53:49 -05:00
Dan Williams
e89a50d779 agents: don't complain when there aren't any system secrets 2011-03-16 20:53:13 -05:00
Mu Qiao
08b2fa5401 ifnet: plugin updates and shared connection support 2011-03-16 15:39:13 -05:00
Dan Williams
da4928aea7 core: allow NULL device path in ActivateConnection 2011-03-15 17:04:35 -05:00
Dan Williams
fcbaa00ed0 core: remove some debug logging 2011-03-15 12:22:54 -05:00
Dan Williams
e64cc4dd7c core: specific object of "/" means NULL
Not sure how this got lost...  but it did.
2011-03-15 11:45:47 -05:00
Dan Williams
b21fc6ed45 wimax: use setting accessor helpers 2011-03-15 10:33:04 -05:00
Kjartan Maraas
b224ab8e56 trivial: let g_free and g_strdup handle NULL (bgo #644763) 2011-03-15 00:08:58 -05:00
Kjartan Maraas
c2b4b9c1d7 build: fix return types (bgo #644665)
Some functions were returning FALSE or 0 instead of NULL.
2011-03-14 01:05:51 -05:00
Kjartan Maraas
6d0d302916 build: do the right thing with config.h (bgo #644664)
1) it shouldn't be included in headers
2) it should be the first thing included in source files
3) it's needed for getting translation right
2011-03-14 01:01:22 -05:00
Dan Williams
af7e26b39a ifcfg-rh: fix reading and writing of Dynamic WEP with EAP-LEAP connections
These are distinct from old-school LEAP (ie, Network EAP) in that
they are standard Dynamic WEP with LEAP as an EAP method and use
open-system authentication.  Old LEAP uses the non-standard LEAP
authentication algorithm.  The config for each is different and thus
we need to make sure we handle both cases.
2011-03-14 00:25:14 -05:00
Dan Williams
eaeb43bfad keyfile: ignore temp files ending with ~ 2011-03-13 23:59:09 -05:00
Dan Williams
32dedf4b83 wifi: fix scanned signal strength calculation for WEXT-based drivers
The new wpa_supplicant D-Bus interface only passes back the 'level'
of the scanned BSS, which with nl80211 drivers is almost always dBm,
which NM handled fine.  But WEXT-based drivers (ipw2x00, other older
ones, and some vendor drivers) use a mix of values for the 'level'
parameter, including the old WEXT 8-bit signed-value-in-unsigned-int
scheme.  Handle that.

Alternatively, we could have the supplicant expose the 'flags' value
from its internal BSS list over the bus.
2011-03-13 12:50:40 -05:00
Dan Williams
0e348bad2a wimax: fix dbus-glib assert when wimax strength overflows
Ran into a case right underneath a tower where the Intel WiMAX SDK
reported signal strength as 32767 which is pretty clearly an overflow
of a 16-bit int somewhere in the WiMAX stack.  Since our internal
NMWimaxNsp GObject property for signal-quality has a max value of
100, dbus-glib got pretty angry when GObject failed to return
something because the property value was out of range, leading
to malformed message and the bus daemon kicking us off, and finally
an assertion somewhere.  Fun.
2011-03-11 18:20:42 -06:00
Michael Biebl
8119ecf6b0 core: depend on stable gudev API 2011-03-10 03:19:27 -06:00
Dan Williams
62ad893a4a core: bring the state() D-Bus method back
It's a lot easier for app developers to use.
2011-03-08 12:57:35 -06:00
Dan Williams
5c260cfe4a keyfile: trivial whitespace fixes 2011-03-08 10:19:01 -06:00
Dan Williams
9a3d762fdf settings: fix some code that required glib 2.26
Should be compatible with at least glib 2.24 and hopefully earlier.
2011-03-08 10:18:29 -06:00
Jiří Klimeš
c207d020f7 core: use nm_connection_get_id() and nm_connection_get_uuid() to simplify code 2011-03-08 13:41:28 +01:00
Jiří Klimeš
78b1260af4 ifcfg-rh: fix testcases now that timestamps are no longer written
Timestamps are no longer written to the connection file itself, but
are kept in a lookaside file in /var to allow for read-only or
stateless /etc and to ease system administration and deployment.
2011-03-08 11:58:00 +01:00
Jiří Klimeš
2e3e171171 ifcfg-rh: get rid of LAST_CONNECT
timestamps are now kept in /var/log/NetworkManager/timestamps
2011-03-08 11:49:49 +01:00
Jiří Klimeš
83d8677963 core: timestamp handling - don't touch /etc when updating timestamps (bgo #637825)
NM updates timestamp for active connections every 5 min. We don't
want to touch files in /etc due to this. This commit solves that
by not updating timestamp in the connection's property. Rather it
updates the timestamp internally. All timestamps are also kept track
of in /var/lib/NetworkManager/timestamps file.
When settings are requested via D-Bus GetSettings(), the proper
timestamp is put in the connection setting before returning.
2011-03-08 10:37:44 +01:00
Michael Biebl
f2f406b316 ifupdown: normalize keys using - or _ (debian #609831) 2011-03-07 09:57:05 -06:00
Daniel Trebbien
96044f7d94 ifupdown: use new logging API 2011-03-05 00:02:43 -06:00
Mikhail Efremov
77eeb95233 settings: don't dereference NULL pointer on request removal 2011-03-05 00:01:01 -06:00
Dan Williams
ffb9457d55 dnsmasq: remove unused variable 2011-03-04 23:56:19 -06:00
Dan Williams
b05b77fa26 dnsmasq: send no config file instead of a bogus one (lp:725041) (debian #615082) 2011-03-04 23:52:44 -06:00
Dan Williams
5bce4adfbb ifcfg-rh: clean up temporary testcase certificates files (fixes distcheck) 2011-03-03 19:15:49 -06:00
Dan Williams
e9165c0f7d trivial: turn off debug messages in /etc/hosts testcase 2011-03-03 18:51:55 -06:00
Dan Williams
9e45d7c8cb ifcfg-rh: test writing TLS secret flags and reading them back in 2011-03-03 18:37:33 -06:00
Dan Williams
e165511ed8 ifcfg-rh: fix handling of private key password secret flags 2011-03-03 17:32:29 -06:00
Dan Williams
bc4d3482e1 ifcfg-rh: handle secret flags
Some adjustments need to be made to read and write secret flags, and
to ensure that connections that don't have system-owned secrets are
still parsed as expected.  testcases for 802.1x connections to come
shortly.
2011-03-03 12:29:37 -06:00
Dan Williams
ecca85066d keyfile: write certificates and keys in a human-readable manner
Write actual paths, not integer lists.  When given certificate or
key blobs, write those out to files and put the path to that file
into the keyfile.
2011-03-03 00:03:44 -06:00
Dan Williams
e3cddc8d9f keyfile: allow paths to be used for certificates and private keys
No reason it should have to be bare byte arrays, ick.
2011-03-02 23:44:27 -06:00
Dan Williams
c9246b53ff ifcfg-rh: fix leak writing out certificate blobs 2011-03-02 23:39:42 -06:00
Dan Williams
eae0e8de61 dhcp: don't quit if we can't find a usable DHCP client (bgo #643575) 2011-03-02 23:33:43 -06:00
Dan Williams
97053ccb8c ifcfg-rh: consolidate variable definition 2011-02-28 17:51:52 -06:00
Dan Williams
781ea63209 ifcfg-rh: clean up EAP-TLS reading 2011-02-28 17:50:32 -06:00
Dan Williams
223b87b28d ifcfg-rh: shorten variable name 2011-02-28 17:47:29 -06:00
Dan Williams
a1a5342f1d ifcfg-rh: get rid of useless EAP reader prototypes
Shuffle stuff around and we don't need these prototypes.
2011-02-28 16:07:36 -06:00