Some stuff we build (the DHCP manager) gets built independently
so that we can use it for unit tests. For that, we need to build
the logging bits separately too, since the independent DHCP
library can't use them if they are embedded in NM.
This ensures that the next time secrets are required, they will
be requested from the settings service. Internally, NM shouldn't
be caching secrets; it should always request them from the settings
service so that the settings service can enforce policy about
password lifetime, if it wants to.
The original OLPC mesh patch confused g_source_remove() with
g_signal_handler_disconnect(), so the signal handler that the
mesh device creates for the managers 'device-added' signal was
getting called after that mesh device instance was freed, which
is of course bad.
After the DUN branch merge (I think?) a number of NM_IS_MODEM
calls were left around which now always return FALSE since
NMDeviceCdma and NMDeviceGsm aren't subclasses of NMModem anymore.
But we still need generic "is this a modem subclass" checks in
a few places, so add a modem base class that both the GSM and
CDMA device classes inherit from and use that.
Plus, we want to consolidate a ton of the common code in
nm-device-gsm.c and nm-device-cdma.c into the base class in the
future anyway.
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.
If the hostname was changed while NM wasn't running, and thus /etc/hosts
was out of sync with the new hostname, NM wouldn't make sure that
the new hostname was mapped in /etc/hosts. Make sure that happens
and add a bunch of testcases for /etc/hosts rewriting.
'--plugins' option now takes preference over [plugins] from config file.
This change fixes a bug caused by accepting default config file when no
'--config' was specified and thus effectively disabling '--plugins'.
