GitLab

The signal was emitted in case the removed connection was managed instead of
for unmanaged connection. Thus the signal had no effect.
That caused incorrect behaviour in case of changing NM_CONTROLLED=no to yes.
That didn't enable the device; only after the file was changed for the second time.

We need to remember to clear out keys that aren't getting used since
we're not rewriting the whole ifcfg file.

At some point we'll be passing other info like whether we need
the 802.1x identity too, or unknown CA certificate data for the
Agent to accept, etc.  Basically state that instead of only
hints from the setting, we can pass other stuff as well.

Make sure the dispose won't run twice for the same code and
make sure we never schedule a handler for monitor_cb() more
than once, though it's really hard to see how that could ever
happen anyway.

Another attempt to blindly fix lp:752143

While this should never happen while the PPP manager is alive, modems
can switch their IP method while alive, since the net port is sometimes
discovered after the serial ports have been.  This happens for some
devices that have separate drivers for the net and serial sides, like
ZTE Icera-based devices (cdc-ether and cdc-acm) and newer Sierra
devices (sierra and sierra-net).  Just be paranoid here and ensure
that the PPP manager gets cleaned up.

Partial attempt at fixing lp:752143

Pointless, since we're not going to activate it anyway when it's
in one of those states.

Instead of having new/delete code sprinkled around, consolidate it.

The default wired connection should own a reference to the device
it's made for, but that got dropped in
78df8c49, which used to use a
set_property handler with g_value_dup_object() which obviously
increments the reference count.  But that ref got dropped when
the object initialization was simplified.

If any ethernet devices were left up (because we can assume control
over them seamlessly when NM starts up again) make sure we write
out a usable resolv.conf for the device on shutdown, otherwise the
users networking is broken with an empty resolv.conf.  This only
happened when DNS plugins were active, in which case the user
would be left with a localhost-pointing resolv.conf but no
local caching nameserver running since NM shut it down when NM
terminated.

We've got types for this stuff in NetworkManager.h, let's use them.

Casting to (void) still doesn't do it, so just assign it to
something and ignore it, in conjunction with
-Wno-unused-but-set-variable

These interfaces are a proprietary USB-ethernet-style virtual interface
that of course does not have proper driver links.  Given that it's so
easy to support, just do it.

Due to a stupid error the cache wasn't actually getting used and
way more than one SessionMonitor was getting created.

Now that initscripts also support IPADDRn syntax, update the implementation
to match the intitscripts' one (see rh #633984)
Basically, writer produces IPADDR0 .. IPADDR255. reader is more tolerant and
supports older configs too: IPADDR, IPADDR0, IPADDR1 could be missing, from
IPADDR2 up the indexes have to be contiguous.

Since the user state stuff got committed in 0.8.2, WWAN enable
state has been somewhat broken.  The problem is that we want two
things: (1) that the current modem enabled state is reflected
in the WwanEnabled property, and (2) that enabled state should not
affect the user's ability to enable the modem via the UI.

The code did not properly separate these two.  For all automatic
decisions and properties (ie the WwanEnabled property, setting the
initial enabled state on startup or hotplug, etc) the ModemManager
enabled state should be respected.  But the user should be able
to override that state by turn WWAN on.

This calls for a fourth enabled check that modems have, the 'daemon'
state, distinct from the hardware and software kernel rfkill states
and from the user's chosen enabled/disabled state.  Add that new
check.

The actual problem was in manager_radio_user_toggled() where after
updating the user enabled state, new_enabled still equaled
old_enabled, because the kernel rfkill state was a combination of
both the kernel rfkill state *and* the ModemManager enabled state,
so the manager_update_radio_enabled() call would never happen and
the modem would never become enabled as a result of a user request.

It's an acronym and should thus be capitalized.

It's not required for anything.

Pointless to get/unref it when we already have a pointer to it
sitting around in the private data.

Make sure we catch everything before generating a tarball for release.