If the platform signaled that the external configuration changed (and
thus update_ipX_config() is scheduled) and we are doing a commit of
the new configuration, update priv->ext_ipX_config. Without this, the
commit will remove addresses added externally but not yet captured in
the external configuration.
https://bugzilla.redhat.com/show_bug.cgi?id=1449873
This basically reverts commit 31fe84e467 "core: Add host route for
DHCP4 server if outside assigned subnet (bgo #721767)" because the
additional route added by NM does more harm than good.
First, the code does not consider routes pushed by the server and thus
it can add a route conflicting with the ones from the network
administrator.
Second, there is no specification on what a DHCP client should do when
the server is not reachable via unicast, and adding arbitrary logic
into the client is likely to break things in specific cases. If
network administrators want to make the DHCP server reachable from a
client in a different subnet, they should push proper routes with the
lease.
In any case, if the DHCP server is not reachable through unicast,
before the lease expiration (after timeout T2) the client will resort
to broadcast and so there won't be any network disruption; the renewal
will only happen at a later time.
Fixes: 31fe84e467https://bugzilla.redhat.com/show_bug.cgi?id=1448987
platform: signal: link changed: 2: eth0 <DOWN;broadcast,multicast> mtu ...
...
device[0x7f90c29c64d0] (eth0): bringing up device
...
platform: signal: link changed: 2: eth0 <UP,LOWER_UP;broadcast,multicast,up,running,lowerup> mtu ...
...
device (eth0): link connected
...
device[0x7f90c29c64d0] (eth0): add_pending_action (2): 'carrier wait'
Note how we schedule the pending action 'carrier-wait', although the device
already has carrier. That means, the pending action will not be removed
until timeout, 5 seconds later.
Avoid scheduling 'carrier-wait' if we already have carrier.
However, don't just add the pending action 'carrier-wait' only during
nm_device_bring_up(). Instead, always schedule the carrier_wait timeout.
This gives a grace period during which we keep setting 'carrier-wait' whenever
we have no carrier. This should prevent two cases:
- during nm_device_bring_up() the platform state might not yet have
caught up. If we don't add the pending action there, we will add
it a moment later when carrier goes away.
- bringing the interface up might cause carrier to get lost for a
moment (flapping). If that happens within the timeout, also add the
pending action.
nm_device_set_carrier_from_platform() is only called from two places.
- both check for NM_DEVICE_CAP_CARRIER_DETECT, so move that check
inside the function.
- drop the logging in realize_start_setup(). nm_device_set_carrier() already
does logging.
- always set the fake carrier in nm_device_set_carrier_from_platform().
For the fake carrer, we anyway expect it to be already TRUE in most
case, so usually this should have no effect.
Also emit a property changed signal. That is necessary to refresh the
D-Bus property.
Adding/Removing a pending action with assert_not_yet_pending/
assert_is_pending means that we expect that no action is taken.
Downgrade the logging level in those cases to <trace>.
Don't give the subclass the ability to override the parents
behavior. The parent implementation is not intended to allow
for that. Instead, restrict the flexibility of how the virtual
function integrates with the larger picture. That means, the
virtual function is only called at one place, and there is only
one implementation in NMDeviceEthernet (and it doesn't really
matter whether the implementation chains up the parent implementation
or not).
This makes it a lot more convenient to deal with long outputs (such as
"nmcli c show id ...").
The implementation is essentially jacked from systemd. The bugs are
mine.
This basically uses "wifi" in place of "802-11-wireless" and analogously
for other setting types that have aliases unless a --terse option was
specified.
Apart from looking nicer it makes the default output less wide,
hopefully avoiding line breaks on 80 column terminals in basic use
cases.
The agents may used this to learn that WPS PBC enrollment is active and
suggest that user pushes a button on the router instead of supplying a
network key.
It's sort of a very very lightweight version of get_secrets_done_cb()
that allows as to add secrets to the connection that didn't come from
the agent manager.
Exposure of this property on D-Bus would be useful for the GUIs to encourage
the users to enter a PIN or push a button.
We'll also use it to decide whether a connection will initiate WPS
enrollment unless the user overrides the default.
It serves no purpose, because the function always allocates a new
result and returns it. It would make sense, if the output string
would only be cloned when we need to allocate a new string. But
since that optimization is not done, the argument serves no purpose.
Don't log with level <info>. It's not important enough and only
noise. Also, move the logging line outside of nm_policy_init(). At that
point, the policy instance is not yet initialized (much), so we shouldn't
use the logging macro _LOGD().
A property preferably only emits a notify-changed signal when
the value actually changes and it caches the value (so that
between property-changed signals the value is guaranteed not to change).
NMSettings and NMManager both already cache the hostname, because
NMHostnameManager didn't guarantee this basic concept.
Implement it and rely on it from NMSettings and NMPolicy.
And remove the copy of the property from NMManager.
Move the call for nm_dispatcher_call_hostname() from NMHostnameManager
to NMManager. Note that NMPolicy also has a call to the dispatcher
when set-transient-hostname returns. This should be cleaned up later.
Hostname management is complicated. At least, how it is implemented currently.
For example, NMPolicy also sets the hostname (NMPolicy calls
nm_settings_set_transient_hostname() to have hostnamed set the hostname,
but then falls back to sethostname() in settings_set_hostname_cb()).
Also, NMManager tracks the hostname in NM_MANAGER_HOSTNAME too, and
NMPolicy listens to changes from there -- instead of changes from
NMSettings.
Eventually, NMHostnameManager should contain the hostname parts from NMSettings
and NMPolicy.
- anticipate missing callback/ctx->result
- always invoke the result callback when given
- fix leaking GVariant in disconnect_done()
- fix crash due to non-initialized ctx->result
- pass cancellable to g_dbus_proxy_call()
And some fixes:
- proxy creation may fail. Don't handle it by retrying,
but at least don't access the invalid proxy instance.
- quering "DefaultAdapter" did not take a reference to @self
nor was the operation cancellable. This leads to crash
if the instance gets destroyed early.
