Possible 1.22 nmcli regression with DISCONNECTED device state
I am trying to make sense of a python-dbusmock regression in rawhide. I have a NetworkManager template which emulates parts of NM's D-Bus API for doing integration tests of NM client software.
In one test I add and remove an active connection, after which the underlying device should be in state "disconnected". This also seems to be the case on the D-Bus level:
$ busctl introspect org.freedesktop.NetworkManager /org/freedesktop/NetworkManager/Devices/mock_WiFi1
org.freedesktop.NetworkManager.Device
.State property u 30
30
means DISCONNECTED
, so I think the mock is ok. But nmcli
claims otherwise:
$ nmcli dev
DEVICE TYPE STATE CONNECTION
wlan0 wifi connected --
Did nmcli get some changes so that it looks at something else than just the org.freedesktop.NetworkManager.Device.State property for its STATE column? Or is that an actual regression?