NetworkManager 1.2-beta has a bug activating VLAN on a virtual parent interface.
This is upstream bug https://bugzilla.gnome.org/show_bug.cgi?id=762731
I open this BZ so we don't forget to test the usecase for rhel-7.3.
I didn't confirm whether the same issue exists on nm-1-0/rhel-7.2. Possibly not, because master could behave quite differently due to the "devices-for-all" branch.
Copied from upstream:
>>>>>>>>>>>>>
create a team-slave, a team, and a vlan on top of the team.
nmcli connection delete t-slave t-team t-vlan
DEV=enp0s25
nmcli connection add type team con-name t-team autoconnect no -- \
connection.interface-name t-i-team \
ipv6.method ignore \
ipv4.method manual \
ipv4.addresses 192.168.103.5/24 \
ipv4.gateway 192.168.103.1
nmcli connection add type ethernet con-name t-slave slave-type team \
ifname $DEV autoconnect no master t-team
nmcli connection add type vlan con-name t-vlan autoconnect no id 42 \
dev t-i-team -- \
ipv6.method ignore \
ipv4.method manual \
ipv4.addresses 192.168.104.5/24 \
ipv4.gateway 192.168.104.1
nmcli connection up t-vlan
# fails with:
# Error: Connection activation failed: failed to determine interface
# name: error determine name for vlan
# but
nmcli connection up t-slave
nmcli connection up t-vlan
# succeeds.
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.
https://access.redhat.com/errata/RHSA-2017:2299