feature: support use of bridges via veth pairs for openvswitch data port configuration

[Matt Rae]

When configuring vlan provider networks in neutron, it is currently required to set data-port=br-ex:(some unconfigured interface).

Some users don't have an additional interface to leave unconfigured to use as data-port.

When using neutorn vlan provider networks which are trunked to the management interface, we found that the neutron-openvswitch charm runs 'ovs-vsctl add-port br-ex bond0', but that command fails with 'device busy'

We found the following solution which creates a veth pair between veth-br-bond0 and veth-tenant which allows both containers to br-bond0 but also neutron vlan provider networks to function.

on each compute node
# create veth pair between br-bond0 and veth-tenant
ip l add name veth-br-bond0 type veth peer name veth-tenant

# set mtu if needed on veth interfaces and turn up
#ip l set dev veth-br-bond0 mtu 9000
#ip l set dev veth-tenant mtu 9000
ip l set dev veth-br-bond0 up
ip l set dev veth-tenant up

# add br-bond0 as master for veth-br-bond0
ip l set veth-br-bond0 master br-bond0

juju set neutron-openvswitch data-port="br-ex:veth-tenant"

Should something like this be added to the neutron-openvswitch charm to allow neutron vlan provider networks on the management interface?

[James Page (james-page)]

I have a similar 'dodgy wiring' script that I use for this type of problem that takes the same approach; I think:

  https://review.openstack.org/#/c/392212/

implements this feature.