Wrapping libvirt Virtual Networks Around Open vSwitch Fake Bridges

 

In a previous article, I talked about how to use Open vSwitch (OVS) fake bridges to bring VLAN support into your environment. In this article, I show you how to wrap a libvirt virtual network around your OVS fake bridge.

You might ask, “Why wrap a libvirt virtual network around an OVS fake bridge when you can just use the OVS bridge directly?” That’s a good question, and—to be perfectly honest—I don’t have a great answer. At first glance, it seems like it might make things easier if you are mixing both OVS-based networks and other types of networks, but I don’t know that for certain. If anyone has any feedback one way or the other (why this is a good idea or why it’s not a good idea), please speak up in the comments.

Now that we have that out of the way, the process for using a libvirt virtual network with an OVS fake bridge is actually pretty straightforward. First, create the appropriate OVS fake bridges using the instructions in this article. So, for example, you might create a fake bridge for VLAN 100 like this:

ovs-vsctl add-br vlan100 ovsbr0 100

Next, create an XML definition for a libvirt virtual network. For a fake bridge named vlan100, your XML definition might look something like this:

<network>
  <name>vlan100-net</name>
  <forward mode='bridge'/>
  <bridge name='vlan100'/>
</network>

Then, in the guest domain configuration, you reference the libvirt virtual network instead of the underlying bridge directly, like this:

<interface type='network'>
  <mac address='11:22:33:aa:bb:cc'/>
  <source network='vlan100-net'/>
  ...
</interface>

And that’s it! Based on my testing, it even appears that you can make this change on the fly, without having to reboot the guest domain. However, I could be wrong—if anyone knows definitively, please speak up in the comments. Any other corrections, clarifications, or questions are also welcome in the comments below.