Qrux wrote: > I'm also not a bridging expert, but Nathan's explanation sounds very > reasonable: saying "br0 becomes a switch" is a good analogy with a > nice visual image. I don't know the specifics of what's happening in > the kernel, but when you form a bridge, the original hardware > interface (e.g., eth0) turns into a "port", and is no longer a > configurable interface.
I do have some experience with HW bridges/switches. In theory, a bridge works at layer 2 to translate from one protocol to another, say 802.3 (ethernet) to 802.5 (token ring). A switch also works at layer 2 to 'route' from one network segment to another. It also can implement virtual networks (vlans). Generally, if you want to transition from one network to another, it needs to be done with a router at layer 3. My problem with kvm is that I don't know what it is trying to do. I would think that creating a virtual machine (VM) would create a network connection on both the host and VM sides, say tap0 on the host and eth0 on the client, and connect them. Then these should be connected like the iptables masquerade capability if on a different network or through normal switching if on the same network. Looking at the kvm man page and web site documentation, I seem to get more confused, not less. I'll keep trying. -- Bruce -- http://linuxfromscratch.org/mailman/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
