On Fri, 1 Jul 2005, Stephen Hemminger wrote:
LAN______________________________________________
|
| eth0 (no IP)
_____________________|________________________
| | Linux Box (PC)|
| ________|________ |
| | | |
| | br0 | |
| | (no IP) | |
| |_________________| |
| tap0 | | tap1 |
| 192.168.40.1/24 | | 192.168.30.1/24 |
| MAC_0 | | MAC_1 |
| ___|________|____ |
| | | |
| | OS TCP-IP stack | |
| | | |
| |_________________| |
| | | | | |
| | | | | |
| Processes |
| |
| |
|______________________________________________|
[snip]
It should work if the tap interface looks sufficiently like Ethernet.
You probably need filter rules to make sure and drop packets intended for
the other network get dropped and to prevent broadcast leakage.
The way I interpret the drawing, ISTM that Antonio has the bridge a layer
below the tap devices (even though it's drawn a layer above). I don't
think that's a very sane idea...
The crux of the problem seems to be that Antonio wants a single physical
ethernet card to use two different MAC adresses, which I don't think
briding is ever going to solve. This is, perhaps, possible by putting the
card into promiscuous mode, and using some clever ebtables mangling.
Alexey
_______________________________________________
LARTC mailing list
[email protected]
http://mailman.ds9a.nl/cgi-bin/mailman/listinfo/lartc
