Andi Kleen wrote:
> That sounds broken. Assuming you select the source address and start
> communicating from it on the local net, and the other host tries to do
> ARP probing on you later -- you're dead.
The idea is that when you are in your home network everything works "the normal way". However, when you leave
your home network, you disable ARP and your home agent kicks in. It answers ARP requests for you (with
ARP Proxy), and sends your IP packets to you via Foreign Agent. When you're not in your home network you only
communicate with your FA, and you don't need ARP to do that, since both ends know each others ethernet
address.
In practise, we wouldn't have to disable ARP, and most of the Mobile IP solutions on Linux don't do that, but
anyway it's in the RFC and it might be a good thing since it *forces* all the traffic to go via HA and FA.
> If you changed the routing tables correctly (you only don't want to
> answer ARP when you have a more specific routing table entry), then
> I have a better patch: It controls ARP behaviour via the routing
> table, which has a few other interesting advantages too. And unlike
> yours it is error tolerant (when the mobile link goes away things work
> again)
That sounds good; I'll have to take a look into it. Currently the Mobile IP software I'm developing for my
job takes care of "everything", but it sure sounds like worth looking at.
-- | vnummela@lut.fi work: +358-5-4125389 home: +358-40-5075560 | | So Linus, what are we doing tonight? | | The same thing we every night Tux. Try to take over the world! |- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun May 07 2000 - 21:00:14 EST