Re: Kernel Routing sequence

From: Al Boldi
Date: Sun Aug 14 2005 - 16:19:02 EST


Thanks for your input!
Meelis Roos wrote:
> >> No, it's part of the 10.0.1.0/24 network, since that's the more
> >> specific route.
>
> AB>
> AB> It's part of 10.0.0.0/8 because it came in on eth0.
>
> Your terminology is clearly different than the networking people
> here use - try to use the same terminology or expect to never get a
> sensible response.

Yes, it really does feel that we speak on different WaveLengths.
When talking to an int'l audience it is important to keep an open mind
and relax any strict interpretations. The objective should be to
come to an understanding any which way.

Please bear with me!

> IP address being a part of a IP network is just plain bitwise
> calculation and has nothing to do with interfaces.

This is obvious, but the fact that the Kernel accepted the packet on
eth0 implies it is either part of 10.0.0.0/8 or 10.0.0.0/24.
Throwing this information away is strange.

> AB> You are right! The Kernel is not actively rerouting, but it is
> AB> rerouting passively by blindly following the route table.
>
> It is following the routing table because that's how people expect
> it to behave. Just set up your routing tables as you like.

Yes, but what if your routing requirements are dynamic based on
src/dest-net/mask? Are you saying it is not possible to make such a
routing decision?

> There is one additional thing you might want to have a look at when
> you are building routers with asymmetric routing: rp_filter. You
> might to want to turn it off for the interfaces in questions.

Isn't that the default?

--
Al

-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html