[Olsr-dev] More information about problems with same IPs on multiple interfaces

[Henning Rogge]

Hello,

I'm currently doing a clean rewrite of the Hello-processing code for the 
development tip (which will be able to handle both OLSRv1 and v2 hellos) and 
had some time to think about the "same IP on many interfaces" scenario.

There is a discussion about the same thing on the IETF manet WG list, so this 
is really an important topic.

At first, let's talk about limitations. I think anyone could agree to the 
following assumptions:
1) no originator-IP is used by more than one OLSRd instance
2) no interface-IP is used by two stations in two-hop range at the same time
   In mobile OLSR scenarios number 2) will become similar to number one.

But what is about the "using the same IP twice on a single node" szenario ?

With the current protocol we have a problem here, which always happens when a 
node has BOTH interfaces with the same IP as 2-hop neighbors.

In this case you get linkquality data for two different links, but no 
information to keep them appart ! This already happened in Vienna, so it's not 
a purely theoretical thing. You cannot control the size of your 2-hop 
neighborhood, especially if someone adds wireless bridges and Layer 2 VPN-
Tunnels.

There is a possible sollution for this scenario, but I don't think it would 
work without a change in the Hello message structure:

If a node running multiple interfaces with the same IP would add it's 
interface INDEX into the hello packages, it would be possible to keep the two 
links appart in any kind of 2-hop neighborhood.

My question is, do we really need this ? We hope that future Freifunk OLSR 
networks will only need ONE public IP for the whole node, which can be reached 
through all of it's interfaces. With this it would be easier to choose 
different IPs for the interfaces itself, with IPv6 we could just use 
autogenerated linklocal-IPs.

Henning Rogge

*************************************************
Diplom-Informatiker Henning Rogge
Forschungsgesellschaft für
Angewandte Naturwissenschaften e. V. (FGAN) 
Neuenahrer Str. 20, 53343 Wachtberg, Germany
Tel.: 0049 (0)228 9435-961
Fax: 0049 (0)228 9435-685
E-Mail: (spam-protected)
Web: www.fgan.de
************************************************
Sitz der Gesellschaft: Bonn
Registergericht: Amtsgericht Bonn VR 2530
Vorstand: Dr. rer. nat. Ralf Dornhaus (Vors.), Prof. Dr. Joachim Ender 
(Stellv.)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.olsr.org/pipermail/olsr-dev/attachments/20090907/ea86347e/attachment.sig>