-----Original Message----- From: Aled Morris [mailto:aledm@qix.co.uk] Sent: Friday, October 12, 2001 8:59 AM To: Stephane Bortzmeyer Cc: Nipper, Arnold; Dave Pratt; lir-wg@ripe.net; routing-wg@ripe.net Subject: Re: Multihoming - Resilience or Independence

On Thu, 11 Oct 2001, Stephane Bortzmeyer wrote:

...

Or, said in an other way: RIPE policies make very difficult to be multihomed, therefore people in RIPEland take a lot of time to explain that multihoming is bad.

RIPE doesn't make up the policies - the members make the policies.

RIPE didn't invent IP either, nor did they have a say in the design of router architecture, nor in the pricing of RAM chips.

Multihoming which relies on long prefixes being carried in the default free zone, and/or more AS numbers, simply doesn't scale. That isn't a policy matter, it's simple math.

Which policy do you think should be changed in order to "solve" the multihoming problem?

Aled -- QiX Limited

I do think it is policy related. For example, Customer Alpha want to be multi-homed from the very beginning so he requests a /24 block from RIPE. What RIPE should do is asking which TWO ISPs Alpha is going to use. Let's say the policy requires Alpha to be under the CIDR block between these two ISPs according to a table like: UUnet & CW : 123.45.32.0/20 UUnet & Quest : 123.45.64.0/20 Ebone & Sprint: 123.45.96.0/20 ... and IX range: 123.45.128.0/20 Some note about this example: 1. The major ISP can always filter these range at /20 without worring the holes inside them. 2. It won't need to cover 100% of multi-homed customer but if major ISPs are covered it will reduce the routing table dramatically. 3. For IX people(more than 2 ISPs), it won't be aggregate possible but by allocating them in the same range we might find some small blocks (/23 /22 are highly possible ) that are aggregate possible. This is just one example what RIPE CAN do. And because it involves different ISPs it certainly looks like a community affair that belongs to the RIR level. Ping Lu Cable & Wireless USA Network Tools and Analysis Group W: +1-703-292-2359 E: plu@cw.net