APNIC IPv6 transit exchange

[Terry Manderson]

On 05/12/2007, at 1:44 AM, Bernhard Schmidt wrote:
>
> @Terry: Thanks for fixing this, but by saying you fixed the export 
> filter for 1221->38610 you basically tell me that v6TE is not 
> operating in the state you want it to be and the only cause why it is 
> not leaking such routes at the moment is because it doesn't receive a 
> fulltable. Correct?

I think there are several parts to answer that.

I typo'd the path filter that prevented some routes going (leaking) 
from 1221 to 30071. So no it wasn't operating in the way I wanted by 
config. (5 or so routes were erroneously leaked)

This really begs the question as when is considered a customer, a peer, 
or a transit in v6 space? I mean you can certainly intuit the state. 
(as contracts, and money exchange, are mostly non-existent and changes 
to the relationship happen without changes to any other mechanisms). In 
intuition good enough here? (I'm guessing on your reaction it isn't)

Is there any well defined work in the community to nail down such 
behaviours? and perhaps any BGP communities that are well known to 
apply such so that these errors are less likely to happen, and less 
likely to receive the ire of some operators when germany -> australia 
-> germany paths tend not to exist? Maybe (thinking aloud as a bad idea 
fairy) country based communities so that an international transit won't 
advertise a German prefix to a German network. But then that cycles 
back to the statement above.

Terry
--
Terry Manderson                         email:      terry at apnic.net
Network Operations Manager, APNIC       sip:    info at voip.apnic.net
http://www.apnic.net                    phone:      +61 7 3858 3100