Re: [Idr] Joel Jaeggli's No Objection on draft-ietf-idr-ix-bgp-route-server-11: (with COMMENT)

[joel jaeggli <[e-mail]>]

On 6/14/16 9:52 AM, Nick Hilliard wrote:
> joel jaeggli wrote:
>> IF L is a customer of A and B sure.
> 
> you mean, if L is a downstream of A and B - not necessarily a customer.

if it's a peer we'd view either of them sending it to the exchange as a
leak. The difference between in the downstream cone and customer route
is defintional

>> If it's L's policy that's doing the blocking
> 
> This is the point - it's not L's policy that's doing the blocking.  If A
> blocks L's prefixes at the RS, it will block all paths, via both A and
> B, even though it should have no control over B's paths if it decides to
> take itself out of the equation.

sure, easy enough to concede.

Albiet I rather prefer to take b out of the equation when it's my prefix.

> Nick
>