Re: [v6ops] I-D Action: draft-townsley-troan-ipv6-ce-transitioning-00.txt

Carl,

> Small Q: so you consider "BFD-support" to be mandatory on the CPE ?

as Gert pointed out this isn't really BFD.
it is whatever packet you'd like to send yourself forwarded through the PE.

with regards to your question, too early to say.

cheers,
Ole

> -----Original Message-----
> From: v6ops-bounces@ietf.org [mailto:v6ops-bounces@ietf.org] On Behalf Of Ole Troan
> Sent: woensdag 14 december 2011 14:41
> To: Gert Doering
> Cc: IPv6 Operations; draft-townsley-troan-ipv6-ce-transitioning@tools.ietf.org
> Subject: Re: [v6ops] I-D Action: draft-townsley-troan-ipv6-ce-transitioning-00.txt
> 
> Gert,
> 
>>> in the case of a link to ISP A was down, the CPE could of course forward traffic with SA = ISP A out ISP B's link. if it was stopped by ingress filtering the host would get an ICMP back from the PE instead of the CPE.
>> 
>> Would it?
>> 
>> ISP B's PE has no route to send ISP A's prefix to that CPE, so how 
>> would the ICMP reach the host?
> 
> good point.
> 
> OK, for auto-detection of ingress filtering. we could use "BFD echo mode". like what we do for 6rd BR reachability check.
> the CPE generates a packet with SA = DA = one of its own addresses from ISP A. this packet it forwards to the ISP B.
> 
> cheers,
> Ole
> _______________________________________________
> v6ops mailing list
> v6ops@ietf.org
> https://www.ietf.org/mailman/listinfo/v6ops