Yuchi,

> IMHO doing LPM with the lwAFTR's address is more straightforward than with a 
> "Domain v6 prefix".
> 
> In addition, I don't see why Ian's proposal cannot cover the case you 
> mentioned, the case in which an address out of the prefix domain can be 
> chosen as the tunnel endpoint address. If lwB4 has been provisioned with such 
> an address, and if this address does have a LPM with lwAFTR's address, lwB4 
> can still use it as the tunnel endpoint address. Please correct me if I'm 
> missing anything.

there are two issues here.

1) in the unified CPE context. is there a benefit in having the same algorithm 
to choose the CE tunnel endpoint address?
    are the requirements different?
2) pick the right mechanism for tunnel end point determination. in your above 
scheme you do not have the same flexibility as you have with a provisioned 
prefix selector 

I though we had covered one in previous discussions, but there might be 
something I've missed.
are we in agreement on this point? that it is beneficial to use the same 
mechanism for tunnel endpoint address determination on the CE.

cheers,
Ole

Attachment: signature.asc
Description: Message signed with OpenPGP using GPGMail

_______________________________________________
Softwires mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/softwires

Reply via email to