Re: [Idr] Early Questions for the draft-ietf-idr-mpbgp-extension-4map6

[Chongfeng Xie <[e-mail]>]

Hi Sue,

Thank you for your questions, please see my feedback inline,

 
From: Susan Hares
Date: 2024-03-18 07:06
To: [e-mail]
Subject: [Idr] Early Questions for the draft-ietf-idr-mpbgp-extension-4map6
Chongfeng: 
 
I have a few questions on the tunnel encapsulation section: 
 
Have you considered putting the fields on page 7 – into the SubTLV type for the tunnel type “4map6” ?
                    [Chongfeng]:  I think this is a good idea, and we will consider to put them in the subTLV type in next version.
 
What type of validation do you need for the node sourcing this information? 
                    [Chongfeng]:  Yes, the receiving PE device needs validation of the authenticity of the mapping between IPv4 address block and IPv6 mapping prefix in IPv6-only network. In small-scale controlled network this maybe be not to stringent, but in larger networks, such as multi-operator environment, validation is needed absolutely, so we are working on this kind of work under the RPKI framework. 


One way to secure the information is to validate the tunnel egress ending point. 
 
 
Here’s the diagram: 
 
Peer-1 on Edge (sends update)      BGP Peer 2 (edge)  
Originate the (mapping) ----------- receive Update [     
[Edge 2] [unmap ]                            receives mapping [Edge 1] 
 
Why do you not want to use the Egress Tunnel Endpoint to validate that you have
Peer 2 has a connection to the egress Peer. 
 
I hope these questions help you to prepare for the IDR meeting today. 
 
Sue

I hope the feedback above have addressed your concern. 

Best regards
Chongfeng