Re: [spring] We don't seem to be following our processes (Re: Network Programming - Penultimate Segment Popping)

["Darren Dukes (ddukes)" <[e-mail]>]

Hi Robert, thank you for injecting clarity to this thread.

draft-ietf-spring-network-programming defines PSP, and the only relevant portion of this thread to that draft is Discussion #3.

As I stated in another post, I consider #3 closed as this is clearly complying with RFC8200.
https://mailarchive.ietf.org/arch/msg/ipv6/6Uy_JuMm7W66kTql_o7FYisxkg0

The remainder of the discussion in this thread is unrelated to draft-ietf-spring-network-programming.

Thanks
  Darren

On Dec 7, 2019, at 6:47 AM, Robert Raszuk <[e-mail]<mailto:[e-mail]>> wrote:

Hey Fernando,

(pop when you are the destination but SL!=0 is essentially 'in the
network removal')

I was trying to stay out of this but I have one fundamental question or observation this entire debate seems to be about.

In the context of SRv6 there are two parallel discussions

Discussion #1 - It is about inserting, modifying or deleting SRH by nodes which are not in the outer IPv6 header of the packet

Discussion #2 - It is about RFC8200 compliance when the node doing insertion of SRH is *the* destination of the packet as read verbatim from the outer IPv6 header.

Discussion #3 - It is about RFC8200 compliance when the node doing modification or removal of SRH is *the* destination of the packet as read verbatim from the outer IPv6 header.

First let's observe that RFC8200 is only defining the behaviour regarding EH processing in the context of destination address of the IPv6 outer header: "identified in the Destination Address field of the IPv6 header.identified in the Destination Address field of the IPv6 header. "

Therefore stating that SL value before local decrement matters in this in respect to being compliant to the IPv6 RFC is at best just an individual interpretation. Besides the pseudocode says it black and white "S14.1.   If (updated SL == 0) {". We do all sort of processing decision after decrementing the values ... think TTL :)

So back to reality ...

Discussion #1 - I think we all agree that to accomplish that RFC8200 would need to be updated.

Discussion #2 - I think we also all agree here that to accomplish this RFC8200 would need to be updated as it does says clearly that "Each extension header should occur at most once, ..."

Discussion #3 - It seems clearly that there is no issue with compliance with RFC8200 and that if penultimate segment midpoint decides or is instructed to pop SRH it sure can and still be 100% compliant with current wording of RFC8200.

So other then so much foam what is this debate all about ?

Cheers,
Robert.

--------------------------------------------------------------------
IETF IPv6 working group mailing list
[e-mail]<mailto:[e-mail]>
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------