Hello Juliusz,

Thanks for your quick response.

Indeed, the intended status is very important in the publication process.

First, please note that it is perfectly acceptable for a Proposed Standard to 
have a normative reference to an informational / experimental document, it is 
called a "down ref" and is quite common. It needs to be explicitly listed in 
the IETF Last Call and the IESG has to approve the downref (this is really 
routine) if not already in https://datatracker.ietf.org/doc/downref

As written in my AD review, I am unsure that the section about intermediate 
routers sending ICMP errors belongs to this I-D, a simple reference to RFC 7404 
would suffice. If this section is kept, then the title & abstract should be 
updated accordingly as the I-D would be broader than simply next hop routing.

Regards

-éric

From: Juliusz Chroboczek <[email protected]>
Date: Thursday, 4 December 2025 at 13:13
To: Eric Vyncke (evyncke) <[email protected]>
Cc: [email protected] <[email protected]>, Warren Kumari <[email protected]>, 
[email protected] <[email protected]>, Luigi Iannone <[email protected]>
Subject: Re: AD review of draft-ietf-intarea-v4-via-v6-05

Hello Éric, and thanks for your review.

We'll address the rest of your review promptly, for now please let me
answer your point about the intended status.

> ### Intended publication status
>
> It seems to me that it is rather an informational document than a proposed
> standard as this I-D does not specify anything but merely describe how 
> existing
> implementations can be used.

The very reason this draft exists is that we need a document describing
v4-via-v6 that can be normatively cited in Standards Track documents.  The
lack of such a document is the reason why we had to downgrade RFC 9229 to
Experimental, even though the protocol extension described therein has
been deployed in production for years.

This leaves us with either Standards Track or BCP.  The reason we have
chosen Standards Track is this sentence in Section 4:

  If a router does not have any IPv4 addresses assigned, the router MUST
  use the dummy address 192.0.0.8 as the source address of outgoing ICMP
  packets

This behaviour has been discussed in the Babel working group, and the WG
came to the following conclusions:

  1. the source address of the ICMP packets cannot depend on the routing
     protocol: when a router needs to send a destination unreachable
     error, then it has no way to determine whether it's an OSPF or an
     IS-IS route that's missing;

  2. it is best to agree early on a single dummy address that is used
     across implementations, since it simplifies both operations and the
     implementation of ICMP tools.

Hence, we have chosen to normatively mandate a single dummy address.  For
this reason, we request publication of this document as Standards Track.

I hope you find this argument convincing.  If you do not, then please
suggest a way to proceed that will not prevent this document from being
normatively cited in Standards Track RFCs.

Thanks,

-- Juliusz
_______________________________________________
Int-area mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to