Re: [Technical Errata Reported] RFC4443 (6153)

[神明達哉 <[e-mail]>]

At Wed, 21 Oct 2020 04:09:46 -0300,
Fernando Gont <[e-mail]> wrote:

> > On Fri, May 1, 2020 at 2:04 PM RFC Errata System
> > <[e-mail]> wrote:
> >> The following errata report has been submitted for RFC4443,
> >> "Internet Control Message Protocol (ICMPv6) for the Internet Protocol Version 6 (IPv6) Specification".
> >
> > Folks, we're all busy and that's fine, but it's been early May, any follow-ups?
>
> FWIW, it would seem to me this erratum should be approved.
>
> One might argue that the "It should say" part could be simplified by
> simply s/scope/scope zone/, since RFC4443 need not rehash what's in
> RFC4007. But that's mostly a matter of personal preference, and doesn't
> affect the correctness of the erratum you've submitted.

I don't have a strong opinion on whether to approve the erratum, but
if we do so, I'd keep this part of the original text: "This condition
can occur only when the scope of the source address is smaller than
the scope of the destination address (e.g., when a packet has a
link-local source address and a global-scope destination address)".
Technically, this text is not necessary to implement this ICMPv6 code
correctly, but it will still help the understanding.

Also, if the motivation of this erratum is this:

  Consider a case when the source IP is link-local and the destination
  is global, yet the routing happens in the same VLAN. Per RFC 4007,
  the packet should be transmitted; however, RFC 4443 allows for an
  ambiguity which is already causing vendors to reject packets in this
  case.

then I'm skeptical about how much the new text helps improve the
situation.  The misunderstanding that "a packet with link-local src or
dst address can never be forwarded by a router in any way" (it can, as
long as the packet would still stay in the same link zone) is quite
common, but I suspect the current text of RFC4443 Section 3.1 is not
the main source of this misunderstanding.  Even if so, I suspect the
new text will help correct the misunderstanding very much; if
correcting the well-known confusion is the purpose of the erratum, I'd
rather make the erratum text more explicit about that point.

--
JINMEI, Tatuya