A) Is there any good written summary (draft or RFC) for actual examples of recently attempted
"IPv4 only" work in the IETF - (go read arbitrary ietf mailing archives does not count) ?
I am a bit disappointed that no examples are included in draft-ietf-sunset-4-ipv6-ietf.
B) The way i read the charter of sunset4, it only talks about sunsetting IPv4 for the Internet.
What i am unclear about is how the IETF cares about other IP networks. I would not be
surprised if the mayority of revenue from IETF technologies on this planet is made across
other networks than the Internet, and for those network no discussion about future requirements has
happened in the same manner as for the Internet.
C) I wonder how much the IETF is willing to entertain groups interested in IPv4 only
work in a way that spreads the pain equally between them and IPv6 evangelists:
1. Define the technology to support both IPv4 and IPv6.
2. Implementations meant to be useable across the Internet MUST support IPv6 only,
MAY support dual-stack.
3. Requirements for deployments meant for other networks wrt to stack support are
subject to the WG decision (including an IPv4 only profile).
Point 1 would create pain for folks interested only in IPv4 work. I think that for
the benefit of getting an RFC out of it (including all the expert community review
going along with it), they would be willing to do this.
Point 2 is my best understanding of current IETF thinking, so its neutral.
Point 3 would be the pain point for the IPv6 evangelists in IETF.
Of course, this attempt to get best of both worlds would not work for extensions
to existing IPv4 only protocols like OSPFv2 or ARP or the like, which is why i am
asking A) - to understand how much of a problem those protocols are.
Cheers
Toerless
On Mon, Oct 02, 2017 at 10:57:01AM -0400, R. Atkinson wrote:
> All,
>
> 1. My guess is that most folks agree with the principle of not undertaking new IPv4-only work.
>
> 2. Similarly, if the IESG were to adopt a practice/policy (if they haven???t already done so)
> that IPv4-only work on the IETF-Track required an IESG exception, then I think
> most folks here would be fine with that.
>
> 3. Many networks will need to be dual-stack (IPv4 + IPv6) for many years to come.
> This is a practical reality, even though it is inconvenient and politically incorrect.
> So there likely will be specific areas where work that applies both to IPv4 and IPv6
> will need to be undertaken both now and in the foreseeable future. This ought not
> conflict with (1) above.
>
>
> Discussion:
> ????????????????????????
>
> The particular I-D (draft-ietf-sunset-4-ipv6-ietf-01.txt) is not sufficiently clearly written,
> if it is trying to achieve either of the objectives (1-3) above. I think many folks would like
> to see at least both the title and abstract re-worked, probably the whole draft reworked,
> to make it more clear that the objective is to discontinue most IETF IPv4-only work.
> Certainly, I would want the title, the abstract, and the rest of the document to be edited
> to have a consistent, clear, and non-inflammatory message consistent with (1-3) above.
>
> I have to agree with Stephen Farrell that the best we can hope for in this are is to avoid
> ???most??? IPv4-only work, on grounds that if a major issue (e.g., security) arose in some
> IPv4-specific specification, then the IETF ought to address/resolve that IPv4-unique issue.
>
> However, if the IETF tries to take a hard line that no new IPv4 work is allowed, then the
> practical result will be that some other standards body will simply do IPv4-unique work
> outside the IETF (in practice; de facto) ??? which would be a very bad outcome, both for
> interoperability and for global standards cooperation.
>
> Yours,
>
> Ran