On 10/2/19 9:51 PM, Mark Andrews wrote:
What part of BCP-38 do you think needs to be updated to support IPv6?
Changing the examples to use IPv6 documentation prefixes instead of IPv4 documentation prefixes?
For a start, *add* IPv6 examples in parallel with the IPv4 examples. As RFCs are peer reviewed, if the examples expose a hole or problem then the hole can be filled or the problem resolved. BCP-38 should be reviewed in whole for "IPv6" completeness, and the preamble of BCP-38 add that the Best Practices include complete recommendations for both IPv4 and IPv6. One specific example: BCP-38 currently references RFC1812, _Requirements for IP Version 4 Routers_. It appears that the only parallel paper for IPv6 is draft-ietf-v6ops-ipv6rtr-reqs-04, _Requirements for IPv6 Routers_, which currently carries a copyright of 2018. It's a shame that this document is still in limbo; witness this quote: "It is inappropriate to use Internet-Drafts as reference material or to cite them other than as 'work in progress.' Someone else mentioned that "IPv6 has been around for 25 years, and why is it taking so long for everyone to adopt it?" I present as evidence the lack of a formally-released requirements RFC for IPv6. It suggests that the "science" of IPv6 is not "settled" yet. That puts the deployment of IPv6 in the category of "experiment" and not "production". Is that really true? There may be more issues. And, yes, I understand that a new BCP paper may result -- I don't care how it's labeled, as long as it's done. Or has such a BCP document already been released? If so, why do I not see any references to it here on NANOG, or anywhere else? Why do I care, you ask. I'm working on a BCP-38 module proposal for firewalld, one that can be peer-reviewed for accuracy and completeness. Servers running that firewall package can then be easily configured to conform to the requirements of BCP-38 and can easily become good net-citizens in their own right. So I call for draft-ietf-v6ops-ipv6rtr-reqs-04 to be finished and released as a formal RFC, and that BCP-38 be updated to refer to that finished RFC. Until that is done, my BCP-38 module will have to carry a "work in progress" disclaimer.