Re: [v6ops] The need for local-ipv4 socket transition solutions -- NAT64/DNS64 remains insufficient

[Mark Andrews <[e-mail]>]

In message <CADhXe52acGYR-y7VNpfXt=[e-mail]>, James Woodya
tt writes:
> 
> I'm not optimistic that consensus would emerge for that recommendation.
> 
> Again: I think if 464XLAT promoters want to direct their energies on this
> topic in the most useful direction possible, then the best course of action
> would be to prepare drafts for updating RFC 3493 and RFC 3542 accordingly.
> In particular, I think it's critical to have a clear understanding of
> precisely what features of the PF_INET interfaces are expected to be
> handled by the CLAT function. If the IP_PKTINFO socket option, for example,
> is expected to be unsupported, then it's important to know that. What does
> the CLAT do with IPv4 link-local scope addresses? What about the
> SO_REUSEADDR option? A good place to start would probably be to go through
> the entire POSIX.1 standard, and for all the IPv4 networking interfaces,
> describe what is and isn't supported by the CLAT function recommended by
> IETF.
> 
> If this stuff is going to be a standard feature of host operating system
> for the foreseeable future, then I for one would like it to be standardized
> as well as the current IPv6 programming interfaces are standardized. I
> don't think that's an unreasonable position to start.

Part of that is getting the Advanced Socket API recognised by POSIX.
As a developer it is a right pain in the backside that it isn't and
you have to do different magic incantations to make the Advanced
Socket API visible on different platforms.

Making "socket(AF_INET,...)" fail would also be a interesting idea.
You can talk to IPv4 nodes over socket(AF_INET6, ...) using mapped
addresses.

A boring but useful Google Code project would be to go through the
OSS that is still IPv4 only and add IPv6 support.   Adding fast
failover to the next address would also be useful at the same time.
Pick a something like FreeBSD ports and bring all current ports up
to scratch.

> -- 
> james woodyatt <[e-mail]>
> Nest Labs, Communications Engineering
> 
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: [phone]                 INTERNET: [e-mail]