[as112-ops] Section 4.4 of RFC 4193 and AS112

[Paul Vixie]

> > What about the other IPv6 zones mentioned in Mark's draft that might
> > need a "sacrificial server" (see section 4)? Should AS112 cover them as
> > well?
> 
> I'd like to see those as well.  I focused on the ULA stuff largely due to 
> the sudden increase in message traffic on the NANOG mailing list in 
> response to ARIN's statement about IPv6 and questions sent to me privately 
> about ULA space.

i think ULA is a stupid idea, as those of you on ppml@ know.  but if it
passes through the ietf then AS112 definitely ought to serve it up.  i'd
be happier to see the old site-local prefix used, so that private networks
can be ambiguous like they are in IPv4.  so perhaps AS112 ought to serve
up the site-local in-addr prefix also, just in case somebody uses it.

> There's also the issue of whether or not AS112 servers should get an IPv6 
> address for anycast purposes, i.e. getting AS112 onto IPv6 infrastructure 
> where available and answering on IPv6 sockets.

i think this is a very good idea.  any objection to me requesting a block
from ARIN for this, using their /48 "critical infrastructure" policy?

> Of course, if Marc's draft moves to RFC and DNS implementors follow it, 

haaaaaaaaaaaaaaaaaaaaa haaaaaaaaaaaaaaa haaaaaaaaaa haaaaaaaa haaaa ha ha HA!

> maybe the day will come that AS112 becomes obsolete...but I think it would 
> be a long time before we see AS112 traffic drop in proportion to people 
> moving to newer versions of DNS server software.

i think AS112 will still see a minimum of 10K PPS in the year 2100.