Re: I-D ACTION:draft-ietf-ipv6-unique-local-addr-08.txt

["Stephen Sprunk" <stephen@xxxxxxxxxx>]

Thus spake "Brian McGehee" <doc@xxxxxxxxxx>
> I have to agree with this MUST NOT.
>
> |+    Advertising locally assigned ULA AAAA records in the global DNS
> |+    MUST NOT occur as they are not globally unique and will lead
> |+    to unexpected connections.
>
> Although there is a good chance that someone else in the world has my same
> name, they don't receive their mail at my home address.  OK, Bad example
> (reversed).  But if there is a chance of overlap; allowing this insures
> there is a chance of misdirection.

My original thought in allowing ULA AAAA records in the global DNS was that 
it is not operationally feasible to expose those records to all of the hosts 
that have connectivity to those addresses and no others, particularly when 
you're using them to communicate privately to another organization.

There is certainly a failure mode where two parties may inadvertently select 
the same prefix and one may get traffic intended for the other due to 
entries in the global DNS, but that is no different from someone publishing 
a ULA IP address on a web page instead of a DNS name that points to that 
address.  The failure is the collision, not the DNS entry.

I did think about this and considered the odds of the above happening to be 
significantly lower than the risk of a host not being able to resolve names 
for machines it can reach via a ULA because it (or its nameserver) is 
inadvertently using the global DNS instead of a local zone file.

S

Stephen Sprunk         "God does not play dice."  --Albert Einstein
CCIE #3723         "God is an inveterate gambler, and He throws the
K5SSS        dice at every possible opportunity." --Stephen Hawking 


--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@xxxxxxxx
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------