[99398] in North American Network Operators' Group
RE: ipv6/v4 naming nomenclature [Was: Apple Air...]
daemon@ATHENA.MIT.EDU (michael.dillon@bt.com)
Thu Sep 20 07:01:28 2007
Date: Thu, 20 Sep 2007 11:41:11 +0100
In-Reply-To: <865DF4F4-FFF8-43ED-9041-6105D09250DF@blyon.com>
From: <michael.dillon@bt.com>
To: <nanog@merit.edu>
Errors-To: owner-nanog@merit.edu
> >> Don't come up with any other variants. The above form is=20
> what is in=20
> >> general use around the internet and what some people will at least=20
> >> try to use in cases where a DNS label has both an AAAA and=20
> A and one=20
> >> of them doesn't work.=20
> Where did the www.ipv6 and www.ipv4 "standard" come from?
He already explained that as quoted above. It is a de-facto
standard since that is what is in general use around the
Internet. Standards are not always created by standards groups,
sometimes they just grows, like Topsy.
The key thing here is that adding AAAA records for a host that also
has an A record, can cause strange things to happen. If this would=20
be bad for the services offered by the host with the A record, then
you can create two new pseudo-hosts ipv6.host and ipv4.host. Put the
AAAA record only on the ipv6.host entry, and make the ipv4.host entry
either a duplicate of, or a CNAME to the original host with the A
record.
That way, you can still get some IPv6 traffic from IPv6 knowledgeable=20
people for testing purposes. Or you can solicit people to help with
your testing by using the ipv6.host variant.
--Michael Dillon
