This is an OpenPGP/MIME signed message (RFC 2440 and 3156)
--Apple-Mail-3--526549645
Content-Type: multipart/alternative; boundary=Apple-Mail-2--526549718


--Apple-Mail-2--526549718
Content-Type: text/plain;
charset=US-ASCII;
format=flowed;
delsp=yes
Content-Transfer-Encoding: 7bit

Hello list-namedroppers@pwns.ms,

My name is Olaf Kolkman, what name do you go by in day to day life?

You wrote:
>
> arpa is not delegated anywhere; dig NS ARPA should return NODATA,


Allow me to correct that: .arpa is an independent zone and currently
all of the root servers except for J.root-servers.net are
authoritative for .arpa.

> but if your algorithim is used it would return NXDOMAIN (nonce.arpa
> doesn't exist). arpa could be special cased, though, but I dislike
> that - it's likely that at some point there will be another (indeed,
> there may already be) non-delegated domain in the root; let's not
> require a resolver upgrade to let it work.


George wrote:
> Interestingly, the root servers appear to not behave uniformly for
> arpa queries.
> j.root-servers.net returns a referral for .arpa but a.root-
> servers.net returns NXDOMAIN.
> Whatever, arpa is apparently (and maybe not suprisingly) a special
> case.



This is not unexpected, given that the all-but-J root-servers are
authoritative they can authoritatively provide you with an NXDOMAIN
while J gives a referal.


> I don't see any advantage to having non-delegated domains in the
> root - indeed treating arpa as non-delegated seems quite perverse to
> me, creating special cases is not usually a good idea.


From a protocol perspective the fact that an authoritative server
serves more than one zone is not a special case. From a protocol
perspective the servers that serve the root, or TLD zones, are not
special cases.

If the protocol design starts to make special considerations for root
or TLD servers we should treat that as an indication that we should do
a better job.

Oh, by the way from an operational point of view the fact that the
root servers serve .arpa is somewhat special (see RFC2870 section 1.2
and 2.5), but that is outside the scope of this group, me thinks.



--Olaf
--Apple-Mail-2--526549718
Content-Type: text/html;
charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

-webkit-line-break: after-white-space; ">

My name is Olaf Kolkman, what name do you =
go by in day to day life?

You wrote:
type=3D"cite">
arpa is not delegated anywhere; dig NS ARPA should =
return NODATA,

Allow me to correct =
that:  .arpa is an independent zone and currently all of the root =
servers except for J.root-servers.net are authoritative for =
..arpa.

but if your algorithim is used =
it would return NXDOMAIN (nonce.arpa doesn't exist). arpa could be =
special cased, though, but I dislike that - it's likely that at some =
point there will be another (indeed, there may already be) non-delegated =
domain in the root; let's not require a resolver upgrade to let it =
work.

George wrote:
type=3D"cite">Interestingly, the root servers appear to not behave =
uniformly for arpa queries.
j.root-servers.net returns a referral for =
<nonce>.arpa but a.root-servers.net returns NXDOMAIN.
Whatever, =
arpa is apparently (and maybe not suprisingly) a special =
case.

This is not unexpected, =
given that the all-but-J root-servers are authoritative they can =
authoritatively provide you with an NXDOMAIN while J gives a =
referal.


I don't =
see any advantage to having non-delegated domains in the root - indeed =
treating arpa as non-delegated seems quite perverse to me, creating =
special cases is not usually a good =
idea.

=46rom a protocol perspective the =
fact that an authoritative server serves more than one zone is not a =
special case. =46rom a protocol perspective the servers that serve the =
root, or TLD zones, are not special cases.

If =
the protocol design starts to make special considerations for root or =
TLD servers we should treat that as an indication that we should do a =
better job.

Oh, by the way from an operational =
point of view the fact that the root servers serve .arpa is somewhat =
special (see RFC2870 section 1.2 and 2.5), but that is outside the scope =
of this group, me =
thinks.



--Olaf&nbsp=
;
=

--Apple-Mail-2--526549718--

--Apple-Mail-3--526549645
content-type: application/pgp-signature; x-mac-type=70674453;
name=PGP.sig
content-description: This is a digitally signed message part
content-disposition: inline; filename=PGP.sig
content-transfer-encoding: 7bit

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Darwin)
Comment: This message is locally signed.

iEYEARECAAYFAki+iF4ACgkQtN/ca3YJIodvjwCfdwZuOsR/h9AS7UkzPhgBwb7n
eIMAni/5Os7jO8WrMPSoaEc0sohRYwpo
=QKLW
-----END PGP SIGNATURE-----

--Apple-Mail-3--526549645--

--
to unsubscribe send a message to namedroppers-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: