IPv6 rDNS

Mark Andrews marka at isc.org
Mon Nov 1 22:36:39 CDT 2010


In message <7E9AF5E9-7B3A-4767-A1D3-8EAB64031DAF at delong.com>, Owen DeLong write
s:
> 
> On Nov 1, 2010, at 4:40 PM, Mark Andrews wrote:
> 
> >=20
> > In message =
> <AANLkTinUMZYp9qe0i5pHYZ72aL3XyCtvaqHjzHuTkpo2 at mail.gmail.com>, Mich
> > el de Nostredame writes:
> >> On Fri, Oct 29, 2010 at 6:06 PM, Jeroen van Aart <jeroen at mompl.net> =
> wrote:
> >>> I battled for a few hours getting IPv6 rDNS to work. The following =
> tool
> >>> proved to be quite helpful:
> >>> http://www.fpsn.net/?pg=3Dtools&tool=3Dipv6-inaddr
> >>> Just in case anyone else would run into similar problems. It's not =
> as
> >>> straightforward as IPv4 rDNS.
> >>> Greetings,
> >>> Jeroen
> >>> --
> >>> http://goldmark.org/jeff/stupid-disclaimers/
> >>> http://linuxmafia.com/~rick/faq/plural-of-virus.html
> >>=20
> >> Forgive me if this is a stupid question.
> >>=20
> >> I am curious that if BIND ever tried to make the DB file easier to
> >> operate under pure text-based environment.
> >> For example, allow something like following format inside zone file,
> >>=20
> >>  $ORIGIN 1.0.0.0.3.f.8.0.3.1.4.8.8.7.d.f.ip6.arpa.
> >>  48ff:fe35:d1bc         PTR     server.example.com.
> >=20
> > Firstly you don't have enough bits for a IPv6 address specified and
> > secondly how would you distingish that from wanting the following?
> >=20
> > 48ff:fe35:d1bc.1.0.0.0.3.f.8.0.3.1.4.8.8.7.d.f.ip6.arpa. PTR =
> server.example.com.
> >=20
> What would be the point of wanting that?

The point is that you are CHANGING the meaning of something that already
exists.

> It's not a legitimate ip6.arpa name.

Why not?  It's a perfectly legal name in the DNS.

> While I realize BIND will dutifully parse it in its current state
> and happily await a query for that particular name, it's not a =
> legitimate
> ip6.arpa entry and no such query is going to arise from anything
> other than an error or abuse.

> In other words, while it is syntactically within the bounds of the
> RFC, it is semantically useless.

How do you know I don't have a use for it?

> > If you feel like writing a $6REVERSE directive please go ahead.  We
> > would be happy to accept such a patch.  I would however make it
> > take full IPv6 addresses and also take prefix syntax ((prefixlen % 4) =
> =3D=3D 0,
> > as only nibble boundaries make sense) and allow $ORIGIN to be =
> specified.
> >=20
> > e.g.
> > $6REVERSE fd78:8413:830:1::/64 SOA ....
> > $6REVERSE fd78:8413:830:1::/64 NS ....
> > $6REVERSE fd78:8413:830:1::/64 NS ....
> > $6REVERSE fd78:8413:830:1::48ff:fe35:d1bc PTR server.example.com.
> >=20
> > $6REVERSE $ORIGIN fd78:8413:830:1::/64
> > @	SOA	...
> > @	NS	...
> > @	NS	...
> >=20
> > one could make it more general and do both IPv4 and IPv6 ($REVERSE).
> >=20
> I agree that a $REVERSE directive would be a better solution. (v4 and =
> v6)
> 
> Owen
> >=20
> 
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742                 INTERNET: marka at isc.org




More information about the NANOG mailing list