|
|
|
|
|
|
|
|
|
|
|
|
| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |
From: =?utf-8?Q?Bj=C3=B8rn_Mork?= <bjorn@mork.no> To: "Rodolfo \(kix\)" <kix@kix.es> Date: Thu, 19 May 2011 14:54:05 +0200 In-Reply-To: <BANLkTikUg4KmQ7OsLe0Lr=d_F92DFWUH4A@mail.gmail.com> (Rodolfo's message of "Thu, 19 May 2011 12:30:08 +0200") Cc: nanog@nanog.org Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org "Rodolfo (kix)" <kix@kix.es> writes: > what is the status of the reverse DNS in IPv6? The draft > "http://tools.ietf.org/html/draft-howard-isp-ip6rdns-04" is expired. > > What are the network operators doing? For interfaces: same as for IPv4 (autogenerate forward and reverse) For residential sites:=20=20 nothing For business sites:=20 same as for IPv4 (rDNS management interface, optional delegation, nothing by default) The assumption is that only mail servers, web servers and other hosts you'd like to be verifiable within a domain, will need rDNS. That's the bare minimum. Having names on interfaces is simple to do (for us at least), and makes traceroutes etc easier to read, so we do those. For the rest of the addresses, rDNS really doesn't matter. Personally I would love to provide delegation and/or some rDNS management interface for residential sites as well, but time will tell if we get around to that. There are plenty of items on the wishlist and this is not on the top... We will not auto-generate bulk names like we do for IPv4 pools. In theory this could be done using something like e.g. http://member.wide.ad.jp/~fujiwara/v6rev.pl but I just don't see the use. The generated names will provide exactly no additional value to the address they are supposed to name. We will not use wildcards either. They provide no value as long as they cannot be mapped forward. In fact, they will rather break than fix any application actually caring about rDNS. Bj=C3=B8rn
|
|
|
|
|
|
|
|
|
|
|
|
| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |