To:
Randy Bush <randy@psg.com>
Cc:
ipng@sunroof.eng.sun.com, dnsop@cafax.se, ngtrans@sunroof.eng.sun.com
From:
JINMEI Tatuya / $B?@L@C#:H(B
<jinmei@isl.rdc.toshiba.co.jp>
Date:
Tue, 04 Dec 2001 13:06:52 +0900
In-Reply-To:
<E16AtC7-0005IH-00@rip.psg.com>
Sender:
owner-dnsop@cafax.se
User-Agent:
Wanderlust/2.7.5 (Too Funky) Emacs/21.1 Mule/5.0 (SAKAKI)
Subject:
Re: (ngtrans) Re: reverse delegation under ip6.arpa.?
>>>>> On Mon, 03 Dec 2001 05:31:39 -0800, >>>>> Randy Bush <randy@psg.com> said: >> Anyway, through the responses so far, I feel we must definitely move >> to ip6.arpa. (regardless of the bitstring vs nibble issue). The >> migration should cause additional pain to deploy IPv6, but, with the >> reality, we should start the migration now... > agreed. but the pain is minimal. Since resolver implementations that do not use ip6.arpa have been widely deployed, I don't think the pain is minimal. I'm afraid we should maintain both ip6.arpa. and ip6.int. for same contents forever. But..., > note that, initially, the content of > ip6.arpa is directly that of ip6.int. in fact, one could have the same > zone file pointed to by both names. yes, that's the way that I configured ip6.arpa zones in my v6 network. > the big pain in the transition is > that of the registries, whois, etc. and they've been working on this > for some months. I know, I realized some portion of the ip6.arpa (nibble) domain has already been delegated. So, while I still don't think the migration overhead is minimal, it seems to me that all I have to do now is to implement new resolver code, deploy it, and start operation with ip6.arpa. If the migration is inevitable, we cannot delay it. JINMEI, Tatuya Communication Platform Lab. Corporate R&D Center, Toshiba Corp. jinmei@isl.rdc.toshiba.co.jp