>>>>> On Fri, 16 Dec 2005 00:58:16 +0100, >>>>> Jeroen Massar <[EMAIL PROTECTED]> said:
>> We are considering four possible implementations: >> 1. try only ip6.arpa > This is the only one that should exist in current and new implementations. Yes, of course. But the fact is that there are implementations deployed that still use (or rely on) ip6.int. That's why I asked the question. > For deployed stacks, the ones people don't want/forget to upgrade there > is a very simple solution to all of this: DNAME Yes, this can generally be a helpful workaround at the authoritative server's side that manages an IPv6 reverse zone in order to provide both ip6.arpa and ip6.int reverse-map zones for the same address block. It's actually pretty known approached first (IIRC) proposed by Mark [EMAIL PROTECTED] in 2002: http://www.isc.org/index.pl?/pubs/tn/index.pl?tn=isc-tn-2002-1.html But I guess this is not 100% relevant to the point I'm talking about. The thing which is going to happen is that an RIR (likely with other RIRs) will simply remove the ip6.int tree, and my interest is the effect at the resolver (end user) side with the change. The DNAME approach may be relevant to this issue if the RIR adopts the workaround (as you may have indicated in your message), but I suspect it will unlikely happen. JINMEI, Tatuya Communication Platform Lab. Corporate R&D Center, Toshiba Corp. [EMAIL PROTECTED] --------------------------------------------------------------------- The IPv6 Users Mailing List Unsubscribe by sending "unsubscribe users" to [EMAIL PROTECTED]
