To:
Edward Lewis <edlewis@arin.net>
Cc:
"Hollenbeck, Scott" <shollenbeck@verisign.com>, "'ietf-provreg@cafax.se'" <ietf-provreg@cafax.se>
From:
Joe Abley <jabley@isc.org>
Date:
Thu, 10 Apr 2003 16:50:28 -0400
In-Reply-To:
<a05111b05babb86d0a04c@[192.149.252.108]>
Sender:
owner-ietf-provreg@cafax.se
Subject:
Re: [ietf-provreg] References for Today's Host Object Discussion
On Thursday, Apr 10, 2003, at 16:45 Canada/Eastern, Edward Lewis wrote: > Speaking as a DNS person - you don't need glue for MX's because you > get that from the A records sitting at the NS-named machines. You > need NS glue because there's nowhere else to get the A records other > than NS-named machines. (Speaking grossly.) > > In the example below, "smtp.example2 A" isn't glue, it's part of the > "something." zone. Sure. But I thought the case we were talking about was that where the "something." zone contains bare MX records, and there *is* no delegation (hence no NS-named machines). I agree they're not glue, though. The idea I was trying to convey was that they were records whose function was to make the MX records useful, in the same kind of way that glue records' function is to make NS records useful. I've seen more than one ccTLD zone that started off looking like that (indeed, that started off with NS records in the zone being a rarity, and A, MX and TXT records being much more common). Joe