Hi John, On 1 May 2020, at 14:23, John R Levine <jo...@taugh.com> wrote:
>> On Thu, Apr 30, 2020 at 9:44 PM John Levine <jo...@taugh.com> wrote: >>> I think it's benign to allow any sort of record as an immediate child >>> of the domain, since you need to go two levels down for split zones. >>> That handes the nominet and zz--zz cases. > >> Is there any chance that a user trying to reach https://example.com could >> get the orphan glue A record for example.com instead of the A record in the >> real zone? >> (Just trying to think of cases where orphan glue might make a difference.) > > Only if the zone had NS and A at the same name, which would be pretty broken. Oh, interesting. In a sense, a glue record with the same owner name as a zone cut could be equivalent to a glue record with an owner name that is subordinate to a zone cut. I don't have enough of the spec in my head to know why they would definitively be different from the protocol perspective. I realise it's not normal, but I don't know that it's prohibited. I definitely don't know operationally how different DNS or registrar software implementations treat that case. I don't think the registry systems I'm familiar with allow host and domain objects with the same name to coexist, but I realise I could quite well be wrong. If I had any more energy to spend at the keyboard today I might be tempted to find out :-) Joe
signature.asc
Description: Message signed with OpenPGP
_______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
