神明達哉 <jin...@wide.ad.jp> wrote:
>
> As long as those records are generated from the target name that would
> probably be okay. But the current draft doesn't seem to enforce it,
> and, (probably unintentionally/implicitly) allows the following setup:
>
> aaaa.example.com. ANAME aaaa.example.net.
> aaaa.example.com. AAAA 2001:db8::aaaa ; not populated from ANAME target
> aaaa.example.net. AAAA 2001:db8::bbbb
>
> This looks more like a prohibited "CNAME + other AAAA for the same
> name" situation to me.
From an implementation's point of view, it can't tell the difference
between this kind of misconfiguration, and legitimate mismatches due to
things like stale address records or addresses obtained from different
views, etc. In all these cases it can either use the addresses it has been
given or replace them with the addresses from the ANAME target.
Hostmasters who set up inconsistent ANAME and addresses will suffer the
consequences :-) Maybe their suffering could be avoided by a suitably
intelligent master file loader or other provisioning thing.
Tony.
--
f.anthony.n.finch <d...@dotat.at> http://dotat.at/ - I xn--zr8h punycode
Lundy, Fastnet: Variable 3 or 4. Smooth or slight, occasionally moderate in
southwest Fastnet. Fair. Good.
_______________________________________________
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop
