[ Quoting <a...@anvilwalrusden.com> in "Re: [DNSOP] Order of CNAME and A in..."
]
On Tue, Aug 11, 2015 at 05:00:05PM +0100, Ralph Corderoy wrote:
That still leaves open the question of whether the stub resolvers can
assume, as many have apparently been doing for years, that they will be
given CNAME before A.
In general, RRsets aren't ordered. Now, I think a CNAME response that
also chases the CNAME (and there's a live question about whether the
authoritative-only server ought to do that anyway, but let's assume
it's all in-zone to make this simple) is not returning one RRset in
the answer section, but I don't think there's any promise anywhere
about what order the RRsets come in. Therefore any client that is
depending on an order has made a mistake.
So this discussion stems from this issue:
https://github.com/skynetservices/skydns/issues/217
And apparently the glibc resolver assume this is ordering is in effect.
This btw, has nothing to with the Go language (as implied on the ietf@ thread),
but is just an issue in SkyDNS.
It is also true that (stub)resolvers directly talk to SkyDNS, any recursive
queries are proxied through to a real recursor, but it will authorative answer
for the (single) domain is has configured.
/Miek
--
Miek Gieben
_______________________________________________
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop
