Joe Abley wrote:
... I don't think it's sensible to say absolutely that there will never be a need to disambiguate NXDOMAIN or NOERROR since never is an awfully long time, and who knows or dares to dream?
that outcome depends on scope. if you imagine a protocol speaker behaving differently based on fine-grained differences in cause that relate in small ways to the larger NXDOMAIN/NOERROR status code, then i think we should not allow for that -- it would be madness, even compared to the present day.
if you believe that it will appear in "dig" output and syslogs and dnstap traces, so as to help diagnosticians and academicians decide the whichness of what, after the fact, then i agree, we should not prohibit the fine detail just because we think the gross details are obvious.
-- P Vixie _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
