> On Dec 26, 2019, at 1:55 PM, Dick Franks <[email protected]> wrote:
>
> Is there an implicit assumption here that all RRs beyond the UDP limit
> are to be discarded without requiring inspection of their internal
> content?
That makes for RRs that carry relevant DNS data from zone files, but
the OPT RR is "chrome" (channel metadata), and required for EDNS.
> If so, then I have a problem with the current EDE draft which appears
> to involve preferentially discarding the EDE option from the OPT RDATA
> (as distinct from the entire OPT RR).
So it may be reasonable to drop EDNS options, rather the whole OPT RR.
And if necessary drop answer and/or auth RRs, while retaining at least
a minimal OPT RR.
But I have not thought this through further, so perhaps some crucial
neglected factor could change the conclusion.
--
Viktor.
_______________________________________________
DNSOP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsop
