[ietf-dkim] Re: New issue: Upward query vs. wildcard publication

[Jim Fenton]

Frank Ellermann wrote:
> Hi Jim,
> I'm lost with these DNS considerations.  Your option 3 is apparently
> the "CSV method" (or rather the CSV method was an elaborated variant
> of a naive tree walk, for starters it never hit the root servers)
>   

I don't remember offhand how CSV did this.  My algorithm wouldn't
directly hit the root servers either, but might do so indirectly in
resolving some TLDs.  There was specific concern from some people who
ran root servers (or perhaps TLD servers) about the impact of the tree
walk, which I can't say I fully understood.  I would expect subsequent
queries to a parent domain to be satisfied by cached answers, at least
as far as root and/or TLD servers are concerned.
> Anything using wildcards where they're not already used today could
> also run into trouble.  If folks like it anyway I hope they've read
> I-D.iab-dns-synthesis-concerns-00
>
>   

I wasn't aware of it; but I guess it's new.  Thanks for pointing it out.

-Jim