Woodworth, John R <[email protected]> wrote: > Apologies but I did not hear the full question regarding BULK RR's and > the perl like back-references. If you could please repeat the question > we would be happy to comment.
I didn't ask the question, but based on a quick look at the draft I think there might be a fun problem. Frequently, when you have string rewriting with substitution and iteration, you have a Turing-complete esoteric programming system. NAPTR very cunningly avoids this trap (IIRC) by constraining the rewrites to be on a strictly monotonically reducing suffix of the input. BULK does not itself have built-in iteration like NAPTR, but because it synthesizes records it interacts with the DNS's existing resolver loops: chasing delegations, and chasing CNAMEs and DNAMEs. Note that CNAMEs and DNAMEs are also chased within authoritative servers. So my question is, how does the BULK rewriting system interact with DNS loops? Is there a CPU-eating tarpit in there? Tony. -- f.anthony.n.finch <[email protected]> http://dotat.at/ - I xn--zr8h punycode South Utsire: Northwesterly 4 or 5 backing southwesterly 3 or 4. Slight or moderate. Fog patches, drizzle. Moderate or good, occasionally very poor. _______________________________________________ DNSOP mailing list [email protected] https://www.ietf.org/mailman/listinfo/dnsop
