Notices by Patrick Mevzek (pmevzek@framapiaf.org), page 2

@shaft @lanodan "1 personne = 1 voix" à l'IETF est un mythe à ranger dans la même boiîte que "les startups qui naissent dans un garage avec 2 écus et une miche de pain pour voir 1 milliard 2 ans après". Ca semble superficiellement vrai, mais ne tient pas une analyse approfondie. Pour QUIC, ca a aidé "un peu" que ca vienne de Google :-)

@lanodan @bortzmeyer @duponin All nameservers have test suites . There is https://github.com/dns-violations/dns-violations/ that can give list of edge cases/common problems. There are tools to parse and replay DNS packets and more at https://www.dns-oarc.net/oarc/software (see dnscap, dnsjit, drool, bad-packets projects)

@duponin @bortzmeyer "I can't work out how to make sure the query is well-formed, neither find a validation tool anywhere" That part should be easy. Any good DNS library should have the equivalent of `to_wire/from_wire` methods (they are called like that in DNSPython), which allows to either pass a buffer of bytes as captured in theory from network and see how they are parsed, or the opposite.

@duponin @bortzmeyer But then how can you compare/validate your results (from need X to bytes ABCDE...)? With just the RFC? tough luck, there are 1) too many of them for DNS 2) contradictory between themselves or ambiguous at times and 3) just impossible to draw a concise exhaustive view of what is "the DNS". Hence my suggestions to use external existing libraries (at least Python and Go ones are very good DNS wise) to help building you own case. But YMMV.