[dns-operations] DNS Violations (Was: Hall of DNS Shame (?))

[Jan Včelák]

> So both combined (textual + dnstap) might work the best.

Agreed.

> still have no idea how that could be used for automated testing unless we start editing those dnstap files to have a "correct" response from those servers. But any ideas are welcomed.

I have one idea. Most (all) of the violations are query-response
pairs. So for instance, if you want to test if Knot Resolves is able
to cope with the malformed responses, you can translate the .dnstap
files to Deckard's .rpl templates. I believe this could be already
done (semi-)automatically by a script.

Jan