Re: Bogus nic.fr behavior

["D. J. Bernstein" <djb@cr.yp.to>]

Jerry Scharf, former Executive Director of ISC, writes:
> Having tests that forces people get thier nameservers right is a good thing.

Especially when the tests prohibit the perfectly valid default behavior
of DNS software that competes with yours, eh?

> Answering for the roots in case of an errant query seems reasonable to me,

Refusing to answer is also reasonable.

By the way, BIND 9 will also fail the nic.fr check, because nic.fr looks
for the root NS records in the _answer_ section of the packet. If your
reaction is ``well, they should check the other sections too,'' you are
missing the point. THEY SHOULD NOT EVEN BE ASKING THE QUESTION.

> If I turn off TCP, then put in a record that is too 
> long to be answered in 512 wire bytes, what happens?

If you have an answer that doesn't fit into 65535 bytes, what happens?
If you use a new record type that most BIND installations don't support,
what happens? If you use IP addresses in MX records, what happens? If
you mistype an IP address, what happens?

> fail to load the zone

Bad design.

---Dan