[dns-operations] DNS software test!

[Phil Regnauld]

Stephane Bortzmeyer (bortzmeyer) writes:
> > 	Testing one tool at a time, on a known dataset would be the
> > 	way to go for recursion (unbound, bind) and authoritative
> > 	(bind, nsd).
> 
> For an authoritative name server, I fully agree with you. However, for
> a recursive, it is more complicated: the performances of an isolated
> recursive name server are not interesting, unless you access
> already-cached data.

	Exactly.

> I'm not sure we have a nice, sound and realistic methodology to test
> recursive name servers.

	Fake root + ensure that you're loading the same data set on a number of 
	virtual upstream auth servers, with a large number of zones. May not
	look very real, but at least you're testing against the same data set
	every time.