[dns-operations] extra records in resolver answer, any benefit?

[Marek Vavruša]

On 27 January 2015 at 11:46, bert hubert <bert.hubert at netherlabs.nl> wrote:
> On Tue, Jan 27, 2015 at 09:40:42PM +1100, Mark Andrews wrote:
>> > It is all optional, and nobody does anything with that data. In fact stub
>> > resolvers do very little with what they receive. So for example, even the
>> > additional processing for an MX record is completely ignored mostly.
>>
>> That is really dependent on the application.
>
> Can you name me one client side application that benefits from anything
> other than the answer section?
>
>> > It also saves CPU for the resolver.
>>
>> Named will save, validate and return additional data without
>> performing additional queries if it is signed (assuming the DNSKEY
>> records exist in the cache and have already been validated).
>
> It will still be digging through caches and creating larger packets though.
>
>         Bert

That is what I think as well.  But both Named and NSD do this, and I'm
struggling to understand the original motivation for this.
I wonder if anyone could share how their traffic really looks like,
this could give us a better idea of what is actually useful for
clients.
The recursive resolution is a wee bit under-documented (compared to
authoritative), so I'd like to lean on the op experience in this.

Thanks!

Marek