[dns-operations] TC=1 with RA=0 from a recursive resolver

[Florian Weimer]

* Mark Andrews:

> In message <87io0f6i78.fsf at mid.deneb.enyo.de>, Florian Weimer writes:
>> * Mark Andrews:
>> 
>> > That's what RFC 1034 says.
>> >
>> >    - The recursion available, or RA bit, is set or cleared by a
>> >      name server in all responses.  The bit is true if the name
>> >      server is willing to provide recursive service for the client,
>> >      regardless of whether the client requested recursive service.
>> >      That is, RA signals availability rather than use.
>> >
>> > If RD effects RA then the server is broken.
>> 
>> But RFC 1034 does not reflect current practice.
>
> Nothing has changed this behaviour since RFC 1034 was written.

What I'm trying to say is that some (many?) currently deployed
recursors do not send RA=1 responses to RD=0 queries even if they
would have performed recursion for the same query with RD=1.

RFCs from the 1980s are rarely a good reference to determine current
operational practice.