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

[Mark Andrews]

In message <56F05A95.8080107 at dougbarton.email>, Doug Barton writes:
> On 03/21/2016 07:24 AM, Tony Finch wrote:
> > bert hubert <bert.hubert at powerdns.com> wrote:
> >>
> >> On a philosophical note, the glibc position is defensible but since both
> >> Google DNS and dnsdist made the mistake of setting RA=RD *after* judging i
> f
> >> the question merited a TC=0 response or not means it may be better to be
> >> 'liberal in what you accept'.  Others might make the same mistake.
> >
> > Shouldn't RA be set independently of the query, purely based on the server
> > configuration? If I make a RD=0 query against a recursive server I should
> > get an RD=0 RA=1 response.
> 
> FWIW, I was looking at the problem the same way Tony did here.

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.

Mark

> That said, if you get TC=1 it's fair for the resolver to retry even if 
> RA=0.
>
> Doug
> 
> _______________________________________________
> dns-operations mailing list
> dns-operations at lists.dns-oarc.net
> https://lists.dns-oarc.net/mailman/listinfo/dns-operations
> dns-jobs mailing list
> https://lists.dns-oarc.net/mailman/listinfo/dns-jobs
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742                 INTERNET: marka at isc.org