Re: Issue: Add a new QTYPE

[Michael Milligan <milli@acmeps.com>]

Paul Vixie wrote:
> > We actually discussed and rejected the additional processing suggestion
> > -- it was present in the original draft of 1886. The argument was pretty
> > much the one Robert just gave: we will have to remove the processing
> > requirement eventually. It was also not clear that it would be useful in
> > most cases.
>
>
> i believed i had shown during the debate over 1886, and i believe i have
> shown again here today, that the lifetime of the data will be the same
> as the lifetime of its usefulness.  this really is not a bad idea, folks.

I agree.  This seems to solve the problem nicely and is in the same vein 
as adding A records to the additional data section when MX records are 
asked for.  Think of it as an optimization.  Resolvers only have to know 
to look for the A records and not start another query.  Once all A 
records go away, there's just an extra bit of code path taken that looks 
for records that won't be found.  And in the case of BIND, the execution 
of that extra code path can eventually go away like IQUERY did from a 
configuration standpoint.

Regards,
Mike

--
Michael Milligan                                   -> milli@acmeps.com


--
to unsubscribe send a message to namedroppers-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/namedroppers/>