----- Original Message ----- 
From: "Dean Anderson" <[EMAIL PROTECTED]>


> How can the response be trusted, if the resolver did not perform
> validation?
>
> "Hi, I'm the president of the United States. See, I've set the 'response
> obtained through a secure channel' bit. Surely there is no need to
> validate my claim."
>

I was thinking of the case where a resolver has a secure channel set up with
a middlebox resolver that performs validation and sends back the result.
The end client could re-do the validation (if it is capable), or could trust
the results from the upstream validating resolver.  I should have been more
clear on that.

I am thinking how necessary it is for an application to know if it is
dealing with a resolver that cannot perform validation, but uses some other
trusted (or untrusted) method of resolving DNS queries.

Scott


> --Dean
>
> On Tue, 17 Feb 2004, Scott Rose wrote:
>
> > First off, I think this draft is long overdue - thanks for publishing
it.
> >
> > I was thinking that there might be a need for passing on the settings of
the
> > AD/CD bits, or a bit to indicate that the response was obtained through
a
> > secure channel.  So three bits of the array in total:
> >
> > x   AD bit set
> > x+1 CD bit set
> > x+2 response obtained through a secure channel
> >
> > I don't know if every application may care about this, but I can imagine
a
> > response array with the secure channel bit set, the AD bit set,
indicating
> > it could be trusted, even if the resolver did not perform validation
itself.
> >
> >
> > Scott
> >
> > .
> > dnsop resources:_____________________________________________________
> > web user interface: http://darkwing.uoregon.edu/~llynch/dnsop.html
> > mhonarc archive: http://darkwing.uoregon.edu/~llynch/dnsop/index.html
> >
>
>

.
dnsop resources:_____________________________________________________
web user interface: http://darkwing.uoregon.edu/~llynch/dnsop.html
mhonarc archive: http://darkwing.uoregon.edu/~llynch/dnsop/index.html

Reply via email to