> On 6 Jun 2017, at 19:55, Suresh Krishnan <[email protected]> wrote:
> 
> Hi Stephane,
>> 
>> It seems there was no reply to this DISCUSS?
> 
> Yes. There was no response to this DISCUSS.

Apologies, I indirectly responded to a similar comment another review (in that 
we would need to update the text given that there were several comments on it) 
but hadn’t proposed any new text. The revised I-D is still being worked on.


> On 6 Jun 2017, at 20:21, Stephane Bortzmeyer <[email protected]> wrote:
>> 
>> The strict profile requires the DHCP server to be “secure and
>> trustworthy” but declines to define what it means by “secure and
>> trusted”. So I cannot tell if it requires anything from current DHCP
>> servers or not.

Stephane was correct in that this section was not requesting or proposing 
anything specific with regard to DHCP, only outlining a potential future use 
case. 

I have proposed some minor modifications to Stephane’s text below which could 
make this even clearer, however if this isn’t sufficient then I think moving 
this to an appendix would be the best option. 

Sara. 



> Proposal for a new 7.3:

7.3 No configuration - dynamic discovery of ADN 

This section discusses the general case of a DNS client discovering both the 
authentication domain name and IP address dynamically. This is not possible at 
the time of writing by any standard means. However since, for example, a future 
DHCP extension could (in principle) provide this mechanism the required 
security properties of such mechanisms are outlined here. 

When using a Strict profile the dynamic discovery technique used as a source of 
authentication domain names MUST be
considered secure and trustworthy. This requirement does not apply when using 
an Opportunistic profile given the
security expectation of that profile. 

7.3.1 DHCP

In the typical case today, a DHCP server [RFC2131] [RFC3315] provides a list
of IP addresses for DNS resolvers (see section 3.8 of [RFC2132]), but
does not provide a domain name for the DNS resolver, thus preventing
the use of most of the authentication methods described here. (All
those that are based on a mechanism with ADN in table 2.)

This document does not specify or request any DHCP extension to provide 
authentication domain names. However, if one is developed in future work the 
issues outlined in section 8 of [RFC7227] should be taken into account as 
should the Security Considerations in section 23 of [RFC3315]) . 

This document does not attempt to describe secured and
trusted relationships to DHCP servers, which is a purely DHCP
issue (still open, at the time of writing.)  Whilst some
implementation work is in progress to secure IPv6 connections for
DHCP, IPv4 connections have received little to no implementation
attention in this area.




_______________________________________________
dns-privacy mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dns-privacy

Reply via email to