Re: [dns-privacy] Roman Danyliw's No Objection on draft-ietf-dprive-xfr-over-tls-11: (with COMMENT)

[Roman Danyliw <[e-mail]>]

Hi Sara!

> -----Original Message-----
> From: iesg <[e-mail]> On Behalf Of Sara Dickinson
> Sent: Thursday, May 6, 2021 6:13 AM
> To: Roman Danyliw <[e-mail]>
> Cc: [e-mail]; [e-mail]; The IESG <[e-mail]>; draft-
> [e-mail]; [e-mail]
> Subject: Re: Roman Danyliw's No Objection on draft-ietf-dprive-xfr-over-tls-11:
> (with COMMENT)
> 
> 
> 
> > On 4 May 2021, at 22:44, Roman Danyliw via Datatracker <[e-mail]>
> wrote:
> >
> > Roman Danyliw has entered the following ballot position for
> > draft-ietf-dprive-xfr-over-tls-11: No Objection
> >
> > When responding, please keep the subject line intact and reply to all
> > email addresses included in the To and CC lines. (Feel free to cut
> > this introductory paragraph, however.)
> >
> >
> > Please refer to
> > https://www.ietf.org/iesg/statement/discuss-criteria.html
> > for more information about DISCUSS and COMMENT positions.
> >
> >
> > The document, along with other ballot positions, can be found here:
> > https://datatracker.ietf.org/doc/draft-ietf-dprive-xfr-over-tls/
> >
> >
> >
> > ----------------------------------------------------------------------
> > COMMENT:
> > ———————————————————————————————————
> 
> Hi Roman,
> 
> Many thanks for the review.
> 
> >
> > Section 1.  s/can make reconnaissance trivial/can make reconnaissance
> > and attack targeting easier/
> 
> Yes.
> 
> >
> > Section 1.  Per “… zone walking is now possible with NSEC3 due to
> > crypto-breaking advances”, a reference here would be helpful.
> 
> Agreed and requested by another reviewer - let me dig a good one out.
> 
> >
> > Section 1.  As far as I can tell the work on draft-vcelak-nsec5 has
> > not been adopted and the draft is expired.  Perhaps this should be
> > signaled via s/This has prompted further work on an alternative
> > mechanism/This promoted work on an alternative mechanism/
> 
> That seems reasonable.

Thanks for all of the above edits.

> >
> > Section 1.  Per “It is noted that in all of the common open source
> > implementations …”, as this is a point in time assessment, it would be
> > helpful to at least mention parenthetically the
> > implementations/version numbers assessed informally for this conclusion.
> 
> Another review suggested just adding “(at the time of writing)” to qualify that
> statement. Would that be enough (there are at least 5 implementations we
> could name)?

That works for me.

> >
> > Section 1.  Editorial.  “… must cater for accepting …” doesn’t parse for me.
> 
> “must therefore accept”?
>
> >
> > Section 4.
> >
> >   The threat model does not, however, consider the existence of a zone,
> >   the act of zone transfer between two entities, nor the identities of
> >   the nameservers hosting a zone
> >
> > To further document the assumptions, consider adding that this threat
> > model doesn’t consider/protect the mechanisms to decide on triggering
> > the zone transfer (e.g., protecting NOTIFY messages from an active
> > attacker)
> 
> Thats a reasonable point - I’ll add it. FYI - some operators do protect the NOTIFY
> with a TSIG for _some_ added protection.
>
> >
> > Section 6.2.  Per “However it is noted that most widely used open
> > source authoritative nameserver implementations (including both [BIND]
> > and [NSD]) do IXFR using TCP by default in their latest releases”, as
> > this document ages, “latest release” may not be meaningful.  Consider
> > providing a version number for [BIND] and [NSD].
> 
> Yes - that makes sense here.

Thanks.

> >
> > Section 8.4 and 10.4.  As already mentioned by Martin and John -- It
> > seems like a strong statement to say that IP ACLs are in the same
> > class of “channel authentication” as mTLS.
> 
> Hopefully addressed in the thread with Ben.

No problem.  I know a number of us made a related or identical comment.  I'll follow along in the DISCUSS made by Ben.

> >
> > Section 8.8.1.  It’s difficult to assess how effective this notional
> > padding approach would be for providing traffic analysis protection.
> > A few thoughts on the existing text realizing the details are out of scope:
> >
> > -- Does padding for AXoT need to be coordinated with the padding on IXoT?
> 
> I think that any zone that uses IXoT and pads would also always apply AXoT
> padding (because IXoT can fall back to AXoT). It would expect the draft on the
> specific padding policy would address that in more detail.
> 
> >
> > -- Is keeping state required to ensure that padding provides the
> > appropriate obfuscation over time?
> 
> Interesting question. Are you thinking about AXoT where the zone could e.g.
> grow then shrink? If so, that does seem like a good idea (again - input for the
> follow up padding policy draft - thanks!).

Exactly. Changes in sizes would be a prime traffic analysis metric.  This obfuscation would have to be consistently maintained to provide protection (to the degree that this obfuscation is robust) against a persistent observer who can take multiple measurements at different points in time.

With this comment and the one above it, I appreciate there is a fine line between balancing what's out of scope for a future, detailed specification; and adding design considerations or cautions in a notional architecture specified here.  I leave it to you and the rest of author team to assess the right balance as these were just (optional) ballot comments.

Regards,
Roman

> Thanks and regards
> 
> Sara.