Re: [secdir] secdir review of draft-housley-suite-b-to-historic-04

[Russ Housley <[e-mail]>]

> On Apr 24, 2018, at 11:08 AM, Paul Wouters <[e-mail]> wrote:
> 
> On Tue, 24 Apr 2018, Taylor Yu wrote:
> 
>> "7.  Security Considerations
>> 
>>  The CNSA Suite includes algorithms using the larger key sizes that
>>  are included in Suite B.  There are no interoperability or security
>>  concerns raised by reclassifying the Suite-B-related RFCs to Historic
>>  Status."
> 
> This text is interesting as I see two statements by the US government
> on the CNSA Suite.
> 
> The first one is at:
> 
> https://www.iad.gov/iad/programs/iad-initiatives/cnsa-suite.cfm
> 
> Which claims the CNSA Suite is not published yet, but there is a
> transition set of algorithms to use.
> 
> The second one is at:
> 
> https://www.iad.gov/iad/library/ia-guidance/ia-solutions-for-classified/algorithm-guidance/cnsa-suite-and-quantum-computing-faq.cfm
> 
> Which lists the same algorithms, but no longer calls these a transition
> list but the actual CNSA Suite.
> 
> Regardless, since I think the IETF should not be advertising a single
> governments crypto standards, I think this document would be better of
> not mentioning CNSA in Section 7 at all. We shouldn't have stamped Suite B
> in the IETF either, but we were young and naive at the time. So let's
> not stamp CNSA as a "successor" in any kind of way. The only thing this
> document should do is move Suite B to Historic.
> 
> I would suggest something like this for Section 7:
> 
> 	The algorithms and key sizes from Suite B, where these algorithms
> 	and key sizes were published by the IETF, have been obsoleted or
> 	updated by new and more secure algorithms and key sizes. Please
> 	see the respective IANA registries and RFC updates for the
> 	specific algorithm usage within their specific protocols.

I think that last sentence of the existing Security Considerations should remain:

   ... There are no interoperability or security
   concerns raised by reclassifying the Suite-B-related RFCs to Historic
   Status.

In fact, I think it should be said first.

> I'm fine with mentioning CNSA in Section 2.

Good.  I think it is useful information.

Russ