-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 07/23/2011 08:59 PM, Bruce Lowekamp wrote:
> On Fri, Jul 22, 2011 at 5:09 PM, Marc Petit-Huguenin <[email protected]> wrote:
> On 07/22/2011 01:54 PM, Bruce Lowekamp wrote:
>>>> On Fri, Jul 22, 2011 at 11:55 AM, Marc Petit-Huguenin <[email protected]> 
>>>> wrote:
>>>> On 07/01/2011 04:38 AM, Gonzalo Camarillo wrote:
>>>>>>> Folks,
>>>>>>>
>>>>>>> the P2PSIP WG chairs are in the process of requesting the publication of
>>>>>>> the following draft:
>>>>>>>
>>>>>>> https://datatracker.ietf.org/doc/draft-ietf-p2psip-base/
>>>>>>>
>>>>>>> At this point, they are working with the secretary to fix the state of
>>>>>>> the document in the tracker (it should be publication requested) and to
>>>>>>> upload the document's PROTO write up.
>>>>>>>
>>>>>>> While that gets fixed, I have done my AD review (see below) in order to
>>>>>>> gain some time. As soon as these comments are addressed, I will start
>>>>>>> the IETF LC on this draft.
>>>>>>>
>>>>
>>>> The IETF LC ends today but Michael Chen and I sent multiple questions to 
>>>> the
>>>> mailing-list that are still without answer.  I find it difficult to decide 
>>>> if
>>>> this specification is ready to be published until this questions are 
>>>> answered.
>>>>
>>>>
>>>>> Marc,
>>>>
>>>>> The other authors and I greatly appreciate the time you have put into
>>>>> implementing and evaluating the protocol, and into providing comments
>>>>> on the document.  However, it's difficult to keep track of them since
>>>>> they are generally in separate emails, so I would appreciate it if you
>>>>> could be specific listing any unanswered questions.  I've looked
>>>>> through my email archive and responded to the two questions (one from
>>>>> you and one from Michael Chen) that I found, both of which I think are
>>>>> simple to resolve in the text.   If there are any others, please list
>>>>> them so we can be sure to address them.
> 
> http://www.ietf.org/mail-archive/web/p2psip/current/msg05937.html
> 
> 
>> Thanks for pointing that out.  I didn't go back beyond the beginning
>> of last call when I looked yesterday.
> 
> 
> You may also want to look at my draft containing stuff I plan to publish as
> extensions or in RELOAD 2.0:
> 
> 
>> If you have a comment to make about the current draft, please make it
>> clearly on-list.  I sure wouldn't expect a 2.0 document anytime soon,
>> and if you believe there are issues that somehow need to be resolved,
>> you should be making clear technical statements about them.  I've read
>> that draft, but I'm not going to go through guessing what you think
>> you want to do later as extensions and what you think should be
>> changed in the current draft.

If there was something in it that I thought was necessary to the base draft, I
would have send it directly.

> 
> 
> http://www.ietf.org/internet-drafts/draft-petithuguenin-p2psip-reload-bis-00.txt
> 
> Version -01 will contains at least one additional section, here's the current
> draft (which is related to the TURN-SERVICE question):
> 
> <section title="Certificate Processing">
>  <t>
>    An original store is able to verify the signature of the items to store
> because the signer of the request is the same than the signer of the stored
> values, but it is not true for replica stores, where the request signer and 
> the
> stored value signer are different.
>    Even for the Access Control Policies where the certificate is stored at the
> same peer than the stored value, there is no guarantee that during a 
> replication
> the certificate will be available for verification of the value.
>  </t>
> 
>  <t>
>    This document adds a rule to the list in section 6.4.1.1 that says that 
> when
> processing a StoreReq, none of the modifications is visible to the validation
> process until the whole content of the StoreReq is validated.
>    This means for example that sending a certificate with a 
> CERTIFICATE_BY_NODE
> kind in the same StoreReq than a a value signed by this certificate will not
> permit to validate this value.
>    The certificate has to be stored before in a different Store request or has
> to be sent in the GenericCertificate field of the SecurityBlock of the 
> request.
>  </t>
> 
>  <t>
>    This document also requires that each peer stores all the certificates
> needed to verify all the stored values.
> 
>> None of this seems to add anything beyond what's already in 5.3.4,
>> that the certificates necessary to validate a request SHOULD be in the
>> message.  Quoting again from that section:
> 
>>   In systems which have some alternate
>>   certificate distribution mechanism, some certificates MAY be omitted.
> 
>> that's the exception, clearly stated.  Otherwise, the certs are included.

I can assure you that it is not what implementers understand (and I am not
talking about myself).  This is the problem when a SHOULD is used instead of a
conditional MUST.

> 
>    When replicating the stored values, the peer MUST also send the matching
> certificates in the GenericCertificate field of the SecurityBlock request.
> 
>> The exception was requested by some people with specific scenarios
>> where they had (or could get) certificates quite easily.  The
>> exception also applies to the shared-secret security mechanism in the
>> document (12.4).  Since that specific case where certs don't need to
>> be included is already in the base document, I would be against
>> changing the current approach by making including certs a MUST.

OK.

- -- 
Marc Petit-Huguenin
Personal email: [email protected]
Professional email: [email protected]
Blog: http://blog.marc.petit-huguenin.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iEYEARECAAYFAk4rHRoACgkQ9RoMZyVa61flsACfb10o/deRx4Q24yV/rE+KIjns
thYAnRdr8OeI1CWv1mpzZZM6ablxnpZJ
=UFHi
-----END PGP SIGNATURE-----
_______________________________________________
P2PSIP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/p2psip

Reply via email to