Hiya,

On 17/05/16 20:10, Eric Voit (evoit) wrote:
> + more

Looks like we're sorted. Just shoot out a rev with those
changes and I'll clear,

Thanks,
S.

> 
>> From: Stephen Farrell, May 17, 2016 2:24 PM
>>
>>
>> Hiya,
>>
>> On 17/05/16 18:45, Eric Voit (evoit) wrote:
>>> Hi Stephen,
>>>
>>>> Stephen Farrell, May 03, 2016 8:23 PM
>>>>
>>>> Stephen Farrell has entered the following ballot position for
>>>> draft-ietf-i2rs-pub-sub-requirements-06: Discuss
>>>>
>>>> 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 IESG DISCUSS and COMMENT positions.
>>>>
>>>>
>>>> The document, along with other ballot positions, can be found
>>>> here:
>>>> https://datatracker.ietf.org/doc/draft-ietf-i2rs-pub-sub-requirements/
>>>>
>>>>
>>>>
>>>>
>>>>
>> ----------------------------------------------------------------------
>>>> DISCUSS:
>>>> ----------------------------------------------------------------------
>>>>
>>>>
>>>>
>>>>
>> I have what I hope are two very easily sorted things that I'd like to
>> chat about:
>>>>
>>>> (1) 4.2.5, para2:
>>
>> It's para 3 now. Not sure if that changed in this rev or if
>> I goofed. If the latter, apologies;-)
> 
> Yes.   That is what caught me too.   I was answering for Paragraph 2.
>  
>>>> Hang on - this is 2016 isn't it? :-) Why would we
>>>> ever have a pub/sub service whose subscribers can pretend to be the
>>>> service?
>>>
>>> I am not sure I understand.  The Subscription Service is located on
>>> the Publisher.  Having the Publisher and Subscriber mutually
>>> authenticate each other makes sense.
>>>
>>> If you mean something else and you are wondering if whether a
>>> Subscriber can also be a Publisher for the information it receives,
>>> yes it can.  Having a collector/controller as a middle-man is a very
>>> valid implementation.  But there is nothing which binds such
>>> independent subscriptions together.
>>
>> The problem is that the text says "it SHOULD be possible to provide
>>    cryptographic authentication in such a way that no Subscriber can
>>    pose as the original Subscription Service"
>>
>> I see no reason for that SHOULD. The implication I take from it
>> is that someone wants to use purely symmetric keying which seems
>> pretty odd. I'd say better text would be to say that "Subscribers
>> MUST NOT be able to pose as the original Subscription Service"
>> maybe.
> 
> Your text is better.   Changed.
> 
>>>
>>>> (2) Don't you need a statement somewhere that commensurate security
>>>> needs to be provided for pushed notifications as was used for the
>>>> original subscription? That might be a little hard to phrase
>>>> correctly but I hope we agree that the notifications ought not be
>>>> significantly less secure than the subscription.
>>>
>>> We had some interactions with Ben Campbell on this topic.    In
>>> general we are doing our best to decouple the subscription
>>> establishment and maintenance from the underlying transport options.
>>> This is because there are implementations (for example wholly within
>>> an MSDC) which don't require payload encryption.  Still most
>>> implementations should have transport encryption.  So after several
>>> interactions with Ben, the text leading off Section 4.2.5 now reads:
>>>
>>> Some uses of this Subscription Service will push privacy-sensitive
>>> updates and metadata.  Good deployment practices will bind this
>>> generated information within secure, encrypted transport layer
>>> mechanisms.  For example if NETCONF is used as transport, then
>>> [RFC5539] would be a valid option to secure the transported
>>> information.  The Subscription Service can also be used with
>>> emerging deployment contexts as well.  As an example, deployments
>>> based on [i2rs-usecase] can apply these requirements in conjunction
>>> with those documented within [i2rs-protocol-security] to secure
>>> ephemeral state information being pushed from a Network Element.
>>>
>>> Does this hit your objective?
>>
>> Not quite, though it is related.
>>
>> If you added "Commensurate strength security mechanisms MUST
>> be defined (and SHOULD be used) for all of the stages of
>> the pub/sub process, for example using the same algorithms
>> and endpoints for keying for both subscription and publication
>> stages would seem advisable."
>>
>> The thing to avoid is e.g. using some really strong mechanism
>> for the subscription but something much weaker or with worse
>> endpoints for the publication (or vice versa).
>>
>> Sorry if that wording is a bit vague, I didn't re-read the
>> entire doc to get the context right.
> 
> I see what you are getting at now.   How about the requirement text:
> 
> "For any encrypted information exchanges, commensurate strength security 
> mechanisms MUST be available and SHOULD be used.  This includes all stages of 
> the Subscription and update push process."
> 
> Thanks,
> Eric
> 
> 
>  
>> Cheers,
>> S.
>>
>>>
>>> Eric
>>>
>>>>
>>>> ----------------------------------------------------------------------
>>>>
>>>>
>> COMMENT:
>>>> ----------------------------------------------------------------------
>>>>
>>>>
>>>>
>>>>
>> - I wondered if this was maybe of interest to more than just i2rs, and
>> if so,
>>>> whether any effort had been made to try figure out if these
>>>> requirements work for folks who don't care about i2rs? It'd seem a
>>>> shame to work on this but stop one step short of being
>>>> appropriately general. (But you probably already checked that I
>>>> guess.)
>>>>
>>>> - 4.2.2, last para: The MUST here seems like it may be quite
>>>> onerous, in general. Did someone think all of that through? For
>>>> example, what if the reason for declining is that the Subscriber
>>>> doesn't have sufficient privilege? Saying what privilege is needed
>>>> would be a breach of least-privilege. Transient errors may also
>>>> make this very hard to do well. I'd suggest s/MUST/MAY/ and to also
>>>> turn the information returned into a hint and not a promise.
>>>>
>>>> - 4.2.5, para 1: saying there "MUST be mutual authentication" is
>>>> odd - the usual terms would be "MUST implement" or "MUST use" which
>>>> of those does "MUST be" mean?
>>>>
>>>> - 4.2.8: when you say fetch... by whom? Is there an implicit
>>>> requirement in the title of the subsection?
>>>>
>>>>
>>>> _______________________________________________ i2rs mailing list
>>>> [email protected] https://www.ietf.org/mailman/listinfo/i2rs
> 

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

_______________________________________________
i2rs mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/i2rs

Reply via email to