Re: [Dtls-iot] DTLS multicast security

[Michael StJohns <[e-mail]>]

On 9/19/2014 1:10 PM, Dorothy Gellert wrote:
> HI Mike-
>
> What we want to do as Chairs is to reach consensus on how to progress 
> the work items.   If there consensus to drop the work item,  that 
> would be progress.
>
> If the consensus of the group is to rework the Charter Item, so that 
> it does not depend on DTLS, we can also do that.    So far there is 
> willingness to re-work the charter item.
>
> Can we get more feedback from the WG?

Hi Dorothy -

My preference is to remove this entirely from the WG - I don't think its 
a good fit for DTLS.

If the group wants to do asymmetric control multicast either at CoAP or 
as a profile of one of the other existing multicast protocols, I 
wouldn't resist modifying the charter to do the work here.

If the group continues to want to do symmetric control multicast, I 
would continue to argue strongly against doing it here (or for that 
matter anywhere in the IETF) for all of the reasons I've reiterated many 
times.

Later, Mike

>
> Thanks,
> Dorothy
>
>
> On Sep 19, 2014, at 8:01 AM, Michael StJohns <[e-mail] 
> <mailto:[e-mail]>> wrote:
>
>> On 9/19/2014 7:27 AM, Rahman, Akbar wrote:
>>> Hi Dorothy,
>>>
>>>
>>> I agree with Ludwig that having a secure multicast is considered a 
>>> benefit by many.
>> My problem with this statement is that I would consider anti-gravity 
>> to be a benefit to many, but that AG has exactly the same scientific 
>> basis as symmetric key multicast security - none.  There's long, long 
>> experience on this topic that the document writers have ignored.
>>
>>>   For example, during the recent IESG review of the base CoAP Group 
>>> Communication spec there were several comments made by AD's 
>>> reflecting the need for a secure multicast solution to be developed 
>>> by IETF.  See for example:
>>>
>>> http://www.ietf.org/mail-archive/web/core/current/msg05566.html
>>>
>>> "The lack of security controls is an issue, experimental
>>> would be good until it is resolved as there is a lot of work to be done
>>> in this space and it is active."
>>
>> I think you're misrepresenting that message.  It's by Kathleen 
>> Moriarty.  She notes the lack of security in the body of the message, 
>> but the comment on "experimental" isn't on security specifically, but 
>> on the whole idea of CoAP group communications. Cf the other messages 
>> which object to the document as informational without mentioning 
>> security.
>>
>>>
>>>
>>> So, I think we still need to have a Work Item to develop a secure 
>>> group communication solution.  However, perhaps we can modify the 
>>> description of the Work Item and not have it exclusively linked to a 
>>> DTLS-based approach for secure group communication.  We should allow 
>>> for other approaches if people want to propose them.  But we should 
>>> still definitely keep working on this topic (i.e. secure group 
>>> communication).
>>
>> There's two things here:  1) The group is supposed to be profiling 
>> work done elsewhere to shrink it for use with IOT, not creating new 
>> stuff; 2) My objections to secure multicast are specifically in the 
>> area of the use of multicast as a control protocol; symmetric key 
>> systems are NOT secure enough for control systems and there appears 
>> to be deep and abiding resistance to the use of asymmetric systems 
>> (e.g. signed control messages) leading us to an impasse.
>>
>> I agree that DTLS is probably not the appropriate protocol for signed 
>> control messages, but there also seems to be a deep and abiding 
>> resistance to adding it to CoAP where it might make the most sense.
>>
>> Dorothy has proposed the withdrawal of multicast DTLS and I think 
>> that's the correct decision.   If someone wants to propose an 
>> asymmetric system that works with CoAP and run it through this group, 
>> I won't object (but the AD's might given the current charter).
>>
>>>
>>> A separate thought is that we may also want to progress the 
>>> existinghttp://datatracker.ietf.org/doc/draft-keoh-dice-multicast-security/but 
>>> put it on an Experimental track.  That way we can get experience 
>>> with the solution but not put it directly on Standards track.
>>
>> Instead, place it as a company informational like hundreds of other 
>> documents.  Philips can provide experimental results in a year or so. 
>>  There's only a reason to place it on the experimental track if more 
>> than one company is planning on using it and modifying it.
>>
>> Later, Mike
>>
>>
>>>
>>>
>>> Best Regards,
>>>
>>>
>>> Akbar
>>>
>>> -----Original Message-----
>>> From: dtls-iot [mailto:[e-mail]] On Behalf Of 
>>> Ludwig Seitz
>>> Sent: Friday, September 19, 2014 2:45 AM
>>> To: [e-mail] <mailto:[e-mail]>
>>> Subject: Re: [Dtls-iot] DTLS multicast security
>>>
>>> On 09/18/2014 10:41 PM, Dorothy Gellert wrote:
>>>> Dear WG,
>>>>
>>>> Last week our AD and the WG chairs, myself and Zach, met to discuss 
>>>> the progress of the DTLS multicast security Work Item.
>>>> it seems as though we have reach an impasse with regards to the 
>>>> issues raised on the mailing list with multicast security and DTLS.
>>>>
>>>> If this is the consensus of the WG  we can progress the WG without 
>>>> this Work item and move forward with the other 2 work items, the 
>>>> dtls profile and practical issues around the DTLS handshake.
>>>>
>>>> I'd like to request feedback from the WG on this plan.
>>>>
>>>> Thanks,
>>>> Dorothy
>>>>
>>> When making a decision on this, please note that secure multicast 
>>> would be considered a considerable benefit by some. See e.g.
>>> http://www.ietf.org/mail-archive/web/ace/current/msg00826.html
>>>
>>> Regards,
>>>
>>> Ludwig
>>>
>>> --
>>> Ludwig Seitz, PhD
>>> SICS Swedish ICT AB
>>> Ideon Science Park
>>> Building Beta 2
>>> Scheelevägen 17
>>> SE-223 70 Lund
>>>
>>> Phone [phone]
>>> http://www.sics.se
>>>
>>> _______________________________________________
>>> dtls-iot mailing list
>>> [e-mail]
>>> https://www.ietf.org/mailman/listinfo/dtls-iot
>>
>> _______________________________________________
>> dtls-iot mailing list
>> [e-mail] <mailto:[e-mail]>
>> https://www.ietf.org/mailman/listinfo/dtls-iot
>