Re: [solace] Notes from the SOLACE ad-hoc at IETF85

[Carsten Bormann <[e-mail]>]

On Nov 5, 2012, at 10:17, Michael Richardson <[e-mail]> wrote:

> 
> Carsten Bormann <[e-mail]> wrote:
> 
>    CB> -- Is there anything done to secure group communications
>    CB> (multicast/broadcast)?  
> 
> I think that any specific session keying is out of scope.

Good thing we don't have a scope yet :-)
Good point, though.
But I have a slightly different view of how we should be running this first phase.

> This is about building trust between objects and network controllers.

(I don't know what a network controller is (we still need to define this kind of term) -- that would be one probably of the "roles" that we find out more about.)

I see the objective of soliciting the contributions as follows: to understand the lifecycle management in a number of solutions for specific scenarios down to a level of technical detail that we can learn from that and start to draw technical conclusions about them.

For the soutions, it is certainly interesting to hear whether they can manage trust relationships that make use of groups, and, if yes, how they do that.
So I would encourage the submitter to highlight this in their contributions.
Indeed, I would like to know how the solutions do the session keying, and/or how flexible they are with respect to the protocols they use for that.

Now I agree that the specific ways to do session keying are likely to be out of scope for any architecture we develop out of these contributions, but without an existence proof I would feel uneasy about any solution that claims to be able to manage lifecycles involving groups.

Grüße, Carsten