On Fri, Aug 29, 2014 at 08:01:32AM -0400, Sam Hartman wrote:
> >>>>> "Juergen" == Juergen Schoenwaelder 
> >>>>> <[email protected]> writes:
> 
>     Juergen> At least, we should not confuse 'Abstract Service
>     Juergen> Interfaces', 'Subsystems', 'Models' and 'extension points'
>     Juergen> (which is a new concept since so far Models do not have
>     Juergen> such plugin extension points).
> 
> Hmm, I actually do think USM has several such extension points.
> There does seem to be an extension point for an authentication algorithm
> in the model already.
> 
> It's been my experience that adding security algorithms without such
> extension points does tend to cause problems both in interoperability
> because you tend to use much less pprecision when you don't need to
> define a clear extension point and in security because that lack of
> precision tends to lead to security analysis problems.  I've seen this
> both in the routing area and with core security protocols.
> 
> I don't have enough SNMP experience to figure out whether the results
> will be different here.
> 

Let me try to clarify my statement. I was trying to say is that we
should not use references to RFC 3011 architectural modularity in this
discussion since the RFC 3011 modularity concerns subsystems not what
happens in modules implementing subsystems. In particular, ISMS was
struggling with the fact that the subsystems did not forsee security
provided by the transport. What we are discussing here is different
from that.

/js

-- 
Juergen Schoenwaelder           Jacobs University Bremen gGmbH
Phone: +49 421 200 3587         Campus Ring 1, 28759 Bremen, Germany
Fax:   +49 421 200 3103         <http://www.jacobs-university.de/>

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

Reply via email to