Re: [Anima-signaling] Comments needed: issues in draft-ietf-anima-grasp-09

[Brian E Carpenter <[e-mail]>]

On 03/03/2017 03:15, Michael Richardson wrote:
> 
> Brian E Carpenter <[e-mail]> wrote:
>     > - Split the document? [Charlie Perkins]
> 
>     > "parts of the document seem more philosophical than
>     > prescriptive... It should be considered to break the document
>     > into a Requirements document and a more rigorously defined
>     > protocol solution document."
> 
>     > Proposed resolution: writing a separate requirements document
>     > was essentially excluded when the WG was chartered. Unless the
>     > WG and AD want to backtrack on that, the proposed resolution
>     > is to *not* do this. Of course, all the specific review comments
>     > about non-rigorous text will be actioned.
> 
> Mark the parts of the document which are requirements.
> not having seperate documents doesn't mean that we can't be clear.

It's already a separate and suitably titled section. I think we're fine.

> 
>     > - Clarify security [Charlie Perkins]
> 
>     > "In some
>     > places, ACP seems to be mandated, and in other places that is relaxed
>     > to mean "a sufficient security mechanism".  It would be better to
>     > identify the security requirements, and put them unmistakably in the
>     > Security Considerations section, which deserves to have teeth."
> 
>     > (and various detailed comments in the text)
> 
>     > Of course we will deal with the detailed comments. The larger
>     > issue is whether we should move most of the security discussion
>     > to the Security Considerations section.
> 
>     > Proposed resolution: *Waiting for WG Chair and AD guidance.*
> 
> I think I already expressed my opinion:  "must be secured by ACP".
> I would like no other exceptions or allusions to DTLS, etc.

I'm not convinced we have WG consensus for that. I think that when
we've tightened up the text in several ways suggested by Charlie,
at least the intention will be clear.

> 
>     > - Text on circular dependencies [Joel Halpern]
> 
> ...
>     > Proposed resolution: We should be more explicit that this is a complex
>     > issue that is simply not handled by the GRASP layer and that must be
>     > considered by ASA designers.
> 
> +1
> 
>     > Proposed resolution: No change.
> 
>     > "3.8.13.  No Operation Message
> 
>     > In fragmentary CDDL, a No Operation message follows the pattern:
> 
>     > noop-message = [M_NOOP]
> 
>     > This message MAY be sent by an implementation that for practical
>     > reasons needs to activate a socket.  It MUST be silently ignored by a
>     > recipient.
> 
>     > CEP: ?? what does it mean to "activate" a socket?  Is there something
>     > about
>     > how long a socket may remain open but not used?"
> 
>     > Not that I know of. But I discovered that to find out the port number
>     > assigned to a socket, you pretty much have to perform sendto() and
>     > getsockname(), so the ability to send a no-op during initialisation
>     > seemed reasonable. I could have just sent a zero, but a well-defined
>     > message seemed better.
> 
> Also, we may need traffic to keep ACP links alive, or discovery that they are
> in fact dead.  Or just for debugging.  It just seems useful to have to me.

Yes. It only got added because I needed to send one to make my code work ;-).

   Brian