RE: draft-ietf-ccamp-lmp-02.txt

["Wijnen, Bert (Bert)" <bwijnen@lucent.com>]

Further comments (I have not checked the whole document,
just reporting as I see thibgs while looking for specific
stuff).

The IANA considerations section is setting aside 4% of the
name space for Message Type and 8% of Object Class name space for
OIF UNI. Is that the proper thing to do?

I understand OIF has already "provisionally claimed" a few
of the "reserved" numbers. We can honor that while we move forward.
But I think it would be better to keep the whole name space assignments
in one place, and under control of IANA.

The OIF (and other organisations) can then request assignments through
the normal process, similar to how 
  http://www.ietf.org/internet-drafts/draft-bala-uni-signaling-extensions-00.txt
is trying to do right now.

By the way, you use these terms:

  - Message Type
  - Msg Type
  - MsgType

Probably better to settle for one "spelling" and be consistent I think

In IANA Considerations section you also talk about
"class and class type name spaces" while you only elaborate on the
control of the "Object Class Name Space". So are there multiple names spaces
for class and class-type or just one. There seem to be two, i.e. Object Class
and Class Type (within a class). Does the Class Type name space not have
IANA considerations?

Bert 

> -----Original Message-----
> From: Wijnen, Bert (Bert) 
> Sent: Tuesday, February 26, 2002 11:52 AM
> To: ccamp@ops.ietf.org
> Subject: draft-ietf-ccamp-lmp-02.txt
> 
> 
> From Kireeti's "wg document status" email
> 
> > -----Original Message-----
> > From: Kireeti Kompella [mailto:kireeti@juniper.net]
> > Sent: Monday, February 25, 2002 2:52 AM
> > To: ccamp@ops.ietf.org
> > Subject: WG dcoument status
> > 
> > 
> > Here's a status update.
> > 
> ... snip ...
> 
> > The LMP draft:
> > 	draft-ietf-ccamp-lmp-02.txt
> > has gone through one round of WG Last Call comments and, once a
> > new version has been produced incorporating these comments, will
> > go through a final WG Last Call.  This is also targeted as a
> > Proposed Standard.
> > 
> I will note that in my view, the security section will need
> serious work. I doubt that the Security ADs will sign off on the
> current text. The security section should address the risks
> and treats. It should then specify how to protect against them.
> A text like "LMP exchanges may be authenticated with MD5" (which
> is basically what you write now) seems not sufficient to me.
> 
> Bert
>