Hi Brian,

Apologies for the late response.

On 3/31/17 6:40 AM, Brian Weis (bew) wrote:
> I support advancing this document, and have the following minor comments.
>
> (1) Section 1.3. LLDP should be referenced at first use. The wording
> at the beginning of Section 11 is nice: "IEEE802.1AB Link Layer
> Discovery Protocol (LLDP) [IEEE8021AB]”.

Fixed.
>
> (2) Section 1.4. An interesting policy example is:
>
> Allow access to host controller.example.com
> <http://controller.example.com> with QoS AF11


>
> I don’t see how QoS markings are defined in MUD. If not, then “ with
> QoS AF11” should be removed. Supporting QoS markings would be a good
> idea though.

Removed for now.  This might be something that we use an extension for.

>
> (3) Section 1.4 and beyond. The word “authority”, “authority
> component”, “authority section” are all used to mean the same thing,
> but without further explanation as to what this means. The
> inconsistency is confusing. But more importantly, it isn’t clear to
> the reader in those earlier sections that each “authority *" is a
> forward reference to the “authority element” in the URL definition
> (Section 5)  It would clearer if these references said something like
> “authority element of the MUD URL”.
I normalized to "authority component" where appropriate, and added a
brief definition.  That component is a domain name, roughly speaking.

>
> (4) Section 1.6. Regarding this text: "In the case of IEEE 802.1X, the
> switch would tunnel the URI via a certificate ….”.  The URI isn’t
> really “tunneled”. How about, "In the case of IEEE 802.1X, the switch
> would present a certificate containing the URI …”.

Corrected.
>
> (5) Section 3.2. For "it should not be necessary to resign a MUD file
> when a new one is released”, I believe you mean that the MUD file does
> not not need to be signed again (“re-sign”). Or do you really mean
> that the manufacturer will “resign” (e.g., “leave” or “stand down”)
> the MUD file by moving it to the archival location? If so, this needs
> to be clearer.
Right.  Re-sign.
>
> (6) Section 3.12. This section mentions "standard classes”. There’s
> should be a reference to what those classes would be, or more text
> describing what is a “standard class”. Should there be an IANA
> registry of “standard classes”?

Yes.  I don't actually have a registry for this, and maybe we need one.

Regards,

Eliot
>
>

Attachment: signature.asc
Description: OpenPGP digital signature

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

Reply via email to