Re: [Last-Call] Rtgdir last call review of draft-ietf-idr-rfc7752bis-11

Ketan Talaulikar <ketant.ietf@xxxxxxxxx> · Mon, 7 Nov 2022 18:20:11 +0530

Hi Joel
Please check inline below for responses with KT2.

On Mon, Nov 7, 2022 at 4:58 PM Joel Halpern <jmh@xxxxxxxxxxxxxxx> wrote:

    Slightly trimmed, two comments.  Joel

    On 11/7/2022 1:03 AM, Ketan Talaulikar
      wrote:

        Hi Joel,

          Thanks for your review and please check inline below for
            responses.

          On Thu, Oct 27, 2022 at 7:32
            PM Joel Halpern via Datatracker <noreply@xxxxxxxx>
            wrote:

          ..

                None

            Nits:

              At the end of the first paragraph of section 4, could we
            add a sentence

              saying "the BGP-LS attributes appear within the
            corresponding new BGP NLRI"

              or similar?  While that is explained later in section 4,
            the length of the

              section means that a new reader is left wondering for
            quite some time.

          KT> The BGP-LS Attribute TLVs do not appear within the
            NLRI. Sec 4.2 introduces BGP-LS NLRIs and indicates that
            they are carried within the MP_REACH/UNREACH. Further Sec
            4.3 introduces the BGP-LS Attribute and indicates that they
            are carried as part of the BGP update along with the Link
            State NLRIs (and other attributes). Perhaps we can clarify a
            bit upfront that the new NLRI types are carried in the
            MP_REACH/UNREACH? 

    <jmh> Apparently I still managed to misread it.  Sorry.  It
      would help if you could put a little more context at the front of
      section 4. </jmh>

KT2> How about the following?

   The link-state information is carried in BGP UPDATE messages as:
   (1) BGP NLRI information carried within MP_REACH_NLRI and 
   MP_UNREACH_NLRI attributes that describes link, node, or 
   prefix object, and (2) a new BGP path attribute (BGP-LS
   Attribute) that carries properties of the link, node, or 
   prefix objects such as the link and prefix metric or 
   auxiliary Router-IDs of nodes, etc..

             Section 4.1 has the paragraph:

               All TLVs within the NLRI that are not specified as
            mandatory are

               considered optional.  All TLVs within the BGP-LS
            Attribute are

               considered optional unless specified otherwise.

              As far as I can tell, those two sentences are saying,
            about two different

              aspects of the encoding, the same thing.  But they say it
            in different ways. 

              If there is some subtle difference in meaning taht is
            intended, please

              clarify.  If the meaning is indeed the same, could we use
            parallel

              construction to avoid readers thinking there is a
            difference?

          KT> They are talking about two different "containers"
            - the NLRI and the BGP-LS Attribute. The "default" is
            different for them.

    <jmh>I am having trouble parsing the distinction, even if the
    external default is different.  The two sentences both seem to say
    that the only mandatory items are those which are specified as
    mandatory.  They simply say it in two different ways.  Why is the
    sentence construction different?  If there is an actual difference
    in intended effect, I think a few more words would be helpful.
    </jmh>

KT2> How about the following?

   For both the NLRI and BGP-LS Attribute parts, all TLVs are 
   considered as optional except where explicitly specified as
   mandatory or required in specific conditions.

Thanks,
Ketan

          Thanks,
          Ketan

-- 
last-call mailing list
last-call@xxxxxxxx
https://www.ietf.org/mailman/listinfo/last-call