Re: [CCAMP] Still have issues in WSON Processing HOP Attribute Encoding in draft-ietf-ccamp-wson-signaling-08

Hi,

Lou, here's my comment on your comment. In a nutshell replacing [RSVP-RO] with [RFC5420] will solve the confusion. 

Please see in-line for details.

Thanks,

Young

-----Original Message-----
From: CCAMP [mailto:ccamp-bounces@ietf.org] On Behalf Of Lou Berger
Sent: Saturday, August 02, 2014 10:30 AM
To: CCAMP; draft-ietf-ccamp-wson-signaling@tools.ietf.org
Subject: [CCAMP] Still have issues in WSON Processing HOP Attribute Encoding in draft-ietf-ccamp-wson-signaling-08

All, (WG, Authors, Young, Giovanni),
	It looks like an unintended encoding change made it into -08 of the draft, or there remains some confusion in the intended encoding.  The topic traces back to a discussion in March, see http://www.ietf.org/mail-archive/web/ccamp/current/msg15894.html.  To move forward, I'd like to ensure that the authors and any other interested party, i.e., existing and future implementors, are on the same page.

I'd like to focus on the intent and the what document should says as opposed to how we got here, i.e., past confusion.  We need to ensure that the draft accurately documents the Authors/WG intent.  I am not trying to propose any technical changes at this very late date.

YOUNG>> Agree. 

The area of the document in question is sections 4.2-4.4 and relates to how the so called sub-TLVs are encoded and processed.  The key text in question, from rev -07 is:
   To target a specific node, this section defines a WSON_Processing
   object as part of the LSP_REQUIRED_ATTRIBUTE and follows procedures
   defined in [RSVP-RO].

Looking back on the -07 and the March mail thread, the "sub-TLVs"
defined in Section 4.3 and 4.4 should be carried as Attribute TLVs in LSP_REQUIRED_ATTRIBUTES Objects, and not in ERO_HOP_ATTRIBUTE Subobjects. Also that there actually is no WSON_Processing object carried on the wire.

Authors, Young, Giovanni:

    Please confirm if this is correct.  If not, please clarify what was
    intended in the -07 rev of the document.

YOUNG>> Correct. It looks like there was a confusion between [RFC5420] and [RSVP-RO]. It seems that it makes no sense associating LSP_REQUIRED_ATTRIBUTES Object with [RSVP-RO] as [RSVP-RO] talks only about ERO_HOP_ATTRIBUTE subobject which is completely irrelevant with LSP_REQUIRED_ATTRIBUTES object. 

Assuming it is correct I think it remains unclear what it means for a LSP_REQUIRED_ATTRIBUTES Object (and not an ERO_HOP_ATTRIBUTE subobject) to "follows procedures defined in [RSVP-RO]."

YOUNG>> Yes. Please see my comment below. 

Authors, Young, Giovanni:

   Can you clarify what specific procedures defined in [RSVP-RO] are
   to be applied to the LSP_REQUIRED_ATTRIBUTES Objects?

I think it should be possible to align the draft text fairly easily once the intent is clear, and I'm happy/plan to propose specific text.

YOUNG>> Replacing [RSVP-RO] with [RFC5420] will solve the problem discussed here. 

Thanks,
Lou

_______________________________________________
CCAMP mailing list
CCAMP@ietf.org
https://www.ietf.org/mailman/listinfo/ccamp