Re: [Gen-art] Gen-ATR review of draft-ietf-ccamp-ospf-availability-extension-07

["Yemin (Amy)" <[e-mail]>]

Hi all,

After discussing with Jouni, authors, chairs, Lou and our AD, we will update the draft with the following changes to clarify Jouni's concern:

In the Abstract and Introduction, add:
"It is intended that technology-specific documents will reference this document to describe specific uses."

Section 3.2 is updated as the following:
"A node advertising an interface with a Switching Capability which supports variable bandwidth attached SHOULD contain one or more ISCD Availability sub-TLVs in its OSPF TE LSA messages. Each ISCD Availability sub-TLV provides the information about how much bandwidth a link can support for a specified availability. This information MAY be used for path calculation by the node(s).

The ISCD Availability sub-TLV MUST NOT be sent in ISCDs with Switching Capability field values that have not been defined to support the ISCD Availability sub-TLV. Non-supporting nodes would see such as a malformed ISCD/LSA.

Absence of the ISCD Availability sub-TLV in an ISCD containing Switching Capability field values that have been defined to support the ISCD Availability sub-TLV, SHALL be interpreted as representing fixed-bandwidth link with the highest availability value. 

Only one ISCD Availability sub-TLVs for the specific availability level SHOULD be sent. If multiple are present, only the first ISCD Availability sub-TLV for an availability level carried in the same ISCD SHALL be processed."


In the IANA considerations, add:
"Technology-specific documents will reference this document to describe specific use of this Availability sub-TLV."

The draft will be revised by a new version. 

BR,
Amy
-----Original Message-----
From: jouni.nospam [mailto:[e-mail]] 
Sent: Wednesday, October 19, 2016 5:26 AM
To: Daniele Ceccarelli
Cc: Yemin (Amy); [e-mail]; [e-mail]
Subject: Re: Gen-ATR review of draft-ietf-ccamp-ospf-availability-extension-07

Hi Daniele,


> On Oct 17, 2016, at 11:28 PM, Daniele Ceccarelli <[e-mail]> wrote:
> 
> Hi Jouni,
> 
> Thanks a lot for your review and your thoughts. I tend to agree with you, publishing a document referencing a future one doesn't make much sense.
> We had a long discussion inside the WG on how to progress this draft with many alternative options and this one happened to be the less painful one. 
> 
> What I would suggest to do is to ask Amy to publish that document ASAP, reference it and then progress the two drafts as a cluster (i.e. this one needs to wait for the other one to become an RFC).
> Would this work for you? Deborah, Fatai, what's your opinion?


I assume “that document” above refers to a new draft that would amend RFC4203 to handle generic TLVs in scsi field. If this is the case it would be a preferable outcome to me.

I know having a normative reference in draft-ietf-ccamp-ospf-availability-extension to a new work is a bit of pain but I guess there’s no harm as you were already fine publishing an extension without describing how to implement it in the first place.

- Jouni


> 
> BR
> Daniele
> 
>> -----Original Message-----
>> From: jouni.nospam [mailto:[e-mail]]
>> Sent: martedì 18 ottobre 2016 08:23
>> To: Yemin (Amy) <[e-mail]>
>> Cc: [e-mail]; draft-ietf-ccamp-ospf-availability-
>> [e-mail]
>> Subject: Re: Gen-ATR review of 
>> draft-ietf-ccamp-ospf-availability-extension-
>> 07
>> 
>> Hi,
>> 
>>> On Oct 17, 2016, at 8:16 PM, Yemin (Amy) <[e-mail]>
>> wrote:
>>> 
>>> Hi Jouni,
>>> 
>>> You're right that the current draft text doesn't provide any 
>>> information
>> about the discussion.
>>> So how about add the following text at the end of section 3.1:
>>> This document only defines the Availability TLV, how the existing 
>>> Switching
>> Capability makes use of the Availability TLV will be addressed in a 
>> different document. An example is to define a new type code for the 
>> Availability TLV, if the Switching Capability-specific information (SCSI) supports TLV format.
>> 
>> With some rewording I could be OK.. I am not going to be stubborn 
>> trying to block this document. Just add that the “different document” 
>> is “different future document” or something like that.
>> 
>> Anyway, I am still somewhat surprised there is now a document (this 
>> draft under review) defining a new extension to RFC4203 before there 
>> is a document that actually describes how to do that in a proper way, 
>> specifically if this draft does not want to do that. The order is imho wrong.. just saying..
>> 
>> - Jouni
>> 
>> 
>> 
>> 
>>> 
>>> Section 3 of RFC7688 provides clear definition for new SC and SCSI. 
>>> But
>> since the conclusion is to use another different document, other than 
>> this document to explain the technology specific usage(including the 
>> SC/SCSI allocation), it’s preferred not to include the such text in this document.
>>> 
>>> BR,
>>> Amy
>>> -----Original Message-----
>>> From: jouni.nospam [mailto:[e-mail]]
>>> Sent: Monday, October 17, 2016 11:21 PM
>>> To: Yemin (Amy)
>>> Cc: [e-mail];
>>> [e-mail]
>>> Subject: Re: Gen-ATR review of
>>> draft-ietf-ccamp-ospf-availability-extension-07
>>> 
>>> Hi,
>>> 
>>> 
>>>> On Oct 17, 2016, at 1:06 AM, Yemin (Amy) <[e-mail]>
>> wrote:
>>>> 
>>>> Hi Jouni,
>>>> 
>>>> Thanks very much for the comments. I fixed the nits in the draft.
>>>> 
>>>> Regarding the Switching Capability-specific information field, we 
>>>> had the
>> discussion in WG, and here's the summary/conclusion:
>>>> It's decided that this document will just define the availability 
>>>> TLV, and a
>> new draft will define its technology specific usage.
>>> 
>>> Ok, thanks for sharing this. This document had no mention or 
>>> guidance of
>> this kind of decision. I would have assumed some text since the 
>> extension defined in this document is seemingly backwards 
>> incompatible without some glue. As the text is now, is not sufficient.
>>> 
>>> I would appreciate text along lines that can be found in Section 3 
>>> of
>> RFC7688 with an addition of the details you point out below regarding 
>> the allocation of new Switching-capability types. Also, it would not 
>> hurt describing (with the Switching-capability types text) how the 
>> situation where an existing non-TLV Switching Capability-specific 
>> information and the new TLV-based information have to coexist.. or 
>> whether that kind format is not supported.
>>> 
>>> - Jouni
>>> 
>>> 
>>>> For example, for the SCSI which supports TLV(e.g., OTN/WSON), a new
>> type code is needed to make use of availability TLV.
>>>> For the SCSI who doesn’t support TLV(e.g., PSC), a new SC types is
>> needed.
>>>> 
>>>> BR,
>>>> Amy
>>>> 
>>>> -----Original Message-----
>>>> From: jouni.nospam [mailto:[e-mail]]
>>>> Sent: Monday, October 17, 2016 5:50 AM
>>>> To: [e-mail]
>>>> Cc: [e-mail]
>>>> Subject: Gen-ATR review of
>>>> draft-ietf-ccamp-ospf-availability-extension-07
>>>> 
>>>> I am the assigned Gen-ART reviewer for this draft. The General Area
>> Review Team (Gen-ART) reviews all IETF documents being processed by 
>> the IESG for the IETF Chair.  Please treat these comments just like 
>> any other last call comments.
>>>> 
>>>> For more information, please see the FAQ at
>>>> 
>>>> <http://wiki.tools.ietf.org/area/gen/trac/wiki/GenArtfaq>.
>>>> 
>>>> Document: draft-ietf-ccamp-ospf-availability-extension-07
>>>> Reviewer: Jouni Korhonen
>>>> Review Date:        2016-10-16
>>>> IETF LC End Date:   2016-10-24
>>>> IESG Telechat date: 2016-11-03
>>>> 
>>>> Summary:
>>>> 
>>>> Document is ready with nits.
>>>> 
>>>> Major issues:
>>>> 
>>>> None.
>>>> 
>>>> Minor issues:
>>>> 
>>>> It is not clear to me how the ISCD Availability sub-TLV is encoded 
>>>> into RFC4203 Switching Capability-specific information field. This 
>>>> is because RFC4203 lists specific encodings depending on “Switching Cap”
>>>> field and those encoded information fields seem not to be TLVs. I 
>>>> would like to see some text that deals with switching cap, its 
>>>> relation to the TLV described in this document and the coexistence 
>>>> with existing capability specific information fields described in 
>>>> RFC4203. If I did not understand something regarding the encoding 
>>>> that is supposed to be trivial I am happy to told that ;)
>>>> 
>>>> Nits/editorial comments:
>>>> 
>>>> o Line 21: ISCD is not expanded.
>>>> o Line 142: unnecessary extra space in "a < availability”.
>>>> o Line 150: Space needed before the reference "protocol[ETPAI].”
>>>> o Line 142-.. TE is never expanded or part of the list acronyms.
>>>> o Lines 176-178: formatting issue with indentation, line spacing 
>>>> and line endings (not a fullstop but ‘;’).
>>>> o Line 162: TLV is never expanded or  part of the list acronyms.
>>>> 
>>>> __________________________________________
>