Re: [Teas] Review and thoughts on draft-zheng-teas-gmpls-controller-inter-work

Hi Lou, Pavan,
Thanks for your feedback, we were waiting for this as discussed during the presentation in Bangkok.
As said by Haomian, we surely will try to address point 1 and 2 for next update. 

Thanks
Sergio

-----Original Message-----
From: Zhenghaomian (Zhenghaomian, Optical &Microwave Technology Research Dept) <zhenghaomian@huawei.com> 
Sent: Monday, December 10, 2018 2:15 AM
To: Lou Berger <lberger@labn.net>; adrian@olddog.co.uk; teas@ietf.org
Cc: draft-zheng-teas-gmpls-controller-inter-work@ietf.org
Subject: 答复: [Teas] Review and thoughts on draft-zheng-teas-gmpls-controller-inter-work

Hi, Lou and Pavan,

Thank you for the advices, we will address 1) and 2) in the next update. For 3), I agree with you that section 7 needed to be updated consecutively, but probably after WG adoption. 

Best wishes,
Haomian

-----邮件原件-----
发件人: Lou Berger [mailto:lberger@labn.net]
发送时间: 2018年12月8日 21:54
收件人: Zhenghaomian (Zhenghaomian, Optical &Microwave Technology Research Dept) <zhenghaomian@huawei.com>; adrian@olddog.co.uk; teas@ietf.org
抄送: draft-zheng-teas-gmpls-controller-inter-work@ietf.org
主题: Re: [Teas] Review and thoughts on draft-zheng-teas-gmpls-controller-inter-work

Hi,

Since you asked....

I think Pavan and I have no reservations about this work progressing in the WG.  We do have a couple of things that think it would be worth clarifying before adoption poll.

1) it would be good to clarify that ACTN is one (example) method for implementing centralized control.  The document is good in some places, e.g., section 1, but less so in others. It would be good to ensure that the reader, particularly those new to the topic, understand that ACTN is just one way to provide centralized control, e.g., section 2.3 reads as if ACTN is the sole supported method; a PCE based approach is mentioned in the document but this is buried at end of section 5.2.

2) It would best if the examples lined up with what we are covering in the WG.  For example, I don't believe we've had any suggestion or discussion use of OSPF-TE as an SBI as shown in figure 1.  (A nit, it should be NetConf/YANG.) If your intent is to propose this, than this should be clarified and perhaps added to section 7.

3) We see Section 7 as the (future) core of the document and clearly a work in progress.  This is can be addressed pre or post WG adoption.

Thanks,
Pavan and Lou

On 12/6/2018 7:38 AM, Zhenghaomian (Zhenghaomian, Optical &Microwave Technology Research Dept) wrote:
> Hi, Adrian,
>
> Thank you for the comments, we just updated this work to -02 and hopefully all the comments are addressed. We are also looking forward to see the feedback from the chairs, plans and/or advices are highly appreciated, thank you.
>
> FYI, the -02 draft is now available at:
> URL:            https://www.ietf.org/internet-drafts/draft-zheng-teas-gmpls-controller-inter-work-02.txt
> Status:         https://datatracker.ietf.org/doc/draft-zheng-teas-gmpls-controller-inter-work/
> Htmlized:       https://tools.ietf.org/html/draft-zheng-teas-gmpls-controller-inter-work-02
> Htmlized:       https://datatracker.ietf.org/doc/html/draft-zheng-teas-gmpls-controller-inter-work
> Diff:           https://www.ietf.org/rfcdiff?url2=draft-zheng-teas-gmpls-controller-inter-work-02
>
> Best wishes,
> Haomian
>
> -----邮件原件-----
> 发件人: Adrian Farrel [mailto:adrian@olddog.co.uk]
> 发送时间: 2018年12月5日 17:33
> 收件人: teas@ietf.org
> 抄送: draft-zheng-teas-gmpls-controller-inter-work@ietf.org
> 主题: Review and thoughts on
> draft-zheng-teas-gmpls-controller-inter-work
>
> Hi,
>
> I just had cause to read this draft, and that generated a few small comments (below). Otherwise, it seems a fairly sound document.
>
> I wonder what the plans are for this document. IIRC it was moved from CCAMP to TEAS after discussion between the chairs and considering that it had a broader TE scope. Does that mean there is a plan for this to become a TEAS WG document? Chairs?
>
> Thanks,
> Adrian
>
> ===
>
> Right hand margin alignment on the top of the front page seems to be a bit messed up.
>   
> ---
>   
> There seem to be two carriage returns at the bottom of each page.
>   
> ---
>   
> idnits shows a few issues with references
>    == Missing Reference: 'TE-TOPO' is mentioned on line 247, but not 
> defined
>   
>    == Missing Reference: 'I-D.ietf-pce-pce-initiated-lsp' is mentioned on line
>       349, but not defined
>   
>    == Missing Reference: 'PCEP-LS' is mentioned on line 397, but not 
> defined
>   
>    == Unused Reference: 'RFC8281' is defined on line 593, but no explicit
>       reference was found in the text
>   
> ---
>   
> Abstract
> I think, in order to avoid having to expand "LMP" and to make the text a little bit more brief, you might leave out the three examples in parentheses.
>   
> ---
>   
> Abstract
> You aren't allowed citations in the Abstract, so OLD is ACTN 
> controller hierarchy [RFC8453] NEW is the ACTN controller hierarchy 
> described in RFC 8453 END
>   
> ---
>   
> Abstract
> s/advantage/advantages/
>   
> ---
>   
> If you think you need the 2119 language (which you currently don't use and probably don't need), you should move "Conventions used in this document" down to become Section 2 and replace it with the new text...
>   
>     The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
>     "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
>     "OPTIONAL" in this document are to be interpreted as described in BCP
>     14 [RFC2119] [RFC8174] when, and only when, they appear in all
>     capitals, as shown here.
>   
> That requires you to make an addition to the normative references for RFC 8174.
>   
> Of course, better is to entirely remove this text and to remove the reference to 2119 from 11.1.
>   
> ---
>   
> Introduction
> Need to expand PNC, MDSC, and CNC.
> Similarly PCE.
>   
> ---
>   
> 2.1
> s/establish LSP/establish Label Switched Paths (LSPs)/ s/the same 
> network topology/the same view of the network topology/
>   
> ---
>   
> 2.2
> s/In centralized controller system/In centralized controller systems/
>   
> ---
>   
> 2.3
> s/Following figure/Figure 1/
>   
> ---
>   
> 3.
> OLD
>     In addition to setup and
>     maintain control channels,
> NEW
>     In addition to the setup and
>     maintenance of control channels,
> END
>   
> ---
>   
> 5.
> s/Once a controller learn/Once a controller learns/
>   
> ---
>   
> 6.
> s/Signaling mechanism is/Signaling mechanisms are/
>   
> ---
>   
> 6.
> You have....
>     Signaling protocols such as
>     RSVP-TE [RFC3473] and CR-LDP [RFC3472] have been extended to support
>     different interfaces in GMPLS.
>   
> I think it is time to stop mentioning CR-LDP. It is now substantially historic and has fallen so far behind RSVP-TE as to be basically unusable.
>   
> ---
>   
> 7.1
> s/method has/methods have/
>   
> ---
>   
> Obviously sections 8 and 9 need some text.
>   
> ---
>   
> I think you need to sort out the references. Currently you have them all as Normative, and that isn't right. Your guide should be "Which documents *must* you read to understand the basis of this document (normative), and which documents provide additional detail that a dedicated reader would want to read (informative)?"
>
> _______________________________________________
> Teas mailing list
> Teas@ietf.org
> https://www.ietf.org/mailman/listinfo/teas