Re: [Teas] WG Last Call: draft-ietf-teas-yang-path-computation-18

[Dhruv Dhody <[e-mail]>]

Hi WG,

Please find my review comments -

# Review of draft-ietf-teas-yang-path-computation-18

## Minor

### Abstract

* Abstract only mentions "intra-domain paths", whereas the documents do
talk about inter-layer and inter-domain path computation as well.

### Section 1.3

* Remove inet from the table

### Section 2

* You state "Use cases d) and e)..." without stating what d) and e) is.
Perhaps use section numbers instead?

### Section 2.1

* Figure 1 and Figure 2 are completely different network topologies, but
the text gives an impression that they are related. I suggest making it
clear that Figure 2 shows a different network topology.

### Section 2.4

* Please update "[RFC5441] has defined the Virtual Source Path Tree (VSPT)
TLV within PCE Reply Object..." to "[RFC5441] has defined the Virtual
Source Path Tree (VSPT) flag within RP (Request Parameters) object..."

### Section 5.3.1

* Could you explain the rationale for the fact that the requested path can
only be a primary path for non-transit cases?

### Section 5.3.2

* The phrase "empty ERO" makes sense for PCEP, but not for YANG. Please
rephrase.

### Section 6.2 (YANG Model)

* WG Web: should point to datatracker instead of tools page
* Authors/Editors mismatch between the front page and YANG
* Copyright is incorrect, it should be Revised BSD.
https://trustee.ietf.org/documents/trust-legal-provisions/tlp-5/
* Reference for svec-metric-cumul-hop should be RFC5541
* Is the use of the term "configuration" inside grouping requested-state
correct? After all, this is part of the RPC input, can that be called
configuration?
* For objective-function-type in grouping synchronization-optimization, the
use of "default "te-types:of-minimize-cost-path" is a problem, because this OF
is not applicable to SVEC. I don't think you need a default here!
* I am not able to make sense of this when statement. Is this when the
statement suppose to tell it is transit? How?
```
          choice path-role {
            when 'not (./source) and not (./destination) and
                  not (./src-tunnel-tp-id) and
                  not (./dst-tunnel-tp-id)' {
```
* This description is incorrect in the context of the container, it is the
description of the leaf primary-path-name that follows -
```
            container secondary-path {
              presence
                "Indicates that the requested path is a secondary
                path.";
              description
                "The name of the primary path which the requested
                primary reverse path belongs to.";
```


## Appendix A

* The IP address in the example should be from the range reserved for
documentation. See RFC5737. I see this issue exists in
draft-ietf-teas-yang-te-29 as well.

## Nits

* s/PCE request message/PCReq message/
* s/PCE Response message/PCRep message/
* Section 3.2.1, Figure 8 has cost 65, the text states the cose as 60
* s/srlgs/SRLGs/

Thanks,
Dhruv

On Tue, May 24, 2022 at 11:15 PM Vishnu Pavan Beeram <[e-mail]>
wrote:

> All,
>
> This starts working group last call on
> https://datatracker.ietf.org/doc/draft-ietf-teas-yang-path-computation/
>
> Given the size of the document, this will be an extended
> LC (3 weeks). The working group last call ends on June 14th.
> Please send your comments to the working group mailing list.
>
> Positive comments, e.g., "I've reviewed this document
> and believe it is ready for publication", are welcome!
> This is useful and important, even from authors.
>
> Thank you,
> Pavan and Lou
> _______________________________________________
> Teas mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/teas
>