HI Julien,
On Thu, May 16, 2024 at 7:57 PM <[email protected]> wrote:
> Dear authors of draft-ietf-pce-pcep-yang,
>
> I've reviewed the aforementioned document to prepare its publication
> request. The I-D is almost ready to move forward and only has minor
> issues and nits that should be addressed before sending it to the IESG.
>
> Minor issues:
> - The introduction doesn't mention the ietf-pcep-stats module though
> it's defined in the body of the I-D; a brief additional sentence would
> be welcome.
>
Dhruv: Added "Further, this document also includes the PCEP statistics
YANG module
"ietf-pcep-stats" which provides statistics, counters and telemetry data."
> - For SR, the leaf "msd-limit" (page 45) is a boolean that should be
> renamed to be understandable, e.g. into "no-msd-limit" or "ignore-msd".
>
Dhruv: changed to "no-msd-limit".
> - On page 46, in the H-PCE section, there's a "if Stateful GMPLS is
> enabled" left instead of "if H-PCE is enabled.
>
Dhruv: thanks for spotting that
> - Section 7.1 about TLS should be deeply summarized and rather point to
> the referenced document (pointer to be updated).
>
Dhruv: Changed to -- "The PCC acting as the TLS client opens the TLS
connection and the PCE acting as the TLS server listens for incoming
connections as per TLS specifications ([RFC8446] and [RFC5246]). [RFC8253]
specifies the StartTLS procedure in PCEP that initiates the TLS connection
before exchanging PCEP messages thus the identity verification is completed
before the PCEP session is established."
> - On page 51, in the description of the hexadecimal case, I don't think
> the 2 long sentences about the rationale should be included there; if
> the authors consider it necessary, it may be included in the text body
> of the draft.
>
Dhruv: The text/approach is borrowed from RFC 8177. I prefer to keep it.
> - On page 59, similar comment about the sync-timer description, which
> I'd shorten into the following:
> "The value of SyncTimer in seconds is used in the
> case of synchronized path computation request
> using the SVEC object. If after the expiration of
> the SyncTimer all the path computation requests
> have not been received, a protocol error is
> triggered and the PCE must cancel the whole set
> of path computation requests.
> Zero means that the PCEP entity does not use the
> SyncTimer."
>
Dhruv: Thanks. Updated.
> - On page 69, about path key, the name of the leaf "pcc-original" feels
> odd, how about "originator-pcc" instead?
>
>
Dhruv: I changed it to pcc-requester and used "original" in the description
to march the text in RFC 5520.
leaf pcc-requester {
type leafref {
path "/pcep/entity/peers/peer/addr";
}
description
"Reference to PCC peer address that
issued the original request that led
to the creation of the path-key.";
}
> Nits:
> - Page 11: s/system generated entity index/system-generated entity index
> - P.11: s/the local entity is PCE it/the local entity is a PCE, it
> - P.11: s/dead-timer in YANG is called DeadTimer in the protocol
> specification/DeadTimer in the protocol specification is called
> dead-timer in YANG/
> - P.16: s/learn PCE in the network via IGP discovery/learn a PCE address
> in the network via the IGP discovery/
> - P.28-30: There are several bullets points in the descriptions fields
> that would benefit from semicolons at each line end.
>
Dhruv: I used a comma instead for readability
> - P.40: s/maybe relevant/may be relevant/
> - P.44: s/PCE triggered/PCE-triggered/ [twice]
> - P.49: s/instance specific data/instance-specific data/
> - P.105: s/this document also include/this document also includes/
>
>
Dhruv: Ack!
Thanks for your review! New version -24 is posted.
Diff:
https://author-tools.ietf.org/iddiff?url1=draft-ietf-pce-pcep-yang-23&url2=draft-ietf-pce-pcep-yang-24&difftype=--html
Regards,
Dhruv
>
> Best regards,
>
> Julien
>
> _______________________________________________
> Pce mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
>
_______________________________________________
Pce mailing list -- [email protected]
To unsubscribe send an email to [email protected]
