[spring] Re: Draft Update & Request for Reviews: draft-i etf-spring-dhc-distribute-srv6-locator-dhcp

Hi  Min,
Thanks a lot for your 
Please see my responses inline.

Best Regards,
Weiqiang
 
From: [email protected]
Date: 2025-05-16 14:14
To: [email protected]
CC: spring; buraglio; spring-chairs; dhcwg; dhc-chairs; linchangwang.04414; danvoyerwork; zhanggeng; hanruibo
Subject: [spring] Re: Draft Update & Request for Reviews: draft-ietf-spring-dhc-distribute-srv6-locator-dhcp
Dear authors,

I have a few comments on your draft "Distribute SRv6 Locator by DHCP".
* Section 4.1, the IA_SRV6_LOCATOR option is a new DHCPv6 option introduced by this draft, I suggest to provide an explicit list of DHCP messages in which this new DHCPv6 option can be carried in. A typo needs to be fixed, s/Identify Association/Identity Association, there are three places for this typo.

[Weiqiaing] Thanks - we'll address this in the next revision.

* Section 4.2, the field definition for "IALocator-Options" is as following:
"     -  IALocator-Options: Options associated with this SRv6 locator.
         A variable-length field (28 octets less than the value in the
         Option-Len field)."
[Weiqiaing] The length of IALocator-Options is indeed calculated as ("Option-Len" - 12 - length of SRv6-Locator). The length of SRv6-Locator is determined by the sum of "LB-Len" and "LN-Len". This correction will be made in the next version. 
It defines some options associated with this SRv6 locator can be placed here, I suggest to provide an explicit list of options if possible.

[Weiqiaing] Thanks - we'll add a list of options in the next revision.

Furthermore, it says the length of IALocator-Options is "28 octets less than the value in the Option-Len field", however the length of SRv6-Locator is between 1 octet and 16 octets, it seems to me the length of IALocator-Options is ("Option-Len" - 12 - "length of SRv6-Locator") but not ("Option-Len" - 28).
[Weiqiaing] Good catch - we'll address this in the next revision.

* Section 4.2, there are four 1-octet fields LB-Len, LN-Len, Fun-Len, and Arg-Len defined for the IA SRv6 Locator Option. As I understand it, the value of these fields can't be larger than 128, if that's the case, I suggest to add value range for each of these fields and specify how to handle it if the value is over the range.

[Weiqiaing] OK, we'll update the text in the next revision as your suggest.

* Section 5.1, it says "Upon receiving the Release message, the server removes the lease and frees the locator...", suggest to add normative language MUST into this sentence, that means it should read "Upon receiving the Release message, the server MUST remove the lease and frees the locator...".

[Weiqiaing] Thanks - we'll address this in the next revision.

* Section 5.2, it says "In a message sent by a client to a server, the preferred-lifetime and valid-lifetime fields SHOULD be set to 0", why SHOULD but not MUST? Within the next paragraph it says "The client SHOULD NOT send an IA SRv6 Locator option with 0 in the "LB-Len" and "LN-Len" fields", why SHOULD NOT but not MUST NOT?

[Weiqiang] This refers to: https://www.ietf.org/archive/id/draft-ietf-dhc-rfc8415bis-10.html#IAPREFIX-option.
The srv6 locator option is consistent with the IA prefix option, so there is no need to change it to MUST.

* Section 5.4, it says "After receiving the DHCPv6 Release and Decline messages from the client...", suggest to replace "and" with "or", that means it should read "After receiving the DHCPv6 Release or Decline messages from the client...".

[Weiqiaing] Thanks - we'll address this in the next revision.

Hope these comments can help to improve this draft.

Best Regards,
Xiao Min
Original
From: WeiqiangCheng <[email protected]>
To: spring <[email protected]>;
Cc: buraglio <[email protected]>;spring-chairs <[email protected]>;dhcwg <[email protected]>;dhc-chairs <[email protected]>;linchangwang.04414 <[email protected]>;danvoyerwork <[email protected]>;zhanggeng <[email protected]>;hanruibo <[email protected]>;
Date: 2025年05月07日 12:56
Subject: [spring] Draft Update & Request for Reviews: draft-ietf-spring-dhc-distribute-srv6-locator-dhcp
_______________________________________________
spring mailing list -- [email protected]
To unsubscribe send an email to [email protected]
Dear All,

The co-authors have updated draft-ietf-spring-dhc-distribute-srv6-locator-dhcp version -08. 
Draft link：https://www.ietf.org/archive/id/draft-ietf-spring-dhc-distribute-srv6-locator-dhcp-08.html
This revision addresses all comments received to date, and we believe the document is now ready for Working Group Last Call (WGLC).

We would like to express our sincere appreciation to Nick for agreeing to serve as the document shepherd.

Backgroud of the draft:
- Proposes a mechanism for SRv6 locator assignment to Segment Endpoint Nodes via DHCPv6, applicable within trusted domains.
- Successfully implemented by H3C and Raisecom.
- Validated through field testing by China Mobile, confirming full functional and interoperability compliance with requirements.

Request for Feedback
To facilitate comprehensive alignment with the broader SPRING and DHC WGs, we kindly request feedback on the following aspects:
- Further Technical Review
- Use Case Expansion, such as:
Are there additional deployment scenarios that should be explicitly addressed?
- Any other editorial or technical suggestions to improve clarity and interoperability.

Best regards,
Weiqiang Cheng
On behalf of the co-authors

_______________________________________________
spring mailing list -- [email protected]
To unsubscribe send an email to [email protected]