Skip to main content

Early Review of draft-ietf-opsawg-ucl-acl-10
review-ietf-opsawg-ucl-acl-09-intdir-early-pelov-2025-12-13-01

Request Review of draft-ietf-opsawg-ucl-acl
Requested revision No specific revision (document currently at 12)
Type Early Review
Team Internet Area Directorate (intdir)
Deadline 2026-01-03
Requested 2025-12-12
Requested by Joe Clarke
Authors Qiufang Ma , Qin Wu , Mohamed Boucadair , Daniel King
I-D last updated 2026-02-28 (Latest revision 2026-02-03)
Completed reviews Opsdir Early review of -08 by Dhruv Dhody (diff)
Intdir Early review of -10 by Alexander Pelov (diff)
Secdir IETF Last Call review of -11 by Valery Smyslov (diff)
Yangdoctors IETF Last Call review of -12 by Acee Lindem
Opsdir IETF Last Call review of -11 by Dhruv Dhody (diff)
Assignment Reviewer Alexander Pelov
State Completed
Request Early review on draft-ietf-opsawg-ucl-acl by Internet Area Directorate Assigned
Posted at https://mailarchive.ietf.org/arch/msg/int-dir/YztCtma59QP9OOXeH6f5fo6mErw
Reviewed revision 10 (document currently at 12)
Result Ready w/nits
Completed 2025-12-22
review-ietf-opsawg-ucl-acl-09-intdir-early-pelov-2025-12-13-01
I previously reviewed version -09 of this draft and raised concerns regarding
inconsistent terminology and the scope of the group definitions. The authors
have significantly improved the document in version -10. All questions and
issues I identified in my previous review were addressed or answered. Thanks to
the authors for the rapid reaction!

The restructuring of the Terminology section and the clarification of the
relationship between the generic "Endpoint Group" and its specific subtypes
(User, Device, Application) has resolved the architectural ambiguity I
previously noted. Furthermore, the added examples clarify that the defined
RADIUS attribute (User-Access-Group-ID) is intended specifically for
user-centric authentication scenarios, while other group types are provisioned
differently. This distinction makes the attribute naming logical and clarify
the indented use for the different group types.

The choice of 'group-id' being a string has been discussed on github and in the
WG, so I don't see any issue with it.

Minor nits:
In the terminology section, there are leading spaces missing before the bullets
"* device group:" and "* Application group:"