[pim] Yangdoctors last call review of draft-ietf-pim-msdp-yang-12

Reshad Rahman via Datatracker <noreply@ietf.org> Wed, 29 January 2020 03:05 UTC

MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
From: Reshad Rahman via Datatracker <noreply@ietf.org>
To: yang-doctors@ietf.org
Cc: last-call@ietf.org, draft-ietf-pim-msdp-yang.all@ietf.org, pim@ietf.org
Auto-Submitted: auto-generated
Precedence: bulk
Reply-To: Reshad Rahman <rrahman@cisco.com>
Message-ID: <158026714472.4559.17171044234392202915@ietfa.amsl.com>
Date: Tue, 28 Jan 2020 19:05:44 -0800
Archived-At: <https://mailarchive.ietf.org/arch/msg/pim/i8q8FGRsHIiCLa0GJOPZy4SlxLI>
Subject: [pim] Yangdoctors last call review of draft-ietf-pim-msdp-yang-12

Reviewer: Reshad Rahman
Review result: Almost Ready

YANG Doctor review of draft-ietf-pim-msdp-yang-12 by Reshad Rahman

1 module in this draft:
- ietf-msdp@2020-01-24.yang

No YANG validation errors or warnings

Thank you for addressing comments which were provided on rev-01 of the document.

- Page 21, the YANG module has the augmentation of
“/rt:…/rt:control-plane-protocol” but there is a missing when statement. This
means that on a device which supports this YANG model, the “msdp” container
node will appear in all instances of “rt:control-plane-protocol”, even the
non-MSDP ones. If you need an example of how to fix this with when statement,
please take a look at OSPF and BFD YANG models.

- There are no examples. Just a couple of simple XML examples would help a lot.

- There are IMO still too many (14?) features for a fairly straight-forward
YANG model. An explanation for this is provided in section 3.1, but this does
not comply with 4.17 of RFC8407:
   The set of YANG features defined in a module should be considered
   carefully.  Very fine granular features increase interoperability
   complexity and should be avoided.  A likely misuse of the feature
   mechanism is the tagging of individual leafs (e.g., counters) with
   separate features.

- Page 12, the if-feature was taken out of password (to address a comment
regarding duplicate if-feature), but I believe the remaining one should be
moved up from case key-chain to container authentication.

- There is an as-number leaf in the YANG model, but no such thing in RFC3618.
Do we need a reference here?

- Page 24, RPC clear-sa-cache has source-addr using type
ipv4-multi-cast-source-address. But in the operational model (container
sa-cache), source-addr is a union of either * or ipv4-address. Why the
difference? Same question, wrt inconsistency, for leaf group
(ipv4-multicast-group-address in RPC and ipv4-address in operational model).

- Page 22, rp-address is of type ip-address, should that be ipv4-address just
like rpf-peer? Or am I misunderstanding this?

- Many of the descriptions are still very terse, e.g. up-time, expire.

- I’m not an MSDP expert, but I believe adding, where appropriate in the YANG
module, more references to the appropriate sections of RFC3618 or other PIM
RFCs would improve the document. e.g. this could help for RPF-related nodes.

- Section 6: s/one new URIs/one new URI/

[pim] Yangdoctors last call review of draft-ietf-… Reshad Rahman via Datatracker