Last Call Review of draft-ietf-6man-grand-04
review-ietf-6man-grand-04-iotdir-lc-gomez-2021-06-18-00

In my opinion, the document is well written. I found no major issues.

I have reviewed the document from the perspective of constrained nodes or
constrained-node networks.

I understand that the document is intended for devices that do not use RFC
6775/8505. Section 8.2 explicitly discusses why the registration-based approach
in 6775/8505 is not followed in draft-ietf-6man-grand. However, 6775/8505
shares the proactive spirit of the mechanisms defined in draft-ietf-6man-grand.
Regarding  registration-based ND, the document states that "This option
requires some investigation and discussion.". While it makes sense to defer
such discussion beyond the timeline for this document, it would appear that
6775/8505 would allow to achieve goals similar to those allowed by
draft-ietf-6man-grand, while offering additional benefits for constrained-node
networks. Indeed, this area offers opportunities for promising future
discussion/work.

I am not sure about the "implementation complexity" attributed to RFC 6775/8505
in section 8.2, in the sense that such documents have been precisely created
for devices that are typically characterized by significant resource
constraints. Perhaps some clarification or rephrasing might help.

Please find below a number of nits/comments:

- Section 2, bullet 3: "... by creating an INCOMPLETE cache entry...". Perhaps
adding a reference for 'INCOMPLETE cache entry' at the end of the sentence
would be helpful.

- Section 2, bullet 4: s/it would consider/in the worst case, it would consider

- Section 4.1, first sentence: s/Neighbor Advertisement/Neighbor Advertisements

- There are several instances of e.g. "NA" and "Neighbor Advertisement"
throughout the document. It might be good to define the acronym on its first
use and use the acronym thereafter.

- Section 5.1, title: s/Other That/Other Than

- Section 5.3.2, bullet 6: s/and wait up/and waits up

- Section 5.3.2: "within tens of milliseconds". Perhaps this time could be
slightly greater in some cases (e.g. with RTTs in the order of ~100 ms).

- Section 6: shouldn't the last dotted line need to be placed right before
Section 7?

- Section 10: s/layer2/layer-2