I am an assigned INT directorate reviewer fordraft-ietf-6man-icmpv6-reflection. These comments were written primarily for the benefit of the Internet Area Directors. Document editors and shepherd(s) should treat these comments just like they would treat comments from any other IETF contributors and resolve them along with any other Last Call comments that have been received. For more details on the INT Directorate, see https://datatracker.ietf.org/group/intdir/about/

I found the document well written and easy to read. I do have some concerns that probably need to be addressed in the document before publication and they are listed below

High Level:

It would be good to have some text in this document that describes how the probing node correlates the received response to the request it originally sent out, verifies that the response is well formed and presents the results to the users. 

Major:

* Section 4

"Middle boxes must not modify the Reflect All extension object."

Unsure if it is desirable or even useful to put in a "must not" level requirement for middleboxes, while middleboxes messing with contents of other ICMPv6 messages is claimed as a reason for this new mechanism (Section 1 comment below).

Minor:

* Section 1

"However, middle boxes modify the invoking packet in ICMPv6 Destination Unreachable messages (e.g. [RFC3022])"

Is this something that has been implemented/observed? If so, this might need a better citation. The very thought of IPv6 middleboxes modifying the *contents* of IPv6 packets like NAT44 ALGs stresses me out

* Section 4

"The length of both the request and reply packets SHOULD NOT exceed the IPv6 minimum MTU (1280 bytes)"

I think this document should simply refer to RFC8200 and not repeat the 1280 number.

* Section 5

"The object payload field in a request message MAY contain all zeros. Alternatively, this field MAY contain arbitrary data."

Unsure what this text is trying to accomplish. Wouldn't the second sentence "this field MAY contain arbitrary data" be sufficient to express the same intent (and cover the first case as well)?