Last Call Review of draft-ietf-rtgwg-uloop-delay-06
review-ietf-rtgwg-uloop-delay-06-secdir-lc-shore-2017-10-04-00

Request Review of draft-ietf-rtgwg-uloop-delay
Requested rev. no specific revision (document currently at 09)
Type Last Call Review
Team Security Area Directorate (secdir)
Deadline 2017-10-04
Requested 2017-09-20
Authors Stephane Litkowski, Bruno Decraene, Clarence Filsfils, Pierre Francois
Draft last updated 2017-10-04
Completed reviews Rtgdir Early review of -01 by Matthew Bocci (diff)
Secdir Last Call review of -06 by Melinda Shore (diff)
Genart Last Call review of -06 by Roni Even (diff)
Opsdir Last Call review of -08 by Linda Dunbar (diff)
Assignment Reviewer Melinda Shore 
State Completed
Review review-ietf-rtgwg-uloop-delay-06-secdir-lc-shore-2017-10-04
Reviewed rev. 06 (document currently at 09)
Review result Has Nits
Review completed: 2017-10-04

Review
review-ietf-rtgwg-uloop-delay-06-secdir-lc-shore-2017-10-04

This document describes a mechanism to mitigate against failures stemming from the formation of "microloops" during a re-routing convergence, as described in RFC 5715.  Modulo some mechanical problems with language usage (i.e. grammatical errors) and some missing definitions, the document clearly describes the problem it is addressing and the proposed solution.

The security considerations section is very clear about why the authors believe no new attacks are introduced by this mechanism, and it is credible

Sections 4 and 5 represent the core of the document and are very clear - a very nice piece of specification.

It would be helpful to have a terminology section, or to expand some of the acronyms in-line (LFA, for example).

For some reason the grammatical errors are clustered towards the front of the document but there are many scattered throughout:

Section 1, first paragraph singular/plural mismatch: "Based on network analysis, local failure make up a significant portion of the micro-forwarding loops"

Section 1, second paragraph unidiomatic use of "the topology"

Section 2, first paragraph unidiomatic use of "high damages"

Section 2.1, first paragraph needs an article on "IGP shortcut"

Same paragraph, doesn't need an article on "the router C"

Same paragraph, "nexthop" should be two words

Item 1 in 2.1, needs an article before "preprogrammed FRR path", also run-on sentence needs to be split or a conjunction inserted

Item 3 in 2.1, "no more" should be "no longer", and "encapsulate anymore" should be "does not continue to encapsulate"

Section 2.1, last paragraph: "The protection enabled by fast-reroute is working perfectly, but ensures a protection, by definition, only until the PLR has converged." is somewhat unclear

Section 3, third paragraph: first comma is unnecessary.  Also, "local only" should be "local-only" 

Section 8.2, first paragraph: "associating timing" should be "associated timing".

Also in section 8.2, the message chart header is separated from the actual contents by a page break, and that should be remedied

Section 8.3, first paragraph: "that happens" should be "that happen".  Also, "without further delaying route insertion" would be more idiomatic than "without delaying route insertion anymore"

Section 9.1, throughout: "nexthop" should be "next hop"

Section 9.1, first bullet item: "only have one" should be "only has one" (or "has only one")

Section 10: "a good behavior" should be "good behavior"