Last Call Review of draft-ietf-mpls-self-ping-04
review-ietf-mpls-self-ping-04-opsdir-lc-wijnen-2015-10-12-00

Request Review of draft-ietf-mpls-self-ping
Requested rev. no specific revision (document currently at 06)
Type Last Call Review
Team Ops Directorate (opsdir)
Deadline 2015-10-13
Requested 2015-09-28
Authors Ron Bonica, Ina Minei, Michael Conn, Dante Pacella, Luis Tomotaki
Draft last updated 2015-10-12
Completed reviews Genart Last Call review of -04 by Russ Housley (diff)
Secdir Last Call review of -04 by Leif Johansson (diff)
Opsdir Last Call review of -04 by Bert Wijnen (diff)
Rtgdir Early review of -04 by John Drake (diff)
Assignment Reviewer Bert Wijnen 
State Completed
Review review-ietf-mpls-self-ping-04-opsdir-lc-wijnen-2015-10-12
Reviewed rev. 04 (document currently at 06)
Review result Has Nits
Review completed: 2015-10-12

Review
review-ietf-mpls-self-ping-04-opsdir-lc-wijnen-2015-10-12

Hi,

I did OPS-DIR review fordraft-ietf-mpls-self-ping-05

Summary document is in good shape and ready for publication.

However, when I read (in the abstract and also in the text itself):

   When certain RSVP-TE optimizations are implemented, ingress LSRs can
   receive RSVP RESV messages before forwarding state has been installed
   on all downstream nodes.

And I then see that this document describes an extra protocol (self-ping)
in order to correct that problem, then I really wonder why one would
not change:
   Referring to any LSR, RFC 3209 says, ""The node SHOULD be prepared to
   forward packets carrying the assigned label prior to sending the RESV
   message".  However, RFC 3209 does not strictly require this behavior.

Why not make the SHOULD in RFC3209 a MUST ??
Does the additional self-ping not add a similar delay as that what is
avoided with a SHOULD??

Bert