Skip to main content

Early Review of draft-ietf-bfd-secure-sequence-numbers-18
review-ietf-bfd-secure-sequence-numbers-18-opsdir-early-qu-2025-04-06-00

Request Review of draft-ietf-bfd-secure-sequence-numbers-18
Requested revision 18 (document currently at 20)
Type Early Review
Team Ops Directorate (opsdir)
Deadline 2025-04-04
Requested 2025-03-06
Requested by Reshad Rahman
Authors Alan DeKok , Mahesh Jethanandani , Sonal Agarwal , Ashesh Mishra , Ankur Saxena
I-D last updated 2025-05-15 (Latest revision 2025-05-05)
Completed reviews Secdir Early review of -18 by Rich Salz (diff)
Rtgdir IETF Last Call review of -18 by Ben Niven-Jenkins (diff)
Opsdir Early review of -18 by Yingzhen Qu (diff)
Assignment Reviewer Yingzhen Qu
State Completed
Request Early review on draft-ietf-bfd-secure-sequence-numbers by Ops Directorate Assigned
Posted at https://mailarchive.ietf.org/arch/msg/ops-dir/vb21FUcQ9kuPxR8pBW-wc72axS8
Reviewed revision 18 (document currently at 20)
Result Has issues
Completed 2025-04-06
review-ietf-bfd-secure-sequence-numbers-18-opsdir-early-qu-2025-04-06-00
I'm assigned to do an early OPS DIR review of this document.

Thanks to the authors for working on this document. However I think
the following major issues should be addressed before progressing the
document.

Major:

My understanding is that this document introduces a new authentication 
type for BFD, and it's called Meticulous Keyed ISAAC. However I don't
see where this term is defined.
The following text is from the Introduction:
102	   This document proposes the use of an Authentication method which
103	   provides meticulous keying, but which has less impact on resource
104	   constrained systems.  

"Meticulous Keyed" is a name given by the authors. Please explicitly
define it in the document, and add reference to RFC5880.

760	13.  IANA Considerations
For the "IANA Considerations" in this document, it says to reference
I-D.ietf-bfd-optimizing-authentication, however the referenced  document
doesn't specify "TBD1 (Meticulous Keyed ISAAC)".

704	   the isaac() function.c.  Both pages should be maintained at all
705	   times.
719	   page, it can simply swap the "current" and "next" pages (updating the
720	   BFD variables as appropriate), and then notify an asynchronous system
721	   to calculate the "next" page.  Such asynchronous calculations are
722	   preferable to calculating the next page in the BFD fast path.
Question: About page flipping, the draft says it's asynchronous calculation.
After swapping to the next page, an asynchronous system is to calculate
the next page, which means the next page won't be available until the
calculation is done. Is this considered maintained at all times?

nits:
474	   The Seed field is used to is used to identify and secure different
please remove one "is used to"

Thanks,
Yingzhen