Matt,

Thanks for all your efforts and time in keeping this I-D updated.
Some technical and some editorial comments on the new version 10 follow.

Technical comments:

1. At the top of sec. 5.1, say something like – 
Here we focus only on validation of BGPSEC_Path or Secure_Path attribute. 
And refer to RFC 6811 for the recommended origin validation algorithm. 

2. In sec. 5.1 and 5.2, wherever there is mention of validation of *BGPSEC 
update message*, 
it should be replaced by “validation of BGPSEC_Path (or Secure_Path or 
Signature_Block).

3. In sec. 7 (Security Considerations), the vulnerability to replay attacks is 
currently 
not mentioned? Should it be mentioned?

Editorial comments:

1. Some references to Internet Drafts have old dates (not showing the actual 
updated dates).

2. Perhaps RFC 6811 should be a normative reference.

3. p. 5, 1st para: 

“A BGP speaker SHOULD
   NOT advertise the capability of BGPSEC support for a particular AFI
   unless it has also advertised the multiprotocol extension capability
   for the same AFI combination [3].”

s/ a particular AFI/ a particular set of AFIs/

4. p. 7, para 1: s/in the same way/ in a similar way/

5. The following sentence is repeated on pages 6 and 7:

“A BGPSEC update message
   containing the BGPSEC_Path attribute MUST NOT contain the AS_PATH attribute.”

6. p. 7, para 2: s/for one AS number/for each AS number/

7. p. 7, para 2: s/secure path segment/ Secure_Path segment/

8. p. 8, para 4: 

“(The pCount field is also
   useful in managing AS Number migrations, see [18] for details.)”
Could be replaced with:
(The pCount field is also
   useful in managing route servers (see Section 4.2) and AS Number migrations, 
see [18] for details.)

9. p. 10, para 2:

“When originating a new route
   advertisement and sending it to an internal peer, the BGPSEC speaker
   creates a new BGPSEC_Path attribute with zero Secure_Path segments
   and zero Signature Segments.”

This sentence is a bit verbose. There is no BGPSEC_Path attribute if it has 
zero 
Secure_Path segments  and zero Signature Segments.

So why not reword it simply as:
“When originating a new route
   advertisement and sending it to an internal peer, the BGPSEC speaker
   sends only the NLRI (to an iBGP peer).”

10. The last paragraph on p.11 and the top two paragraphs on p.12 can be moved 
to the preamble of 
Section 4 (to a spot just before start of Sec. 4.1). 
This is because these three paragraphs are generally applicable to both 
Sections 4.1 and 4.2.

11. p.13, 2nd last para: s/ to the RPKI router corresponding to the
   BGPSEC speaker / to the RPKI router *certificate* corresponding to the
   BGPSEC speaker/

12. p.16, para 2: s/ in the Subject Key Identifier extension of
   the RPKI router corresponding to / in the Subject Key Identifier extension of
   the RPKI router *certificate* corresponding to/

13. p.17, last para: s/ the most recently added Secure_Path *segments*/ the 
most recently 
added Secure_Path *segment*/  (make “segments” singular)

14. p 18, para 3: s/ for a signature
   produced by a BGPSEC speaker outside of a confederation/ for a signature
   produced by a *peer* BGPSEC speaker outside of a confederation/

15. In Sections 5.1 and 5.2, search “update message” and see if it should be 
replaced by 
“BGPSEC_Path attribute” or Secure_Path” or “Signature_Block” 
as the case may be. (In the context of validation)

16. p.28, 3 rd last para: s/ A BGPSEC speaker/ a BGPSEC speaker/  (should be 
lower case A)

Sriram
_______________________________________________
sidr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/sidr

Reply via email to