1st draft of notes

[[e-mail]]

Hi all:

Here's my first rough draft on the working group notes.
Please let me know is wrong.  I'll be glad to fix it.

Sue Hares

=========== 
Notes


Agenda:

1) Agenda Bashing
2) Status on Current Documents
3) Route Flap Damping (Curtis Villamizar)
4) BGP-4+ practice and experience for multicast
	(Dave Meyer)

5) To be Multihomed: Requirements and Definitions
	Howard Berkowitz

6) MD5 TCP for BGP (Pedro Marques) 

7) BGP Security Analysis  (Sandy Murphy)

8) Use of MD5 for BGP Authentication
		(T. Przygienda)

9) BGP-4+ for IPv6
	draft-ietf-idr-bgp4-ipv6-00.txt
	(Pedro Marques, F. Dupont)

10) Capability Negotiation values 
	draft-marques-bgp-cap-mp-00.txt
	(Pedro Marques, F. Dupont)



More Detailed Notes on the working group

1) Agenda Bashing
2) Status of Current Documents:
	Last Call - Multiprotocol Extensions for BGP-4 
	Sent to RFC Editor - Guidelines for AS


3) BGP Route Damping 
	draft-ietf-idr-route-damp-00.txt

	presentation at: 
		http://engr.ans.net/route-damp/route-damp-ietf.ps
	

	Discussion:	
	Move to Working Group last call for 2 weeks,
	Release as Proposed standard.

	Change "fixed" timer in BGP-4 draft specification from
	"must" to may.


4) BGP-4+ practice and experience for multicast
	(Dave Meyer)

	No document, slides at: http
	Discussion about the needs of multicast:

	a) referred to "multicast" group for discussion of SAFI
		containing MRIB (unicast for Multicast routes)
		and  GRIB (Class D addresses)

	b) Capability Negotiation request for the particular
		support of SAFI in multiprotocol option.

5) To Be Multihomed: Requirements and Definitions
	(H. Berkowitz)	 

	[presentation location needed]    

	draft-berkowitz-multirqmt-00.txt

	Suggestion that the application documents be forwarded to
	the Operations area for review and publication of RFCs.
	
	Other reviews are possible in the following exterior 
	communities: NANOG, IOPS, IEPG.

  
6) MD5 TCP for BGP (Pedro Marques) 
     draft-heffenan-bgp-tcp-md5-00.txt
	will be re-issued as a working group draft
	for progression to Proposed Standard.
	A two week Last call will be issued after
	the draft has been publish on the mailing 
	list.


7) BGP Security Analysis  (Sandy Murphy)

  Sandy Murphy presented a security analysis of
  BGP as presented in RFC 1771.  Discussion ensued
  on presentation.  The note taker's high points were:

     A) for protecting the BGP to bgp authentication the
	   options were:

		IPSEC  
		TCP over MD5 (item 6)
		MD5 for BGP Authentication (item 8)
	 
	  IPSEC/TCP over MD5 provided protection against the
	  RESET and SYN attacks.  MD5 for BGP Authentication
	  does not protect against RESET and SYN attacks.

     B) Any security of BGP from creating ISP to final
	ISP by means of signatures on UPdate packet information
	(ASPATH, nlri grouping) does:
	
	a) also require filtering at each point to ensure the 
		necessary  aspath-nlri is correct

	b) require a substantial investment in authentication
	   administration for "signatures"


   The note taker requests additional feedback on this section
   of the meeting prior to producing the final minutes of the
   meeting.  			

8) Use of MD5 for BGP Authentication
		(T. Przygienda)
	draft-przygienda-bgp-md5-00.txt
	(presentation location required) 

9) BGP-4+ for IPv6
	draft-ietf-idr-bgp4-ipv6-00.txt
	(Pedro Marques, F. Dupont)

    Discussion:
	Discussion on the mailing list has raised Dimitry Haskin's
	concerns about the requirement for Global addresses.
	There is no clear consensus that this is a problem.
	Code implementation require the global addresses. 

	The Document will go to Last Call on the Working Group list 
	prior to being forward on Proposed Standard track. 

10) Capability Negotiation Option for IPv6

	draft-marques-bgp-cap-mp-00.txt

   Discussion:
	Capability option seems reasonable, but must await
	the forwarding of the Capability Options processing
	to proceed forward along the standards track. 
	Once the Capability Options have been forwarded, this
	document will go to the working group for last call.