7/18/95 Wg Meeting notes (IETF)

[[e-mail]]

Agenda for the 7/18/95 Meeting

1) Agenda bashing
2) BGP-4 to full standard
3) DPA proposal (Tony Bates, MCI)
4) Route Server to Experimental
5) Route Damping proposal


The meeting followed the above agenda and was done within
1 hour.

2) BGP-4 to full standard

Summary of discussion:
 
The BGP-4 document has been out for a few weeks with it's final
changes.  Paul Traina should change the BGP-4 documents he's
author/co-author for:

	-  Experience with the BGP-4 protocol, and 
	-  IDRP for version 6

(I'm not sure what documents you were referring to.) 

3) Destination Preference Attribute proposal   (Tony Bates, MCI) 

The following documents are related to the DPA proposal:
 
1)protocol document: draft-ietf-idr-bgp-dpa-01.txt

  This document is intended to move along the Proposed standards track.

2)Current Practice of Implementing Symmetric Routing
   and Load Sharing in Multi-Provider Internet.
 	 
   draft-ietf-idr-symm-multi-prov-01.txt

	This document will become an Informational RFC.

3) Application of the BGP Destination Preference Attribute
    in Implementing Symmetric Routing
     	
   draft-ietf-idr-dpa-application-01.txt

   This document will become Best Current Practices RFC.


Summary of discussion:

Tony Bates gave a presentation about DPA.
(Tony can you post your notes to the mailing group??)

The notes simply reviewed the DPA document. 
 
Two vendors indicated their plans to implement this feature.  One of the
vendors is ISI,  the creator of the  Route Server.


One difficulty with the document is that the code for the 
attribute (10) has been used by another implementor.
The IANA should be handing out the code points.  Bill Manning
will talk to Jon Postel (IANA) on  getting code points
assigned, recorded and posted.  At the bottom of these notes,
the current information on the code points have been included. 

Once two implementations of DPA have been created, this specification
will move on to Proposed Standard. 

4) Route Server Document to Experimental

Document:

Summary of Discussion:
The route server document will move on to experimental.  The
route server document has been implemented by Bay Networks.

However, a different Route Server has been implemented by
cisco systems to solve the same problem of the large IBGP mesh.

Ravi Chandra of cisco and Tony Bates promised to write up the
Cisco implementation for the next IETF.  
 
In the long run, only one solution to the IBGP problem  will
be forwarded as a standard. In the future, the group will review 
the experience of users of these two implementations  to decide which
solution to select. 
	

5) Route Damping protocol

Curtis Villamizar wrote up a Route Damping protocol two years
ago.  The Internet Draft received little review during the last time.
Comments from the audience indicate that this was due to the 
intense focus on getting "the network" to work during the 
last year or so.  Now, the route flapping is causing problems and
the group wants to re-examine this document.

Bill Manning volunteered Ramesh Govindan to talk to Curtis,
and get the Route Damping Document written up by the RA's.
The RA route server has implemented this function.  
  


Additional information:
	 
1) Notes from Bill on the current Code points


I was told that the current definitive list was in the
BGP-4 spec.  Looking, I found no enumerated list.
This is from the cisco headers & some clustering around.

1       origin
2       as path
3       next hop
4       multi exit disc.
5       local preference
6       atomic aggregate
7       aggregator
8       community                               cisco ID
9       originator                              cisco ID
10      cluster list                            cisco ID
11      destination preference                  MCI ID
??      bay route server proposal?              BAY ID