Network Working Group                                          G. Huston
Request for Comments: 3765                                       Telstra
Category: Informational                                       April 2004


          NOPEER Community for Border Gateway Protocol (BGP)
                         Route Scope Control

Status of this Memo

  This memo provides information for the Internet community.  It does
  not specify an Internet standard of any kind.  Distribution of this
  memo is unlimited.

Copyright Notice

  Copyright (C) The Internet Society (2004).  All Rights Reserved.

Abstract

  This document describes the use of a scope control Border Gateway
  Protocol (BGP) community.  This well-known advisory transitive
  community allows an origin AS to specify the extent to which a
  specific route should be externally propagated.  In particular this
  community, NOPEER, allows an origin AS to specify that a route with
  this attribute need not be advertised across bilateral peer
  connections.

1.  Introduction

  BGP today has a limited number of commonly defined mechanisms that
  allow a route to be propagated across some subset of the routing
  system.  The NOEXPORT community allows a BGP speaker to specify that
  redistribution should extend only to the neighbouring AS.  Providers
  commonly define a number of communities that allow their neighbours
  to specify how advertised routes should be re-advertised.  Current
  operational practice is that such communities are defined on as AS by
  AS basis, and while they allow an AS to influence the re-
  advertisement behaviour of routes passed from a neighbouring AS, they
  do not allow this scope definition ability to be passed in a
  transitive fashion to a remote AS.

  Advertisement scope specification is of most use in specifying the
  boundary conditions of route propagation.  The specification can take
  on a number of forms, including as AS transit hop count, a set of
  target ASs, the presence of a particular route object, or a
  particular characteristic of the inter-AS connection.




Huston                       Informational                      [Page 1]

RFC 3765                         NOPEER                       April 2004


  There are a number of motivations for controlling the scope of
  advertisement of route prefixes, including support of limited transit
  services where advertisements are restricted to certain transit
  providers, and various forms of selective transit in a multi-homed
  environment.

  This memo does not attempt to address all such motivations of scope
  control, and addresses in particular the situation of both multi-
  homing and traffic engineering.  The commonly adopted operational
  technique is that the originating AS advertises an encompassing
  aggregate route to all multi-home neighbours, and also selectively
  advertises a collection of more specific routes.  This implements a
  form of destination-based traffic engineering with some level of fail
  over protection.  The more specific routes typically cease to lever
  any useful traffic engineering outcome beyond a certain radius of
  redistribution, and a means of advising that such routes need not to
  be distributed beyond such a point is of some value in moderating one
  of the factors of continued route table growth.

  Analysis of the BGP routing tables reveals a significant use of the
  technique of advertising more specific prefixes in addition to
  advertising a covering aggregate.  In an effort to ameliorate some of
  the effects of this practice, in terms of overall growth of the BGP
  routing tables in the Internet and the associated burden of global
  propagation of dynamic changes in the reachability of such more
  specific address prefixes, this memo describes the use of a
  transitive BGP route attribute that allows more specific route tables
  entries to be discarded from the BGP tables under appropriate
  conditions.  Specifically, this attribute, NOPEER, allows a remote AS
  not to advertise a route object to a neighbour AS when the two AS's
  are interconnected under the conditions of some form of sender keep
  all arrangement, as distinct from some form of provider / customer
  arrangement.

2.  NOPEER Attribute

  This memo defines the use a new well-known bgp transitive community,
  NOPEER.

  The semantics of this attribute is to allow an AS to interpret the
  presence of this community as an advisory qualification to
  readvertisement of a route prefix, permitting an AS not to
  readvertise the route prefix to all external bilateral peer neighbour
  AS's.  It is consistent with these semantics that an AS may filter
  received prefixes that are received across a peering session that the
  receiver regards as a bilateral peer sessions.





Huston                       Informational                      [Page 2]

RFC 3765                         NOPEER                       April 2004


3.  Motivation

  The size of the BGP routing table has been increasing at an
  accelerating rate since late 1998.  At the time of publication of
  this memo the BGP forwarding table contains over 118,000 entries, and
  the three year growth rate of this table shows a trend rate which can
  be correlated to a compound growth rate of no less than 10% per year
  [2].

  One of the aspects of the current BGP routing table is the widespread
  use of the technique of advertising both an aggregate and a number of
  more specific address prefixes.  For example, the table may contain a
  routing entry for the prefix 10.0.0.0/23 and also contain entries for
  the prefixes 10.0.0.0/24 and 10.0.1.0/24.  In this example the
  specific routes fully cover the aggregate announcement.  Sparse
  coverage of aggregates with more specifics is also observed, where,
  for example, routing entries for 10.0.0.0/8 and 10.0.1.0/24 both
  exist in the routing table.  In total, these more specific route
  entries occupy some 51% of the routing table, so that more than one
  half of the routing table does not add additional address
  reachability information into the routing system, but instead is used
  to impose a finer level of detail on existing reachability
  information.

  There are a number of motivations for having both an aggregate route
  and a number of more specific routes in the routing table, including
  various forms of multi-homed configurations, where there is a
  requirement to specify a different reachability policy for a part of
  the advertised address space.

  One of the observed common requirements in the multi-homed network
  configuration is that of undertaking some form of load balancing of
  incoming traffic across a number of external connections to a number
  of different neighbouring ASs.  If, for example, an AS wishes to use
  a multi-homed configuration for routing-based load balancing and some
  form of mutual fail over between the multiple access connections for
  incoming traffic, then one approach is for the AS to advertise the
  same aggregate address prefix to a number of its upstream transit
  providers, and then advertise a number of more specifics to
  individual upstream providers.  In such a case all of the traffic
  destined to the more specific address prefixes will be received only
  over those connections where the more specific has been advertised.
  If the neighbour BGP peering session of the more specific
  advertisement fails, the more specific will cease to be announced and
  incoming traffic will then be passed to the originating network based
  on the path associated with the advertisement of the encompassing
  aggregate.  In this situation the more specific routes are not
  automatically subsumed by the presence of the aggregate at any remote



Huston                       Informational                      [Page 3]

RFC 3765                         NOPEER                       April 2004


  AS.  Both the aggregate and the associated more specific routes are
  redistributed across the entire external BGP routing domain.  In many
  cases, particularly those associated with desire to undertake traffic
  engineering and service resilience, the more specific routes are
  redistributed well beyond the scope where there is any outcomes in
  terms of traffic differentiation.

  To the extent that remote analysis of BGP tables can observe this
  form of configuration, the number of entries in the BGP forwarding
  table where more specific entries share a common origin AS with their
  immediately enclosing aggregates comprise some 20% of the total
  number of FIB entries.  Using a slightly stricter criteria where the
  AS path of the more specific route matches the immediately enclosing
  aggregate, the number of more specific routes comprises some 14% of
  the number of FIB entries.

  One protocol mechanism that could be useful in this context is to
  allow the originator of an advertisement to state some additional
  qualification on the redistribution of the advertisement, allowing a
  remote AS to suppress further redistribution under some originator-
  specified criteria.

  The redistribution qualification condition can be specified either by
  enumeration or by classification.  Enumeration would encompass the
  use of a well-known transitive extended community to specify a list
  of remote AS's where further redistribution is not advised.  The
  weakness of this approach is that the originating AS would need to
  constantly revise this enumerated AS list to reflect the changes in
  inter-AS topology, as, otherwise, the more specific routes would leak
  beyond the intended redistribution scope.  An approach of
  classification allows an originating AS to specify the conditions
  where further redistribution is not advised without having to refer
  to the particular AS's where a match to such conditions are
  anticipated.

  The approach described here to specifying the redistribution boundary
  condition is one based on the type of bilateral inter-AS peering.
  Where one AS can be considered as a customer, and the other AS can be
  considered as a contracted agent of the customer, or provider, then
  the relationship is one where the provider, as an agent of the
  customer, carries the routes and associated policy associated with
  the routes.  Where neither AS can be considered as a customer of the
  other, then the relationship is one of bilateral peering, and neither
  AS can be considered as an agent of the other in redistributing
  policies associated with routes.  This latter arrangement is commonly
  referred to as a "sender keep all peer" relationship, or "peering".





Huston                       Informational                      [Page 4]

RFC 3765                         NOPEER                       April 2004


  This peer boundary can be regarded as a logical point where the
  redistribution of additional reachability policy imposed by the
  origin AS on a route is no longer an imposed requirement.

  This approach allows an originator of a prefix to attach a commonly
  defined policy to a route prefix, indicate that a route should be
  re-advertised conditionally, based on the characteristics of the
  inter-AS connection.

4.  IANA Considerations

  The IANA has registered NOPEER as a well-known community, as defined
  in [1], as having global significance.

     NOPEER (0xFFFFFF04)

  This is an advisory qualification to readvertisement of a route
  prefix, permitting an AS not to readvertise the route prefix to all
  external bilateral peer neighbour AS's.  It is consistent with these
  semantics that an AS may filter received prefixes that are received
  across a peering session that the receiver regards as a bilateral
  peer sessions

5.  Security Considerations

  BGP is an instance of a relaying protocol, where route information is
  received, processed and forwarded.  BGP contains no specific
  mechanisms to prevent the unauthorized modification of the
  information by a forwarding agent, allowing routing information to be
  modified, deleted or false information to be inserted without the
  knowledge of the originator of the routing information or any of the
  recipients.

  The NOPEER community does not alter this overall situation concerning
  the integrity of BGP as a routing system.

  Use of the NOPEER community has the capability to introduce
  additional attack mechanisms into BGP by allowing the potential for
  man-in-the-middle, session-hijacking, or denial of service attacks
  for an address prefix range being launched by a remote AS.

  Unauthorized addition of this community to a route prefix by a
  transit provider where there is no covering aggregate route prefix
  may cause a denial of service attack based on denial of reachability
  to the prefix.  Even in the case that there is a covering aggregate,
  if the more specific route has a different origin AS than the





Huston                       Informational                      [Page 5]

RFC 3765                         NOPEER                       April 2004


  aggregate, the addition of this community by a transit AS may cause a
  denial of service attack on the origin AS of the more specific
  prefix.

  BGP is already vulnerable to a denial of service attack based on the
  injection of false routing information.  It is possible to use this
  community to limit the redistribution of a false route entry such
  that its visibility can be limited and detection and rectification of
  the problem can be more difficult under the circumstances of limited
  redistribution.

6.  References

6.1.  Normative References

  [1] Chandrasekeran, R., Traina, P. and T. Li, "BGP Communities
      Attribute", RFC 1997, August 1996.

6.2.  Informative References

  [2] Huston, G., "Commentary on Inter-Domain Routing in the Internet",
      RFC 3221, December 2001.

7.  Author's Address

  Geoff Huston
  Telstra

  EMail: [email protected]






















Huston                       Informational                      [Page 6]

RFC 3765                         NOPEER                       April 2004


8.  Full Copyright Statement

  Copyright (C) The Internet Society (2004).  This document is subject
  to the rights, licenses and restrictions contained in BCP 78 and
  except as set forth therein, the authors retain all their rights.

  This document and the information contained herein are provided on an
  "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
  OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET
  ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED,
  INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE
  INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
  WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Intellectual Property

  The IETF takes no position regarding the validity or scope of any
  Intellectual Property Rights or other rights that might be claimed to
  pertain to the implementation or use of the technology described in
  this document or the extent to which any license under such rights
  might or might not be available; nor does it represent that it has
  made any independent effort to identify any such rights.  Information
  on the procedures with respect to rights in RFC documents can be
  found in BCP 78 and BCP 79.

  Copies of IPR disclosures made to the IETF Secretariat and any
  assurances of licenses to be made available, or the result of an
  attempt made to obtain a general license or permission for the use of
  such proprietary rights by implementers or users of this
  specification can be obtained from the IETF on-line IPR repository at
  http://www.ietf.org/ipr.

  The IETF invites any interested party to bring to its attention any
  copyrights, patents or patent applications, or other proprietary
  rights that may cover technology that may be required to implement
  this standard.  Please address the information to the IETF at ietf-
  [email protected].

Acknowledgement

  Funding for the RFC Editor function is currently provided by the
  Internet Society.









Huston                       Informational                      [Page 7]