[Docs] [txt|pdf] [Tracker] [WG] [Email] [Diff1] [Diff2] [Nits]

Versions: 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 RFC 5291

Network Working Group                                          Enke Chen
Internet Draft                                    Redback Networks, Inc.
Expiration Date: February 2003                             Yakov Rekhter
                                                        Juniper Networks

            Cooperative Route Filtering Capability for BGP-4

                   draft-ietf-idr-route-filter-09.txt


1. Status of this Memo

   This document is an Internet-Draft and is in full conformance with
   all provisions of Section 10 of RFC2026 except that the right to
   produce derivative works is not granted.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF), its areas, and its working groups.  Note that
   other groups may also distribute working documents as Internet-
   Drafts.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as ``work in progress.''

   The list of current Internet-Drafts can be accessed at
   http://www.ietf.org/ietf/1id-abstracts.txt

   The list of Internet-Draft Shadow Directories can be accessed at
   http://www.ietf.org/shadow.html.


2. Specification of Requirements

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in RFC2119 [RFC2119].













Chen & Rekhter                                                  

[Page 1]
Internet Draft     draft-ietf-idr-route-filter-09.txt        August 2003


3. Abstract

   This document defines a BGP-based mechanism that allows a BGP speaker
   to send to its BGP peer a set of route filters that the peer would
   use to constrain/filter its outbound routing updates to the speaker.


4. Introduction

   Currently it is not uncommon for a BGP speaker to receive, and then
   filter out some unwanted routes from its peers based on its local
   routing policy. Since the generation and transmission of routing
   updates by the sender, as well as the processing of routing updates
   by the receiver consume resources, it may be beneficial if the
   generation of such unwanted routing updates can be avoided in the
   first place.

   This document defines a BGP-based mechanism that allows a BGP speaker
   to send to its BGP peer a set of Outbound Route Filters (ORFs).  The
   peer would then apply these filters, in addition to its locally
   configured outbound filters (if any), to constrain/filter its
   outbound routing updates to the speaker.


5. Outbound Route Filter (ORF)

   Conceptually an ORF entry is a tuple of the form <AFI/SAFI, ORF-Type,
   Action, Match, ORF-value>; an ORF consists of one or more ORF entries
   that have a common AFI/SAFI and ORF-Type. An ORF is identified by
   <AFI/SAFI, ORF-Type>.

   The "AFI/SAFI" component provides a coarse granularity control by
   limiting the ORF to only the routes whose NLRI matches the "AFI/SAFI"
   component of the ORF.

   The "ORF-Type" component determines the content of the ORF-value.

   The "Action" component controls handling of the ORF Request by the
   remote peer.  Action can be one of ADD, REMOVE, REMOVE-ALL. ADD adds
   an ORF entry to the ORF on the remote peer; REMOVE deletes a
   previously installed ORF entry on the remote peer; REMOVE-ALL deletes
   the previously installed entries in the specified ORF on the remote
   peer.

   The "Match" component is used if support matching granularity on a
   per ORF entry basis is needed, in which case the "Match" component
   can be one of PERMIT or DENY. The semantics of PERMIT is to ask the
   peer to pass updates for the set of routes that match the ORF entry.



Chen & Rekhter                                                  

[Page 2]
Internet Draft     draft-ietf-idr-route-filter-09.txt        August 2003


   The semantics of DENY is to ask the peer not to pass updates for the
   set of routes that match the ORF entry.


5.1. Communities ORF-Type

   The Community ORF-Type allows to express ORFs in terms of BGP
   Communities [BGP-COMMUNITIES]. That is, the Communities ORF-Type
   provides Communities-based route filtering.

   Conceptually the ORF-value of the Communities ORF-Type consists of a
   single Community.

   The sender SHOULD set the value of the Match field to PERMIT; the
   receiver SHOULD ignore the value of the Match field.

   The remote peer should consider only those routes whose Communities
   attribute has at least one Community in common with the Communities
   list specified in the ORF.


5.2. Extended Communities ORF-Type

   The Extended Community ORF-Type allows to express ORFs in terms of
   BGP Extended Communities [BGP-EXT-COMMUNITIES]. That is, the Extended
   Communities ORF-Type provides Extended Communities-based route
   filtering.

   Conceptually the ORF-value of the Extended Communities ORF-Type
   consists of a single Extended Community.

   The sender SHOULD set the value of the Match field to PERMIT; the
   receiver SHOULD ignore the value of the Match field.

   The remote peer should consider only those routes whose Extended
   Communities attribute has at least one Extended Community in common
   with the Extended Communities list specified in the ORF.














Chen & Rekhter                                                  

[Page 3]
Internet Draft     draft-ietf-idr-route-filter-09.txt        August 2003


6. Carrying ORF entries in BGP

   ORF entries are carried in the BGP ROUTE-REFRESH message [BGP-RR].

   A BGP speaker can distinguish an incoming ROUTE-REFRESH message that
   carries one or more ORF entries from an incoming plain ROUTE-REFRESH
   message by using the Message Length field in the BGP message header.

   A single ROUTE-REFRESH message could carry multiple ORF entries, as
   long as all these entries share the same AFI/SAFI.

   From the encoding point of view each ORF entry consists of a common
   part and type-specific part.

   The common part consists of <AFI/SAFI, ORF-Type, Action, Match>, and
   is encoded as follows:

      The AFI/SAFI component of an ORF entry is encoded in the AFI/SAFI
      field of the ROUTE-REFRESH message.

      Following the AFI/SAFI component is the one-octet When-to-refresh
      field.  The value of this field can be one of IMMEDIATE (0x01) or
      DEFER (0x02). The semantics of IMMEDIATE and DEFER are discussed
      in the "Operation" section of this document.

      Following the When-to-refresh field is a collection of one or more
      ORFs, grouped by ORF-Type.

      The ORF-Type component is encoded as a one-octet field.

      The Length of ORFs component is a two-octets field that contains
      the length (in octets) of the ORF entries that follows.


            +--------------------------------------------------+
            | Address Family Identifier (2 octets)             |
            +--------------------------------------------------+
            | Reserved (1 octet)                               |
            +--------------------------------------------------+
            | Subsequent Address Family Identifier (1 octet)   |
            +--------------------------------------------------+
            | When-to-refresh (1 octet)                        |
            +--------------------------------------------------+
            | ORF Type (1 octet)                               |
            +--------------------------------------------------+
            | Length of ORFs (2 octets)                        |
            +--------------------------------------------------+
            | First ORF entry (variable)                       |



Chen & Rekhter                                                  

[Page 4]
Internet Draft     draft-ietf-idr-route-filter-09.txt        August 2003


            +--------------------------------------------------+
            | Second ORF entry (variable)                      |
            +--------------------------------------------------+
            .........
            +--------------------------------------------------+
            | N-th ORF entry (variable)                        |
            +--------------------------------------------------+
            | ORF Type (1 octet)                               |
            +--------------------------------------------------+
            | Length of ORFs (2 octets)                        |
            +--------------------------------------------------+
            | First ORF entry (variable)                       |
            +--------------------------------------------------+
            | Second ORF entry (variable)                      |
            +--------------------------------------------------+
            .........
            +--------------------------------------------------+
            | N-th ORF entry (variable)                        |
            +--------------------------------------------------+
            .........

            Fig 1. Carrying ORF entries in the ROUTE-REFRESH message

   The rest of the components in the common part are encoded in first
   octet of each ORF-entry as follows (from the most significant to the
   least significant bit):

      Action is a two-bit field. The value of this field is 0 for ADD, 1
      for REMOVE, and 2 for REMOVE-ALL.

      Match is a one-bit field. The value of this field is 0 for PERMIT
      and 1 for DENY. This field is significant only when the value of
      the Action field is either ADD or REMOVE.

      Reserved is a 5-bit field. It is set to 0 on transmit and ignored
      on receive.


         +---------------------------------+
         |   Action (2 bit)                |
         +---------------------------------+
         |   Match (1 bit)                 |
         +---------------------------------+
         |   Reserved (5 bits)             |
         +---------------------------------+
         |   Type specific part (variable) |
         +---------------------------------+




Chen & Rekhter                                                  

[Page 5]
Internet Draft     draft-ietf-idr-route-filter-09.txt        August 2003


         Fig 2. ORF entry encoding


      When the Action component of an ORF entry specifies REMOVE-ALL,
      the entry consists of only the common part.


6.1. Type specific encoding (Communities ORF-Type)

   The value of the ORF-Type for the Communities ORF-Type is 2.

   The type-specific part of Communities ORF-Type consists of single
   Community encoded as a four-octets field.


6.2. Type specific encoding (Extended Communities ORF-Type)

   The value of the ORF-Type for the Extended Communities ORF-Type is 3.

   The type-specific part of Extended Communities ORF-Type consists of a
   single Extended Community encoded as an eight-octets field.


7. Cooperative Route Filtering Capability

   A BGP speaker that is willing to receive ORF entries from its peer,
   or a BGP speaker that would like to send ORF entries to its peer
   advertises this to the peer by using the Cooperative Route Filtering
   Capability, as described below.

   The Cooperative Route Filtering Capability is a new BGP capability
   [BGP-CAP] defined as follows:

      Capability code: 3

      Capability length: variable

      Capability value: one or more of the following entries:


            +--------------------------------------------------+
            | Address Family Identifier (2 octets)             |
            +--------------------------------------------------+
            | Reserved (1 octet)                               |
            +--------------------------------------------------+
            | Subsequent Address Family Identifier (1 octet)   |
            +--------------------------------------------------+
            | Number of ORFs (1 octet)                         |



Chen & Rekhter                                                  

[Page 6]
Internet Draft     draft-ietf-idr-route-filter-09.txt        August 2003


            +--------------------------------------------------+
            | ORF Type (1 octet)                               |
            +--------------------------------------------------+
            | Send/Receive (1 octet)                           |
            +--------------------------------------------------+
            | ...                                              |
            +--------------------------------------------------+
            | ORF Type (1 octet)                               |
            +--------------------------------------------------+
            | Send/Receive (1 octet)                           |
            +--------------------------------------------------+

            Fig 4. Capability encoding


   The use and meaning of these fields are as follows:

      Address Family Identifier (AFI):

         This field carries the identity of the Network Layer protocol
         associated with the Network Address that follows. Presently
         defined values for this field are specified in RFC1700 (see the
         Address Family Numbers section).

      Subsequent Address Family Identifier (SAFI):

         This field provides additional information about the type of
         the Network Layer Reachability Information carried in the
         attribute.

      Number of ORF Types:

         This field contains the number of Filter Types to be listed in
         the following fields.

      ORF Type:

         This field contains the value of an ORF Type.

      Send/Receive:

         This field indicates whether the sender is (a) willing to
         receive ORF entries from its peer (value 1), (b) would like to
         send ORF entries to its peer (value 2), or (c) both (value 3)
         for the ORF Type that follows.






Chen & Rekhter                                                  

[Page 7]
Internet Draft     draft-ietf-idr-route-filter-09.txt        August 2003


8. Operation

   A BGP speaker that is willing to receive ORF entries from its peer,
   or would like to send ORF entries to its peer SHOULD advertise the
   Cooperative Route Filtering Capability to the peer using BGP
   Capabilities advertisement [BGP-CAP].

   A BGP speaker that implements the Cooperative Route Filtering
   Capability must support BGP ROUTE-REFRESH message, as defined in
   [BGP-RR]. A BGP speaker that advertises the Cooperative Route
   Filtering Capability to a peer using BGP Capabilities advertisement
   [BGP-CAP] doesn't have to advertise the BGP Route Refresh capability
   to that peer.

   Consider a BGP speaker that advertises the Cooperative Route
   Filtering Capability indicating its willingness to receive a
   particular set of <AFI, SAFI, ORF-Type> from its peer, and that
   receives the Cooperative Route Filtering Capability indicating the
   desire of the peer to send a particular set <AFI, SAFI, ORF-Type> to
   the speaker. If for a given <AFI, SAFI> the intersection between
   these two sets are not-empty, the speaker SHOULD NOT advertise to the
   peer any routes with that <AFI, SAFI> prior to receiving from the
   peer any ROUTE-REFRESH message carrying that <AFI, SAFI>, where the
   message could be either without any ORF entries, or with one or more
   ORF entry and When-to-refresh field set to IMMEDIATE. If, on the
   other hand, for a given <AFI, SAFI> the intersection between these
   two sets is empty, the speaker SHOULD follow normal BGP procedures.

   A BGP speaker may send a ROUTE-REFRESH message with one or more ORF
   entries to its peer only if the peer advertises to the speaker the
   Cooperative Route Filtering Capability indicating its willingness to
   receive ORF entries from the speaker, and the speaker advertises to
   the peer the Cooperative Route Filtering Capability indicating its
   desire to send ORF entries to the peer.  The message may contain only
   ORF entries of <AFI, SAFI, ORF-type> that the peer is willing to
   receive, as advertised to the speaker in the Cooperative Route
   Filtering Capability.

   When a BGP speaker receives a ROUTE-REFRESH message with one or more
   ORF entries from its peer, then the speaker performs the following
   actions. If the <AFI, SAFI, ORF-type> carried by the message doesn't
   match <AFI, SAFI, ORF-type> that the speaker is willing to receive
   from the peer (as advertised to the peer in the Cooperative Route
   Filtering Capability), the specified ORF is ignored.  Otherwise, the
   speaker modifies the specified ORF, as specified in the ORF entries
   carried by the message. If any of the fields within an ORF entry
   contain an unrecognized value, the whole specified ORF is removed.




Chen & Rekhter                                                  

[Page 8]
Internet Draft     draft-ietf-idr-route-filter-09.txt        August 2003


   If the Action component of an ORF entry is REMOVE, but the ORF
   doesn't contain the specified entry, the entry is ignored.

   ORF entries with either REMOVE or REMOVE-ALL can not remove locally
   configured outbound route filters.

   If the When-to-Refresh indicates IMMEDIATE, then after processing all
   the ORF entries carried in the message the speaker re-advertises to
   the peer routes from the Adj-RIB-Out associated with the peer that
   have the same AFI/SAFI as what is carried in the message, and taking
   into account all the ORF entries received from the peer.  However,
   the routes that have not be affected by the ORF entries carried in
   the message SHOULT NOT be re-advertised to the peer.

   If the When-to-Refresh indicates DEFER, then after processing all the
   ORF entries carried in the message the speaker defers re-
   advertisement to the peer routes from the Adj-RIB-Out associated with
   the peer that have the same AFI/SAFI as what is carried in the
   message, and taking into account all the ORF entries received from
   the peer until the speaker receives a subsequent ROUTE-REFRESH
   message for the same AFI/SAFI either without any ORF entries, or with
   one or more ORF entries and When-to-refresh set to IMMEDIATE.

   If the speaker receives from the peer a ROUTE-REFRESH message without
   any ORF entries, then the speaker sends to the peer all routes from
   the Adj-RIB-Out associated with the peer whose AFI/SAFI is the same
   as what is carried in the message and taking into account the ORF
   received from the peer.

   The set of ORF entries that the speaker sends to the peer expresses
   the speaker's local preference, that the peer may or may not decide
   to honor.

   During a single BGP session the speaker may pass multiple ORF entries
   to the peer.

   The lifetime of an ORF is the duration of the BGP session during
   which the ORF is exchanged.

   An ORF is removed when the last ORF entry is remove (either via
   REMOVE-ALL, or via a sequence of REMOVE).

   If a particular route maintained by a BGP speaker doesn't match any
   of the ORF entries of any of the (non-empty) ORFs associated with a
   particular peer, then this route SHOULD NOT be advertised to the
   peer.

   If a BGP speaker maintains multiple ORFs of different ORF-Types for a



Chen & Rekhter                                                  

[Page 9]
Internet Draft     draft-ietf-idr-route-filter-09.txt        August 2003


   particular peer, then the decision by the speaker to advertise a
   route to the peer is determined by passing the route through each
   such ORF, and and-ing the results (and-ing of PERMIT and DENY results
   in DENY).


9. IANA Considerations

   As specified in this document, an ORF enty contains the ORF-Type
   field.  ORF-Type value 0 is reserved. ORF-Type values 1 through 63
   are to be assigned by IANA using the "IETF Consensus" policy defined
   in RFC2434.  ORF-Type values 64 through 127 are to be assigned by
   IANA, using the "First  Come First Served" policy defined in RFC2434.
   ORF-Type values 128 through 255 are vendor-specific, and values in
   this range are not to be assigned by IANA.


10. Security Considerations

   This extension to BGP does not change the underlying security issues.


11. Acknowledgements

   Some of the material in the document is "borrowed" from a proposal
   for selective updates by Yakov Rekhter, Kannan Varadhan, and Curtis
   Villamizar.


12. Normative References

   [BGP-4] Rekhter, Y., and T. Li, "A Border Gateway Protocol 4
   (BGP-4)", RFC 1771, March 1995.

   [BGP-MP] Bates, T., Chandra, R., Katz, D., and Rekhter, Y.,
   "Multiprotocol Extensions for BGP-4", RFC 2858, June 2000

   [BGP-CAP] Chandra, R., Scudder, J., "Capabilities Advertisement with
   BGP-4", RFC2842, May 2000

   [BGP-COMMUNITIES] Chandra, R., Traina, P., and Li, T., "BGP
   Communities Attribute", RFC1997, August 1996.

   [BGP-EXT-COMMUNITIES] Ramachandra, S., Tappan, D., "BGP Extended
   Communities Attribute", draft-ramachandra-bgp-ext-communities-02.txt

   [BGP-RR] Chen, E., "Route Refresh Capability for BGP-4", RFC2918,
   September 2000



Chen & Rekhter                                                 

[Page 10]
Internet Draft     draft-ietf-idr-route-filter-09.txt        August 2003


   [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
   Requirement Levels", BCP 14, RFC 2119, March 1997.



13. Author Information


   Enke Chen
   Redback Networks, Inc.
   350 Holger Way
   San Jose, CA 95134
   e-mail: enke@redback.com

   Yakov Rekhter
   Juniper Networks
   e-mail: yakov@juniper.net


































Chen & Rekhter                                                 

[Page 11]


Html markup produced by rfcmarkup 1.107, available from http://tools.ietf.org/tools/rfcmarkup/