mLDP in a VRF: Multipoint Label Distribution Protocol In-Band Signaling in a Virtual Routing and Forwarding (VRF) Table Context

Internet Engineering Task Force (IETF)                 IJ. Wijnands, Ed.
Request for Comments: 7246                                 Cisco Systems
Category: Standards Track                                     P. Hitchen
ISSN: 2070-1721                                                       BT
                                                              N. Leymann
                                                        Deutsche Telekom
                                                           W. Henderickx
                                                          Alcatel-Lucent
                                                                A. Gulko
                                                         Thomson Reuters
                                                             J. Tantsura
                                                                Ericsson
                                                               June 2014


      Multipoint Label Distribution Protocol In-Band Signaling in
          a Virtual Routing and Forwarding (VRF) Table Context

Abstract

   An IP Multicast Distribution Tree (MDT) may traverse both label
   switching (i.e., Multiprotocol Label Switching, or MPLS) and non-
   label switching regions of a network.  Typically, the MDT begins and
   ends in non-MPLS regions, but travels through an MPLS region.  In
   such cases, it can be useful to begin building the MDT as a pure IP
   MDT, then convert it to an MPLS Multipoint Label Switched Path
   (MP-LSP) when it enters an MPLS-enabled region, and then convert it
   back to a pure IP MDT when it enters a non-MPLS-enabled region.
   Other documents specify the procedures for building such a hybrid
   MDT, using Protocol Independent Multicast (PIM) in the non-MPLS
   region of the network, and using Multipoint Label Distribution
   Protocol (mLDP) in the MPLS region.  This document extends those
   procedures to handle the case where the link connecting the two
   regions is a Virtual Routing and Forwarding (VRF) table link, as
   defined in the "BGP IP/MPLS VPN" specification.  However, this
   document is primarily aimed at particular use cases where VRFs are
   used to support multicast applications other than multicast VPN.

Status of This Memo

   This is an Internet Standards Track document.

   This document is a product of the Internet Engineering Task Force
   (IETF).  It represents the consensus of the IETF community.  It has
   received public review and has been approved for publication by the
   Internet Engineering Steering Group (IESG).  Further information on
   Internet Standards is available in Section 2 of RFC 5741.

   Information about the current status of this document, any errata,
   and how to provide feedback on it may be obtained at
   http://www.rfc-editor.org/info/rfc7246.

Copyright Notice

   Copyright (c) 2014 IETF Trust and the persons identified as the
   document authors.  All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   (http://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must
   include Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.

Table of Contents

   1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  3
     1.1.  Conventions Used in This Document  . . . . . . . . . . . .  4
     1.2.  Terminology  . . . . . . . . . . . . . . . . . . . . . . .  5
   2.  VRF In-Band Signaling for MP LSPs  . . . . . . . . . . . . . .  5
   3.  Encoding the Opaque Value of an LDP MP FEC . . . . . . . . . .  7
     3.1.  Transit VPNv4 Source TLV . . . . . . . . . . . . . . . . .  7
     3.2.  Transit VPNv6 Source TLV . . . . . . . . . . . . . . . . .  8
     3.3.  Transit VPNv4 Bidir TLV  . . . . . . . . . . . . . . . . .  9
     3.4.  Transit VPNv6 Bidir TLV  . . . . . . . . . . . . . . . . . 10
   4.  Security Considerations  . . . . . . . . . . . . . . . . . . . 10
   5.  IANA Considerations  . . . . . . . . . . . . . . . . . . . . . 11
   6.  Acknowledgments  . . . . . . . . . . . . . . . . . . . . . . . 11
   7.  References . . . . . . . . . . . . . . . . . . . . . . . . . . 11
     7.1.  Normative References . . . . . . . . . . . . . . . . . . . 11
     7.2.  Informative References . . . . . . . . . . . . . . . . . . 12

1.  Introduction

   Sometimes an IP Multicast Distribution Tree (MDT) traverses both
   MPLS-enabled and non-MPLS-enabled regions of a network.  Typically,
   the MDT begins and ends in non-MPLS regions, but travels through an
   MPLS region.  In such cases, it can be useful to begin building the
   MDT as a pure IP MDT, then convert it to an MPLS Multipoint Label
   Switched Path (LSP) when it enters an MPLS-enabled region, and then
   convert it back to a pure IP MDT when it enters a non-MPLS-enabled
   region.  Other documents specify the procedures for building such a
   hybrid MDT, using Protocol Independent Multicast (PIM) in the non-
   MPLS region of the network, and using Multipoint Label Distribution
   Protocol (mLDP) in the MPLS region.  This document extends the
   procedures from [RFC6826] to handle the case where the link
   connecting the two regions is a Virtual Routing and Forwarding (VRF)
   table link, as defined in the "BGP IP/MPLS VPN" specification
   [RFC6513].  However, this document is primarily aimed at particular
   use cases where VRFs are used to support multicast applications other
   than multicast VPN.

   In PIM, a tree is identified by a source address (or in the case of
   bidirectional trees [RFC5015], a rendezvous point address or "RPA")
   and a group address.  The tree is built from the leaves up, by
   sending PIM control messages in the direction of the source address
   or the RPA.  In mLDP, a tree is identified by a root address and an
   "opaque value", and is built by sending mLDP control messages in the
   direction of the root.  The procedures of [RFC6826] explain how to
   convert a PIM <source address or RPA, group address> pair into an
   mLDP <root node, opaque value> pair and how to convert the mLDP <root
   node, opaque value> pair back into the original PIM <source address
   or RPA, group address> pair.

   However, the procedures in [RFC6826] assume that the routers doing
   the PIM/mLDP conversion have routes to the source address or RPA in
   their global routing tables.  Thus, the procedures cannot be applied
   exactly as specified when the interfaces connecting the non-MPLS-
   enabled region to the MPLS-enabled region are interfaces that belong
   to a VRF as described in [RFC4364].  This specification extends the
   procedures of [RFC6826] so that they may be applied in the VRF
   context.

   As in [RFC6826], the scope of this document is limited to the case
   where the multicast content is distributed in the non-MPLS-enabled
   regions via PIM-created source-specific or bidirectional trees.
   Bidirectional trees are always mapped onto multipoint-to-multipoint
   LSPs, and source-specific trees are always mapped onto point-to-
   multipoint LSPs.

   Note that the procedures described herein do not support non-
   bidirectional PIM Any-Source Multicast (ASM) groups, do not support
   the use of multicast trees other than mLDP multipoint LSPs in the
   core, and do not provide the capability to aggregate multiple PIM
   trees onto a single multipoint LSP.  If any of those features are
   needed, they can be provided by the procedures of [RFC6513] and
   [RFC6514].  However, there are cases where multicast services are
   offered through interfaces associated with a VRF, and where mLDP is
   used in the core, but where aggregation is not desired.  For example,
   some service providers offer multicast content to their customers,
   but have chosen to use VRFs to isolate the various customers and
   services.  This is a simpler scenario than one in which the customers
   provide their own multicast content, out of the control of the
   service provider, and can be handled with a simpler solution.  Also,
   when PIM trees are mapped one-to-one to multipoint LSPs, it is
   helpful for troubleshooting purposes to have the PIM source/group
   addresses encoded into the mLDP FEC (Forwarding Equivalence Class)
   element in what this document terms "mLDP in-band signaling".

   In order to use the mLDP in-band signaling procedures for a
   particular group address in the context of a particular set of VRFs,
   those VRFs MUST be configured with a range of multicast group
   addresses for which mLDP in-band signaling is to be enabled.  This
   configuration is per VRF defined in [RFC4364]).  For those groups,
   and those groups only, the procedures of this document are used.  For
   other groups, the general-purpose multicast VPN procedures MAY be
   used, although it is more likely this VRF is dedicated to mLDP in-
   band signaling procedures and all other groups are discarded.  The
   configuration MUST be present in all PE routers that attach to sites
   containing senders or receivers for the given set of group addresses.
   Note that because the provider most likely owns the multicast content
   and the method of transportation across the network, which are both
   transparent to the end user, no coordination needs to happen between
   the end user and the provider.

1.1.  Conventions Used in This Document

   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 RFC 2119 [RFC2119].

1.2.  Terminology

   In-band signaling:  Using the opaque value of an mLDP FEC element to
      encode the (S,G) or (*,G) identifying a particular IP multicast
      tree.

   Ingress LSR:  Source of a P2MP LSP, also referred to as root node.

   IP multicast tree:  An IP multicast distribution tree identified by a
      source IP address and/or IP multicast destination address, also
      referred to as (S,G) and (*,G).

   mLDP:  Multipoint LDP.

   MP LSP:  A multipoint LSP, either a P2MP or an MP2MP LSP.

   MP2MP LSP:  An LSP that connects a set of leaf nodes, acting
      indifferently as ingress or egress (see [RFC6388]).

   P2MP LSP:  An LSP that has one Ingress LSR and one or more Egress
      LSRs (see [RFC6388]).

   RPA: Rendezvous Point Address, the address that is used as the root
      of the distribution tree for a range of multicast groups.

   RD: Route Distinguisher, an identifier that makes a route unique in
      the context of a VRF.

   UMH: Upstream Multicast Hop, the upstream router in that is in the
      path to reach the source of the multicast flow.

   VRF:  Virtual Routing and Forwarding table.

2.  VRF In-Band Signaling for MP LSPs

   Suppose that a PE router, PE1, receives a PIM Join(S,G) message over
   one of its interfaces that is associated with a VRF.  Following the
   procedure of Section 5.1 of [RFC6513], PE1 determines the "upstream
   RD", the "upstream PE", and the "upstream multicast hop" (UMH) for
   the source address S.

   In order to transport the multicast tree via a multipoint (MP) LSP
   using VRF in-band signaling, an mLDP Label Mapping message is sent by
   PE1.  This message will contain either a P2MP FEC or an MP2MP FEC
   (see [RFC6388]), depending upon whether the PIM tree being
   transported is a source-specific tree, or a bidirectional tree,
   respectively.  The FEC contains a "root" and an "opaque value".

   If the UMH and the upstream PE have the same IP address (i.e., the
   upstream PE is the UMH), then the root of the multipoint FEC is set
   to the IP address of the upstream PE.  If, in the context of this
   VPN, (S,G) refers to a source-specific MDT, then the values of S, G,
   and the upstream RD are encoded into the opaque value.  If, in the
   context of this VPN, G is a bidirectional group address, then S is
   replaced with the value of the RPA associated with G.

   The encoding details are specified in Section 3.  Conceptually, the
   multipoint FEC can be thought of as an ordered pair:
   {root=<Upstream-PE>; opaque_value=<S or RPA , G, Upstream-RD>}.  The
   mLDP Label Mapping message is then sent by PE1 on its LDP session to
   the "next hop" on the message's path to the upstream PE.  The "next
   hop" is usually the directly connected next hop, but see [RFC7060]
   for cases in which the next hop is not directly connected.

   If the UMH and the upstream PE do not have the same IP address, the
   procedures of Section 2 of [RFC6512] should be applied.  The root
   node of the multipoint FEC is set to the UMH.  The recursive opaque
   value is then set as follows: the root node is set to the upstream
   PE, and the opaque value is set to the multipoint FEC described in
   the previous paragraph.  That is, the multipoint FEC can be thought
   of as the following recursive ordered pair: {root=<UMH>;
   opaque_value=<root=Upstream-PE, opaque_value=<S or RPA, G,
   Upstream-RD>>}.

   The encoding of the multipoint FEC also specifies the "type" of PIM
   MDT being spliced onto the multipoint LSP.  Four opaque encodings are
   defined in [RFC6826]: IPv4 source-specific tree, IPv6 source-specific
   tree, IPv4 bidirectional tree, and IPv6 bidirectional tree.

   When a PE router receives an mLDP message with a P2MP or MP2MP FEC,
   where the PE router itself is the root node, and the opaque value is
   one of the types defined in Section 3, then it uses the RD encoded in
   the opaque value field to determine the VRF context.  (This RD will
   be associated with one of the PEs VRFs.)  Then, in the context of
   that VRF, the PE follows the procedure specified in section 2 of
   [RFC6826].

3.  Encoding the Opaque Value of an LDP MP FEC

   This section documents the different transit opaque encodings.

3.1.  Transit VPNv4 Source TLV

   This opaque value type is used when transporting a source-specific
   mode multicast tree whose source and group addresses are IPv4
   addresses.

    0                   1                   2                   3
    0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   | Type          | Length                        | Source
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
                                                   | Group
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
                                                   |               ~
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   ~                   RD                          |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Type:  250

   Length:  16

   Source:  IPv4 multicast source address, 4 octets.

   Group:  IPv4 multicast group address, 4 octets.

   RD:  Route Distinguisher, 8 octets.

3.2.  Transit VPNv6 Source TLV

   This opaque value type is used when transporting a source-specific
   mode multicast tree whose source and group addresses are IPv6
   addresses.

    0                   1                   2                   3
    0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   | Type          | Length                        | Source        ~
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   ~                                               | Group         ~
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   ~                                               |               ~
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   ~                 RD                            |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Type:  251

   Length:  40

   Source:  IPv6 multicast source address, 16 octets.

   Group:  IPv6 multicast group address, 16 octets.

   RD:  Route Distinguisher, 8 octets.

3.3.  Transit VPNv4 Bidir TLV

   This opaque value type is used when transporting a bidirectional
   multicast tree whose group address is an IPv4 address.  The RP
   address is also an IPv4 address in this case.

    0                   1                   2                   3
    0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   | Type          | Length                        | Mask Len      |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                              RP                               |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                            Group                              |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   ~                              RD                               |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Type:  9

   Length:  17

   Mask Len:  The number of contiguous one bits that are left justified
      and used as a mask, 1 octet.

   RP:  Rendezvous Point (RP) IPv4 address used for the encoded Group, 4
      octets.

   Group:  IPv4 multicast group address, 4 octets.

   RD:  Route Distinguisher, 8 octets.

3.4.  Transit VPNv6 Bidir TLV

   This opaque value type is used when transporting a bidirectional
   multicast tree whose group address is an IPv6 address.  The RP
   address is also an IPv6 address in this case.

    0                   1                   2                   3
    0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   | Type          | Length                        | Mask Len      |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                              RP                               ~
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   ~                                                               |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                            Group                              ~
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   ~                                                               |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   ~                              RD                               |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Type:  10

   Length:  41

   Mask Len:  The number of contiguous one bits that are left justified
      and used as a mask, 1 octet.

   RP:  Rendezvous Point (RP) IPv6 address used for the encoded group,
      16 octets.

   Group:  IPv6 multicast group address, 16 octets.

   RD:  Route Distinguisher, 8 octets.

4.  Security Considerations

   The same security considerations apply as for the base LDP
   specification, described in [RFC5036], and the base mLDP
   specification, described in [RFC6388].

   Operators MUST configure packet filters to ensure that the mechanism
   described in this memo does not cause non-global-scoped IPv6
   multicast packets to be tunneled outside of their intended scope.

5.  IANA Considerations

   [RFC6388] defines a registry for the "LDP MP Opaque Value Element
   basic type".  IANA has assigned four new code points in this
   registry:

      Transit VPNv4 Source TLV type - 250

      Transit VPNv6 Source TLV type - 251

      Transit VPNv4 Bidir TLV type - 9

      Transit VPNv6 Bidir TLV type - 10

6.  Acknowledgments

   Thanks to Eric Rosen, Andy Green, Yakov Rekhter, and Eric Gray for
   their comments on the document.

7.  References

7.1.  Normative References

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

   [RFC4364]  Rosen, E. and Y. Rekhter, "BGP/MPLS IP Virtual Private
              Networks (VPNs)", RFC 4364, February 2006.

   [RFC5015]  Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano,
              "Bidirectional Protocol Independent Multicast (BIDIR-
              PIM)", RFC 5015, October 2007.

   [RFC5036]  Andersson, L., Ed., Minei, I., Ed., and B. Thomas, Ed.,
              "LDP Specification", RFC 5036, October 2007.

   [RFC6388]  Wijnands, IJ., Ed., Minei, I., Ed., Kompella, K., and B.
              Thomas, "Label Distribution Protocol Extensions for Point-
              to-Multipoint and Multipoint-to-Multipoint Label Switched
              Paths", RFC 6388, November 2011.

   [RFC6512]  Wijnands, IJ., Rosen, E., Napierala, M., and N. Leymann,
              "Using Multipoint LDP When the Backbone Has No Route to
              the Root", RFC 6512, February 2012.

   [RFC6826]  Wijnands, IJ., Ed., Eckert, T., Leymann, N., and M.
              Napierala, "Multipoint LDP In-Band Signaling for Point-to-
              Multipoint and Multipoint-to-Multipoint Label Switched
              Paths", RFC 6826, January 2013.

7.2.  Informative References

   [RFC6513]  Rosen, E., Ed., and R. Aggarwal, Ed., "Multicast in
              MPLS/BGP IP VPNs", RFC 6513, February 2012.

   [RFC6514]  Aggarwal, R., Rosen, E., Morin, T., and Y. Rekhter, "BGP
              Encodings and Procedures for Multicast in MPLS/BGP IP
              VPNs", RFC 6514, February 2012.

   [RFC7060]  Napierala, M., Rosen, E., and IJ. Wijnands, "Using LDP
              Multipoint Extensions on Targeted LDP Sessions", RFC 7060,
              November 2013.

Authors' Addresses

   IJsbrand Wijnands (editor)
   Cisco Systems
   De kleetlaan 6a
   Diegem  1831
   Belgium
   EMail: ice@cisco.com

   Paul Hitchen
   BT
   BT Adastral Park
   Ipswich  IP53RE
   United Kingdom
   EMail: paul.hitchen@bt.com

   Nicolai Leymann
   Deutsche Telekom
   Winterfeldtstrasse 21
   Berlin  10781
   Germany
   EMail: n.leymann@telekom.de

   Wim Henderickx
   Alcatel-Lucent
   Copernicuslaan 50
   Antwerp  2018
   Belgium
   EMail: wim.henderickx@alcatel-lucent.com

   Arkadiy Gulko
   Thomson Reuters
   195 Broadway
   New York, NY  10007
   United States
   EMail: arkadiy.gulko@thomsonreuters.com

   Jeff Tantsura
   Ericsson
   300 Holger Way
   San Jose, CA  95134
   United States
   EMail: jeff.tantsura@ericsson.com