Network Working Group                                        K. Kompella
Request for Comments: 4379                        Juniper Networks, Inc.
Updates: 1122                                                 G. Swallow
Category: Standards Track                            Cisco Systems, Inc.
                                                          February 2006


  Detecting Multi-Protocol Label Switched (MPLS) Data Plane Failures

Status of This Memo

  This document specifies an Internet standards track protocol for the
  Internet community, and requests discussion and suggestions for
  improvements.  Please refer to the current edition of the "Internet
  Official Protocol Standards" (STD 1) for the standardization state
  and status of this protocol.  Distribution of this memo is unlimited.

Copyright Notice

  Copyright (C) The Internet Society (2006).

Abstract

  This document describes a simple and efficient mechanism that can be
  used to detect data plane failures in Multi-Protocol Label Switching
  (MPLS) Label Switched Paths (LSPs).  There are two parts to this
  document: information carried in an MPLS "echo request" and "echo
  reply" for the purposes of fault detection and isolation, and
  mechanisms for reliably sending the echo reply.

Table of Contents

  1. Introduction ....................................................2
     1.1. Conventions ................................................3
     1.2. Structure of This Document .................................3
     1.3. Contributors ...............................................3
  2. Motivation ......................................................4
     2.1. Use of Address Range 127/8 .................................4
  3. Packet Format ...................................................6
     3.1. Return Codes ..............................................10
     3.2. Target FEC Stack ..........................................11
          3.2.1. LDP IPv4 Prefix ....................................12
          3.2.2. LDP IPv6 Prefix ....................................13
          3.2.3. RSVP IPv4 LSP ......................................13
          3.2.4. RSVP IPv6 LSP ......................................14
          3.2.5. VPN IPv4 Prefix ....................................14
          3.2.6. VPN IPv6 Prefix ....................................15
          3.2.7. L2 VPN Endpoint ....................................16



Kompella & Swallow          Standards Track                     [Page 1]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


          3.2.8. FEC 128 Pseudowire (Deprecated) ....................16
          3.2.9. FEC 128 Pseudowire (Current) .......................17
          3.2.10. FEC 129 Pseudowire ................................18
          3.2.11. BGP Labeled IPv4 Prefix ...........................19
          3.2.12. BGP Labeled IPv6 Prefix ...........................20
          3.2.13. Generic IPv4 Prefix ...............................20
          3.2.14. Generic IPv6 Prefix ...............................21
          3.2.15. Nil FEC ...........................................21
     3.3. Downstream Mapping ........................................22
          3.3.1. Multipath Information Encoding .....................26
          3.3.2. Downstream Router and Interface ....................28
     3.4. Pad TLV ...................................................29
     3.5. Vendor Enterprise Number ..................................29
     3.6. Interface and Label Stack .................................29
     3.7. Errored TLVs ..............................................31
     3.8. Reply TOS Byte TLV ........................................31
  4. Theory of Operation ............................................32
     4.1. Dealing with Equal-Cost Multi-Path (ECMP) .................32
     4.2. Testing LSPs That Are Used to Carry MPLS Payloads .........33
     4.3. Sending an MPLS Echo Request ..............................33
     4.4. Receiving an MPLS Echo Request ............................34
          4.4.1. FEC Validation .....................................40
     4.5. Sending an MPLS Echo Reply ................................41
     4.6. Receiving an MPLS Echo Reply ..............................42
     4.7. Issue with VPN IPv4 and IPv6 Prefixes .....................42
     4.8. Non-compliant Routers .....................................43
  5. References .....................................................43
     5.1. Normative References ......................................43
     5.2. Informative References ....................................44
  6. Security Considerations ........................................44
  7. IANA Considerations ............................................46
     7.1. Message Types, Reply Modes, Return Codes ..................46
     7.2. TLVs ......................................................47
  8. Acknowledgements ...............................................48

1.  Introduction

  This document describes a simple and efficient mechanism that can be
  used to detect data plane failures in MPLS Label Switched Paths
  (LSPs).  There are two parts to this document: information carried in
  an MPLS "echo request" and "echo reply", and mechanisms for
  transporting the echo reply.  The first part aims at providing enough
  information to check correct operation of the data plane, as well as
  a mechanism to verify the data plane against the control plane, and
  thereby localize faults.  The second part suggests two methods of
  reliable reply channels for the echo request message for more robust
  fault isolation.




Kompella & Swallow          Standards Track                     [Page 2]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  An important consideration in this design is that MPLS echo requests
  follow the same data path that normal MPLS packets would traverse.
  MPLS echo requests are meant primarily to validate the data plane,
  and secondarily to verify the data plane against the control plane.
  Mechanisms to check the control plane are valuable, but are not
  covered in this document.

  This document makes special use of the address range 127/8.  This is
  an exception to the behavior defined in RFC 1122 [RFC1122] and
  updates that RFC.  The motivation for this change and the details of
  this exceptional use are discussed in section 2.1 below.

1.1.  Conventions

  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 [KEYWORDS].

  The term "Must Be Zero" (MBZ) is used in object descriptions for
  reserved fields.  These fields MUST be set to zero when sent and
  ignored on receipt.

  Terminology pertaining to L2 and L3 Virtual Private Networks (VPNs)
  is defined in [RFC4026].

  Since this document refers to the MPLS Time to Live (TTL) far more
  frequently than the IP TTL, the authors have chosen the convention of
  using the unqualified "TTL" to mean "MPLS TTL" and using "IP TTL" for
  the TTL value in the IP header.

1.2.  Structure of This Document

  The body of this memo contains four main parts: motivation, MPLS echo
  request/reply packet format, LSP ping operation, and a reliable
  return path.  It is suggested that first-time readers skip the actual
  packet formats and read the Theory of Operation first; the document
  is structured the way it is to avoid forward references.

1.3.  Contributors

  The following made vital contributions to all aspects of this
  document, and much of the material came out of debate and discussion
  among this group.

     Ronald P. Bonica, Juniper Networks, Inc.
     Dave Cooper, Global Crossing
     Ping Pan, Hammerhead Systems




Kompella & Swallow          Standards Track                     [Page 3]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


     Nischal Sheth, Juniper Networks, Inc.
     Sanjay Wadhwa, Juniper Networks, Inc.

2.  Motivation

  When an LSP fails to deliver user traffic, the failure cannot always
  be detected by the MPLS control plane.  There is a need to provide a
  tool that would enable users to detect such traffic "black holes" or
  misrouting within a reasonable period of time, and a mechanism to
  isolate faults.

  In this document, we describe a mechanism that accomplishes these
  goals.  This mechanism is modeled after the ping/traceroute paradigm:
  ping (ICMP echo request [ICMP]) is used for connectivity checks, and
  traceroute is used for hop-by-hop fault localization as well as path
  tracing.  This document specifies a "ping" mode and a "traceroute"
  mode for testing MPLS LSPs.

  The basic idea is to verify that packets that belong to a particular
  Forwarding Equivalence Class (FEC) actually end their MPLS path on a
  Label Switching Router (LSR) that is an egress for that FEC.  This
  document proposes that this test be carried out by sending a packet
  (called an "MPLS echo request") along the same data path as other
  packets belonging to this FEC.  An MPLS echo request also carries
  information about the FEC whose MPLS path is being verified.  This
  echo request is forwarded just like any other packet belonging to
  that FEC.  In "ping" mode (basic connectivity check), the packet
  should reach the end of the path, at which point it is sent to the
  control plane of the egress LSR, which then verifies whether it is
  indeed an egress for the FEC.  In "traceroute" mode (fault
  isolation), the packet is sent to the control plane of each transit
  LSR, which performs various checks that it is indeed a transit LSR
  for this path; this LSR also returns further information that helps
  check the control plane against the data plane, i.e., that forwarding
  matches what the routing protocols determined as the path.

  One way these tools can be used is to periodically ping an FEC to
  ensure connectivity.  If the ping fails, one can then initiate a
  traceroute to determine where the fault lies.  One can also
  periodically traceroute FECs to verify that forwarding matches the
  control plane; however, this places a greater burden on transit LSRs
  and thus should be used with caution.

2.1.  Use of Address Range 127/8

  As described above, LSP ping is intended as a diagnostic tool.  It is
  intended to enable providers of an MPLS-based service to isolate
  network faults.  In particular, LSP ping needs to diagnose situations



Kompella & Swallow          Standards Track                     [Page 4]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  where the control and data planes are out of sync.  It performs this
  by routing an MPLS echo request packet based solely on its label
  stack.  That is, the IP destination address is never used in a
  forwarding decision.  In fact, the sender of an MPLS echo request
  packet may not know, a priori, the address of the router at the end
  of the LSP.

  Providers of MPLS-based services also need the ability to trace all
  of the possible paths that an LSP may take.  Since most MPLS services
  are based on IP unicast forwarding, these paths are subject to
  equal-cost multi-path (ECMP) load sharing.

  This leads to the following requirements:

  1. Although the LSP in question may be broken in unknown ways, the
     likelihood of a diagnostic packet being delivered to a user of an
     MPLS service MUST be held to an absolute minimum.

  2. If an LSP is broken in such a way that it prematurely terminates,
     the diagnostic packet MUST NOT be IP forwarded.

  3. A means of varying the diagnostic packets such that they exercise
     all ECMP paths is thus REQUIRED.

  Clearly, using general unicast addresses satisfies neither of the
  first two requirements.  A number of other options for addresses were
  considered, including a portion of the private address space (as
  determined by the network operator) and the newly designated IPv4
  link local addresses.  Use of the private address space was deemed
  ineffective since the leading MPLS-based service is an IPv4 Virtual
  Private Network (VPN).  VPNs often use private addresses.

  The IPv4 link local addresses are more attractive in that the scope
  over which they can be forwarded is limited.  However, if one were to
  use an address from this range, it would still be possible for the
  first recipient of a diagnostic packet that "escaped" from a broken
  LSP to have that address assigned to the interface on which it
  arrived and thus could mistakenly receive such a packet.
  Furthermore, the IPv4 link local address range has only recently been
  allocated.  Many deployed routers would forward a packet with an
  address from that range toward the default route.

  The 127/8 range for IPv4 and that same range embedded in as IPv4-
  mapped IPv6 addresses for IPv6 was chosen for a number of reasons.

  RFC 1122 allocates the 127/8 as "Internal host loopback address" and
  states: "Addresses of this form MUST NOT appear outside a host."
  Thus, the default behavior of hosts is to discard such packets.  This



Kompella & Swallow          Standards Track                     [Page 5]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  helps to ensure that if a diagnostic packet is misdirected to a host,
  it will be silently discarded.

  RFC 1812 [RFC1812] states:

     A router SHOULD NOT forward, except over a loopback interface, any
     packet that has a destination address on network 127.  A router
     MAY have a switch that allows the network manager to disable these
     checks.  If such a switch is provided, it MUST default to
     performing the checks.

  This helps to ensure that diagnostic packets are never IP forwarded.

  The 127/8 address range provides 16M addresses allowing wide
  flexibility in varying addresses to exercise ECMP paths.  Finally, as
  an implementation optimization, the 127/8 provides an easy means of
  identifying possible LSP packets.

3.  Packet Format

  An MPLS echo request is a (possibly labeled) IPv4 or IPv6 UDP packet;
  the contents of the UDP packet have the following format:

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |         Version Number        |         Global Flags          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |  Message Type |   Reply mode  |  Return Code  | Return Subcode|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                        Sender's Handle                        |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                        Sequence Number                        |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                    TimeStamp Sent (seconds)                   |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                  TimeStamp Sent (microseconds)                |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                  TimeStamp Received (seconds)                 |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                TimeStamp Received (microseconds)              |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                            TLVs ...                           |
     .                                                               .
     .                                                               .
     .                                                               .
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+



Kompella & Swallow          Standards Track                     [Page 6]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  The Version Number is currently 1.  (Note: the version number is to
  be incremented whenever a change is made that affects the ability of
  an implementation to correctly parse or process an MPLS echo
  request/reply.  These changes include any syntactic or semantic
  changes made to any of the fixed fields, or to any Type-Length-Value
  (TLV) or sub-TLV assignment or format that is defined at a certain
  version number.  The version number may not need to be changed if an
  optional TLV or sub-TLV is added.)

  The Global Flags field is a bit vector with the following format:

      0                   1
      0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |             MBZ             |V|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

  One flag is defined for now, the V bit; the rest MUST be set to zero
  when sending and ignored on receipt.

  The V (Validate FEC Stack) flag is set to 1 if the sender wants the
  receiver to perform FEC Stack validation; if V is 0, the choice is
  left to the receiver.

  The Message Type is one of the following:

     Value    Meaning
     -----    -------
         1    MPLS echo request
         2    MPLS echo reply

  The Reply Mode can take one of the following values:

     Value    Meaning
     -----    -------
         1    Do not reply
         2    Reply via an IPv4/IPv6 UDP packet
         3    Reply via an IPv4/IPv6 UDP packet with Router Alert
         4    Reply via application level control channel

  An MPLS echo request with 1 (Do not reply) in the Reply Mode field
  may be used for one-way connectivity tests; the receiving router may
  log gaps in the Sequence Numbers and/or maintain delay/jitter
  statistics.  An MPLS echo request would normally have 2 (Reply via an
  IPv4/IPv6 UDP packet) in the Reply Mode field.  If the normal IP
  return path is deemed unreliable, one may use 3 (Reply via an
  IPv4/IPv6 UDP packet with Router Alert).  Note that this requires
  that all intermediate routers understand and know how to forward MPLS



Kompella & Swallow          Standards Track                     [Page 7]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  echo replies.  The echo reply uses the same IP version number as the
  received echo request, i.e., an IPv4 encapsulated echo reply is sent
  in response to an IPv4 encapsulated echo request.

  Some applications support an IP control channel.  One such example is
  the associated control channel defined in Virtual Circuit
  Connectivity Verification (VCCV) [VCCV].  Any application that
  supports an IP control channel between its control entities may set
  the Reply Mode to 4 (Reply via application level control channel) to
  ensure that replies use that same channel.  Further definition of
  this codepoint is application specific and thus beyond the scope of
  this document.

  Return Codes and Subcodes are described in the next section.

  The Sender's Handle is filled in by the sender, and returned
  unchanged by the receiver in the echo reply (if any).  There are no
  semantics associated with this handle, although a sender may find
  this useful for matching up requests with replies.

  The Sequence Number is assigned by the sender of the MPLS echo
  request and can be (for example) used to detect missed replies.

  The TimeStamp Sent is the time-of-day (in seconds and microseconds,
  according to the sender's clock) in NTP format [NTP] when the MPLS
  echo request is sent.  The TimeStamp Received in an echo reply is the
  time-of-day (according to the receiver's clock) in NTP format that
  the corresponding echo request was received.

  TLVs (Type-Length-Value tuples) have the following format:

      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             |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                             Value                             |
     .                                                               .
     .                                                               .
     .                                                               .
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

  Types are defined below; Length is the length of the Value field in
  octets.  The Value field depends on the Type; it is zero padded to
  align to a 4-octet boundary.  TLVs may be nested within other TLVs,
  in which case the nested TLVs are called sub-TLVs.  Sub-TLVs have
  independent types and MUST also be 4-octet aligned.



Kompella & Swallow          Standards Track                     [Page 8]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  Two examples follow.  The Label Distribution Protocol (LDP) IPv4 FEC
  sub-TLV has the following format:

      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 = 1 (LDP IPv4 FEC)    |          Length = 5           |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                          IPv4 prefix                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Prefix Length |         Must Be Zero                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

  The Length for this TLV is 5.  A Target FEC Stack TLV that contains
  an LDP IPv4 FEC sub-TLV and a VPN IPv4 prefix sub-TLV has the
  following format:

      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 = 1 (FEC TLV)       |          Length = 12          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |  sub-Type = 1 (LDP IPv4 FEC)  |          Length = 5           |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                          IPv4 prefix                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Prefix Length |         Must Be Zero                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | sub-Type = 6 (VPN IPv4 prefix)|          Length = 13          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                      Route Distinguisher                      |
     |                          (8 octets)                           |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                         IPv4 prefix                           |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Prefix Length |                 Must Be Zero                  |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+














Kompella & Swallow          Standards Track                     [Page 9]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  A description of the Types and Values of the top-level TLVs for LSP
  ping are given below:

         Type #                  Value Field
         ------                  -----------
              1                  Target FEC Stack
              2                  Downstream Mapping
              3                  Pad
              4                  Not Assigned
              5                  Vendor Enterprise Number
              6                  Not Assigned
              7                  Interface and Label Stack
              8                  Not Assigned
              9                  Errored TLVs
             10                  Reply TOS Byte

  Types less than 32768 (i.e., with the high-order bit equal to 0) are
  mandatory TLVs that MUST either be supported by an implementation or
  result in the return code of 2 ("One or more of the TLVs was not
  understood") being sent in the echo response.

  Types greater than or equal to 32768 (i.e., with the high-order bit
  equal to 1) are optional TLVs that SHOULD be ignored if the
  implementation does not understand or support them.

3.1.  Return Codes

  The Return Code is set to zero by the sender.  The receiver can set
  it to one of the values listed below.  The notation <RSC> refers to
  the Return Subcode.  This field is filled in with the stack-depth for
  those codes that specify that.  For all other codes, the Return
  Subcode MUST be set to zero.

         Value    Meaning
         -----    -------

             0    No return code

             1    Malformed echo request received

             2    One or more of the TLVs was not understood

             3    Replying router is an egress for the FEC at stack-
                  depth <RSC>

             4    Replying router has no mapping for the FEC at stack-
                  depth <RSC>




Kompella & Swallow          Standards Track                    [Page 10]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


             5    Downstream Mapping Mismatch (See Note 1)

             6    Upstream Interface Index Unknown (See Note 1)

             7    Reserved

             8    Label switched at stack-depth <RSC>

             9    Label switched but no MPLS forwarding at stack-depth
                  <RSC>

            10    Mapping for this FEC is not the given label at stack-
                  depth <RSC>

            11    No label entry at stack-depth <RSC>

            12    Protocol not associated with interface at FEC stack-
                  depth <RSC>

            13    Premature termination of ping due to label stack
                  shrinking to a single label

  Note 1

     The Return Subcode contains the point in the label stack where
     processing was terminated.  If the RSC is 0, no labels were
     processed.  Otherwise the packet would have been label switched at
     depth RSC.

3.2.  Target FEC Stack

  A Target FEC Stack is a list of sub-TLVs.  The number of elements is
  determined by looking at the sub-TLV length fields.

     Sub-Type       Length            Value Field
     --------       ------            -----------
            1            5            LDP IPv4 prefix
            2           17            LDP IPv6 prefix
            3           20            RSVP IPv4 LSP
            4           56            RSVP IPv6 LSP
            5                         Not Assigned
            6           13            VPN IPv4 prefix
            7           25            VPN IPv6 prefix
            8           14            L2 VPN endpoint
            9           10            "FEC 128" Pseudowire (deprecated)
           10           14            "FEC 128" Pseudowire
           11          16+            "FEC 129" Pseudowire
           12            5            BGP labeled IPv4 prefix



Kompella & Swallow          Standards Track                    [Page 11]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


           13           17            BGP labeled IPv6 prefix
           14            5            Generic IPv4 prefix
           15           17            Generic IPv6 prefix
           16            4            Nil FEC

  Other FEC Types will be defined as needed.

  Note that this TLV defines a stack of FECs, the first FEC element
  corresponding to the top of the label stack, etc.

  An MPLS echo request MUST have a Target FEC Stack that describes the
  FEC Stack being tested.  For example, if an LSR X has an LDP mapping
  [LDP] for 192.168.1.1 (say, label 1001), then to verify that label
  1001 does indeed reach an egress LSR that announced this prefix via
  LDP, X can send an MPLS echo request with an FEC Stack TLV with one
  FEC in it, namely, of type LDP IPv4 prefix, with prefix
  192.168.1.1/32, and send the echo request with a label of 1001.

  Say LSR X wanted to verify that a label stack of <1001, 23456> is the
  right label stack to use to reach a VPN IPv4 prefix [see section
  3.2.5] of 10/8 in VPN foo.  Say further that LSR Y with loopback
  address 192.168.1.1 announced prefix 10/8 with Route Distinguisher
  RD-foo-Y (which may in general be different from the Route
  Distinguisher that LSR X uses in its own advertisements for VPN foo),
  label 23456 and BGP next hop 192.168.1.1 [BGP].  Finally, suppose
  that LSR X receives a label binding of 1001 for 192.168.1.1 via LDP.
  X has two choices in sending an MPLS echo request: X can send an MPLS
  echo request with an FEC Stack TLV with a single FEC of type VPN IPv4
  prefix with a prefix of 10/8 and a Route Distinguisher of RD-foo-Y.
  Alternatively, X can send an FEC Stack TLV with two FECs, the first
  of type LDP IPv4 with a prefix of 192.168.1.1/32 and the second of
  type of IP VPN with a prefix 10/8 with Route Distinguisher of RD-
  foo-Y.  In either case, the MPLS echo request would have a label
  stack of <1001, 23456>.  (Note: in this example, 1001 is the "outer"
  label and 23456 is the "inner" label.)

3.2.1.  LDP IPv4 Prefix

  The IPv4 Prefix FEC is defined in [LDP].  When an LDP IPv4 prefix is
  encoded in a label stack, the following format is used.  The value
  consists of 4 octets of an IPv4 prefix followed by 1 octet of prefix
  length in bits; the format is given below.  The IPv4 prefix is in
  network byte order; if the prefix is shorter than 32 bits, trailing
  bits SHOULD be set to zero.  See [LDP] for an example of a Mapping
  for an IPv4 FEC.






Kompella & Swallow          Standards Track                    [Page 12]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                          IPv4 prefix                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Prefix Length |         Must Be Zero                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

3.2.2.  LDP IPv6 Prefix

  The IPv6 Prefix FEC is defined in [LDP].  When an LDP IPv6 prefix is
  encoded in a label stack, the following format is used.  The value
  consists of 16 octets of an IPv6 prefix followed by 1 octet of prefix
  length in bits; the format is given below.  The IPv6 prefix is in
  network byte order; if the prefix is shorter than 128 bits, the
  trailing bits SHOULD be set to zero.  See [LDP] for an example of a
  Mapping for an IPv6 FEC.

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                          IPv6 prefix                          |
     |                          (16 octets)                          |
     |                                                               |
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Prefix Length |         Must Be Zero                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

3.2.3.  RSVP IPv4 LSP

  The value has the format below.  The value fields are taken from RFC
  3209, sections 4.6.1.1 and 4.6.2.1.  See [RSVP-TE].

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                 IPv4 tunnel end point address                 |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |          Must Be Zero         |     Tunnel ID                 |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                       Extended Tunnel ID                      |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                   IPv4 tunnel sender address                  |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |          Must Be Zero         |            LSP ID             |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+




Kompella & Swallow          Standards Track                    [Page 13]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


3.2.4.  RSVP IPv6 LSP

  The value has the format below.  The value fields are taken from RFC
  3209, sections 4.6.1.2 and 4.6.2.2.  See [RSVP-TE].

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                 IPv6 tunnel end point address                 |
     |                                                               |
     |                                                               |
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |          Must Be Zero         |          Tunnel ID            |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                       Extended Tunnel ID                      |
     |                                                               |
     |                                                               |
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                   IPv6 tunnel sender address                  |
     |                                                               |
     |                                                               |
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |          Must Be Zero         |            LSP ID             |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

3.2.5.  VPN IPv4 Prefix

  VPN-IPv4 Network Layer Routing Information (NLRI) is defined in
  [RFC4365].  This document uses the term VPN IPv4 prefix for a VPN-
  IPv4 NLRI that has been advertised with an MPLS label in BGP.  See
  [BGP-LABEL].

















Kompella & Swallow          Standards Track                    [Page 14]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  When a VPN IPv4 prefix is encoded in a label stack, the following
  format is used.  The value field consists of the Route Distinguisher
  advertised with the VPN IPv4 prefix, the IPv4 prefix (with trailing 0
  bits to make 32 bits in all), and a prefix length, as follows:

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                      Route Distinguisher                      |
     |                          (8 octets)                           |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                         IPv4 prefix                           |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Prefix Length |                 Must Be Zero                  |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

  The Route Distinguisher (RD) is an 8-octet identifier; it does not
  contain any inherent information.  The purpose of the RD is solely to
  allow one to create distinct routes to a common IPv4 address prefix.
  The encoding of the RD is not important here.  When matching this
  field to the local FEC information, it is treated as an opaque value.

3.2.6.  VPN IPv6 Prefix

  VPN-IPv6 Network Layer Routing Information (NLRI) is defined in
  [RFC4365].  This document uses the term VPN IPv6 prefix for a VPN-
  IPv6 NLRI that has been advertised with an MPLS label in BGP.  See
  [BGP-LABEL].

  When a VPN IPv6 prefix is encoded in a label stack, the following
  format is used.  The value field consists of the Route Distinguisher
  advertised with the VPN IPv6 prefix, the IPv6 prefix (with trailing 0
  bits to make 128 bits in all), and a prefix length, as follows:

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                      Route Distinguisher                      |
     |                          (8 octets)                           |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                         IPv6 prefix                           |
     |                                                               |
     |                                                               |
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Prefix Length |                 Must Be Zero                  |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+




Kompella & Swallow          Standards Track                    [Page 15]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  The Route Distinguisher is identical to the VPN IPv4 Prefix RD,
  except that it functions here to allow the creation of distinct
  routes to IPv6 prefixes.  See section 3.2.5.  When matching this
  field to local FEC information, it is treated as an opaque value.

3.2.7.  L2 VPN Endpoint

  VPLS stands for Virtual Private LAN Service.  The terms VPLS BGP NLRI
  and VE ID (VPLS Edge Identifier) are defined in [VPLS-BGP].  This
  document uses the simpler term L2 VPN endpoint when referring to a
  VPLS BGP NLRI.  The Route Distinguisher is an 8-octet identifier used
  to distinguish information about various L2 VPNs advertised by a
  node.  The VE ID is a 2-octet identifier used to identify a
  particular node that serves as the service attachment point within a
  VPLS.  The structure of these two identifiers is unimportant here;
  when matching these fields to local FEC information, they are treated
  as opaque values.  The encapsulation type is identical to the PW Type
  in section 3.2.8 below.

  When an L2 VPN endpoint is encoded in a label stack, the following
  format is used.  The value field consists of a Route Distinguisher (8
  octets), the sender (of the ping)'s VE ID (2 octets), the receiver's
  VE ID (2 octets), and an encapsulation type (2 octets), formatted as
  follows:

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                      Route Distinguisher                      |
     |                          (8 octets)                           |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |         Sender's VE ID        |       Receiver's VE ID        |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |      Encapsulation Type       |         Must Be Zero          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

3.2.8.  FEC 128 Pseudowire (Deprecated)

  FEC 128 (0x80) is defined in [PW-CONTROL], as are the terms PW ID
  (Pseudowire ID) and PW Type (Pseudowire Type).  A PW ID is a non-zero
  32-bit connection ID.  The PW Type is a 15-bit number indicating the
  encapsulation type.  It is carried right justified in the field below
  termed encapsulation type with the high-order bit set to zero.  Both
  of these fields are treated in this protocol as opaque values.







Kompella & Swallow          Standards Track                    [Page 16]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  When an FEC 128 is encoded in a label stack, the following format is
  used.  The value field consists of the remote PE address (the
  destination address of the targeted LDP session), the PW ID, and the
  encapsulation type as follows:

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                      Remote PE Address                        |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                             PW ID                             |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |            PW Type            |          Must Be Zero         |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

  This FEC is deprecated and is retained only for backward
  compatibility.  Implementations of LSP ping SHOULD accept and process
  this TLV, but SHOULD send LSP ping echo requests with the new TLV
  (see next section), unless explicitly configured to use the old TLV.

  An LSR receiving this TLV SHOULD use the source IP address of the LSP
  echo request to infer the sender's PE address.

3.2.9.  FEC 128 Pseudowire (Current)

  FEC 128 (0x80) is defined in [PW-CONTROL], as are the terms PW ID
  (Pseudowire ID) and PW Type (Pseudowire Type).  A PW ID is a non-zero
  32-bit connection ID.  The PW Type is a 15-bit number indicating the
  encapsulation type.  It is carried right justified in the field below
  termed encapsulation type with the high-order bit set to zero.

  Both of these fields are treated in this protocol as opaque values.
  When matching these field to the local FEC information, the match
  MUST be exact.

















Kompella & Swallow          Standards Track                    [Page 17]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  When an FEC 128 is encoded in a label stack, the following format is
  used.  The value field consists of the sender's PE address (the
  source address of the targeted LDP session), the remote PE address
  (the destination address of the targeted LDP session), the PW ID, and
  the encapsulation type as follows:

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                     Sender's PE Address                       |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                      Remote PE Address                        |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                             PW ID                             |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |            PW Type            |          Must Be Zero         |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

3.2.10.  FEC 129 Pseudowire

  FEC 129 (0x81) and the terms PW Type, Attachment Group Identifier
  (AGI), Attachment Group Identifier Type (AGI Type), Attachment
  Individual Identifier Type (AII Type), Source Attachment Individual
  Identifier (SAII), and Target Attachment Individual Identifier (TAII)
  are defined in [PW-CONTROL].  The PW Type is a 15-bit number
  indicating the encapsulation type.  It is carried right justified in
  the field below PW Type with the high-order bit set to zero.  All the
  other fields are treated as opaque values and copied directly from
  the FEC 129 format.  All of these values together uniquely define the
  FEC within the scope of the LDP session identified by the source and
  remote PE addresses.




















Kompella & Swallow          Standards Track                    [Page 18]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  When an FEC 129 is encoded in a label stack, the following format is
  used.  The Length of this TLV is 16 + AGI length + SAII length + TAII
  length.  Padding is used to make the total length a multiple of 4;
  the length of the padding is not included in the Length field.

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                     Sender's PE Address                       |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                      Remote PE Address                        |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |            PW Type            |   AGI Type    |  AGI Length   |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     ~                           AGI Value                           ~
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |   AII Type    |  SAII Length  |      SAII Value               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     ~                    SAII Value (continued)                     ~
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |   AII Type    |  TAII Length  |      TAII Value               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     ~                    TAII Value (continued)                     ~
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |  TAII (cont.) |  0-3 octets of zero padding                   |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

3.2.11.  BGP Labeled IPv4 Prefix

  BGP labeled IPv4 prefixes are defined in [BGP-LABEL].  When a BGP
  labeled IPv4 prefix is encoded in a label stack, the following format
  is used.  The value field consists the IPv4 prefix (with trailing 0
  bits to make 32 bits in all), and the prefix length, as follows:

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                          IPv4 Prefix                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Prefix Length |                 Must Be Zero                  |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+








Kompella & Swallow          Standards Track                    [Page 19]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


3.2.12.  BGP Labeled IPv6 Prefix

  BGP labeled IPv6 prefixes are defined in [BGP-LABEL].  When a BGP
  labeled IPv6 prefix is encoded in a label stack, the following format
  is used.  The value consists of 16 octets of an IPv6 prefix followed
  by 1 octet of prefix length in bits; the format is given below.  The
  IPv6 prefix is in network byte order; if the prefix is shorter than
  128 bits, the trailing bits SHOULD be set to zero.

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                          IPv6 prefix                          |
     |                          (16 octets)                          |
     |                                                               |
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Prefix Length |         Must Be Zero                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

3.2.13.  Generic IPv4 Prefix

  The value consists of 4 octets of an IPv4 prefix followed by 1 octet
  of prefix length in bits; the format is given below.  The IPv4 prefix
  is in network byte order; if the prefix is shorter than 32 bits,
  trailing bits SHOULD be set to zero.  This FEC is used if the
  protocol advertising the label is unknown or may change during the
  course of the LSP.  An example is an inter-AS LSP that may be
  signaled by LDP in one Autonomous System (AS), by RSVP-TE [RSVP-TE]
  in another AS, and by BGP between the ASes, such as is common for
  inter-AS VPNs.

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                          IPv4 prefix                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Prefix Length |         Must Be Zero                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+












Kompella & Swallow          Standards Track                    [Page 20]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


3.2.14.  Generic IPv6 Prefix

  The value consists of 16 octets of an IPv6 prefix followed by 1 octet
  of prefix length in bits; the format is given below.  The IPv6 prefix
  is in network byte order; if the prefix is shorter than 128 bits, the
  trailing bits SHOULD be set to zero.

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                          IPv6 prefix                          |
     |                          (16 octets)                          |
     |                                                               |
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Prefix Length |         Must Be Zero                          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

3.2.15.  Nil FEC

  At times, labels from the reserved range, e.g., Router Alert and
  Explicit-null, may be added to the label stack for various diagnostic
  purposes such as influencing load-balancing.  These labels may have
  no explicit FEC associated with them.  The Nil FEC Stack is defined
  to allow a Target FEC Stack sub-TLV to be added to the Target FEC
  Stack to account for such labels so that proper validation can still
  be performed.

  The Length is 4.  Labels are 20-bit values treated as numbers.

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                 Label                 |          MBZ          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

  Label is the actual label value inserted in the label stack; the MBZ
  fields MUST be zero when sent and ignored on receipt.













Kompella & Swallow          Standards Track                    [Page 21]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


3.3.  Downstream Mapping

  The Downstream Mapping object is a TLV that MAY be included in an
  echo request message.  Only one Downstream Mapping object may appear
  in an echo request.  The presence of a Downstream Mapping object is a
  request that Downstream Mapping objects be included in the echo
  reply.  If the replying router is the destination of the FEC, then a
  Downstream Mapping TLV SHOULD NOT be included in the echo reply.
  Otherwise the replying router SHOULD include a Downstream Mapping
  object for each interface over which this FEC could be forwarded.
  For a more precise definition of the notion of "downstream", see
  section 3.3.2, "Downstream Router and Interface".

  The Length is K + M + 4*N octets, where M is the Multipath Length,
  and N is the number of Downstream Labels.  Values for K are found in
  the description of Address Type below.  The Value field of a
  Downstream Mapping has the following format:

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |               MTU             | Address Type  |    DS Flags   |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |             Downstream IP Address (4 or 16 octets)            |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |         Downstream Interface Address (4 or 16 octets)         |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Multipath Type| Depth Limit   |        Multipath Length       |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     .                                                               .
     .                     (Multipath Information)                   .
     .                                                               .
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |               Downstream Label                |    Protocol   |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     .                                                               .
     .                                                               .
     .                                                               .
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |               Downstream Label                |    Protocol   |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

  Maximum Transmission Unit (MTU)

     The MTU is the size in octets of the largest MPLS frame (including
     label stack) that fits on the interface to the Downstream LSR.





Kompella & Swallow          Standards Track                    [Page 22]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  Address Type

     The Address Type indicates if the interface is numbered or
     unnumbered.  It also determines the length of the Downstream IP
     Address and Downstream Interface fields.  The resulting total for
     the initial part of the TLV is listed in the table below as "K
     Octets".  The Address Type is set to one of the following values:

        Type #        Address Type           K Octets
        ------        ------------           --------
             1        IPv4 Numbered                16
             2        IPv4 Unnumbered              16
             3        IPv6 Numbered                40
             4        IPv6 Unnumbered              28

  DS Flags

     The DS Flags field is a bit vector with the following format:

         0 1 2 3 4 5 6 7
        +-+-+-+-+-+-+-+-+
        | Rsvd(MBZ) |I|N|
        +-+-+-+-+-+-+-+-+

     Two flags are defined currently, I and N.  The remaining flags
     MUST be set to zero when sending and ignored on receipt.

     Flag  Name and Meaning
     ----  ----------------

        I  Interface and Label Stack Object Request

           When this flag is set, it indicates that the replying
           router SHOULD include an Interface and Label Stack
           Object in the echo reply message.

        N  Treat as a Non-IP Packet

           Echo request messages will be used to diagnose non-IP
           flows.  However, these messages are carried in IP
           packets.  For a router that alters its ECMP algorithm
           based on the FEC or deep packet examination, this flag
           requests that the router treat this as it would if the
           determination of an IP payload had failed.







Kompella & Swallow          Standards Track                    [Page 23]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  Downstream IP Address and Downstream Interface Address

     IPv4 addresses and interface indices are encoded in 4 octets; IPv6
     addresses are encoded in 16 octets.

     If the interface to the downstream LSR is numbered, then the
     Address Type MUST be set to IPv4 or IPv6, the Downstream IP
     Address MUST be set to either the downstream LSR's Router ID or
     the interface address of the downstream LSR, and the Downstream
     Interface Address MUST be set to the downstream LSR's interface
     address.

     If the interface to the downstream LSR is unnumbered, the Address
     Type MUST be IPv4 Unnumbered or IPv6 Unnumbered, the Downstream IP
     Address MUST be the downstream LSR's Router ID, and the Downstream
     Interface Address MUST be set to the index assigned by the
     upstream LSR to the interface.

     If an LSR does not know the IP address of its neighbor, then it
     MUST set the Address Type to either IPv4 Unnumbered or IPv6
     Unnumbered.  For IPv4, it must set the Downstream IP Address to
     127.0.0.1; for IPv6 the address is set to 0::1.  In both cases,
     the interface index MUST be set to 0.  If an LSR receives an Echo
     Request packet with either of these addresses in the Downstream IP
     Address field, this indicates that it MUST bypass interface
     verification but continue with label validation.

     If the originator of an Echo Request packet wishes to obtain
     Downstream Mapping information but does not know the expected
     label stack, then it SHOULD set the Address Type to either IPv4
     Unnumbered or IPv6 Unnumbered.  For IPv4, it MUST set the
     Downstream IP Address to 224.0.0.2; for IPv6 the address MUST be
     set to FF02::2.  In both cases, the interface index MUST be set to
     0.  If an LSR receives an Echo Request packet with the all-routers
     multicast address, then this indicates that it MUST bypass both
     interface and label stack validation, but return Downstream
     Mapping TLVs using the information provided.














Kompella & Swallow          Standards Track                    [Page 24]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  Multipath Type

     The following Multipath Types are defined:

     Key   Type                  Multipath Information
     ---   ----------------      ---------------------
      0    no multipath          Empty (Multipath Length = 0)
      2    IP address            IP addresses
      4    IP address range      low/high address pairs
      8    Bit-masked IP         IP address prefix and bit mask
             address set
      9    Bit-masked label set  Label prefix and bit mask

     Type 0 indicates that all packets will be forwarded out this one
     interface.

     Types 2, 4, 8, and 9 specify that the supplied Multipath
     Information will serve to exercise this path.

  Depth Limit

     The Depth Limit is applicable only to a label stack and is the
     maximum number of labels considered in the hash; this SHOULD be
     set to zero if unspecified or unlimited.

  Multipath Length

     The length in octets of the Multipath Information.

  Multipath Information

     Address or label values encoded according to the Multipath Type.
     See the next section below for encoding details.

  Downstream Label(s)

     The set of labels in the label stack as it would have appeared if
     this router were forwarding the packet through this interface.
     Any Implicit Null labels are explicitly included.  Labels are
     treated as numbers, i.e., they are right justified in the field.

     A Downstream Label is 24 bits, in the same format as an MPLS label
     minus the TTL field, i.e., the MSBit of the label is bit 0, the
     LSBit is bit 19, the EXP bits are bits 20-22, and bit 23 is the S
     bit.  The replying router SHOULD fill in the EXP and S bits; the
     LSR receiving the echo reply MAY choose to ignore these bits.





Kompella & Swallow          Standards Track                    [Page 25]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  Protocol

     The Protocol is taken from the following table:

     Protocol #        Signaling Protocol
     ----------        ------------------
              0        Unknown
              1        Static
              2        BGP
              3        LDP
              4        RSVP-TE

3.3.1.  Multipath Information Encoding

  The Multipath Information encodes labels or addresses that will
  exercise this path.  The Multipath Information depends on the
  Multipath Type.  The contents of the field are shown in the table
  above.  IPv4 addresses are drawn from the range 127/8; IPv6 addresses
  are drawn from the range 0:0:0:0:0:FFFF:127/104.  Labels are treated
  as numbers, i.e., they are right justified in the field.  For Type 4,
  ranges indicated by Address pairs MUST NOT overlap and MUST be in
  ascending sequence.

  Type 8 allows a more dense encoding of IP addresses.  The IP prefix
  is formatted as a base IP address with the non-prefix low-order bits
  set to zero.  The maximum prefix length is 27.  Following the prefix
  is a mask of length 2^(32-prefix length) bits for IPv4 and 2^(128-
  prefix length) bits for IPv6.  Each bit set to 1 represents a valid
  address.  The address is the base IPv4 address plus the position of
  the bit in the mask where the bits are numbered left to right
  beginning with zero.  For example, the IPv4 addresses 127.2.1.0,
  127.2.1.5-127.2.1.15, and 127.2.1.20-127.2.1.29 would be encoded as
  follows:

   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
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
  |0 1 1 1 1 1 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0|
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
  |1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 0 0|
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+










Kompella & Swallow          Standards Track                    [Page 26]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  Those same addresses embedded in IPv6 would be encoded as follows:

   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
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
  |0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0|
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
  |0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0|
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
  |0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0|
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
  |0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1|
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
  |0 1 1 1 1 1 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0|
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
  |1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 0 0|
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

  Type 9 allows a more dense encoding of labels.  The label prefix is
  formatted as a base label value with the non-prefix low-order bits
  set to zero.  The maximum prefix (including leading zeros due to
  encoding) length is 27.  Following the prefix is a mask of length
  2^(32-prefix length) bits.  Each bit set to one represents a valid
  label.  The label is the base label plus the position of the bit in
  the mask where the bits are numbered left to right beginning with
  zero.  Label values of all the odd numbers between 1152 and 1279
  would be encoded as follows:

   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 +-
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |0 0
  0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0| +-+-+-
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |0 1 0 1
  0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1| +-+-+-+-+-
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |0 1 0 1 0 1
  0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1| +-+-+-+-+-+-+-
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |0 1 0 1 0 1 0 1
  0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1| +-+-+-+-+-+-+-+-+-
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |0 1 0 1 0 1 0 1 0 1
  0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1| +-+-+-+-+-+-+-+-+-
  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

  If the received Multipath Information is non-null, the labels and IP
  addresses MUST be picked from the set provided.  If none of these
  labels or addresses map to a particular downstream interface, then
  for that interface, the type MUST be set to 0.  If the received
  Multipath Information is null (i.e., Multipath Length = 0, or for
  Types 8 and 9, a mask of all zeros), the type MUST be set to 0.



Kompella & Swallow          Standards Track                    [Page 27]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  For example, suppose LSR X at hop 10 has two downstream LSRs, Y and
  Z, for the FEC in question.  The received X could return Multipath
  Type 4, with low/high IP addresses of 127.1.1.1->127.1.1.255 for
  downstream LSR Y and 127.2.1.1->127.2.1.255 for downstream LSR Z.
  The head end reflects this information to LSR Y.  Y, which has three
  downstream LSRs, U, V, and W, computes that 127.1.1.1->127.1.1.127
  would go to U and 127.1.1.128-> 127.1.1.255 would go to V.  Y would
  then respond with 3 Downstream Mappings: to U, with Multipath Type 4
  (127.1.1.1->127.1.1.127); to V, with Multipath Type 4
  (127.1.1.127->127.1.1.255); and to W, with Multipath Type 0.

  Note that computing Multipath Information may impose a significant
  processing burden on the receiver.  A receiver MAY thus choose to
  process a subset of the received prefixes.  The sender, on receiving
  a reply to a Downstream Mapping with partial information, SHOULD
  assume that the prefixes missing in the reply were skipped by the
  receiver, and MAY re-request information about them in a new echo
  request.

3.3.2.  Downstream Router and Interface

  The notion of "downstream router" and "downstream interface" should
  be explained.  Consider an LSR X.  If a packet that was originated
  with TTL n>1 arrived with outermost label L and TTL=1 at LSR X, X
  must be able to compute which LSRs could receive the packet if it was
  originated with TTL=n+1, over which interface the request would
  arrive and what label stack those LSRs would see.  (It is outside the
  scope of this document to specify how this computation is done.)  The
  set of these LSRs/interfaces consists of the downstream
  routers/interfaces (and their corresponding labels) for X with
  respect to L.  Each pair of downstream router and interface requires
  a separate Downstream Mapping to be added to the reply.

  The case where X is the LSR originating the echo request is a special
  case.  X needs to figure out what LSRs would receive the MPLS echo
  request for a given FEC Stack that X originates with TTL=1.

  The set of downstream routers at X may be alternative paths (see the
  discussion below on ECMP) or simultaneous paths (e.g., for MPLS
  multicast).  In the former case, the Multipath Information is used as
  a hint to the sender as to how it may influence the choice of these
  alternatives.









Kompella & Swallow          Standards Track                    [Page 28]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


3.4.  Pad TLV

  The value part of the Pad TLV contains a variable number (>= 1) of
  octets.  The first octet takes values from the following table; all
  the other octets (if any) are ignored.  The receiver SHOULD verify
  that the TLV is received in its entirety, but otherwise ignores the
  contents of this TLV, apart from the first octet.

     Value        Meaning
     -----        -------
         1        Drop Pad TLV from reply
         2        Copy Pad TLV to reply
     3-255        Reserved for future use

3.5.  Vendor Enterprise Number

  SMI Private Enterprise Numbers are maintained by IANA.  The Length is
  always 4; the value is the SMI Private Enterprise code, in network
  octet order, of the vendor with a Vendor Private extension to any of
  the fields in the fixed part of the message, in which case this TLV
  MUST be present.  If none of the fields in the fixed part of the
  message have Vendor Private extensions, inclusion of this TLV is
  OPTIONAL.  Vendor Private ranges for Message Types, Reply Modes, and
  Return Codes have been defined.  When any of these are used, the
  Vendor Enterprise Number TLV MUST be included in the message.

3.6.  Interface and Label Stack

  The Interface and Label Stack TLV MAY be included in a reply message
  to report the interface on which the request message was received and
  the label stack that was on the packet when it was received.  Only
  one such object may appear.  The purpose of the object is to allow
  the upstream router to obtain the exact interface and label stack
  information as it appears at the replying LSR.

















Kompella & Swallow          Standards Track                    [Page 29]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  The Length is K + 4*N octets; N is the number of labels in the label
  stack.  Values for K are found in the description of Address Type
  below.  The Value field of a Downstream Mapping has the following
  format:

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Address Type  |             Must Be Zero                      |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                   IP Address (4 or 16 octets)                 |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                   Interface (4 or 16 octets)                  |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     .                                                               .
     .                                                               .
     .                          Label Stack                          .
     .                                                               .
     .                                                               .
     .                                                               .
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

  Address Type

     The Address Type indicates if the interface is numbered or
     unnumbered.  It also determines the length of the IP Address and
     Interface fields.  The resulting total for the initial part of the
     TLV is listed in the table below as "K Octets".  The Address Type
     is set to one of the following values:

        Type #        Address Type           K Octets
        ------        ------------           --------
             1        IPv4 Numbered                12
             2        IPv4 Unnumbered              12
             3        IPv6 Numbered                36
             4        IPv6 Unnumbered              24

  IP Address and Interface

     IPv4 addresses and interface indices are encoded in 4 octets; IPv6
     addresses are encoded in 16 octets.

     If the interface upon which the echo request message was received
     is numbered, then the Address Type MUST be set to IPv4 or IPv6,
     the IP Address MUST be set to either the LSR's Router ID or the
     interface address, and the Interface MUST be set to the interface
     address.




Kompella & Swallow          Standards Track                    [Page 30]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


     If the interface is unnumbered, the Address Type MUST be either
     IPv4 Unnumbered or IPv6 Unnumbered, the IP Address MUST be the
     LSR's Router ID, and the Interface MUST be set to the index
     assigned to the interface.

  Label Stack

     The label stack of the received echo request message.  If any TTL
     values have been changed by this router, they SHOULD be restored.

3.7.  Errored TLVs

  The following TLV is a TLV that MAY be included in an echo reply to
  inform the sender of an echo request of mandatory TLVs either not
  supported by an implementation or parsed and found to be in error.

  The Value field contains the TLVs that were not understood, encoded
  as sub-TLVs.

      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 = 9          |            Length             |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                             Value                             |
     .                                                               .
     .                                                               .
     .                                                               .
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

3.8.  Reply TOS Byte TLV

  This TLV MAY be used by the originator of the echo request to request
  that an echo reply be sent with the IP header TOS byte set to the
  value specified in the TLV.  This TLV has a length of 4 with the
  following value field.

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Reply-TOS Byte|                 Must Be Zero                  |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+








Kompella & Swallow          Standards Track                    [Page 31]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


4.  Theory of Operation

  An MPLS echo request is used to test a particular LSP.  The LSP to be
  tested is identified by the "FEC Stack"; for example, if the LSP was
  set up via LDP, and is to an egress IP address of 10.1.1.1, the FEC
  Stack contains a single element, namely, an LDP IPv4 prefix sub-TLV
  with value 10.1.1.1/32.  If the LSP being tested is an RSVP LSP, the
  FEC Stack consists of a single element that captures the RSVP Session
  and Sender Template that uniquely identifies the LSP.

  FEC Stacks can be more complex.  For example, one may wish to test a
  VPN IPv4 prefix of 10.1/8 that is tunneled over an LDP LSP with
  egress 10.10.1.1.  The FEC Stack would then contain two sub-TLVs, the
  bottom being a VPN IPv4 prefix, and the top being an LDP IPv4 prefix.
  If the underlying (LDP) tunnel were not known, or was considered
  irrelevant, the FEC Stack could be a single element with just the VPN
  IPv4 sub-TLV.

  When an MPLS echo request is received, the receiver is expected to
  verify that the control plane and data plane are both healthy (for
  the FEC Stack being pinged) and that the two planes are in sync.  The
  procedures for this are in section 4.4 below.

4.1.  Dealing with Equal-Cost Multi-Path (ECMP)

  LSPs need not be simple point-to-point tunnels.  Frequently, a single
  LSP may originate at several ingresses, and terminate at several
  egresses; this is very common with LDP LSPs.  LSPs for a given FEC
  may also have multiple "next hops" at transit LSRs.  At an ingress,
  there may also be several different LSPs to choose from to get to the
  desired endpoint.  Finally, LSPs may have backup paths, detour paths,
  and other alternative paths to take should the primary LSP go down.

  To deal with the last two first: it is assumed that the LSR sourcing
  MPLS echo requests can force the echo request into any desired LSP,
  so choosing among multiple LSPs at the ingress is not an issue.  The
  problem of probing the various flavors of backup paths that will
  typically not be used for forwarding data unless the primary LSP is
  down will not be addressed here.

  Since the actual LSP and path that a given packet may take may not be
  known a priori, it is useful if MPLS echo requests can exercise all
  possible paths.  This, although desirable, may not be practical,
  because the algorithms that a given LSR uses to distribute packets
  over alternative paths may be proprietary.

  To achieve some degree of coverage of alternate paths, there is a
  certain latitude in choosing the destination IP address and source



Kompella & Swallow          Standards Track                    [Page 32]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  UDP port for an MPLS echo request.  This is clearly not sufficient;
  in the case of traceroute, more latitude is offered by means of the
  Multipath Information of the Downstream Mapping TLV.  This is used as
  follows.  An ingress LSR periodically sends an MPLS traceroute
  message to determine whether there are multipaths for a given LSP.
  If so, each hop will provide some information how each of its
  downstream paths can be exercised.  The ingress can then send MPLS
  echo requests that exercise these paths.  If several transit LSRs
  have ECMP, the ingress may attempt to compose these to exercise all
  possible paths.  However, full coverage may not be possible.

4.2.  Testing LSPs That Are Used to Carry MPLS Payloads

  To detect certain LSP breakages, it may be necessary to encapsulate
  an MPLS echo request packet with at least one additional label when
  testing LSPs that are used to carry MPLS payloads (such as LSPs used
  to carry L2VPN and L3VPN traffic.  For example, when testing LDP or
  RSVP-TE LSPs, just sending an MPLS echo request packet may not detect
  instances where the router immediately upstream of the destination of
  the LSP ping may forward the MPLS echo request successfully over an
  interface not configured to carry MPLS payloads because of the use of
  penultimate hop popping.  Since the receiving router has no means to
  differentiate whether the IP packet was sent unlabeled or implicitly
  labeled, the addition of labels shimmed above the MPLS echo request
  (using the Nil FEC) will prevent a router from forwarding such a
  packet out unlabeled interfaces.

4.3.  Sending an MPLS Echo Request

  An MPLS echo request is a UDP packet.  The IP header is set as
  follows: the source IP address is a routable address of the sender;
  the destination IP address is a (randomly chosen) IPv4 address from
  the range 127/8 or IPv6 address from the range
  0:0:0:0:0:FFFF:127/104.  The IP TTL is set to 1.  The source UDP port
  is chosen by the sender; the destination UDP port is set to 3503
  (assigned by IANA for MPLS echo requests).  The Router Alert option
  MUST be set in the IP header.

  An MPLS echo request is sent with a label stack corresponding to the
  FEC Stack being tested.  Note that further labels could be applied
  if, for example, the normal route to the topmost FEC in the stack is
  via a Traffic Engineered Tunnel [RSVP-TE].  If all of the FECs in the
  stack correspond to Implicit Null labels, the MPLS echo request is
  considered unlabeled even if further labels will be applied in
  sending the packet.

  If the echo request is labeled, one MAY (depending on what is being
  pinged) set the TTL of the innermost label to 1, to prevent the ping



Kompella & Swallow          Standards Track                    [Page 33]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  request going farther than it should.  Examples of where this SHOULD
  be done include pinging a VPN IPv4 or IPv6 prefix, an L2 VPN endpoint
  or a pseudowire.  Preventing the ping request from going too far can
  also be accomplished by inserting a Router Alert label above this
  label; however, this may lead to the undesired side effect that MPLS
  echo requests take a different data path than actual data.  For more
  information on how these mechanisms can be used for pseudowire
  connectivity verification, see [VCCV].

  In "ping" mode (end-to-end connectivity check), the TTL in the
  outermost label is set to 255.  In "traceroute" mode (fault isolation
  mode), the TTL is set successively to 1, 2, and so on.

  The sender chooses a Sender's Handle and a Sequence Number.  When
  sending subsequent MPLS echo requests, the sender SHOULD increment
  the Sequence Number by 1.  However, a sender MAY choose to send a
  group of echo requests with the same Sequence Number to improve the
  chance of arrival of at least one packet with that Sequence Number.

  The TimeStamp Sent is set to the time-of-day (in seconds and
  microseconds) that the echo request is sent.  The TimeStamp Received
  is set to zero.

  An MPLS echo request MUST have an FEC Stack TLV.  Also, the Reply
  Mode must be set to the desired reply mode; the Return Code and
  Subcode are set to zero.  In the "traceroute" mode, the echo request
  SHOULD include a Downstream Mapping TLV.

4.4.  Receiving an MPLS Echo Request

  Sending an MPLS echo request to the control plane is triggered by one
  of the following packet processing exceptions: Router Alert option,
  IP TTL expiration, MPLS TTL expiration, MPLS Router Alert label, or
  the destination address in the 127/8 address range.  The control
  plane further identifies it by UDP destination port 3503.

  For reporting purposes the bottom of stack is considered to be
  stack-depth of 1.  This is to establish an absolute reference for the
  case where the actual stack may have more labels than there are FECs
  in the Target FEC Stack.

  Furthermore, in all the error codes listed in this document, a
  stack-depth of 0 means "no value specified".  This allows
  compatibility with existing implementations that do not use the
  Return Subcode field.

  An LSR X that receives an MPLS echo request then processes it as
  follows.



Kompella & Swallow          Standards Track                    [Page 34]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  1. General packet sanity is verified.  If the packet is not well-
     formed, LSR X SHOULD send an MPLS Echo Reply with the Return Code
     set to "Malformed echo request received" and the Subcode to zero.
     If there are any TLVs not marked as "Ignore" that LSR X does not
     understand, LSR X SHOULD send an MPLS "TLV not understood" (as
     appropriate), and the Subcode set to zero.  In the latter case,
     the misunderstood TLVs (only) are included as sub-TLVs in an
     Errored TLVs TLV in the reply.  The header fields Sender's Handle,
     Sequence Number, and Timestamp Sent are not examined, but are
     included in the MPLS echo reply message.

  The algorithm uses the following variables and identifiers:


  Interface-I:       the interface on which the MPLS echo request was
                     received.

  Stack-R:           the label stack on the packet as it was received.

  Stack-D:           the label stack carried in the Downstream Mapping
                     TLV (not always present)

  Label-L:           the label from the actual stack currently being
                     examined.  Requires no initialization.

  Label-stack-depth: the depth of label being verified.  Initialized to
                     the number of labels in the received label stack
                     S.

  FEC-stack-depth:   depth of the FEC in the Target FEC Stack that
                     should be used to verify the current actual label.
                     Requires no initialization.

  Best-return-code:  contains the return code for the echo reply packet
                     as currently best known.  As algorithm progresses,
                     this code may change depending on the results of
                     further checks that it performs.

  Best-rtn-subcode:  similar to Best-return-code, but for the Echo
                     Reply Subcode.

  FEC-status:        result value returned by the FEC Checking
                     algorithm described in section 4.4.1.

  /* Save receive context information */






Kompella & Swallow          Standards Track                    [Page 35]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  2. If the echo request is good, LSR X stores the interface over
     which the echo was received in Interface-I, and the label stack
     with which it came in Stack-R.

  /* The rest of the algorithm iterates over the labels in Stack-R,
     verifies validity of label values, reports associated label
     switching operations (for traceroute), verifies correspondence
     between the Stack-R and the Target FEC Stack description in the
     body of the echo request, and reports any errors. */

  /* The algorithm iterates as follows. */

  3. Label Validation:

     If Label-stack-depth is 0 {

     /* The LSR needs to report its being a tail-end for the LSP */

        Set FEC-stack-depth to 1, set Label-L to 3 (Implicit Null).
        Set Best-return-code to 3 ("Replying router is an egress for
        the FEC at stack depth"), set Best-rtn-subcode to the
        value of FEC-stack-depth (1) and go to step 5 (Egress
        Processing).
     }

     /* This step assumes there is always an entry for well-known
        label values */

     Set Label-L to the value extracted from Stack-R at depth
     Label-stack-depth.  Look up Label-L in the Incoming Label Map
     (ILM) to determine if the label has been allocated and an
     operation is associated with it.

     If there is no entry for L {

     /* Indicates a temporary or permanent label synchronization
        problem the LSR needs to report an error */

        Set Best-return-code to 11 ("No label entry at stack-depth")
        and Best-rtn-subcode to Label-stack-depth.  Go to step 7
        (Send Reply Packet).
     }

     Else {

        Retrieve the associated label operation from the
        corresponding NLFE and proceed to step 4 (Label Operation
        check).



Kompella & Swallow          Standards Track                    [Page 36]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


     }

  4. Label Operation Check

     If the label operation is "Pop and Continue Processing" {

     /* Includes Explicit Null and Router Alert label cases */

        Iterate to the next label by decrementing Label-stack-depth
        and loop back to step 3 (Label Validation).
     }

     If the label operation is "Swap or Pop and Switch based on Popped
        Label" {

        Set Best-return-code to 8 ("Label switched at stack-depth")
        and Best-rtn-subcode to Label-stack-depth to report transit
        switching.

        If a Downstream Mapping TLV is present in the received echo
        request {

           If the IP address in the TLV is 127.0.0.1 or 0::1 {
              Set Best-return-code to 6 ("Upstream Interface Index
              Unknown").  An Interface and Label Stack TLV SHOULD be
              included in the reply and filled with Interface-I and
              Stack-R.
           }

           Else {

              Verify that the IP address, interface address, and label
              stack in the Downstream Mapping TLV match Interface-I
              and Stack-R.  If there is a mismatch, set
              Best-return-code to 5, "Downstream Mapping Mismatch".
              An Interface and Label Stack TLV SHOULD be included in
              the reply and filled in based on Interface-I and
              Stack-R.  Go to step 7 (Send Reply Packet).
           }
        }


        For each available downstream ECMP path {

           Retrieve output interface from the NHLFE entry.

           /* Note: this return code is set even if Label-stack-depth
              is one */



Kompella & Swallow          Standards Track                    [Page 37]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


           If the output interface is not MPLS enabled {

              Set Best-return-code to Return Code 9, "Label switched
              but no MPLS forwarding at stack-depth" and set
              Best-rtn-subcode to Label-stack-depth and goto
              Send_Reply_Packet.
           }

           If a Downstream Mapping TLV is present {

             A Downstream Mapping TLV SHOULD be included in the echo
             reply (see section 3.3) filled in with information about
             the current ECMP path.
           }
        }

        If no Downstream Mapping TLV is present, or the Downstream IP
           Address is set to the ALLROUTERS multicast address,
              go to step 7 (Send Reply Packet).

        If the "Validate FEC Stack" flag is not set and the LSR is not
        configured to perform FEC checking by default, go to step 7
        (Send Reply Packet).

     /* Validate the Target FEC Stack in the received echo request.

        First determine FEC-stack-depth from the Downstream Mapping
        TLV.  This is done by walking through Stack-D (the Downstream
        labels) from the bottom, decrementing the number of labels
        for each non-Implicit Null label, while incrementing
        FEC-stack-depth for each label.  If the Downstream Mapping TLV
        contains one or more Implicit Null labels, FEC-stack-depth
        may be greater than Label-stack-depth.  To be consistent with
        the above stack-depths, the bottom is considered to entry 1.
        */

        Set FEC-stack-depth to 0.  Set i to Label-stack-depth.

        While (i > 0 ) do {
           ++FEC-stack-depth.
           if Stack-D[FEC-stack-depth] != 3 (Implicit Null)
              --i.
        }

        If the number of labels in the FEC stack is greater
           than or equal to FEC-stack-depth {





Kompella & Swallow          Standards Track                    [Page 38]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


           Perform the FEC Checking procedure (see subsection 4.4.1
           below).

           If FEC-status is 2, set Best-return-code to 10 ("Mapping
           for this FEC is not the given label at stack-depth").

           If the return code is 1, set Best-return-code to
           FEC-return-code and Best-rtn-subcode to FEC-stack-depth.
        }

        Go to step 7 (Send Reply Packet).
     }

  5. Egress Processing:

     /* These steps are performed by the LSR that identified itself
        as the tail-end LSR for an LSP. */

     If received echo request contains no Downstream Mapping TLV, or
        the Downstream IP Address is set to 127.0.0.1 or 0::1
           go to step 6 (Egress FEC Validation).

     Verify that the IP address, interface address, and label stack in
     the Downstream Mapping TLV match Interface-I and Stack-R.  If
     not, set Best-return-code to 5, "Downstream Mapping
     Mis-match".  A Received Interface and Label Stack TLV SHOULD be
     created for the echo response packet.  Go to step 7 (Send Reply
     Packet).

  6. Egress FEC Validation:

     /* This is a loop for all entries in the Target FEC Stack
        starting with FEC-stack-depth. */

     Perform FEC checking by following the algorithm described in
     subsection 4.4.1 for Label-L and the FEC at FEC-stack-depth.

     Set Best-return-code to FEC-code and Best-rtn-subcode to the
     value in FEC-stack-depth.

     If FEC-status (the result of the check) is 1,
        go to step 7 (Send Reply Packet).

     /* Iterate to the next FEC entry */

     ++FEC-stack-depth.





Kompella & Swallow          Standards Track                    [Page 39]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


     If FEC-stack-depth > the number of FECs in the FEC-stack,
        go to step 7 (Send Reply Packet).

     If FEC-status is 0 {
        ++Label-stack-depth.
        If Label-stack-depth > the number of labels in Stack-R,
           Go to step 7 (Send Reply Packet).

        Label-L = extracted label from Stack-R at depth
           Label-stack-depth.
        Loop back to step 6 (Egress FEC Validation).
     }

  7. Send Reply Packet:

     Send an MPLS echo reply with a Return Code of Best-return-code,
     and a Return Subcode of Best-rtn-subcode.  Include any TLVs
     created during the above process.  The procedures for sending
     the echo reply are found in subsection 4.4.1.

4.4.1.  FEC Validation

  /* This subsection describes validation of an FEC entry within the
     Target FEC Stack and accepts an FEC, Label-L, and Interface-I.
     The algorithm performs the following steps. */

  1. Two return values, FEC-status and FEC-return-code, are initialized
     to 0.

  2. If the FEC is the Nil FEC {
        If Label-L is either Explicit_Null or Router_Alert, return.

        Else {
           Set FEC-return-code to 10 ("Mapping for this FEC is not
           the given label at stack-depth").
           Set FEC-status to 1
           Return.
        }
     }

  3. Check the FEC label mapping that describes how traffic received on
     the LSP is further switched or which application it is associated
     with.  If no mapping exists, set FEC-return-code to Return 4,
     "Replying router has no mapping for the FEC at stack-depth".  Set
     FEC-status to 1.  Return.

  4. If the label mapping for FEC is Implicit Null, set FEC-status to 2
     and proceed to step 5.  Otherwise, if the label mapping for FEC is



Kompella & Swallow          Standards Track                    [Page 40]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


     Label-L, proceed to step 5.  Otherwise, set FEC-return-code to 10
     ("Mapping for this FEC is not the given label at stack-depth"),
     set FEC-status to 1, and return.

  5. This is a protocol check.  Check what protocol would be used to
     advertise FEC.  If it can be determined that no protocol
     associated with Interface-I would have advertised an FEC of that
     FEC-Type, set FEC-return-code to 12 ("Protocol not associated with
     interface at FEC stack-depth").  Set FEC-status to 1.

  6. Return.

4.5.  Sending an MPLS Echo Reply

  An MPLS echo reply is a UDP packet.  It MUST ONLY be sent in response
  to an MPLS echo request.  The source IP address is a routable address
  of the replier; the source port is the well-known UDP port for LSP
  ping.  The destination IP address and UDP port are copied from the
  source IP address and UDP port of the echo request.  The IP TTL is
  set to 255.  If the Reply Mode in the echo request is "Reply via an
  IPv4 UDP packet with Router Alert", then the IP header MUST contain
  the Router Alert IP option.  If the reply is sent over an LSP, the
  topmost label MUST in this case be the Router Alert label (1) (see
  [LABEL-STACK]).

  The format of the echo reply is the same as the echo request.  The
  Sender's Handle, the Sequence Number, and TimeStamp Sent are copied
  from the echo request; the TimeStamp Received is set to the time-of-
  day that the echo request is received (note that this information is
  most useful if the time-of-day clocks on the requester and the
  replier are synchronized).  The FEC Stack TLV from the echo request
  MAY be copied to the reply.

  The replier MUST fill in the Return Code and Subcode, as determined
  in the previous subsection.

  If the echo request contains a Pad TLV, the replier MUST interpret
  the first octet for instructions regarding how to reply.

  If the replying router is the destination of the FEC, then Downstream
  Mapping TLVs SHOULD NOT be included in the echo reply.

  If the echo request contains a Downstream Mapping TLV, and the
  replying router is not the destination of the FEC, the replier SHOULD
  compute its downstream routers and corresponding labels for the
  incoming label, and add Downstream Mapping TLVs for each one to the
  echo reply it sends back.




Kompella & Swallow          Standards Track                    [Page 41]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  If the Downstream Mapping TLV contains Multipath Information
  requiring more processing than the receiving router is willing to
  perform, the responding router MAY choose to respond with only a
  subset of multipaths contained in the echo request Downstream
  Mapping.  (Note: The originator of the echo request MAY send another
  echo request with the Multipath Information that was not included in
  the reply.)

  Except in the case of Reply Mode 4, "Reply via application level
  control channel", echo replies are always sent in the context of the
  IP/MPLS network.

4.6.  Receiving an MPLS Echo Reply

  An LSR X should only receive an MPLS echo reply in response to an
  MPLS echo request that it sent.  Thus, on receipt of an MPLS echo
  reply, X should parse the packet to ensure that it is well-formed,
  then attempt to match up the echo reply with an echo request that it
  had previously sent, using the destination UDP port and the Sender's
  Handle.  If no match is found, then X jettisons the echo reply;
  otherwise, it checks the Sequence Number to see if it matches.

  If the echo reply contains Downstream Mappings, and X wishes to
  traceroute further, it SHOULD copy the Downstream Mapping(s) into its
  next echo request(s) (with TTL incremented by one).

4.7.  Issue with VPN IPv4 and IPv6 Prefixes

  Typically, an LSP ping for a VPN IPv4 prefix or VPN IPv6 prefix is
  sent with a label stack of depth greater than 1, with the innermost
  label having a TTL of 1.  This is to terminate the ping at the egress
  PE, before it gets sent to the customer device.  However, under
  certain circumstances, the label stack can shrink to a single label
  before the ping hits the egress PE; this will result in the ping
  terminating prematurely.  One such scenario is a multi-AS Carrier's
  Carrier VPN.

  To get around this problem, one approach is for the LSR that receives
  such a ping to realize that the ping terminated prematurely, and send
  back error code 13.  In that case, the initiating LSR can retry the
  ping after incrementing the TTL on the VPN label.  In this fashion,
  the ingress LSR will sequentially try TTL values until it finds one
  that allows the VPN ping to reach the egress PE.








Kompella & Swallow          Standards Track                    [Page 42]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


4.8.  Non-compliant Routers

  If the egress for the FEC Stack being pinged does not support MPLS
  ping, then no reply will be sent, resulting in possible "false
  negatives".  If in "traceroute" mode, a transit LSR does not support
  LSP ping, then no reply will be forthcoming from that LSR for some
  TTL, say, n.  The LSR originating the echo request SHOULD try sending
  the echo request with TTL=n+1, n+2, ..., n+k to probe LSRs further
  down the path.  In such a case, the echo request for TTL > n SHOULD
  be sent with Downstream Mapping TLV "Downstream IP Address" field set
  to the ALLROUTERs multicast address until a reply is received with a
  Downstream Mapping TLV.  The label stack MAY be omitted from the
  Downstream Mapping TLV.  Furthermore, the "Validate FEC Stack" flag
  SHOULD NOT be set until an echo reply packet with a Downstream
  Mapping TLV is received.

5.  References

5.1.  Normative References

  [BGP]          Rekhter, Y., Li, T., and S. Hares, "A Border Gateway
                 Protocol 4 (BGP-4)", RFC 4271, January 2006.

  [IANA]         Narten, T. and H. Alvestrand, "Guidelines for Writing
                 an IANA Considerations Section in RFCs", BCP 26, RFC
                 2434, October 1998.

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

  [LABEL-STACK]  Rosen, E., Tappan, D., Fedorkow, G., Rekhter, Y.,
                 Farinacci, D., Li, T., and A. Conta, "MPLS Label Stack
                 Encoding", RFC 3032, January 2001.

  [NTP]          Mills, D., "Simple Network Time Protocol (SNTP)
                 Version 4 for IPv4, IPv6 and OSI", RFC 2030, October
                 1996.

  [RFC1122]      Braden, R., "Requirements for Internet Hosts -
                 Communication Layers", STD 3, RFC 1122, October 1989.

  [RFC1812]      Baker, F., "Requirements for IP Version 4 Routers",
                 RFC 1812, June 1995.

  [RFC4026]      Andersson, L. and T. Madsen, "Provider Provisioned
                 Virtual Private Network (VPN) Terminology", RFC 4026,
                 March 2005.




Kompella & Swallow          Standards Track                    [Page 43]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


5.2.  Informative References

  [BGP-LABEL]    Rekhter, Y. and E. Rosen, "Carrying Label Information
                 in BGP-4", RFC 3107, May 2001.

  [ICMP]         Postel, J., "Internet Control Message Protocol", STD
                 5, RFC 792, September 1981.

  [LDP]          Andersson, L., Doolan, P., Feldman, N., Fredette, A.,
                 and B. Thomas, "LDP Specification", RFC 3036, January
                 2001.

  [PW-CONTROL]   Martini, L., El-Aawar, N., Heron, G., Rosen, E.,
                 Tappan, D., and  T. Smith, "Pseudowire Setup and
                 Maintenance using the Label Distribution Protocol",
                 Work in Progress.

  [RFC4365]      Rosen, E., "Applicability Statement for BGP/MPLS IP
                 Virtual Private Networks (VPNs)", RFC 4365, February
                 2006.

  [RSVP-TE]      Awduche, D., Berger, L., Gan, D., Li, T., Srinivasan,
                 V., and G. Swallow, "RSVP-TE: Extensions to RSVP for
                 LSP Tunnels", RFC 3209, December 2001.

  [VCCV]         Nadeau, T. and R. Aggarwal, "Pseudo Wire Virtual
                 Circuit Connectivity Verification (VCCV), Work in
                 Progress, August 2005.

  [VPLS-BGP]     Kompella, K. and Y. Rekhter, "Virtual Private LAN
                 Service", Work in Progress.

6.  Security Considerations

  Overall, the security needs for LSP ping are similar to those of ICMP
  ping.

  There are at least three approaches to attacking LSRs using the
  mechanisms defined here.  One is a Denial-of-Service attack, by
  sending MPLS echo requests/replies to LSRs and thereby increasing
  their workload.  The second is obfuscating the state of the MPLS data
  plane liveness by spoofing, hijacking, replaying, or otherwise
  tampering with MPLS echo requests and replies.  The third is an
  unauthorized source using an LSP ping to obtain information about the
  network.






Kompella & Swallow          Standards Track                    [Page 44]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  To avoid potential Denial-of-Service attacks, it is RECOMMENDED that
  implementations regulate the LSP ping traffic going to the control
  plane.  A rate limiter SHOULD be applied to the well-known UDP port
  defined below.

  Unsophisticated replay and spoofing attacks involving faking or
  replaying MPLS echo reply messages are unlikely to be effective.
  These replies would have to match the Sender's Handle and Sequence
  Number of an outstanding MPLS echo request message.  A non-matching
  replay would be discarded as the sequence has moved on, thus a spoof
  has only a small window of opportunity.  However, to provide a
  stronger defense, an implementation MAY also validate the TimeStamp
  Sent by requiring and exact match on this field.

  To protect against unauthorized sources using MPLS echo request
  messages to obtain network information, it is RECOMMENDED that
  implementations provide a means of checking the source addresses of
  MPLS echo request messages against an access list before accepting
  the message.

  It is not clear how to prevent hijacking (non-delivery) of echo
  requests or replies; however, if these messages are indeed hijacked,
  LSP ping will report that the data plane is not working as it should.

  It does not seem vital (at this point) to secure the data carried in
  MPLS echo requests and replies, although knowledge of the state of
  the MPLS data plane may be considered confidential by some.
  Implementations SHOULD, however, provide a means of filtering the
  addresses to which echo reply messages may be sent.

  Although this document makes special use of 127/8 address, these are
  used only in conjunction with the UDP port 3503.  Furthermore, these
  packets are only processed by routers.  All other hosts MUST treat
  all packets with a destination address in the range 127/8 in
  accordance to RFC 1122.  Any packet received by a router with a
  destination address in the range 127/8 without a destination UDP port
  of 3503 MUST be treated in accordance to RFC 1812.  In particular,
  the default behavior is to treat packets destined to a 127/8 address
  as "martians".












Kompella & Swallow          Standards Track                    [Page 45]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


7.  IANA Considerations

  The TCP and UDP port number 3503 has been allocated by IANA for LSP
  echo requests and replies.

  The following sections detail the new name spaces to be managed by
  IANA.  For each of these name spaces, the space is divided into
  assignment ranges; the following terms are used in describing the
  procedures by which IANA allocates values: "Standards Action" (as
  defined in [IANA]), "Specification Required", and "Vendor Private
  Use".

  Values from "Specification Required" ranges MUST be registered with
  IANA.  The request MUST be made via an Experimental RFC that
  describes the format and procedures for using the code point; the
  actual assignment is made during the IANA actions for the RFC.

  Values from "Vendor Private" ranges MUST NOT be registered with IANA;
  however, the message MUST contain an enterprise code as registered
  with the IANA SMI Private Network Management Private Enterprise
  Numbers.  For each name space that has a Vendor Private range, it
  must be specified where exactly the SMI Private Enterprise Number
  resides; see below for examples.  In this way, several enterprises
  (vendors) can use the same code point without fear of collision.

7.1.  Message Types, Reply Modes, Return Codes

  The IANA has created and will maintain registries for Message Types,
  Reply Modes, and Return Codes.  Each of these can take values in the
  range 0-255.  Assignments in the range 0-191 are via Standards
  Action; assignments in the range 192-251 are made via "Specification
  Required"; values in the range 252-255 are for Vendor Private Use,
  and MUST NOT be allocated.

  If any of these fields fall in the Vendor Private range, a top-level
  Vendor Enterprise Number TLV MUST be present in the message.

  Message Types defined in this document are the following:

     Value    Meaning
     -----    -------
         1    MPLS echo request
         2    MPLS echo reply








Kompella & Swallow          Standards Track                    [Page 46]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  Reply Modes defined in this document are the following:

     Value    Meaning
     -----    -------
         1    Do not reply
         2    Reply via an IPv4/IPv6 UDP packet
         3    Reply via an IPv4/IPv6 UDP packet with Router Alert
         4    Reply via application level control channel

  Return Codes defined in this document are listed in section 3.1.

7.2.  TLVs

  The IANA has created and will maintain a registry for the Type field
  of top-level TLVs as well as for any associated sub-TLVs.  Note the
  meaning of a sub-TLV is scoped by the TLV.  The number spaces for the
  sub-TLVs of various TLVs are independent.

  The valid range for TLVs and sub-TLVs is 0-65535.  Assignments in the
  range 0-16383 and 32768-49161 are made via Standards Action as
  defined in [IANA]; assignments in the range 16384-31743 and
  49162-64511 are made via "Specification Required" as defined above;
  values in the range 31744-32767 and 64512-65535 are for Vendor
  Private Use, and MUST NOT be allocated.

  If a TLV or sub-TLV has a Type that falls in the range for Vendor
  Private Use, the Length MUST be at least 4, and the first four octets
  MUST be that vendor's SMI Private Enterprise Number, in network octet
  order.  The rest of the Value field is private to the vendor.






















Kompella & Swallow          Standards Track                    [Page 47]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


  TLVs and sub-TLVs defined in this document are the following:

        Type       Sub-Type        Value Field
        ----       --------        -----------
           1                       Target FEC Stack
                        1          LDP IPv4 prefix
                        2          LDP IPv6 prefix
                        3          RSVP IPv4 LSP
                        4          RSVP IPv6 LSP
                        5          Not Assigned
                        6          VPN IPv4 prefix
                        7          VPN IPv6 prefix
                        8          L2 VPN endpoint
                        9          "FEC 128" Pseudowire (Deprecated)
                       10          "FEC 128" Pseudowire
                       11          "FEC 129" Pseudowire
                       12          BGP labeled IPv4 prefix
                       13          BGP labeled IPv6 prefix
                       14          Generic IPv4 prefix
                       15          Generic IPv6 prefix
                       16          Nil FEC
           2                       Downstream Mapping
           3                       Pad
           4                       Not Assigned
           5                       Vendor Enterprise Number
           6                       Not Assigned
           7                       Interface and Label Stack
           8                       Not Assigned
           9                       Errored TLVs
                   Any value       The TLV not understood
          10                       Reply TOS Byte

8.  Acknowledgements

  This document is the outcome of many discussions among many people,
  including Manoj Leelanivas, Paul Traina, Yakov Rekhter, Der-Hwa Gan,
  Brook Bailey, Eric Rosen, Ina Minei, Shivani Aggarwal, and Vanson
  Lim.

  The description of the Multipath Information sub-field of the
  Downstream Mapping TLV was adapted from text suggested by Curtis
  Villamizar.









Kompella & Swallow          Standards Track                    [Page 48]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


Authors' Addresses

  Kireeti Kompella
  Juniper Networks
  1194 N.Mathilda Ave
  Sunnyvale, CA 94089

  EMail:  [email protected]


  George Swallow
  Cisco Systems
  1414 Massachusetts Ave,
  Boxborough, MA 01719

  Phone:  +1 978 936 1398
  EMail:  [email protected]


































Kompella & Swallow          Standards Track                    [Page 49]

RFC 4379           Detecting MPLS Data Plane Failures      February 2006


Full Copyright Statement

  Copyright (C) The Internet Society (2006).

  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
  [email protected].

Acknowledgement

  Funding for the RFC Editor function is provided by the IETF
  Administrative Support Activity (IASA).







Kompella & Swallow          Standards Track                    [Page 50]