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

Versions: 00 01 02

Network Working Group                                     George Swallow
Internet Draft                                       Cisco Systems, Inc.
Category: Standards Track
Expiration Date: August 2005
                                                          Mickey Spiegel
                                                     Cisco Systems, Inc.

                                                           February 2005


    Soft Permanent Virtual Circuit Interworking between PWE3 and ATM


                   draft-swallow-pwe3-spvc-iw-02.txt

Status of this Memo

   By submitting this Internet-Draft, the authors certify that any
   applicable patent or other IPR claims of which we are aware have been
   disclosed, and any of which we become aware will be disclosed, in
   accordance with RFC 3668.

   This document is an Internet-Draft and is in full conformance with
   all provisions of Section 5 of RFC3667.  Internet-Drafts are working
   documents of the Internet Engineering Task Force (IETF), its areas,
   and its working groups.  Note that other groups may also distribute
   working documents as Internet-Drafts.

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

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

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

Copyright Notice

   Copyright (C) The Internet Society (2005).

Abstract


   This document defines interworking between Private Network Node
   Interface (PNNI) SPVC signaling and the Label Distribution Protocol



Swallow & Spiegel            Standards Track                    [Page 1]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


   [LDP] as extended by [PWE3-CP] and [L2VPN-SIG].




Contents

 1      Introduction  ..............................................   3
 1.1    Conventions  ...............................................   3
 1.2    Terminology  ...............................................   3
 2      Problem Statement  .........................................   4
 2.1    Reference Network Diagram  .................................   5
 3      Requirements  ..............................................   5
 3.1    Configuration  .............................................   6
 3.2    Redundant ATM/PSN Interfaces  ..............................   6
 3.3    Re-assembly  ...............................................   6
 3.4    No Change to Existing ATM Signaling Protocols  .............   6
 3.5    Frame Relay and ATM Circuits at the MPLS Network Edge  .....   6
 4      Non-Requirements  ..........................................   7
 4.1    Frame Relay / ATM Interworking  ............................   7
 5      Background on identifiers  .................................   7
 5.1    Pseudowire Identifiers  ....................................   7
 5.2    ATM SPVC Identifiers  ......................................   8
 6      Proposed Solution  .........................................   9
 6.1    PSN / ATM Interface  .......................................   9
 6.2    Signaling  .................................................   9
 6.3    Mapping Identifiers  .......................................  10
 6.3.1  Mapping Addresses  .........................................  10
 6.3.2  Mapping Port and SPVC IEs to AIs  ..........................  11
 6.4    Configuration  .............................................  12
 6.5    Procedures  ................................................  12
 6.5.1  Procedures within the ATM Network  .........................  12
 6.5.2  Procedures for the ME  .....................................  13
 6.5.3  Procedures for the PME  ....................................  13
 6.5.4  Call Completion  ...........................................  14
 6.6    Handling Re-assembly  ......................................  14
 7      Issues  ....................................................  14
 8      Security Considerations  ...................................  15
 9      Acknowledgments  ...........................................  15
10      References  ................................................  15
10.1    Normative References  ......................................  15
10.2    Informative References  ....................................  15
/Users/swallow/ietf/pwe3/pwe/atmpwe050220:679: warning: escape character ignored before `'
11      Authors' Addresses  ........................................  16
12      Full Copyright and Intellectual Property Statements  .......  16






Swallow & Spiegel            Standards Track                    [Page 2]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


1. Introduction

   In current ATM deployments, Soft Permanent Virtual Connections
   (SPVCs) are used to provision both Asynchronous Transfer Mode (ATM)
   Permanent Virtual Channel Connections (PVCC) and Permanent Virtual
   Path Connections (PVPC) and Frame Relay (FR) PVCCs.

   Pseudowires over Multiprotocol Label Switching (MPLS) and L2TPv3 PSNs
   are current being introduced as backbone technologies to carry these
   same services.  Mechanisms are being developed to enable a flexible
   provisioning model which incorporates elements of the SPVC model in
   that configuration of the end service exists only at the end-points.
   These mechanisms are described in [PWE3-CP] and [L2VPN-SIG].

   As services are migrated from ATM to PSNs, any reasonable deployment
   scenario mandates that there be a period of time (possibly protracted
   or permanent) in which services will need to be established and main-
   tained between end-users where one end of a circuit is attached to an
   ATM network and the other end is attached to a PSN.


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].


1.2. Terminology

   PAE     Provider ATM Edge, a customer facing node which is part of
           the ATM network

   AE      ATM Edge, a node of the ATM network which is attached to a
           node of the MPLS/IP network

   PSN     Packet Switched Network

   PME     Provider MPLS/IP Edge, a customer facing node which is part
           of the MPLS/IP PSN

   ME      MPLS/IP Edge, a node of the MPLS/IP PSN which is attached to
           a node of the ATM network

   AI      Attachment Identifier

   AGI     Attachment Group Identifier




Swallow & Spiegel            Standards Track                    [Page 3]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


   AII     Attachment Individual Identifier

   SAI     Source Attachment Identifier

   TAI     Target Attachment Identifier

   SAII    Source Attachment Individual Identifier

   TAII    Target Attachment Individual Identifier

   PNNI    Private Network-Network Interface

   UNI     User Network Interface

   AINI    ATM Inter-Network Interface

   PVCC    Permanent Virtual Channel Connection

   PVPC    Permanent Virtual Path Connection

   SPVC    Soft Permanent Virtual Connection

   IE      Information Element



2. Problem Statement

   To facilitate the migration of ATM and Frame Relay SPVCs to a PSN
   carrying pseudowires, a means of interoperating ATM and LDP signaling
   needs to be defined.  Further this interoperation must preserve the
   essential reasons for using SPVCs, namely, keeping configuration lim-
   ited to the edge nodes supporting a particular connection and allow-
   ing the network to be able to recover in the event of the failure of
   any facility between those two edge nodes.

   It is also useful to perform reassembly of AAL5 frames of Frame Relay
   connections at the boundary between the ATM network and the PSN.
   This serves to reduce dataplane protocol overhead in the PSN.

   Finally, any solution must not preclude any existing services.  In
   particular, Frame Relay to ATM interworking is recognized to be in
   wide use.








Swallow & Spiegel            Standards Track                    [Page 4]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


2.1. Reference Network Diagram

   The diagram below shows an ATM network interfaced to a PSN.  A soft
   PVC is to be set up between the two customer facing edge nodes, PAE,
   and PME.  Two interconnections are shown, to indicate that the con-
   nection must be routable over either interconnection in the event of
   the failure of the preferred interconnection.  The 'M' (for MPLS/IP
   PSN) was chosen over 'P' (for PSN) to allow 'P' to stand for
   provider.

               _______                          _______
              /       %                        /       %
             /         %   +-----+  +-----+   /         %
            /           %  |     |  |     |  /           %
            |           |==| AE1 |==| ME1 |==|           |
   +-----+  |           |  |     |  |     |  |           |  +-----+
   |     |  |    ATM    |  +-----+  +-----+  |  MPLS/IP  |  |     |
   | PAE |==|           |                    |    PSN    |==| PME |
   |     |  |           |  +-----+  +-----+  |           |  |     |
   +-----+  |           |  |     |  |     |  |           |  +-----+
            |           |==| AE2 |==| ME2 |==|           |
            %           /  |     |  |     |  %           /
             %         /   +-----+  +-----+   %         /
              %_______/                        %_______/

   Key:

       All node acronyms are composed of the following
       letters

         P - Provider
         M - MPLS/IP PSN
         A - ATM
         E - Edge

                     Figure 1: Reference Network



3. Requirements

   The purpose of Soft Permanent Virtual Connections is two-fold.  First
   they confine circuit specific configuration to the edge nodes where
   the user access circuits are terminated.  Second they allow the net-
   work to automatically reroute / re-establish the circuit when fail-
   ures are encountered.  The requirements for this solution follow
   directly from this as well as some of the common uses of SPVCs.




Swallow & Spiegel            Standards Track                    [Page 5]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


3.1. Configuration

   All per circuit configuration must be limited to the edge nodes.  In
   particular the solution must not necessitate any per circuit configu-
   ration on the nodes that support the ATM/PSN interface.


3.2. Redundant ATM/PSN Interfaces

   The solution must support multiple inter-connections between the ATM
   network and the PSN.  Upon failure of an ATM/PSN interface, it must
   be possible to re-route the SPVCs that had been traversing that
   interface over other ATM/PSN interfaces.


3.3. Re-assembly

   It must be possible to locate the AAL5 re-assembly at the ME.  That
   is, a ME by default will perform AAL5 re-assembly for Frame Relay
   SPVCs.  The ME's ATM/PSN interface may be configured to not perform
   re-assembly and leave the job to the target PME, when the target PME
   supports AAL5 re-assembly for Frame Relay circuits.  No per circuit
   selection need be supported.


3.4. No Change to Existing ATM Signaling Protocols

   It is highly desirable that no changes be required to existing ATM
   signaling protocols.


3.5. Frame Relay and ATM Circuits at the MPLS Network Edge

   The solution must support both ATM and Frame Relay circuits at the
   Provider MPLS Edge.  For the case of ATM at the PME and Frame Relay
   at the PAE, Frame Relay to ATM interworking is the responsibility of
   the ATM network.  For the case of Frame Relay at both the PME and the
   PAE, FRF5 or FRF8.1 Frame Relay to ATM interworking capability is
   required at the PME (to be specific, Frame Relay to AAL5 SDU Frame
   encapsulation over PW) and is optional at the ME (Frame Relay over
   Pseudo Wire to ATM).  For the case of Frame Relay at the PME and ATM
   at the PAE, Frame Relay to ATM interworking capability is required at
   the PME and is optional at the ME.








Swallow & Spiegel            Standards Track                    [Page 6]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


4. Non-Requirements

4.1. Frame Relay / ATM Interworking

   While Frame Relay to ATM interworking is recognized as an important
   and pervasive service, such a service is deemed to be beyond the
   scope of this document.  This is not to imply that such a service
   cannot be supported by the means specified herein.  It merely implies
   that when Frame Relay to ATM interworking is required in the PSN net-
   work, the interworking function (IWF) is located (and configured) at
   the PME.



5. Background on identifiers

5.1. Pseudowire Identifiers

   Pseudowires serve to connect a pair of attachment circuits (ACs).  In
   the context of this document these ACs are either Frame Relay DLCIs
   or ATM VPI/VCIs.  An AC is identified by an Attachment Identifier
   (AI) and the IP address of the egress node.  AIs are defined in [1].
   An AI is a logical reference for both the physical/logical port as
   well as the virtual circuit.  That is an AC is fully identified by a
   node-ID (IP address) and an AI.

   The AI has further structure to designate groups of identifiers and
   individual identifiers within a group, these are called attachment
   group identifiers (AGI) and attachment individual identifiers (AII),
   respectively.  When pairs of AIs are used in signaling, they are fur-
   ther designated by their role in the call, with the operative terms
   being source and target of the call.  Thus we also have the terminol-
   ogy, source attachment identifier (SAI), source attachment individual
   identifier (SAII), target attachment identifier (TAI), and target
   attachment individual identifier (TAII). The source and target desig-
   nations are only relevant from the perspective of the pseudowire con-
   trol protocol. For example, a node receiving an LDP label mapping
   message from a remote node will swap the SAI and TAI values when it
   sends a label mapping message in the reverse direction back to the
   originating node.

   Attachment Identifiers (AIs) are carried in the Generalized ID FEC
   Element of LDP.  Each AI encoded as two fields, the Attachment Group
   Identifier (AGI) and an Attachment Individual Identifier (AII), each
   encoded using a type-length-value (TLV) format as defined in Section
   5.2.2/[PWE3-CP].  In particular:

   "Note that the interpretation of a particular field as AGI, SAII, or



Swallow & Spiegel            Standards Track                    [Page 7]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


   TAII depends on the order of its occurrence.  The type field identi-
   fies the type of the AGI, SAII, or TAII.

   The rules for constructing the AGI and AII are left to the specifica-
   tion of applications and/or models."



5.2. ATM SPVC Identifiers

   In ATM, the identifying information of the attachment circuit at the
   destination interface consists of an ATM End-System Address (AESA)
   and the DLCI or VPI/VCI.  The AESA identifies both the destination
   node and port where the end-user is attached.  The DLCI or VPI/VCI
   are signaled in the Called party SPVC IE and are carried as literal
   values.  The Called party SPVC IE has two formats depending on
   whether the service being signaled is a Frame Relay SPVC or an ATM
   SPVC.  Furthermore, ATM SPVCCs and ATM SPVPCs are distinguished
   through the bearer class codepoint in the Broadband bearer capability
   IE.

   AESAs are based on the NSAP address format.  Figure 2 shows the
   generic format.


          |--AFI (Authority & Format Identifier)
          |                                   Selector
          |  |--IDP (Initial Domain Part)       |
          V  V                                  V
         +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
         | |   |    H O - D S P     |   E S I  |0|
         +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

         HO-DSP   High Order Domain Specific Part
         ESI      End System Identifier

              Figure 2:  Generic AESA Format


   Although many formats are permitted within AESAs, all ATM Forum
   defined formats include a six byte End System Identifier or ESI.  The
   ESI's role is to identify a host.  Typically, the first 13 bytes of
   the AESA are common for all end systems attached to an ATM node.
   This is the default behavior in PNNI.  In this case, the ESI is used
   to differentiate between end systems attached to the same switch.
   From the point of view of the egress ATM switch, the ESI maps to a
   physical or logical port.




Swallow & Spiegel            Standards Track                    [Page 8]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


   Thus common practice is to use the ESI to carry the port information.



6. Proposed Solution

6.1. PSN / ATM Interface

   The interface between the ATM network and the PSN can be any of the
   following:

       ATM Forum User Network Interface (UNI)
       ITU-T User Network Interface (UNI)
       Private Network-Network Interface (PNNI)
       ATM Inter-Network Interface (AINI)

   In the case of the UNI, there must be extensions to support the
   Called party soft PVPC or PVCC IE defined in [PNNI].  (In this docu-
   ment we refer to the Called party soft PVPC or PVCC IE as simply the
   SPVC IE.)  There may be extensions to support:

     o  the Calling party soft PVPC or PVCC IE,

     o  signaled VPs, using the "transparent VP service" codepoint for
        the bearer class in the Broadband bearer capability IE,

     o  crankback indication by setting the cause location in the Cause
        IE to a value other than "user", and

     o  frame discard indication using either the Frame Discard bits
        or the ATM adaptation layer parameters IE.

   [Note: while the details of various versions of ATM signaling and the
   support for particular IEs is a subject for other Fori and thus
   beyond the scope of the WG (and IETF), an informative appendix
   appropriate references will be added if the WG so desires.]


6.2. Signaling

   In ATM soft PVCs are statically defined across the UNI interfaces,
   but are signaled across the ATM network using PNNI signaling.  For
   the signaling part, one edge node is configured to be active for the
   SPVC and the other is defined to be passive.  That is, one end always
   initiates the call.  ATM signaling creates a bidirectional circuit in
   a single pass.  Bandwidth parameters for each direction of the cir-
   cuit are carried in the setup message.




Swallow & Spiegel            Standards Track                    [Page 9]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


   A paradigm analogous to the active/passive roles in SPVC setup above
   for pseudowires is known as single sided provisioning.  These proce-
   dures are defined in draft-rosen-ppvpn-l2-signaling-03.txt [L2VPN-
   SIG] section 5.1.1.2.  A small difference exists here in that the
   "discovery" process occurs when an ATM SETUP message arrives.

   It should be noted, that the pseudowire setup remains a pair of uni-
   directional labels assigned by two essentially independent label
   requests.  It is only the triggering of the reverse label request
   that is tied to the forward label request.  Further [PWE3-CP] does
   not carry bandwidth parameters at all.


6.3. Mapping Identifiers

   In [PWE3-CP], an IP address identifies the egress node, and the (T)AI
   identifies the egress port and DLCI or VPI/VCI.  In ATM, an AESA
   identifies both the egress node and port, and the DLCI or VPI/VCI is
   carried as a literal in the SPVC IE.

   Two issues must be addressed.  First a mapping between ATM and IP
   addresses is needed.  Second a means of translating between the ATM
   port and SPVC IE and the AI used in [PWE3-CP].


6.3.1. Mapping Addresses

   OSI Network Service Access Point Addresses include an Authority and
   Format Identifier (AFI). The AFI value 35 has been assigned to the
   IANA. Within this format a two-octet Internet Code Point (ICP) field
   is available for assignment by the IANA. Currently there is one code
   point, 0, assigned for embedding IPv6 addresses.  A format and code
   point assignment has been proposed by the ATM Forum in [ATM-ipaddr].
   That format is shown below.

















Swallow & Spiegel            Standards Track                   [Page 10]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


          |--AFI (Authority &
          |       Format Identifier)          Selector
          |  |--IDP (Initial Domain Part)       |
          V  V |<---- HO-DSP ----->|<-- ESI -->|V
         +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
         |3|0 0|I P v 4|0 0 0 0 0 0|           |0|
         |5|0 1|Address|0 0 0 0 0 0|           |0|
         +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

         HO-DSP   High Order Domain Specific Part
         ESI      End System Identifier

         Figure 3: NSAP with Embedded IPv4 Address

   While it is common practice to carry the port number in the ESI
   field, We note that there are six unused bytes in the HO-DSP as well
   as the Selector Byte which could be used in a situation where the ESI
   was not available.

   The format to embed an IPv6 address in an NSAP address is defined in
   [RFC1888].  In this format the only unused space is the Selector
   Byte.  This allows for the identification of 256 ports.  If more
   ports are needed, multiple addresses must be assigned.


          |--AFI (Authority &
          |       Format Identifier)          Selector
          |  |--IDP (Initial Domain Part)       |
          V  V                                  V
         +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
         |3|0 0|    I P v 6   A d d r e s s    | |
         |5|0 0|                               | |
         +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+


         Figure 4: NSAP with Embedded IPv6 Address

   While it is expected that for IPv4 the ESI will commonly be used and
   for IPv6 the Selector Byte must be used, the discussion below will
   simply refers to a generic port field.


6.3.2. Mapping Port and SPVC IEs to AIs

   It is proposed that the Port and SPVC IE values be mapped into the
   AII.  The SPVC IE has three formats, one each for Frame Relay, ATM
   SPVC, and ATM SPVP.  Each of these will be mapped into three [to be
   assigned] AII types.  [If there is an issue with IANA allocating



Swallow & Spiegel            Standards Track                   [Page 11]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


   three, one, with some form of sub-typing will do.]




6.4. Configuration

   PAEs: For each Permanent Virtual Connection, the PAE is configured
         with the target AESA and DLCI or VPI/VCI.

   AEs:  AEs are configured with the AESA prefix representing the set of
         PSN nodes reachable through its link(s) to the PSN.  Multiple
         prefixes may be configured to allow choices of optimum nodes to
         reach and to allow reachability to a larger set of nodes,
         should some other AE or ME fail.  The advertisement into the
         ATM network's routing protocol (e.g.PNNI) should be withdrawn
         if the associated link(s) have failed.

   PMEs: PME require no special configuration.  They are simply
         configured with the (T)AII of each of their ACs.

   MEs:  MEs must be able to encode and parse the [to be assigned] AII
         types.  Associated with each of these AII type is an AII format
         (used to form TAIIs) and rules for how to extract the port from
         the ATM called party address.  .fi


6.5. Procedures

6.5.1. Procedures within the ATM Network

   In an SPVC, one end is designated as the 'owner' and is responsible
   for initiating the connection.  In order to simplify the interwork-
   ing, this solution proposes that SPVCs always be initiated from the
   ATM side.  This obviates any need to communicate bandwidth informa-
   tion across the PSN to the ATM network.

   The AEs as the owner of the connection, initiates PNNI signaling as
   it normally would.  Finding a longest match associated with one or
   more of the AEs it performs normal PNNI routing selection and sends a
   SETUP message which includes the SPVC IE.

   When the SETUP message arrives at the AE, it performs normal PNNI
   signaling processing and forwards the message across the PNNI, AINI
   or UNI to the ME.






Swallow & Spiegel            Standards Track                   [Page 12]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


6.5.2. Procedures for the ME

   When the ME receives a SETUP message, performs ATM admission control.
   Additionally, the ME may perform additional checks to determine if it
   has the necessary resources to support the pseudowire connection in
   the forward direction.  For example, in some network deployments it
   may determine if a PSN tunnel can be established in order to satisfy
   QoS or restoration requirements.

   In the event that the call cannot be admitted, the ME SHOULD set an
   appropriate cause code, assuming that it is capable of supporting
   crankback procedures.

   When the ME has successfully performed ATM admission control, it
   splices the call to a pseudowire using the signaling procedures of
   [L2VPN-SIG].  First, it extracts the destination IP address from the
   ATM called party address.  It then determines if a LDP session exists
   with this node.  If not, one is initiated.  It then examines the SPVC
   IE to determine the type of service which is being requested.  Based
   on the service type it selects AII type and format.  It then extracts
   the port, VPI, VCI, and/or DLCI information as appropriate to the
   service and formats an AII.  It formats an SAI using the same encod-
   ing rules, using the port the setup message was received on and the
   Connection Identifier.  This AI becomes the handle which will be used
   to locate the context for this call.  It then sends a Label Mapping
   message to the target node.

   When it receives a Label mapping message from the target node, it
   uses the TAI to locate the call context and completes the ATM call.

   [Note: while the details of the ATM call processing and crankback
   codes is a subject for other Fori and thus beyond the scope of the WG
   (and IETF), an informative appendix will be added if the WG so
   desires.]


6.5.3. Procedures for the PME

   The procedures for the PME follow [L2VPN-SIG] with no changes.  That
   is, the PME uses the TAI to identify interface and DLCI or VPI/VCI.
   No decoding of the TAII is necessary; the AI and AC are simply con-
   figured as in [L2VPN-SIG].

   If the forward label mapping completed successfully, the PME responds
   with an LDP label mapping message in the reverse direction.  The same
   encapsulation as the forward direction MUST be signaled.





Swallow & Spiegel            Standards Track                   [Page 13]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


6.5.4. Call Completion

   When the ME receives the label mapping message, it uses the TAI to
   (i.e. what it sent as the SAI) locate the context for this call and
   then completes the ATM call by sending a CONNECT message back to the
   PAE.


6.6. Handling Re-assembly

   When an ME detects that the requested service is Frame Relay, by
   default it will perform AAL5 re-assembly for Frame Relay SPVCs.  In
   this case the AAL5 SDU frame mode encapsulation as defined in
   [PWE3-ENCAPS] is RECOMMENDED.

   On a per ATM/PSN interface basis, an ME may be configured to not per-
   form re-assembly for Frame Relay.  No per circuit selection is pro-
   vided.  In this case the RECOMMENDED encapsulation is ATM N-to-1 Cell
   Mode.

   Both ATM SPVCCs and SPVPCs are encapsulated using one of the cell-
   mode encapsulations.  The RECOMMENDED encapsulation is ATM N-to-1
   Cell Mode.

   [As noted in the "Issues" section below, the Frame Relay format of
   the SPVC IE may not be available on some ATM equipment.  Alternative
   means of determining that the service is Frame Relay can be achieved
   in many cases by examining some combination of the .



7. Issues

   The Frame Relay format for the SPVC IE was added later.  Some ATM
   equipment still uses the ATM SPVC format with the DLCI encoded in the
   VPI/VCI fields.  To support these switches without change, and still
   allow AAL5 reassembly to be done at the ME, some other means of indi-
   cating that the circuit is Frame Relay needs to be established.













Swallow & Spiegel            Standards Track                   [Page 14]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


8. Security Considerations

   This document represents a straightforward application of [L2VPN-
   SIG].  It poses no new security considerations over and above those
   identified in that document.



9. Acknowledgments

   The authors would like to thank Chris Metz and Chandra Krishnamurthy
   for their input to this document.



10. References

10.1. Normative References

[LDP]        Andersson, L. et al., "LDP Specification", RFC 3036,
             January 2001.

[L2VPN-SIG]  Rosen, E.& Radoaca, V, "Provisioning Models and Endpoint
             Identifiers in L2VPN Signaling",
             draft-ietf-l2vpn-signaling-00.txt, September 2003.

[PWE3-CP]    Martini, L. & Rosen, E., "Pseudowire Setup and Maintenance
             using LDP", draft-ietf-pwe3-control-protocol-04.txt,
             October 2003.

[PWE3-ENCAP] Martini, et al., "Encapsulation Methods for Transport of
             ATM Cells/Frame Over IP and MPLS Networks",
             draft-ietf-pwe3-atm-encap-03.txt, October 2003.

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


10.2. Informative References

[PWE3-IANA]  Townsley, M., "IANA Allocations for pseudo Wire Edge to
             Edge Emulation (PWE3)",
             draft-ietf-pwe3-iana-allocation-04.txt, April 2004.

[ATM-AINI]   ATM Forum, "ATM Inter-Network Interface Specification
             Version 1.1 (AINI 1.1)", af-cs-0125.002, September 2002

[ATM-UNI]    ATM Forum, "ATM User-Network Interface (UNI) Signalling



Swallow & Spiegel            Standards Track                   [Page 15]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


             Specification Version 4.1", af-sig-0061.002, April 2002

[PNNI]       ATM Forum, "Private Network-Network Interface Specification
             Version 1.1", af-pnni-0055.001, April 2002

[ATM-ipaddr] ATM Forum, "IP-Based Addressing Version 1.0",
             str-cs-ipaddr-01.01, October 2003

[RFC1888]    Bound, J., et al., "OSI NSAPs and IPv6", RFC 1888,
             August 1996.




11. Authors' Addresses

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

   Email:  swallow@cisco.com


   Ethan Mickey Spiegel
   Cisco Systems, Inc.
   3700 Cisco Way
   San Jose, CA 95134

   Email:  mspiegel@cisco.com



12. Full Copyright and Intellectual Property Statements

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

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

Intellectual Property



Swallow & Spiegel            Standards Track                   [Page 16]


Internet Draft      draft-swallow-pwe3-spvc-iw-02.txt      February 2005


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

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

Acknowledgement

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

























Swallow & Spiegel            Standards Track                   [Page 17]


Html markup produced by rfcmarkup 1.129c, available from https://tools.ietf.org/tools/rfcmarkup/