Internet Engineering Task Force (IETF)                          K. Drage
Request for Comments: 6050                                Alcatel-Lucent
Category: Informational                                    November 2010
ISSN: 2070-1721


            A Session Initiation Protocol (SIP) Extension
                  for the Identification of Services

Abstract

  This document describes private extensions to the Session Initiation
  Protocol (SIP) that enable a network of trusted SIP servers to assert
  the service of authenticated users.  The use of these extensions is
  only applicable inside an administrative domain with previously
  agreed-upon policies for generation, transport, and usage of such
  information.  This document does NOT offer a general service
  identification model suitable for use between different trust domains
  or for use in the Internet at large.

  The document also defines a URN to identify both services and User
  Agent (UA) applications.  This URN can be used within the SIP header
  fields defined in this document to identify services, and also within
  the framework defined for caller preferences and callee capabilities
  to identify usage of both services and applications between end UAs.

Status of This Memo

  This document is not an Internet Standards Track specification; it is
  published for informational purposes.

  This document is a product of the Internet Engineering Task Force
  (IETF).  It represents the consensus of the IETF community.  It has
  received public review and has been approved for publication by the
  Internet Engineering Steering Group (IESG).  Not all documents
  approved by the IESG are a candidate for any level of Internet
  Standard; see Section 2 of RFC 5741.

  Information about the current status of this document, any errata,
  and how to provide feedback on it may be obtained at
  http://www.rfc-editor.org/info/rfc6050.










Drage                         Informational                     [Page 1]

RFC 6050               SIP Service Identification          November 2010


Copyright Notice

  Copyright (c) 2010 IETF Trust and the persons identified as the
  document authors.  All rights reserved.

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

Table of Contents

  1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  3
  2.  Applicability Statement  . . . . . . . . . . . . . . . . . . .  5
  3.  Conventions  . . . . . . . . . . . . . . . . . . . . . . . . .  6
  4.  Syntax of the Header Fields  . . . . . . . . . . . . . . . . .  6
    4.1.  The P-Asserted-Service Header  . . . . . . . . . . . . . .  6
    4.2.  The P-Preferred-Service Header . . . . . . . . . . . . . .  7
    4.3.  Service and Application Definition . . . . . . . . . . . .  8
    4.4.  Registration Template  . . . . . . . . . . . . . . . . . .  8
  5.  Usage of the P-Preferred-Service and P-Asserted-Service
      Header Fields  . . . . . . . . . . . . . . . . . . . . . . . . 10
    5.1.  Usage of the P-Preferred-Service and
          P-Asserted-Service Header Fields in Requests . . . . . . . 10
      5.1.1.  Procedures at User Agent Clients (UAC) . . . . . . . . 10
      5.1.2.  Procedures at Intermediate Proxies . . . . . . . . . . 11
      5.1.3.  Procedures at User Agent Servers . . . . . . . . . . . 12
    5.2.  Usage of the P-Preferred-Service and
          P-Asserted-Service Header Fields in Responses  . . . . . . 12
  6.  Examples of Usage  . . . . . . . . . . . . . . . . . . . . . . 12
  7.  Security Considerations  . . . . . . . . . . . . . . . . . . . 15
  8.  IANA Considerations  . . . . . . . . . . . . . . . . . . . . . 16
    8.1.  P-Asserted-Service and P-Preferred-Service Header
          Fields . . . . . . . . . . . . . . . . . . . . . . . . . . 16
    8.2.  Definition of Service-ID Values  . . . . . . . . . . . . . 16
  9.  References . . . . . . . . . . . . . . . . . . . . . . . . . . 17
    9.1.  Normative References . . . . . . . . . . . . . . . . . . . 17
    9.2.  Informative References . . . . . . . . . . . . . . . . . . 18








Drage                         Informational                     [Page 2]

RFC 6050               SIP Service Identification          November 2010


1.  Introduction

  This document describes private extensions to the Session Initiation
  Protocol (SIP) that enable a network of trusted SIP servers to assert
  the service, possibly subject to the user being entitled to that
  service.  The use of these extensions is only applicable inside an
  administrative domain with previously agreed-upon policies for
  generation, transport, and usage of such information.  This document
  does NOT offer a general service model suitable for use between
  different trust domains or for use in the Internet at large.

  The concept of "service" within SIP has no hard and fast rules.  RFC
  5897 [RFC5897] provides general guidance on what constitutes a
  service within SIP and what does not.

  This document also makes use of the terms "derived service
  identification" and "declarative service identification" as defined
  in RFC 5897 [RFC5897].

  It should be noted that RFC 5897 [RFC5897] clearly states that
  declarative service identification -- the process by which a user
  agent inserts a moniker into a message that defines the desired
  service, separate from explicit and well-defined protocol mechanisms
  -- is harmful.

  During a session setup, proxies may need to understand what service
  the request is related to in order to know what application server to
  contact or other service logic to invoke.  The SIP INVITE request
  contains all of the information necessary to determine the service.
  However, the calculation of the service may be computational and
  database intensive.  For example, a given trust domain's definition
  of a service might include request authorization.  Moreover, the
  analysis may require examination of the Session Description Protocol
  (SDP).

  For example, an INVITE request with video SDP directed to a video-on-
  demand Request-URI could be marked as an IPTV session.  An INVITE
  request with push-to-talk over cellular (PoC) routes could be marked
  as a PoC session.  An INVITE request with a Require header field
  containing an option tag of "foogame" could be marked as a foogame
  session.

  NOTE: If the information contained within the SIP INVITE request is
  not sufficient to uniquely identify a service, the remedy is to
  extend the SIP signaling to capture the missing element.  RFC 5897
  [RFC5897] provides further explanation.





Drage                         Informational                     [Page 3]

RFC 6050               SIP Service Identification          November 2010


  By providing a mechanism to compute and store the results of the
  domain-specific service calculation, i.e., the derived service
  identification, this optimization allows a single trusted proxy to
  perform an analysis of the request and authorize the requestor's
  permission to request such a service.  The proxy may then include a
  service identifier that relieves other trusted proxies and trusted
  UAs from performing further duplicate analysis of the request for
  their service identification purposes.  In addition, this extension
  allows user agent clients outside the trust domain to provide a hint
  of the requested service.

  This extension does not provide for the dialog or transaction to be
  rejected if the service is not supported end-to-end.  SIP provides
  other mechanisms, such as the option-tag and use of the Require and
  Proxy-Require header fields, where such functionality is required.
  No explicitly signaled service identification exists, and the session
  proceeds for each node's definition of the service in use, on the
  basis of information contained in the SDP and in other SIP header
  fields.

  This mechanism is specifically for managing the information needs of
  intermediate routing devices between the calling user and the user
  represented by the Request-URI.  In support of this mechanism, a URN
  is defined to identify the services.  This URN has wider
  applicability to additionally identify services and terminal
  applications.  Between end users, caller preferences and callee
  capabilities as specified in RFC 3840 [RFC3840] and RFC 3841
  [RFC3841] provide an appropriate mechanism for indicating such
  service and application identification.  These mechanisms have been
  extended by RFC 5688 [RFC5688] to provide further capabilities in
  this area.

  The mechanism proposed in this document relies on a new header field
  called 'P-Asserted-Service' that contains a URN.  This is supported
  by a further new header field called 'P-Preferred-Service' that also
  contains a URN and that allows the UA to express preferences
  regarding the decisions made on service within the trust domain.

  An example of the P-Asserted-Service header field is:

  P-Asserted-Service: urn:urn-7:3gpp-service.exampletelephony.version1

  A proxy server that handles a request can, after authenticating the
  originating user in some way (for example: digest authentication) to
  ensure that the user is entitled to that service, insert such a
  P-Asserted-Service header field into the request and forward it to





Drage                         Informational                     [Page 4]

RFC 6050               SIP Service Identification          November 2010


  other trusted proxies.  A proxy that is about to forward a request to
  a proxy server or UA that it does not trust removes all the
  P-Asserted-Service header field values.

  This document labels services by means of an informal URN.  This
  provides a hierarchical structure for defining services and
  subservices, and provides an address that can be resolvable for
  various purposes outside the scope of this document, e.g., to obtain
  information about the service so described.

2.  Applicability Statement

  This document describes private extensions to SIP (see RFC 3261
  [RFC3261]) that enable a network of trusted SIP servers to assert the
  service of end users or end systems.  The use of these extensions is
  only applicable inside a 'trust domain' as defined in "Short Term
  Requirements for Network Asserted Identity" (see RFC 3324 [RFC3324]).
  Nodes in such a trust domain are explicitly trusted by its users and
  end systems to publicly assert the service of each party, and that
  they have common and agreed-upon definitions of services and
  homogeneous service offerings.  The means by which the network
  determines the service to assert is outside the scope of this
  document (though it commonly entails some form of authentication).

  The mechanism for defining a trust domain is to provide a certain set
  of specifications known as 'Spec(T)', and then specify compliance to
  that set of specifications.  Spec(T) MUST specify behavior as
  documented in RFC 3324 [RFC3324].

  This document does NOT offer a general service model suitable for
  inter-domain use or use in the Internet at large.  Its assumptions
  about the trust relationship between the user and the network may not
  apply in many applications.  For example, these extensions do not
  accommodate a model whereby end users can independently assert their
  service by use of the extensions defined here.  End users assert
  their service by including the SIP and SDP parameters that correspond
  to the service they require.  Furthermore, since the asserted
  services are not cryptographically certified, they are subject to
  forgery, replay, and falsification in any architecture that does not
  meet the requirements of RFC 3324 [RFC3324].

  The asserted services also lack an indication of who specifically is
  asserting the service, and so it must be assumed that a member of the
  trust domain is asserting the service.  Therefore, the information is
  only meaningful when securely received from a node known to be a
  member of the trust domain.





Drage                         Informational                     [Page 5]

RFC 6050               SIP Service Identification          November 2010


  Despite these limitations, there are sufficiently useful specialized
  deployments, that meet the assumptions described above and can accept
  the limitations that result, to warrant informational publication of
  this mechanism.

3.  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 BCP 14, RFC 2119
  [RFC2119].

  Throughout this document, requirements for or references to proxy
  servers or proxy behavior apply similarly to other intermediaries
  within a trust domain (for example, back-to-back user agents
  (B2BUAs)).

  The term trust domain in this document has the meaning as defined in
  RFC 3324 [RFC3324].

4.  Syntax of the Header Fields

  The following syntax specification uses the augmented Backus-Naur
  Form (BNF) as described in RFC 5234 [RFC5234].

4.1.  The P-Asserted-Service Header

  The P-Asserted-Service header field is used among trusted SIP
  entities (typically intermediaries) to carry the service information
  of the user sending a SIP message.

  The P-Asserted-Service header field carries information that is
  derived service identification.  While a declarative service
  identification can assist in deriving the value transferred in this
  header field, this should be in the form of streamlining the correct
  derived service identification.

        PAssertedService = "P-Asserted-Service"
                           HCOLON PAssertedService-value

        PAssertedService-value = Service-ID *(COMMA Service-ID)

  See Section 4.4 for the definition of Service-ID in ABNF.

  Proxies can (and will) add and remove this header field.






Drage                         Informational                     [Page 6]

RFC 6050               SIP Service Identification          November 2010


  Table 1 adds the header fields defined in this document to Table 2 in
  SIP [RFC3261], Section 7.1 of the SIP-specific event notification
  [RFC3265], Tables 1 and 2 in the SIP INFO method [RFC2976], Tables 1
  and 2 in the reliability of provisional responses in SIP [RFC3262],
  Tables 1 and 2 in the SIP UPDATE method [RFC3311], Tables 1 and 2 in
  the SIP extension for instant messaging [RFC3428], Table 1 in the SIP
  REFER method [RFC3515], and Tables 2 and 3 in the SIP PUBLISH method
  [RFC3903]:

        Header field          where  proxy  ACK BYE CAN INV OPT REG SUB
        _______________________________________________________________
        P-Asserted-Service      R     admr   -   -   -   o   o   -   o

        Header field                        NOT PRA INF UPD MSG REF PUB
        _______________________________________________________________
        P-Asserted-Service                   -   -   -   -   o   o   o

                             Table 1

  Syntactically, there may be multiple P-Asserted-Service header fields
  in a request.  The semantics of multiple P-Asserted-Service header
  fields appearing in the same request is not defined at this time.
  Implementations of this specification MUST provide only one
  P-Asserted-Service header field value.

4.2.  The P-Preferred-Service Header

  The P-Preferred-Service header field is used by a user agent sending
  the SIP request to provide a hint to a trusted proxy of the preferred
  service that the user wishes to be used for the P-Asserted-Service
  field value that the trusted element will insert.

  The P-Preferred-Service header field carries information that is
  declarative service identification.  Such information should only be
  used to assist in deriving a derived service identification at the
  recipient entity.

        PPreferredService = "P-Preferred-Service"
                            HCOLON PPreferredService-value

        PPreferredService-value = Service-ID *(COMMA Service-ID)

  See Section 4.4 for the definition of Service-ID in ABNF.

  Table 2 adds the header fields defined in this document to Table 2 in
  SIP [RFC3261], Section 7.1 of the SIP-specific event notification
  [RFC3265], Tables 1 and 2 in the SIP INFO method [RFC2976], Tables 1
  and 2 in Reliability of provisional responses in SIP [RFC3262],



Drage                         Informational                     [Page 7]

RFC 6050               SIP Service Identification          November 2010


  Tables 1 and 2 in the SIP UPDATE method [RFC3311], Tables 1 and 2 in
  the SIP extension for Instant Messaging [RFC3428], Table 1 in the SIP
  REFER method [RFC3515], and Tables 2 and 3 in the SIP PUBLISH method
  [RFC3903]:

        Header field          where  proxy  ACK BYE CAN INV OPT REG SUB
        _______________________________________________________________
        P-Preferred-Service     R      dr    -   -   -   o   o   -   o

        Header field                        NOT PRA INF UPD MSG REF PUB
        _______________________________________________________________
        P-Preferred-Service                  -   -   -   -   o   o   o

                             Table 2

  Syntactically, there may be multiple P-Preferred-Service header
  fields in a request.  The semantics of multiple P-Preferred-Service
  header fields appearing in the same request is not defined at this
  time.  Implementations of this specification MUST only provide one
  P-Preferred-Service header field value.

4.3.  Service and Application Definition

  Service definitions and characteristics are outside the scope of this
  document.  Other standards organizations, vendors, and operators may
  define their own services and register them.

  A hierarchical structure is defined consisting of service identifiers
  or application identifiers, and subservice identifiers.

  The service and subservice identifiers are as described in Section 1.
  The URN may also be used to identify a service or an application
  between end users for use within the context of RFC 3840 [RFC3840]
  and RFC 3841 [RFC3841].

  IANA maintains a registry of service identifier values that have been
  assigned.  This registry has been created by the actions of Section
  8.2 of this document.

  subservice identifiers are not managed by IANA.  It is the
  responsibility of the organization that registered the service to
  manage the subservices.

4.4.  Registration Template

  Below, we include the registration template for the URN scheme
  according to RFC 3406 [RFC3406].  The URN scheme is defined as an
  informal Namespace ID (NID).



Drage                         Informational                     [Page 8]

RFC 6050               SIP Service Identification          November 2010


  Namespace ID:  urn-7

  Registration Information:
     Registration version: 1; registration date: 2009-03-22

  Declared registrant of the namespace:  3GPP Specifications Manager
     ([email protected]) (+33 (0)492944200)

  Declaration of syntactic structure:  The URN consists of a
     hierarchical service identifier or application identifier, with a
     sequence of labels separated by periods.  The leftmost label is
     the most significant one and is called 'top-level service
     identifier', while names to the right are called 'subservices' or
     'sub-applications'.  The set of allowable characters is the same
     as that for domain names (see RFC 1123 [RFC1123]) and a subset of
     the labels allowed in RFC 3958 [RFC3958].  Labels are case-
     insensitive and MUST be specified in all lowercase.  For any given
     service identifier, labels can be removed right-to-left and the
     resulting URN is still valid, referring a more generic service,
     with the exception of the top-level service identifier and
     possibly the first subservice or sub-application identifier.
     Labels cannot be removed beyond a defined basic service; for
     example, the label w.x may define a service, but the label w may
     only define an assignment authority for assigning subsequent
     values and not define a service in its own right.  In other words,
     if a service identifier 'w.x.y.z' exists, the URNs 'w.x' and
     'w.x.y' are also valid service identifiers, but w may not be a
     valid service identifier if it merely defines who is responsible
     for defining x.

       Service-ID      = "urn:urn-7:" urn-service-id
       urn-service-id  = top-level *("." sub-service-id)
       top-level       = let-dig [ *26let-dig ]
       sub-service-id  = let-dig [ *let-dig ]
       let-dig         = ALPHA / DIGIT / "-"

     While the naming convention above uses the term "service", all the
     constructs are equally applicable to identifying applications
     within the UA.

  Relevant ancillary documentation:  None

  Identifier uniqueness considerations:  A service identifier
     identifies a service, and an application identifier an application
     indicated in the service or application registration (see IANA
     Considerations (Section 8)).  Uniqueness is guaranteed by the IANA
     registration.




Drage                         Informational                     [Page 9]

RFC 6050               SIP Service Identification          November 2010


  Identifier persistence considerations:  The service or application
     identifier for the same service or application is expected to be
     persistent, although there naturally cannot be a guarantee that a
     particular service will continue to be available globally or at
     all times.

  Process of identifier assignment:  The process of identifier
     assignment is described in the IANA Considerations (Section 8).

  Process for identifier resolution:  There is no single global
     resolution service for service identifiers or application
     identifiers.

  Rules for lexical equivalence:  'service' identifiers are compared
     according to case-insensitive string equality.

  Conformance with URN syntax:  The BNF in the 'Declaration of
     syntactic structure' above constrains the syntax for this URN
     scheme.

  Validation mechanism:  Validation determines whether a given string
     is currently a validly assigned URN (see RFC 3406 [RFC3406]).  Due
     to the distributed nature of usage and since not all services are
     available everywhere, validation in this sense is not possible.

  Scope:  The scope for this URN can be local to a single domain, or
     may be more widely used.

5.  Usage of the P-Preferred-Service and P-Asserted-Service Header
   Fields

5.1.  Usage of the P-Preferred-Service and P-Asserted-Service Header
     Fields in Requests

5.1.1.  Procedures at User Agent Clients (UAC)

  The UAC MAY insert a P-Preferred-Service in a request that creates a
  dialog, or a request outside of a dialog.  This information can
  assist the proxies in identifying appropriate service capabilities to
  apply to the call.  This information MUST NOT conflict with other SIP
  or SDP information included in the request.  Furthermore, the SIP or
  SDP information needed to signal functionality of this service MUST
  be present.  Thus, if a service requires a video component, then the
  SDP has to include the media line associated with that video
  component; it cannot be assumed from the P-Preferred-Service header
  field value.  Similarly, if the service requires particular SIP





Drage                         Informational                    [Page 10]

RFC 6050               SIP Service Identification          November 2010


  functionality for which a SIP extension and a Require header field
  value is defined, then the request has to include that SIP signaling
  as well as the P-Preferred-Service header field value.

  A UAC that is within the same trust domain as the proxy to which it
  sends a request (e.g., a media gateway or application server) MAY
  insert a P-Asserted-Service header field in a request that creates a
  dialog, or a request outside of a dialog.  This information MUST NOT
  conflict with other SIP or SDP information included in the request.
  Furthermore, the SIP or SDP information needed to signal
  functionality of this service MUST be present.

5.1.2.  Procedures at Intermediate Proxies

  A proxy in a trust domain can receive a request from a node that it
  trusts or a node that it does not trust.  When a proxy receives a
  request from a node it does not trust and it wishes to add a
  P-Asserted-Service header field, the proxy MUST identify the service
  appropriate to the capabilities (e.g., SDP) in the request, MAY
  authenticate the originator of the request (in order to determine
  whether the user is subscribed for that service).  Where the
  originator of the request is authenticated, the proxy MUST use the
  identity that results from this checking and authentication to insert
  a P-Asserted-Service header field into the request.

  When a proxy receives a request containing a P-Preferred-Service
  header field, the Proxy MAY use the contents of that header field to
  assist in determining the service to be included in a P-Asserted-
  Service header field (for instance, to prioritize the order of
  comparison of filter criteria for potential services that the request
  could match).  The proxy MUST NOT use the contents of the
  P-Preferred-Service header field to identify the service without
  first checking against the capabilities (e.g., SDP) contained in the
  request.  If the proxy inserts a P-Asserted-Service header field in
  the request, the proxy MUST remove the P-Preferred-Service header
  field before forwarding the request; otherwise, the Proxy SHOULD
  include the P-Preferred-Service header field when forwarding the
  request.

  If the proxy receives a request from a node that it trusts, it can
  use the information in the P-Asserted-Service header field, if any,
  as if it had authenticated the user itself.

  If there is no P-Asserted-Service header field present, or it is not
  possible to match the request to a specific service as identified by
  the service identifier, a proxy MAY add one containing it using its
  own analysis of the information contained in the SIP request.  If the
  proxy received the request from an element that it does not trust and



Drage                         Informational                    [Page 11]

RFC 6050               SIP Service Identification          November 2010


  there is a P-Asserted-Service header present, the proxy MUST replace
  that header field's contents with a new analysis or remove that
  header field.

  The analysis performed to identify such service identifiers is
  outside the scope of this document.  However, it is perfectly valid
  as a result of the analysis not to include any service identifier in
  the forwarded request, and thus not include a P-Asserted-Service
  header field.

  If a proxy forwards a request to a node outside the proxy's trust
  domain, there MUST NOT be a P-Asserted-Service header field in the
  forwarded request.

5.1.3.  Procedures at User Agent Servers

  For a User Agent Server (UAS) outside the trust domain, the
  P-Asserted-Service header is removed before it reaches this entity;
  therefore, there are no procedures for such a device.

  However, if a UAS receives a request from a previous element that it
  does not trust, it MUST NOT use the P-Asserted-Service header field
  in any way.

  If a UA is part of the trust domain from which it received a request
  containing a P-Asserted-Service header field, then it can use the
  value freely, but it MUST ensure that it does not forward the
  information to any element that is not part of the trust domain.

5.2.  Usage of the P-Preferred-Service and P-Asserted-Service Header
     Fields in Responses

  There is no usage of these header fields in responses.

6.  Examples of Usage

  In this example, proxy.example.com creates a P-Asserted-Service
  header field from the user identity it discovered from SIP digest
  authentication, the list of services appropriate to that user, and
  the services that correspond to the SDP information included in the
  request.  Note that F1 and F2 are about identifying the user and do
  not directly form part of the capability provided in this document.
  It forwards this information to a trusted proxy that forwards it to a
  trusted gateway.  Note that these examples consist of partial SIP
  messages that illustrate only those header fields relevant to the
  authenticated identity problem.





Drage                         Informational                    [Page 12]

RFC 6050               SIP Service Identification          November 2010


     * F1   useragent.example.com -> proxy.example.com

     INVITE sip:[email protected] SIP/2.0
     Via: SIP/2.0/TCP useragent.example.com;branch=z9hG4bK-123
     To: <sip:[email protected]>
     From: "Anonymous" <sip:[email protected]>;tag=9802748
     Call-ID: 245780247857024504
     CSeq: 1 INVITE
     Max-Forwards: 70

     v=0
     o=- 2987933615 2987933615 IN IP6 5555::aaa:bbb:ccc:ddd
     s=-
     c=IN IP6 5555::aaa:bbb:ccc:ddd
     t=0 0
     m=audio 3456 RTP/AVPF 97 96
     b=AS:25.4
     a=curr:qos local sendrecv
     a=curr:qos remote none
     a=des:qos mandatory local sendrecv
     a=des:qos mandatory remote sendrecv
     a=sendrecv
     a=rtpmap:97 AMR
     a=fmtp:97 mode-set=0,2,5,7; maxframes


     * F2   proxy.example.com -> useragent.example.com

     SIP/2.0 407 Proxy Authorization
     Via: SIP/2.0/TCP useragent.example.com;branch=z9hG4bK-123
     To: <sip:[email protected]>;tag=123456
     From: "Anonymous" <sip:[email protected]>;tag=9802748
     Call-ID: 245780247857024504
     CSeq: 1 INVITE
     Proxy-Authenticate: .... realm="sip.example.com"


     * F3   useragent.example.com -> proxy.example.com

     INVITE sip:[email protected] SIP/2.0
     Via: SIP/2.0/TCP useragent.example.com;branch=z9hG4bK-124
     To: <sip:[email protected]>
     From: "Anonymous" <sip:[email protected]>;tag=9802748
     Call-ID: 245780247857024504
     CSeq: 2 INVITE
     Max-Forwards: 70
     Proxy-Authorization: realm="sip.example.com" user="fluffy"




Drage                         Informational                    [Page 13]

RFC 6050               SIP Service Identification          November 2010


     v=0
     o=- 2987933615 2987933615 IN IP6 5555::aaa:bbb:ccc:ddd
     s=-
     c=IN IP6 5555::aaa:bbb:ccc:ddd
     t=0 0
     m=audio 3456 RTP/AVPF 97 96
     b=AS:25.4
     a=curr:qos local sendrecv
     a=curr:qos remote none
     a=des:qos mandatory local sendrecv
     a=des:qos mandatory remote sendrecv
     a=sendrecv
     a=rtpmap:97 AMR
     a=fmtp:97 mode-set=0,2,5,7; maxframes


   * F4   proxy.example.com -> proxy.pstn.example (trusted)

   INVITE sip:+14085551212@proxy. pstn.example SIP/2.0
   Via: SIP/2.0/TCP useragent.example.com;branch=z9hG4bK-124
   Via: SIP/2.0/TCP proxy.example.com;branch=z9hG4bK-abc
   To: <sip:[email protected]>
   From: "Anonymous" <sip:[email protected]>;tag=9802748
   Call-ID: 245780247857024504
   CSeq: 2 INVITE
   Max-Forwards: 69
   P-Asserted-Service: urn:urn-7:3gpp-service.exampletelephony.version1

   v=0
   o=- 2987933615 2987933615 IN IP6 5555::aaa:bbb:ccc:ddd
   s=-
   c=IN IP6 5555::aaa:bbb:ccc:ddd
   t=0 0
   m=audio 3456 RTP/AVPF 97 96
   b=AS:25.4
   a=curr:qos local sendrecv
   a=curr:qos remote none
   a=des:qos mandatory local sendrecv
   a=des:qos mandatory remote sendrecv
   a=sendrecv
   a=rtpmap:97 AMR
   a=fmtp:97 mode-set=0,2,5,7; maxframes









Drage                         Informational                    [Page 14]

RFC 6050               SIP Service Identification          November 2010


   * F5   proxy.pstn.example -> gw.pstn.example (trusted)

   INVITE sip:[email protected] SIP/2.0
   Via: SIP/2.0/TCP useragent.example.com;branch=z9hG4bK-124
   Via: SIP/2.0/TCP proxy.example.com;branch=z9hG4bK-abc
   Via: SIP/2.0/TCP proxy.pstn.example;branch=z9hG4bK-a1b2
   To: <sip:[email protected]>
   From: "Anonymous" <sip:[email protected]>;tag=9802748
   Call-ID: 245780247857024504
   CSeq: 2 INVITE
   Max-Forwards: 68
   P-Asserted-Service: urn:urn-7:3gpp-service.exampletelephony.version1

   v=0
   o=- 2987933615 2987933615 IN IP6 5555::aaa:bbb:ccc:ddd
   s=-
   c=IN IP6 5555::aaa:bbb:ccc:ddd
   t=0 0
   m=audio 3456 RTP/AVPF 97 96
   b=AS:25.4
   a=curr:qos local sendrecv
   a=curr:qos remote none
   a=des:qos mandatory local sendrecv
   a=des:qos mandatory remote sendrecv
   a=sendrecv
   a=rtpmap:97 AMR
   a=fmtp:97 mode-set=0,2,5,7; maxframes

7.  Security Considerations

  The mechanism provided in this document is a partial consideration of
  the problem of service identification in SIP.  For example, these
  mechanisms provide no means by which end users can securely share
  service information end-to-end without a trusted service provider.
  This information is secured by transitive trust, which is only as
  reliable as the weakest link in the chain of trust.

  The trust domain provides a set of servers where the characteristics
  of the service are agreed for that service identifier value, and
  where the calling user is entitled to use that service.  RFC 5897
  [RFC5897] identifies the impact of allowing such service identifier
  values to "leak" outside of the trust domain, including implications
  on fraud, interoperability, and stifling of service innovation.








Drage                         Informational                    [Page 15]

RFC 6050               SIP Service Identification          November 2010


8.  IANA Considerations

8.1.  P-Asserted-Service and P-Preferred-Service Header Fields

  This document specifies two new SIP header fields: P-Asserted-Service
  and P-Preferred-Service.  Their syntax is given in Section 3.  These
  header fields are defined by the following information, which has
  been added to the header sub-registry under http://www.iana.org.

       Header Name        compact    Reference
       -----------------  -------    ---------
       P-Asserted-Service            RFC 6050
       P-Preferred-Service           RFC 6050

8.2.  Definition of Service-ID Values

  Top-level identifiers are identified by labels managed by IANA,
  according to the processes outlined in RFC 5226 [RFC5226], in a new
  registry called "Service-ID/Application-ID Labels".  Thus, creating a
  new service at the top-level requires IANA action.  The policy for
  adding service labels is 'specification required'.  The following two
  identifiers are initially defined:

     3gpp-service

     3gpp-application

  subservice identifiers are not managed by IANA.  It is the
  responsibility of the organization that registered the service to
  manage the subservices.

  Application identifiers are not managed by IANA.  It is the
  responsibility of the organization that registered the service to
  manage the applicable applications.

















Drage                         Informational                    [Page 16]

RFC 6050               SIP Service Identification          November 2010


  Entries in the registration table have the following format:

  Service/Application   Description                          Reference
  --------------------------------------------------------------------
  3gpp-service          Communication services defined by    RFC 6050
                        3GPP for use by the IM CN subsystem
                        and its attached UAs.  This value
                        in itself does not define a service
                        and requires subsequent labels to
                        define the service.

  3gpp-application      Applications defined by 3GPP for     RFC 6050
                        use by UAs attached to the IM CN
                        subsystem. This value in itself
                        does not define a service and
                        requires subsequent labels to define
                        the service.

  Here, the IM CN subsystem stands for the IP Multimedia Core Network
  subsystem.

9.  References

9.1.  Normative References

  [RFC1123]  Braden, R., "Requirements for Internet Hosts - Application
             and Support", STD 3, RFC 1123, October 1989.

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

  [RFC3261]  Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston,
             A., Peterson, J., Sparks, R., Handley, M., and E.
             Schooler, "SIP: Session Initiation Protocol", RFC 3261,
             June 2002.

  [RFC3324]  Watson, M., "Short Term Requirements for Network Asserted
             Identity", RFC 3324, November 2002.

  [RFC3406]  Daigle, L., van Gulik, D., Iannella, R., and P. Faltstrom,
             "Uniform Resource Names (URN) Namespace Definition
             Mechanisms", BCP 66, RFC 3406, October 2002.

  [RFC3958]  Daigle, L. and A. Newton, "Domain-Based Application
             Service Location Using SRV RRs and the Dynamic Delegation
             Discovery Service (DDDS)", RFC 3958, January 2005.





Drage                         Informational                    [Page 17]

RFC 6050               SIP Service Identification          November 2010


  [RFC5226]  Narten, T. and H. Alvestrand, "Guidelines for Writing an
             IANA Considerations Section in RFCs", BCP 26, RFC 5226,
             May 2008.

  [RFC5234]  Crocker, D. and P. Overell, "Augmented BNF for Syntax
             Specifications: ABNF", STD 68, RFC 5234, January 2008.

9.2.  Informative References

  [RFC2976]  Donovan, S., "The SIP INFO Method", RFC 2976,
             October 2000.

  [RFC3262]  Rosenberg, J. and H. Schulzrinne, "Reliability of
             Provisional Responses in Session Initiation Protocol
             (SIP)", RFC 3262, June 2002.

  [RFC3265]  Roach, A., "Session Initiation Protocol (SIP)-Specific
             Event Notification", RFC 3265, June 2002.

  [RFC3311]  Rosenberg, J., "The Session Initiation Protocol (SIP)
             UPDATE Method", RFC 3311, October 2002.

  [RFC3428]  Campbell, B., Rosenberg, J., Schulzrinne, H., Huitema, C.,
             and D. Gurle, "Session Initiation Protocol (SIP) Extension
             for Instant Messaging", RFC 3428, December 2002.

  [RFC3515]  Sparks, R., "The Session Initiation Protocol (SIP) Refer
             Method", RFC 3515, April 2003.

  [RFC3840]  Rosenberg, J., Schulzrinne, H., and P. Kyzivat,
             "Indicating User Agent Capabilities in the Session
             Initiation Protocol (SIP)", RFC 3840, August 2004.

  [RFC3841]  Rosenberg, J., Schulzrinne, H., and P. Kyzivat, "Caller
             Preferences for the Session Initiation Protocol (SIP)",
             RFC 3841, August 2004.

  [RFC3903]  Niemi, A., "Session Initiation Protocol (SIP) Extension
             for Event State Publication", RFC 3903, October 2004.

  [RFC5688]  Rosenberg, J., "A Session Initiation Protocol (SIP) Media
             Feature Tag for MIME Application Subtypes", RFC 5688,
             January 2010.

  [RFC5897]  Rosenberg, J., "Identification of Communications Services
             in the Session Initiation Protocol (SIP)", RFC 5897,
             June 2010.




Drage                         Informational                    [Page 18]

RFC 6050               SIP Service Identification          November 2010


Author's Address

  Keith Drage
  Alcatel-Lucent
  Quadrant, Stonehill Green, Westlea
  Swindon, Wilts
  UK

  EMail: [email protected]










































Drage                         Informational                    [Page 19]