Requirements for an End-to-End Session Identification in IP-Based Multimedia Communication Networks

Internet Engineering Task Force (IETF)                          P. Jones
Request for Comments: 7206                                  G. Salgueiro
Category: Informational                                          J. Polk
ISSN: 2070-1721                                            Cisco Systems
                                                                L. Liess
                                                        Deutsche Telekom
                                                               H. Kaplan
                                                                  Oracle
                                                                May 2014


           Requirements for an End-to-End Session Identifier
             in IP-Based Multimedia Communication Networks

Abstract

   This document specifies the requirements for an end-to-end session
   identifier in IP-based multimedia communication networks.  This
   identifier would enable endpoints, intermediate devices, and
   management and monitoring systems to identify a session end-to-end
   across multiple SIP devices, hops, and administrative domains.

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/rfc7206.

Copyright Notice

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

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

Table of Contents

   1. Introduction ....................................................3
   2. Conventions Used in This Document ...............................3
   3. Terminology .....................................................4
      3.1. What Does the Session Identifier Identify? .................4
      3.2. Communication Session ......................................5
      3.3. End-to-End .................................................6
   4. Session Identifier Use Cases ....................................6
      4.1. End-to-End Identification of a Communication Session .......6
      4.2. Protocol Interworking ......................................6
      4.3. Traffic Monitoring .........................................7
      4.4. Tracking Transferred Sessions ..............................7
      4.5. Session Signal Logging .....................................8
      4.6. Identifier Syntax ..........................................8
      4.7. 3PCC Use Case ..............................................9
   5. Requirements for the End-to-End Session Identifier ..............9
   6. Related Work in Other Standards Organizations ..................10
      6.1. Coordination with the ITU-T ...............................10
      6.2. Requirements within 3GPP ..................................11
   7. Security Considerations ........................................11
   8. Acknowledgments ................................................12
   9. Contributors ...................................................12
   10. References ....................................................12
      10.1. Normative References .....................................12
      10.2. Informative References ...................................12

1.  Introduction

   IP-based multimedia communication systems like SIP [1] and H.323 [2]
   have the concept of a "call identifier" that is globally unique.  The
   identifier is intended to represent an end-to-end communication
   session from the originating device to the terminating device.  Such
   an identifier is useful for troubleshooting, session tracking, and
   so forth.

   Unfortunately, there are a number of factors that mean that the
   current call identifiers defined in SIP and H.323 are not suitable
   for end-to-end session identification.  Perhaps most significant is
   the fact that the syntax for the call identifier in SIP and H.323 is
   different between the two protocols.  This important fact makes it
   impossible for call identifiers to be exchanged end-to-end when a
   network uses both of these session protocols.

   Another reason why the current call identifiers are not suitable to
   identify the session end-to-end is that in real-world deployments,
   devices like Back-to-Back User Agents (B2BUAs) often change the
   values as the session signaling passes through.  This is true even
   when a single session protocol is employed and is not a byproduct of
   protocol interworking.

   Lastly, identifiers that might have been used to identify a session
   end-to-end fail to meet that need when sessions are manipulated
   through supplementary service interactions.  For example, when a
   session is transferred or if a private branch exchange (PBX) joins or
   merges two communication sessions together locally, the end-to-end
   properties of currently defined identifiers are lost.

   This document specifies the requirements for an end-to-end session
   identifier in IP-based multimedia communication networks.  This
   identifier would enable endpoints, intermediate devices, and
   management and monitoring systems to identify a session end-to-end
   across multiple SIP devices, hops, and administrative domains.

2.  Conventions Used in This Document

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in RFC 2119 [3] when they
   appear in ALL CAPS.  These words may also appear in this document in
   lower case as plain English words, absent their normative meanings.

3.  Terminology

3.1.  What Does the Session Identifier Identify?

   The identifier on which this document places requirements, the
   session identifier, identifies a set of signaling messages associated
   with exactly two endpoints that, from each endpoint's perspective,
   are related to a single invocation of a communication application.

   How the endpoints determine which signaling messages share a given
   identifier (that is, what constitutes a single invocation of a
   communication application) is intentionally left loosely defined.

   The term "call" is often used as an example of such an invocation for
   voice and video communication, but different protocols and
   deployments define the scope of a "call" in different ways.  For
   instance, some systems would associate all of the activity between
   all three parties involved in a transfer as a single "call".

   Similarly, the term "session" is often used as an example of such an
   invocation, but this term is overloaded to describe both signaling
   and media-level interaction.  A single invocation of the
   communication application, as described above, may involve multiple
   RTP "sessions" as described by RFC 3550 [4], and possibly even
   multiple concurrent sessions.

   In this document, unless otherwise qualified, the term "communication
   session", or simply "session", will refer only to the set of
   signaling messages identified by the common session identifier.  That
   is, a "session" is a set of signaling messages associated with
   exactly two endpoints that, from each endpoint's perspective, are
   related to a single invocation of a communication application.

   The requirements in this document put some constraints on what an
   endpoint will consider the same, or a different, invocation of a
   communication session.  They also ensure that related sessions (as
   this document is using the term) can be correlated using only the
   session identifiers for each session.  Again, what constitutes a
   "related" session is intentionally left loosely defined.

   The definition considers messages associated with exactly two
   endpoints instead of messages sent between two endpoints to allow for
   intermediaries that create messages on an endpoint's behalf.  It is
   possible that an endpoint may not see all of the messages in a
   session (as this document is using the term) associated with it.

   This definition, along with the constraints imposed by the
   requirements in this document, facilitates specifying an identifier
   that allows the two endpoints to use two entirely different protocols
   (and hence to potentially have different ideas of what a single
   invocation means) or use two applications that have a different idea
   of what a single invocation means.

3.2.  Communication Session

   A communication session may exist between two SIP User Agents (UAs)
   and may pass through one or more intermediary devices, including
   B2BUAs or SIP proxies.  For example:

            UA A               Middlebox(es)                UA B

            SIP message(s) -------[]---[]-------> SIP message(s)
            SIP message(s)  <-----[]---[]-------  SIP message(s)

           Figure 1: Communication Session through Middlebox(es)

   The following are examples of acceptable communication sessions as
   described in Section 3.1 and are not exhaustive:

   o  A call directly between two user agents

   o  A call between two user agents with one or more SIP middleboxes in
      the signaling path

   o  A call between two user agents that was initiated using third-
      party call control (3PCC) [5]

   o  A call between two user agents (e.g., between Alice and Carol)
      that results from a different communication session (e.g., Alice
      and Bob) wherein one of those user agents (Alice) is transferred
      to another user agent (Carol) using a REFER request or a re-INVITE
      request

   The following are not considered communication sessions:

   o  A call between any two user agents wherein two or more user agents
      are engaged in a conference call via a conference focus:

      - each call between the user agent and the conference focus would
        be a communication session, and

      - each of these is a distinct communication session.

   o  A call between three user agents (e.g., Alice, Bob, and Carol)
      wherein the first user agent (Alice) ad hoc conferences the other
      two user agents (Bob and Carol):

      - The call between Alice and Bob would be one communication
        session.

      - The call between Alice and Carol would be a different
        communication session.

3.3.  End-to-End

   The term "end-to-end" in this document means the communication
   session from the point of origin, passing through any number of
   intermediaries, to the ultimate point of termination.  It is
   recognized that legacy devices may not support the end-to-end session
   identifier.  Since such an endpoint will not create a session
   identifier, an intermediary device that supports this identifier can
   inject an identifier into the session signaling.

4.  Session Identifier Use Cases

4.1.  End-to-End Identification of a Communication Session

   For SIP messaging that either does not involve SIP servers or only
   involves SIP proxies, the Call-ID header field value sufficiently
   identifies each SIP message within a transaction (see Section 17 of
   [1]) or dialog (see Section 12 of [1]).  This is not the case when
   either B2BUAs or Session Border Controllers (SBCs) [6] are in the
   signaling path between User Agents (UAs).  Therefore, we need the
   ability to identify each communication session through a single SIP
   header field, regardless of which types of SIP servers are in the
   signaling path between UAs.  For messages that create a dialog, each
   message within the same dialog MUST use the same session identifier.

   Derived Requirements: All Requirements in Section 5.

4.2.  Protocol Interworking

   A communication session might originate on an H.323 [2] endpoint and
   pass through an SBC before ultimately reaching a terminating SIP user
   agent.  Likewise, a call might originate on a SIP user agent and
   terminate on an H.323 endpoint.  It MUST be possible to identify such
   sessions end-to-end across the plurality of devices, networks, or
   administrative domains.

   It is anticipated that the ITU-T will define protocol elements for
   H.323 to make the end-to-end signaling possible.

   Derived Requirements: REQ5, REQ7 (Section 5).

4.3.  Traffic Monitoring

   UA A and UA B communicate using SIP messaging with a SIP B2BUA acting
   as a middlebox that belongs to a SIP service provider.  For privacy
   reasons, the B2BUA changes the SIP header fields that reveal
   information related to the SIP users, devices, or domain identities.
   The service provider uses an external device to monitor and log all
   SIP traffic coming to and from the B2BUA.  In the case of failures
   reported by the customer or when security issues arise (e.g., theft
   of service), the service provider has to analyze the logs from the
   past several days or weeks and then correlates those messages that
   were messages for a single end-to-end SIP session.

   For this scenario, we must consider three particular use cases:

   a) UAs A and B support the end-to-end session identifier.

      Derived Requirements: REQ1, REQ3, REQ4, REQ6.

   b) Only UA A supports the end-to-end session identifier; UA B
      does not.

      Derived Requirements: REQ1, REQ3, REQ4, REQ5, REQ6.

   c) UAs A and B do not support the end-to-end session identifier.

      Derived Requirements: REQ1, REQ3, REQ4, REQ5, REQ6.

4.4.  Tracking Transferred Sessions

   It is difficult to track which SIP messages were involved in the same
   call across transactions, especially when invoking supplementary
   services such as call transfer or call join.  There exists a need for
   the ability to track communication sessions as they are transferred,
   one side at a time, until completion of the session (i.e., until a
   BYE is sent).

   Derived Requirements: REQ1, REQ2, REQ9.

4.5.  Session Signal Logging

   An after-the-fact search of SIP messages to determine which messages
   were part of the same transaction or call is difficult when B2BUAs
   and SBCs are involved in the signaling between UAs.  Mapping more
   than one Call-ID together can be challenging because all of the
   values in SIP header fields on one side of the B2BUA or SBC will
   likely be different than those on the other side.  If multiple B2BUAs
   and/or SBCs are in the signaling path, more than two sets of header
   field values will exist, creating more of a challenge.  Creating a
   common header field value through all SIP entities will greatly
   reduce any challenge for the purposes of debugging, communication
   tracking (such as for security purposes in case of theft of
   service), etc.

   Derived Requirements: REQ1, REQ3, REQ5, REQ6.

4.6.  Identifier Syntax

   A syntax that is too lax (e.g., one that allows special characters or
   a very long identifier) would make it difficult to encode the
   identifier in other protocols.  Therefore, the syntax of the
   identifier should be reasonably constrained.

   Derived Requirement: REQ8.

4.7.  3PCC Use Case

   Third-party call control refers to the ability of an entity to create
   a call in which communication is actually between two or more parties
   other than the one setting up the call.  For example, a B2BUA acting
   as a third-party controller could establish a call between two SIP
   UAs using 3PCC procedures as described in Section 4.1 of RFC 3725
   [5], the flow for which is reproduced below.

                A              Controller               B
                |(1) INVITE no SDP  |                   |
                |<------------------|                   |
                |(2) 200 offer1     |                   |
                |------------------>|                   |
                |                   |(3) INVITE offer1  |
                |                   |------------------>|
                |                   |(4) 200 OK answer1 |
                |                   |<------------------|
                |                   |(5) ACK            |
                |                   |------------------>|
                |(6) ACK answer1    |                   |
                |<------------------|                   |
                |(7) RTP            |                   |
                |.......................................|

                Figure 2: Session Identifier 3PCC Scenario

   Such a flow must result in a single session identifier being used for
   the communication session between UA A and UA B.  This use case does
   not extend to three SIP UAs.

   Derived Requirement: REQ9.

5.  Requirements for the End-to-End Session Identifier

   The following requirements are derived from the use cases and
   additional constraints regarding the construction of the identifier.

   REQ1: It MUST be possible for an administrator or an external device
      that monitors the SIP traffic to use the identifier to identify
      those dialogs, transactions, and messages that were at some point
      in time components of a single end-to-end SIP session (e.g., parts
      of the same call).

   REQ2: It MUST be possible to correlate two end-to-end sessions when a
      session is transferred or if two different sessions are joined
      together via an intermediary (e.g., a PBX).

   REQ3: The solution MUST require that the identifier, if present, pass
      unchanged through SIP B2BUAs or other intermediaries.

   REQ4: The identifier MUST NOT reveal any information related to any
      SIP user, device, or domain identity.  Additionally, it MUST NOT
      be possible to correlate a set of session identifiers produced
      over a period of time with one another, or with a particular user
      or device.  This includes any IP address, port, hostname, domain
      name, username, Address-of-Record, Media Access Control (MAC)
      address, IP address family, transport type, subscriber ID,
      Call-ID, tags, or other SIP header field or body parts.

   REQ5: It MUST be possible to identify SIP traffic with an end-to-end
      session identifier from and to end devices that do not support
      this new identifier, such as by allowing an intermediary to inject
      an identifier into the session signaling.

   REQ6: The identifier SHOULD be unique in time and space, similar to
      the Call-ID.

   REQ7: The identifier SHOULD be constructed in such a way as to make
      it suitable for transmission in SIP [1] and H.323 [2].

   REQ8: The identifier SHOULD use a restricted syntax and length so as
      to allow the identifier to be used in other protocols.

   REQ9: It MUST be possible to correlate two end-to-end sessions when
      the sessions are created by a third-party controller using 3PCC
      procedures as shown in Figure 1 of RFC 3725 [5].

6.  Related Work in Other Standards Organizations

6.1.  Coordination with the ITU-T

   IP multimedia networks are often comprised of a mix of session
   protocols like SIP [1] and H.323 [2].  A benefit of the session
   identifier is that it uniquely identifies a communication session
   end-to-end across session protocol boundaries.  Therefore, the need
   for coordinated standardization activities across Standards
   Development Organizations (SDOs) is imperative.

   To facilitate this, a parallel effort is underway in the ITU-T to
   introduce the session identifier for H.323 in such a way as to be
   interoperable with the procedures defined by the IETF.

6.2.  Requirements within 3GPP

   The Third Generation Partnership Project (3GPP) identified in their
   Release 9 the need for a session identifier for operation and
   maintenance purposes to correlate flows in an end-to-end
   communication session.  3GPP TS24.229 [7] points to the fact that the
   session identifier can be used to correlate SIP messages belonging to
   the same session.  In the case where signaling passes through SIP
   entities like B2BUAs, the end-to-end session identifier indicates
   that these dialogs belong to the same end-to-end SIP communication
   session.

7.  Security Considerations

   The security vulnerabilities, attacks, and threat models affecting
   other similar SIP identifiers are well documented in RFC 3261 [1] and
   are equally applicable to the end-to-end session identifier and
   subject to the same mitigating security best practices.  Further,
   storage of the session identifier in a log file is also subject to
   the security considerations specified in RFC 6872 [8].

   An end-to-end identifier, if not properly constructed, could provide
   confidential information that would allow one to identify the
   individual, device, or domain initiating or terminating a
   communication session.  In adhering to REQ4, the solution produced in
   accordance with these requirements MUST take appropriate measures to
   properly secure and obfuscate sensitive or private information that
   might allow one to identify a person, device, or domain.  This means
   that the end-to-end session identifier MUST NOT reveal information
   elements such as the MAC address or IP address.  It is outside the
   scope of this document to specify the implementation details of such
   security and privacy measures.  Those details may vary with the
   specific construction mechanism selected for the end-to-end session
   identifier and therefore will be discussed in the document specifying
   the actual end-to-end identifier.

   A key security consideration is to ensure that an attacker cannot
   surreptitiously spoof the identifier and effectively render it
   useless to diagnostic equipment that cannot properly correlate
   signaling messages due to the duplicate session identifiers that
   exist in the same space and time.  In accordance with REQ6, this
   end-to-end identifier MUST be sufficiently long and random to prevent
   it from being guessable as well as avoid collision with another
   identifier.  The secure transport of the identifier, need for
   authentication, encryption, etc. should be appropriately evaluated
   based on the network infrastructure, transport domain, and usage
   scenarios for the end-to-end session identifier.

8.  Acknowledgments

   The authors would like to acknowledge Paul Kyzivat, Christer
   Holmberg, Charles Eckel, Andy Hutton, Salvatore Loreto, Keith Drage,
   and Chris Pearce for their contribution and collaboration in
   developing this document.

9.  Contributors

   Roland Jesske and Parthasarathi Ravindran provided substantial
   contributions to this document during its initial creation.

10.  References

10.1.  Normative References

   [1]  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.

   [2]  Recommendation ITU-T H.323, "Packet-based multimedia
        communications systems", December 2009.

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

10.2.  Informative References

   [4]  Schulzrinne, H., Casner, S., Frederick, R., and V. Jacobson,
        "RTP: A Transport Protocol for Real-Time Applications", STD 64,
        RFC 3550, July 2003.

   [5]  Rosenberg, J., Peterson, J., Schulzrinne, H., and G. Camarillo,
        "Best Current Practices for Third Party Call Control (3pcc) in
        the Session Initiation Protocol (SIP)", BCP 85, RFC 3725,
        April 2004.

   [6]  Hautakorpi, J., Ed., Camarillo, G., Penfield, R., Hawrylyshen,
        A., and M. Bhatia, "Requirements from Session Initiation
        Protocol (SIP) Session Border Control (SBC) Deployments",
        RFC 5853, April 2010.

   [7]  3GPP TS 24.229, "IP multimedia call control protocol based on
        Session Initiation Protocol (SIP) and Session Description
        Protocol (SDP); Stage 3".

   [8]  Gurbani, V., Ed., Burger, E., Ed., Anjali, T., Abdelnur, H., and
        O. Festor, "The Common Log Format (CLF) for the Session
        Initiation Protocol (SIP): Framework and Information Model",
        RFC 6872, February 2013.

Authors' Addresses

   Paul E. Jones
   Cisco Systems, Inc.
   7025 Kit Creek Rd.
   Research Triangle Park, NC  27709
   USA

   Phone: +1 919 476 2048
   EMail: paulej@packetizer.com
   IM: xmpp:paulej@packetizer.com


   Gonzalo Salgueiro
   Cisco Systems, Inc.
   7025 Kit Creek Rd.
   Research Triangle Park, NC  27709
   USA

   Phone: +1 919 392 3266
   EMail: gsalguei@cisco.com
   IM: xmpp:gsalguei@cisco.com


   James Polk
   Cisco Systems, Inc.
   3913 Treemont Circle
   Colleyville, TX
   USA

   Phone: +1 817 271 3552
   EMail: jmpolk@cisco.com
   IM: xmpp:jmpolk@cisco.com


   Laura Liess
   Deutsche Telekom NP
   64295 Darmstadt
   Heinrich-Hertz-Str. 3-7
   Germany

   Phone: +49 6151 268 2761
   EMail: laura.liess.dt@gmail.com
   Hadriel Kaplan
   Oracle
   71 Third Ave.
   Burlington, MA  01803
   USA

   EMail: hadriel.kaplan@oracle.com