Network Working Group                                     Nevil Brownlee
INTERNET-DRAFT                                The University of Auckland
Category: Informational                                      Alan Blount
<draft-ietf-aaa-accounting-attributes-03.txt>            MetraTech Corp.
14 April 2000

                Accounting Attributes and Record Formats

Status of this Memo

   This document is an Internet-Draft, and is in full conformance with
   all provisions of Section 10 of RFC 2026.

   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-

   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 .

   The list of Internet-Draft Shadow Directories can be accessed at .

   The distribution of this memo is unlimited.  It is filed as <draft-
   ietf-aaa-accounting-attributes-03.txt>, and expires October 31, 2000.
   Please send comments to the authors.


   This draft summarises IETF and ITU-T documents related to Accounting.
   A classification scheme for the Accounting Attributes in the
   summarised documents is presented.  Exchange formats for Accounting
   data records are discussed, as are advantages and disadvantages of
   integrated versus separate record formats and transport protocols.
   This draft discusses service definition independence, extensibility,
   and versioning.  Compound service definition capabilities are

Table of Contents

   1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . .   3
   2. Terminology and Notation . . . . . . . . . . . . . . . . . . .   3
   3. Architecture Model . . . . . . . . . . . . . . . . . . . . . .   4
   4. IETF Documents . . . . . . . . . . . . . . . . . . . . . . . .   4
   4.1. RADIUS . . . . . . . . . . . . . . . . . . . . . . . . . . .   4
   4.1.1. RADIUS Attributes  . . . . . . . . . . . . . . . . . . . .   5
   4.2. DIAMETER . . . . . . . . . . . . . . . . . . . . . . . . . .   7

Brownlee, Blount                                                [Page 1]

INTERNET-DRAFT                                                March 2000

   4.2.1. DIAMETER Attributes  . . . . . . . . . . . . . . . . . . .   7
   4.3. ROAMOPS  . . . . . . . . . . . . . . . . . . . . . . . . . .   8
   4.4. RTFM . . . . . . . . . . . . . . . . . . . . . . . . . . . .   8
   4.4.1. RTFM Attributes  . . . . . . . . . . . . . . . . . . . . .   9
   4.5. ISDN MIB . . . . . . . . . . . . . . . . . . . . . . . . . .  10
   4.5.1. ISDN Attributes  . . . . . . . . . . . . . . . . . . . . .  10
   4.6. AToMMIB  . . . . . . . . . . . . . . . . . . . . . . . . . .  10
   4.6.1. AToMMIB Attributes . . . . . . . . . . . . . . . . . . . .  11
   4.7. QoS: RSVP and DIFFSERV . . . . . . . . . . . . . . . . . . .  11
   4.7.1. QoS: RSVP and DIFFSERV Attributes  . . . . . . . . . . . .  12
   5. ITU-T Documents  . . . . . . . . . . . . . . . . . . . . . . .  13
   5.1. Q.825: Call Detail Recording . . . . . . . . . . . . . . . .  13
   5.2. Q.825 Attributes . . . . . . . . . . . . . . . . . . . . . .  13
   6. Other Documents  . . . . . . . . . . . . . . . . . . . . . . .  17
   6.1. TIPHON: ETSI TS 101 321  . . . . . . . . . . . . . . . . . .  17
   6.2. MSIX . . . . . . . . . . . . . . . . . . . . . . . . . . . .  18
   7. Accounting File and Record Formats . . . . . . . . . . . . . .  18
   7.1. ASN.1 Records  . . . . . . . . . . . . . . . . . . . . . . .  18
   7.1.1. RTFM and AToMMIB . . . . . . . . . . . . . . . . . . . . .  18
   7.1.2. Q.825  . . . . . . . . . . . . . . . . . . . . . . . . . .  18
   7.2. Binary Records . . . . . . . . . . . . . . . . . . . . . . .  18
   7.2.1. RADIUS . . . . . . . . . . . . . . . . . . . . . . . . . .  18
   7.2.2. DIAMETER . . . . . . . . . . . . . . . . . . . . . . . . .  19
   7.3. Text Records . . . . . . . . . . . . . . . . . . . . . . . .  19
   7.3.1. ROAMOPS  . . . . . . . . . . . . . . . . . . . . . . . . .  20
   8. AAA Requirements . . . . . . . . . . . . . . . . . . . . . . .  21
   8.1. A Well-defined Set of Attributes . . . . . . . . . . . . . .  21
   8.2. A Simple Interchange Format  . . . . . . . . . . . . . . . .  21
   9. Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . .  22
   9.1. Record Format vs. Protocol . . . . . . . . . . . . . . . . .  22
   9.2. Tagged, Typed Data . . . . . . . . . . . . . . . . . . . . .  23
   9.2.1. Standard Type Definitions  . . . . . . . . . . . . . . . .  23
   9.3. Transaction Identifiers  . . . . . . . . . . . . . . . . . .  24
   9.4. Service Definitions  . . . . . . . . . . . . . . . . . . . .  24
   9.4.1. Service Independence . . . . . . . . . . . . . . . . . . .  24
   9.4.2. Versioned Service Definitions  . . . . . . . . . . . . . .  26
   9.4.3. Relationships Among Usage Events . . . . . . . . . . . . .  26
   9.4.4. Service Namespace Management . . . . . . . . . . . . . . .  27
   10. Encodings . . . . . . . . . . . . . . . . . . . . . . . . . .  27
   11. Security Considerations . . . . . . . . . . . . . . . . . . .  28
   12. References  . . . . . . . . . . . . . . . . . . . . . . . . .  28
   13. Authors' Addresses  . . . . . . . . . . . . . . . . . . . . .  31

Brownlee, Blount                                                [Page 2]

INTERNET-DRAFT                                                March 2000

1.  Introduction

   This draft summarises IETF and ITU-T documents related to Accounting.
   For those documents which describe Accounting Attributes (i.e. quan-
   tities which can be measured and reported), an Attribute Summary is
   given.  Although several of the documents describe Attributes which
   are similar, no attempt is made to identify those which are the same
   in several documents.  An extensible classification scheme for AAA
   Accounting Attributes is proposed; it is a superset of the attributes
   in all the documents summarised.

   Many existing accounting record formats and protocols [RAD-ACT]
   [TIPHON] are of limited use due to their single-service descriptive
   facilities and lack of extensibility.  While some record formats and
   protocols support extensible attributes [RAD-ACT], none provide iden-
   tification, type checking, or versioning support for defined group-
   ings of attributes (service definitions).  This draft makes a case
   for well-defined services.

   Advantages and disadvantages of integrated versus separate record
   formats and transport protocols are discussed.  This draft discusses
   service definition independence, extensibility, and versioning.  Com-
   pound service definition capabilities are described.

2.  Terminology and Notation

   The following terms are used throughout the document.

   Accounting Server
      A network element that accepts Usage Events from Service Elements.
      It acts as an interface to back-end rating, billing, and opera-
      tions support systems.

   Attribute-Value Pair (AVP)
      A representation for a Usage Attribute consisting of the name of
      the Attribute and a value.

      A component of a Usage Event.  A Usage Event describing a phone
      call, for instance, might have a "duration" Property.

      A type of task that is performed by a Service Element for a Ser-
      vice Consumer.

   Service Consumer
      Client of a Service Element.  End-user of a network service.

   Service Definition
      A specification for a particular service.  It is composed of a
      name or other identifier, versioning information, and a collection
      of Properties.

   Service Element

Brownlee, Blount                                                [Page 3]

INTERNET-DRAFT                                                March 2000

      A network element that provides a service to Service Consumers.
      Examples include RAS devices, voice and fax gateways, conference

   Usage Attribute
      A component of a Usage Event that describes some metric of service

   Usage Event
      The description of an instance of service usage.

3.  Architecture Model

   Service Elements provide Services to Service Consumers.  Before,
   while, and/or after services are provided, the Service Element
   reports Usage Events to an Accounting Server.  Alternately, the
   Accounting Server may query the Service Element for Usage Events.
   Usage events are sent singly or in bulk.

      +------------+       +-----------+              +------------+
      |  Service   |<----->|  Service  | Usage Events | Accounting |
      |  Consumer  |   +-->|  Element  |------------->|   Server   |
      +------------+   |   +-----------+              +------------+
      +------------+   |
      |  Service   |<--+
      |  Consumer  |

   Accounting Servers may forward Usage Events to other systems, possi-
   bly in other administrative domains.  These transfers are not
   addressed by this document.

4.  IETF Documents

   In March 1999 there were at least 19 Internet Drafts and 8 RFCs con-
   cerned with Accounting.  These are summarised (by working group) in
   the following sections.

4.1.  RADIUS

   The RADIUS protocol [RAD-PROT] carries authentication, authorization
   and configuration information between a Network Access Server (NAS)
   and an authentication server.  Requests and responses carried by the
   protocol are expressed in terms of RADIUS attributes such as User-
   Name, Service-Type, and so on.  These attributes provide the informa-
   tion needed by a RADIUS server to authenticate users and to establish
   authorized network service for them.

   The protocol was extended to carry accounting information between a
   NAS and a shared accounting server.  This was achieved by defining a
   set of RADIUS accounting attributes [RAD-ACT].

   RADIUS packets have a short header containing the RADIUS packet type

Brownlee, Blount                                                [Page 4]

INTERNET-DRAFT                                                March 2000

   and authenticator (sixteen octets) and length, followed by a sequence
   of (Type, Length, Value) triples, one for each attribute.

   RADIUS is very widely used, and a number of significant new exten-
   sions to it have been proposed.  For example [RAD-EXT] discusses
   extensions to implement the Extensible Authentication Protocol (EAP)
   and the Apple Remote Access Protocol (ARAP). [RAD-TACC] discusses
   extensions to permit RADIUS to interwork effectively with tunnels
   using protocols such as PPTP and L2TP.

4.1.1.  RADIUS Attributes

   Each RADIUS attribute is identified by an 8-bit number, referred to
   as the RADIUS Type field.  Up-to-date values of this field are speci-
   fied in the most recent Assigned Numbers RFC [ASG-NBR], but the cur-
   rent list is as follows:

   RADIUS Attributes [RAD-PROT]             36  Login-LAT-Group
                                            37  Framed-AppleTalk-Link
       1  User-Name                         38  Framed-AppleTalk-Network
       2  User-Password                     39  Framed-AppleTalk-Zone
       3  CHAP-Password
       4  NAS-IP-Address                    60  CHAP-Challenge
       5  NAS-Port                          61  NAS-Port-Type
       6  Service-Type                      62  Port-Limit
       7  Framed-Protocol                   63  Login-LAT-Port
       8  Framed-IP-Address
       9  Framed-IP-Netmask              RADIUS Accounting Attributes
      10  Framed-Routing                 [RAD-ACT]
      11  Filter-Id
      12  Framed-MTU                        40  Acct-Status-Type
      13  Framed-Compression                41  Acct-Delay-Time
      14  Login-IP-Host                     42  Acct-Input-Octets
      15  Login-Service                     43  Acct-Output-Octets
      16  Login-TCP-Port                    44  Acct-Session-Id
      17  (unassigned)                      45  Acct-Authentic
      18  Reply-Message                     46  Acct-Session-Time
      19  Callback-Number                   47  Acct-Input-Packets
      20  Callback-Id                       48  Acct-Output-Packets
      21  (unassigned)                      49  Acct-Terminate-Cause
      22  Framed-Route                      50  Acct-Multi-Session-Id
      23  Framed-IPX-Network                51  Acct-Link-Count
      24  State
      25  Class                          RADIUS Extension Attributes
      26  Vendor-Specific                [RAD-EXT]
      27  Session-Timeout
      28  Idle-Timeout                      52  Acct-Input-Gigawords
      29  Termination-Action                53  Acct-Output-Gigawords
      30  Called-Station-Id                 54  Unused
      31  Calling-Station-Id                55  Event-Timestamp
      32  NAS-Identifier
      33  Proxy-State                       70  ARAP-Password
      34  Login-LAT-Service                 71  ARAP-Features
      35  Login-LAT-Node                    72  ARAP-Zone-Access

Brownlee, Blount                                                [Page 5]

INTERNET-DRAFT                                                March 2000

      73  ARAP-Security
      74  ARAP-Security-Data
      75  Password-Retry
      76  Prompt
      77  Connect-Info
      78  Configuration-Token
      79  EAP-Message
      80  Signature

      84  ARAP-Challenge-Response

   RADIUS Tunneling Attributes

      51  Acct-Link-Count

      64  Tunnel-Type
      65  Tunnel-Medium-Type
      66  Tunnel-Client-Endpoint
      67  Tunnel-Server-Endpoint
      68  Acct-Tunnel-Connection
      69  Tunnel-Password

      81  Tunnel-Private-Group-ID
      82  Tunnel-Assignment-ID
      83  Tunnel-Preference

Brownlee, Blount                                                [Page 6]

INTERNET-DRAFT                                                March 2000


   The DIAMETER framework [DIAM-FRAM] defines a policy protocol used by
   clients to perform Policy, AAA and Resource Control.  This allows a
   single server to handle policies for many services.  The DIAMETER
   protocol consists of a header followed by objects.  Each object is
   encapsulated in a header known as an Attribute-Value Pair (AVP).

   DIAMETER defines a base protocol that specifies the header formats,
   security extensions and requirements as well as a small number of
   mandatory commands and AVPs.  A new service can extend DIAMETER by
   extending the base protocol to support new functionality.

   One key differentiator with DIAMETER is its inherent support for
   Inter-Server communication.  Although this can be achieved in a vari-
   ety of ways, the most useful feature is the ability to "proxy" mes-
   sages across a set of DIAMETER servers (known as a proxy chain).

   The DIAMETER Accounting Extension document [DIAM-ACT] extends DIAME-
   TER by defining a protocol for securely transferring accounting
   records over the DIAMETER base protocol.  This includes the case
   where accounting records may be passed through one or more intermedi-
   ate proxies, in accordance with the 'referral broker' model.

   The DIAMETER accounting protocol [DIAM-ACT] defines DIAMETER records
   for transferring an ADIF record (see below).  It introduces five new
   attributes (480..485) which specify the way in which accounting
   information is to be delivered between DIAMETER servers.

4.2.1.  DIAMETER Attributes

   DIAMETER AVPs are identified by a 16-bit number defined in [DIAM-
   AUTH].  Since most of the AVPs found in that document were copied
   from the RADIUS protocol [RAD-PROT], it is possible to have both
   RADIUS and DIAMETER servers read the same dictionary and users files.

   The backward compatibility that DIAMETER offers is intended to facil-
   itate deployment.  To this end, DIAMETER inherits the RADIUS
   attributes, and adds only a few of its own.

   In the list below attribute numbers which are used for RADIUS
   attributes but not for DIAMETER are indicated with a star (*).
   RADIUS attributes used by DIAMETER are not listed again here.

   The DIAMETER attributes are:

       4      (unassigned, *)
      17      (unassigned)
      21      (unassigned)
      24      (unassigned, *)
      25      (unassigned, *)
      27      (unassigned, *)
      32      (unassigned, *)
      33      (unassigned, *)

Brownlee, Blount                                                [Page 7]

INTERNET-DRAFT                                                March 2000

     280      Filter-Rule
     281      Framed-Password-Policy

     480      Accounting-Record-Type
     481      ADIF-Record
     482      Accounting-Interim-Interval
     483      Accounting-Delivery-Max-Batch
     484      Accounting-Delivery-Max-Delay
     485      Accounting-Record-Number

     600      SIP-Sequence
     601      SIP-Call-ID
     602      SIP-To
     603      SIP-From


   [ROAM-IMPL] reviews the design and functionality of existing roaming
   implementations.  "Roaming capability" may be loosely defined as the
   ability to use any one of multiple Internet service providers (ISPs),
   while maintaining a formal customer-vendor relationship with only
   one.  One requirement for successful roaming is the provision of
   effective accounting.

   [ROAM-ADIF] proposes a standard accounting record format, the
   Accounting Data Interchange Format (ADIF), which is designed to com-
   pactly represent accounting data in a protocol-independent manner.
   As a result, ADIF may be used to represent accounting data from any
   protocol using attribute value pairs (AVPs) or variable bindings.

   ADIF does not define accounting attributes of its own.  Instead, it
   gives examples of accounting records using the RADIUS accounting

4.4.  RTFM

   The RTFM Architecture [RTFM-ARC] provides a general method of measur-
   ing network traffic flows between "metered traffic groups."  Each
   RTFM flow has a set of "address" attributes, which define the traffic
   groups at each of the flow's end-points.

   As well as address attributes, each flow has traffic-related
   attributes, e.g. times of first and last packets, counts for packets
   and bytes in each direction.

   RTFM flow measurements are made by RTFM meters [RTFM-MIB] and col-
   lected by RTFM meter readers using SNMP. The MIB uses a "DataPackage"
   convention, which specifies the attribute values to be read from a
   flow table row.  The meter returns the values for each required
   attribute within a BER-encoded sequence.  This means there is only
   one object identifier for the whole sequence, greatly reducing the
   number of bytes required to retrieve the data.

Brownlee, Blount                                                [Page 8]

INTERNET-DRAFT                                                March 2000

4.4.1.  RTFM Attributes

   RTFM attributes are identified by a 16-bit attribute number.

   The RTFM Attributes are:

    0  Null
    1  Flow Subscript                Integer    Flow table info

    4  Source Interface              Integer    Source Address
    5  Source Adjacent Type          Integer
    6  Source Adjacent Address       String
    7  Source Adjacent Mask          String
    8  Source Peer Type              Integer
    9  Source Peer Address           String
   10  Source Peer Mask              String
   11  Source Trans Type             Integer
   12  Source Trans Address          String
   13  Source Trans Mask             String

   14  Destination Interface         Integer    Destination Address
   15  Destination Adjacent Type     Integer
   16  Destination Adjacent Address  String
   17  Destination AdjacentMask      String
   18  Destination PeerType          Integer
   19  Destination PeerAddress       String
   20  Destination PeerMask          String
   21  Destination TransType         Integer
   22  Destination TransAddress      String
   23  Destination TransMask         String

   26  Rule Set Number               Integer    Meter attribute

   27  Forward Bytes                 Integer    Source-to-Dest counters
   28  Forward Packets               Integer
   29  Reverse Bytes                 Integer    Dest-to-Source counters
   30  Reverse Packets               Integer
   31  First Time                    Timestamp  Activity times
   32  Last Active Time              Timestamp
   33  Source Subscriber ID          String     Session attributes
   34  Destination Subscriber ID     String
   35  Session ID                    String

   36  Source Class                  Integer    "Computed" attributes
   37  Destination Class             Integer
   38  Flow Class                    Integer
   39  Source Kind                   Integer
   40  Destination Kind              Integer
   41  Flow Kind                     Integer

   50  MatchingStoD                  Integer    PME variable

   51  v1                            Integer    Meter Variables
   52  v2                            Integer

Brownlee, Blount                                                [Page 9]

INTERNET-DRAFT                                                March 2000

   53  v3                            Integer
   54  v4                            Integer
   55  v5                            Integer

   65-127 "Extended" attributes
             (to be defined by the RTFM working group)

4.5.  ISDN MIB

   The ISDN MIB [ISDN-MIB] defines a minimal set of managed objects for
   SNMP-based management of ISDN terminal interfaces.  It does not
   explicitly define anything related to accounting, however it does
   define isdnBearerChargedUnits as

       The number of charged units for the current or last connection.
       For incoming calls or if charging information is not supplied
       by the switch, the value of this object is zero.

   This allows for an ISDN switch to convert its traffic flow data (such
   as Call Connect Time) into charging data.

4.5.1.  ISDN Attributes

   The relevant object in the MIB is the ISDN bearer table, which has
   entries in the following form:

   IsdnBearerEntry ::=
       SEQUENCE {
           isdnBearerChannelType           INTEGER,
           isdnBearerOperStatus            INTEGER,
           isdnBearerChannelNumber         INTEGER,
           isdnBearerPeerAddress           DisplayString,
           isdnBearerPeerSubAddress        DisplayString,
           isdnBearerCallOrigin            INTEGER,
           isdnBearerInfoType              INTEGER,
           isdnBearerMultirate             TruthValue,
           isdnBearerCallSetupTime         TimeStamp,
           isdnBearerCallConnectTime       TimeStamp,
           isdnBearerChargedUnits          Gauge32

4.6.  AToMMIB

   The "ATM Accounting Information MIB" document [ATM-ACT] describes a
   large set of accounting objects for ATM connections.  An administra-
   tor may select objects from this set using a selector of the form
   (subtree, list) where "subtree" specifies an object identifier from
   the AToMMIB.  For each subtree there is a table holding values for
   each ATM connection.  The required connections are indicated by set-
   ting bits in "list," which is an octet string.  For example, the set
   containing the number of received cells for the first eight ATM con-
   nections would be selected by (atmAcctngReceivedCells, 0xFF).

   The Connection-Oriented Accounting MIB document [ATM-COLL] defines a

Brownlee, Blount                                               [Page 10]

INTERNET-DRAFT                                                March 2000

   MIB providing managed objects used for controlling the collection and
   storage of accounting information for connection-oriented networks
   such as ATM. The accounting data is collected into files for later
   retrieval via a file transfer protocol.  Records within an accounting
   file are stored as BER strings [ASN1, BER].

4.6.1.  AToMMIB Attributes

   Accounting data objects within the AToMMBIB are identified by the
   last integer in their object identifiers.

   The ATM accounting data objects are:

      1   atmAcctngConnectionType
      2   atmAcctngCastType
      3   atmAcctngIfName
      4   atmAcctngIfAlias
      5   atmAcctngVpi
      6   atmAcctngVci
      7   atmAcctngCallingParty
      8   atmAcctngCalledParty
      9   atmAcctngCallReference
     10   atmAcctngStartTime
     11   atmAcctngCollectionTime
     12   atmAcctngCollectMode
     13   atmAcctngReleaseCause
     14   atmAcctngServiceCategory
     15   atmAcctngTransmittedCells
     16   atmAcctngTransmittedClp0Cells
     17   atmAcctngReceivedCells
     18   atmAcctngReceivedClp0Cells
     19   atmAcctngTransmitTrafficDescriptorType
     20   atmAcctngTransmitTrafficDescriptorParam1
     21   atmAcctngTransmitTrafficDescriptorParam2
     22   atmAcctngTransmitTrafficDescriptorParam3
     23   atmAcctngTransmitTrafficDescriptorParam4
     24   atmAcctngTransmitTrafficDescriptorParam5
     25   atmAcctngReceiveTrafficDescriptorType
     26   atmAcctngReceiveTrafficDescriptorParam1
     27   atmAcctngReceiveTrafficDescriptorParam2
     28   atmAcctngReceiveTrafficDescriptorParam3
     29   atmAcctngReceiveTrafficDescriptorParam4
     30   atmAcctngReceiveTrafficDescriptorParam5
     31   atmAcctngCallingPartySubAddress
     32   atmAcctngCalledPartySubAddress
     33   atmAcctngRecordCrc16

4.7.  QoS: RSVP and DIFFSERV

   As we move towards providing more than simple "best effort" connec-
   tivity, there has been a tremendous surge of interest in (and work
   on) protocols to provide managed Quality of Service for Internet ses-
   sions.  This is of particular interest for the provision of "Inte-
   grated Services," i.e. the transport of audio, video, real-time, and

Brownlee, Blount                                               [Page 11]

INTERNET-DRAFT                                                March 2000

   classical data traffic within a single network infrastructure.

   Two approaches to this have emerged so far:

    - the Integrated Services architecture (intserv) [IIS-ARC], with its
      accompanying signaling protocol, RSVP [RSVP-ARC], and RSVP's
      Common Open Policy Service protocol, COPS [RAP-COPS]

    - the Differentiated Services architecture (diffserv) [DSRV-ARC]

   RSVP is a signaling protocol that applications may use to request
   resources from the network.  The network responds by explicitly
   admitting or rejecting RSVP requests.  Certain applications that have
   quantifiable resource requirements express these requirements using
   intserv parameters [IIS-SPEC].

   Diffserv networks classify packets into one of a small number of
   aggregated flows or "classes", based on the diffserv codepoint (DSCP)
   in the packet's IP header.  At each diffserv router, packets are sub-
   jected to a "per-hop behavior" (PHB), which is invoked by the DSCP.
   Since RSVP is purely a requirements signalling protocol it can also
   be used to request connections from a diffserv network [RS-DS-OP].

4.7.1.  RSVP and DIFFSERV Attributes

   A set of parameters for specifying a requested Quality of Service are
   given in [IIS-SPEC]. These have been turned into accounting
   attributes within RTFM [RTFM-NEWA] and within the RSVP MIB [RSVP-

   The RTFM QoS attributes are:

        98      QoSService
        99      QoSStyle
       100      QoSRate
       101      QoSSlackTerm
       102      QoSTokenBucketRate
       103      QoSTokenBucketSize
       104      QoSPeakDataRate
       105      QoSMinPolicedUnit
       106      QoSMaxPolicedUnit

   The RSVP MIB contains a large number of objects, arranged within the
   following sections:

       General Objects
       Session Statistics Table
       Session Sender Table
       Reservation Requests Received Table
       Reservation Requests Forwarded Table
       RSVP Interface Attributes Table
       RSVP Neighbor Table

   The Session tables contain information such as the numbers of senders

Brownlee, Blount                                               [Page 12]

INTERNET-DRAFT                                                March 2000

   and receivers for each session, while the Reservation Requests tables
   contain details of requests handled by the RSVP router.  There are
   too many objects to list here, but many of them could be used for
   accounting.  In particular, RSVP Requests contain the specification
   of the service parameters requested by a user; these, together with
   the actual usage data for the connection make up an accounting record
   for that usage.

5.  ITU-T Documents

5.1.  Q.825: Call Detail Recording

   ITU-T Recommendation Q.825 specifies how CDRs (Call Detail Records)
   are produced and managed in Network Elements for POTS, ISDN and IN
   (Intelligent Networks).

   Uses of Call Detail information for various purposes are discussed.

   Each call produces one or more records describing events that
   occurred during the life of a call.  Data may be produced in real
   time (single CDRs), near real-time (blocks of CDRs), or as batch
   files of CDRs.

   The information model for Call Detail Recording is formally described
   in terms of an Entity-Relationship model, and an object model speci-
   fied in terms of GDMO templates (Guidelines for the Definition of
   Managed Objects).  Note that this model includes the ways in which
   CDRs are transported from the (NE) Network Element where they are
   generated to the OS (Operations System) where they are used.

5.2.  Q.825 Attributes

   The following attributes are defined.  The explanations given are
   very brief summaries only, see [Q-825] for the complete text.

     1  accessDelivery
          Indicates that the call was delivered to the called subscriber

     2  accountCodeInput
          Account code (for billing), supplied by subscriber.

    78  additionalParticipantInfo
          (No details given)

     5  b-PartyCategory
          Subscriber category for called subscriber.

     4  bearerService
          Bearer capability information (only for ISDN calls).

    13  cDRPurpose
          Reason for triggering this Call Data Record.

Brownlee, Blount                                               [Page 13]

INTERNET-DRAFT                                                March 2000

    70  callDetailDataId
          Unique identifier for the CallDetailData object.

    79  callDuration
          Duration of call

     6  callIdentificationNumber
          Identification number for call; all records produced for this
          call have the same callIdenfificationNumber.

    73  callStatus
          Identifies whether the call was answered or not.

     9  calledPartyNumber
          Telephone number of the called subscriber (may be a
          "diverted-to" or "translated" number.

     7  callingPartyCategory
          Calling subscriber category.

     8  callingPartyNumber
          Telephone number of the calling party.

    10  callingPartyNumberNotScreened
          An additional, user-provided (not screened) number to the
          calling party.

    11  callingPartyType
          Calling subscriber type.

    74  carrierId
          Carrier ID to which the call is sent.

    12  cause
          Cause and location value for the termination of the call.

    14  chargedDirectoryNumber
          Charged directory number (where the charged participant
          element can't indicate the number).

    16  chargedParticipant
          Participant to be charged for the usage.

    15  chargingInformation
          Charging information generated by a Network Element which is
          capable of charging.

    17  configurationMask
          Time consumption, e.g. from B-answer to termination time,
          between partial call records, etc.

    18  conversationTime
          Time consumption from B-answer to end of call.

Brownlee, Blount                                               [Page 14]

INTERNET-DRAFT                                                March 2000

    19  creationTriggerList
          List of trigger values which will create Call Detail data

    75  dPC
          Destination point code (for analysis purposes).

    20  dataValidity
          Indicates that the NE is having problems, contents of the
          generated Call Detail record is not reliable.

    23  durationTimeACM
          Time consumption from seizure until received ACM.

    21  durationTimeB-Answer
          Time consumption from seizure until B-answer.

    22  durationTimeNoB-Answer
          Time from seizure to termination when no B-answer was

    25  exchangeInfo
          Identity of exchange where Call Detail record was generated.

    26  fallbackBearerService
          Fallback bearer capability information for a call.

    27  glare
          Indicates if a glare condition was encountered.

    31  iNServiceInformationList
          Contains information about the use of IN (Intelligent Network)

    32  iNSpecificInformation
          Contains information about the use of one IN service.

    33  iSUPPreferred
          Indicate whether an ISUP preference was requested.

    28  immediateNotificationForUsageMetering
          Indicates that the Call Detail records requires
          immediate data transfer to the Operations System.

    34  maxBlockSize
          Maximum number of Call Detail records in a block.

    35  maxTimeInterval
          Maximum latency allowable for near-real-time Call Detail
          data delivery.

    36  networkManagementControls
          Indicates which Traffic Management Control has affected
          the call.

Brownlee, Blount                                               [Page 15]

INTERNET-DRAFT                                                March 2000

    37  networkProviderId
          Indicates the Network Provider for whom the CDR is generated.

    76  oPC
          Originating point code for a failed call (for analysis

    38  operatorSpecific1AdditionalNumber
    40  operatorSpecific2AdditionalNumber
    42  operatorSpecific3AdditionalNumber
          Operator-defined additional participant information.

    39  operatorSpecific1Number
    41  operatorSpecific2Number
    43  operatorSpecific3Number
          Operator-defined participant information.

    44  originalCalledNumber
          Telephone number of the original called party.

    45  partialGeneration
          Included if the CDR (Call Detail record) output is partial.
          Such CDRs have a field indicating their partial record number.

    77  participantInfo
          (No details given).

    46  percentageToBeBilled
          Percentage to be billed when normal billing rules are
          not to be followed.

    47  periodicTrigger
          Defines the intervals at which the CDR file should be created.

    48  personalUserId
          Internationally unique personal User Identity (for UPT calls).

    49  physicalLineCode
          Identifies the call subscriber's physical line.

    50  progress
          Describes an event which occurred during the life of a call.

    51  queueInfo
          Used to record usage of queueing resources with IN calls.

    52  receivedDigits
          The digits dialed by the subscriber.  (Normally only included
          for customer care purposes).

    53  recordExtensions
          Information elements added by network operators and/or
          manufacturers in addition to the standard ones above.

Brownlee, Blount                                               [Page 16]

INTERNET-DRAFT                                                March 2000

6.  Other Documents

6.1.  TIPHON: ETSI TS 101 321

   TIPHON [TIPHON] is an XML-based protocol, carried by HTTP, which han-
   dles accounting and authorization requests and responses.

   The following are elements selected from TIPHON's DTD that are used
   for accounting.

   <!ELEMENT Currency (#PCDATA)> <!ELEMENT Amount (#PCDATA)>
       Identifies a numeric value.  Expressed using the period (.) as a
       decimal separator with no punctuation as the thousands separator.

   <!ELEMENT CallId (#PCDATA)>
       Contains a call's H.323 CallID value, and is thus used to
       uniquely identify individual calls.

   <!ELEMENT Currency (#PCDATA)>
       Defines the financial currency in use for the parent element.

   <!ELEMENT DestinationInfo type ( e164 | h323 | url | email |
                                    transport | international |
                                    national | network | subscriber |
                                    abbreviated | e164prefix )
       Gives the primary identification of the destination for a call.

   <!ELEMENT Increment (#PCDATA)>
       Indicates the number of units being accounted.

   <!ELEMENT Service EMPTY>
       Indicates a type of service being priced, authorized, or
       reported.  An empty Service element indicates basic Internet
       telephony service, which is the only service type defined by
       V1.4.2 of the specification.  The specification notes that "Later
       revisions of this standard are expected to specify more enhanced
       service definitions to represent quality of service,
       availability, payment methods, etc."

   <!ELEMENT DestinationInfo type ( e164 | h323 | url | email |
                                    transport | international |
                                    national | network | subscriber |
                                    abbreviated | e164prefix)
       Gives the primary identification of the source of a call.

   <!ELEMENT Timestamp (#PCDATA)>
       A restricted form of [ISO-DATE] that indicates the time at which
       the component was generated.

   <!ELEMENT TransactionId (#PCDATA)>
       Contains an integer, decimal valued identifier assigned to a
       specific authorized transaction.

Brownlee, Blount                                               [Page 17]

INTERNET-DRAFT                                                March 2000

   <!ELEMENT Unit (#PCDATA)>
       Indicates the units by which pricing is measured or usage
       recorded.  It shall contain one of the following values:
           s      seconds
           p      packets (datagrams)
           byte   bytes

   <!Element UsageDetail ( Service, Amount, Increment, Unit ) >
       Collects information describing the usage of a service.

6.2.  MSIX

   MSIX [MSIX-SPEC] is an XML-based protocol transported by HTTP that is
   used to make accounting service definitions and transmit service
   usage information.  As its service definitions are parameterized and
   dynamic, it makes no definition of services or attributes itself, but
   allows implementors to make their own.  It specifies only the base
   data types that attributes may take: STRING, UNISTRING, INT32, FLOAT,

7.  Accounting File and Record Formats

7.1.  ASN.1 Records

7.1.1.  RTFM and AToMMIB

   RTFM and AToMMIB use ASN.1 Basic Encoding Rules (BER) to encode lists
   of attributes into accounting records.  RTFM uses SNMP to retrieve
   such records as BER strings, thus avoiding having to have an object
   identifier for every object.

   AToMMIB carries this a stage further by defining an accounting file
   format in ASN.1 and making it available for retrieval by a file
   transfer protocol, thereby providing a more efficient alternative to
   simply retrieving the records using SNMP.

7.1.2.  Q.825

   A Q.825 Call Record is an ASN.1 SET containing a specified group of
   the Q.825 attributes.  Call records would presumably be encoded as
   BER strings before being collected for later processing.

7.2.  Binary Records

7.2.1.  RADIUS

   Radius packets carry a sequence of attributes and their values, as
   (Type, Length, Value) triples.  The format of the value field is one
   of four data types.

       string   0-253 octets

Brownlee, Blount                                               [Page 18]

INTERNET-DRAFT                                                March 2000

       address  32 bit value, most significant octet first.

       integer  32 bit value, most significant octet first.

       time     32 bit value, most significant octet first -- seconds
                since 00:00:00 GMT, January 1, 1970.  The standard
                Attributes do not use this data type but it is presented
                here for possible use within Vendor-Specific attributes.

7.2.2.  DIAMETER

   Each DIAMETER message consists of multiple AVP's that are 32-bit
   aligned, with the following format:

      0                   1                   2                   3
      0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
      |                           AVP Code                            |
      |          AVP Length           |     Reserved        |P|T|V|R|M|
      |                        Vendor ID (opt)                        |
      |                           Tag (opt)                           |
      |    Data ...

           The AVP Code identifies the attribute uniquely.  If the Vendor-
           Specific bit is set, the AVP Code is allocated from the vendor's
           private address space.

           The first 256 AVP numbers are reserved for backward
           compatibility with RADIUS and are to be interpreted as per
           RADIUS [RAD-PROT]. AVP numbers 256 and above are used for
           DIAMETER, which are allocated by IANA.

        AVP Length
           A 16-bit field contains the total object length in bytes.
           Must always be a multiple of 4, and at least 8.

        AVP Flags
           P                      Protected bit
           T                      Tag bit
           V                      Vendor-ID bit
           R                      Reserved (MUST be set to 0)
           M                      Mandatory bit

7.3.  Text Records

Brownlee, Blount                                               [Page 19]

INTERNET-DRAFT                                                March 2000

7.3.1.  ROAMOPS

   ADIF (Accounting Data Interchange Format [ROAM-ADIF]) presents a gen-
   eral, text-based format for accounting data files, described in a
   straightforward BNF grammar.  Its file header contains a field indi-
   cating the default protocol from which accounting attributes are
   drawn.  If an attribute from another protocol is to be used, it is
   preceded by its protocol name, for example rtfm//27 would be RTFM's
   "forward bytes" attribute.  Comments in an ADIF file begin with a

   Example: An ADIF file encoding RADIUS accounting data

        version: 1
        device: server3
        description: Accounting Server 3
        date: 02 Mar 1998 12:19:01 -0500
        defaultProtocol: radius

        5: 12
        61: 2
        40: 2
        41: 14
        42: 234732
        43: 15439
        44: 185
        45: 1
        46: 1238
        47: 153
        48: 148
        49: 11
        50: 73
        51: 2

Brownlee, Blount                                               [Page 20]

INTERNET-DRAFT                                                March 2000

8.  AAA Requirements

8.1.  A Well-Defined Set of Attributes

   AAA needs a well-defined set of attributes whose values are to be
   carried in records to or from accounting servers.

   Most of the existing sets of documents described above include a set
   of attributes, identified by small integers.  It is likely that these
   sets overlap, i.e. that some of them have attributes which represent
   the same quantity using different names in different sets.  This sug-
   gests it might be possible to produce a single combined set of "uni-
   versal" accounting attributes, but such a "universal" set does not
   seem worthwhile.

   The ADIF approach of specifying a default protocol (from which
   attributes are assumed to come) and identifying any exceptions seems
   much more practical.  We therefore propose that AAA should use the
   ADIF convention (or something like it) to identify attributes,
   together with all the sets of attributes covered by the [ASG-NBR]

8.2.  A Simple Interchange Format

   AAA needs a simple interchange file format, to be used for accounting
   data.  Several schemes for packaging and transporting such data have
   been described above.

   The SNMP-based ones fit well within the context of an SNMP-based net-
   work management system.  RTFM and AToMMIB provide ways to reduce the
   SNMP overhead for collecting data, and AToMMIB defines a complete
   file format.  Both provide good ways to collect accounting data.

   As an interchange format, however, ASN.1-based schemes suffer from
   being rather complex binary structures.  This means that one requires
   suitable tools to work with them, as compared to plain-text files
   where one can use existing text-based utilities.

   The binary schemes such as RADIUS and DIAMETER have simpler struc-
   tures, but they too need purpose-built tools.  For general use they
   would need to be extended to allow them to use attributes from other

   From the point of view of being easy for humans to understand, ADIF
   seems very promising.  Of course any processing program would need a
   suitable ADIF input parser, but using plain-text files makes them
   much easier to understand.

   TIPHON's record format is specified by an XML DTD.  While XML repre-
   sentations have the advantages of being well-known, they are limited
   by XML's inability to specify type or other validity checking for
   information within the tags.  This situation will likely be improved
   by the XML Schema [XML-SCHM] efforts that are underway, but a stable

Brownlee, Blount                                               [Page 21]

INTERNET-DRAFT                                                March 2000

   reference is not yet available.

9.  Issues

   It is generally agreed that there is a need for a standard record
   format and transport protocol for communication between Service Ele-
   ments and Accounting Servers.

   There is less agreement on the following issues:

      o Separate or integral record format and transport protocol
      o Standard set of base data types
      o Service definitions: part of the protocol or separately defined
      o Service definition namespace management

   The following sections address these issues.

9.1.  Record Format vs. Protocol

   All known Internet-centric billing protocols to date have an integral
   record format.  That is, the collection of Properties that describe a
   Usage Event are specified as an integral part of the protocol, typi-
   cally as a part of a "submit" message that is used to transmit a
   Usage Event from a Service Entity to an Accounting Server.

   It may be advantageous to define a record format that is independent
   of the transport protocol.  Such a record format should support both
   representation of individual records and records in bulk, as Usage
   Events are often aggregated and transmitted in bulk.

   A separate record format is useful for record archiving and temporary
   file storage.  Multiple transport protocols may be defined without
   affecting the record format.  The task of auditing is made easier if
   a standard file format is defined.  If a canonical format is used,
   bulk records may be hashed with MD5 [MD5] or a similar function, for
   reliability and security purposes.

                                    |  transport |
                                    |   header   |
              +------------+        +------------+
              |            |        |            |
              |   Usage    |        |   Usage    |
              |  Event(s)  |        |  Event(s)  |
              |            |        |            |
              |            |        |            |
              +------------+        +------------+
                                    |  trailer   |

              record format       transport protocol

   If the protocol is written such that it can transmit Usage Events in
   the record format, no record rewriting for transport is required.

Brownlee, Blount                                               [Page 22]

INTERNET-DRAFT                                                March 2000

9.2.  Tagged, Typed Data

   Record formats and protocols use a combination of data locality and
   explicit tagging to identify data elements.  Mail [RFC822], for
   instance, defines a header block composed of several Attribute-Value
   Pairs, followed by a message body.  Each header field is explicitly
   tagged, but the order of the AVPs is undefined.  The message body is
   not tagged (except with an additional preceding blank line), and is
   found through its position in the message, which must be after all
   header fields.

   Some record formats make no use of tags--data elements are identified
   only by their position within a record structure.  While this prac-
   tice provides for the least amount of record space overhead, it is
   difficult to later modify the record format by adding or removing
   elements, as all record readers will have to be altered to handle the
   change.  Tagged data allows old readers to detect unexpected tags and
   to detect if required data are missing.  If the overhead of carrying
   explicit tags can be borne, it is advantageous to use explicitly
   tagged data elements where possible.

   An AVP approach has proven useful in accounting.  RADIUS [RADIUS]
   uses numeric data type identifiers.  ETSI's TIPHON [TIPHON] uses XML

   For an AAA accounting record format, the authors suggest that each
   Property be named by a textual or numeric identifier and carry a
   value and a data type indicator, which governs interpretation of the
   value.  It may also be useful for each Property to carry a units of
   measure identifier.  The TIPHON specification takes this approach.
   TS 101 321 also carries an Increment field, which denominates the
   Property's Unit of Measure field.  Whether this additional conve-
   nience is necessary is a matter for discussion.

   It is not strictly necessary for each data record to carry data type,
   units of measure, or increments identifiers.  If this information is
   recorded in a record schema document that is referenced by each data
   record, each record may be validated against the schema without the
   overhead of carrying type information.

9.2.1.  Standard Type Definitions

   It is useful to define a standard set of primitive data types to be
   used by the record format and protocol.  Looking at the prior art,
   DIAMETER supports Data (arbitrary octets), String (UTF-8), Address
   (32 or 128 bit), Integer32, Integer64, Time (32 bits, seconds since
   1970), and Complex.  MSIX [MSIX-SPEC] supports String, Unistring,
   Int32, Float, Double, Boolean, and Timestamp.  SNMP [SNMP] offers
   Counter, Gauge, TimeTicks, and Opaque.

   An appropriate set would likely include booleans, 32 and 64 bit
   signed integers, 32 and 64 bit floats, arbitrary octets, UTF-8 and
   UTF-16 strings, and ISO 8601:1988 [ISO-DATE] timestamps.  Fixed-

Brownlee, Blount                                               [Page 23]

INTERNET-DRAFT                                                March 2000

   precision numbers capable of representing currency amounts (with pre-
   cision specified on both sides of the decimal point) have proven use-
   ful in accounting record formats, as they are immune to the precision
   problems that are encountered when one attempts to represent fixed-
   point amounts with floating point numbers.

9.3.  Transaction Identifiers

   Each Usage Event requires its own unique identifier.

   It is expedient to allow Service Elements to create their own unique
   identifiers.  In this manner, Usage Events can be created and
   archived without the involvement of an Accounting Server or other
   central authority.

   A number of methods for creating unique identifiers are well known.
   One popular identifier is an amalgamation of a monotonically increas-
   ing sequence number, a large random value, a network element identi-
   fier, and a timestamp.  Another possible source of entropy is a hash
   value of all or part of the record itself.

   RFC 822 [MAIL], RFC 1036 [NEWS], and RFC 2445 [ICAL-CORE] give guid-
   ance on the creation of good unique identifiers.

9.4.  Service Definitions

   A critical differentiator in accounting record formats and protocols
   is their capability to account for arbitrary service usage.  To date,
   no accounting record format or protocol that can handle arbitrary
   service definitions has achieved broad acceptance on the Internet.

   This section analyzes the issues in service definition and makes a
   case for a record format and protocol with the capability to carry
   Usage Events for rich, independently-defined services.

9.4.1.  Service Independence

   It is informative to survey a number of popular Internet protocols
   and document encodings and examine their capacities for extension.
   These protocols can be categorized into two broad categories--"fully
   specified" protocols that have little provision for extension and
   "framework" protocols that are incomplete, but provide a basis for
   future extension when coupled with application documents.

   Examples of fully-specified protocols are NTP [NTP], NNTP [NNTP],
   RADIUS Accounting [RAD-ACT], and HTML [HTML].

   Aside from leaving some field values "reserved for future use," all
   of Network Time Protocol's fields are fixed-width and completely
   defined.  This is appropriate for a simple protocol that solves a
   simple problem.

   Network News Transfer Protocol [NEWS-PROT] specifies that further
   commands may be added, and requests that non-standard implementations

Brownlee, Blount                                               [Page 24]

INTERNET-DRAFT                                                March 2000

   use the "X-" experimental prefix so as to not conflict with future
   additions.  The content of news is 7-bit data, with the high-order
   bit cleared to 0.  Nothing further about the content is defined.
   There is no in-protocol facility for automating decoding of content

   We pay particular attention to RADIUS Accounting [RAD-ACT].  Perhaps
   the second most frequently heard complaint (after security shortcom-
   ings) about RADIUS Accounting is its preassigned and fixed set of
   "Types".  These are coded as a range of octets from 40 to 51 and are
   as follows:

         40      Acct-Status-Type
         41      Acct-Delay-Time
         42      Acct-Input-Octets
         43      Acct-Output-Octets
         44      Acct-Session-Id
         45      Acct-Authentic
         46      Acct-Session-Time
         47      Acct-Input-Packets
         48      Acct-Output-Packets
         49      Acct-Terminate-Cause
         50      Acct-Multi-Session-Id
         51      Acct-Link-Count

   These identifiers were designed to account for packet-based network
   access service.  They are ill-suited for describing other services.
   While extension documents have specified additional types, the base
   protocol limits the type identifier to a single octet, limiting the
   total number of types to 256.

   HTML/2.0 [HTML] is mostly a fully-specified protocol, but with W3C's
   HTML/4.0, HTML is becoming more of a framework protocol.  HTML/2.0
   specified a fixed set of markups, with no provision for addition
   (without protocol revision).

   Examples of "framework" protocols and document encodings are HTTP,
   XML, and SNMP.

   HTTP/1.1 [HTTP] is somewhat similar to NNTP in that it is designed to
   transport arbitrary content.  It is different in that it supports
   description of that content through its Content-Type, Content-Encod-
   ing, Accept-Encoding, and Transfer-Encoding header fields.  New types
   of content can be designated and carried by HTTP/1.1 without modifi-
   cation to the HTTP protocol.

   XML [XML] is a preeminent general-purpose framework encoding.  DTD
   publishing is left to users.  There is no standard registry of DTDs.

   SNMP presents a successful example of a framework protocol.  SNMP's
   authors envisioned SNMP as a general management protocol, and allow
   extension through the use of private MIBs.  SNMP's ASN.1 MIBs are
   defined, published, and standardized without the necessity to modify
   the SNMP standard itself.  From "An Overview of SNMP" [SNMP-OVER]:

Brownlee, Blount                                               [Page 25]

INTERNET-DRAFT                                                March 2000

      It can easily be argued that SNMP has become prominent mainly from
      its ability to augment the standard set of MIB objects with new
      values specific for certain applications and devices.  Hence, new
      functionality can continuously be added to SNMP, since a standard
      method has been defined to incorporate that functionality into
      SNMP devices and network managers.

   Most accounting protocols are fully-specified, with either a com-
   pletely defined service or set of services (RADIUS Accounting) or
   with one or more services defined and provision for "extension" ser-
   vices to be added to the protocol later (TIPHON).  While the latter
   is preferable, it may be preferable to take a more SNMP-like
   approach, where the accounting record format and protocol provide
   only a framework for service definition, and leave the task of ser-
   vice definition (and standardization) to separate efforts.  In this
   manner, the accounting protocol itself would not have to be modified
   to handle new services.

9.4.2.  Versioned Service Definitions

   Versioning is a naming and compatibility issue.  Version identifiers
   are useful in service definition because they enable service defini-
   tions to be upgraded without a possibly awkward name change.  They
   also enable possible compatibility between different versions of the
   same service.

   An example could be the service definition of a phone call.  Version
   1 might define Properties for the start time, duration, and called
   and calling party numbers.  Later, version 2 is defined, which aug-
   ments the former service definition with a byte count.  An Accounting
   Server, aware only of Version 1, may accept Version 2 records, dis-
   carding the additional information (forward compatibility).  Alter-
   nately, if an Accounting Server is made aware of version 2, it could
   optionally still accept version 1 records from Service Elements, pro-
   vided the Accounting Sever does not require the additional informa-
   tion to properly account for service usage (backward compatibility).

9.4.3.  Relationships Among Usage Events

   Accounting record formats and protocols to date do not sufficiently
   addressed "compound" service description.

   A compound service is a service that is described as a composition of
   other services.  A conference call, for example, may be described as
   a number of point-to-point calls to a conference bridge.  It is
   important to account for the individual calls, rather than just sum-
   ming up an aggregate, both for auditing purposes and to enable dif-
   ferential rating.  If these calls are to be reported to the Account-
   ing Server individually, the Usage Events require a shared identifier
   that can be used by the Accounting Server and other backend systems
   to group the records together.

   In order for a Service Element to report compound events over time as
   a succession of individual Usage Events, the accounting protocol

Brownlee, Blount                                               [Page 26]

INTERNET-DRAFT                                                March 2000

   requires a facility to communicate that the compound event has
   started and stopped.  The "start" message can be implicit--the trans-
   mission of the first Usage Event will suffice.  An additional
   semaphore is required to tell the Accounting Server that the compound
   service is complete and may be further processed.  This is necessary
   to prevent the Accounting Server from prematurely processing compound
   events that overlap the end of a billing period.

   RADIUS Accounting has some provision for this sort of accounting with
   its "Acct-Multi-Session-Id" field.  Unfortunately, RADIUS Account-
   ing's other shortcomings preclude it from being used in general pur-
   pose service usage description.

9.4.4.  Service Namespace Management

   "Framework" protocols, as previously mentioned, do not define com-
   plete schema for their payload.  For interoperability to be achieved,
   it must be possible for:

      (1) content definers to specify definitions without conflicting with
          the names of other definitions

      (2) protocol users to find and use content definitions

   Condition (1) can be readily managed through IANA assignment or by
   using an existing namespace differentiator (for example, DNS).

   Condition (2) is harder, and places considerable burden on the imple-
   mentors.  Their clients and servers must be able, statically or
   dynamically, to find and validate definitions, and manage versioning

   As previously mentioned, the XML specification provides no facility
   for DTD discovery or namespace management.  XML specifies only a doc-
   ument format, and as such does not need to specify support for more
   "protocol" oriented problems.

   For an accounting record format and protocol, an approach closer to
   SNMP's is useful.  SNMP uses an ISO-managed dotted-decimal namespace.
   An IANA-managed registry of service types is a possibility.  Another
   possibility, used by MSIX [MSIX-SPEC], is for Service Element cre-
   ators to identify their services by concatenation of a new service
   name with existing unique identifier, such as a domain name.

   A standard record format for service definitions would make it possi-
   ble for Service Element creators to directly supply accounting system
   managers with the required definitions, via the network or other

10.  Encodings

   It may be useful to define more than one record encoding.

   A "verbose" XML encoding is easily implemented and records can be

Brownlee, Blount                                               [Page 27]

INTERNET-DRAFT                                                March 2000

   syntactically verified with existing tools.  "Human-readable" proto-
   cols tend to have an edge on "bitfield" protocols where ease of
   implementation is paramount and the application can tolerate any
   additional processing required to generate, parse, and transport the

   A alternative "compressed" encoding that makes minimal use of storage
   and processing may be useful in many contexts.

   There are disadvantages to supporting multiple encodings.  Option-
   ally-supported multiple encodings mandate the requirement for capa-
   bilities exchange between Service Element and Accounting Server.
   Also, implementations can tend to "drift apart," with one encoding
   better-supported than another.  Unless all encodings are mandatory,
   implementors may find they are unable to interoperate because they
   picked the wrong encoding.

11.  Security Considerations

   This draft summarises many existing IETF and ITU documents; please
   refer to the original documents for security considerations for their
   particular protocols.

   It must be possible for the accounting protocol to be carried by a
   secure transport.  A canonical record format is useful so that regen-
   eration of secure record hashes is possible.

   When dealing with accounting data files, one must take care that
   their integrity and privacy are preserved.  This document, however,
   is only concerned with the format of such files.

12.  References

   [ACC-BKG]   Mills, C., Hirsch, G. and Ruth, G., "Internet
               Accounting Background", RFC 1272, November 1991.

   [ASG-NBR]   Reynolds, J., Postel, J., "Assigned Numbers,"
               RFC 1700, October 1994.

   [ASN1]      Information processing systems - Open Systems
               Interconnection - Specification of Abstract Syntax Notation
               One (ASN.1), International Organization for Standardization,
               International Standard 8824, December 1987.

   [ATM-ACT]   McCloghrie, K., Heinanen, J., Greene, W. and Prasad,
               A., "Accounting Information for ATM Networks,"
               RFC 2512, February 1999.

   [ATM-COLL]  McCloghrie, K., Heinanen, J., Greene, W. and Prasad,
               A., " Managed Objects for Controlling the Collection
               and Storage of Accounting Information for Connection-
               Oriented Networks," RFC 2513, February 1999.

   [BER]       Information processing systems - Open Systems

Brownlee, Blount                                               [Page 28]

INTERNET-DRAFT                                                March 2000

               Interconnection - Specification of Basic Encoding Rules for
               Abstract Notation One (ASN.1), International Organization for
               Standardization, International Standard 8825, December 1987.

   [DIAM-ACT]  Arkko, J., Calhoun, P.R., Patel, P. and Zorn, G.,
               "DIAMETER Accounting Extension," Internet Draft (Work
               in progress), draft-calhoun-diameter-accounting- ..

   [DIAM-AUTH] Calhoun, P.R. and Bulley, W., "DIAMETER User
               Authentication Extensions," Internet Draft (Work in
               progress), draft-calhoun-diameter-authent- ..

   [DIAM-FRAM] Calhoun, P.R., Zorn, G. and Pan, P., "DIAMETER
               Framework Document," Internet Draft (Work in
               progress), draft-calhoun-diameter-framework- ..

   [DSRV-ARC]  Blake, S., Black, D., Carlson, M., Davies, E., Wang, Z.
               and W. Weiss, "An Architecture for Differentiated
               Services", RFC 2475, December 1998.

   [HTML]      Berners-Lee, T., D. Connolly, "Hypertext Markup Language -
               2.0", RFC 1866, November 1995.

   [HTTP]      Fielding, R., J. Gettys, J. Mogul, H. Frystyk, and
               T. Berners-Lee. "Hypertext Transfer Protocol--HTTP/1.1",
               RFC 2068, January 1997.

   [ICAL-CORE] Dawson, F., D. Stenerson, "Internet Calendaring and
               Scheduling Core Object Specification", RFC 2445, November

   [IIS-ARC]   Braden, R., Clark, D. and Shenker, S., "Integrated
               Services in the Internet Architecture: an Overview",
               RFC 1633, June 1994

   [IIS-SPEC]  Shenker, S., Partridge, C., Guerin, R.: "Specification of
               Guaranteed Quality of Service," RFC 2212, 1997.

   [ISDN-MIB]  Roeck, G., "ISDN Management Information Base using
               SMIv2," RFC 2127, March 1997.

   [ISO-DATE]  "Data elements and interchange formats -- Information
               interchange -- Representation of dates and times",
               ISO 8601:1988.

               MESSAGES", RFC 822, August 1982.

   [MD5]       Rivest, R., "The MD5 Message-Digest Algorithm", RFC 1321,
               April 1992.

   [MSIX-SPEC] Blount, A., D. Young, "Metered Service Information Exchange
               1.2", Work in Progress,

Brownlee, Blount                                               [Page 29]

INTERNET-DRAFT                                                March 2000

               July 1999.

   [NEWS-MSGS] Horton, M., R. Adams, "Standard for Interchange of USENET
               Messages", RFC 1036, December 1987.

   [NEWS-PROT] Kantor, B., P. Lapsley, "Network News Transfer Protocol"
               RFC 977, February 1986.

   [NTP]       Mills, D.L., "Network Time Protocol (NTP)", RFC 958,
               September 1985.

   [Q-825]     "Specification of TMN applications at the Q3
               interface: Call detail recording,"
               ITU-T Recommendation Q.825, 1998.

   [RAD-ACT]   Rigney, C., "RADIUS Accounting," RFC 2139, April 1997.

   [RAD-EXT]   Rigney, C., Willats, W. and Calhoun, P., "RADIUS
               Extensions," Internet Draft (Work in progress),
               draft-ietf-radius-ext- ..

   [RAD-PROT]  Rigney, C., Rubens, A., Simpson, W. and Willens, S.,
               "Remote Authentication Dial In User Service (RADIUS),"
               RFC 2138, April 1997.

   [RAD-RECX]  Calhoun, P.R. and Beadles, M.., "RADIUS Accounting
               Interim Accounting Record Extension," Internet Draft
               (Work in progress), draft-ietf-radius-acct-interim- ..

   [RAD-TACC]  Zorn, G., Mitton, D. and Aboba, A.,"RADIUS Accounting
               "Modifications for Tunnel Protocol Support,"
               (Work in progress), draft-ietf-radius-tunnel-acct- ..

   [RAP-COPS]  Boyle, J., Cohen, R., Durham, D., Herzog, S., Rajan, R.
               and Sastry, A., "The COPS (Common Open Policy Service)
               draft-ietf-rap-cops- ..

   [ROAM-ADIF] Aboba, B and Lidyard, D., "The Accounting Data
               Interchange Format (ADIF)," Internet Draft  (Work
               in progress), draft-ietf-roamops-actng- ..

   [ROAM-IMPL] Aboba, B., Lu J., Alsop J.,Ding J., and W. Wang,
               "Review of Roaming Implementations",
               RFC 2194, September 1997.

   [RS-DS-OP]  Bernet, Y., Yavatkar, R., Ford, P., Baker, F., Zhang, L.,
               Speer, M. and Braden, R., "Interoperation of RSVP/Intserv
               and Diffserv Networks," Internet Draft (Work in Progress),
               draft-ietf-issll-diffserv-rsvp- ..

   [RSVP-ARC]  Braden, R., Zhang, L., Berson, S., Herzog, S. and
               Jamin, S., "Resource Reservation Protocol (RSVP) Version 1
               Functional Specification", RFC 2205, September 1997

Brownlee, Blount                                               [Page 30]

INTERNET-DRAFT                                                March 2000

   [RSVP-MIB]  Baker, F., Krawczyk, J. and Sastry, A., "RSVP Management
               Information Base," Internet Draft (Work in Progress),
               draft-ietf-rsvp-mib-v2- ..

   [RTFM-ARC]  Brownlee, N., Mills, C., and Ruth, G., "Traffic Flow
               Measurement: Architecture", RFC 2722, October 1999.

   [RTFM-MIB]  Brownlee, N., "Traffic Flow Measurement: Meter MIB",
               Measurement: Architecture", RFC 2720, October 1999.

   [RTFM-NEWA] Handelman, S, Brownlee, N., Ruth, G., and Stibler, S.,
               "New Attributes for Traffic Flow Measurement",
               RFC 2724, October 1999.

   [SIP-PROT]  Handley, M.,Schulzrinne, H.,Schooler, E. and
               Rosenberg, J., "SIP: session initiation protocol,"
               RFC 2543, March 1999.

   [SNMP]      Case, J., M. Fedor, M. Schoffstall, J. Davin, "A Simple
               Network Management Protocol (SNMP)", RFC 1157, May 1990

   [SNMP-OVER] "AN OVERVIEW OF SNMP V2.0", Diversified Data Resources, Inc.,
     , 1999.

   [TIPHON]    "Telecommunications and Internet Protocol Harmonization Over
               Networks (TIPHON); Inter-domain pricing, authorization, and
               usage exchange", TS 101 321 V1.4.2, December 1998.

   [XML]       Bray, T., J. Paoli, and C. Sperberg-McQueen, "Extensible
               Markup Language (XML) 1.0", W3C Recommendation, February

   [XML-SCHM]  "XML Schema Part 1: Structures", W3C Working Draft 17,
               December 1999.

13.  Authors' Addresses

   Alan Blount
   MetraTech Corp.
   330 Bear Hill Road
   Waltham, MA 02451

   Nevil Brownlee
   Information Technology Systems & Services
   The University of Auckland
   Phone: +64 9 373 7599 x8941