RFC 3838






Network Working Group                                          A. Barbir
Request for Comments: 3838                               Nortel Networks
Category: Informational                                       O. Batuner
                                                              Consultant
                                                                 A. Beck
                                                     Lucent Technologies
                                                                 T. Chan
                                                                   Nokia
                                                                H. Orman
                                               Purple Streak Development
                                                             August 2004


          Policy, Authorization, and Enforcement Requirements
               of the Open Pluggable Edge Services (OPES)

Status of this Memo



   This memo provides information for the Internet community.  It does
   not specify an Internet standard of any kind.  Distribution of this
   memo is unlimited.

Copyright Notice



   Copyright (C) The Internet Society (2004).

Abstract



   This document describes policy, authorization, and enforcement
   requirements for the selection of the services to be applied to a
   given Open Pluggable Edge Services (OPES) flow.




















Barbir, et al.               Informational                      [Page 1]

RFC 3838                OPES Policy Requirements             August 2004


Table of Contents



   1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  3
   2.  Terminology  . . . . . . . . . . . . . . . . . . . . . . . . .  3
   3.  Policy Architecture  . . . . . . . . . . . . . . . . . . . . .  4
       3.1.  Policy Components and Functions  . . . . . . . . . . . .  4
       3.2.  Requirements for Policy Decision Points. . . . . . . . .  5
       3.3.  Requirements for Policy Enforcement Points . . . . . . .  5
   4.  Requirements for Interfaces  . . . . . . . . . . . . . . . . .  6
       4.1.  Service Bindings Requirements  . . . . . . . . . . . . .  7
             4.1.1.  Environment Variables  . . . . . . . . . . . . .  7
             4.1.2.  Requirements for Using State Information . . . .  8
             4.1.3.  Requirements for Passing Information Between
                     Services . . . . . . . . . . . . . . . . . . . .  8
       4.2.  Requirements for Rule and Rules Management . . . . . . .  8
             4.2.1.  Requirements for Rule Providers  . . . . . . . .  8
             4.2.2.  Requirements for Rule Formats and Protocols  . .  9
             4.2.3.  Requirements for Rule Conditions . . . . . . . .  9
             4.2.4.  Requirements for Rule Actions  . . . . . . . . .  9
       4.3.  Requirements for Policy Expression . . . . . . . . . . . 10
   5.  Authentication of Principals and Authorization of Services . . 10
       5.1.  End users, Publishers and Other Considerations . . . . . 11
             5.1.1.  Considerations for End Users . . . . . . . . . . 11
             5.1.2.  Considerations for Publishing Sites. . . . . . . 12
             5.1.3.  Other Considerations . . . . . . . . . . . . . . 12
       5.2.  Authentication . . . . . . . . . . . . . . . . . . . . . 12
       5.3.  Authorization  . . . . . . . . . . . . . . . . . . . . . 13
       5.4.  Integrity and Encryption . . . . . . . . . . . . . . . . 14
             5.4.1.  Integrity and Confidentiality of Authentication
                     and Requests/Responses for Service . . . . . . . 14
             5.4.2.  Integrity and Confidentiality of Application
                     Content  . . . . . . . . . . . . . . . . . . . . 14
       5.5.  Privacy. . . . . . . . . . . . . . . . . . . . . . . . . 14
   6.  Security Considerations  . . . . . . . . . . . . . . . . . . . 15
   7.  References . . . . . . . . . . . . . . . . . . . . . . . . . . 15
       7.1.  Normative References . . . . . . . . . . . . . . . . . . 15
       7.2.  Informative References . . . . . . . . . . . . . . . . . 15
   8.  Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 16
   9.  Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 16
   10. Full Copyright Statement . . . . . . . . . . . . . . . . . . . 17











Barbir, et al.               Informational                      [Page 2]

RFC 3838                OPES Policy Requirements             August 2004


1.  Introduction



   The Open Pluggable Edge Services (OPES) [1]  architecture enables
   cooperative application services (OPES services) between a data
   provider, a data consumer, and zero or more OPES processors.  The
   application services under consideration analyze and possibly
   transform application-level messages exchanged between the data
   provider and the data consumer.  The OPES processor can distribute
   the responsibility of service execution by communicating and
   collaborating with one or more remote callout servers.

   The execution of such services is governed by a set of rules
   installed on the OPES processor.  The rule evaluation can trigger the
   execution of service applications local to the OPES processor or on a
   remote callout server.

   Policies express the goals of an OPES processor as a set of rules
   used to administer, manage, and control access to resources.  The
   requirements in this document govern the behavior of OPES entities in
   determining which of the available services are to be applied to a
   given message, if any.

   The scope of OPES policies described in this document are limited to
   those that describe which services to call and, if appropriate, with
   what parameters.  These policies do not include those that prescribe
   the behavior of the called services.  It is desirable to enable a
   common management framework for specifying policies for both the
   calling of and the behavior of a service.  The integration of such a
   function is the domain of policy administration user interaction
   applications.

   The document is organized as follows: Section 2 considers policy
   framework.  Section 3 discusses requirements for interfaces, while
   section 4 examines authentication of principals and authorization of
   services.

2.  Terminology



   The keywords "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in [4].  When used with
   the normative meanings, these keywords will be all uppercase.
   Occurrences of these words in lowercase comprise normal prose usage,
   with no normative implications.







Barbir, et al.               Informational                      [Page 3]

RFC 3838                OPES Policy Requirements             August 2004


3.  Policy Architecture



   This section describes the architectural policy decomposition
   requirements.  It also describes the requirements for the interfaces
   between the policy components.  Many of the rules here were
   determined under the influence of RFC 3238 [2].

3.1.  Policy Components and Functions



   The policy functions are decomposed into three components: a Rule
   Author, a Policy Decision Point (PDP) [6], and a Policy Enforcement
   Point (PEP) [6].  The Rule Author provides the rules to be used by an
   OPES entity.  These rules control the invocation of services on
   behalf of the rule author.  The PDP and the PEP interpret the
   collected rules and appropriately enforce them.  The decomposition is
   illustrated in Figure 1.

         +--------+                         +--------+
         |  Rule  |                         |  Rule  |
         | Author |          ...            | Author |
         +--------+                         +--------+
              |                                 |
              |                                 |
              |          +----------+           |
              |          |  Policy  |           |  <- PDP Interface
              +--------->| Decision |<----------+
                         |  Point   |
                         +----------+
                             | ^
                             | |
                             | |  <- PEP Interface
                             | |
                             V |
                       +--------------+   ...
                  ---> |    Policy    | --->
                       |  Enforcement |       Data Traffic
                  <--- |    Point     | <---
                       +--------------+


                  Figure 1: Policy Components

   The decomposition of policy control into a PDP and a PEP permit the
   offloading of some tasks to an administrative service that may be
   located on a server separate from the real-time enforcement services
   of the PEP that reside on the OPES processor.





Barbir, et al.               Informational                      [Page 4]

RFC 3838                OPES Policy Requirements             August 2004


   The PDP provides for the authentication and authorization of rule
   authors and the validation and compilation of rules.

   The PEP resides in the data filter where the data from an OPES flow
   is evaluated against the compiled rules and appropriate calls to the
   requested services are performed.

   Interfaces between these architectural components are points of
   interoperability.  The interface between rule authors and the policy
   decision points (PDP Interface) MUST use the format that may result
   from the requirements as described in this document.

   The interface between the policy decision points and the policy
   enforcement points (PEP Interface) can be internal to a specific
   vendor implementation of an OPES processor.  Implementations MUST use
   standard interface only if the PDP and the PEP reside on different
   OPES processors.

3.2.  Requirements for Policy Decision Points



   The Policy Decision Point is essentially a policy compiler.  The PDP
   MUST be a service that provides administrative support to the
   enforcement points.  The PDP service MUST authenticate the rule
   authors.

   The PDP MUST verify that the specified rules are within the scope of
   the rule authors authority.  The PDP MUST be a component of the OPES
   Administration Authority.

3.3.  Requirements for Policy Enforcement Points



   In the OPES architecture, the data filter represents a Policy
   Enforcement point (PEP).  At this point, data from an OPES flow is
   evaluated against the compiled rules, and appropriate calls to the
   requested services are performed.

   In the PEP rules MAY chain actions together, where a series of
   services to be called are specified.  Implementation MUST ensure the
   passing of information from one called service to another.
   Implementation MUST NOT prohibit the re-evaluation of a message to
   determine if another service or set of services should be called.

   The execution of an action (i.e., the triggering of a rule) may lead
   to the modification of message property values.  For example, an OPES
   service that under some circumstances converts JPEG images to GIF
   images modifies the content type of the requested web object.





Barbir, et al.               Informational                      [Page 5]

RFC 3838                OPES Policy Requirements             August 2004


   Such modification of message property values may change the behavior
   of subsequently performed OPES actions.  The data filter SHOULD act
   on matched rules before it evaluates subsequent rules.  Multiple
   matched rules can be triggered simultaneously if the data filter can
   determine in advance that there are no side effects from the
   execution of any specific rule.

   A data filter MAY evaluate messages several times in the course of
   handling an OPES flow.  The rule processing points MAY be defined by
   administratively defined names.  The definition of such names can
   serve as a selector for policy rules to determine the applicability
   of a rule or a set of rules at each processing point.

   Policy roles ([5] and [6]) SHOULD be used where they aid in the
   development of the OPES policy model.

   Figure 2 expresses a typical message data flow between a data
   consumer application, an OPES processor, and a data provider
   application.  There are four commonly used processing points
   identified by the numbers 1 through 4.

            +--------+       +-----------+       +---------+
            |        |<------|4         3|<------|         |
            | Data   |       |  OPES     |       | Data    |
            |Consumer|       | Processor |       |Provider |
            |  Appl. |------>|1         2|------>| Appl.   |
            +--------+       +-----------+       +---------+


                 Figure 2: Processing Execution Points

   Any data filter (PEP) or any administrative (PDP) implementation MUST
   support the four rule processing points.

   o  Data Consumer Request handling role: This involves request
      processing when received from a Data Consumer Application.
   o  OPES Processor Request handling role: This involves request
      processing before forwarding to Data Provider Application.
   o  Data Provider Response handling role: This involves response
      processing when forwarding to Data Consumer Application.
   o  OPES Processor Response handling role: This involves response
      processing when forwarding to Data Consumer Application.

4.  Requirements for Interfaces



   The interface between the policy system and OPES services needs to
   include the ability to pass system state information as well as the
   subject message.



Barbir, et al.               Informational                      [Page 6]

RFC 3838                OPES Policy Requirements             August 2004


4.1.  Service Bindings Requirements



   The invoked OPES services MUST be able to be specified in a location
   independent fashion.  That is, the rule authors need not know and
   need not specify the instance of an OPES service in the rules.

   The rule author SHOULD be able to identify the required service at
   the detail level that is appropriate for his or her needs.  The rule
   author SHOULD be able to specify a type of service or be able to
   specify any service that fits a general category of service to be
   applied to its traffic.

   The binding of OPES service names to a specific service MAY be
   distributed between the PDP and the PEP.  As rules are compiled and
   validated by the PDP, they MUST be resolved to a specific
   installations' set of homogeneous OPES service.

   The selection of a specific instance MAY be postponed and left to PEP
   to select at either the rule installation time or at run time.  To
   achieve interoperability, PEP MUST support resolving a generic name
   to a specific instance.  It is possible to use services such as SLP
   or UDDI to resolve generic service names to specific OPES service
   instances.

   The policy system MAY support dynamic discovery of service bindings.
   The rule author may not know specific service bindings, such as
   protocol and parameters, when a rule (as specified on the PDP
   Interface) is general in nature.  The required binding information
   MUST be provided by the PDP and conveyed on the PEP Interface.  A
   service description methodology such as WSDL [8] MUST be present in
   the policy system.

4.1.1.  Environment Variables



   There may be a need to define and support a means for maintaining
   state information that can be used in both condition evaluation and
   action execution.  Depending on the execution environment, OPES
   services MAY have the freedom to define variables that are needed and
   use these variables to further define their service behavior without
   the data filter support.











Barbir, et al.               Informational                      [Page 7]

RFC 3838                OPES Policy Requirements             August 2004


4.1.2.  Requirements for Using State Information



   Policy rules MAY specify that state information be used as part of
   the evaluation of the rules against a given message in an OPES flow.
   Thus, the policy system SHOULD support the maintenance of groups that
   can be used in evaluating rule conditions.  Membership in such groups
   can be used as action triggers.

   For example, an authorized site blocking service might conclude that
   a particular user shouldn't be permitted access to a certain web
   site.  Rather than calling the service for each request sent by such
   a user, a rule might be created to determine whether a user is a
   member of blocked users and if a requested site is a member of
   blocked-sites, and then invoke a local blocking service to return an
   appropriate message to the user.

4.1.3.  Requirements for Passing Information Between Services



   Environment variables can be used to pass state information between
   services.  For example, analysis of the request or modifications to
   the request may need to be captured as state information that can be
   passed to other services on the request path or to services on the
   response(s) associated with that request.

   In the PEP, there SHOULD be provisions to enable setting up variables
   when returning from a service call and passing variables to other
   called services based on policy.

4.2.  Requirements for Rule and Rules Management



   This section provides the requirements for rule management.  The
   rules are divided into two groups.  Some rules are provided by the
   data consumer application, and other rules are provided by the data
   provider application.

4.2.1.  Requirements for Rule Providers



   The requirements for rule providers are:

   o  Rule providers MUST be authenticated and authorized for rules that
      apply to their network role.
   o  Rule providers MUST NOT be able to specify rules that are NOT
      within their scope of authority.
   o  Rule providers SHOULD be able to specify only what is needed for
      their services.
   o  Compilation of rules from different sources MUST NOT lead to
      execution of conflicting rules.
   o  The resolution of such rule conflicts is out of scope.



Barbir, et al.               Informational                      [Page 8]

RFC 3838                OPES Policy Requirements             August 2004


   o  Rules are assumed to be static and applied to current network
      state.

4.2.2.  Requirements for Rule Formats and Protocols



   It is desirable to choose standard technologies like XML to specify
   the rule language format.

   Rules need to be sent from the rule authors to the OPES
   administrative server for service authorization, rule validation, and
   compilation.  The mechanisms for doing that are out of scope of the
   current work.

   Once the rules are authorized, validated, and compiled by the
   administrative server, the rules need to be sent to the OPES
   processor.  The mechanisms for doing that are out of scope of the
   current work.

4.2.3.  Requirements for Rule Conditions



   Rule conditions MUST be matched against attribute values of the
   encapsulated protocol as well as environment variable values.
   Attribute values of the encapsulated protocol include protocol header
   values and possibly also protocol body values.

   Some OPES services may need to be invoked for all user requests or
   server responses, such as services with logging functionality, for
   example.  The rule system SHOULD allow unconditional rules rather
   than requiring rule authors to specify rule conditions that are
   always true.

4.2.4.  Requirements for Rule Actions



   The rule system MUST allow for the specification of rule actions that
   are triggered if the conditions of a rule are met.  Matched rules
   typically lead to the invocation of local or remote services.  Rule
   actions MUST identify the OPES service that is to be executed for the
   current message request or response.

   Rule actions MAY contain run-time parameters which can be used to
   control the behavior of an OPES service.  If specified, these
   parameters MUST be passed to the executed OPES service.









Barbir, et al.               Informational                      [Page 9]

RFC 3838                OPES Policy Requirements             August 2004


4.3.  Requirements for Policy Expression



   OPES processors MUST enforce policy requirements set by data
   consumers and/or data publishers in accordance with the architecture
   [1] and this document.  They cannot do this consistently unless there
   are an unambiguous semantics and representation of the data elements
   mentioned in the policy.  For example, this document mentions
   protection of user "identity" and "profile" information.  If a user
   specifies that his identity must not be shared with other OPES
   administrative trust domains, and later discovers that his family
   name has been shared, he might complain.  If he were told that
   "family names are not considered 'identities' by this site", he would
   probably feel that he had cause for complaint.  Or, he might be told
   that when he selected "do not share identity" on a web form offered
   by the OPES service provider, that this only covered his login name,
   and that a different part of the form had to be filled out to protect
   the family name.  A further breakdown can occur if the configuration
   information provided by such a web form gets translated into
   configuration elements given to an OPES processor, and those
   configuration elements are difficult for a software engineer to
   translate into policy enforcement.  The data elements might have
   confusing names or be split into groupings that are difficult to
   relate to one another.

   The examples illustrate why the OPES policy MUST have definitions of
   data elements, their relationships, and how they relate to
   enforcement.  These semantics of essential items do not require a
   separate protocol, but they MUST be agreed upon by all OPES service
   providers, and the users of OPES services MUST be assured that they
   have the ability to know their settings, to change them if the
   service provider policy allows the changes, and to have reasonable
   assurance that they are enforced with reasonable interpretations.

   The requirements for policy data elements in the OPES specification
   do not have to be all-inclusive, but they MUST cover the minimal set
   of elements that enable the policies that protect the data of end
   users and publishers.

5.  Authentication of Principals and Authorization of Services



   This section considers the authorization and authentication of OPES
   services.









Barbir, et al.               Informational                     [Page 10]

RFC 3838                OPES Policy Requirements             August 2004


5.1.  End Users, Publishers and Other Considerations



5.1.1.  Considerations for End Users



   An OPES rule determines which attributes of traffic will trigger the
   application of OPES services.  The author of the service can supply
   rules, but the author cannot supply the necessary part of the rule
   precondition that determines which network users will have the OPES
   services applied for them.  This section discusses how users are
   identified in the rule preconditions, and how users can select and
   deselect OPES services for their traffic, how an OPES service
   provider SHOULD identify the users, and how they determine whether or
   not to add their service selection to an OPES enforcement point.

   An OPES service provider MUST satisfy these major requirements:

   o  Allow all users to request addition, deletion, or blocking of OPES
      services for their traffic (blocking means "do not use this
      service for my traffic").
   o  Prevent untrusted users from causing OPES services to interfere
      with the traffic of other users.
   o  Allow users to see their OPES service profiles and notify them of
      changes.
   o  Keep a log of all profile activity for audit purposes.
   o  Adhere to a privacy policy guarding users' profiles.

   The administrator of the PDP is a trusted party and can set policy
   for individuals or groups using out-of-band communication and
   configuration files.  However, users MUST always be able to query the
   PDP in order to learn what rules apply to their traffic.

   Rules can be deposited in the PDP with no precondition relating to
   network users.  This is the way rules are packaged with an OPES
   service when it is delivered for installation.  The PDP is
   responsible for binding identities to the rules and transmitting them
   to the PEP.  The identity used by the PDP for policy decisions MUST
   be strictly mapped to the identity used by the PEP.  Thus, if a user
   goes through an identification and authentication procedure with the
   PDP and is known by identity "A", and if the PEP uses IP addresses
   for identities, then the PDP MUST provide the PEP with a binding
   between "A" and A's current IP address.










Barbir, et al.               Informational                     [Page 11]

RFC 3838                OPES Policy Requirements             August 2004


5.1.2.  Considerations for Publishing Sites



   An OPES service provider acting on behalf of different publishing
   sites SHOULD keep all the above considerations in mind when
   implementing an OPES site.  Because each publishing site may be
   represented by only a single identity, the authentication and
   authorization databases may be easier for the PEP to handle.

5.1.3.  Other Considerations



   Authentication may be necessary between PDP's and PEP's, PEP's and
   callout servers, PEP's and other PEP's, and callout servers and other
   callout servers, for purposes of validating privacy policies.  In any
   case where user data or traffic crosses trust domain boundaries, the
   originating trust domain SHOULD have a policy describing which other
   domains are trusted, and it SHOULD authenticate the domains and their
   policies before forwarding information.

5.2.  Authentication



   When an individual selects (or deselects) an OPES service, the
   individual MUST be authenticated by the OPES service provider.  This
   means that a binding between the user's communication channel and an
   identity known to the service provider is made in a secure manner.
   This SHOULD be done using a strong authentication method with a
   public key certificate for the user; this will be helpful in
   resolving later disputes.  It is recommended that the service
   provider keep a log of all requests for OPES services.  The service
   provider SHOULD use public key certificates to authenticate responses
   to requests.

   The service provider may have trusted users who through explicit or
   implicit contract can assign, remove, or block OPES services for
   particular users.  The trusted users MUST be authenticated before
   being allowed to take actions which will modify the policy base, and
   thus, the actions of the PEP's.

   Because of the sensitivity of user profiles, the PEP Interface
   between the PEP and the PDP MUST use a secure transport protocol.
   The PEP's MUST adhere to the privacy preferences of the users.

   When an OPES service provider accepts an OPES service, there MUST be
   a unique name for the service provided by the entity publishing the
   service.  Users MAY refer to the unique name when requesting a
   service.  The unique name MUST be used when notifying users about
   their service profiles.  PEP's MUST be aware of the unique name for
   each service that can be accessed from their domain.  There MUST be a
   cryptographic binding between the unique name and the entity



Barbir, et al.               Informational                     [Page 12]

RFC 3838                OPES Policy Requirements             August 2004


   responsible for the functional behavior of the service, i.e., if it
   is a human language translating service, then the name of company
   that wrote the software SHOULD be bound to the unique name.

5.3.  Authorization



   In addition to requesting or terminating specific services, users MAY
   block particular services, indicating that the services should not be
   applied to their traffic.  The "block all OPES" directive MUST be
   supported on a per user basis.

   A response to a request for an OPES service can be positive or
   negative.  Reasons for a negative response include "service unknown"
   or "service denied by PDP policy".  Positive responses SHOULD include
   the identity of the requestor and the service and the type of
   request.

   As described in the OPES Architecture [1], requests for OPES services
   originate in either the end user or the publisher domain.  The PDP
   bases its authorization decision on the requestor and the domain.
   There are some cases where the decision may be complicated.

   o  The end user has blocked a service, but a trusted user of the PDP
      wants it applied anyway.  In this case, the end user SHOULD
      prevail, unless there are security or legal reasons to leave it in
      place.
   o  The publisher and the end user are in the same domain.  If the
      publisher and end user are both clients of a PDP, can they make
      requests that effect each other's processing?  In this case, the
      PDP MUST have policy rules naming the identities that are allowed
      to set such rules.
   o  The publisher requests a service for an end user.  In this case,
      where the PDP and PEP are in the publisher's administrative
      domain, the publisher has some way of identifying the end user and
      his traffic, and the PDP MUST enable the PEP to enforce the
      policy.  This is allowed, but the PDP MUST use strong methods to
      identify the user and his traffic.  The user MUST be able to
      request and receive information about the service profile that a
      publisher site keeps about him.
   o  The end user requests a service specific to a publisher's identity
      (e.g., nfl.com), but the publisher prohibits the service (e.g.,
      through a "NO OPES" application header).  As in the case above,
      the publisher MUST be able to request and receive profile
      information that a user keeps about a publisher.

   In general, the PDP SHOULD keep its policy base in a manner that
   makes the decision procedure for all cases easy to understand.




Barbir, et al.               Informational                     [Page 13]

RFC 3838                OPES Policy Requirements             August 2004


5.4.  Integrity and Encryption



5.4.1.  Integrity and Confidentiality of Authentication and Requests/
        Responses for Service



   The requests and responses SHOULD be cryptographically tied to the
   identities of the requestor and responder, and the messages SHOULD
   NOT
be alterable without detection.  A certificate-based digital
   signature is strongly recommended as part of the authentication
   process.  A binding between the request and response SHOULD be
   established using a well-founded cryptographic means, to show that
   the response is made in reply to a specific request.

5.4.2.  Integrity and Confidentiality of Application Content



   As directed by the PEP, content will be transformed in whole or in
   part by OPES services.  This means that end-to-end cryptographic
   protections cannot be used.  This is probably acceptable for the vast
   majority of traffic, but in cases where a lesser form of content
   protection is desirable, hop-by-hop protections can be used instead.
   The requirements for such protections are:

   o  Integrity using shared secrets MUST be used between all processing
      points, end-to-end (i.e., the two ends of a "hop" MUST share a
      secret, but the secret can be different between "hops").  The
      processing points include the callout servers.
   o  Encryption can be requested separately, with the same secret
      sharing requirement between "hops".  When requested, encryption
      applies to all processing points, including callout servers.
   o  The signal for integrity (and optionally encryption) MUST
      originate from either the requestor (in which case it is applied
      to the response as well) or the responder (in which case it covers
      only the response).
   o  The shared secrets MUST be unique (to within a very large
      probabilistic certainty) for each requestor/responder pair.  This
      helps to protect the privacy of end user data from insider attacks
      or configuration errors while it transits the provider's network.

5.5.  Privacy



   The PDP MUST have a privacy policy regarding OPES data such as user
   profiles for services.  Users MUST be able to limit the promulgation
   of their profile data and their identities.

   Supported limitations MUST include:

   o  The ability to prevent Identity from being given to callout
      servers.



Barbir, et al.               Informational                     [Page 14]

RFC 3838                OPES Policy Requirements             August 2004


   o  The ability to prevent Profile information from being shared.
   o  The ability to prevent Traffic data from being sent to callout
      servers run by third parties.
   o  The ability to prevent Traffic from particular sites from being
      given to OPES callout servers.

   When an OPES service is provided by a third-party, it MUST have a
   privacy policy and identify itself to upstream and downstream
   parties, telling them how to access its privacy policy.  A mechanism
   is needed to specify these preferences and a protocol to distribute
   them (see section 3.3).

6.  Security Considerations



   This document discusses policy, authorization and enforcement
   requirements of OPES.  In [3]  multiple security and privacy issues
   related to the OPES services are discussed.

7.  References



7.1.  Normative References



   [1]  Barbir, A., Penno, R., Chen, R., Hofmann, M., and H. Orman, "An
        Architecture for Open Pluggable Edge Services (OPES)", RFC 3835,
        August 2004.

   [2]  Floyd, S. and L. Daigle, "IAB Architectural and Policy
        Considerations for Open Pluggable Edge Services", RFC 3238,
        January 2002.

   [3]  Barbir, A., Batuner, O., Srinivas, B., Hofmann, M., and H.
        Orman, "Security Threats and Risks for Open Pluggable Edge
        Services (OPES)", RFC 3837, August 2004.

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

   [5] Moore, B., Ellesson, E., Strassner, J., and A. Westerinen,
        "Policy Core Information Model -- Version 1 Specification", RFC
        3060, February 2001.

7.2.  Informative References



   [6]  Westerinen, A., Schnizlein, J., Strassner, J., Scherling, M.,
        Quinn, B., Herzog, S., Huynh, A., Carlson, M., Perry, J., and S.
        Waldbusser, "Terminology for Policy-Based Management", RFC 3198,
        November 2001.




Barbir, et al.               Informational                     [Page 15]

RFC 3838                OPES Policy Requirements             August 2004


   [7]  Fielding, R., Gettys, J., Mogul, J., Frystyk, H., Masinter, L.,
        Leach, P., and T. Berners-Lee, "Hypertext Transfer Protocol --
        HTTP/1.1", RFC 2616, June 1999.

   [8]  Christensen, et al., Web Services Description Language (WSDL)
        1.1, W3C Note 15 March 2001, http://www.w3.org/TR/wsdl

8.  Acknowledgements



   Many thanks to Andreas Terzis, L. Rafalow (IBM), L. Yang (Intel), M.
   Condry (Intel), Randy Presuhn (Mindspring), and B. Srinivas (Nokia).

9.  Authors' Addresses



   Abbie Barbir
   Nortel Networks
   3500 Carling Avenue
   Nepean, Ontario  K2H 8E9
   Canada
   Phone: +1 613 763 5229
   EMail: abbieb@nortelnetworks.com

   Oskar Batuner
   Consultant
   EMail: batuner@attbi.com

   Andre Beck
   Lucent Technologies
   101 Crawfords Corner Road
   Holmdel, NJ  07733
   USA
   EMail: abeck@bell-labs.com

   Tat Chan
   Nokia
   5 Wayside Road
   Burlington, MA  01803
   USA
   EMail: Tat.Chan@nokia.com

   Hilarie Orman
   Purple Streak Development
   EMail: ho@alum.mit.edu








Barbir, et al.               Informational                     [Page 16]

RFC 3838                OPES Policy Requirements             August 2004


10.  Full Copyright Statement



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

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

Intellectual Property



   The IETF takes no position regarding the validity or scope of any
   Intellectual Property Rights or other rights that might be claimed to
   pertain to the implementation or use of the technology described in
   this document or the extent to which any license under such rights
   might or might not be available; nor does it represent that it has
   made any independent effort to identify any such rights.  Information
   on the procedures with respect to rights in RFC documents can be
   found in BCP 78 and BCP 79.

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

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

Acknowledgement



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









Barbir, et al.               Informational                     [Page 17]