Skip to main content

A Generic Discovery and Negotiation Protocol for Autonomic Networking
draft-carpenter-anima-gdn-protocol-00

The information below is for an old version of the document.
Document Type
This is an older version of an Internet-Draft whose latest revision state is "Replaced".
Authors Brian E. Carpenter , Sheng Jiang , Bing Liu
Last updated 2014-10-13
Replaced by draft-ietf-anima-grasp, RFC 8990
RFC stream (None)
Formats
Stream Stream state (No stream defined)
Consensus boilerplate Unknown
RFC Editor Note (None)
IESG IESG state I-D Exists
Telechat date (None)
Responsible AD (None)
Send notices to (None)
draft-carpenter-anima-gdn-protocol-00
Network Working Group                                       B. Carpenter
Internet-Draft                                         Univ. of Auckland
Intended status: Standards Track                                S. Jiang
Expires: April 16, 2015                                           B. Liu
                                            Huawei Technologies Co., Ltd
                                                        October 13, 2014

 A Generic Discovery and Negotiation Protocol for Autonomic Networking
                 draft-carpenter-anima-gdn-protocol-00

Abstract

   This document defines a new protocol that enables intelligent devices
   to dynamically discover peer devices, to synchronize state with them,
   and to negotiate mutual configurations with them.  This document only
   defines a general protocol as a negotiation platform, while the
   negotiation objectives for specific scenarios are to be described in
   separate documents.

Status of This Memo

   This Internet-Draft is submitted in full conformance with the
   provisions of BCP 78 and BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF).  Note that other groups may also distribute
   working documents as Internet-Drafts.  The list of current Internet-
   Drafts is at http://datatracker.ietf.org/drafts/current/.

   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."

   This Internet-Draft will expire on April 16, 2015.

Copyright Notice

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

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   (http://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must

Carpenter, et al.        Expires April 16, 2015                 [Page 1]
Internet-Draft                GDN Protocol                  October 2014

   include Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.

Table of Contents

   1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . .   3
   2.  Requirements Language and Terminology . . . . . . . . . . . .   4
   3.  Requirement Analysis of Discovery, Synchronization and
       Negotiation . . . . . . . . . . . . . . . . . . . . . . . . .   5
     3.1.  Requirements for Discovery  . . . . . . . . . . . . . . .   5
     3.2.  Requirements for Synchronization and Negotiation
           Capability  . . . . . . . . . . . . . . . . . . . . . . .   6
   4.  Negotiation Capability Analysis of Current Protocols  . . . .   7
   5.  GDNP Protocol Overview  . . . . . . . . . . . . . . . . . . .   9
     5.1.  High-Level Design Choices . . . . . . . . . . . . . . . .   9
     5.2.  GDNP Protocol Basic Properties and Mechanisms . . . . . .  13
       5.2.1.  IP Version Independent  . . . . . . . . . . . . . . .  13
       5.2.2.  Discovery Mechanism and Procedures  . . . . . . . . .  13
       5.2.3.  Certificate-based Security Mechanism  . . . . . . . .  14
       5.2.4.  Negotiation Procedures  . . . . . . . . . . . . . . .  17
     5.3.  GDNP Constants  . . . . . . . . . . . . . . . . . . . . .  18
     5.4.  Device Identifier and Certificate Tag . . . . . . . . . .  18
     5.5.  Session Identifier  . . . . . . . . . . . . . . . . . . .  19
     5.6.  GDNP Messages . . . . . . . . . . . . . . . . . . . . . .  19
       5.6.1.  GDNP Message Format . . . . . . . . . . . . . . . . .  19
       5.6.2.  Discovery Message . . . . . . . . . . . . . . . . . .  20
       5.6.3.  Response Message  . . . . . . . . . . . . . . . . . .  21
       5.6.4.  Request Message . . . . . . . . . . . . . . . . . . .  21
       5.6.5.  Negotiation Message . . . . . . . . . . . . . . . . .  22
       5.6.6.  Negotiation-ending Message  . . . . . . . . . . . . .  22
       5.6.7.  Confirm-waiting Message . . . . . . . . . . . . . . .  22
     5.7.  GDNP General Options  . . . . . . . . . . . . . . . . . .  22
       5.7.1.  Format of GDNP Options  . . . . . . . . . . . . . . .  22
       5.7.2.  Divert Option . . . . . . . . . . . . . . . . . . . .  23
       5.7.3.  Accept Option . . . . . . . . . . . . . . . . . . . .  24
       5.7.4.  Decline Option  . . . . . . . . . . . . . . . . . . .  24
       5.7.5.  Waiting Time Option . . . . . . . . . . . . . . . . .  25
       5.7.6.  Certificate Option  . . . . . . . . . . . . . . . . .  26
       5.7.7.  Signature Option  . . . . . . . . . . . . . . . . . .  26
       5.7.8.  Locator Options . . . . . . . . . . . . . . . . . . .  27
     5.8.  Discovery Objective Option  . . . . . . . . . . . . . . .  29
     5.9.  Negotiation Objective Options and Considerations  . . . .  29
       5.9.1.  Organizing of GDNP Options  . . . . . . . . . . . . .  30
       5.9.2.  Vendor Specific Options . . . . . . . . . . . . . . .  30
       5.9.3.  Experimental Options  . . . . . . . . . . . . . . . .  30
     5.10. Items for Future Work . . . . . . . . . . . . . . . . . .  30
   6.  Security Considerations . . . . . . . . . . . . . . . . . . .  32

Carpenter, et al.        Expires April 16, 2015                 [Page 2]
Internet-Draft                GDN Protocol                  October 2014

   7.  IANA Considerations . . . . . . . . . . . . . . . . . . . . .  32
   8.  Acknowledgements  . . . . . . . . . . . . . . . . . . . . . .  35
   9.  Change log [RFC Editor: Please remove]  . . . . . . . . . . .  35
   10. References  . . . . . . . . . . . . . . . . . . . . . . . . .  35
     10.1.  Normative References . . . . . . . . . . . . . . . . . .  35
     10.2.  Informative References . . . . . . . . . . . . . . . . .  35
   Authors' Addresses  . . . . . . . . . . . . . . . . . . . . . . .  37

1.  Introduction

   The success of the Internet has made IP-based networks bigger and
   more complicated.  Large-scale ISP networks have become more and more
   problematic for human based management.  Also operational costs are
   growing quickly.  Consequently, there are therefore increased
   requirements for autonomy in the networks.  General aspects of
   autonomic networks are discussed in
   [I-D.irtf-nmrg-autonomic-network-definitions] and
   [I-D.irtf-nmrg-an-gap-analysis].  In order to fulfil autonomy,
   devices that are more intelligent need to be able to discover each
   other, to synchronize state with each other, and negotiate directly
   with each other.

   Following this Introduction and the definition of useful terminology,
   Section 3 describes the requirements and application scenarios for
   network device negotiation.  Then the negotiation capabilities of
   various existing protocols are reviewed in Section 4.  State
   synchronization, when needed, can be considered as a special case of
   negotiation.  Prior to negotiation or synchronization, devices must
   discover each other.  Section 5.1 describes a behavior model for a
   protocol intended to support discovery, synchronization and
   negotiation.  The design of Generic Discovery and Negotiation
   Protocol (GDNP) in Section 5 of this document is mainly based on this
   behavior model.

   Although many negotiations may happen between horizontally
   distributed peers, the main target scenarios are still hierarchical
   networks, which is the major structure of current large-scale
   networks.  Thus, where necessary, we assume that each network element
   has a hierarchical superior.  Of course, the protocol itself is
   capable of being used in a small and/or flat network structure such
   as a small office or home network, too.

   This document defines a Generic Discovery and Negotiation Protocol
   (GDNP), that can be used to perform decision process among
   distributed devices or between networks.  The newly defined GDNP in
   this document adapts a tight certificate-based mechanism, which needs
   a Public Key Infrastructure (PKI, [RFC5280]) system.  The PKI may be
   managed by an operator or be autonomic.  The document also introduces

Carpenter, et al.        Expires April 16, 2015                 [Page 3]
Internet-Draft                GDN Protocol                  October 2014

   a new discovery mechanism, which is based on a neighbor learning
   process and is oriented towards negotiation objectives.

   It is understood that in realistic deployments, not all devices will
   support GDNP.  Such mixed scenarios are not discussed in this
   specification.

2.  Requirements Language and Terminology

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
   "OPTIONAL" in this document are to be interpreted as described in
   [RFC2119] when they appear in ALL CAPS.  When these words are not in
   ALL CAPS (such as "should" or "Should"), they have their usual
   English meanings, and are not to be interpreted as [RFC2119] key
   words.

   o  Discovery: a process by which a device discovers peer devices
      according to a specific discovery objective.  The discovery
      results may be different according to the different discovery
      objectives.  The discovered peer devices may later be used as
      negotiation counterparts.

   o  Negotiation: a process by which two (or more) devices interact
      iteratively to agree on parameter settings that best satisfy the
      objectives of one or more devices.

   o  State Synchronization: a process by which two (or more) devices
      interact iteratively to agree on the current state of parameter
      values stored in each device.  This is a special case of
      negotiation in which information is exchanged but the devices do
      not request their peers to change parameter settings.  All other
      definitions apply to both negotiation and synchronization.

   o  Discovery Objective: a specific functionality, role-based network
      element or service agent (TBD) which the discovery initiator
      intends to discover.  One device may support multiple discovery
      objectives.

   o  Discovery Initiator: a device that spontaneously starts discovery
      by sending a discovery message referring to a specific discovery
      objective.

   o  Discovery Responder: a peer device which responds to the discovery
      objective initiated by the discovery initiator.

   o  Negotiation Objective: specific negotiation content, which needs
      to be decided in coordination with another network device.  It is

Carpenter, et al.        Expires April 16, 2015                 [Page 4]
Internet-Draft                GDN Protocol                  October 2014

      naturally based on a specific service or function or action.  It
      could be a logical, numeric, or string value or a more complex
      data structure.

   o  Negotiation Initiator: a device that spontaneously starts
      negotiation by sending a request message referring to a specific
      negotiation objective.

   o  Negotiation Counterpart: a peer device with which the Negotiation
      Initiator negotiates a specific negotiation objective.

   o  Device Identifier: a public key, which identifies the device in
      CDNP messages.  It is assumed that its associated private key is
      maintained in the device only.

   o  Device Certificate: A certificate for a single device, also the
      identifier of the device, further described in Section 5.4.

   o  Device Certificate Tag: a tag, which is bound to the device
      identifier.  It is used to present Device Certificate in short
      form.

3.  Requirement Analysis of Discovery, Synchronization and Negotiation

   This section discusses the requirements for discovery, negotiation
   and synchronization capabilities.

3.1.  Requirements for Discovery

   In an autonomic network we must assume that when a device starts up
   it has no information about any peer devices.  In some cases, when a
   new user session starts up, the device concerned may again lack
   information about relevant peer devices.  It might be necessary to
   set up resources on multiple other devices, coordinated and matched
   to each other so that there is no wasted resource.  Security settings
   might also need updating to allow for the new device or user.
   Therefore a basic requirement is that there must be a mechanism by
   which a device can discover peer devices.  These devices might be
   immediate neighbors on the same layer 2 link or they might be more
   distant and only accessible via layer 3.

   The relevant peer devices may be different for different discovery
   objectives.  Therefore discovery needs to be repeated as often as
   necessary to find peers capable of acting as counterparts for each
   objective that a discovery initiator needs to handle.  In many
   scenarios, discovery process may follow up by negotiation process.
   Correspondently, the discovery objective may associate with the
   negotiation objective.

Carpenter, et al.        Expires April 16, 2015                 [Page 5]
Internet-Draft                GDN Protocol                  October 2014

   In most networks, as mentioned above, there will be some hierarchical
   structure.  A special case of discovery is that each device must be
   able to discover its hierarchical superior for each negotiation
   objective that it is capable of handling.

   During initialisation, a device must be able to discover the
   appropriate trust anchor.  Logically, this is just a specific case of
   discovery.  However, it might be a special case requiring its own
   solution.  This question requires further study.

3.2.  Requirements for Synchronization and Negotiation Capability

   We start by considering routing protocols, the closest approximation
   to autonomic networking in widespread use.  Routing protocols use a
   largely autonomic model based on distributed devices that communicate
   iteratively with each other.  However, routing is mainly based on
   one-way information announcements (in either direction), rather than
   on bi-directional negotiation.  The only focus is reachability, so
   current routing protocols only consider simple link status, as up or
   down.  More information, such as latency, congestion, capacity, and
   particularly unused capacity, would be helpful to get better path
   selection and utilization rate.  Also, autonomic networks need to be
   able to manage many more dimensions, such as security settings, power
   saving, load balancing, etc.  A basic requirement for the protocol is
   therefore the ability to represent, discover, synchronize and
   negotiate almost any kind of network parameter.

   Human intervention in complex situations is costly and error-prone.
   Therefore, a negotiation model without human intervention is
   desirable whenever the coordination of multiple devices can provide
   better overall network performance.  Therefore a requirement for the
   protocol is to be capable of being installed in any device that would
   otherwise need human intervention.

   Human intervention in large networks is often replaced by use of a
   top-down network management system (NMS).  It follows that a
   requirement for the protocol is to be capable of being installed in
   any device that would otherwise be managed by an NMS, and that it can
   co-exist with an NMS.

   Since the goal is no human intervention, it is necessary that the
   network can in effect "think ahead" before changing its parameters.
   In other words there must be a possibility of forecasting the effect
   of a change.  Stated differently, the protocol must be capable of
   supporting a "dry run" of a changed configuration before actually
   installing the change.

Carpenter, et al.        Expires April 16, 2015                 [Page 6]
Internet-Draft                GDN Protocol                  October 2014

   Status information and traffic metrics need to be shared between
   nodes for dynamic adjustment of resources and for monitoring
   purposes.  While this might be achieved by existing protocols when
   they are available, the new protocol needs to be able to support
   parameter exchange, including mutual synchronization, even when no
   negotiation as such is required.

   Recovery from faults and identification of faulty devices should be
   as automatic as possible.  The protocol needs to be capable of
   detecting unexpected events such a negotiation counterpart failing,
   so that all devices concerned can initiate a recovery process.

   The protocol needs to be able to deal with a wide variety of
   negotiation objectives, covering any type of network parameter.
   Therefore the protocol will need either an explicit information model
   describing its messages, or at least a flexible and extensible
   message format.  One design consideration is whether to adopt an
   existing information model or to design a new one.  Another
   consideration is whether to be able to carry some or all of the
   message formats used by existing configuration protocols.

   The protocol needs to be fully secure against forged messages and
   man-in-the middle attacks, and as secure as reasonably possible
   against denial of service attacks.  It needs to be capable of
   encryption in order to resist unwanted monitoring, although this
   capability may not be required in all deployments.

4.  Negotiation Capability Analysis of Current Protocols

   This section discusses various existing protocols with properties
   related to the above negotiation and synchronisation requirements.
   The purpose is to evaluate whether any existing protocol, or a simple
   combination of existing protocols, can meet those requirements.

   The analysis does not include discovery protocols.  While numerous
   protocols include some form of discovery, these all appear to be very
   specific in their applicability.

   Routing protocols are mainly one-way information announcements.  The
   receiver makes independent decisions based on the received
   information and there is no direct feedback information to the
   announcing peer.  This remains true even though the protocol is used
   in both directions between peer routers; there is state
   synchronization, but no negotiation, and each peer runs its route
   calculations independently.

   Simple Network Management Protocol (SNMP) [RFC3416] uses a command/
   response model not well suited for peer negotiation.  Network

Carpenter, et al.        Expires April 16, 2015                 [Page 7]
Internet-Draft                GDN Protocol                  October 2014

   Configuration Protocol (NETCONF) [RFC6241] uses an RPC model that
   does allow positive or negative responses from the target system, but
   this is still not adequate for negotiation.

   There are various existing protocols that have elementary negotiation
   abilities, such as Dynamic Host Configuration Protocol for IPv6
   (DHCPv6) [RFC3315], Neighbor Discovery (ND) [RFC4861], Port Control
   Protocol (PCP) [RFC6887], Remote Authentication Dial In User Service
   (RADIUS) [RFC2865], Diameter [RFC6733], etc.  Most of them are
   configuration or management protocols.  However, they either provide
   only a simple request/response model in a master/slave context or
   very limited negotiation abilities.

   There are also signalling protocols with an element of negotiation.
   For example Resource ReSerVation Protocol (RSVP) [RFC2205] was
   designed for negotiating quality of service parameters along the path
   of a unicast or multicast flow.  RSVP is a very specialised protocol
   aimed at end-to-end flows.  However, it has some flexibility, having
   been extended for MPLS label distribution [RFC3209].  A more generic
   design is General Internet Signalling Transport (GIST) [RFC5971], but
   it is complex, tries to solve many problems, and is also aimed at
   per-flow signalling across many hops rather than at device-to-device
   signalling.  However, we cannot completely exclude extended RSVP or
   GIST as a synchronization and negotiation protocol.  They do not
   appear to be directly useable for peer discovery.

   We now consider two protocols that are works in progress at the time
   of this writing.  Firstly, RESTCONF [I-D.ietf-netconf-restconf] is a
   protocol intended to convey NETCONF information expressed in the YANG
   language via HTTP, including the ability to transit HTML
   intermediaries.  While this is a powerful approach in the context of
   centralised configuration of a complex network, it is not well
   adapted to efficient interactive negotiation between peer devices,
   especially simple ones that are unlikely to include YANG processing
   already.

   Secondly, we consider HomeNet Control Protocol (HNCP)
   [I-D.ietf-homenet-hncp].  This is defined as "a minimalist state
   synchronization protocol for Homenet routers."  Specific features
   are:

   o  Every participating node has a unique node identifier.

   o  "HNCP is designed to operate between directly connected neighbors
      on a shared link using link-local IPv6 addresses."

   o  Currency of state is maintained by spontaneous link-local
      multicast messages.

Carpenter, et al.        Expires April 16, 2015                 [Page 8]
Internet-Draft                GDN Protocol                  October 2014

   o  HNCP discovers and tracks link-local neighbours.

   o  HNCP messages are encoded as a sequence of TLV objects, sent over
      UDP.

   o  Authentication depends on a signature TLV (assuming public keys
      are associated with node identifiers).

   o  The functionality covered initially includes: site border
      discovery, prefix assignment, DNS namespace discovery, and routing
      protocol selection.

   Clearly HNCP does not completely meet the needs of a general
   negotiation protocol, especially due to its limitation to link-local
   messages and its strict dependency on IPv6, but at the minimum it is
   a very interesting test case for this style of interaction between
   devices without needing a central authority.

   A proposal has been made for an IP based Generic Control Protocol
   (IGCP) [I-D.chaparadza-intarea-igcp].  This is aimed at information
   exchange and negotiation but not directly at peer discovery.
   However, it has many points in common with the present work.

   None of the above solutions appears to completely meet the needs of
   discovery, state synchronization and negotiation in the general case.
   Neither is there an obvious combination of protocols that does so.
   Therefore, the remainder of this document proposes the design of a
   protocol that does meet those needs.  However, this proposal needs to
   be confronted with alternatives such as extension and adaptation of
   GIST or HNCP, or combination with IGCP.

5.  GDNP Protocol Overview

5.1.  High-Level Design Choices

   This section describes a behavior model and some considerations for
   designing a generic discovery and negotiation protocol, which would
   act as a platform for different negotiation objectives.

   NOTE: This protocol is described here in a stand-alone fashion as a
   proof of concept.  An elementary version has been prototyped by
   Huawei and the Beijing University of Posts and Telecommunications.
   However, this is not yet a definitive proposal for IETF adoption.  In
   particular, adaptation and extension of one of the protocols
   discussed in Section 4 might be an option.  Also, the security model
   outlined below would in practice be part of a general security
   mechanism in an autonomic control plane.  This whole specification is
   subject to change as a result.

Carpenter, et al.        Expires April 16, 2015                 [Page 9]
Internet-Draft                GDN Protocol                  October 2014

   o  A generic platform

      The design of the network device protocol is desired to be a
      generic platform, which is independent from the negotiation
      contents.  It should only take care of the general
      intercommunication between negotiation counterparts.  The
      negotiation contents will vary according to the various
      negotiation objectives and the different pairs of negotiating
      counterparts.

   o  Security infrastructure and trust relationship

      Because this negotiation protocol may directly cause changes to
      device configurations and bring significant impacts to a running
      network, this protocol must be based on a restrictive security
      infrastructure.  It should be carefully managed and monitored so
      that every device in this negotiation system behaves well and
      remains well protected.

      On the other hand, a limited negotiation model might be deployed
      based on a limited trust relationship.  For example, between two
      administrative domains, devices might also exchange limited
      information and negotiate some particular configurations based on
      a limited conventional or contractual trust relationship.

   o  Discovery and negotiation designed together

      The discovery method and the negotiation method are designed in
      the same way and can be combined when this is useful.

   o  A uniform pattern for negotiation contents

      The negotiation contents should be defined according to a uniform
      pattern.  They could be carried either in TLV (Type, Length and
      Value) format or in payloads described by a flexible language,
      like XML.  A protocol design should choose one of these two.  The
      format must be extensible for unknown future requirements.  As
      noted above, an existing information model and existing message
      format(s) should be considered.

   o  A simple initiator/responder model

      Multi-party negotiations are too complicated to be modeled and
      there may be too many dependencies among the parties to converge

Carpenter, et al.        Expires April 16, 2015                [Page 10]
Internet-Draft                GDN Protocol                  October 2014

      efficiently.  A simple initiator/responder model is more feasible
      and could actually complete multiple-party negotiations by
      indirect steps.  Naturally this process must be guaranteed to
      terminate and must contain tie-breaking rules.

   o  Organizing of negotiation content

      Naturally, the negotiation content should be organized according
      to the relevant function or service.  The content from different
      functions or services should be kept independent from each other.
      They should not be combined into a single option or single session
      because these contents may be negotiated with different
      counterparts or may be different in response time.

   o  Self aware network device

      Every network device should be pre-configured with its role and
      functions and be aware of its own capabilities.  The roles may be
      only distinguished because of network behaviors, which may include
      forwarding behaviors, aggregation properties, topology location,
      bandwidth, tunnel or translation properties, etc.  The role and
      functions may depend on the network planning.  The capability is
      typically decided by the hardware or firmware.  These parameters
      are the foundation of the negotiation behavior of a specific
      device.

   o  Requests and responses in negotiation procedures

      The initiator should be able to negotiate with its relevant
      negotiation counterpart devices, which may be different according
      to the negotiation objective.  It may request relevant information
      from the negotiation counterpart so that it can decide its local
      configuration to give the most coordinated performance.  It may
      request the negotiation counterpart to make a matching
      configuration in order to set up a successful communication with
      it.  It may request certain simulation or forecast results by
      sending some dry run conditions.

      Beyond the traditional yes/no answer, the responder should be able
      to reply with a suggested alternative if its answer is 'no'.  This
      would start a bi-directional negotiation ending in a compromise
      between the two devices.

   o  Convergence of negotiation procedures

Carpenter, et al.        Expires April 16, 2015                [Page 11]
Internet-Draft                GDN Protocol                  October 2014

      The negotiation procedure should move towards convergent results.
      It means that when a responder makes a suggestion of a changed
      condition in a negative reply, it should be as close as possible
      to the original request or previous suggestion.  The suggested
      value of the third or later negotiation steps should be chosen
      between the suggested values from the last two negotiation steps.
      In any case there must be a mechanism to guarantee rapid
      convergence in a small number of steps.

   o  Dependencies of negotiation

      In order to decide a configuration on a device, the device may
      need information from neighbors.  This can be established through
      the above negotiation procedure.  However, a given item in a
      neighbor may depend on other information from its own neighbors,
      which may need another negotiation procedure to obtain or decide.
      Therefore, there are dependencies among negotiation procedures.
      There need to be clear boundaries and convergence mechanisms for
      these negotiation dependencies.  Also some mechanisms are needed
      to avoid loop dependencies.

   o  End of negotiation

      A single negotiation procedure also needs ending conditions if it
      does not converge.  A limited number of rounds, for example three,
      should be set on the devices.  It may be an implementation choice
      or a pre-configurable parameter.  However, the protocol design
      needs to clearly specify this, so that the negotiation can be
      terminated properly.  In some cases, a timeout might be needed to
      end a negotiation.

   o  Failed negotiation

      There must be a well-defined procedure for concluding that a
      negotiation cannot succeed, and if so deciding what happens next
      (deadlock resolution, tie-breaking, or revert to best-effort
      service).

   o  Policy constraints

      There must be provision for general policy rules to be applied by
      all devices in the network (e.g., security rules, prefix length,
      resource sharing rules).  However, policy distribution might not
      use the negotiation protocol itself.

Carpenter, et al.        Expires April 16, 2015                [Page 12]
Internet-Draft                GDN Protocol                  October 2014

   o  Management monitoring, alerts and intervention

      Devices should be able to report to a monitoring system.  Some
      events must be able to generate operator alerts and some provision
      for emergency intervention must be possible (e.g.  to freeze
      negotiation in a mis-behaving device).  These features may not use
      the negotiation protocol itself.

5.2.  GDNP Protocol Basic Properties and Mechanisms

5.2.1.  IP Version Independent

   To be a generic platform, GDNP should be IP version independent.  In
   other words, it should be able to run over IPv6 and IPv4.  Its
   messages and general options are neutral with respect to the IP
   version.

   However, some functions, such as multicasting or broadcasting on a
   link, might need to be IP version dependent.  For these parts, the
   document defines support for both IP versions separately.

5.2.2.  Discovery Mechanism and Procedures

   o  Separated discovery and negotiation mechanisms

         Although discovery and negotiation defined together in the
         GDNP, they are separated mechanisms.  The discovery process
         could run independently from the negotiation process.  Upon
         receiving a discovery (defined in Section 5.6.2) or request
         message (defined in Section 5.6.4) , the recipient device
         should return a message in which it either indicates itself as
         a discovery responder or diverts the initiator towards another
         more suitable device.

         The discovery objective could be network functionalities, role-
         based network elements or service agents (TBD).  The discovery
         results could be utilized by the negotiation protocol to decide
         which device the initiator will negotiate with.

   o  Discovery Procedures

         Discovery starts as on-link operation.  The Divert option can
         tell the discovery initiator to contact an off-link discovery
         objective device.  Every DISCOVERY message is sent by a
         discovery initiator to the ALL_GDNP_NEIGHBOR multicast address
         (Section 5.3).  Every network device that supports the GDNP
         always listens to a well-known (UDP?) port to capture the
         discovery messages.

Carpenter, et al.        Expires April 16, 2015                [Page 13]
Internet-Draft                GDN Protocol                  October 2014

         If the neighbor device supports a proper discovery objective,
         it MAY respond with a Response message (defined in
         Section 5.6.3) with locator option(s).  Otherwise, if the
         neigbor device knows a device that supports the proper
         discovery objective (for example because it discovered the same
         objective before), it SHOULD respond with a Response message
         with a Divert option pointed to the proper discovery objective.

         After a GDNP device successfully discovered a device supporting
         a specific objective, it MUST record this discovery objective.
         This record may be used for future negotiation or to pass to
         another neighbor as a Divert option.  This learning mechanism
         should be able to support most network establishment scenarios

   o  Rapid Mode (Discovery/Negotiation binding)

         A DISCOVERY message MAY includes one or more negotiation
         objective option(s) to indicate to the discovery objective that
         it could directly reply to the discovery initiator with a
         Negotiation message for rapid processing, if the discovery
         objective could act as the corresponding negotiation
         counterpart.  However, the indication is only advisory not
         prescriptive.

         This rapid mode could reduce the interactions between nodes so
         that a higher efficiency could be achieved.  This rapid
         negotiation function SHOULD be configured on or off by the
         administrators.

5.2.3.  Certificate-based Security Mechanism

   A certification based security mechanism provides security properties
   for CDNP:

   o  the identity of a GDNP message sender can be verified by a
      recipient.

   o  the integrity of GDNP message can be checked by the recipient of
      the message.

   o  anti-replay protection on the GDNP message recipient.

   The authority of the GDNP message sender depends on a Public Key
   Infrastructure (PKI) system with a Certification Authority (CA),
   which should normally be run by the network operator.  In the case of
   a network with no operator, such as a small office or home network,
   the PKI itself needs to be established by an autonomic process, which
   is out of scope for this specification.

Carpenter, et al.        Expires April 16, 2015                [Page 14]
Internet-Draft                GDN Protocol                  October 2014

   A Request message MUST carry a Certificate option, defined in
   Section 5.7.6.  The first Negotiation Message, responding to a
   Request message, SHOULD also carry a Certificate option.  Using these
   messages, recipients build their certificate stores, indexed by the
   Device Certificate Tags included in every GDNP message.  This process
   is described in more detail below.

   Every message MUST carry a signature option, defined in
   Section 5.7.7.

   For now, the authors do not think packet size is a problem.  In this
   GDNP specification, there SHOULD NOT be multiple certificates in a
   single message.  The current most used public keys are 1024/2048
   bits, some may reach 4096.  With overhead included, a single
   certificate is less than 500 bytes.  Messages should be far shorter
   than the normal packet MTU within a modern network.

5.2.3.1.  Support for algorithm agility

   Hash functions are used to provide message integrity checks.  In
   order to provide a means of addressing problems that may emerge in
   the future with existing hash algorithms, as recommended in
   [RFC4270], a mechanism for negotiating the use of more secure hashes
   in the future is provided.

   In addition to hash algorithm agility, a mechanism for signature
   algorithm agility is also provided.

   The support for algorithm agility in this document is mainly a
   unilateral notification mechanism from sender to recipient.  If the
   recipient does not support the algorithm used by the sender, it
   cannot authenticate the message.  Senders in a single administrative
   domain are not required to upgrade to a new algorithm simultaneously.

   So far, the algorithm agility is supported by one-way notification,
   rather than negotiation mode.  As defined in Section 5.7.7, the
   sender notifies the recipient what hash/signature algorithms it uses.
   If the responder doesn't know a new algorithm used by the sender, the
   negotiation request would fail.  In order to establish a negotiation
   session, the sender MAY fall back to an older, less preferred
   algorithm.  To avoid downgrade attacks it MUST NOT fall back to an
   algorithm considered weak.

5.2.3.2.  Message validation on reception

   When receiving a GDNP message, a recipient MUST discard the GDNP
   message if the Signature option is absent, or the Certificate option
   is in a Request Message.

Carpenter, et al.        Expires April 16, 2015                [Page 15]
Internet-Draft                GDN Protocol                  October 2014

   For the Request message and the Response message with a Certification
   Option, the recipient MUST first check the authority of this sender
   following the rules defined in [RFC5280].  After successful authority
   validation, an implementation MUST add the sender's certification
   into the local trust certificate record indexed by the associated
   Device Certificate Tag, defined in Section 5.4.

   The recipient MUST now authenticate the sender by verifying the
   Signature and checking a timestamp, as specified in Section 5.2.3.3.
   The order of two procedures is left as an implementation decision.
   It is RECOMMENDED to check timestamp first, because signature
   verification is much more computationally expensive.

   The signature field verification MUST show that the signature has
   been calculated as specified in Section 5.7.7.  The public key used
   for signature validation is obtained from the certificate either
   carried by the message or found from a local trust certificate record
   by searching the message-carried Device Certificate Tag.

   Only the messages that get through both the signature verifications
   and timestamp check are accepted and continue to be handled for their
   contained CDNP options.  Messages that do not pass the above tests
   MUST be discarded as insecure messages.

5.2.3.3.  TimeStamp checking

   Recipients SHOULD be configured with an allowed timestamp Delta
   value, a "fuzz factor" for comparisons, and an allowed clock drift
   parameter.  The recommended default value for the allowed Delta is
   300 seconds (5 minutes); for fuzz factor 1 second; and for clock
   drift, 0.01 second.

   The timestamp is defined in the Signature Option, Section 5.7.7.  To
   facilitate timestamp checking, each recipient SHOULD store the
   following information for each sender:

   o  The receive time of the last received and accepted GDNP message.
      This is called RDlast.

   o  The time stamp in the last received and accepted GDNP message.
      This is called TSlast.

   An accepted GDNP message is any successfully verified (for both
   timestamp check and signature verification) GDNP message from the
   given peer.  It initiates the update of the above variables.
   Recipients MUST then check the Timestamp field as follows:

Carpenter, et al.        Expires April 16, 2015                [Page 16]
Internet-Draft                GDN Protocol                  October 2014

   o  When a message is received from a new peer (i.e., one that is not
      stored in the cache), the received timestamp, TSnew, is checked,
      and the message is accepted if the timestamp is recent enough to
      the reception time of the packet, RDnew:

         -Delta < (RDnew - TSnew) < +Delta

      The RDnew and TSnew values SHOULD be stored in the cache as RDlast
      and TSlast.

   o  When a message is received from a known peer (i.e., one that
      already has an entry in the cache), the timestamp is checked
      against the previously received GDNP message:

         TSnew + fuzz > TSlast + (RDnew - RDlast) x (1 - drift) - fuzz

      If this inequality does not hold, the recipient SHOULD silently
      discard the message.  If, on the other hand, the inequality holds,
      the recipient SHOULD process the message.

      Moreover, if the above inequality holds and TSnew > TSlast, the
      recipient SHOULD update RDlast and TSlast.  Otherwise, the
      recipient MUST NOT update RDlast or TSlast.

   An implementation MAY use some mechanism such as a timestamp cache to
   strengthen resistance to replay attacks.  When there is a very large
   number of nodes on the same link, or when a cache filling attack is
   in progress, it is possible that the cache holding the most recent
   timestamp per sender will become full.  In this case, the node MUST
   remove some entries from the cache or refuse some new requested
   entries.  The specific policy as to which entries are preferred over
   others is left as an implementation decision.

5.2.4.  Negotiation Procedures

   A negotiation initiator sends a negotiation request to counterpart
   devices, which may be different according to different negotiation
   objectives.  It may request relevant information from the negotiation
   counterpart so that it can decide its local configuration to give the
   most coordinated performance.  This would be sufficient in a case
   where the required function is limited to state synchronization.  It
   may additionally request the negotiation counterpart to make a
   matching configuration in order to set up a successful communication
   with it.  It may request a certain simulation or forecast result by
   sending some dry run conditions.  The details will be defined
   separately for each type of negotiation objective.

Carpenter, et al.        Expires April 16, 2015                [Page 17]
Internet-Draft                GDN Protocol                  October 2014

   If the counterpart can immediately apply the requested configuration,
   it will give a positive (yes) answer.  This will normally end the
   negotiation phase immediately.  Otherwise it will give a negative
   (no) answer.  Normally, this will not end the negotiation phase.

   In the negative (no) case, the negotiation counterpart should be able
   to reply with a proposed alternative configuration that it can apply
   (typically, a configuration that uses fewer resources than requested
   by the negotiation initiator).  This will start a bi-directional
   negotiation to reach a compromise between the two network devices.

   The negotiation procedure is ended when one of the negotiation peers
   sends a Negotiation Ending message, which contains an accept or
   decline option and does not need a response from the negotiation
   peer.

   A negotiation procedure concerns one objective and one counterpart.
   Both the initiator and the counterpart may take part in simultaneous
   negotiations with various other devices, or in simultaneous
   negotiations about different objectives.  Thus, GDNP is expected to
   be used in a multi-threaded mode.  Certain negotiation objectives may
   have restrictions on multi-threading, for example to avoid over-
   allocating resources.

5.3.  GDNP Constants

   o  ALL_GDNP_NEIGHBOR (TBD1)

      A link-local scope multicast address used by a GDNP-enabled router
      to discover GDNP-enabled neighbor (i.e., on-link) devices . All
      routers that support GDNP are members of this multicast group.

      *  IPv6 multicast address: TBD1

      *  IPv4 multicast address: TBD2

   o  GDNP Listen Port (TBD3)

      A UDP port that every GDNP-enabled network device always listens
      to.

5.4.  Device Identifier and Certificate Tag

   A GDNP-enabled Device MUST generate a stable public/private key pair
   before it participates in GDNP.  There MUST NOT be any way of
   accessing the private key via the network or an operator interface.
   The device then uses the public key as its identifier, which is

Carpenter, et al.        Expires April 16, 2015                [Page 18]
Internet-Draft                GDN Protocol                  October 2014

   cryptographic in nature.  It is a GDNP unique identifier for a GDNP
   participant.

   It then gets a certificate for this public key, signed by a
   Certificate Authority that is trusted by other network devices.  The
   Certificate Authority SHOULD be managed by the network administrator,
   to avoid needing to trust a third party.  The signed certificate
   would be used for authentication of the message sender.  In a managed
   network, this certification process could be performed at a central
   location before the device is physically installed at its intended
   location.  In an unmanaged network, this process must be autonomic,
   including the bootstrap phase.

   A 128-bit Device Certifcate Tag, which is generated by taking a
   cryptographic hash over the device certificate, is a short
   presentation for GDNP messages.  It is the index key to find the
   device certificate in a recipient's local trusted certificate record.

   The tag value is formed by taking a SHA-1 hash algorithm over the
   corresponding device certificate and taking the leftmost 128 bits of
   the hash result.

5.5.  Session Identifier

   A 24-bit opaque value used to distinguish multiple sessions between
   the same two devices.  A new Session ID SHOULD be generated for every
   new Request message.  All follow-up messages in the same negotiation
   procedure, which is initiated by the request message, SHOULD carry
   the same Session ID.

   The Session ID SHOULD have a very low collision rate locally.  It is
   RECOMMENDED to be generated by a pseudo-random algorithm using a seed
   which is unlikely to be used by any other device in the same network.

5.6.  GDNP Messages

   This document defines the following GDNP message format and types.
   Message types not listed here are reserved for future use.  The
   numeric encoding for each message type is shown in parentheses.

5.6.1.  GDNP Message Format

   All GDNP messages share an identical fixed format header and a
   vaiable format area for options.  Every Message carries the Device
   Certificate Tag of its sender and a Session ID.  Options are
   presented serially in the options field, with no padding between the
   options.  Options are byte-aligned.

Carpenter, et al.        Expires April 16, 2015                [Page 19]
Internet-Draft                GDN Protocol                  October 2014

   The following diagram illustrates the format of GDNP messages:

    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
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   | MESSAGE_TYPE  |                Session ID                     |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                                                               |
   |                   Device Certificate Tag                      |
   |                                                               |
   |                                                               |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                        Options  (variable length)             |
   .                                                               .
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   MESSAGE_TYPE   Identifies the GDNP message type. 8-bit.

   Session ID     Identifies this negotiation session, as defined in
                  Section 6. 24-bit.

   Device Certificate Tag
                  Present the Device Certificate, which identifies
                  the negotiation devices, as defined in Section 5.4.
                  The Device Certificate Tag is 128 bit, also defined
                  in Section 5. It is used as index key to find the
                  device certificate.

   Options        GDNP Options carried in this message. Options are
                  definded in Section 5.7, 5.8 and 5.9.

5.6.2.  Discovery Message

Carpenter, et al.        Expires April 16, 2015                [Page 20]
Internet-Draft                GDN Protocol                  October 2014

DISCOVERY (1)  A discovery initiator sends a DISCOVERY message
               to initiate a discovery process.

               The discovery initiator sends the DISCOVERY
               messages to the link-local ALL_GDNP_NEIGHBOR multicast
               address for discovery, and stores the discovery
               results (including responding discovery objectives and
               corresponding unicast addresses or FQDNs).

               A DISCOVERY message MUST include a discovery objective
               option defined in Section 5.8.

               A DISCOVERY message MAY include one or more negotiation
               objective option(s) (defined in Section 5.9) to indicate
               the discovery objective that it could directly return to
               the discovery initiatior with a Negotiation message for
               rapid processing, if the discovery objective could act as
               the corresponding negotiation counterpart.

5.6.3.  Response Message

RESPONSE (2)   A node which receives a DISCOVERY message sends a
               Response message to respond to a discovery.

               If the responding node itself is the discovery objective
               of the discovery, it MUST include at least one kind of
               locator option (defined in 5.7.8) to indicate its own
               location. A combination of multiple kinds of locator
               options (e.g. IP address option + FQDN option) is also
               valid.

               If the responding node itself is NOT the discovery
               objective, but it knows the locator of the discovery
               objective, then it SHOULD respond to the discovery with a
               divert option (defined in 5.7.2) embedding a locator
               option or a combination of multiple kinds of locator
               options which indicate the locator(s) of the discovery
               objective.

5.6.4.  Request Message

  REQUEST (3)    A negotiation requesting node sends the REQUEST message
                 to the unicast address (directly stored or resolved
                 from the FQDN) of the negotiation counterpart (selected
                 from the discovery results).

Carpenter, et al.        Expires April 16, 2015                [Page 21]
Internet-Draft                GDN Protocol                  October 2014

5.6.5.  Negotiation Message

   NEGOTIATION (4)A negotiation counterpart sends a NEGOTIATION
                  message in response to a REQUEST message, a
                  Negotiation message, or a DISCOVERY message
                  in a negotiation process which may need
                  multiple steps.

5.6.6.  Negotiation-ending Message

   NEGOTIATION-ENDING (5)
                  A negotiation counterpart sends an NEGOTIATION-EDNING
                  message to close the negotiation. It MUST contain
                  one, but only one of accept/decline option,
                  defined in Section 8. It could be sent either by the
                  requesting node or the responding node.

5.6.7.  Confirm-waiting Message

   CONFIRM-WAITING (6)
                  A responding node sends a CONFIRM-WAITING message to
                  indicate the requesting node to wait for a further
                  negotiation response. It might be that the local
                  process needs more time or that the negotiation
                  depends on another triggered negotiation. This
                  message MUST NOT include any other options than the
                  WAITING option defined in Section 8.5.

5.7.  GDNP General Options

   This section defines the GDNP general option for the negotiation
   protocol signalling.  Option type 10~64 is reserved for GDNP general
   options defined in the future.

5.7.1.  Format of GDNP Options

Carpenter, et al.        Expires April 16, 2015                [Page 22]
Internet-Draft                GDN Protocol                  October 2014

    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
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |          option-code          |           option-len          |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                          option-data                          |
   |                      (option-len octets)                      |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Option-code    An unsigned integer identifying the specific option
                  type carried in this option.

   Option-len     An unsigned integer giving the length of the
                  option-data field in this option in octets.

   Option-data    The data for the option; the format of this data
                  depends on the definition of the option.

   GDNP options are scoped by using encapsulation.  If an option
   contains other options, the outer Option-len includes the total size
   of the encapsulated options, and the latter apply only to the outer
   option.

5.7.2.  Divert Option

   The divert option is used to redirect a GDNP request to another node,
   which may be more appropriate for the intended negotiation.  It may
   redirect to an entity that is known as a specific negotiation
   counterpart or a default gateway or a hierarchically upstream
   devices.  The divert option MUST only be encapsulated in Negotiation-
   ending messages.  If found elsewhere, it SHOULD be silently ignored.

Carpenter, et al.        Expires April 16, 2015                [Page 23]
Internet-Draft                GDN Protocol                  October 2014

    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
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |         OPTION_DIVERT         |           option-len          |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |             Locator Option (s) of Diversion Device(s)         |
   .                                                               .
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Option-code    OPTION_DIVERT (1).

   Option-len     The total length of diverted destination
                  sub-option(s) in octets.

   Locator Option (s) of Diverted Device(s)
                  Embedded Locator Option(s), defined in Section 5.7.8,
                  that point to diverted destination device(s).

5.7.3.  Accept Option

   The accept option is used to indicate the negotiation counterpart
   that the proposed negotiation content is accepted.

   The accept option MUST only be encapsulated in Negotiation-ending
   messages.  If found elsewhere, it SHOULD be silently ignored.

    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
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |        OPTION_ACCEPT          |           option-len          |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Option-code    OPTION_ACCEPT (2).

   Option-len     0.

5.7.4.  Decline Option

   The decline option is used to indicate the negotiation counterpart
   the proposed negotiation content is declined and end the negotiation
   process.

   The decline option MUST only be encapsulated in Negotiation-ending
   messages.  If found elsewhere, it SHOULD be silently ignored.

Carpenter, et al.        Expires April 16, 2015                [Page 24]
Internet-Draft                GDN Protocol                  October 2014

    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
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |        OPTION_DECLINE         |           option-len          |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Option-code    OPTION_DECLINE (3).

   Option-len     0.

   Notes: there are scenarios where a negotiation counterpart wants to
   decline the proposed negotiation content and continue the negotiation
   process.  For these scenarios, the negotiation counterpart SHOULD use
   a Response message, with either an objective option that contains at
   least one data field with all bits set to 1 to indicate a meaningless
   initial value, or a specific objective option that provides further
   conditions for convergence.

5.7.5.  Waiting Time Option

   The waiting time option is used to indicate that the negotiation
   counterpart needs to wait for a further negotiation response, since
   the processing might need more time than usual or it might depend on
   another triggered negotiation.

   The waiting time option MUST only be encapsulated in Confirm-waiting
   messages.  If found elsewhere, it SHOULD be silently ignored.

   The counterpart SHOULD send a Response message or another Confirm-
   waiting message before the current waiting time expires.  If not, the
   initiator SHOULD abandon or restart the negotiation procedure, to
   avoid an indefinite wait.

    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
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |       OPTION_WAITING          |           option-len          |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                              Time                             |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Option-code    OPTION_WAITING (4).

   Option-len     4, in octets.

   Time           The time is counted in millisecond as a unit.

Carpenter, et al.        Expires April 16, 2015                [Page 25]
Internet-Draft                GDN Protocol                  October 2014

5.7.6.  Certificate Option

   The Certificate option carries the certificate of the sender.  The
   format of the Certificate option is as follows:

      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
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |       OPTION Certificate      |           option-len          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                                                               |
     .                    Certificate (variable length)              .
     .                                                               .
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     Option-code    OPTION_CERT_PARAMETER (5)

     Option-len     Length of certificate in octets

     Public key     A variable-length field containing a certificate

5.7.7.  Signature Option

   The Signature option allows public key-based signatures to be
   attached to a GDNP message.  The Signature option is REQUIRED in
   every GDNP message and could be any place within the GDNP message.
   It protects the entire GDNP header and options.  A TimeStamp has been
   integrated in the Signature Option for anti-replay protection.  The
   format of the Signature option is described as follows:

    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
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |     OPTION_SIGNATURE          |           option-len          |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |           HA-id               |            SA-id              |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                     Timestamp (64-bit)                        |
   |                                                               |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                                                               |
   .                    Signature (variable length)                .
   .                                                               .
   |                                                               |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Option-code    OPTION_SIGNATURE (6)

Carpenter, et al.        Expires April 16, 2015                [Page 26]
Internet-Draft                GDN Protocol                  October 2014

   Option-len     12 + Length of Signature field in octets.

   HA-id          Hash Algorithm id. The hash algorithm is used for
                  computing the signature result. This design is
                  adopted in order to provide hash algorithm agility.
                  The value is from the Hash Algorithm for GDNP
                  registry in IANA. The initial value assigned
                  for SHA-1 is 0x0001.

   SA-id          Signature Algorithm id. The signature algorithm is
                  used for computing the signature result. This
                  design is adopted in order to provide signature
                  algorithm agility. The value is from the Signature
                  Algorithm for GDNP registry in IANA. The initial
                  value assigned for RSASSA-PKCS1-v1_5 is
                  0x0001.

   Timestamp      The current time of day (NTP-format timestamp
                  [RFC5905] in UTC (Coordinated Universal Time), a
                  64-bit unsigned fixed-point number, in seconds
                  relative to 0h on 1 January 1900.). It can reduce
                  the danger of replay attacks.

   Signature      A variable-length field containing a digital
                  signature. The signature value is computed with
                  the hash algorithm and the signature algorithm, as
                  described in HA-id and SA-id. The signature
                  constructed by using the sender's private key
                  protects the following sequence of octets:

                  1. The GDNP message header.

                  2. All GDNP options including the Signature option
                  (fill the signature field with zeroes).

                  The signature field MUST be padded, with all 0, to
                  the next 16 bit boundary if its size is not an even
                  multiple of 8 bits. The padding length depends on
                  the signature algorithm, which is indicated in the
                  SA-id field.

5.7.8.  Locator Options

   These locator options are used to present a device's or interface's
   reachability information.  They are Locator IPv4 Address Option,
   Locator IPv6 Address Option and Locator FQDN (Fully Qualified Domain
   Name) Option.

Carpenter, et al.        Expires April 16, 2015                [Page 27]
Internet-Draft                GDN Protocol                  October 2014

5.7.8.1.  Locator IPv4 address option

    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
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |    OPTION_LOCATOR_IPV4ADDR    |           option-len          |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                          IPv4-Address                         |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Option-code    OPTION_LOCATOR_IPV4ADDR (7)

   Option-len     4, in octets.

   IPv4-Address   The IPv4 address locator of the device/interface.

5.7.8.2.  Locator IPv6 address option

    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
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |   OPTION_LOCATOR_IPV6ADDR     |           option-len          |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                                                               |
   |                          IPv6-Address                         |
   |                                                               |
   |                                                               |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Option-code    OPTION_LOCATOR_IPV6ADDR (8).

   Option-len     16, in octets.

   IPv6-Address   The IPv6 address locator of the device/interface.

   Note: link-local IPv6 address SHOULD be avoided when this option is
   used in the Divert option.  It may create a connection problem.

5.7.8.3.  Locator FQDN option

Carpenter, et al.        Expires April 16, 2015                [Page 28]
Internet-Draft                GDN Protocol                  October 2014

    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
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |         OPTION_FQDN           |           option-len          |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                   Fully Qualified Domain Name                 |
   |                       (variable length)                       |
   .                                                               .
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Option-code    OPTION_FQDN (9).

   Option-len     Length of Fully Qualified Domain Name in octets.

   Domain-Name    The Fully Qualified Domain Name of the entity.

5.8.  Discovery Objective Option

   The discovery objective option is to express the discovery objectives
   that the initiating node wants to discover.

    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
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |         OPTION_DISOBJ         |           option-len          |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |             Expression of Discovery Objectives (TBD)          |
   .                                                               .
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   Option-code    OPTION_DISOBJ (TBD).

   Option-len     The total length in octets.

   Expression of Discovery Objectives (TBD)
                  This field is to express the discovery objectives
                  that the initiating node wants to discover. It might
                  be network functionality, role-based network element
                  or service agent.

5.9.  Negotiation Objective Options and Considerations

   The Negotiation Objective Options contain negotiation objectives,
   which are various according to different functions/services.  They
   MUST be carried by Discovery, Request or Negotiation Messages only.
   Objective options SHOULD be assigned an option type greater than 64
   in the GDNP option table.

Carpenter, et al.        Expires April 16, 2015                [Page 29]
Internet-Draft                GDN Protocol                  October 2014

   For most scenarios, there SHOULD be initial values in the negotiation
   requests.  Consequently, the Objective options SHOULD always be
   completely presented in a Request message.  If there is no initial
   value, the bits in the value field SHOULD all be set to 1 to indicate
   a meaningless value, unless this is inappropriate for the specific
   negotiation objective.

5.9.1.  Organizing of GDNP Options

   Naturally, a negotiation objective, which is based on a specific
   service or function or action, SHOULD be organized as a single GDNP
   option.  It is NOT RECOMMENDED to organize multiple negotiation
   objectives into a single option.

   A negotiation objective may have multiple parameters.  Parameters can
   be categorized into two class: the obligatory ones presented as fixed
   fields; and the optional ones presented in TLV sub-options.  It is
   NOT RECOMMENDED to split parameters in a single objective into
   multiple options, unless they have different response periods.  An
   exception scenario may also be described by split objectives.

5.9.2.  Vendor Specific Options

   Option codes 128~159 have been reserved for vendor specific options.
   Multiple option codes have been assigned because a single vendor may
   use multiple options simultaneously.  These vendor specific options
   are highly likely to have different meanings when used by different
   vendors.  Therefore, they SHOULD NOT be used without an explicit
   human decision.  They are not suitable for unmanaged networks such as
   home networks.

5.9.3.  Experimental Options

   Option code 176~191 have been reserved for experimental options.
   Multiple option codes have been assigned because a single experiment
   may use multiple options simultaneously.  These experimental options
   are highly likely to have different meanings when used for different
   experiments.  Therefore, they SHOULD NOT be used without an explicit
   human decision.  They are not suitable for unmanaged networks such as
   home networks.

5.10.  Items for Future Work

   There are a few open design questions that are worthy of more work in
   the near future, as listed below:

   o  UDP vs TCP: For now, this specification has chosen UDP as message
      transport mechanism.  However, this is not closed yet.  UDP is

Carpenter, et al.        Expires April 16, 2015                [Page 30]
Internet-Draft                GDN Protocol                  October 2014

      good for short conversations, fitting the divert scenarios well.
      However, it may have issues with large packets.  TCP is good for
      stable and long sessions, with a little bit of time consumption
      during the session establishment stage.  If messages exceed a
      reasonable MTU, a TCP mode may be necessary.

   o  Message encryption: should GDNP messages be (optionally) encrypted
      as well as signed, to protect against internal eavesdropping or
      monitoring within the network?

   o  TLS or DTLS vs built-in security mechanism.  For now, this
      specification has chosen a PKI based build-in security mechanism.
      However, TLS or DTLS might be chosen as security infrastructure
      for simplification reasons.

   o  Timeout for lost Negotiation Ending and other messages to be
      added.

   o  GDNP currently requires every participant to have an NTP-
      synchronized clock.  Is this OK for low-end devices?

   o  Would use of MDNS have any impact on the Locator FQDN option?

   o  Use case.  A use case may help readers to understand the
      applicability of this specification.  However, the authors have
      not yet decided whether to have a separate document or have it in
      this document.  General uses cases for AN have been developed, but
      they are not specific enough for this purpose.

   o  Rules about how data items are defined in a negotiation objective.
      Maybe a formal information model is needed.

   o  We currently assume that there is only one counterpart for each
      discovery action.  If this is false or one negotiation request
      receives multiple different responses, how does the initiator
      choose between them?  Could it split them into multiple follow-up
      negotiations?

   o  Alternatives to TLV format.  It may be useful to provide a generic
      method of carrying negotiation objectives in a high-level format
      such as YANG or XML schema.  It may also be useful to provide a
      generic method of carrying existing configuration information such
      as DHCP(v6) or IPv6 RA messages.  These features could be provided
      by encapsulating such messages in their own TLVs.

Carpenter, et al.        Expires April 16, 2015                [Page 31]
Internet-Draft                GDN Protocol                  October 2014

6.  Security Considerations

   It is obvious that a successful attack on negotiation-enabled nodes
   would be extremely harmful, as such nodes might end up with a
   completely undesirable configuration.  Security considerations are in
   the following aspects as the following.

   - Authentication

      A cryptographically authenticated identity for each device is
      needed in an autonomic network.  It is not safe to assume that a
      large network is physically secured against interference or that
      all personnel are trustworthy.  Each autonomic device should be
      capable of proving its identity and authenticating its messages.
      One approach for the negotiation protocol is using certificate-
      based security mechanism and its verification mechanism in GDNP
      message exchanging provides the authentication and data integrity
      protection.

      The timestamp mechanism provides an anti-replay function.

      Since GDNP is intended to be deployed in a single administrative
      domain recommended to operate its own trust anchor and CA, there
      is no need for a trusted public third party.

   - Privacy

      Generally speaking, no personal information is expected to be
      involved in the negotiation protocol, so there should be no direct
      impact on personal privacy.  Nevertheless, traffic flow paths,
      VPNs, etc. may be negotiated, which could be of interest for
      traffic analysis.  Also, carriers generally want to conceal
      details of their network topology and traffic density from
      outsiders.  Therefore, since insider attacks cannot be prevented
      in a large carrier network, the security mechanism for the
      negotiation protocol needs to provide message confidentiality.

   - DoS Attack Protection

      TBD.

7.  IANA Considerations

   Section 5.3 defines the following multicast addresses, which have
   been assigned by IANA for use by GDNP:

   ALL_GDNP_NEIGHBOR multicast address  (IPv6): (TBD1)

Carpenter, et al.        Expires April 16, 2015                [Page 32]
Internet-Draft                GDN Protocol                  October 2014

   ALL_GDNP_NEIGHBOR multicast address  (IPv4): (TBD2)

   Section 5.3 defines the following UDP port, which have been assigned
   by IANA for use by GDNP:

   GDNP Listen Port:  (TBD3)

   This document defined a new General Discovery and Negotiation
   Protocol.  The IANA is requested to create a new GDNP registry.  The
   IANA is also requested to add two new registry tables to the newly-
   created GDNP registry.  The two tables are the GDNP Messages table
   and GDNP Options table.

   Initial values for these registries are given below.  Future
   assignments are to be made through Standards Action or Specification
   Required [RFC5226].  Assignments for each registry consist of a type
   code value, a name and a document where the usage is defined.

   GDNP Messages table.  The values in this table are 16-bit unsigned
   integers.  The following initial values are assigned in Section 5.6
   in this document:

         Type  |          Name               |   RFCs
      ---------+-----------------------------+------------
           0   |Reserved                     | this document
           1   |Request Message              | this document
           2   |Negotiation Message          | this document
           3   |Negotiation-end Message      | this document
           4   |Confirm-waiting Message      | this document

   GDNP Options table.  The values in this table are 16-bit unsigned
   integers.  The following initial values are assigned in Section 5.7
   and Section 5.9 in this document:

Carpenter, et al.        Expires April 16, 2015                [Page 33]
Internet-Draft                GDN Protocol                  October 2014

         Type  |          Name               |   RFCs
      ---------+-----------------------------+------------
           0   |Reserved                     | this document
           1   |Divert Option                | this document
           2   |Accept Option                | this document
           3   |Decline Option               | this document
           4   |Waiting Time Option          | this document
           5   |Certificate Option           | this document
           6   |Sigature Option              | this document
           7   |Device IPv4 Address Option   | this document
           8   |Device IPv6 Address Option   | this document
           9   |Device FQDN Option           | this document
        10~64  |Reserved for future CDNP     | this document
               |General Options              |
       128~159 |Vendor Specific Options      | this document
       176~191 |Experimental Options         | this document

   The IANA is also requested to create two new registry tables in the
   GDNP Parameters registry.  The two tables are the Hash Algorithm for
   GDNP table and the Signature Algorithm for GDNP table.

   Initial values for these registries are given below.  Future
   assignments are to be made through Standards Action or Specification
   Required [RFC5226].  Assignments for each registry consist of a name,
   a value and a document where the algorithm is defined.

   Hash Algorithm for GDNP.  The values in this table are 16-bit
   unsigned integers.  The following initial values are assigned for
   Hash Algorithm for GDNP in this document:

             Name          |  Value    |  RFCs
      ---------------------+-----------+------------
            Reserved       |   0x0000  | this document
            SHA-1          |   0x0001  | this document
            SHA-256        |   0x0002  | this document

   Signature Algorithm for GDNP.  The values in this table are 16-bit
   unsigned integers.  The following initial values are assigned for
   Signature Algorithm for GDNP in this document:

             Name          |   Value   |  RFCs
      ---------------------+-----------+------------
            Reserved       |   0x0000  | this document
       RSASSA-PKCS1-v1_5   |   0x0001  | this document

Carpenter, et al.        Expires April 16, 2015                [Page 34]
Internet-Draft                GDN Protocol                  October 2014

8.  Acknowledgements

   Valuable comments were received from Zhenbin Li, Dacheng Zhang, Rene
   Struik, Dimitri Papadimitriou, and other participants in the ANIMA
   and NMRG working group.

   This document was produced using the xml2rfc tool [RFC2629].

9.  Change log [RFC Editor: Please remove]

   draft-carpenter-anima-discovery-negotiation-protocol-00, combination
   of draft-jiang-config-negotiation-ps-03 and draft-jiang-config-
   negotiation-protocol-02, 2014-10-08.

10.  References

10.1.  Normative References

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

   [RFC5280]  Cooper, D., Santesson, S., Farrell, S., Boeyen, S.,
              Housley, R., and W. Polk, "Internet X.509 Public Key
              Infrastructure Certificate and Certificate Revocation List
              (CRL) Profile", RFC 5280, May 2008.

10.2.  Informative References

   [I-D.chaparadza-intarea-igcp]
              Behringer, M., Chaparadza, R., Petre, R., Li, X., and H.
              Mahkonen, "IP based Generic Control Protocol (IGCP)",
              draft-chaparadza-intarea-igcp-00 (work in progress), July
              2011.

   [I-D.ietf-homenet-hncp]
              Stenberg, M. and S. Barth, "Home Networking Control
              Protocol", draft-ietf-homenet-hncp-01 (work in progress),
              June 2014.

   [I-D.ietf-netconf-restconf]
              Bierman, A., Bjorklund, M., and K. Watsen, "RESTCONF
              Protocol", draft-ietf-netconf-restconf-02 (work in
              progress), October 2014.

   [I-D.irtf-nmrg-an-gap-analysis]
              Jiang, S., Carpenter, B., and M. Behringer, "Gap Analysis
              for Autonomic Networking", draft-irtf-nmrg-an-gap-
              analysis-02 (work in progress), October 2014.

Carpenter, et al.        Expires April 16, 2015                [Page 35]
Internet-Draft                GDN Protocol                  October 2014

   [I-D.irtf-nmrg-autonomic-network-definitions]
              Behringer, M., Pritikin, M., Bjarnason, S., Clemm, A.,
              Carpenter, B., Jiang, S., and L. Ciavaglia, "Autonomic
              Networking - Definitions and Design Goals", draft-irtf-
              nmrg-autonomic-network-definitions-04 (work in progress),
              October 2014.

   [RFC2205]  Braden, B., Zhang, L., Berson, S., Herzog, S., and S.
              Jamin, "Resource ReSerVation Protocol (RSVP) -- Version 1
              Functional Specification", RFC 2205, September 1997.

   [RFC2629]  Rose, M., "Writing I-Ds and RFCs using XML", RFC 2629,
              June 1999.

   [RFC2865]  Rigney, C., Willens, S., Rubens, A., and W. Simpson,
              "Remote Authentication Dial In User Service (RADIUS)", RFC
              2865, June 2000.

   [RFC3209]  Awduche, D., Berger, L., Gan, D., Li, T., Srinivasan, V.,
              and G. Swallow, "RSVP-TE: Extensions to RSVP for LSP
              Tunnels", RFC 3209, December 2001.

   [RFC3315]  Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C.,
              and M. Carney, "Dynamic Host Configuration Protocol for
              IPv6 (DHCPv6)", RFC 3315, July 2003.

   [RFC3416]  Presuhn, R., "Version 2 of the Protocol Operations for the
              Simple Network Management Protocol (SNMP)", STD 62, RFC
              3416, December 2002.

   [RFC4270]  Hoffman, P. and B. Schneier, "Attacks on Cryptographic
              Hashes in Internet Protocols", RFC 4270, November 2005.

   [RFC4861]  Narten, T., Nordmark, E., Simpson, W., and H. Soliman,
              "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861,
              September 2007.

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

   [RFC5905]  Mills, D., Martin, J., Burbank, J., and W. Kasch, "Network
              Time Protocol Version 4: Protocol and Algorithms
              Specification", RFC 5905, June 2010.

   [RFC5971]  Schulzrinne, H. and R. Hancock, "GIST: General Internet
              Signalling Transport", RFC 5971, October 2010.

Carpenter, et al.        Expires April 16, 2015                [Page 36]
Internet-Draft                GDN Protocol                  October 2014

   [RFC6241]  Enns, R., Bjorklund, M., Schoenwaelder, J., and A.
              Bierman, "Network Configuration Protocol (NETCONF)", RFC
              6241, June 2011.

   [RFC6733]  Fajardo, V., Arkko, J., Loughney, J., and G. Zorn,
              "Diameter Base Protocol", RFC 6733, October 2012.

   [RFC6887]  Wing, D., Cheshire, S., Boucadair, M., Penno, R., and P.
              Selkirk, "Port Control Protocol (PCP)", RFC 6887, April
              2013.

Authors' Addresses

   Brian Carpenter
   Department of Computer Science
   University of Auckland
   PB 92019
   Auckland  1142
   New Zealand

   Email: brian.e.carpenter@gmail.com

   Sheng Jiang
   Huawei Technologies Co., Ltd
   Q14, Huawei Campus
   No.156 Beiqing Road
   Hai-Dian District, Beijing  100095
   P.R. China

   Email: jiangsheng@huawei.com

   Bing Liu
   Huawei Technologies Co., Ltd
   Q14, Huawei Campus
   No.156 Beiqing Road
   Hai-Dian District, Beijing  100095
   P.R. China

   Email: leo.liubing@huawei.com

Carpenter, et al.        Expires April 16, 2015                [Page 37]