Skip to main content

Reactive Discovery of Point-to-Point Routes in Low Power and Lossy Networks
draft-ietf-roll-p2p-rpl-08

The information below is for an old version of the document.
Document Type
This is an older version of an Internet-Draft that was ultimately published as RFC 6997.
Authors Mukul Goyal , Emmanuel Baccelli , Matthias Philipp , Anders Brandt , Jerry Martocci
Last updated 2012-03-02
RFC stream Internet Engineering Task Force (IETF)
Formats
Reviews
Additional resources Mailing list discussion
Stream WG state WG Document
Document shepherd (None)
IESG IESG state Became RFC 6997 (Experimental)
Consensus boilerplate Unknown
Telechat date (None)
Responsible AD (None)
Send notices to (None)
draft-ietf-roll-p2p-rpl-08
Internet Engineering Task Force                            M. Goyal, Ed.
Internet-Draft                                   University of Wisconsin
Intended status: Experimental                                  Milwaukee
Expires: September 3, 2012                                   E. Baccelli
                                                              M. Philipp
                                                                   INRIA
                                                               A. Brandt
                                                           Sigma Designs
                                                             J. Martocci
                                                        Johnson Controls
                                                           March 2, 2012

   Reactive Discovery of Point-to-Point Routes in Low Power and Lossy
                                Networks
                       draft-ietf-roll-p2p-rpl-08

Abstract

   This document specifies a point-to-point route discovery mechanism,
   complementary to the RPL core functionality.  This mechanism allows
   an IPv6 router to discover "on demand" routes to one or more IPv6
   routers in the LLN such that the discovered routes meets specified
   metrics constraints.

Status of this Memo

   This Internet-Draft is submitted to IETF 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 September 3, 2012.

Copyright Notice

   Copyright (c) 2012 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

Goyal, et al.           Expires September 3, 2012               [Page 1]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

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

Table of Contents

   1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  3
   2.  The Use Cases  . . . . . . . . . . . . . . . . . . . . . . . .  3
   3.  Terminology  . . . . . . . . . . . . . . . . . . . . . . . . .  4
   4.  Applicability  . . . . . . . . . . . . . . . . . . . . . . . .  5
   5.  Functional Overview  . . . . . . . . . . . . . . . . . . . . .  5
   6.  P2P Route Discovery Mode Of Operation  . . . . . . . . . . . .  8
     6.1.  Setting a P2P Mode DIO . . . . . . . . . . . . . . . . . .  8
   7.  New RPL Control Message Options  . . . . . . . . . . . . . . . 11
     7.1.  P2P Route Discovery Option (P2P-RDO) . . . . . . . . . . . 11
     7.2.  Data Option  . . . . . . . . . . . . . . . . . . . . . . . 14
   8.  The Discovery Reply Object (DRO) . . . . . . . . . . . . . . . 14
     8.1.  Secure DRO . . . . . . . . . . . . . . . . . . . . . . . . 16
     8.2.  Setting a P2P-RDO Carried in a Discovery Reply Object  . . 16
   9.  P2P-RPL Route Discovery By Creating a Temporary DAG  . . . . . 17
     9.1.  Joining a Temporary DAG  . . . . . . . . . . . . . . . . . 17
     9.2.  Trickle Operation For P2P Mode DIOs  . . . . . . . . . . . 18
     9.3.  Processing a P2P Mode DIO  . . . . . . . . . . . . . . . . 19
     9.4.  Additional Processing of a P2P Mode DIO At An
           Intermediate Router  . . . . . . . . . . . . . . . . . . . 20
     9.5.  Additional Processing of a P2P Mode DIO At The Target  . . 21
     9.6.  Processing a DRO At An Intermediate Router . . . . . . . . 22
     9.7.  Processing a DRO At The Origin . . . . . . . . . . . . . . 23
   10. The Discovery Reply Object Acknowledgement (DRO-ACK) . . . . . 24
   11. Packet Forwarding Along a Route Discovered Using P2P-RPL . . . 25
   12. Constants  . . . . . . . . . . . . . . . . . . . . . . . . . . 25
   13. Interoperability with Core RPL . . . . . . . . . . . . . . . . 26
   14. Security Considerations  . . . . . . . . . . . . . . . . . . . 26
   15. IANA Considerations  . . . . . . . . . . . . . . . . . . . . . 27
     15.1. Additions to DIO Mode of Operation . . . . . . . . . . . . 27
     15.2. Additions to RPL Control Message Options . . . . . . . . . 27
     15.3. Additions to RPL Control Codes . . . . . . . . . . . . . . 27
   16. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 28
   17. References . . . . . . . . . . . . . . . . . . . . . . . . . . 28
     17.1. Normative References . . . . . . . . . . . . . . . . . . . 28
     17.2. Informative References . . . . . . . . . . . . . . . . . . 29
   Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 29

Goyal, et al.           Expires September 3, 2012               [Page 2]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

1.  Introduction

   Targeting Low power and Lossy Networks (LLNs), the RPL routing
   protocol [I-D.ietf-roll-rpl] provides paths along a Directed Acyclic
   Graph (DAG) rooted at a single router in the network.  Establishment
   and maintenance of a DAG is performed by routers in the LLN using
   DODAG Information Object (DIO) messages.  When two arbitrary routers
   (neither of which is the DAG's root) need to communicate, the data
   packets are restricted to travel only along the links in the DAG.
   Such point-to-point (P2P) routing functionality may not be sufficient
   for several Home and Building Automation applications [RFC5826]
   [RFC5867] due to the following reasons:

   o  The need to pre-establish routes: each potential destination in
      the network must declare itself as such ahead of the time a source
      needs to reach it.

   o  The need to route only along the links in the DAG: A DAG is built
      to optimize the routing cost to reach the root.  Restricting P2P
      routes to use only the in-DAG links may result in significantly
      suboptimal routes and severe traffic congestion near the DAG root.

   This document describes an extension to core RPL that enables an IPv6
   router in the LLN to discover routes to one or more IPv6 routers in
   the LLN "on demand", such that the discovered routes meet the
   specified metrics constraints, without necessarily going along the
   links in an existing DAG.  This reactive P2P route discovery
   mechanism is henceforth referred to as P2P-RPL.  P2P-RPL does not
   guarantee discovery of a route.  Also, the discovered routes may not
   be the best available.  However, any discovered routes are guaranteed
   to satisfy the desired constraints in terms of the routing metrics
   and are thus considered "good enough" from the application's
   perspective.

   A mechanism to measure the end-to-end cost of an existing route has
   been specified in [I-D.ietf-roll-p2p-measurement].  As discussed in
   Section 4, measuring the end-to-end cost of an existing route may
   help decide whether to initiate the discovery of a better route using
   P2P-RPL and the metric constraints to be used for this purpose.

2.  The Use Cases

   One use case, common in home and commercial building environments,
   involves a device (say a remote control or an airduct controller)
   that suddenly needs to communicate with another device (say a lamp or
   a humidity sensor) to which it does not already have a route.  In
   this case, the remote control (or the airduct controller) must be

Goyal, et al.           Expires September 3, 2012               [Page 3]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   able to discover a route to the lamp (or the humidity sensor) "on
   demand".

   Another use case, common in a commercial building environment,
   involves a large LLN deployment where P2P communication along a
   particular DAG among hundreds (or thousands) of routers creates
   severe traffic congestion near that DAG's root, and thus routes
   across this DAG are desirable.

   Other use cases involve scenarios where energy or latency constraints
   are not satisfied by the P2P routes along an existing DAG because
   they involve traversing many more intermediate routers than necessary
   to reach the destination.

3.  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].

   Additionally, this document uses terminology from
   [I-D.ietf-roll-terminology] and [I-D.ietf-roll-rpl].  This document
   introduces the following terms:

   Origin : The IPv6 router initiating the P2P-RPL route discovery.

   Target : The IPv6 router at the other end point of the P2P route(s)
   to be discovered.  A P2P-RPL route discovery can discover routes to
   multiple targets at the same time.

   Intermediate Router: An IPv6 router that is neither the origin nor a
   target.

   Forward Route: A route in the forward direction, i.e., from the
   origin to the target.

   Backward Route: A route in the backward direction, i.e., from the
   target to the origin.

   Bidirectional Route: A route that can be used in both forward and
   backward directions.

   Source Route: A complete and ordered list of routers that can be used
   by a packet to travel from a source to a destination node.

   Hop-by-hop Route: The route characterized by each router on the route

Goyal, et al.           Expires September 3, 2012               [Page 4]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   using its routing table to determine the next hop on the route.

4.  Applicability

   A route discovery using P2P-RPL may be performed by an origin when no
   route exists between itself and the target(s) or when the existing
   routes do not satisfy the application requirements.  P2P-RPL is
   designed to discover hop-by-hop or source routes to one or more
   targets such that the discovered routes meet the specified
   constraints.  In some application contexts, the constraints that the
   discovered routes must satisfy are intrinsically known or can be
   specified by the application.  For example, an origin that expects
   its targets to be less than 5 hops away may use "hop-count < 5" as
   the constraint.  In other application contexts, the origin may need
   to measure the cost of the existing route to a target to determine
   the constraints.  For example, an origin that measures the total ETX
   along its current route to a target to be 20 may use "ETX < x*20",
   where x is a fraction that the origin decides, as the constraint.  A
   mechanism to measure the cost of an existing route between two IPv6
   routers is specified in [I-D.ietf-roll-p2p-measurement].  If there is
   no existing route between the origin and the target(s) or the cost
   measurement for the existing routes fails, the origin will have to
   guess the constraints to be used in the initial route discovery.
   Once, the initial route discovery succeeds or fails, the origin will
   have a better estimate for the constraints to be used in the
   subsequent route discovery.

   P2P-RPL may result in discovery of better P2P routes than the ones
   available along a DAG designed to optimize routing cost to the DAG's
   root.  The improvement in route quality depends on a number of
   factors including the network topology, the routing metrics in use
   and the prevalent conditions in the network.  A network designer may
   take into consideration both the benefits (potentially better routes;
   no need to maintain routes proactively) and costs (control messages
   generated during the route discovery process) when using P2P-RPL.

5.  Functional Overview

   This section contains a high level description of P2P-RPL.

   A P2P-RPL route discovery takes place by forming a DAG rooted at the
   origin.  As is the case with core RPL, P2P-RPL uses IPv6 link-local
   multicast DIO messages to establish a DAG.  However, unlike core RPL,
   this DAG is temporary in nature and routers in the DAG leave once the
   DAG's life time is over.  The sole purpose of DAG creation is to
   discover routes to the target(s) and DIOs serve as the route

Goyal, et al.           Expires September 3, 2012               [Page 5]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   discovery messages.  Each router joining the DAG determines a rank
   for itself in the DAG and ignores the subsequent DIOs received from
   lower (higher in numerical value) ranked neighbors.  Thus, the route
   discovery messages propagate away from the origin rather than return
   back to it.  As in core RPL, DIO generation at a router is controlled
   by a Trickle timer [RFC6206] that allows a router to avoid generating
   unnecessary messages while providing protection against unreliable
   wireless communication.  P2P-RPL also uses the routing metrics
   [I-D.ietf-roll-routing-metrics], objective functions and packet
   forwarding framework
   [I-D.ietf-6man-rpl-routing-header][I-D.ietf-6man-rpl-option]
   developed for core RPL.

   An origin may use P2P-RPL to discover routes to one or more targets
   identified by one or more unicast/multicast addresses.  P2P-RPL
   allows for the discovery of one hop-by-hop route or upto four source
   routes per target.  P2P-RPL allows an origin to piggyback time-
   critical application data on the DIO messages for delivery to the
   target(s).  P2P-RPL does not guarantee discovery of a route to a
   target.  Also, the discovered routes may not be the best available.
   However, any discovered routes are guaranteed to satisfy the desired
   constraints in terms of the routing metrics and are thus considered
   "good enough" from the application's perspective.

   A P2P-RPL route discovery takes place by forming a temporary DAG
   rooted at the origin.  The DIOs, used to create the temporary DAG,
   are identified by a new Mode of Operation (P2P Route Discovery mode
   defined in Section 6).  The DIOs, listing the P2P Route Discovery
   mode as the Mode of Operation, are henceforth referred to as the P2P
   mode DIOs.  A P2P mode DIO always carries one P2P Route Discovery
   Option (defined in Section 7.1) in which the origin specifies the
   following information:

   o  The IPv6 address of a target.  This could be a unicast address or
      a multicast one.  Any additional targets may be specified by
      including one or more RPL Target Options [I-D.ietf-roll-rpl]
      inside the DIO.

   o  The nature of the route(s) to be discovered: hop-by-hop or source
      routes.  This specification allows for the discovery of one hop-
      by-hop route or up to four source routes per target.

   o  The desired number of routes (if source routes are being
      discovered).

   o  Whether the target(s) should send Discovery Reply Object (DRO)
      messages (defined in Section 8) back to the origin on receiving a
      DIO message.  A DRO message carries a discovered source route back

Goyal, et al.           Expires September 3, 2012               [Page 6]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

      to the origin or establishes a hop-by-hop route between the origin
      and the target.  By not allowing the generation of DRO messages,
      an origin can use P2P-RPL as purely a mechanism to deliver time-
      critical application data to the target(s).

   A P2P Route Discovery Option also accumulates a route from the origin
   to a target as the routers join the temporary DAG.

   A P2P mode DIO MAY also carry:

   o  One or more Metric Container Options to specify:

      *  The relevant routing metrics.

      *  The constraints that the discovered route must satisfy.  These
         constraints also limit how far the DIOs message may travel.

   o  One or more RPL Target options to specify additional unicast or
      multicast targets.

   o  One or more Data Options (defined in Section 7.2) to carry time-
      critical application-level data to be delivered to the target(s).

   As the routers join the temporary DAG, they keep track of the best
   (partial) route(s) they have seen and advertise these routes, along
   with the corresponding routing metrics, in their P2P mode DIOs.  A
   router, including the target(s), discards a received P2P mode DIO if
   the aggregated routing metrics on the route advertised by the DIO do
   not satisfy the listed constraints.  These constraints can be used to
   limit the propagation of P2P mode DIO messages.  A router may also
   discard a received P2P mode DIO if it does not wish to be a part of
   the discovered route due to limited resources or due to policy
   reasons.

   When a target receives a P2P mode DIO, it forwards the data in any
   Data Options to the higher layer.  The target may remember the
   discovered route for use as a source route to reach the origin.  If
   the origin has requested DRO messages to be sent back, the target may
   select the route contained in the received DIO for further processing
   as described next.  This document does not specify a particular
   method for the target to use to select a route for further
   processing.  Example methods include selecting any route that meets
   the constraints or selecting the best route(s) discovered over a
   certain time period.

   If one or more source routes are being discovered, the target sends
   the selected source routes to the origin via DRO messages with one
   DRO message carrying one discovered route.  On receiving a DRO

Goyal, et al.           Expires September 3, 2012               [Page 7]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   message, the origin stores the discovered route in its memory.  If a
   hop-by-hop route is being discovered, the target sends a DRO message
   containing the selected route to the origin.  The DRO message travels
   back to the origin along the selected route, establishing state for
   this route in the routers on the path.  The target may include one or
   more Data Options in a DRO message to deliver any time-critical
   application data to the origin.

   The target may request the origin to acknowledge the receipt of a DRO
   message by sending back a DRO Acknowledgement (DRO-ACK) message
   (defined in Section 10).  The origin unicasts a DRO-ACK message to
   the target.  When the target does not receive the requested DRO-ACK
   within a certain time interval of sending a DRO, it resends the DRO
   message (up to a certain number of times) carrying the same route as
   before.

   The use of trickle timers to delay the propagation of DIO messages
   may cause some nodes to generate these messages even when the desired
   routes have already been discovered.  In order to preempt the
   generation of such unnecessary messages, the target may set a "stop"
   flag in the DRO message to let the nodes in the LLN know about the
   completion of the route discovery process.  The routers receiving
   such a DRO should not generate any more DIOs for this temporary DAG.
   Neither should they process any received DIOs for this temporary DAG
   in future.  However, such routers must still process the DROs
   received for this temporary DAG.

6.  P2P Route Discovery Mode Of Operation

   This section specifies a new RPL Mode of Operation (MOP), P2P Route
   Discovery mode (or P2P mode, for short), with value 4 (to be
   confirmed by IANA).  A DIO message, listing P2P mode as the MOP, is
   identified as performing a P2P-RPL route discovery by creating a
   temporary DAG.  A P2P mode DIO MUST carry one and only one P2P Route
   Discovery Option (specified in Section 7.1).

6.1.  Setting a P2P Mode DIO

   The Base Object in a P2P mode DIO message MUST be set in the
   following manner:

   o  RPLInstanceID: RPLInstanceID MUST be a local value as described in
      Section 5.1 of [I-D.ietf-roll-rpl].  The origin MUST NOT use the
      same RPLInstanceID in two or more concurrent route discoveries.

   o  Version Number: MUST be set to zero.  The temporary DAG used for
      P2P-RPL route discovery does not exist long enough to have new

Goyal, et al.           Expires September 3, 2012               [Page 8]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

      versions.

   o  Grounded (G) Flag: MUST be cleared since this DAG is temporary in
      nature, is created solely for the purpose of P2P-RPL route
      discovery and MUST NOT be used for packet routing.

   o  Mode of Operation (MOP): MUST be set to 4, corresponding to P2P
      Route Discovery mode.

   o  DTSN: MUST be set to value zero on transmission and ignored on
      reception.

   o  DODAGPreference (Prf): This field MUST be set to value 0 (least
      preferred).

   o  DODAGID: This field MUST be set to an IPv6 address of the origin.

   o  The other fields in the DIO Base Object can be set in the desired
      fashion as per the rules described in [I-D.ietf-roll-rpl].

   The DODAG Configuration Option, inside a P2P mode DIO MUST be set in
   the following manner:

   o  MaxRankIncrease: This field MUST be set to 0 to disable local
      repair of the temporary DAG.

   o  Trickle parameters (DIOIntervalDoublings, DIOIntervalMin,
      DIORedundancyConstant) SHOULD be set as described in Section 9.2.

   o  The Default Lifetime and Lifetime Unit parameters in DODAG
      Configuration option indicate the life time of the state the
      routers maintain for a hop-by-hop route established using P2P-RPL
      and may be set as desired.

   o  The other fields in the DODAG Configuration Option, including the
      OCP identifying the Objective function, can be set in the desired
      fashion as per the rules described in [I-D.ietf-roll-rpl].

   A default DODAG Configuration Option comes in effect if a P2P mode
   DIO does not carry an explicit one.  The default DODAG Configuration
   Option has the following parameter values:

   o  Authentication Enabled: 0

   o  DIOIntervalMin: 6, which translates to 64ms as the value for Imin
      parameter in Trickle operation.

Goyal, et al.           Expires September 3, 2012               [Page 9]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   o  DIORedundancyConstant: 1

   o  MaxRankIncrease: 0

   o  Default Lifetime: 0xFF

   o  Lifetime Unit: 0xFFFF

   o  Objective Code Point: 0, i.e., OF0 [I-D.ietf-roll-of0] is the
      default objective function.

   o  The remaining parameters have default values as specified in
      [I-D.ietf-roll-rpl].

   The routing metrics and constraints [I-D.ietf-roll-routing-metrics]
   used in P2P-RPL route discovery are included in one or more Metric
   Container Options [I-D.ietf-roll-rpl] inside the P2P mode DIO.  Note
   that a DIO need not include a Metric Container if OF0 is the
   objective function in effect.  In that case, a P2P mode DIO may still
   specify an upper limit on the maximum rank, that a router may have in
   the temporary DAG, inside the P2P Route Discovery Option (described
   in Section 7.1).

   A P2P mode DIO:

   o  MUST carry one (and only one) P2P Route Discovery Option
      (described in Section 7.1).  The P2P Route Discovery Option allows
      for the specification of one unicast or multicast address for the
      target.

   o  MAY carry one or more RPL Target Options to specify additional
      unicast/multicast addresses for the target.

   o  MAY carry one or more Metric Container Options to specify routing
      metrics and constraints.

   o  MAY carry one or more Data Options (described in Section 7.2)
      containing time-critical application data to be delivered to the
      target(s).

   o  MAY carry one or more Route Information or Prefix Information
      Options (described in [I-D.ietf-roll-rpl]).

   A router MUST discard a received P2P mode DIO if it violates any of
   the rules listed above.

Goyal, et al.           Expires September 3, 2012              [Page 10]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

7.  New RPL Control Message Options

   This document defines two new RPL control message options: the P2P
   Route Discovery Option and the Data Option.

7.1.  P2P Route Discovery Option (P2P-RDO)

           -
       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
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
       |   Type = 10   | Option Length |R|H| N | Compr | L |MaxRank/NH |
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
       |                                                               |
       |                           Target                              |
       |                                                               |
       |                                                               |
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
       |                                                               |
       |                       Address[1..n]                           |
       |                                                               |
       |                                                               |
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

         Figure 1: Format of P2P Route Discovery Option (P2P-RDO)

   The format of a P2P Route Discovery Option (P2P-RDO) is illustrated
   in Figure 1.  A P2P mode DIO and a DRO (defined in Section 8) message
   MUST carry one and at most one P2P-RDO.  A P2P-RDO consists of the
   following fields:

   o  Option Type: 0x0A (to be confirmed by IANA).

   o  Option Length: 8-bit unsigned integer, representing the length in
      octets of the option, not including the Option Type and Option
      Length fields.

   o  Reply (R): The origin sets this flag to 1 to allow the target(s)
      to send DRO messages back to the origin.  If this flag is 0, a
      target MUST NOT generate any DRO message.

   o  Hop-by-hop (H): This flag is valid only if the R flag is set to 1.
      The origin sets this flag to 1 if it desires hop-by-hop routes.
      The origin sets this flag to 0 if it desires source routes.  This
      specification allows for the establishment of one hop-by-hop route
      or up to four source routes per target.  The hop-by-hop route is
      established in the forward direction, i.e. from the origin to the

Goyal, et al.           Expires September 3, 2012              [Page 11]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

      target.  This specification does not allow for the establishment
      of hop-by-hop routes in the backward direction.

   o  Number of Routes (N): This flag is valid only if the R flag is 1
      and H flag is 0, i.e. the targets are allowed to generate DRO
      messages carrying discovered source routes back to the origin.  In
      this case, the value in the N field plus one indicates the number
      of source routes that each target should convey to the origin.
      When hop-by-hop routes are being discovered, the N field MUST be
      set to zero on transmission and ignored on reception.

   o  Compr: 4-bit unsigned integer indicating the number of prefix
      octets that are elided from the Target field and the Address
      vector.  For example, Compr value will be 0 if full IPv6 addresses
      are carried in the Target field and the Address vector.

   o  Life Time (L): A 2-bit field that indicates the suggested life
      time of the temporary DAG, i.e., the suggested duration a router
      joining the temporary DAG SHOULD maintain its membership in the
      DAG.  The mapping between the values in this field and the life
      time of the temporary DAG is as follows:

      *  0x00: 1 second;

      *  0x01: 4 seconds;

      *  0x02: 16 seconds;

      *  0x03: 64 seconds;

      The origin sets this field based on its expectation regarding the
      time required for the DIOs to reach the target(s).

   o  MaxRank/NH:

      *  When a P2P-RDO is included in a P2P mode DIO, this field
         indicates the upper limit on the integer portion of the rank
         (calculated using the DAGRank() macro defined in
         [I-D.ietf-roll-rpl]) that a router may have in the temporary
         DAG being created.  An intermediate router MUST NOT join a
         temporary DAG being created by a P2P mode DIO if the integer
         portion of its rank would be equal to or higher (in numerical
         value) than the MaxRank limit.  A target can join the temporary
         DAG at a rank whose integer portion is equal to the MaxRank.  A
         router MUST discard a received P2P mode DIO if the integer part
         of the advertized rank equals or exceeds the MaxRank limit.  A
         value 0 in this field indicates that the MaxRank is infinity.

Goyal, et al.           Expires September 3, 2012              [Page 12]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

      *  When a P2P-RDO is included in a DRO message, this field
         indicates the index of the next hop address inside the Address
         vector.

   o  Target: An IPv6 address of the target after eliding Compr number
      of prefix octets.  When the P2P-RDO is included in a P2P mode DIO,
      this field may contain a unicast address or a multicast one.  Any
      additional target addresses can be specified by including one or
      more RPL Target Options [I-D.ietf-roll-rpl] in the DIO.  When the
      P2P-RDO is included in a DRO, this field MUST contain a unicast
      IPv6 address of the target generating the DRO.

   o  Address[1..n]: A vector of IPv6 addresses representing a (partial)
      route in the forward direction:

      *  Each element in the Address vector has size (16 - Compr) octets
         and MUST contain a valid IPv6 address with first Compr octets
         elided.

      *  The total number of elements inside the Address vector is given
         by n = (Option Length - 2 - (16 - Compr))/(16 - Compr).

      *  The IPv6 addresses in the Address vector MUST be accessible in
         both forward and backward directions.  Accessibility in the
         backward direction allows a DRO message to use the route
         accumulated in the Address vector to travel from the target to
         the origin.

      *  The Address vector MUST carry the accumulated route in the
         forward direction, i.e., the first element in the Address
         vector must contain the IPv6 address of the router next to the
         origin and so on.

      *  The origin and target addresses MUST NOT be included in the
         Address vector.

      *  A router adding its address to the vector MUST ensure that its
         address does not already exist in the vector.  A router
         specifying a complete route in the Address vector MUST ensure
         that the vector does not contain any address more than once.

      *  The Address vector MUST NOT contain any multicast addresses.

Goyal, et al.           Expires September 3, 2012              [Page 13]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

7.2.  Data 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
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
       |   Type = 11   | Option Length |    Data                       |
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+...                          |

                      Figure 2: Format of Data Option

   The format of a Data Option is illustrated in Figure 2.  A P2P mode
   DIO and a DRO (defined in Section 8) message MAY carry one or more
   Data Options.  A P2P-RDO consists of the following fields:

   o  Option Type: 0x0B (to be confirmed by IANA).

   o  Option Length: 8-bit unsigned integer, representing the length in
      octets of the option, not including the Option Type and Option
      Length fields.

   o  Data: If the Data Option is contained in a DIO, this field
      contains application data to be delivered to the target(s).  If
      the Data Option is contained in a DRO, this field contains
      application data to be delivered to the origin.

8.  The Discovery Reply Object (DRO)

   This section defines two new RPL Control Message types, the Discovery
   Reply Object (DRO), with code 0x04 (to be confirmed by IANA), and the
   Secure DRO, with code 0x84 (to be confirmed by IANA).  A DRO serves
   one of the following functions:

   o  Carry a discovered source route from a target to the origin;

   o  Establish a hop-by-hop route as it travels from a target to the
      origin.

   A DRO message MAY serve the function of letting the routers in the
   LLN know that a P2P-RPL route discovery is complete and no more DIO
   messages need to be generated for the corresponding temporary DAG.  A
   DRO message MAY also carry time-critical application data from the
   target to the origin in one or more Data Options.  A DRO message MUST
   carry one P2P-RDO whose Target field MUST contain a unicast IPv6
   address of the target that generated the DRO.  A DRO message travels
   from the target to the origin via link-local multicast along the
   route specified inside the Address vector in the P2P-RDO.

Goyal, et al.           Expires September 3, 2012              [Page 14]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

       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
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
       | RPLInstanceID |    Version    |S|A|Seq|     Reserved          |
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
       |                                                               |
       |                         DODAGID                               |
       |                                                               |
       |                                                               |
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
       | Option(s)...
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+...

         Figure 3: Format of the base Discovery Reply Object (DRO)

   The format of the base Discovery Reply Object (DRO) is shown in
   Figure 3.  A base DRO consists of the following fields:

   o  RPLInstanceID: The RPLInstanceID of the temporary DAG used for
      route discovery.

   o  Version: The Version of the temporary DAG used for route
      discovery.  Since a temporary DAG always has value zero for the
      Version, this field MUST always be set to zero.

   o  Stop (S): This flag, when set by a target, indicates that the P2P-
      RPL route discovery is over.  All the routers receiving such a
      DRO, including the ones not listed in the route carried inside
      P2P-RDO,

      *  SHOULD NOT process any more DIOs received for this temporary
         DAG;

      *  SHOULD NOT generate any more DIOs for this temporary DAG;

      *  SHOULD cancel any pending DIO transmission for this temporary
         DAG.

      Note that the stop flag serves to stop further DIO transmissions
      for a P2P-RPL route discovery but it does not affect the
      processing of DRO messages at either the origin or the
      intermediate routers.  In other words, a router (the origin or an
      intermediate router) MUST continue to process the DRO messages
      even if an earlier DRO message (with same RPLInstanceID and
      DODAGID fields) had the stop flag set to 1.

Goyal, et al.           Expires September 3, 2012              [Page 15]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   o  Ack Required (A): This flag, when set by the target, indicates
      that the origin MUST unicast a DRO-ACK message (defined in
      Section 10) to the target when it receives the DRO.

   o  Sequence Number (Seq): This 2-bit field indicates the sequence
      number for the DRO.  This field is relevant when the A flag is
      set, i.e., the target requests an acknowledgement from the origin
      for a received DRO.  The origin includes the RPLInstanceID, the
      DODAGID and the Sequence Number of the received DRO inside the
      DRO-ACK message it sends back to the target.

   o  Reserved: These bits are reserved for future use.  These bits MUST
      be set to zero on transmission and MUST be ignored on reception.

   o  DODAGID: The DODAGID of the temporary DAG used for route
      discovery.  The DODAGID also identifies the origin.  The
      RPLInstanceID, the Version and the DODAGID together uniquely
      identify the temporary DAG used for route discovery and can be
      copied from the DIO message advertizing the temporary DAG.

   o  Options: The DRO message:

      *  MUST carry one P2P-RDO that MUST specify a complete route
         between the target and the origin;

      *  MAY carry one or more Metric Container Options that contains
         the aggregated routing metrics values for the route specified
         in P2P-RDO;

      *  MAY carry one or more Data Options to carry any time-critical
         application data to the origin.

8.1.  Secure DRO

   A Secure DRO message follows the format in Figure 7 of
   [I-D.ietf-roll-rpl], where the base format is the base DRO shown in
   Figure 3.

8.2.  Setting a P2P-RDO Carried in a Discovery Reply Object

   A Discovery Reply Object MUST carry one P2P-RDO, which MUST be set as
   defined in Section 7.1.  Specifically, the following fields MUST be
   set as specified next:

   o  Reply (R): This flag MUST be set to zero on transmission and
      ignored on reception.

Goyal, et al.           Expires September 3, 2012              [Page 16]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   o  Hop-by-Hop (H): The H flag in the P2P-RDO included in a DRO
      message MUST have the same value as the H flag in the P2P-RDO
      inside the corresponding DIO message.

   o  Number of Routes (N): This field MUST be set to zero on
      transmission and ignored on reception.

   o  Life Time (L): This field MUST be set to zero on transmission and
      ignored on reception.

   o  MaxRank/NH: This field indicates the index of the next hop address
      in the Address vector.  When a target generates a DRO message, the
      NH field is set to n = (Option Length - 2 - (16 - Compr))/(16 -
      Compr).

   o  Target: This field MUST contain a unicast IPv6 address of the
      target generating the DRO.

   o  Address[1..n]: The Address vector MUST contain a complete route
      between the origin and the target such that the first element in
      the vector contains the IPv6 address of the router next to the
      origin and the last element contains the IPv6 address of the
      router next to the target.

9.  P2P-RPL Route Discovery By Creating a Temporary DAG

   This section details the functioning of P2P-RPL route discovery by
   creating a temporary DAG, using the P2P mode DIO, DRO and DRO-ACK
   messages.

9.1.  Joining a Temporary DAG

   All the routers participating in a P2P-RPL route discovery, including
   the origin and the target(s), MUST join the temporary DAG being
   created for the purpose.  When a router joins a temporary DAG
   advertized by a P2P mode DIO, it SHOULD maintain its membership in
   the temporary DAG for the suggested Life Time duration listed in the
   P2P-RDO.  The only purpose of a temporary DAG's existence is to
   facilitate the P2P-RPL route discovery process.  The temporary DAG
   MUST NOT be used to route packets.  A router SHOULD detach from the
   temporary DAG once the duration of its membership in the DAG has
   exceeded the DAG's suggested life time.  A router SHOULD NOT send or
   receive any more DIOs for the temporary DAG and SHOULD cancel any
   pending DIO transmission when it receives a DRO about the temporary
   DAG with the stop flag set to 1.

Goyal, et al.           Expires September 3, 2012              [Page 17]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

9.2.  Trickle Operation For P2P Mode DIOs

   An RPL router uses a Trickle timer [RFC6206] to control DIO
   transmissions.  The Trickle control of DIO transmissions provides
   quick resolution of any "inconsistency" while avoiding redundant DIO
   transmissions.  The Trickle algorithm also imparts protection against
   loss of DIOs due to inherent lack of reliability in wireless
   communication.  When controlling the transmissions of a P2P mode DIO,
   a Trickle timer SHOULD follow the following rules:

   o  The receipt of a P2P mode DIO, that allows the router to advertise
      a better route (in terms of the routing metrics and the OF in use)
      than before, is considered "inconsistent" and hence resets the
      Trickle timer.  Note that the first receipt of a P2P mode DIO
      advertising a particular temporary DAG is always considered an
      "inconsistent" event.

   o  The receipt of a P2P mode DIO from a parent in the temporary DAG
      is considered neither "consistent" nor "inconsistent" if it does
      not allow the router to advertise a better route than before.
      Thus, the receipt of such DIOs has no impact on the Trickle
      operation.  Note that this document does not impose any
      requirements on how a router might choose its parents in the
      temporary DAG.

   o  The receipt of a P2P mode DIO is considered "consistent" if the
      source of the DIO is not a parent in the temporary DAG and either
      of the following conditions is true:

      *  The DIO advertises a better route than the router but does not
         allow the router to advertise a better route itself; or

      *  The DIO advertises a route as good as the route (to be)
         advertised by the router.

      Note that Trickle algorithm's DIO suppression rules are in effect
      at all times.  Hence, a P2P-RPL router may suppress a DIO
      transmission even if it has not made any DIO transmission yet.

   o  The receipt of a P2P mode DIO, that advertises a worse route than
      what the router advertises (or would advertise when it gets a
      chance to generate its DIO), is considered neither "consistent"
      nor "inconsistent", i.e., the receipt of such a DIO has no impact
      on the Trickle operation.

   o  The Imin parameter SHOULD be set taking in account the
      connectivity within the network.  For highly connected networks, a
      small Imin value (of the order of the typical transmission delay

Goyal, et al.           Expires September 3, 2012              [Page 18]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

      for a DIO) may lead to congestion in the network as a large number
      of routers reset their Trickle timers in response to the first
      receipt of a DIO from the origin.  These routers would generate
      their DIOs within Imin interval and cause additional routers to
      reset their trickle timers and generate more DIOs.  Thus, for
      highly connected networks, the Imin parameter SHOULD be set to a
      value at least one order of magnitude larger than the typical
      transmission delay for a DIO.  For sparsely connected networks,
      the Imin parameter can be set to a value that is a small multiple
      of the typical transmission delay for a DIO.  Note that the Imin
      value has a direct impact on the time required for a P2P-RPL route
      discovery to complete.  In general, the time required for a P2P-
      RPL route discovery would increase approximately linearly with the
      value of the Imin parameter.

   o  The Imax parameter SHOULD be set to a large value (several orders
      of magnitude higher than the Imin value) and is unlikely to be
      critical for P2P-RPL operation.  This is because the first receipt
      of a P2P mode DIO for a particular temporary DAG is considered an
      inconsistent event and would lead to resetting of Trickle timer
      duration to the Imin value.  Given the temporary nature of the
      DAGs used in P2P-RPL, Trickle timer may not get a chance to
      increase much.

   o  The recommended value of redundancy constant "k" is 1.  With this
      value of "k", a DIO transmission will be suppressed if the router
      receives even a single "consistent" DIO during a timer interval.
      This setting for the redundancy constant is designed to reduce the
      number of messages generated during a route discovery process and
      is suitable for environments with low or moderate packet loss
      rates.  In environments with high packet loss rates, a higher
      value for the redundancy constant may be more suitable.

9.3.  Processing a P2P Mode DIO

   The rules for DIO processing and transmission, described in Section 8
   of RPL [I-D.ietf-roll-rpl], apply to P2P mode DIOs as well except as
   modified in this document.

   The following rules for processing a received P2P mode DIO apply to
   both intermediate routers and the target.

   A router SHOULD discard a received P2P mode DIO with no further
   processing if it does not have bidirectional reachability with the
   neighbor that generated the received DIO.  Note that bidirectional
   reachability does not mean that the link must have the same values
   for a routing metric in both directions.  A router SHOULD calculate
   the values of the link-level routing metrics included in the received

Goyal, et al.           Expires September 3, 2012              [Page 19]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   DIO taking in account the metric's value in both forward and backward
   directions.  Bidirectional reachability along a discovered route
   allows the target to use this route to reach the origin.  In
   particular, the DRO messages travel from the target to the origin
   along a discovered route.

   A router MUST discard a received P2P mode DIO with no further
   processing:

   o  If the DIO advertises INFINITE_RANK as defined in
      [I-D.ietf-roll-rpl].

   o  If the integer part of the rank advertised in the DIO equals or
      exceeds the MaxRank limit listed in the P2P Route Discovery
      Option.

   o  If the router cannot evaluate the mandatory route constraints
      listed in the DIO or if the routing metric values do not satisfy
      one or more of the mandatory constraints.

   o  If the router previously received a DRO message with same
      RPLInstanceID and DODAGID as the received DIO and with the stop
      flag set to 1.

   The router MUST check the target addresses listed in the P2P-RDO and
   any RPL Target Options included in the received DIO.  If one of its
   IPv6 addresses is listed as a target address or if it belongs to the
   multicast group specified as one of the target addresses, the router
   considers itself a target and processes the received DIO as specified
   in Section 9.5.  Otherwise, the router considers itself an
   intermediate router and processes the received DIO as specified in
   Section 9.4.

9.4.  Additional Processing of a P2P Mode DIO At An Intermediate Router

   An intermediate router MUST discard a received P2P mode DIO with no
   further processing if the router cannot elide Compr (as specified in
   the P2P-RDO) prefix octets from its IPv6 address.

   On receiving a P2P mode DIO, an intermediate router MUST determine
   whether this DIO advertises a better route than the router itself and
   whether the receipt of the DIO would allow the router to advertise a
   better route than before.  Accordingly, the router SHOULD consider
   this DIO as consistent/inconsistent from Trickle perspective as
   described in Section 9.2.  Note that the route comparison in a P2P-
   RPL route discovery is performed using the parent selection rules of
   the OF in use as specified in Section 14 of RPL [I-D.ietf-roll-rpl].
   If the received DIO would allow the router to advertise a better

Goyal, et al.           Expires September 3, 2012              [Page 20]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   route, the router MUST remember the route advertised (inside the P2P-
   RDO) in the DIO (after adding its own IPv6 address to the route) as
   well as any Data Options for inclusion in its future DIOs.  When an
   intermediate router adds itself to a route, it MUST ensure that the
   IPv6 address added to the route is reachable in both forward and
   backward directions.  To improve the diversity of the routes being
   discovered, an intermediate router SHOULD keep track of multiple
   partial routes to be advertised in the P2P-RDO inside its DIO.  When
   the router generates its DIO, it SHOULD randomly select the partial
   route to be included in the P2P-RDO.

9.5.  Additional Processing of a P2P Mode DIO At The Target

   The target MUST deliver the data contained in any Data Options in the
   received DIO to the application layer.  The target MAY store the
   route contained in the P2P-RDO in the received DIO for use as a
   source route to reach the origin.  If the Reply flag inside the P2P-
   RDO is 0, the target MUST discard the received DIO with no further
   processing.  Otherwise, the target MAY select the route contained in
   the P2P-RDO to send a DRO message back to the origin.  If the H flag
   inside the P2P-RDO is 1, the target needs to select one route and
   send a DRO message along this route back to the origin.  If the H
   flag is 0, the number of routes to be selected (and the number of DRO
   messages to be sent back) is given by one plus the value of the N
   field in the P2P-RDO.  This document does not prescribe a particular
   method for the target to select the routes.  Example methods include
   selecting each route that meets the specified routing constraints
   until the desired number have been selected or selecting the best
   routes discovered over a certain time period.  If multiple routes are
   to be selected, the target SHOULD avoid selecting routes that have
   large segments in common.

   If the target selects the route contained in the P2P-RDO in the
   received DIO, it sends a DRO message back to the origin (identified
   by the DODAGID field in the DIO).  The DRO message MUST include a
   P2P-RDO that contains the selected route inside the Address vector.
   Various fields inside the P2P-RDO MUST be set as specified in
   Section 8.2.  The target MAY set the A flag inside the DRO message if
   it desires the origin to send back a DRO-ACK message on receiving the
   DRO.  In this case, the target waits for DRO_ACK_WAIT_TIME duration
   for the DRO-ACK message to arrive.  Failure to receive the DRO-ACK
   message within this time duration causes the target to retransmit the
   DRO message.  The target MAY retransmit the DRO message in this
   fashion up to MAX_DRO_RETRANSMISSIONS times.  The values of
   DRO_ACK_WAIT_TIME and MAX_DRO_RETRANSMISSIONS are defined in
   Section 12.

   The target MAY set the stop flag inside the DRO message if

Goyal, et al.           Expires September 3, 2012              [Page 21]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   o  this router is the only target specified in the corresponding DIO,
      i.e., the corresponding DIO specified a unicast address of the
      router as the Target inside the P2P-RDO with no additional targets
      specified via RPL Target Options; and

   o  the target has already selected the desired number of routes.

   The target MAY include a Metric Container Option in the DRO message.
   This Metric Container contains the end-to-end routing metric values
   for the route specified in the P2P-RDO.  The target MAY include one
   or more Data Options in the DRO message to carry time-critical
   application data for the origin.  The target MUST transmit the DRO
   message via a link-local multicast.

   A target MUST NOT forward a P2P mode DIO any further.

9.6.  Processing a DRO At An Intermediate Router

   When a router receives a DRO message that does not list its IPv6
   address in the DODAGID field, the router MUST process the received
   message in the following manner:

   o  If the stop flag inside the received DRO is set, the router SHOULD
      NOT send or receive any more DIOs for this temporary DAG and
      SHOULD cancel any pending DIO transmission.

   o  An intermediate router MUST ignore any Metric Container and Data
      Options contained in the DRO message.

   o  If Address[NH] element inside the Route Discovery Option lists the
      router's own IPv6 address, the router is a part of the route
      carried in the P2P-RDO.  In this case, the router MUST do the
      following:

      *  If the H flag inside the P2P-RDO is 0 (i.e., the P2P-RDO is
         carrying a source route), the router MAY make a note of the
         RPLInstanceID and the DODAGID values listed in the DRO.  The
         router may need this information to forward a packet traveling
         along the discovered source route using a Source Routing Header
         (SRH) [I-D.ietf-6man-rpl-routing-header] (see Section 11 for
         details).

      *  If the H flag inside the P2P-RDO is 1, the router MUST store
         the state for the forward hop-by-hop route carried inside the
         P2P-RDO.  This state consists of:

         +  The RPLInstanceID and the DODAGID fields of the DRO.

Goyal, et al.           Expires September 3, 2012              [Page 22]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

         +  The route's destination, the target (identified by Target
            field inside P2P-RDO).

         +  The IPv6 address of the next hop, Address[NH+1] (unless NH
            value equals the number of elements in the Address vector,
            in which case the target itself is the next hop).

      *  If the router already maintains a hop-by-hop state listing the
         target as the destination and carrying same RPLInstanceID and
         DODAGID fields as the received DRO and the next hop information
         in the state does not match the next hop indicated in the
         received DRO, the router MUST drop the DRO message with no
         further processing.

      *  The router MUST decrement the NH field inside the P2P-RDO and
         send the DRO further via link-local multicast.

9.7.  Processing a DRO At The Origin

   When a router receives a DRO message that lists its IPv6 address in
   the DODAGID field, the router recognizes itself as the origin for the
   corresponding P2P-RPL route discovery and processes the message in
   the following manner.

   The origin MUST deliver data in any Data Options in the received DRO
   to the application layer.

   If the stop flag inside the received DRO is set, the origin SHOULD
   NOT generate any more DIOs for this temporary DAG and SHOULD cancel
   any pending DIO transmission.

   If the P2P-RDO inside the DRO identifies the discovered route as a
   source route (H=0), the origin MUST store in its memory the
   discovered route contained in the Address vector.

   If the P2P-RDO inside the DRO identifies the discovered route as a
   hop-by-hop route (H=1), the origin MUST store in its memory the state
   for the discovered route in the manner described in Section 9.6.

   If the received DRO message contains one or more Metric Container
   Options, the origin MAY store the values of the routing metrics
   associated with the discovered route in its memory.  This information
   may be useful in formulating the constraints for any future P2P-RPL
   route discovery to the target.

   If the A flag is set to one in the received DRO message, the origin
   MUST generate a DRO-ACK message as described in Section 10 and
   unicast the message to the target (identified by the Target field

Goyal, et al.           Expires September 3, 2012              [Page 23]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   inside the P2P-RDO).  The origin MAY use the route just discovered to
   send the DRO-ACK message to the target.  Section 11 describes how a
   packet may be forwarded along a source/hop-by-hop route discovered
   using P2P-RPL.

10.  The Discovery Reply Object Acknowledgement (DRO-ACK)

       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
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
       | RPLInstanceID |    Version    |Seq|        Reserved           |
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
       |                                                               |
       |                         DODAGID                               |
       |                                                               |
       |                                                               |
       +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

    Figure 4: Format of the base Discovery Reply Object Acknowledgement
                                 (DRO-ACK)

   A DRO message may fail to reach the origin due to a number of
   reasons.  Unlike the DIO messages that benefit from Trickle-
   controlled retransmissions, the DRO messages are prone to loss due to
   unreliable wireless communication.  Since a DRO message travels via
   link-local multicast, it cannot use link-level acknowledgements to
   improve the reliability of its transmission.  Also, an intermediate
   router may drop the DRO message (e.g., because of its inability to
   store the state for the hop-by-hop route the DRO is establishing).
   To protect against the potential failure of a DRO message to reach
   the origin, the target MAY request the origin to send back a DRO
   Acknowledgement (DRO-ACK) message on receiving a DRO message.
   Failure to receive such an acknowledgement within the
   DRO_ACK_WAIT_TIME interval of sending the DRO message forces the
   target to resend the message.

   This section defines two new RPL Control Message types: DRO
   Acknowledgement (DRO-ACK; with code 0x05; to be confirmed by IANA)
   and Secure DRO-ACK (with code 0x85; to be confirmed by IANA).  A DRO-
   ACK message MUST travel as a unicast message from the origin to the
   target.  The format of a base DRO-ACK message is shown in Figure 4.
   Various fields in a DRO-ACK message MUST have the same values as the
   corresponding fields in the DRO message.  The field marked as
   "Reserved" MUST be set to zero on transmission and MUST be ignored on
   reception.  A Secure DRO-ACK message follows the format in Figure 7
   of [I-D.ietf-roll-rpl], where the base format is same as the base

Goyal, et al.           Expires September 3, 2012              [Page 24]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   DRO-ACK shown in Figure 4.

11.  Packet Forwarding Along a Route Discovered Using P2P-RPL

   An origin MAY use a Source Routing Header (SRH)
   [I-D.ietf-6man-rpl-routing-header] to send a packet along a source
   route discovered using P2P-RPL.  For this purpose, the origin MUST
   set the DODAGID of the temporary DAG used for the source route
   discovery as the source IPv6 address of the packet.  Further, the
   origin MUST specify inside the SRH the RPLInstanceID of the temporary
   DAG used for the source route discovery.  An intermediate router
   verifies being on the source route (it must have noted the
   RPLInstanceID and DODAGID before forwarding the DRO message, carrying
   the source route, towards the origin) before forwarding the packet
   further.

   Travel along a hop-by-hop route, established using P2P-RPL, requires
   specifying the RPLInstanceID and the DODAGID (of the temporary DAG
   used for the route discovery) to identify the route.  This is because
   P2P-RPL route discovery does not use globally unique RPLInstanceID
   values and hence both the RPLInstanceID, a local value assigned by
   the origin, and the DODAGID, an IPv6 address of the origin, are
   required to uniquely identify a P2P-RPL hop-by-hop route to a
   particular destination.

   An origin MAY include an RPL option [I-D.ietf-6man-rpl-option] inside
   the IPv6 hop-by-hop options header of a packet to send it along a
   hop-by-hop route established using P2P-RPL.  For this purpose, the
   origin MUST set the DODAGID of the temporary DAG used for the route
   discovery as the source IPv6 address of the packet.  Further, the
   origin MUST specify inside the RPL option the RPLInstanceID of the
   temporary DAG used for the route discovery and set the O flag inside
   the RPL option to 1.  On receiving this packet, an intermediate
   router checks the O flag and correctly infer the source IPv6 address
   of the packet as the DODAGID of the hop-by-hop route.  The router
   then uses the DODAGID, the RPLInstanceID and the destination address
   to identify the routing state to be used to forward the packet
   further.

12.  Constants

   This document defines the following constants:

   o  DRO_ACK_WAIT_TIME: The time duration a target waits for the DRO-
      ACK before retransmitting a DRO message.  DRO_ACK_WAIT_TIME has a
      value of 1 second.

Goyal, et al.           Expires September 3, 2012              [Page 25]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   o  MAX_DRO_RETRANSMISSIONS: The maximum number of times a DRO message
      may be retransmitted if the target does not receive a DRO-ACK in
      response.  MAX_DRO_RETRANSMISSIONS has a value 2.

13.  Interoperability with Core RPL

   This section describes how RPL routers that implement P2P-RPL
   interact with RPL routers that do not.  In general, P2P-RPL operation
   does not affect core RPL operation and vice versa.  However, core RPL
   does allow a router to join a DAG as a leaf node even if it does not
   understand the Mode of Operation (MOP) used in the DAG.  Thus, an RPL
   router that does not implement P2P-RPL may conceivably join a
   temporary DAG being created for a P2P-RPL route discovery as a leaf
   node and maintain its membership even though the DAG no longer
   exists.  This may impose a drain on the router's memory.  However,
   such RPL-only leaf nodes do not interfere with P2P-RPL route
   discovery since a leaf node may only generate a DIO advertising an
   INFINITE_RANK and all routers implementing P2P-RPL are required to
   discard such DIOs.  Note that core RPL does not require a router to
   join a DAG whose MOP it does not understand.  Moreover, RPL routers
   in a particular deployment may have strict restrictions on the DAGs
   they may join, thereby mitigating the problem.

   The P2P-RPL mechanism described in this document works best when all
   the RPL routers in the LLN implement P2P-RPL.  In general, the
   ability to discover routes as well as the quality of discovered
   routes would deteriorate with the fraction of RPL routers that
   implement P2P-RPL.

14.  Security Considerations

   The security considerations for the operation of P2P-RPL are similar
   to the ones for the operation of RPL (as described in Section 19 of
   [I-D.ietf-roll-rpl]).  Section 10 of RPL specification
   [I-D.ietf-roll-rpl] describes a variety of security mechanisms that
   provide data confidentiality, authentication, replay protection and
   delay protection services.  Each RPL control message has a secure
   version that allows the specification of the level of security and
   the algorithms used to secure the message.  The mechanism defined in
   this document is based on the use of DIOs to form a temporary DAG and
   discover P2P routes.  These DIOs can be used in their secure versions
   if desired.  New RPL control messages defined in this document (DRO
   and DRO-ACK) have secure versions as well.  Thus, a particular P2P-
   RPL deployment can analyze its security requirements and use the
   appropriate set of RPL security mechanisms that meet those
   requirements.

Goyal, et al.           Expires September 3, 2012              [Page 26]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

15.  IANA Considerations

15.1.  Additions to DIO Mode of Operation

   IANA is requested to allocate a new value in the "DIO Mode of
   Operation" registry for the "P2P Route Discovery Mode" described in
   this document.

   +----------+-----------------------------------------+--------------+
   |    MOP   |               Description               |   Reference  |
   |   Value  |                                         |              |
   +----------+-----------------------------------------+--------------+
   |     4    |   Reactive P2P route discovery mode of  |     This     |
   |          |                operation                |   document   |
   +----------+-----------------------------------------+--------------+

                           DIO Mode of Operation

15.2.  Additions to RPL Control Message Options

   IANA is requested to allocate new values in the "RPL Control Message
   Options" registry for the "P2P Route Discovery Option" and the "Data
   Option" described in this document.

              +-------+---------------------+---------------+
              | Value |       Meaning       |   Reference   |
              +-------+---------------------+---------------+
              |   10  | P2P Route Discovery | This document |
              |   11  |         Data        | This document |
              +-------+---------------------+---------------+

                        RPL Control Message Options

15.3.  Additions to RPL Control Codes

   IANA is requested to allocate new code points in the "RPL Control
   Codes" registry for the "Discovery Reply Object" and "Discovery Reply
   Object Acknowledgement" (and their secure versions) described in this
   document.

Goyal, et al.           Expires September 3, 2012              [Page 27]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

   +------+--------------------------------------------+---------------+
   | Code |                 Description                |   Reference   |
   +------+--------------------------------------------+---------------+
   | 0x04 |           Discovery Reply Object           | This document |
   | 0x05 |   Discovery Reply Object Acknowledgement   | This document |
   | 0x84 |        Secure Discovery Reply Object       | This document |
   | 0x85 |        Secure Discovery Reply Object       | This document |
   |      |               Acknowledgement              |               |
   +------+--------------------------------------------+---------------+

                             RPL Control Codes

16.  Acknowledgements

   Authors gratefully acknowledge the contributions of the following
   individuals (in alphabetical order) in the development of this
   document: Dominique Barthel, Jakob Buron, Thomas Clausen, Richard
   Kelsey, Phil Levis, Zach Shelby, Pascal Thubert, Hristo Valev and JP
   Vasseur.

17.  References

17.1.  Normative References

   [I-D.ietf-roll-routing-metrics]
              Barthel, D., Vasseur, J., Pister, K., Kim, M., and N.
              Dejean, "Routing Metrics used for Path Calculation in Low
              Power and Lossy Networks",
              draft-ietf-roll-routing-metrics-19 (work in progress),
              March 2011.

   [I-D.ietf-roll-rpl]
              Brandt, A., Vasseur, J., Hui, J., Pister, K., Thubert, P.,
              Levis, P., Struik, R., Kelsey, R., Clausen, T., and T.
              Winter, "RPL: IPv6 Routing Protocol for Low power and
              Lossy Networks", draft-ietf-roll-rpl-19 (work in
              progress), March 2011.

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

   [RFC6206]  Levis, P., Clausen, T., Hui, J., Gnawali, O., and J. Ko,
              "The Trickle Algorithm", RFC 6206, March 2011.

Goyal, et al.           Expires September 3, 2012              [Page 28]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

17.2.  Informative References

   [I-D.ietf-6man-rpl-option]
              Hui, J. and J. Vasseur, "RPL Option for Carrying RPL
              Information in Data-Plane Datagrams",
              draft-ietf-6man-rpl-option-06 (work in progress),
              December 2011.

   [I-D.ietf-6man-rpl-routing-header]
              Culler, D., Hui, J., Vasseur, J., and V. Manral, "An IPv6
              Routing Header for Source Routes with RPL",
              draft-ietf-6man-rpl-routing-header-07 (work in progress),
              December 2011.

   [I-D.ietf-roll-of0]
              Thubert, P., "RPL Objective Function Zero",
              draft-ietf-roll-of0-20 (work in progress), September 2011.

   [I-D.ietf-roll-p2p-measurement]
              Goyal, M., Baccelli, E., Brandt, A., and J. Martocci, "A
              Mechanism to Measure the Quality of a Point-to-point Route
              in a Low Power and Lossy Network",
              draft-ietf-roll-p2p-measurement-02 (work in progress),
              October 2011.

   [I-D.ietf-roll-terminology]
              Vasseur, J., "Terminology in Low power And Lossy
              Networks", draft-ietf-roll-terminology-06 (work in
              progress), September 2011.

   [RFC5826]  Brandt, A., Buron, J., and G. Porcu, "Home Automation
              Routing Requirements in Low-Power and Lossy Networks",
              RFC 5826, April 2010.

   [RFC5867]  Martocci, J., De Mil, P., Riou, N., and W. Vermeylen,
              "Building Automation Routing Requirements in Low-Power and
              Lossy Networks", RFC 5867, June 2010.

Goyal, et al.           Expires September 3, 2012              [Page 29]
Internet-Draft         draft-ietf-roll-p2p-rpl-08             March 2012

Authors' Addresses

   Mukul Goyal (editor)
   University of Wisconsin Milwaukee
   3200 N Cramer St
   Milwaukee, WI  53201
   USA

   Phone: +1 414 2295001
   Email: mukul@uwm.edu

   Emmanuel Baccelli
   INRIA

   Phone: +33-169-335-511
   Email: Emmanuel.Baccelli@inria.fr
   URI:   http://www.emmanuelbaccelli.org/

   Matthias Philipp
   INRIA

   Phone: +33-169-335-511
   Email: Matthias.Philipp@inria.fr

   Anders Brandt
   Sigma Designs
   Emdrupvej 26A, 1.
   Copenhagen, Dk-2100
   Denmark

   Phone: +45-29609501
   Email: abr@sdesigns.dk

   Jerald Martocci
   Johnson Controls
   507 E Michigan St
   Milwaukee, WI  53202
   USA

   Phone: +1 414-524-4010
   Email: jerald.p.martocci@jci.com

Goyal, et al.           Expires September 3, 2012              [Page 30]