RPL: IPv6 Routing Protocol for Low power and Lossy Networks
draft-ietf-roll-rpl-04

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), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts.

Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as “work in progress.”

The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt.

The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html.

This Internet-Draft will expire on April 29, 2010.

Copyright Notice

Copyright (c) 2009 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 in effect on the date of publication of this document (http://trustee.ietf.org/license-info). Please review these documents carefully, as they describe your rights and restrictions with respect to this document.

Abstract

Low power and Lossy Networks (LLNs) are a class of network in which both the routers and their interconnect are constrained: LLN routers typically operate with constraints on (any subset of) processing power, memory and energy (battery), and their interconnects are characterized by (any subset of) high loss rates, low data rates and instability. LLNs are comprised of anything from a few dozen and up to thousands of LLN routers, and support point-to- point traffic (between devices inside the LLN), point-to-multipoint traffic (from a central control point to a subset of devices inside the LLN) and multipoint-to- point traffic (from devices inside the LLN towards a central control point). This document specifies the IPv6 Routing Protocol for LLNs (RPL), which provides a mechanism whereby multipoint-to-point traffic from devices inside the LLN towards a central control point, as well as point-to-multipoint traffic from the central control point to the devices inside the LLN, is supported. Support for point-to-point traffic is also available.

Table of Contents

1.  Introduction
    1.1.  Design Principles
    1.2.  Expectations of Link Layer Type
2.  Terminology
3.  Protocol Model
    3.1.  Overview
        3.1.1.  Topology Instance and Objectives
        3.1.2.  Multipoint-to-Point Traffic Flows and DAGs
        3.1.3.  Point-to-Multipoint Traffic Flows
        3.1.4.  Point-to-Point Traffic Flows
    3.2.  Protocol Operation
        3.2.1.  DAG Construction
        3.2.2.  Destination Advertisement
    3.3.  Loop Avoidance and Stability
        3.3.1.  Greediness and Rank-based Instabilities
        3.3.2.  DAG Loops
        3.3.3.  DAO Loops
        3.3.4.  Sibling Loops
4.  Routing Metrics and Constraints Used By RPL
5.  RPL Protocol Specification
    5.1.  RPL Messages
        5.1.1.  ICMPv6 RPL Control Message
        5.1.2.  DAG Information Solicitation (DIS)
        5.1.3.  DAG Information Object (DIO)
        5.1.4.  Destination Advertisement Object (DAO)
    5.2.  Conceptual Data Structures
        5.2.1.  Candidate Neighbors Data Structure
        5.2.2.  Directed Acyclic Graphs (DAGs) Data Structure
    5.3.  DAG Rank
    5.4.  DAG Discovery and Maintenance
        5.4.1.  DAG Discovery Rules
        5.4.2.  Reception and Processing of DIO messages
        5.4.3.  DIO Transmission
        5.4.4.  Trickle Timer for DIO Transmission
    5.5.  DAG Sequence Number Increment
    5.6.  DAG Selection
    5.7.  Administrative rank
    5.8.  Collision
    5.9.  Guidelines for Objective Functions
        5.9.1.  Objective Function
        5.9.2.  Objective Function 0 (OF0)
    5.10.  Establishing Routing State Outward Along the DAG
        5.10.1.  Destination Advertisement Operation
    5.11.  Loop Detection
        5.11.1.  Host Basic Operation
        5.11.2.  Instance Forwarding
        5.11.3.  DAG Inconsistency Loop Detection
        5.11.4.  Sibling Loop Avoidance
        5.11.5.  DAO Inconsistency Loop Detection and Recovery
    5.12.  Multicast Operation
    5.13.  Maintenance of Routing Adjacency
    5.14.  Packet Forwarding
6.  RPL Constants and Variables
7.  Manageability Considerations
    7.1.  Control of Function and Policy
        7.1.1.  Initialization Mode
        7.1.2.  DIO Base option
        7.1.3.  Trickle Timers
        7.1.4.  DAG Sequence Number Increment
        7.1.5.  Destination Advertisement Timers
        7.1.6.  Policy Control
        7.1.7.  Data Structures
    7.2.  Information and Data Models
    7.3.  Liveness Detection and Monitoring
        7.3.1.  Candidate Neighbor Data Structure
        7.3.2.  Directed Acyclic Graph (DAG) Table
        7.3.3.  Routing Table
        7.3.4.  Other RPL Monitoring Parameters
        7.3.5.  RPL Trickle Timers
    7.4.  Verifying Correct Operation
    7.5.  Requirements on Other Protocols and Functional Components
    7.6.  Impact on Network Operation
8.  Security Considerations
9.  IANA Considerations
    9.1.  RPL Control Message
    9.2.  New Registry for RPL Control Codes
    9.3.  New Registry for the Control Field of the DIO Base Option
    9.4.  DAG Information Object (DIO) Suboption
    9.5.  Objective Code Point for the Default Objective Function OF0
10.  Acknowledgements
11.  Contributors
12.  References
    12.1.  Normative References
    12.2.  Informative References
Appendix A.  Requirements
    A.1.  Protocol Properties Overview
        A.1.1.  IPv6 Architecture
        A.1.2.  Typical LLN Traffic Patterns
        A.1.3.  Constraint Based Routing
    A.2.  Deferred Requirements
Appendix B.  Examples
    B.1.  Destination Advertisement
    B.2.  Example: DAG Parent Selection
    B.3.  Example: DAG Maintenance
    B.4.  Example: Greedy Parent Selection and Instability
Appendix C.  Outstanding Issues
    C.1.  Additional Support for P2P Routing
    C.2.  Loop Detection
    C.3.  Destination Advertisement / DAO Fan-out
    C.4.  Source Routing
    C.5.  Address / Header Compression
§  Authors' Addresses

1.  Introduction

Low power and Lossy Networks (LLNs) are made largely of constrained nodes (with limited processing power, memory, and sometimes energy when they are battery operated). These routers are interconnected by lossy links, typically time supporting only low data rates, that are usually unstable with relatively low packet delivery rates. Another characteristic of such networks is that the traffic patterns are not simply unicast, but in many cases point-to-multipoint or multipoint-to-point. Furthermore such networks may potentially comprise up to thousands of nodes. These characteristics offer unique challenges to a routing solution: the IETF ROLL Working Group has defined application-specific routing requirements for a Low power and Lossy Network (LLN) routing protocol, specified in [I‑D.ietf‑roll‑building‑routing‑reqs] (Martocci, J., Riou, N., Mil, P., and W. Vermeylen, “Building Automation Routing Requirements in Low Power and Lossy Networks,” September 2009.), [I‑D.ietf‑roll‑home‑routing‑reqs] (Brandt, A., Buron, J., and G. Porcu, “Home Automation Routing Requirements in Low Power and Lossy Networks,” September 2009.), [RFC5673] (Pister, K., Thubert, P., Dwars, S., and T. Phinney, “Industrial Routing Requirements in Low-Power and Lossy Networks,” October 2009.), and [RFC5548] (Dohler, M., Watteyne, T., Winter, T., and D. Barthel, “Routing Requirements for Urban Low-Power and Lossy Networks,” May 2009.). This document specifies the IPv6 Routing Protocol for Low power and Lossy Networks (RPL).

1.1.  Design Principles

RPL was designed with the objective to meet the requirements spelled out in [I‑D.ietf‑roll‑building‑routing‑reqs] (Martocci, J., Riou, N., Mil, P., and W. Vermeylen, “Building Automation Routing Requirements in Low Power and Lossy Networks,” September 2009.), [I‑D.ietf‑roll‑home‑routing‑reqs] (Brandt, A., Buron, J., and G. Porcu, “Home Automation Routing Requirements in Low Power and Lossy Networks,” September 2009.), [RFC5673] (Pister, K., Thubert, P., Dwars, S., and T. Phinney, “Industrial Routing Requirements in Low-Power and Lossy Networks,” October 2009.), and [RFC5548] (Dohler, M., Watteyne, T., Winter, T., and D. Barthel, “Routing Requirements for Urban Low-Power and Lossy Networks,” May 2009.). Because those requirements are heterogeneous and sometimes incompatible in nature, the approach is first taken to design a protocol capable of supporting a core set of functionalities corresponding to the intersection of the requirements. (Note: it is intended that as this design evolves optional features may be added to address some application specific requirements). This is a key protocol design decision providing a granular approach in order to restrict the core of the protocol to a minimal set of functionalities, and to allow each implementation of the protocol to be optimized in terms of, e.g., minimizing required code space and use of limited computation resources.

Multiple instances of the protocol can be operated at the same time in order to serve different and potentially antagonistic constraints. Instances run independently of one another with no required interaction. A node might participate to multiple instances and route independently along the associated topologies. This specification defines only the protocol operation for the node within one instance. Consideration is given to default behavior that enables future extensions for the multiple instances and related policies.

It must be noted that RPL is not restricted to the aforementioned applications and is expected to be used in other environments. All "MUST" application requirements that cannot be satisfied by RPL will be specifically listed in the Appendix A, accompanied by a justification.

The core set of functionalities is to be capable of operating in the most severely constrained environments, with minimal requirements for memory, energy, processing, communication, and other consumption of limited resources from nodes. Trade-offs inherent in the provisioning of protocol features will be exposed to the implementer in the form of configurable parameters, such that the implementer can further tweak and optimize the operation of RPL as appropriate to a specific application and implementation. Finally, RPL is designed to consult implementation specific policies to determine, for example, the evaluation of routing metrics.

A set of companion documents to this specification will provide further guidance in the form of applicability statements specifying a set of operating points appropriate to the Building Automation, Home Automation, Industrial, and Urban application scenarios.

1.2.  Expectations of Link Layer Type

This specification does not rely on any particular features of a specific link layer technologies. It is anticipated that an implementer should be able to operate RPL over a variety of different link layers, including but not limited to low power wireless or PLC (Power Line Communication) technologies.

Implementers may find RFC 3819 (Karn, P., Bormann, C., Fairhurst, G., Grossman, D., Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. Wood, “Advice for Internet Subnetwork Designers,” July 2004.) [RFC3819] a useful reference when designing a link layer interface between RPL and a particular link layer technology.

2.  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 RFC 2119 (Bradner, S., “Key words for use in RFCs to Indicate Requirement Levels,” March 1997.) [RFC2119].

This document requires readers to be familiar with the terminology described in `Terminology in Low power And Lossy Networks' [I‑D.ietf‑roll‑terminology] (Vasseur, J., “Terminology in Low power And Lossy Networks,” March 2010.).

DAG:

Directed Acyclic Graph. A directed graph having the property that all edges are oriented in such a way that no cycles exist. In the RPL context, all edges are contained in paths oriented toward and terminating at one or more root nodes (a DAG root, or sink- typically a Low power and Lossy Network Border Router (LBR)). For the purpose of this document, the term DAG is often used to refer to a DAG Iteration as defined below.

DAG Instance:

A DAG Instance is a set of possibly multiple Destination Oriented DAGs. A network may have more than one DAG Instance, and a RPL router can participate to multiple DAG instances. Each DAG Instance operates independently of other DAG Instances. This document describes operation within a single DAG instance.

InstanceID:

Unique identifier of a DAG Instance.

Destination Oriented DAG:

A DAG rooted at a single destination, which is a node with no outgoing edges. The tuple (InstanceID, DAGID) uniquely identifies a Destination Oriented DAG. In the RPL context, a router can can belong to at most one Destination Oriented DAG per DAG Instance.

DAGID:

The identifier of a DAG root. The DAGID must be unique within the scope of a DAG Instance in the LLN.

DAG Iteration:

The DAG that results from the iterative process that reshapes the Destination Oriented DAG upon a stimulation by the root.

DAGSequenceNumber:

A sequential counter that is incremented by the root to form a new Iteration of a DAG. A DAG Iteration is identified uniquely by the (InstanceID, DAGID, DAGSequenceNumber) tuple.

DAG parent:

A parent of a node within a DAG is one of the immediate successors of the node on a path towards the DAG root.

DAG sibling:

A sibling of a node within a DAG is defined in this specification to be any neighboring node which is located at the same rank within a DAG. Note that siblings defined in this manner do not necessarily share a common parent.

DAG root:

A DAG root is a node within the DAG that has no outgoing edges. Because the graph is acyclic, by definition all DAGs must have at least one DAG root and all paths terminate at a DAG root.

Sub-DAG

The sub-DAG of a node is the set of other nodes in the DAG that might use a path towards the DAG root that contains the node. Nodes in the sub-DAG of a node have a greater rank (although not all nodes of greater rank are in the sub-DAG).

Grounded:

A DAG is grounded if it contains a DAG root offering connectivity to an external routed infrastructure such as the public Internet or a private core (non-LLN) IP network.

Floating:

A DAG is floating if is not grounded. A floating DAG is not expected to reach any additional external routed infrastructure such as the public Internet or a private core (non-LLN) IP network.

Inward:

Inward refers to the direction from leaf nodes towards DAG roots, following the orientation of the edges within the DAG.

Outward:

Outward refers to the direction from DAG roots towards leaf nodes, going against the orientation of the edges within the DAG.

OCP:

Objective Code Point. The Objective Code Point is used to indicate which Objective Function is in use in a DAG. The Objective Code Point is further described in [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.).

OF:

Objective Function. The Objective Function (OF) defines which routing metrics, optimization objectives, and related functions are in use in a DAG. The Objective Function is further described in [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.).

Note that in this document, the terms `node' and `LLN router' are used interchangeably.

3.  Protocol Model

The aim of this section is to describe RPL in the spirit of [RFC4101] (Rescorla, E. and IAB, “Writing Protocol Models,” June 2005.). Protocol details can be found in further sections.

3.1.  Overview

3.1.1.  Topology Instance and Objectives

A topology instance of RPL exists over the scope of an LLN in support of a particular application, or service, and is optimized according to a certain objective, as determined by an Objective Function (OF), and may be characterized by certain destination prefixes as well. A topology instance, or DAG Instance, may be administratively associated with an InstanceID.

A single topology instance may comprise:

• a single Destination Oriented DAG with a single DAG root
  • For example, a DAG optimized to minimize latency rooted at a single centralized lighting controller in a home automation application.
• multiple uncoordinated Destination Oriented DAGs with independent DAG roots (differing DAGIDs)
  • For example, multiple data collection points in an urban data collection application that do not have an always-on backbone suitable to coordinate to form a single DAG, and further use the formation of multiple DAGs as a means to dynamically and autonomously partition the network.
• a single Destination Oriented DAG with multiple DAG roots coordinating over some backbone
  • For example, multiple border routers operating with a reliable backbone, e.g. in support of a 6LowPAN application, that are capable to act as logically equivalent sinks to the same DAG.
• a combination of one of the above as suited to some application scenario

The exact deployment scenario is determined as appropriate to the application and capabilities of the LLN nodes. What is suitable for one deployment may not be possible or necessary for another.

Traffic is bound to a specific DAG Instance by a marking in the flow label of the IPv6 header. Traffic originating in support of a particular application may be tagged to follow an appropriate instance, for example to follow paths optimized for low latency or low energy. The provisioning or automated discovery of a mapping between an InstanceID and a type or service of application traffic is beyond the scope of this specification.

Conceptually a node running RPL may capable to maintain a membership in multiple DAG Instances in support of different application services and/or optimization objectives. For example, one instance may optimize for minimizing latency and a separate orthogonal instance may optimize for minimizing energy. This scenario does introduce some additional considerations, for example loop avoidance and default routing behavior. These considerations are beyond the scope of this specification and are intended to be elaborated on in a future revision of this or a companion specification. As such, this specification will deal exclusively with the scenario where a node implements RPL in support of a single DAG Instance.

3.1.2.  Multipoint-to-Point Traffic Flows and DAGs

Many of the dominant traffic flows in support of the LLN application scenarios are MP2P flows ([I‑D.ietf‑roll‑building‑routing‑reqs] (Martocci, J., Riou, N., Mil, P., and W. Vermeylen, “Building Automation Routing Requirements in Low Power and Lossy Networks,” September 2009.), [I‑D.ietf‑roll‑home‑routing‑reqs] (Brandt, A., Buron, J., and G. Porcu, “Home Automation Routing Requirements in Low Power and Lossy Networks,” September 2009.), [RFC5673] (Pister, K., Thubert, P., Dwars, S., and T. Phinney, “Industrial Routing Requirements in Low-Power and Lossy Networks,” October 2009.), and [RFC5548] (Dohler, M., Watteyne, T., Winter, T., and D. Barthel, “Routing Requirements for Urban Low-Power and Lossy Networks,” May 2009.)). These flows are rooted at designated nodes that have some application significance, such as providing connectivity to an external routed infrastructure. The term "external" is used to refer to the public Internet or a core private (non-LLN) IP network.

LLN nodes running RPL will construct Directed Acyclic Graphs (DAGs) rooted at DAG roots, which may be naturally designated according to their application significance. This structure provides the routing solution for the dominant MP2P traffic flows. The DAG structure further provides each node potentially multiple successors for MP2P flows, which may be used for, e.g., local route repair or load balancing.

Nodes running RPL are able to further restrict the scope of the routing problem by using the DAG as a reference topology. By referencing a rank property that is related to the positions in the DAG, nodes are able to determine their positions in a DAG relative to each other. This information is used by RPL in part to construct rules for movement relative to the DAG that endeavor to avoid loops. It is important to note that the rank property is derived from metrics, and not directly from the position in the DAG (Section 5.3 (DAG Rank)).

3.1.3.  Point-to-Multipoint Traffic Flows

As DAGs are organized, RPL will use a destination advertisement mechanism to build up routing tables in support of outward P2MP traffic flows. This mechanism, using the DAG as a reference, distributes routing information across intermediate nodes (between the DAG leaves and the root), guided along the DAG, such that the routes toward destination prefixes in the outward direction may be set up. As the DAG undergoes modification during DAG maintenance, the destination advertisement mechanism can be triggered to update the outward routing state.

3.1.4.  Point-to-Point Traffic Flows

A baseline support for P2P traffic in RPL is provided by the DAG, as P2P traffic may flow inward along the DAG until a common parent is reached that has stored an entry for the destination in its routing table and is capable of directing the traffic outward along the correct outward path. RPL also provides support for the trivial case where a P2P destination may be a `one-hop' neighbor. In the present document RPL does not specify nor preclude any additional mechanisms that may be capable to compute and install more optimal routes into LLN nodes in support of arbitrary P2P traffic according to some routing metric.

3.2.  Protocol Operation

3.2.1.  DAG Construction

3.2.1.1.  DAG Information Object (DIO)

A DAG Information Object is defined and used by RPL in order to build and maintain a DAG. This document defines an ICMPv6 Message Type RPL Control Message, which is capable to carry the DIO. The DIO message conveys information about the DAG, including:

• A DAGID used to identify the DAG as sourced from the DAG root. The DAGID must be unique to a single DAG in the scope of the LLN.
• Objective Function identified by an Objective Code Point (OCP) as described below.
• Rank information used by nodes to determine their positions in the DAG relative to each other.
• Sequence number originated from the DAG root, used to aid in identification of dependent sub-DAGs and coordinate topology changes in a manner so as to avoid loops.
• Indications and configuration for the DAG, e.g. grounded or floating, administrative preference, ...
• A set of path metrics and constraints, as further described in [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.).
• List of additional destination prefixes reachable inwards along the DAG.

The DIO messages are issued whenever a change is detected to the DAG such that a node is able to determine that a region of the DAG has become inconsistent. As the DAG stabilizes the period at which RA messages occur is configured to taper off, reducing the steady-state overhead of DAG maintenance. The periodic issue of DIO messages, along with the triggered DIO messages in response to inconsistency, is one feature that enables RPL to operate in the presence of unreliable links.

3.2.1.2.  Grounded and Floating DAGs

Certain LLN nodes may offer connectivity to an external routed infrastructure in support of an application scenario. These nodes are designated `grounded', and may serve as the DAG roots of a grounded DAG. DAGs that do not have a grounded DAG root are floating DAGs. In either case routes may be provisioned toward the DAG root, although in the floating case there is no expectation to reach an external infrastructure. Some applications will include permanent floating DAGs.

3.2.1.3.  Administrative Preference

An administrative preference may be associated with each DAG root, and thereby each DAG, in order that some DAGs in the LLN may be more preferred over other DAGs. For example, a DAG root that is sinking traffic in support of a data collection application may be configured by the application to be very preferred. A transient DAG, e.g. a DAG that is only existing until a permanent DAG is found, may be configured to be less preferred. The administrative preference offers a way to engineer the formation of the DAG in support of the application.

3.2.1.4.  Objective Function (OF)

The Objective Function (OF) conveys and controls the optimization objectives in use within the DAG. The Objective Function is indicated by an Objective Code Point (OCP), and is further specified in [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.). Each instance of an allocated OF indicates:

• The set of metrics used within the DAG
• The method used for least cost path determination.
• The method used to compute DAG Rank
• The methods used to prepare metrics for propagation within a DIO message

By using defined OCPs that are understood by all nodes in a particular implementation, and by conveying them in the DIO message, RPL nodes may work to build optimized LLN using a variety of application and implementation specific metrics and goals.

A default OF, OF0 (designated by OCP value of 0x0000), is specified with a well-defined default behavior. OF0 may be used to define RPL behaviors in the case where a node encounters a DIO message containing a code point that it does not support, if allowed by policy.

3.2.1.5.  Distributed Algorithm Operation

A high level overview of the distributed algorithm which constructs the DAG is as follows:

• Some nodes may be initially provisioned to act as DAG roots, either permanent or transient, with associated preferences.
• Nodes will maintain a data structure containing their candidate (viable) neighbors, as determined by the implementation. This data structure will also track DAG information as learned from and associated with each neighbor.
• Nodes that are members of a DAG, including DAG roots, will multicast DIO messages as needed (when inconsistency is detected), to their link-local neighbors. Nodes will also respond to DIS messages.
• Nodes that receive DIO messages may either discard the DIO based on several criteria, including rank-based loop avoidance rules, or process the DIO to maintain a position in an existing DAG or improve a position as according to an Objective Function (OF) and current path cost.
• Nodes manage a set of DAG Parents according to the rules specified by RPL. This set is also augmented to include DAG siblings.
• DIO messages may be emitted more or less frequently as a function of DAG consistency.
• As less preferred DAGs encounter more preferred DAGs that offer equivalent or better optimization objectives for the same InstanceID, the nodes in the less preferred DAGs may leave to join the more preferred DAGs, finally leaving only the more preferred DAGs. This is an illustration of the mechanism by which an application may engineer DAG construction.
• The nodes provision routing table entries for the destinations specified by the DIO towards their DAG Parents. Nodes may provision a DAG Parent as a default gateway.

3.2.2.  Destination Advertisement

As RPL constructs DAGs, nodes may provision routes toward destinations advertised through DIO messages through their selected parents, and are thus able to send traffic inward along the DAG by forwarding to their selected parents. However, this mechanism alone is not sufficient to support P2MP traffic flowing outward along the DAG from the DAG root toward nodes. A destination advertisement mechanism is employed by RPL to build up routing state in support of these outward flows. The destination advertisement mechanism may not be supported in all implementations, as appropriate to the application requirements. A DAG root that supports using the destination advertisement mechanism to build up routing state will indicate such in the DIO message. A DAG root that supports using the destination advertisement mechanism must be capable of allocating enough state to store the routing state received from the LLN.

3.2.2.1.  Destination Advertisement Object (DAO)

A Destination Advertisement Object is defined and used by RPL in order to convey the destination information inward along the DAG toward the DAG root. This document defines an ICMPv6 Message Type RPL Control Message, which is capable to carry the DAO. The information conveyed in the DAO message includes the following:

• A lifetime and sequence counter to determine the freshness of the destination advertisement.
• Depth information used by nodes to determine how far away the destination (the source of the destination advertisement) is
• Prefix information to identify the destination, which may be a prefix, an individual host, or multicast listeners
• Reverse Route information to record the nodes visited (along the outward path) when the intermediate nodes along the path cannot support storing state for Hop-By-Hop routing.

3.2.2.2.  Destination Advertisement Operation

As the DAG is constructed and maintained, nodes are capable to emit DAO messages to a subset of their DAG parents.

3.2.2.2.1.  `One-Hop' Neighbors

As a special case, a node may periodically emit a link-local multicast IPv6 DAO message advertising its locally available destination prefixes. This mechanism allows for the one-hop neighbors of a node to learn explicitly of the prefixes on the node, and in some application specific scenarios this is desirable in support of provisioning a trivial `one-hop' route. In this case, nodes that receive the multicast destination advertisement may use it to provision the one-hop route only, and not engage in any additional processing (so as not to engage the mechanisms used by a DAG parent).

3.2.2.2.2.  Operation in Support of Stateful Nodes

When a (unicast) DAO message reaches a node capable of storing routing state, the node extracts information from the DAO message and updates its local database with a record of the DAO message and the neighbor that it was received from. When the node later propagates DAO messages, it selects the best (least depth) information for each destination and conveys this information again in the form of DAO messages to a subset of its own DAG parents. At this time the node may perform route aggregation if it is able, thus reducing the overall number of DAO messages.

3.2.2.2.3.  Operation in Support of Stateless Nodes

When a (unicast) DAO message reaches a node incapable of storing additional state, the node must append the next-hop address (from which neighbor the DAO message was received) to a Reverse Route Stack carried within the DAO message. The node then passes the DAO message on to one or more of its DAG parents without storing any additional state.

When a node that is capable of storing routing state encounters a (unicast) DAO message with a Reverse Route Stack that has been populated, the node knows that the DAO message has traversed a region of nodes that did not record any routing state. The node is able to detach and store the Reverse Route State and associate it with the destination described by the DAO message. Subsequently the node may use this information to construct a source route in order to bridge the region of nodes that are unable to support Hop-By-Hop routing to reach the destination.

3.2.2.2.4.  Additional Considerations

Further aggregations of DAO messages prefix reachability information by destinations are possible in order to support additional scalability.

A special case of an DAO message, termed a `no-DAO', may be used to tear down the routing state that has been established by the destination advertisement mechanism in case of, e.g., unreachability or other events that affect the outward routing state.

A further example of the operation of the destination advertisement mechanism is available in Appendix B.1 (Destination Advertisement)

3.3.  Loop Avoidance and Stability

The goal of a guaranteed consistent and loop free global routing solution for an LLN may not be practically achieved given the real behavior and volatility of the underlying metrics. The trade offs to achieve a stable approximation of global convergence may be too restrictive with respect to the need of the LLN to react quickly in response to the lossy environment. Globally the LLN may be able to achieve a weak convergence, in particular as link changes are able to be handled locally and result in minimal changes to global topology.

RPL does not aim to guarantee loop free path selection, or strong global convergence. In order to reduce control overhead, in particular the expense of mechanisms such as count-to-infinity, RPL does try to avoid the creation of loops when undergoing topology changes.

RPL includes rank-based mechanisms for detecting loops to ensure that packets make forward progress within the DAG and trigger DAG repair if necessary.

3.3.1.  Greediness and Rank-based Instabilities

Once a node has joined a DAG, RPL disallows certain behaviors, including greediness, in order to prevent resulting instabilities in the DAG.

If a node is allowed to be greedy and attempts to move deeper in the DAG, beyond its most preferred parent, in order to increase the size of the DAG parent set, then an instability can result. This is illustrated in Figure 14 (Greedy DAG Parent Selection).

Suppose a node is willing to receive and process a DIO messages from a node in its own sub-DAG, and in general a node deeper than it. In such cases a chance exists to create a feedback loop, wherein two or more nodes continue to try and move in the DAG in order to optimize against each other. In some cases this will result in an instability. It is for this reason that RPL mandates that a node never receive and process DIO messages from deeper nodes. This rule creates an `event horizon', whereby a node cannot be influenced into an instability by the action of nodes that may be in its own sub-DAG.

A further example of the consequences of greedy operation, and instability related to processing DIO messages from nodes of greater rank, may be found in Appendix B.4 (Example: Greedy Parent Selection and Instability)

3.3.2.  DAG Loops

A DAG loop may occur when a node detaches from the DAG and reattaches to a device in its prior sub-DAG. This may happen in particular when DIO messages are missed. Strict use of the DAG sequence number can eliminate this type of loop.

3.3.3.  DAO Loops

A DAO loop may occur when the parent has a route installed upon receiving and processing a DAO message from a child, but the child has subsequently cleaned up the state. This loop happens when a no-DAO was missed till a heartbeat cleans up all states. RPL includes loop detection mechanisms that may mitigate the impact of DAO loops and trigger their repair.

In the case where stateless DAO operation is used, i.e. source routing specifies the outwards routes, then DAO Loops should not occur on the stateless portions of the path.

3.3.4.  Sibling Loops

Sibling loops could occur if a group of siblings kept choosing amongst themselves as successors such that a packet does not make forward progress. This specification limits the number of times that sibling forwarding may be used at a given rank to prevent sibling loops.

4.  Routing Metrics and Constraints Used By RPL

Routing metrics are used by routing protocols to compute the shortest paths. Interior Gateway Protocols (IGPs) such as IS-IS ([RFC5120] (Przygienda, T., Shen, N., and N. Sheth, “M-ISIS: Multi Topology (MT) Routing in Intermediate System to Intermediate Systems (IS-ISs),” February 2008.)) and OSPF ([RFC4915] (Psenak, P., Mirtorabi, S., Roy, A., Nguyen, L., and P. Pillay-Esnault, “Multi-Topology (MT) Routing in OSPF,” June 2007.)) use static link metrics. Such link metrics can simply reflect the bandwidth or can also be computed according to a polynomial function of several metrics defining different link characteristics; in all cases they are static metrics. Some routing protocols support more than one metric: in the vast majority of the cases, one metric is used per (sub)topology. Less often, a second metric may be used as a tie-breaker in the presence of Equal Cost Multiple Paths (ECMP). The optimization of multiple metrics is known as an NP complete problem and is sometimes supported by some centralized path computation engine.

In contrast, LLNs do require the support of both static and dynamic metrics. Furthermore, both link and node metrics are required. In the case of RPL, it is virtually impossible to define one metric, or even a composite, that will satisfy all use cases.

In addition, RPL supports constrained-based routing where constraints may be applied to link and nodes. If a link or a node does not satisfy a required constraint, it is `pruned' from the candidate list thus leading to a constrained shortest path.

The set of supported link/node constraints and metrics is specified in [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.).

The role of the Objective Function is to advertise routing metrics and constraints in addition to the objectives used to compute the (constrained) shortest path.

Example 1:

Shortest path: path offering the shortest end-to-end delay

Example 2:

Constrained shortest path: the path that does traverse any battery-operated node and that optimizes the path reliability

5.  RPL Protocol Specification

5.1.  RPL Messages

5.1.1.  ICMPv6 RPL Control Message

This document defines the RPL Control Message, a new ICMPv6 message. The RPL Control Message has the following general format, in accordance with [RFC4443] (Conta, A., Deering, S., and M. Gupta, “Internet Control Message Protocol (ICMPv6) for the Internet Protocol Version 6 (IPv6) Specification,” March 2006.):

     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      |     Code      |          Checksum             |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |                                                               |
    +                         Message Body                          +
    |                                                               |

The RPL Control message is an ICMPv6 information message with a requested Type of 155.

The Code will be used to identify RPL Control Messages as follows:

• 0x01: DAG Information Solicitation (Section 5.1.2 (DAG Information Solicitation (DIS)))
• 0x02: DAG Information Object (Section 5.1.3 (DAG Information Object (DIO)))
• 0x04: Destination Advertisement Object (Section 5.1.4 (Destination Advertisement Object (DAO)))

5.1.2.  DAG Information Solicitation (DIS)

The DAG Information Solicitation (DIS) message may be used to solicit a DAG Information Object from a RPL node. Its use is analogous to that of a Router Solicitation; a node may use DIS to probe its neighborhood for nearby DAGs. The DAG Information Solicitation carries no additional message body.

5.1.3.  DAG Information Object (DIO)

The DAG Information Object carries a number of metrics and other information that allows a node to discover a DAG, select its DAG parents, and identify its siblings while employing loop avoidance strategies.

5.1.3.1.  DIO Base Option

The DIO Base Option is a container option, which is always present, and might contain a number of suboptions. The base option regroups the minimum information set that is mandatory in all cases.

     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
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |G|D|A|0|0| Prf |   Sequence    |  InstanceID   |    DAGRank    |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |                                                               |
    +                                                               +
    |                            DAGID                              |
    +                                                               +
    |                                                               |
    +                                                               +
    |                                                               |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |   sub-option(s)...
    +-+-+-+-+-+-+-+-+-+-+-+-+-+

Control Field:

The DAG Control Field is currently allocated as follows:

Grounded (G):

The Grounded (G) flag is set when the DAG root is offering connectivity to an external routed infrastructure such as the Internet.

Destination Advertisement Trigger (D):

The Destination Advertisement Trigger (D) flag is set when the DAG root or another node in the successor chain decides to trigger the sending of destination advertisements in order to update routing state for the outward direction along the DAG, as further detailed in Section 5.10 (Establishing Routing State Outward Along the DAG). Note that the use and semantics of this flag are still under investigation.

Destination Advertisement Supported (A):

The Destination Supported (A) bit is set when the DAG root is capable to support the collection of destination advertisement related routing state and enables the operation of the destination advertisement mechanism within the DAG.

DAGPreference (Prf):

3-bit unsigned integer set by the DAG root to its preference and unchanged at propagation. DAGPreference ranges from 0x00 (least preferred) to 0x07 (most preferred). The default is 0 (least preferred). The DAG preference provides an administrative mechanism to engineer the self-organization of the LLN, for example indicating the most preferred LBR. If a node has the option to join a more preferred DAG while still meeting other optimization objectives, then the node will generally seek to join the more preferred DAG as determined by the OF.

Unassigned bits of the Control Field are considered as reserved. They MUST be set to zero on transmission and MUST be ignored on receipt.

Sequence Number:

8-bit unsigned integer set by the DAG root, incremented according to a policy provisioned at the DAG root, and propagated with no change outwards along the DAG. Each increment SHOULD have a value of 1 and may cause a wrap back to zero.

InstanceID:

8-bit field indicating the topology instance associated with the DAG, as provisioned at the DAG root.

DAGRank:

8-bit unsigned integer indicating the DAG rank of the node sending the DIO message. The DAGRank of the DAG root is ROOT_RANK. DAGRank is further described in Section 5.4 (DAG Discovery and Maintenance).

DAGID:

128-bit unsigned integer which uniquely identify a DAG. This value is set by the DAG root. The global IPv6 address of the DAG root can be used, however. the DAGID MUST be unique per DAG within the scope of the LLN. In the case where a DAG root is rooting multiple DAGs the DAGID MUST be unique for each DAG rooted at a specific DAG root.

The following values MUST NOT change during the propagation of DIO messages outwards along the DAG:

Grounded (G)

Destination Advertisement Supported (A)

DAGPreference (Prf)

Sequence

InstanceID

DAGID

All other fields of the DIO message may be updated at each hop of the propagation.

5.1.3.1.1.  DIO Suboptions

In addition to the minimum options presented in the base option, several suboptions are defined for the DIO message:

5.1.3.1.1.1.  Format

     0                   1                   2
     0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - -
    |  Subopt. Type |         Subopt Length         | Subopt Data
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - -

Suboption Type:

8-bit identifier of the type of suboption. When processing a DIO message containing a suboption for which the Suboption Type value is not recognized by the receiver, the receiver MUST silently ignore the unrecognized option, continue to process the following suboption, correctly handling any remaining options in the message.

Suboption Length:

16-bit unsigned integer, representing the length in octets of the suboption, not including the suboption Type and Length fields.

Suboption Data:

A variable length field that contains data specific to the option.

The following subsections specify the DIO message suboptions which are currently defined for use in the DAG Information Object.

Implementations MUST silently ignore any DIO message suboptions options that they do not understand.

DIO message suboptions may have alignment requirements. Following the convention in IPv6, these options are aligned in a packet such that multi-octet values within the Option Data field of each option fall on natural boundaries (i.e., fields of width n octets are placed at an integer multiple of n octets from the start of the header, for n = 1, 2, 4, or 8).

5.1.3.1.1.2.  Pad1

The Pad1 suboption does not have any alignment requirements. Its format is as follows:

     0
     0 1 2 3 4 5 6 7
    +-+-+-+-+-+-+-+-+
    |   Type = 0    |
    +-+-+-+-+-+-+-+-+

NOTE! the format of the Pad1 option is a special case - it has neither Option Length nor Option Data fields.

The Pad1 option is used to insert one or two octets of padding in the DIO message to enable suboptions alignment. If more than two octets of padding is required, the PadN option, described next, should be used rather than multiple Pad1 options.

5.1.3.1.1.3.  PadN

The PadN option does not have any alignment requirements. Its format is as follows:

     0                   1                   2
     0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - -
    |   Type = 1    |         Subopt Length         | Subopt Data
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - -

The PadN option is used to insert three or more octets of padding in the DIO message to enable suboptions alignment. For N (N > 2) octets of padding, the Option Length field contains the value N-3, and the Option Data consists of N-3 zero-valued octets. PadN Option data MUST be ignored by the receiver.

5.1.3.1.1.4.  DAG Metric Container

The DAG Metric Container suboption may be aligned as necessary to support its contents. Its format is as follows:

     0                   1                   2
     0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - -
    |   Type = 2    |       Container Length        | DAG Metric Data
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - -

The DAG Metric Container is used to report aggregated path metrics along the DAG. The DAG Metric Container may contain a number of discrete node, link, and aggregate path metrics as chosen by the implementer. The Container Length field contains the length in octets of the DAG Metric Data. The order, content, and coding of the DAG Metric Container data is as specified in [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.).

The processing and propagation of the DAG Metric Container is governed by implementation specific policy functions.

5.1.3.1.1.5.  Destination Prefix

The Destination Prefix suboption does not have any alignment requirements. Its format 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
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |   Type = 3    |            Length             |Resvd|Prf|Resvd|
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |                        Prefix Lifetime                        |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    | Prefix Length |                                               |
    +-+-+-+-+-+-+-+-+                                               |
    |             Destination Prefix (Variable Length)              |
    .                                                               .
    .                                                               .
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

The Destination Prefix suboption is used when the DAG root, or another node located inwards along the DAG on the path to the DAG root, needs to indicate that it offers connectivity to destination prefixes other than the default. This may be useful in cases where more than one LBR is operating within the LLN and offering connectivity to different administrative domains, e.g. a home network and a utility network. In such cases, upon observing the Destination Prefixes offered by a particular DAG, a node MAY decide to join multiple DAGs in support of a particular application.

The Length is coded as the length of the suboption in octets, excluding the Type and Length fields.

Prf is the Route Preference as in [RFC4191] (Draves, R. and D. Thaler, “Default Router Preferences and More-Specific Routes,” November 2005.). The reserved fields MUST be set to zero on transmission and MUST be ignored on receipt.

The Prefix Lifetime is a 32-bit unsigned integer representing the length of time in seconds (relative to the time the packet is sent) that the Destination Prefix is valid for route determination. A value of all one bits (0xFFFFFFFF) represents infinity. A value of all zero bits (0x00000000) indicates a loss of reachability.

The Prefix Length is an 16-bit unsigned integer that indicates the number of leading bits in the destination prefix.

The Destination Prefix contains Prefix Length significant bits of the destination prefix. The remaining bits of the Destination Prefix, as required to complete the trailing octet, are set to 0.

In the event that a DIO message may need to specify connectivity to more than one destination, the Destination Prefix suboption may be repeated.

5.1.3.1.1.6.  DAG Timer Configuration

The DAG Timer Configuration suboption does not have any alignment requirements. Its format 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
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |   Type = 4    |            Length             | DIOIntDoubl.  |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |  DIOIntMin.   |
    +-+-+-+-+-+-+-+-+

The DAG Timer Configuration suboption is used to distribute configuration information for DAG Timer Operation through the DAG. The information communicated in this suboption is generally static and unchanging within the DAG, therefore it is not necessary to include in every DIO. This suboption MAY be included periodically by the DAG Root, and SHOULD be included in response to a unicast request, e.g. a DAG Information Solicitation (DIS) message.

The Length is coded as 2.

DIOIntervalDoublings is an 8-bit unsigned integer. Configured on the DAG root and used to configure the trickle timer governing when DIO message should be sent within the DAG. DIOIntervalDoublings is the number of times that the DIOIntervalMin is allowed to be doubled during the trickle timer operation.

DIOIntervalMin is an 8-bit unsigned integer. Configured on the DAG root and used to configure the trickle timer governing when DIO message should be sent within the DAG. The minimum configured interval for the DIO trickle timer in units of ms is 2^DIOIntervalMin. For example, a DIOIntervalMin value of 16ms is expressed as 4.

5.1.4.  Destination Advertisement Object (DAO)

The Destination Advertisement Object (DAO) is used to propagate destination information inwards along the DAG. The RPL use of the DAO allows the nodes in the DAG to build up routing state for nodes contained in the sub-DAG in support of traffic flowing outward along the DAG.

     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
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |         DAO Sequence          |  InstanceID   |   DAO Rank    |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |                          DAO Lifetime                         |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |                           Route Tag                           |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    | Prefix Length |    RRCount    |                               |
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+                               |
    |                   Prefix (Variable Length)                    |
    .                                                               .
    .                                                               .
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
    |             Reverse Route Stack (Variable Length)             |
    .                                                               .
    .                                                               .
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

DAO Sequence:

Incremented by the node that owns the prefix for each new DAO message for that prefix.

InstanceID:

8-bit field indicating the topology instance associated with the DAG, as learned from the DIO.

DAO Rank:

Set by the node that owns the prefix and first issues the DAO message to its rank.

DAO Lifetime:

32-bit unsigned integer. The length of time in seconds (relative to the time the packet is sent) that the prefix is valid for route determination. A value of all one bits (0xFFFFFFFF) represents infinity. A value of all zero bits (0x00000000) indicates a loss of reachability.

Route Tag:

32-bit unsigned integer. The Route Tag may be used to give a priority to prefixes that should be stored. This may be useful in cases where intermediate nodes are capable of storing a limited amount of routing state. The further specification of this field and its use is under investigation.

Prefix Length:

Number of valid leading bits in the IPv6 Prefix.

RRCount:

8-bit unsigned integer. This counter is used to count the number of entries in the Reverse Route Stack. A value of `0' indicates that no Reverse Route Stack is present.

Prefix:

Variable-length field containing an IPv6 address or a prefix of an IPv6 address. The Prefix Length field contains the number of valid leading bits in the prefix. The bits in the prefix after the prefix length (if any) are reserved and MUST be set to zero on transmission and MUST be ignored on receipt.

Reverse Route Stack:

Variable-length field containing a sequence of RRCount (possibly compressed) IPv6 addresses. A node that adds on to the Reverse Route Stack will append to the list and increment the RRCount.

5.2.  Conceptual Data Structures

The RPL implementation MUST maintain the following conceptual data structures in support of DAG discovery:

• A set of candidate neighbors
• For each DAG:
• A set of DAG parents and siblings

5.2.1.  Candidate Neighbors Data Structure

The set of candidate neighbors is to be populated by neighbors that are discovered by the neighbor discovery mechanism and further qualified as statistically stable as per the mechanisms discussed in [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.). The candidate neighbors, and related metrics, should demonstrate stability/reliability beyond a certain threshold, and it is recommended that a local confidence value be maintained with respect to the neighbor in order to track this. Implementations MAY choose to bound the maximum size of the candidate neighbor set, in which case a local confidence value will assist in ordering neighbors to determine which ones should remain in the candidate neighbor set and which should be evicted.

If Neighbor Unreachability Detection (NUD) determines that a candidate neighbor is no longer reachable, then it shall be removed from the candidate neighbor set. In the case that the candidate neighbor has associated states in the DAG parent set or active DA entries, then the removal of the candidate neighbor shall be coordinated with tearing down these states. All provisioned routes associated with the candidate neighbor should be removed.

5.2.2.  Directed Acyclic Graphs (DAGs) Data Structure

At a given point of time, a DAG Iteration is uniquely identified by the tuple (DagID, InstanceID, DAGSequenceNumber) where a change in the sequence denotes the iteration of a given DAG over time. When a single device is capable to root multiple DAGs in support of an application need for multiple optimization objectives it MUST produce a different and unique (DagID, InstanceID) pair for each of the multiple DAGs.

For each DAG that a node is, or may become, a member of, the implementation MUST keep a DAG table with the following entries:

• InstanceID
• DAGID
• DAGSequenceNumber
• DAG Metric Container, including DAGObjectiveCodePoint
• A set of Destination Prefixes offered inwards along the DAG
• A set of DAG parents and siblings
• A timer to govern the sending of DIO messages for the DAG

When a DAG is discovered for which no DAG data structure is instantiated, and the node wants to join, then the DAG data structure is instantiated.

When the DAG parent set is depleted (i.e. the last DAG is removed), then the DAG data structure SHOULD be suppressed after the expiration of an implementation-specific local timer. An implementation SHOULD delay before deallocating the DAG data structure in order to observe that the DAGSequenceNumber has incremented should any new DAG parents appear for the DAG.

5.2.2.1.  DAG Parents/Siblings Structure

When the DAG is self-rooted, the set of DAG parents/siblings is empty.

In all other cases, for each node in the set, the implementation MUST keep a record of:

• a reference to the neighboring device which is the DAG parent or sibling
• a record of most recent information taken from the DAG Information Object last processed from the DAG parent

DAG parents may be ordered, according to the OF. When ordering DAG parents, in consultation with the OF, the most preferred DAG parent may be identified. All current DAG parents must have a rank less than self. All current DAG siblings must have a rank equal to self.

When nodes are added to or removed from the DAG set the most preferred DAG parent may have changed. The role of all the nodes in the list should be reevaluated. In particular, any nodes having a rank greater than self after such a change must be evicted from the set.

An implementation may choose to keep these records as an extension of the Default Router List (DRL).

5.3.  DAG Rank

Based on the selection of DAG Parents, the metrics conveyed by the most preferred DAG parent, the nodes own metrics and configuration, and a related function defined by the OF, a node will be able to compute a value for its rank as a consequence of selecting a most preferred DAG parent.

The rank value feeds back into the DAG parent selection according to a loop-avoidance strategy. Once a DAG parent has been added, and a rank value for the node within the DAG has been computed, the nodes further options with regard to DAG parent selection and movement within the DAG are restricted in favor of loop avoidance.

It is important to note that the DAG Rank is not itself a metric, although its value is derived from and influenced by the use of metrics to select DAG parents and take up a position in the DAG. The only aim of the rank is to inform loop avoidance and detection.

The computation of the DAG Rank MUST be done in such a way so as to maintain the following properties for any nodes M and N that are neighbors in the LLN:

DAGRank(M) is less than DAGRank(N):

In this case, M is probably located in a more preferred position than N in the DAG with respect to the metrics and optimizations defined by the objective code point. In any fashion, Node M may safely be a DAG parent for Node N without risk of creating a loop. Further, for a node N, all parents in the DAG parent set must be of rank less than self's DAGRank(N). In other words, the rank presented by a node N MUST be greater (deeper) than that presented by any of its parents.

DAGRank(M) equals DAGRank(N):

In this case M and N are located positions of relatively the same optimality within the DAG. In some cases, Node M may be used as a successor by Node N, but with related chance of creating a loop that must be detected and broken by some other means.

DAGRank(M) is greater than DAGRank(N):

In this case, then node M is located in a less preferred position than N in the DAG with respect to the metrics and optimizations defined by the objective code point. Further, Node (M) may in fact be in Node (N)'s sub-DAG. There is a higher risk to Node (N) selecting Node (M) as a DAG parent, as such a selection may create a loop.

As an example, the DAG Rank could be computed in such a way so as to closely track ETX when the objective function is to minimize ETX, or latency when the objective function is to minimize latency, or in a more complicated way as appropriate to the objective code point being used within the DAG.

5.4.  DAG Discovery and Maintenance

DAG discovery locates the nearest sink (aka root), as determined according to some metrics and constraints, and forms a Directed Acyclic Graph towards that sink, by identifying a set of DAG parents. During this process DAG discovery also identifies siblings, which may be used later to provide additional path diversity towards the DAG root. DAG discovery enables nodes to implement different policies for selecting their DAG parents in the DAG by using implementation specific policy functions. DAG discovery specifies a set of rules to be followed by all implementations in order to ensure interoperation. DAG discovery also standardizes the format that is used to advertise the most common information that is used in order to select DAG parents.

One of these information, the DAG rank, is used by DAG discovery to provide loop avoidance even if nodes implement different policies. The DAG Rank is computed as specified by the OF in use by the DAG, demonstrating the properties described in Section 5.3 (DAG Rank). The rank should be computed in such a way so as to provide a comparable basis with other nodes which may not use the same metric at all.

The DAG discovery procedures take into account a number of factors, including:

• RPL rules for loop avoidance based on DAGs and ranks
• The Objective Function
• The advertised metrics
• Local policy functions (e.g. a bounded number of candidate neighbors).

5.4.1.  DAG Discovery Rules

In order to organize and maintain loopless structure, the DAG discovery implementation in the nodes MUST obey to the following rules and definitions:

5.4.1.1.  DAGs

1. DAG discovery instantiates LLN topologies that are each optimized for specific constraints and goals. A topology assumes the shape of a DAG, and a DAG Instance is uniquely identified by its instanceID.
2. For reasons of scalability and operations of the protocol, a DAG Instance is partitioned into a set of DAGs rooted at a destination, aka Destination Oriented DAGs. A destination is uniquely identified by a DAGID so a DAG rooted at a destination is uniquely identified by the pair (InstanceID, DAGID).
3. A Destination Oriented DAG is periodically reconstructed from the root, by incrementing a DAGSequenceNumber. An Iteration of a Destination Oriented DAG is thus uniquely identified by the tuple (InstanceID, DAGID, DAGSequenceNumber). Through this document, the graph formed by this iterative process is referred to as the DAG Iteration, or in short, the DAG.
4. The rank is defined within the scope of a DAG Iteration as an abstract coordinate to compare the relative position of nodes and ensure forward progress of the traffic.
5. A node MUST belong at most to one DAG Iteration per InstanceID and MUST select all its parents and siblings within that same DAG Iteration.

5.4.1.2.  DAG Sequence Number

1. The DAGSequenceNumber is incremented by the root and flooded through DIOs.
2. The root floods a new DAGSequenceNumber periodically, at a rate that depends on the deployment. This rate can be set to 0 if other methods such as loop detection are considered sufficient to solve the routing issues in that deployment.
3. The root MAY also flood a new DAGSequenceNumber on-demand. The details of the mechanism to signal the root to do so are to be specified in a future revision of this document.
4. A parent that advertises the new DAGSequenceNumber can not possibly belong to the sub-DAG of a node that still advertises an older DAGSequenceNumber. The node MAY thus attach to that parent regardless of the relative rank, and this situation is equivalent to jumping onto a different Destination Oriented DAG.
5. Thus, as a new DAGSequenceNumber spreads, a new DAG Iteration forms that supersedes the previous one. During a DAGSequenceNumber transition, a node MAY decide to forward packets via 'future parents' that belong to the same Destination Oriented DAG (same InstanceID and DagID), but a more recent (incremented) DAGSequenceNumber.

5.4.1.3.  DAG Root

1. A node that does not have any DAG parent MAY become the root of its own floating DAG. It's rank is ROOT_RANK.
2. A (non-LLN) router is considered connected to a grounded infrastructure at rank BASE_RANK. A LLN node that is attached to such an infrastructure router is the DAG root of its own grounded DAG. It's rank is ROOT_RANK.
3. In a deployment that uses a backbone link to federate a number of LLN roots, it is possible to run RPL over the backbone and use one router as a backbone root. The backbone root exposes a rank of BASE_RANK over the backbone. All the LLN roots that are parented to that backbone root, including the backbone root if it also serves as LLN root, expose a rank of ROOT_RANK over the LLN and act as multiple roots for a same DAG, coordinated by the backbone root.
4. The DAG root exposes the DAG in the DIO message and LLN nodes propagate the DIO message outwards along the DAG.

5.4.1.4.  Moving Inside a DAG

1. A node moves when it changes its parent selection within the same DAG Iteration. When a node moves (within its DAG) in a fashion that cause its rank to decrease, the node MUST abandon all parents and siblings with a rank larger than self, and MAY adopt as siblings nodes with the same rank.
2. A node MAY move at any time, with no delay, within its DAG when the move does not cause the node to increase its own DAG rank, as per the rank calculation indicated by the OF.
3. A node MUST NOT move outwards along a DAG that it is attached to, causing the DAG rank to increase. If a node cannot stay within the DAG without a rank increase, then it MUST poison its routes as described in Section 5.4.1.6 (Poisoning a Broken Path).
4. When DIO messages are received from other routers located at lesser rank in the same DAG, those routers are eligible for consideration as DAG parents. DIO messages received from other routers located at the same rank in the same DAG may be considered as coming from siblings. DIO messages that are received from other routers located at greater rank within the same DAG might cause greedy behaviors and loops; such a DIO is ignored unless:
   1. The DIO comes from an existing parent or sibling; in which case that parent must be removed.
   2. The DIO comes from a node that has better OF ratings than any parent known at this point; in that case, this potential parent MAY be remembered in order to jump at a better position when the next sequence is flooded.

5.4.1.5.  Jumping Onto Another DAG

1. A node jumps when it performs a new parent selection whereby its DAG Iteration changes within the same DAG Instance. When a node jumps onto a new DAG Iteration, it MUST abandon all parents and siblings from its previous position.
2. A node MAY jump from its current DAG onto any other DAG that provides service for the same InstanceID if it is preferred by the OF, for example for reasons such as connectivity, configured preference, free medium time, size, security, bandwidth, DAG rank, or whatever metrics the LLN uses. This is allowed regardless of the rank that the node reaches in the new DAG.
3. A node that jumps should attempt to transmit all the packets received as part of the previous DAG along the previous DAG. In other words, it should switch the parent set only after the outstanding packet queue of packets received prior to announcing the jump is exhausted.
4. Jumping back onto a previous DAG is equivalent to moving inside that DAG and obeys the same rules. To satisfy this, a node detaching from a DAG SHOULD remember its DAG as identified by the tuple (InstanceID, DagID, DAGSequenceNumber) as well as its rank within that DAG for long as that DAG exists.

5.4.1.6.  Poisoning a Broken Path

1. A node SHOULD poison its inwards routes when it looses all of its current feasible parents, i.e. the set of DAG parents becomes depleted, and it can not jump onto an alternate DAG.
2. In order to poison its inwards routes, a node MAY stay at its position within its DAG (that is maintain its InstanceID, DagID, DAGSequenceNumber and Rank) but it SHOULD immediately advertise a rank of INFINITE_RANK in a DIO so as to force all its children to remove it from their parent list and try an alternate path. The node SHOULD then wait for a new DAG Iteration (DAGSequenceNumber increment) before resuming its operation in the same Destination Oriented DAG.
3. Alternatively, a node MAY detach from its DAG. A node that detaches becomes root of its own floating DAG and MUST immediately advertise its new situation in a DIO.
4. Either way, the route poisoning will recursively be flooded throughout the impacted sub-DAG as children lose their last parent in the original DAG.
5. The loss of a DIO message may interrupt the flooding. This can be compensated by cheer repetition through the trickle algorithm. If that also fails, packet loops will be prevented by the detection mechanism described in Section 5.11 (Loop Detection).

5.4.1.7.  Following a Parent

1. If a node that receives a DIO from one of its DAG parents indicating that the parent has left the DAG, it may either follow that parent or stay in its current DAG through an alternate DAG parent if that is possible.
2. If a DAG parent increases its rank such that the node rank would have to change, and if the node does not wish to follow (e.g. it has alternate options), then the DAG parent SHOULD be evicted from the DAG parent set. If the DAG parent is the last in the DAG parent set, then the node SHOULD chose to follow it.

5.4.1.8.  DAG Inconsistency

1. When a node detects or causes a DAG inconsistency, as described in Section 5.4.4.2 (Determination of Inconsistency), then the node SHOULD send an unsolicited DIO message to its one-hop neighbors. The DIO is updated to propagate the new DAG information. Such an event MUST also cause the trickle timer governing the periodic sending of DIO messages to be reset.

5.4.2.  Reception and Processing of DIO messages

When an DIO message is received from a source device named SRC, the receiving node must first determine whether or not the DIO message should be accepted for further processing, and subsequently present the DIO message for further processing if eligible.

1. If the DIO message is malformed, then the DIO message is not eligible for further processing and is silently discarded. A RPL implementation MAY log the reception of a malformed DIO message.
2. If SRC is not a member of the candidate neighbor set, then the DIO is not eligible for further processing. (Further evaluation/confidence of this neighbor is necessary)
3. If the DIO message advertises a DAG that the node is already a member of, then:
• If the rank of SRC as reported in the DIO message is lesser than that of the node within the DAG, then the DIO message MUST be considered for further processing.
• If the rank of SRC as reported in the DIO message is equal to that of the node within the DAG, then SRC is marked as a sibling and the DIO message is not eligible for further processing.
• If the rank of SRC as reported in the DIO message is higher than that of the node within the DAG, and SRC is not a DAG parent, then the DIO message MUST NOT be considered for further processing
4. Even if not processed further, information from a DIO might be remembered for instance if SRC is preferable to the current parents per the OF selection process.
5. If SRC is a DAG parent for any other DAG that the node is attached to, then the DIO message MUST be considered for further processing (the DAG parent may have jumped).
6. If the DIO message advertises a DAG that offers a better (new or alternate) solution to an optimization objective desired by the node, then the DIO message MUST be considered for further processing.

5.4.2.1.  Overview of DIO Message Processing

If the received DIO message is for a new/alternate DAG:

If the node has sent an DIO message within the risk window as described in Section 5.8 (Collision) then a collision has occurred; do not process the DIO message any further.

If the SRC node is also a DAG parent for another DAG that the node is a member of, and if the new/alternate DAG is the same InstanceID as the other DAG, then the DAG parent is known to have jumped.

Remove SRC as a DAG parent from the other DAG

If the other DAG is now empty of candidate parents, then prepare to directly follow SRC into the new DAG by adding it as a DAG parent for the new DAG, else ignore the DIO message (do not follow the parent).

Instantiate a data structure for the new/alternate DAG if necessary

If the new/alternate DAG offers a better solution to the optimization objectives, then jump: copy the DIO information place the neighbor into the DAG parent set.

If the DIO message is for a known/existing DAG:

Process the DIO message as per the rules in Section 5.4 (DAG Discovery and Maintenance)

As DIO messages are received from candidate neighbors, the neighbors may be promoted to DAG parents by following the rules of DAG discovery as described in Section 5.4 (DAG Discovery and Maintenance). When a node places a neighbor into the DAG Parent set, the node becomes attached to the DAG through the new parent node.

In the DAG discovery implementation, the most preferred parent should be used to restrict which other nodes may become DAG parents. Some nodes in the DAG parent set may be of a rank less than or equal to the most preferred DAG parent. (This case may occur, for example, if an energy constrained device is at a lesser rank but should be avoided as per an optimization objective, resulting in a more preferred parent at a greater rank).

5.4.3.  DIO Transmission

Each node maintains a timer that governs when to multicast DIO messages. This timer is implemented as a trickle timer operating over a variable interval. Trickle timers are further detailed in Section 5.4.4 (Trickle Timer for DIO Transmission). The governing parameters for the timer should be configured consistently across the DAG, and are provided by the DAG root in the DIO message. In addition to periodic DIO messages, each node may respond to a DIS message with a DIO message.

• When a node detects an inconsistency, it SHOULD reset the interval of the trickle timer to a minimum value, causing DIO messages to be emitted more frequently as part of a strategy to quickly correct the inconsistency. Such inconsistencies may be, for example, an update to a key parameter (e.g. sequence number) in the DIO message or a loop detected when a node located inwards along the DAG forwards traffic outwards. Inconsistencies are further detailed in Section 5.4.4.2 (Determination of Inconsistency).
• When a node enters a mode of consistent operation within a DAG, i.e. DIO messages from its DAG parents are consistent and no other inconsistencies are detected, it may begin to open up the interval of the trickle timer towards a maximum value, causing DIO messages to be emitted less frequently, thus reducing network maintenance overhead and saving energy consumption.
• When a node is initialized, it MAY be configured to remain silent and not multicast any DIO messages until it has encountered and joined a DAG (perhaps initially probing for a nearby DAG with an DIS message). Alternately, it may choose to root its own floating DAG and begin multicasting DIO messages using a default trickle configuration. The second case may be advantageous if it is desired for independent nodes to begin aggregating into scattered floating DAGs in the absence of a grounded node, for example in support of LLN installation and commissioning.

Note that if multiple DAG roots are participating in the same DAG, i.e. offering DIO messages with the same DAGID, then they must coordinate with each other to ensure that their DIO messages are consistent when they emit DIO messages. In particular the Sequence number must be identical from each DAG root, regardless of which of the multiple DAG roots issues the DIO message, and changes to the Sequence number should be issued at the same time. The specific mechanism of this coordination, e.g. along a non-LLN network between DAG roots, is beyond the scope of this specification.

5.4.4.  Trickle Timer for DIO Transmission

RPL treats the construction of a DAG as a consistency problem, and uses a trickle timer [Levis08] (Levis, P., Brewer, E., Culler, D., Gay, D., Madden, S., Patel, N., Polastre, J., Shenker, S., Szewczyk, R., and A. Woo, “The Emergence of a Networking Primitive in Wireless Sensor Networks,” July 2008.) to control the rate of control broadcasts.

For each DAG that a node is part of, the node must maintain a single trickle timer. The required state contains the following conceptual items:

I:

The current length of the communication interval

T:

A timer with a duration set to a random value in the range [I/2, I]

C:

Redundancy Counter

I_min:

The smallest communication interval in milliseconds. This value is learned from the DIO message as (2^DIOIntervalMin)ms. The default value is DEFAULT_DIO_INTERVAL_MIN.

I_doublings:

The number of times I_min should be doubled before maintaining a constant rate, i.e. I_max = I_min * 2^I_doublings. This value is learned from the DIO message as DIOIntervalDoublings. The default value is DEFAULT_DIO_INTERVAL_DOUBLINGS.

5.4.4.1.  Resetting the Trickle Timer

The trickle timer for a DAGID is reset by:

1. Setting I_min and I_doublings to the values learned from the DIO message.
2. Setting C to zero.
3. Setting I to I_min.
4. Setting T to a random value as described above.
5. Restarting the trickle timer to expire after a duration T

When node learns about a DAG through a DIO message and makes the decision to join it, it initializes the state of the trickle timer by resetting the trickle timer and listening. Each time it hears a consistent DIO message for this DAG from a DAG parent, it MAY increment C.

When the timer fires at time T, the node compares C to the redundancy constant, DEFAULT_DIO_REDUNDANCY_CONSTANT. If C is less than that value, the node generates a new DIO message and multicasts it. When the communication interval I expires, the node doubles the interval I so long as it has previously doubled it fewer than I_doubling times, resets C, and chooses a new T value.

5.4.4.2.  Determination of Inconsistency

The trickle timer is reset whenever an inconsistency is detected within the DAG, for example:

• The node joins a new DAGID
• The node moves within a DAGID
• The node receives a modified DIO message from a DAG parent
• A DAG parent forwards a packet intended to move inwards, indicating an inconsistency and possible loop.
• A metric communicated in the DIO message is determined to be inconsistent, as according to a implementation specific path metric selection engine.
• The rank of a DAG parent has changed.

5.5.  DAG Sequence Number Increment

The DAG root makes the sole determination of when to revise the DAGSequenceNumber by incrementing it upwards. When the DAGSequenceNumber is increased an inconsistency results, causing DIO messages to be sent back outwards along the DAG to convey the change. The degree to which this mechanism is relied on may be determined by the implementation- on one hand it may serve as a periodic heartbeat, refreshing the DAG states, and on the other hand it may result in a constant steady-state control cost overhead which is not desirable.

Some implementations may provide an administrative interface, such as a command line, at the DAG root whereby the DAGSequenceNumber may be caused to increment in response to some policy outside of the scope of RPL.

Other implementations may make use of a periodic timer to automatically increment the DAGSequenceNumber, resulting in a periodic DAG iteration at a rate appropriate to the application and implementation. Other automated mechanisms to determine DAGSequenceNumber increments are also possible as appropriate to a deployment.

5.6.  DAG Selection

The DAG selection is implementation and algorithm dependent. Nodes SHOULD prefer to join DAGs for InstanceIDs advertising OCPs and destinations compatible with their implementation specific objectives. In order to limit erratic movements, and all metrics being equal, nodes SHOULD keep their previous selection. Also, nodes SHOULD provide a means to filter out a candidate parent whose availability is detected as fluctuating, at least when more stable choices are available.

When connection to a fixed network is not possible or preferable for security or other reasons, scattered DAGs MAY aggregate as much as possible into larger DAGs in order to allow connectivity within the LLN.

A node SHOULD verify that bidirectional connectivity and adequate link quality is available with a candidate neighbor before it considers that candidate as a DAG parent.

5.7.  Administrative rank

When the DAG is formed under a common administration, or when a node performs a certain role within a community, it might be beneficial to associate a range of acceptable rank with that node. For instance, a node that has limited battery should be a leaf unless there is no other choice, and may then augment the rank computation specified by the OF in order to expose an exaggerated rank.

5.8.  Collision

A race condition occurs if 2 nodes send DIO messages at the same time and then attempt to join each other. This might happen, for example, between nodes which act as DAG root of their own DAGs. In order to detect the situation, LLN Nodes time stamp the sending of DIO message. Any DIO message received within a short link-layer-dependent period introduces a risk. It is up to the implementation to define the duration of the risk window.

There is risk of a collision when a node receives and processes a DIO within the risk window. For example, it may occur that two nodes are associated with different DAGs and near-simultaneously send DIO messages, which are received and processed by both, and possibly result in both nodes simultaneously deciding to attach to each other. As a remedy, in the face of a potential collision, as determined by receiving a DIO within the risk window, the DIO message is not processed. It is expected that subsequent DIOs would not cross.

5.9.  Guidelines for Objective Functions

5.9.1.  Objective Function

An Objective Function (OF) allows for the selection of a DAG to join, and a number of peers in that DAG as parents. The OF is used to compute an ordered list of parents. The OF is also responsible to compute the rank of the device within the DAG.

The Objective Function is specified in the DIO message within a DAG Metric Container using an Objective Code Point (OCP), as specified in [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.), and indicates the method that must be used to compute the DAG (e.g. "minimize the path cost using the ETX metric and avoid `Blue' links"). The Objective Code Points are specified in [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.). This document specifies an Objective Function, OF0, in support of default operation. In the case where the DIO does not include an OCP specification in the DAG Metric Container, OF0 MAY be presumed.

Most Objective Functions are expected to follow the same abstract behavior:

• The parent selection is triggered each time an event indicates that a potential next hop information is updated. This might happen upon the reception of a DIO message, a timer elapse, or a trigger indicating that the state of a candidate neighbor has changed.
• An OF scans all the interfaces on the device. Although there may typically be only one interface in most application scenarios, there might be multiple of them and an interface might be configured to be usable or not for RPL operation. An interface can also be configured with a preference or dynamically learned to be better than another by some heuristics that might be link-layer dependent and are out of scope. Finally an interface might or not match a required criterion for an Objective Function, for instance a degree of security. As a result some interfaces might be completely excluded from the computation, while others might be more or less preferred.
• An OF scans all the candidate neighbors on the possible interfaces to check whether they can act as a router for a DAG. There might be multiple of them and a candidate neighbor might need to pass some validation tests before it can be used. In particular, some link layers require experience on the activity with a router to enable the router as a next hop.
• An OF computes self's rank by adding the step of rank to that candidate to the rank of that candidate. The step of rank is computed by estimating the link as follows:
• The step of rank might vary from 1 to 16.
• 1 indicates a unusually good link, for instance a link between powered devices in a mostly battery operated environment.
• 4 indicates a `normal'/typical link, as qualified by the implementation.
• 16 indicates a link that can hardly be used to forward any packet, for instance a radio link with quality indicator or expected transmission count that is close to the acceptable threshold.
• Candidate neighbors that would cause self's rank to increase are ignored
• Candidate neighbors that advertise an OF incompatible with the set of OF specified by the policy functions are ignored.
• As it scans all the candidate neighbors, the OF keeps the current best parent and compares its capabilities with the current candidate neighbor. The OF defines a number of tests that are critical to reach the objective. A test between the routers determines an order relation.
• If the routers are roughly equal for that relation then the next test is attempted between the routers,
• Else the best of the 2 becomes the current best parent and the scan continues with the next candidate neighbor
• Some OFs may include a test to compare the ranks that would result if the node joined either router
• When the scan is complete, the preferred parent is elected and self's rank is computed as the preferred parent rank plus the step in rank with that parent.
• Other rounds of scans might be necessary to elect alternate parents and siblings. In the next rounds:
• Candidate neighbors that are not in the same DAG are ignored
• Candidate neighbors that are of greater rank than self are ignored
• Candidate neighbors of an equal rank to self (siblings) are ignored
• Candidate neighbors of a lesser rank than self (non-siblings) are preferred

5.9.2.  Objective Function 0 (OF0)

This document specifies a default objective function, called OF0, indicated by an OCP value of 0x0000. OF0 is the default objective function of RPL, and can be used if allowed by the policy of the processing node when the OF indicated in the DIO message is unknown to the node. If not allowed, then the DIO message is simply ignored and not processed by the node. OF0 is notable in that it does not use physical metrics as described in [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.), but is only based on abstract information from the DIO message such as rank and administrative preference.

OF0 favors connectivity. That is, the Objective Function is designed to find the nearest sink into a 'grounded' topology, and if there is none then join any network per order of administrative preference. The metric in use is the rank.

OF0 selects a preferred parent and a backup next hop if one is available. The backup next hop might be a parent or a sibling. All the traffic is routed via the preferred parent. When the link conditions do not let a packet through to the preferred parent, the packet is passed to the backup next hop.

The step of rank is 4 for each hop.

5.9.2.1.  Selection of the Preferred Parent

As it scans all the candidate neighbors, OF0 keeps the parent that is the best for the following criteria (in order):

1. The interface must be usable and any administrative preference associated with the interface applies first.
2. A candidate that would cause the node to augment the rank in the current DAG is not considered.
3. A router that has been validated as usable, e.g. with a local confidence that has exceeded some pre-configured threshold, is better.
4. If none are grounded then a DAG with a more preferred administrative preference (DAGPreference) is better.
5. A router that offers connectivity to a grounded DAG is better.
6. A lesser resulting rank is better.
7. A DAG for which there is an alternate parent is better. This check is optional. It is performed by computing the backup next hop while assuming that this router won.
8. The DAG that was in use already is preferred.
9. The preferred parent that was in use already is better.
10. A router that has announced a DIO message more recently is preferred.

5.9.2.2.  Selection of the Backup Next Hop

• The interface must be usable and the administrative preference (if any) applies first.
• The preferred parent is ignored.
• Candidate neighbors that are not in the same DAG are ignored.
• Candidate neighbors with a higher rank are ignored.
• Candidate neighbors of a better rank than self (non-siblings) are preferred.
• A router that has been validated as usable, e.g. with a local confidence that has exceeded some pre-configured threshold, is better.
• The router with a better router preference wins.
• The backup next hop that was in use already is better.

5.10.  Establishing Routing State Outward Along the DAG

The destination advertisement mechanism supports the dissemination of routing state required to support traffic flows outward along the DAG, from the DAG root toward nodes.

As a result of destination advertisement operation:

• DAG discovery establishes a DAG oriented toward a DAG root along which inward routes toward the DAG root are set up.
• Destination advertisement establishes outward routes along the DAG. Such paths consist of:
• Hop-By-Hop routing state within islands of `stateful' nodes.
• Source Routing `bridges' across nodes that do not retain state.

Destinations disseminated with the destination advertisement mechanism may be prefixes, individual hosts, or multicast listeners. The mechanism supports nodes of varying capabilities as follows:

• When nodes are capable of storing routing state, they may inspect destination advertisements and learn hop-by-hop routing state toward destinations by populating their routing tables with the routes learned from nodes in their sub-DAG. In this process they may also learn necessary piecewise source routes to traverse regions of the LLN that do not maintain routing state. They may perform route aggregation on known destinations before emitting Destination Advertisements.
• When nodes are incapable of storing routing state, they may forward destination advertisements, recording the reverse route as the go in order to support the construction of piecewise source routes.

Nodes that are capable of storing routing state, and finally the DAG roots, are able to learn which destinations are contained in the sub-DAG below the node, and via which next-hop neighbors. The dissemination and installation of this routing state into nodes allows for Hop-By-Hop routing from the DAG root outwards along the DAG. The mechanism is further enhance by supporting the construction of source routes across stateless `gaps' in the DAG, where nodes are incapable of storing additional routing state. An adaptation of this mechanism allows for the implementation of loose-source routing.

A special case, the reception of a destination advertisement addressed to a link-local multicast address, allows for a node to learn destinations directly available from its one-hop neighbors.

A design choice behind advertising routes via destination advertisements is not to synchronize the parent and children databases along the DAG, but instead to update them regularly to recover from the loss of packets. The rationale for that choice is time variations in connectivity across unreliable links. If the topology can be expected to change frequently, synchronization might be an excessive goal in terms of exchanges and protocol complexity. The approach used here results in a simple protocol with no real peering. The destination advertisement mechanism hence provides for periodic updates of the routing state, as cued by occasional RAs and other mechanisms, similarly to other protocols such as RIP [RFC2453] (Malkin, G., “RIP Version 2,” November 1998.).

5.10.1.  Destination Advertisement Operation

5.10.1.1.  Overview

According to implementation specific policy, a subset or all of the feasible parents in the DAG may be selected to receive prefix information from the destination advertisement mechanism. This subset of DAG parents shall be designated the set of DA parents.

As DAO messages for particular destinations move inwards along the DAG, a sequence counter is used to guarantee their freshness. The sequence counter is incremented by the source of the DAO message (the node that owns the prefix, or learned the prefix via some other means), each time it issues a DAO message for its prefix. Nodes that receive the DAO message and, if scope allows, will be forwarding a DAO message for the unmodified destination inwards along the DAG, will leave the sequence number unchanged. Intermediate nodes will check the sequence counter before processing a DAO message, and if the DAO is unchanged (the sequence counter has not changed), then the DAO message will be discarded without additional processing. Further, if the DAO message appears to be out of synch (the sequence counter is 2 or more behind the present value) then the DAO state is considered to be stale and may be purged, and the DAO message is discarded. A depth is also added for tracking purposes; the depth is incremented at each hop as the DAO message is propagated up the DAG. Nodes that are storing routing state may use the depth to determine which possible next-hops for the destination are more optimal.

If destination advertisements are activated in the DIO message as indicated by the `D' bit, the node sends unicast destination advertisements to one of its DA parents, that is selected as most favored for incoming outwards traffic. The node only accepts unicast destination advertisements from any nodes but those contained in the DA parent subset.

Receiving a DIO message with the `D' destination advertisement bit set from a DAG parent stimulates the sending of a delayed destination advertisement back, with the collection of all known prefixes (that is the prefixes learned via destination advertisements for nodes lower in the DAG, and any connected prefixes). If the Destination Advertisement Supported (A) bit is set in the DIO message for the DAG, then a destination advertisement is also sent to a DAG parent once it has been added to the DA parent set after a movement, or when the list of advertised prefixes has changed.

A node that modifies its DAG Parent set may set the `D' bit in subsequent DIO propagation in order to trigger destination advertisements to be updated to its DAG Parents and other inward nodes on the DAG. Additional recommendations and guidelines regarding the use of this mechanism are still under consideration and will be elaborated in a future revision of this specification.

Destination advertisements may advertise positive (prefix is present) or negative (removed) DAO messages, termed as no-DAOs. A no-DAO is stimulated by the disappearance of a prefix below. This is discovered by timing out after a request (a DIO message) or by receiving a no-DAO. A no-DAO is a conveyed as a DAO message with a DAO Lifetime of ZERO_LIFETIME.

A node that is capable of recording the state information conveyed in a unicast DAO message will do so upon receiving and processing the DAO message, thus building up routing state concerning destinations below it in the DAG. If a node capable of recording state information receives a DAO message containing a Reverse Route Stack, then the node knows that the DAO message has traversed one or more nodes that did not retain any routing state as it traversed the path from the DAO source to the node. The node may then extract the Reverse Route Stack and retain the included state in order to specify Source Routing instructions along the return path towards the destination. The node MUST set the RRCount back to zero and clear the Reverse Route Stack prior to passing the DAO message information on.

A node that is unable to record the state information conveyed in the DAO message will append the next-hop address to the Reverse Route Stack, increment the RRCount, and then pass the destination advertisement on without recording any additional state. In this way the Reverse Route Stack will contain a vector of next hops that must be traversed along the reverse path that the DAO message has traveled. The vector will be ordered such that the node closest to the destination will appear first in the list. In such cases, if it is useful to the implementation to try and build up redundant paths, the node may choose to convey the destination advertisement to one or more DAG parents in order of preference as guided by an implementation specific policy.

In some cases (called hybrid cases), some nodes along the path a destination advertisement follows inward along the DAG may store state and some may not. The destination advertisement mechanism allows for the provisioning of routing state such that when a packet is traversing outwards along the DAG, some nodes may be able to directly forward to the next hop, and other nodes may be able to specify a piecewise source route in order to bridge spans of stateless nodes within the path on the way to the desired destination.

In the case where no node is able to store any routing state as destination advertisements pass by, and the DAG root ends up with DAO messages that contain a completely specified route back to the originating node in the form of the inverted Reverse Route Stack. A DAG root should not request (Destination Advertisement Trigger) nor indicate support (Destination Advertisement Supported) for destination advertisements if it is not able to store the Reverse Route Stack information in this case.

The destination advertisement mechanism requires stateful nodes to maintain lists of known prefixes. A prefix entry contains the following abstract information:

• A reference to the ND entry that was created for the advertising neighbor.
• The IPv6 address and interface for the advertising neighbor.
• The logical equivalent of the full destination advertisement information (including the prefixes, depth, and Reverse Route Stack, if any).
• A 'reported' Boolean to keep track whether this prefix was reported already, and to which of the DA parents.
• A counter of retries to count how many DIO messages were sent on the interface to the advertising neighbor without reachability confirmation for the prefix.

Note that nodes may receive multiple information from different neighbors for a specific destination, as different paths through the DAG may be propagating information inwards along the DAG for the same destination. A node that is recording routing state will keep track of the information from each neighbor independently, and when it comes time to propagate the DAO message for a particular prefix to the DA parents, then the DAO information will be selected from among the advertising neighbors who offer the least depth to the destination.

The destination advertisement mechanism stores the prefix entries in one of 3 abstract lists; the Connected, the Reachable and the Unreachable lists.

The Connected list corresponds to the prefixes owned and managed by the local node.

The Reachable list contains prefixes for which the node keeps receiving DAO messages, and for those prefixes which have not yet timed out.

The Unreachable list keeps track of prefixes which are no longer valid and in the process of being deleted, in order to send DAO messages with zero lifetime (also called no-DAO) to the DA parents.

5.10.1.1.1.  Destination Advertisement Timers

The destination advertisement mechanism requires 2 timers; the DelayDAO timer and the RemoveTimer.

• The DelayDAO timer is armed upon a stimulation to send a destination advertisement (such as a DIO message from a DA parent). When the timer is armed, all entries in the Reachable list as well as all entries for Connected list are set to not be reported yet for that particular DA parent.
• The DelayDAO timer has a duration that is DEF_DAO_LATENCY divided by a multiple of the DAG rank of the node. The intention is that nodes located deeper in the DAG should have a shorter DelayDAO timer, allowing DAO messages a chance to be reported from deeper in the DAG and potentially aggregated along sub-DAGs before propagating further inwards.
• The RemoveTimer is used to clean up entries for which DAO messages are no longer being received from the sub-DAG.
• When a DIO message is sent that is requesting destination advertisements, a flag is set for all DAO entries in the routing table.
• If the flag has already been set for a DAO entry, the retry count is incremented.
• If a DAO message is received to confirm the entry, the entry is refreshed and the flag and count may be cleared.
• If at least one entry has reached a threshold value and the RemoveTimer is not running, the entry is considered to be probably gone and the RemoveTimer is started.
• When the RemoveTimer elapse, DAO messages with lifetime 0, i.e. no-DAOs, are sent to explicitly inform DA parents that the entries which have reached the threshold are no longer available, and the related routing states may be propagated and cleaned up.
• The RemoveTimer has a duration of min (MAX_DESTROY_INTERVAL, TBD(DIO Trickle Timer Interval)).

5.10.1.2.  Multicast Destination Advertisement Messages

It is also possible for a node to multicast a DAO message to the link-local scope all-nodes multicast address FF02::1. This message will be received by all node listening in range of the emitting node. The objective is to enable direct P2P communication, between destinations directly supported by neighboring nodes, without needing the RPL routing structure to relay the packets.

A multicast DAO message MUST be used only to advertise information about self, i.e. prefixes in the Connected list or addresses owned by this node. This would typically be a multicast group that this node is listening to or a global address owned by this node, though it can be used to advertise any prefix owned by this node as well. A multicast DAO message is not used for routing and does not presume any DAG relationship between the emitter and the receiver; it MUST NOT be used to relay information learned (e.g. information in the Reachable list) from another node; information obtained from a multicast DAO MAY be installed in the routing table and MAY be propagated by a router in unicast DAOs.

A node receiving a multicast DAO message addressed to FF02::1 MAY install prefixes contained in the DAO message in the routing table for local use. Such a node MUST NOT perform any other processing on the DAO message (i.e. such a node does not presume it is a DA parent).

5.10.1.3.  Unicast Destination Advertisement Messages from Child to Parent

When sending a destination advertisement to a DA parent, a node includes the DAOs for prefix entries not already reported (since the last DA Trigger from an DIO message) in the Reachable and Connected lists, as well as no-DAOs for all the entries in the Unreachable list. Depending on its policy and ability to retain routing state, the receiving node SHOULD keep a record of the reported DAO message. If the DAO message offers the best route to the prefix as determined by policy and other prefix records, the node SHOULD install a route to the prefix reported in the DAO message via the link local address of the reporting neighbor and it SHOULD further propagate the information in a DAO message.

The DIO message from the DAG root is used to synchronize the whole DAG, including the periodic reporting of destination advertisements back up the DAG. Its period is expected to vary, depending on the configuration of the trickle timer that governs the RAs.

When a node receives a DIO message over an LLN interface from a DA parent, the DelayDAO is armed to force a full update.

When the node broadcasts a DIO message on an LLN interface, for all entries on that interface:

• If the entry is CONFIRMED, it goes PENDING with the retry count set to 0.
• If the entry is PENDING, the retry count is incremented. If it reaches a maximum threshold, the entry goes ELAPSED If at least one entry is ELAPSED at the end of the process: if the RemoveTimer is not running then it is armed with a jitter.

Since the DelayDAO timer has a duration that decreases with the depth, it is expected to receive all DAO messages from all children before the timer elapses and the full update is sent to the DA parents.

Once the RemoveTimer is elapsed, the prefix entry is scheduled to be removed and moved to the Unreachable list if there are any DA parents that need to be informed of the change in status for the prefix, otherwise the prefix entry is cleaned up right away. The prefix entry is removed from the Unreachable list when no more DA parents need to be informed. This condition may be satisfied when a no-DAO is sent to all current DA parents indicating the loss of the prefix, and noting that in some cases parents may have been removed from the set of DA parents.

5.10.1.4.  Other Events

Finally, the destination advertisement mechanism responds to a series of events, such as:

• Destination advertisement operation stopped: All entries in the abstract lists are freed. All the routes learned from DAO messages are removed.
• Interface going down: for all entries in the Reachable list on that interface, the associated route is removed, and the entry is scheduled to be removed.
• Loss of routing adjacency: When the routing adjacency for a neighbor is lost, as per the procedures described in Section 5.13 (Maintenance of Routing Adjacency), and if the associated entries are in the Reachable list, the associated routes are removed, and the entries are scheduled to be destroyed.
• Changes to DA parent set: all entries in the Reachable list are set to not 'reported' and DelayDAO is armed.

5.10.1.5.  Aggregation of Prefixes by a Node

There may be number of cases where a aggregation may be shared within a group of nodes. In such a case, it is possible to use aggregation techniques with destination advertisements and improve scalability.

Other cases might occur for which additional support is required:

1. The aggregating node is attached within the sub-DAG of the nodes it is aggregating for.
2. A node that is to be aggregated for is located somewhere else within the DAG, not in the sub-DAG of the aggregating node.
3. A node that is to be aggregated for is located somewhere else in the LLN.

Consider a node M that is performing an aggregation, and a node N that is to be a member of the aggregation group. A node Z situated above the node M in the DAG, but not above node N, will see the advertisements for the aggregation owned by M but not that of the individual prefix for N. Such a node Z will route all the packets for node N towards node M, but node M will have no route to the node N and will fail to forward.

Additional protocols may be applied beyond the scope of this specification to dynamically elect/provision an aggregating node and groups of nodes eligible to be aggregated in order to provide route summarization for a sub-DAG.

5.11.  Loop Detection

RPL loop avoidance mechanisms are kept simple and designed to minimize churn and states. Loops may form for a number of reasons, from control packet loss to sibling forwarding. RPL includes a reactive loop detection technique that protects from meltdown and triggers repair of broken paths.

RPL loop detection uses information that is placed into the packet in the flow label. It assumes that the flow label may be overloaded for this purpose. The flow label is constructed 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
    +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
                            |O|S|R|D|  SenderRank   |  InstanceID   |
                            +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Outwards 'O' bit:

1-bit flag indicating whether the packet is expected to progress inwards or outwards. A router sets the 'O' bit when the packet is expect to progress outwards (using DAO routes), and resets it when forwarding towards the root of the DAG. A host MUST set the bit to 0.

Sibling 'S' bit:

1-bit flag indicating whether the packet has been forwarded via a sibling at the present rank, and denotes a risk of a sibling loop. A host sets the bit to 0.

Rank-Error 'R' bit:

1-bit flag indicating whether a rank error was detected. A rank error is detected when there is a mismatch in the relative ranks and the direction as indicated in the 'O' bit. A host MUST set the bit to 0.

DAO-Error 'D' bit:

1-bit flag indicating whether a DAO error was detected. An undetected DAO error would have resulted in an inward to outward transition that is not expected with this spec. A host MUST set the bit to 0.

SenderRank:

8-bit field indicating the rank of the sender. A host MUST set the rank to INFINITE_RANK. A router MUST place its own rank in the flow label when forwarding.

InstanceID:

8-bit field indicating the DAG instance along which the packet is sent.

5.11.1.  Host Basic Operation

It is expected that a host that does not participate to RPL in any fashion is configured to set the flow label to all zeroes in its outgoing packets. The host MAY send a packet to any router regardless of the DAG and RPL operations at large.

A host that participates to RPL SHOULD zero out all the flags, and it MUST set the sender rank to INFINITE_RANK. If the host can map a flow to a given InstanceID then it MUST set the flow label accordingly. Forwarding rules are the same for this host and a router, and are described in the next section.

5.11.2.  Instance Forwarding

Instance IDs is used to avoid loops between DAGs from different origins. DAGs that constructed for antagonistic constraints might contain paths that, if mixed together, would yield loops. Those loops are avoided by forwarding a packet along the DAG that is associated to a given instance.

The InstanceID is placed by the source in the flow label. It is not meaningful if the packet has the flow label set to all zeroes. Otherwise it MUST match the DAG instance onto which the packet is placed by any node, be it a host or router.

When a router receives a packet that is flagged with a given instance ID and the node can forward the packet along the DAG associated to that instance, then the router MUST do so and leave the instance ID flag unchanged.

If any node can not forward a packet along the DAG associated to the instance ID in the flow label, then the node MAY either change the InstanceID to match a DAG that it is using for this packet or discard the packet. That decision is based on a policy.

The default policy is as follows: if the node can forward along the DAG associated to the instance RPL_DEFAULT_INSTANCE then it should do so. Otherwise it should drop the packet.

5.11.3.  DAG Inconsistency Loop Detection

The DAG is inconsistent is the direction of a packet does not match the rank relationship. A receiver detects an inconsistency if it receives a packet with either:

the 'O' bit set (to outwards) from a node of a higher rank.

the 'O' bit reset (for inwards) from a node of a lesser rank.

the 'S' bit set (to sibling) from a node of a different rank.

The propagation of a new sequence creates local inconsistencies. In particular, it is possible for a router to forward a packet to a future parent (same instance, same DAGID, higher sequence) without a loop, regardless of the rank of that parent. In that case, the sending router MUST present itself as a host on the future DAG and use a rank of INFINITE_RANK as it forwards the packets via a future parent to avoid a false positive.

One inconsistency along the path is not considered as a critical error and the packet may continue. But a second detection along the path of a same packet should not occur and the packet is dropped.

This process is controlled by the Rank-Error bit in the Flow Label. When an inconsistency, is detected on a packet, if the Rank-Error bit was not set then the Rank-Error bit is set. If it was set the packet is discarded and the trickle timer is reset.

5.11.4.  Sibling Loop Avoidance

When a packet is forwarded along siblings, it cannot be checked for forward progress and may loop between siblings. Experimental evidence has shown that one sibling hop can be very useful but is generally sufficient to avoid loops. Based on that evidence, this specification enforces the simple rule that a packet may not make 2 sibling hops in a row.

When a host issues a packet or when a router forwards a packet to a non sibling, the Sibling bit in the packet must be reset. When a router forwards to a sibling: if the Sibling bit was not set then the Sibling bit is set. If the Sibling bit was set then the packet is discarded. This does not denote a graph inconsistency but indicates that a new graph should probably be formed with a new sequence.

5.11.5.  DAO Inconsistency Loop Detection and Recovery

A DAO inconsistency happens when router that has an outwards DAO route via a child that is a remnant from an obsolete state that is not matched in the child. With DAO inconsistency loop recovery, a packet can be used to recursively explore and cleanup the obsolete DAO states along a sub-DAG.

In a general manner, a packet that goes outwards should never go inwards again. So rather than routing inwards a packet with the Outwards bit set, the router MUST discard the packet. If DAO inconsistency loop recovery is applied, then the router SHOULD send the packet to the parent that passed it with the DAO-Error bit set.

Upon a packet with a DAO bit set, the parent MUST remove the routing states that caused forwarding to that child, clear DAO-Error bit and send the packet again. The packet will make its way either to an alternate child or inwards to a parent. If that parent still has an inconsistent DAO state via self, the process will recurse and that state will be cleaned up as well.

5.12.  Multicast Operation

This section describes further the multicast routing operations over an IPv6 RPL network, and specifically how unicast DAOs can be used to relay group registrations inwards. Wherever the following text mentions MLD, one can read MLDv2 or v3.

As is traditional, a listener uses a protocol such as MLD with a router to register to a multicast group.

Along the path between the router and the root of the DAG, MLD requests are mapped and transported as DAO messages within the RPL protocol; each hop coalesces the multiple requests for a same group as a single DAO message to the parent(s), in a fashion similar to proxy IGMP, but recursively between child router and parent up to the root.

A router might select to pass a listener registration DAO message to its preferred parent only, in which case multicast packets coming back might be lost for all of its sub-DAG if the transmission fails over that link. Alternatively the router might select to copy additional parents as it would do for DAO messages advertising unicast destinations, in which case there might be duplicates that the router will need to prune.

As a result, multicast routing states are installed in each router on the way from the listeners to the root, enabling the root to copy a multicast packet to all its children routers that had issued a DAO message including a DAO for that multicast group, as well as all the attached nodes that registered over MLD.

For unicast traffic, it is expected that the grounded root of an RPL DAG terminates RPL and MAY redistribute the RPL routes over the external infrastructure using whatever routing protocol is used there. For multicast traffic, the root MAY proxy MLD for all the nodes attached to the RPL routers (this would be needed if the multicast source is located in the external infrastructure). For such a source, the packet will be replicated as it flows outwards along the DAG based on the multicast routing table entries installed from the DAO message.

For a source inside the DAG, the packet is passed to the preferred parents, and if that fails then to the alternates in the DAG. The packet is also copied to all the registered children, except for the one that passed the packet. Finally, if there is a listener in the external infrastructure then the DAG root has to further propagate the packet into the external infrastructure.

As a result, the DAG Root acts as an automatic proxy Rendezvous Point for the RPL network, and as source towards the Internet for all multicast flows started in the RPL LLN. So regardless of whether the root is actually attached to the Internet, and regardless of whether the DAG is grounded or floating, the root can serve inner multicast streams at all times.

5.13.  Maintenance of Routing Adjacency

The selection of successors, along the default paths inward along the DAG, or along the paths learned from destination advertisements outward along the DAG, leads to the formation of routing adjacencies that require maintenance.

In IGPs such as OSPF [RFC4915] (Psenak, P., Mirtorabi, S., Roy, A., Nguyen, L., and P. Pillay-Esnault, “Multi-Topology (MT) Routing in OSPF,” June 2007.) or IS-IS [RFC5120] (Przygienda, T., Shen, N., and N. Sheth, “M-ISIS: Multi Topology (MT) Routing in Intermediate System to Intermediate Systems (IS-ISs),” February 2008.), the maintenance of a routing adjacency involves the use of Keepalive mechanisms (Hellos) or other protocols such as BFD ([I‑D.ietf‑bfd‑base] (Katz, D. and D. Ward, “Bidirectional Forwarding Detection,” February 2009.)) and MANET Neighborhood Discovery Protocol (NHDP [I‑D.ietf‑manet‑nhdp] (Clausen, T., Dearlove, C., and J. Dean, “MANET Neighborhood Discovery Protocol (NHDP),” July 2009.)). Unfortunately, such an approach is not desirable in constrained environments such as LLN and would lead to excessive control traffic in light of the data traffic with a negative impact on both link loads and nodes resources. Overhead to maintain the routing adjacency should be minimized. Furthermore, it is not always possible to rely on the link or transport layer to provide information of the associated link state. The network layer needs to fall back on its own mechanism.

Thus RPL makes use of a different approach consisting of probing the neighbor using a Neighbor Solicitation message (see [RFC4861] (Narten, T., Nordmark, E., Simpson, W., and H. Soliman, “Neighbor Discovery for IP version 6 (IPv6),” September 2007.)). The reception of a Neighbor Advertisement (NA) message with the "Solicited Flag" set is used to verify the validity of the routing adjacency. Such mechanism MAY be used prior to sending a data packet. This allows for detecting whether or not the routing adjacency is still valid, and should it not be the case, select another feasible successor to forward the packet.

5.14.  Packet Forwarding

When forwarding a packet to a destination, precedence is given to selection of a next-hop successor as follows:

1. In the scope of this specification, it is preferred to select a successor from a DAG that matches the InstanceID marked in the IPv6 header of the packet being forwarded.
2. If a local administrative preference favors a route that has been learned from a different routing protocol than RPL, then use that successor.
3. If there is an entry in the routing table matching the destination that has been learned from a multicast destination advertisement (e.g. the destination is a one-hop neighbor), then use that successor.
4. If there is an entry in the routing table matching the destination that has been learned from a unicast destination advertisement (e.g. the destination is located outwards along the sub-DAG), then use that successor.
5. If there is a DAG offering a route to a prefix matching the destination, then select one of those DAG parents as a successor.
6. If there is a DAG parent offering a default route then select that DAG parent as a successor.
7. If there is a DAG offering a route to a prefix matching the destination, but all DAG parents have been tried and are temporarily unavailable (as determined by the forwarding procedure), then select a DAG sibling as a successor.
8. Finally, if no DAG siblings are available, the packet is dropped. ICMP Destination Unreachable may be invoked. An inconsistency is detected.

TTL MUST be decremented when forwarding. If the packet is being forwarded via a sibling, then the TTL MAY be decremented more aggressively (by more than one) to limit the impact of possible loops.

Note that the chosen successor MUST NOT be the neighbor that was the predecessor of the packet (split horizon), except in the case where it is intended for the packet to change from an inward to an outward flow, such as switching from DIO routes to DAO routes as the destination is neared.

6.  RPL Constants and Variables

ZERO_LIFETIME

This is the special value of a lifetime that indicates immediate death and removal. ZERO_LIFETIME has a value of 0.

BASE_RANK

This is the rank for a virtual root that might be used to coordinate multiple roots. BASE_RANK has a value of 0.

ROOT_RANK

This is the rank for a DAG root. ROOT_RANK has a value of 1.

INFINITE_RANK

This is the constant maximum for the rank. INFINITE_RANK has a value of 0xFF.

RPL_DEFAULT_INSTANCE

This is the instance ID that is used by this protocol by a node without a policy to know any better. RPL_DEFAULT_INSTANCE has a value of 0.

DEFAULT_DIO_INTERVAL_MIN

To be determined

DEFAULT_DIO_INTERVAL_DOUBLINGS

To be determined

DEF_DAO_LATENCY

To be determined

MAX_DESTROY_INTERVAL

To be determined

DIO Timer

One instance per DAG that a node is a member of. Expiry triggers DIO message transmission. Trickle timer with variable interval in [0, DIOIntervalMin..2^DIOIntervalDoublings]. See Section 5.4.4 (Trickle Timer for DIO Transmission)

DAG Sequence Number Increment Timer

Up to one instance per DAG that the node is acting as DAG root of. May not be supported in all implementations. Expiry triggers revision of DAGSequenceNumber, causing a new series of updated DIO message to be sent. Interval should be chosen appropriate to propagation time of DAG and as appropriate to application requirements (e.g. response time vs. overhead). See Section 5.5 (DAG Sequence Number Increment)

DelayDAO Timer

Up to one instance per DA parent (the subset of DAG parents chosen to receive destination advertisements) per DAG. Expiry triggers sending of DAO message to the DA parent. The interval is to be proportional to DEF_DAO_LATENCY/(node rank), such that nodes of greater rank (further outward along the DAG) expire first, coordinating the sending of DAO messages to allow for a chance of aggregation. See Section 5.10.1.1.1 (Destination Advertisement Timers)

RemoveTimer

Up to one instance per DA entry per neighbor (i.e. those neighbors that have given DAO messages to this node as a DAG parent) Expiry triggers a change in state for the DA entry, setting up to do unreachable (No-DAO) advertisements or immediately deallocating the DA entry if there are no DA parents. The interval is min(MAX_DESTROY_INTERVAL, TBD(DIO Trickle Timer Interval)). See Section 5.10.1.1.1 (Destination Advertisement Timers)

7.  Manageability Considerations

The aim of this section is to give consideration to the manageability of RPL, and how RPL will be operated in LLN beyond the use of a MIB module. The scope of this section is to consider the following aspects of manageability: fault management, configuration, accounting and performance.

7.1.  Control of Function and Policy

7.1.1.  Initialization Mode

When a node is first powered up, it may either choose to stay silent and not send any multicast DIO message until it has joined a DAG, or to immediately root a transient DAG and start sending multicast DIO messages. A RPL implementation SHOULD allow configuring whether the node should stay silent or should start advertising DIO messages.

Furthermore, the implementation SHOULD to allow configuring whether or not the node should start sending an DIS message as an initial probe for nearby DAGs, or should simply wait until it received RA messages from other nodes that are part of existing DAGs.

7.1.2.  DIO Base option

RPL specifies a number of protocol parameters.

A RPL implementation SHOULD allow configuring the following routing protocol parameters, which are further described in Section 5.1.3.1 (DIO Base Option):

DAGPreference

InstanceID

DAGObjectiveCodePoint

DAGID

Destination Prefixes

DIOIntervalDoublings

DIOIntervalMin

DAG Root behavior:

In some cases, a node may not want to permanently act as a DAG root if it cannot join a grounded DAG. For example a battery-operated node may not want to act as a DAG root for a long period of time. Thus a RPL implementation MAY support the ability to configure whether or not a node could act as a DAG root for a configured period of time.

DAG Table Entry Suppression

A RPL implementation SHOULD provide the ability to configure a timer after the expiration of which the DAG table that contains all the records about a DAG is suppressed, to be invoked if the DAG parent set becomes empty.

7.1.3.  Trickle Timers

A RPL implementation makes use of trickle timer to govern the sending of DIO message. Such an algorithm is determined a by a set of configurable parameters that are then advertised by the DAG root along the DAG in DIO messages.

For each DAG, a RPL implementation MUST allow for the monitoring of the following parameters, further described in Section 5.4.4 (Trickle Timer for DIO Transmission):

I

T

C

I_min

I_doublings:

A RPL implementation SHOULD provide a command (for example via API, CLI, or SNMP MIB) whereby any procedure that detects an inconsistency may cause the trickle timer to reset.

7.1.4.  DAG Sequence Number Increment

A RPL implementation may allow by configuration at the DAG root to refresh the DAG states by updating the DAGSequenceNumber. A RPL implementation SHOULD allow configuring whether or not periodic or event triggered mechanism are used by the DAG root to control DAGSequenceNumber change.

7.1.5.  Destination Advertisement Timers

The following set of parameters of the DAO messages SHOULD be configurable:

• The DelayDAO timer
• The Remove timer

7.1.6.  Policy Control

DAG discovery enables nodes to implement different policies for selecting their DAG parents.

A RPL implementation SHOULD allow configuring the set of acceptable or preferred Objective Functions (OF) referenced by their Objective Codepoints (OCPs) for a node to join a DAG, and what action should be taken if none of a node's candidate neighbors advertise one of the configured allowable Objective Functions.

A node in an LLN may learn routing information from different routing protocols including RPL. It is in this case desirable to control via administrative preference which route should be favored. An implementation SHOULD allow for specifying an administrative preference for the routing protocol from which the route was learned.

A RPL implementation SHOULD allow for the configuration of the "Route Tag" field of the DAO messages according to a set of rules defined by policy.

7.1.7.  Data Structures

Some RPL implementation may limit the size of the candidate neighbor list in order to bound the memory usage, in which case some otherwise viable candidate neighbors may not be considered and simply dropped from the candidate neighbor list.

A RPL implementation MAY provide an indicator on the size of the candidate neighbor list.

7.2.  Information and Data Models

The information and data models necessary for the operation of RPL will be defined in a separate document specifying the RPL SNMP MIB.

7.3.  Liveness Detection and Monitoring

The aim of this section is to describe the various RPL mechanisms specified to monitor the protocol.

As specified in Section 5.2 (Conceptual Data Structures), an implementation must maintain a set of data structures in support of DAG discovery:

• The candidate neighbors data structure
• For each DAG:
• A set of DAG parents

7.3.1.  Candidate Neighbor Data Structure

A node in the candidate neighbor list is a node discovered by the some means and qualified to potentially become of neighbor or a sibling (with high enough local confidence). A RPL implementation SHOULD provide a way monitor the candidate neighbors list with some metric reflecting local confidence (the degree of stability of the neighbors) measured by some metrics.

A RPL implementation MAY provide a counter reporting the number of times a candidate neighbor has been ignored, should the number of candidate neighbors exceeds the maximum authorized value.

7.3.2.  Directed Acyclic Graph (DAG) Table

For each DAG, a RPL implementation MUST keep track of the following DAG table values:

• DAGID
• DAGObjectiveCodePoint
• A set of Destination Prefixes offered inwards along the DAG
• A set of DAG Parents
• timer to govern the sending of DIO messages for the DAG
• DAGSequenceNumber

The set of DAG parents structure is itself a table with the following entries:

• A reference to the neighboring device which is the DAG parent
• A record of most recent information taken from the DAG Information Object last processed from the DAG Parent
• A flag reporting if the Parent is a DA Parent as described in Section 5.10 (Establishing Routing State Outward Along the DAG)

7.3.3.  Routing Table

For each route provisioned by RPL operation, a RPL implementation MUST keep track of the following:

• Destination Prefix
• Destination Prefix Length
• Lifetime Timer
• Next Hop
• Next Hop Interface
• Flag indicating that the route was provisioned from one of:
  • Unicast DAO message
  • DIO message
  • Multicast DAO message

7.3.4.  Other RPL Monitoring Parameters

A RPL implementation SHOULD provide a counter reporting the number of a times the node has detected an inconsistency with respect to a DAG parent, e.g. if the DAGID has changed.

A RPL implementation MAY log the reception of a malformed DIO message along with the neighbor identification if avialable.

7.3.5.  RPL Trickle Timers

A RPL implementation operating on a DAG root MUST allow for the configuration of the following trickle parameters:

• The DIOIntervalMin expressed in ms
• The DIOIntervalDoublings

A RPL implementation MAY provide a counter reporting the number of times an inconsistency (and thus the trickle timer has been reset).

7.4.  Verifying Correct Operation

This section has to be completed in further revision of this document to list potential Operations and Management (OAM) tools that could be used for verifying the correct operation of RPL.

7.5.  Requirements on Other Protocols and Functional Components

RPL does not have any impact on the operation of existing protocols.

7.6.  Impact on Network Operation

To be completed.

8.  Security Considerations

Security Considerations for RPL are to be developed in accordance with recommendations laid out in, for example, [I‑D.tsao‑roll‑security‑framework] (Tsao, T., Alexander, R., Daza, V., and A. Lozano, “A Security Framework for Routing over Low Power and Lossy Networks,” March 2010.).

9.  IANA Considerations

9.1.  RPL Control Message

The RPL Control Message is an ICMP information message type that is to be used carry DAG Information Objects, DAG Information Solicitations, and Destination Advertisement Objects in support of RPL operation.

IANA has defined a ICMPv6 Type Number Registry. The suggested type value for the RPL Control Message is 155, to be confirmed by IANA.

9.2.  New Registry for RPL Control Codes

IANA is requested to create a registry, RPL Control Codes, for the Code field of the ICMPv6 RPL Control Message.

New codes may be allocated only by an IETF Consensus action. Each code should be tracked with the following qualities:

• Code
• Description
• Defining RFC

Three codes are currently defined:

9.3.  New Registry for the Control Field of the DIO Base Option

IANA is requested to create a registry for the Control field of the DIO Base Option.

New bit numbers may be allocated only by an IETF Consensus action. Each bit should be tracked with the following qualities:

• Bit number (counting from bit 0 as the most significant bit)
• Capability description
• Defining RFC

Four groups are currently defined:

9.4.  DAG Information Object (DIO) Suboption

IANA is requested to create a registry for the DIO Base Option Suboptions

9.5.  Objective Code Point for the Default Objective Function OF0

This specification specifies the Default Objective Function (called OF0) for which the OCP field of the OF object, as defined in [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.), is equal to 0x0000

10.  Acknowledgements

The authors would like to acknowledge the review, feedback, and comments from Emmanuel Baccelli, Dominique Barthel, Yusuf Bashir, Mathilde Durvy, Manhar Goindi, Mukul Goyal, Anders Jagd, Quentin Lampin, Jerry Martocci, Alexandru Petrescu, and Don Sturek.

The authors would like to acknowledge the guidance and input provided by the ROLL Chairs, David Culler and JP Vasseur.

The authors would like to acknowledge prior contributions of Robert Assimiti, Mischa Dohler, Julien Abeille, Ryuji Wakikawa, Teco Boot, Patrick Wetterwald, Bryan Mclaughlin, Carlos J. Bernardos, Thomas Watteyne, Zach Shelby, Caroline Bontoux, Marco Molteni, Billy Moon, and Arsalan Tavakoli, which have provided useful design considerations to RPL.

11.  Contributors

RPL is the result of the contribution of the following members of the ROLL Design Team, including the editors, and additional contributors as listed below:

JP Vasseur
Cisco Systems, Inc
11, Rue Camille Desmoulins
Issy Les Moulineaux,   92782
France

Email: jpv@cisco.com


Jonathan W. Hui
Arch Rock Corporation
501 2nd St. Ste. 410
San Francisco, CA  94107
USA

Email: jhui@archrock.com


Thomas Heide Clausen
LIX, Ecole Polytechnique, France

Phone: +33 6 6058 9349
EMail: T.Clausen@computer.org
URI:   http://www.ThomasClausen.org/


Richard Kelsey
Ember Corporation
Boston, MA
USA

Phone: +1 617 951 1225
Email: kelsey@ember.com


Philip Levis
Stanford University
358 Gates Hall, Stanford University
Stanford, CA  94305-9030
USA

Email: pal@cs.stanford.edu


Stephen Dawson-Haggerty
UC Berkeley
Soda Hall, UC Berkeley
Berkeley, CA  94720
USA

Email: stevedh@cs.berkeley.edu


Kris Pister
Dust Networks
30695 Huntwood Ave.
Hayward,   94544
USA

Email: kpister@dustnetworks.com


Anders Brandt
Zensys, Inc.
Emdrupvej 26
Copenhagen, DK-2100
Denmark

Email: abr@zen-sys.com

12.  References

12.1. Normative References

12.2. Informative References

Appendix A.  Requirements

A.1.  Protocol Properties Overview

RPL demonstrates the following properties, consistent with the requirements specified by the application-specific requirements documents.

A.1.1.  IPv6 Architecture

RPL is strictly compliant with layered IPv6 architecture.

Further, RPL is designed with consideration to the practical support and implementation of IPv6 architecture on devices which may operate under severe resource constraints, including but not limited to memory, processing power, energy, and communication. The RPL design does not presume high quality reliable links, and operates over lossy links (usually low bandwidth with low packet delivery success rate).

A.1.2.  Typical LLN Traffic Patterns

Multipoint-to-Point (MP2P) and Point-to-multipoint (P2MP) traffic flows from nodes within the LLN from and to egress points are very common in LLNs. Low power and lossy network Border Router (LBR) nodes may typically be at the root of such flows, although such flows are not exclusively rooted at LBRs as determined on an application-specific basis. In particular, several applications such as building or home automation do require P2P (Point-to-Point) communication.

As required by the aforementioned routing requirements documents, RPL supports the installation of multiple paths. The use of multiple paths include sending duplicated traffic along diverse paths, as well as to support advanced features such as Class of Service (CoS) based routing, or simple load balancing among a set of paths (which could be useful for the LLN to spread traffic load and avoid fast energy depletion on some, e.g. battery powered, nodes). Conceptually, multiple instances of RPL can be used to send traffic along different topology instances, the construction of which is governed by different Objective Functions (OF). Details of RPL operation in support of multiple instances are beyond the scope of the present specification.

A.1.3.  Constraint Based Routing

The RPL design supports constraint based routing, based on a set of routing metrics and constraints. The routing metrics and constraints for links and nodes with capabilities supported by RPL are specified in a companion document to this specification, [I‑D.ietf‑roll‑routing‑metrics] (Vasseur, J., Kim, M., Networks, D., and H. Chong, “Routing Metrics used for Path Calculation in Low Power and Lossy Networks,” April 2010.). RPL signals the metrics, constraints, and related Objective Functions (OFs) in use in a particular implementation by means of an Objective Code Point (OCP). Both the routing metrics, constraints, and the OF help determine the construction of the Directed Acyclic Graphs (DAG) using a distributed path computation algorithm.

A.2.  Deferred Requirements

NOTE: RPL is still a work in progress. At this time there remain several unsatisfied application requirements, but these are to be addressed as RPL is further specified.

Appendix B.  Examples

Consider the example LLN physical topology in Figure 11 (Example LLN Topology). In this example the links depicted are all usable L2 links. Suppose that all links are equally usable, and that the implementation specific policy function is simply to minimize hops. This LLN physical topology then yields the DAG depicted in Figure 12 (Example DAG), where the links depicted are the edges toward DAG parents. This topology includes one DAG, rooted by an LBR node (LBR) at rank 1. The LBR node will issue DIO messages, as governed by a trickle timer. Nodes (11), (12), (13), have selected (LBR) as their only parent, attached to the DAG at rank 2, and periodically multicast DIOs. Node (22) has selected (11) and (12) in its DAG parent set, and advertises itself at rank 3. Node (22) thus has a set of DAG parents {(11), (12)} and siblings {((21), (23)}.

                                  (LBR)
                                  / | \
                             .---`  |  `----.
                            /       |        \
                         (11)------(12)------(13)
                          | \       | \       | \
                          |  `----. |  `----. |  `----.
                          |        \|        \|        \
                         (21)------(22)------(23)      (24)
                          |        /|        /|         |
                          |  .----` |  .----` |         |
                          | /       | /       |         |
                         (31)------(32)------(33)------(34)
                          |        /| \       | \       | \
                          |  .----` |  `----. |  `----. |  `----.
                          | /       |        \|        \|        \
                .--------(41)      (42)      (43)------(44)------(45)
               /         /         /| \       | \
         .----`    .----`    .----` |  `----. |  `----.
        /         /         /       |        \|        \
     (51)------(52)------(53)------(54)------(55)------(56)

Note that the links depicted represent the usable L2 connectivity available in the LLN. For example, Node (31) can communicate directly with its neighbors, Nodes (21), (22), (32), and (41). Node (31) cannot communicate directly with any other nodes, e.g. (33), (23), (42). In this example these links offer bidirectional communication, and `bad' links are not depicted.

                                  (LBR)
                                  / | \
                             .---`  |  `----.
                            /       |        \
                         (11)      (12)      (13)
                          | \       | \       | \
                          |  `----. |  `----. |  `----.
                          |        \|        \|        \
                         (21)      (22)      (23)      (24)
                          |        /|        /|         |
                          |  .----` |  .----` |         |
                          | /       | /       |         |
                         (31)      (32)      (33)      (34)
                          |        /| \       | \       | \
                          |  .----` |  `----. |  `----. |  `----.
                          | /       |        \|        \|        \
                .--------(41)      (42)      (43)      (44)      (45)
               /         /         /| \       | \
         .----`    .----`    .----` |  `----. |  `----.
        /         /         /       |        \|        \
     (51)      (52)      (53)      (54)      (55)      (56)

Note that the links depicted represent directed links in the DAG overlaid on top of the physical topology depicted in Figure 11 (Example LLN Topology). As such, the depicted edges represent the relationship between nodes and their DAG parents, wherein all depicted edges are directed and oriented `up' on the page toward the DAG root (LBR). The DAG may provide default routes within the LLN, and serves as the foundation on which RPL builds further routing structure, e.g. through the destination advertisement mechanism.

B.1.  Destination Advertisement

Consider the example DAG depicted in Figure 12 (Example DAG). Suppose that Nodes (22) and (32) are unable to record routing state. Suppose that Node (42) is able to perform prefix aggregation on behalf of Nodes (53), (54), and (55).

• Node (53) would send a DAO message to Node (42), indicating the availability of destination (53).
• Node (54) and Node (55) would similarly send DAO messages to Node (42) indicating their own destinations.
• Node (42) would collect and store the routing state for destinations (53), (54), and (55).
• In this example, Node (42) may then be capable of representing destinations (42), (53), (54), and (55) in the aggregation (42').
• Node (42) sends a DAO message advertising destination (42') to Node 32.
• Node (32) does not want to maintain any routing state, so it adds onto to the Reverse Route Stack in the DAO message and passes it on to Node (22) as (42'):[(42)]. It may send a separate DAO message to indicate destination (32).
• Node (22) does not want to maintain any routing state, so it adds on to the Reverse Route Stack in the DAO message and passes it on to Node (12) as (42'):[(42), (32)]. It also relays the DAO message containing destination (32) to Node 12 as (32):[(32)], and finally may send a DAO message for itself indicating destination (22).
• Node (12) is capable to maintain routing state again, and receives the DAO messages from Node (22). Node (12) then learns:
• Destination (22) is available via Node (22)
• Destination (32) is available via Node (22) and the piecewise source route to (32)
• Destination (42') is available via Node (22) and the piecewise source route to (32), (42').
• Node (12) sends DAO messages to (LBR), allowing (LBR) to learn routes to the destinations (12), (22), (32), and (42'). (42), (53), (54), and (55) are available via the aggregation (42'). It is not necessary for Node (12) to propagate the piecewise source routes to (LBR).

B.2.  Example: DAG Parent Selection

For example, suppose that a node (N) is not attached to any DAG, and that it is in range of nodes (A), (B), (C), (D), and (E). Let all nodes be configured to use an OCP which defines a policy such that ETX is to be minimized and paths with the attribute `Blue' should be avoided. Let the rank computation indicated by the OCP simply reflect the ETX aggregated along the path. Let the links between node (N) and its neighbors (A-E) all have an ETX of 1 (which is learned by node (N) through some implementation specific method). Let node (N) be configured to send RPL DIS messages to probe for nearby DAGs.

• Node (N) transmits a RPL DIS message.
• Node (B) responds. Node (N) investigates the DIO message, and learns that Node (B) is a member of DAGID 1 at rank 4, and not `Blue'. Node (N) takes note of this, but is not yet confident.
• Similarly, Node (N) hears from Node (A) at rank 9, Node (C) at rank 5, and Node (E) at rank 4.
• Node (D) responds. Node (D) has a DIO message that indicates that it is a member of DAGID 1 at rank 2, but it carries the attribute `Blue'. Node (N)'s policy function rejects Node (D), and no further consideration is given.
• This process continues until Node (N), based on implementation specific policy, builds up enough confidence to trigger a decision to join DAGID 1. Let Node (N) determine its most preferred parent to be Node (E).
• Node (N) adds Node (E) (rank 4) to its set of DAG parents for DAGID 1. Following the mechanisms specified by the OCP, and given that the ETX is 1 for the link between (N) and (E), Node (N) is now at rank 5 in DAGID 1.
• Node (N) adds Node (B) (rank 4) to its set of DAG parents for DAGID 1.
• Node (N) is a sibling of Node (C), both are at rank 5.
• Node (N) may now forward traffic intended for the default destination inward along DAGID 1 via nodes (B) and (E). In some cases, e.g. if nodes (B) and (E) are tried and fail, node (N) may also choose to forward traffic to its sibling node (C), without making inward progress but with the intention that node (C) or a following successor can make inward progress. Should Node (C) not have a viable parent, it should never send the packet back to Node (N) (to avoid a 2-node loop).

B.3.  Example: DAG Maintenance

       :                      :                      :
       :                      :                      :
      (A)                    (A)                    (A)
       |\                     |                      |
       | `-----.              |                      |
       |        \             |                      |
      (B)       (C)          (B)       (C)          (B)
                 |                      |             \
                 |                      |              `-----.
                 |                      |                     \
                (D)                    (D)                    (C)
                                                               |
                                                               |
                                                               |
                                                              (D)

           -1-                    -2-                    -3-

Consider the example depicted in Figure 13 (DAG Maintenance)-1. In this example, Node (A) is attached to a DAG at some rank d. Node (A) is a DAG parent of Nodes (B) and (C). Node (C) is a DAG parent of Node (D). There is also an undirected sibling link between Nodes (B) and (C).

In this example, Node (C) may safely forward to Node (A) without creating a loop. Node (C) may not safely forward to Node (D), contained within it's own sub-DAG, without creating a loop. Node (C) may forward to Node (B) in some cases, e.g. the link (C)->(A) is temporarily unavailable, but with some chance of creating a loop (e.g. if multiple nodes in a set of siblings start forwarding `sideways' in a cycle) and requiring the intervention of additional mechanisms to detect and break the loop.

Consider the case where Node (C) hears a DIO message from a Node (Z) at a lesser rank and superior position in the DAG than node (A). Node (C) may safely undergo the process to evict node (A) from its DAG parent set and attach directly to Node (Z) without creating a loop, because its rank will decrease.

Now consider the case where the link (C)->(A) becomes nonviable, and node (C) must move to a deeper rank within the DAG:

• Node (C) must first detach from the DAG by removing Node (A) from its DAG parent set, leaving an empty DAG parent set. Node (C) may become the root of its own floating, less preferred, DAG.
• Node (D), hearing a modified DIO message from Node (C), follows Node (C) into the floating DAG. This is depicted in Figure 13 (DAG Maintenance)-2. In general, any node with no other options in the sub-DAG of Node (C) will follow Node (C) into the floating DAG, maintaining the structure of the sub-DAG.
• Node (C) hears a DIO message with an incremented DAGSequenceNumber from Node (B) and determines it is able to rejoin the grounded DAG by reattaching at a deeper rank to Node (B). Node (C) adds Node (B) to its DAG parent set. Node (C) has now safely moved deeper within the grounded DAG without creating any loops.
• Node (D), and any other sub-DAG of Node (C), will hear the modified DIO message sourced from Node (C) and follow Node (C) in a coordinated manner to reattach to the grounded DAG. The final DAG is depicted in Figure 13 (DAG Maintenance)-3

B.4.  Example: Greedy Parent Selection and Instability

      (A)                    (A)                    (A)
       |\                     |\                     |\
       | `-----.              | `-----.              | `-----.
       |        \             |        \             |        \
      (B)       (C)          (B)        \            |        (C)
                               \        |            |        /
                                `-----. |            | .-----`
                                       \|            |/
                                       (C)          (B)

           -1-                    -2-                    -3-

Consider the example depicted in Figure 14 (Greedy DAG Parent Selection). A DAG is depicted in 3 different configurations. A usable link between (B) and (C) exists in all 3 configurations. In Figure 14 (Greedy DAG Parent Selection)-1, Node (A) is a DAG parent for Nodes (B) and (C), and (B)--(C) is a sibling link. In Figure 14 (Greedy DAG Parent Selection)-2, Node (A) is a DAG parent for Nodes (B) and (C), and Node (B) is also a DAG parent for Node (C). In Figure 14 (Greedy DAG Parent Selection)-3, Node (A) is a DAG parent for Nodes (B) and (C), and Node (C) is also a DAG parent for Node (B).

If a RPL node is too greedy, in that it attempts to optimize for an additional number of parents beyond its preferred parent, then an instability can result. Consider the DAG illustrated in Figure 14 (Greedy DAG Parent Selection)-1. In this example, Nodes (B) and (C) may most prefer Node (A) as a DAG parent, but are operating under the greedy condition that will try to optimize for 2 parents.

When the preferred parent selection causes a node to have only one parent and no siblings, the node may decide to insert itself at a slightly higher rank in order to have at least one sibling and thus an alternate forwarding solution. This does not deprive other nodes of a forwarding solution and this is considered acceptable greediness.

• Let Figure 14 (Greedy DAG Parent Selection)-1 be the initial condition.
• Suppose Node (C) first is able to leave the DAG and rejoin at a lower rank, taking both Nodes (A) and (B) as DAG parents as depicted in Figure 14 (Greedy DAG Parent Selection)-2. Now Node (C) is deeper than both Nodes (A) and (B), and Node (C) is satisfied to have 2 DAG parents.
• Suppose Node (B), in its greediness, is willing to receive and process a DIO message from Node (C) (against the rules of RPL), and then Node (B) leaves the DAG and rejoins at a lower rank, taking both Nodes (A) and (C) as DAG parents. Now Node (B) is deeper than both Nodes (A) and (C) and is satisfied with 2 DAG parents.
• Then Node (C), because it is also greedy, will leave and rejoin deeper, to again get 2 parents and have a lower rank then both of them.
• Next Node (B) will again leave and rejoin deeper, to again get 2 parents
• And again Node (C) leaves and rejoins deeper...
• The process will repeat, and the DAG will oscillate between Figure 14 (Greedy DAG Parent Selection)-2 and Figure 14 (Greedy DAG Parent Selection)-3 until the nodes count to infinity and restart the cycle again.
• This cycle can be averted through mechanisms in RPL:
  • Nodes (B) and (C) stay at a rank sufficient to attach to their most preferred parent (A) and don't go for any deeper (worse) alternate parents (Nodes are not greedy)
  • Nodes (B) and (C) do not process DIO messages from nodes deeper than themselves (because such nodes are possibly in their own sub-DAGs)

Appendix C.  Outstanding Issues

This section enumerates some outstanding issues that are to be addressed in future revisions of the RPL specification.

C.1.  Additional Support for P2P Routing

In some situations the baseline mechanism to support arbitrary P2P traffic, by flowing inward along the DAG until a common parent is reached and then flowing outward, may not be suitable for all application scenarios. A related scenario may occur when the outward paths setup along the DAG by the destination advertisement mechanism are not be the most desirable outward paths for the specific application scenario (in part because the DAG links may not be symmetric). It may be desired to support within RPL the discovery and installation of more direct routes `across' the DAG. Such mechanisms need to be investigated.

C.2.  Loop Detection

It is under investigation to complement the loop avoidance strategies provided by RPL with a loop detection mechanism that may be employed when traffic is forwarded.

C.3.  Destination Advertisement / DAO Fan-out

When DAO messages are relayed to more than one DAG parent, in some cases a situation may be created where a large number of DAO messages conveying information about the same destination flow inward along the DAG. It is desirable to bound/limit the multiplication/fan-out of DAO messages in this manner. Some aspects of the Destination Advertisement mechanism remain under investigation, such as behavior in the face of links that may not be symmetric.

In general, the utility of providing redundancy along outwards routes by sending DAO messages to more than one parent is under investigation.

The use of suitable triggers, such as the `D' bit, to trigger DA operation within an affected sub-DAG, is under investigation. Further, the ability to limit scope of the affected depth within the sub-DAG is under investigation (e.g. if a stateful node can proxy for all nodes `behind' it, then there may be no need to propagate the triggered `D' bit further).

C.4.  Source Routing

In support of nodes that maintain minimal routing state, and to make use of the collection of piecewise source routes from the destination advertisement mechanism, there needs to be some investigation of a mechanism to specify, attach, and follow source routes for packets traversing the LLN.

C.5.  Address / Header Compression

In order to minimize overhead within the LLN it is desirable to perform some sort of address and/or header compression, perhaps via labels, addresses aggregation, or some other means. This is still under investigation.

Authors' Addresses