idnits 2.17.1 draft-ietf-roll-rpl-08.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Sep 2009 rather than the newer Notice from 28 Dec 2009. (See https://trustee.ietf.org/license-info/) Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (May 28, 2010) is 5080 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: 'TBDREF' is mentioned on line 3739, but not defined == Outdated reference: A later version (-01) exists of draft-hui-6man-rpl-option-00 == Outdated reference: A later version (-02) exists of draft-hui-6man-rpl-routing-header-00 == Outdated reference: A later version (-15) exists of draft-ietf-manet-nhdp-12 == Outdated reference: A later version (-20) exists of draft-ietf-roll-of0-01 == Outdated reference: A later version (-19) exists of draft-ietf-roll-routing-metrics-06 == Outdated reference: A later version (-13) exists of draft-ietf-roll-terminology-03 == Outdated reference: A later version (-08) exists of draft-ietf-roll-trickle-01 Summary: 1 error (**), 0 flaws (~~), 9 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 ROLL T. Winter, Ed. 3 Internet-Draft 4 Intended status: Standards Track P. Thubert, Ed. 5 Expires: November 29, 2010 Cisco Systems 6 RPL Author Team 7 IETF ROLL WG 8 May 28, 2010 10 RPL: IPv6 Routing Protocol for Low power and Lossy Networks 11 draft-ietf-roll-rpl-08 13 Abstract 15 Low power and Lossy Networks (LLNs) are a class of network in which 16 both the routers and their interconnect are constrained: LLN routers 17 typically operate with constraints on (any subset of) processing 18 power, memory and energy (battery), and their interconnects are 19 characterized by (any subset of) high loss rates, low data rates and 20 instability. LLNs are comprised of anything from a few dozen and up 21 to thousands of routers, and support point-to-point traffic (between 22 devices inside the LLN), point-to-multipoint traffic (from a central 23 control point to a subset of devices inside the LLN) and multipoint- 24 to-point traffic (from devices inside the LLN towards a central 25 control point). This document specifies the IPv6 Routing Protocol 26 for LLNs (RPL), which provides a mechanism whereby multipoint-to- 27 point traffic from devices inside the LLN towards a central control 28 point, as well as point-to-multipoint traffic from the central 29 control point to the devices inside the LLN, is supported. Support 30 for point-to-point traffic is also available. 32 Status of this Memo 34 This Internet-Draft is submitted to IETF in full conformance with the 35 provisions of BCP 78 and BCP 79. 37 Internet-Drafts are working documents of the Internet Engineering 38 Task Force (IETF), its areas, and its working groups. Note that 39 other groups may also distribute working documents as Internet- 40 Drafts. 42 Internet-Drafts are draft documents valid for a maximum of six months 43 and may be updated, replaced, or obsoleted by other documents at any 44 time. It is inappropriate to use Internet-Drafts as reference 45 material or to cite them other than as "work in progress." 47 The list of current Internet-Drafts can be accessed at 48 http://www.ietf.org/ietf/1id-abstracts.txt. 50 The list of Internet-Draft Shadow Directories can be accessed at 51 http://www.ietf.org/shadow.html. 53 This Internet-Draft will expire on November 29, 2010. 55 Copyright Notice 57 Copyright (c) 2010 IETF Trust and the persons identified as the 58 document authors. All rights reserved. 60 This document is subject to BCP 78 and the IETF Trust's Legal 61 Provisions Relating to IETF Documents 62 (http://trustee.ietf.org/license-info) in effect on the date of 63 publication of this document. Please review these documents 64 carefully, as they describe your rights and restrictions with respect 65 to this document. Code Components extracted from this document must 66 include Simplified BSD License text as described in Section 4.e of 67 the Trust Legal Provisions and are provided without warranty as 68 described in the BSD License. 70 Table of Contents 72 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 6 73 1.1. Design Principles . . . . . . . . . . . . . . . . . . . . 6 74 1.2. Expectations of Link Layer Type . . . . . . . . . . . . . 7 75 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 7 76 3. Protocol Overview . . . . . . . . . . . . . . . . . . . . . . 9 77 3.1. Topology . . . . . . . . . . . . . . . . . . . . . . . . . 9 78 3.1.1. Topology Identifiers . . . . . . . . . . . . . . . . . 10 79 3.2. Instances, DODAGs, and DODAG Versions . . . . . . . . . . 10 80 3.3. Upward Routes and DODAG Construction . . . . . . . . . . . 12 81 3.3.1. DAG Repair . . . . . . . . . . . . . . . . . . . . . . 12 82 3.3.2. Grounded and Floating DODAGs . . . . . . . . . . . . . 12 83 3.3.3. Administrative Preference . . . . . . . . . . . . . . 13 84 3.3.4. Objective Function (OF) . . . . . . . . . . . . . . . 13 85 3.3.5. Distributed Algorithm Operation . . . . . . . . . . . 13 86 3.4. Downward Routes and Destination Advertisement . . . . . . 14 87 3.5. Routing Metrics and Constraints Used By RPL . . . . . . . 14 88 3.5.1. Loop Avoidance . . . . . . . . . . . . . . . . . . . . 15 89 3.5.2. Rank Properties . . . . . . . . . . . . . . . . . . . 16 90 3.6. Traffic Flows Supported by RPL . . . . . . . . . . . . . . 19 91 3.6.1. Multipoint-to-Point Traffic . . . . . . . . . . . . . 19 92 3.6.2. Point-to-Multipoint Traffic . . . . . . . . . . . . . 19 93 3.6.3. Point-to-Point Traffic . . . . . . . . . . . . . . . . 19 94 4. RPL Instance . . . . . . . . . . . . . . . . . . . . . . . . . 20 95 4.1. RPL Instance ID . . . . . . . . . . . . . . . . . . . . . 20 96 5. ICMPv6 RPL Control Message . . . . . . . . . . . . . . . . . . 21 97 5.1. RPL Security Fields . . . . . . . . . . . . . . . . . . . 23 98 5.2. DODAG Information Solicitation (DIS) . . . . . . . . . . . 26 99 5.2.1. Format of the DIS Base Object . . . . . . . . . . . . 26 100 5.2.2. Secure DIS . . . . . . . . . . . . . . . . . . . . . . 27 101 5.2.3. DIS Options . . . . . . . . . . . . . . . . . . . . . 27 102 5.3. DODAG Information Object (DIO) . . . . . . . . . . . . . . 27 103 5.3.1. Format of the DIO Base Object . . . . . . . . . . . . 27 104 5.3.2. Secure DIO . . . . . . . . . . . . . . . . . . . . . . 29 105 5.3.3. DIO Options . . . . . . . . . . . . . . . . . . . . . 29 106 5.4. Destination Advertisement Object (DAO) . . . . . . . . . . 30 107 5.4.1. Format of the DAO Base Object . . . . . . . . . . . . 30 108 5.4.2. Secure DAO . . . . . . . . . . . . . . . . . . . . . . 31 109 5.4.3. DAO Options . . . . . . . . . . . . . . . . . . . . . 31 110 5.5. Destination Advertisement Object Acknowledgement 111 (DAO-ACK) . . . . . . . . . . . . . . . . . . . . . . . . 31 112 5.5.1. Format of the DAO-ACK Base Object . . . . . . . . . . 31 113 5.5.2. Secure DAO-ACK . . . . . . . . . . . . . . . . . . . . 32 114 5.5.3. DAO-ACK Options . . . . . . . . . . . . . . . . . . . 32 115 5.6. RPL Control Message Options . . . . . . . . . . . . . . . 32 116 5.6.1. RPL Control Message Option Generic Format . . . . . . 32 117 5.6.2. Pad1 . . . . . . . . . . . . . . . . . . . . . . . . . 33 118 5.6.3. PadN . . . . . . . . . . . . . . . . . . . . . . . . . 33 119 5.6.4. Metric Container . . . . . . . . . . . . . . . . . . . 34 120 5.6.5. Route Information . . . . . . . . . . . . . . . . . . 35 121 5.6.6. DODAG Configuration . . . . . . . . . . . . . . . . . 36 122 5.6.7. RPL Target . . . . . . . . . . . . . . . . . . . . . . 37 123 5.6.8. Transit Information . . . . . . . . . . . . . . . . . 39 124 5.6.9. Solicited Information . . . . . . . . . . . . . . . . 40 125 5.6.10. Prefix Information . . . . . . . . . . . . . . . . . . 42 126 6. Upward Routes . . . . . . . . . . . . . . . . . . . . . . . . 44 127 6.1. DIO Base Rules . . . . . . . . . . . . . . . . . . . . . . 45 128 6.2. Upward Route Discovery and Maintenance . . . . . . . . . . 45 129 6.2.1. Neighbors and Parents within a DODAG Version . . . . . 45 130 6.2.2. Neighbors and Parents across DODAG Versions . . . . . 46 131 6.2.3. DIO Message Communication . . . . . . . . . . . . . . 51 132 6.3. DIO Transmission . . . . . . . . . . . . . . . . . . . . . 52 133 6.3.1. Trickle Parameters . . . . . . . . . . . . . . . . . . 52 134 6.4. DODAG Selection . . . . . . . . . . . . . . . . . . . . . 53 135 6.5. Operation as a Leaf Node . . . . . . . . . . . . . . . . . 53 136 6.6. Administrative Rank . . . . . . . . . . . . . . . . . . . 53 137 7. Downward Routes . . . . . . . . . . . . . . . . . . . . . . . 54 138 7.1. Downward Route Discovery and Maintenance . . . . . . . . . 54 139 7.1.1. Overview . . . . . . . . . . . . . . . . . . . . . . . 54 140 7.1.2. Mode of Operation . . . . . . . . . . . . . . . . . . 55 141 7.1.3. Destination Advertisement Parents . . . . . . . . . . 56 142 7.1.4. DAO Operation on Storing Nodes . . . . . . . . . . . . 56 143 7.1.5. Operation of DAO Non-storing Nodes . . . . . . . . . . 60 144 7.1.6. Scheduling to Send DAO (or No-Path) . . . . . . . . . 61 145 7.1.7. Triggering DAO Message from the Sub-DODAG . . . . . . 61 146 7.1.8. Sending DAO Messages to DAO Parents . . . . . . . . . 62 147 7.1.9. Multicast Destination Advertisement Messages . . . . . 63 148 8. Packet Forwarding and Loop Avoidance/Detection . . . . . . . . 64 149 8.1. Suggestions for Packet Forwarding . . . . . . . . . . . . 64 150 8.2. Loop Avoidance and Detection . . . . . . . . . . . . . . . 65 151 8.2.1. Source Node Operation . . . . . . . . . . . . . . . . 66 152 8.2.2. Router Operation . . . . . . . . . . . . . . . . . . . 66 153 9. Multicast Operation . . . . . . . . . . . . . . . . . . . . . 68 154 10. Maintenance of Routing Adjacency . . . . . . . . . . . . . . . 69 155 11. Guidelines for Objective Functions . . . . . . . . . . . . . . 70 156 11.1. Objective Function Behavior . . . . . . . . . . . . . . . 70 157 12. RPL Constants and Variables . . . . . . . . . . . . . . . . . 72 158 13. Manageability Considerations . . . . . . . . . . . . . . . . . 73 159 13.1. Control of Function and Policy . . . . . . . . . . . . . . 73 160 13.1.1. Initialization Mode . . . . . . . . . . . . . . . . . 73 161 13.1.2. DIO Base option . . . . . . . . . . . . . . . . . . . 74 162 13.1.3. Trickle Timers . . . . . . . . . . . . . . . . . . . . 74 163 13.1.4. DAG Version Number Increment . . . . . . . . . . . . . 75 164 13.1.5. Destination Advertisement Timers . . . . . . . . . . . 75 165 13.1.6. Policy Control . . . . . . . . . . . . . . . . . . . . 75 166 13.1.7. Data Structures . . . . . . . . . . . . . . . . . . . 75 167 13.2. Information and Data Models . . . . . . . . . . . . . . . 76 168 13.3. Liveness Detection and Monitoring . . . . . . . . . . . . 76 169 13.3.1. Candidate Neighbor Data Structure . . . . . . . . . . 76 170 13.3.2. Directed Acyclic Graph (DAG) Table . . . . . . . . . . 76 171 13.3.3. Routing Table . . . . . . . . . . . . . . . . . . . . 77 172 13.3.4. Other RPL Monitoring Parameters . . . . . . . . . . . 77 173 13.3.5. RPL Trickle Timers . . . . . . . . . . . . . . . . . . 78 174 13.4. Verifying Correct Operation . . . . . . . . . . . . . . . 78 175 13.5. Requirements on Other Protocols and Functional 176 Components . . . . . . . . . . . . . . . . . . . . . . . . 78 177 13.6. Impact on Network Operation . . . . . . . . . . . . . . . 78 178 14. Security Considerations . . . . . . . . . . . . . . . . . . . 78 179 14.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . 78 180 14.2. Functional Description of Packet Protection . . . . . . . 80 181 14.2.1. Transmission of Outgoing Packets . . . . . . . . . . . 80 182 14.2.2. Reception of Incoming Packets . . . . . . . . . . . . 81 183 14.2.3. Cryptographic Mode of Operation . . . . . . . . . . . 81 184 14.3. Protecting RPL ICMPv6 messages . . . . . . . . . . . . . . 82 185 14.4. Security State Machine . . . . . . . . . . . . . . . . . . 83 186 15. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 83 187 15.1. RPL Control Message . . . . . . . . . . . . . . . . . . . 83 188 15.2. New Registry for RPL Control Codes . . . . . . . . . . . . 84 189 15.3. New Registry for the Mode of Operation (MOP) DIO 190 Control Field . . . . . . . . . . . . . . . . . . . . . . 84 191 15.4. RPL Control Message Option . . . . . . . . . . . . . . . . 85 192 16. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 85 193 17. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 86 194 18. References . . . . . . . . . . . . . . . . . . . . . . . . . . 88 195 18.1. Normative References . . . . . . . . . . . . . . . . . . . 88 196 18.2. Informative References . . . . . . . . . . . . . . . . . . 88 197 Appendix A. Requirements . . . . . . . . . . . . . . . . . . . . 90 198 A.1. Protocol Properties Overview . . . . . . . . . . . . . . . 90 199 A.1.1. IPv6 Architecture . . . . . . . . . . . . . . . . . . 90 200 A.1.2. Typical LLN Traffic Patterns . . . . . . . . . . . . . 90 201 A.1.3. Constraint Based Routing . . . . . . . . . . . . . . . 91 202 A.2. Deferred Requirements . . . . . . . . . . . . . . . . . . 91 203 Appendix B. Outstanding Issues . . . . . . . . . . . . . . . . . 91 204 B.1. Additional Support for P2P Routing . . . . . . . . . . . . 91 205 B.2. Address / Header Compression . . . . . . . . . . . . . . . 91 206 B.3. Managing Multiple Instances . . . . . . . . . . . . . . . 92 207 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 92 209 1. Introduction 211 Low power and Lossy Networks (LLNs) consist of largely of constrained 212 nodes (with limited processing power, memory, and sometimes energy 213 when they are battery operated). These routers are interconnected by 214 lossy links, typically supporting only low data rates, that are 215 usually unstable with relatively low packet delivery rates. Another 216 characteristic of such networks is that the traffic patterns are not 217 simply point-to-point, but in many cases point-to-multipoint or 218 multipoint-to-point. Furthermore such networks may potentially 219 comprise up to thousands of nodes. These characteristics offer 220 unique challenges to a routing solution: the IETF ROLL Working Group 221 has defined application-specific routing requirements for a Low power 222 and Lossy Network (LLN) routing protocol, specified in 223 [I-D.ietf-roll-building-routing-reqs], [RFC5826], [RFC5673], and 224 [RFC5548]. 226 This document specifies the IPv6 Routing Protocol for Low power and 227 lossy networks (RPL). Note that although RPL was specified according 228 to the requirements set forth in the aforementioned requirement 229 documents, its use is in no way limited to these applications. 231 1.1. Design Principles 233 RPL was designed with the objective to meet the requirements spelled 234 out in [I-D.ietf-roll-building-routing-reqs], [RFC5826], [RFC5673], 235 and [RFC5548]. 237 A network may run multiple instances of RPL concurrently. Each such 238 instance may serve different and potentially antagonistic constraints 239 or performance criteria. This document defines how a single instance 240 operates. 242 In order to be useful in a wide range of LLN application domains, RPL 243 separates packet processing and forwarding from the routing 244 optimization objective. Examples of such objectives include 245 minimizing energy, minimizing latency, or satisfying constraints. 246 This document describes the mode of operation of RPL. Other 247 companion documents specify routing objective functions. A RPL 248 implementation, in support of a particular LLN application, will 249 include the necessary objective function(s) as required by the 250 application. 252 A set of companion documents to this specification will provide 253 further guidance in the form of applicability statements specifying a 254 set of operating points appropriate to the Building Automation, Home 255 Automation, Industrial, and Urban application scenarios. 257 1.2. Expectations of Link Layer Type 259 In compliance with the layered architecture of IP, RPL does not rely 260 on any particular features of a specific link layer technology. RPL 261 is designed to be able to operate over a variety of different link 262 layers, including but not limited to, low power wireless or PLC 263 (Power Line Communication) technologies. 265 Implementers may find [RFC3819] a useful reference when designing a 266 link layer interface between RPL and a particular link layer 267 technology. 269 2. Terminology 271 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 272 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 273 "OPTIONAL" in this document are to be interpreted as described in RFC 274 2119 [RFC2119]. 276 Additionally, this document uses terminology from 277 [I-D.ietf-roll-terminology], and introduces the following 278 terminology: 280 DAG: Directed Acyclic Graph. A directed graph having the property 281 that all edges are oriented in such a way that no cycles exist. 282 All edges are contained in paths oriented toward and 283 terminating at one or more root nodes. 285 DAG root: A DAG root is a node within the DAG that has no outgoing 286 edges. Because the graph is acyclic, by definition all DAGs 287 must have at least one DAG root and all paths terminate at a 288 DAG root. 290 Destination Oriented DAG (DODAG): A DAG rooted at a single 291 destination, i.e. at a single DAG root (the DODAG root) with no 292 outgoing edges. 294 DODAG root: A DODAG root is the DAG root of a DODAG. 296 Rank: The rank of a node in a DAG identifies the nodes position with 297 respect to a DODAG root. The farther away a node is from a 298 DODAG root, the higher is the rank of that node. The rank of a 299 node may be a simple topological distance, or may more commonly 300 be calculated as a function of other properties as described 301 later. 303 DODAG parent: A parent of a node within a DODAG is one of the 304 immediate successors of the node on a path towards the DODAG 305 root. The DODAG parent of a node will have a lower rank than 306 the node itself. (See Section 3.5.2.1). 308 DODAG sibling: A sibling of a node within a DODAG is defined in this 309 specification to be any neighboring node which is located at 310 the same rank within a DODAG. Note that siblings defined in 311 this manner do not necessarily share a common DODAG parent. 312 (See Section 3.5.2.1). 314 Sub-DODAG The sub-DODAG of a node is the set of other nodes in the 315 DODAG that might use a path towards the DODAG root that 316 contains that node. Nodes in the sub-DODAG of a node have a 317 greater rank than that node itself (although not all nodes of 318 greater rank are necessarily in the sub-DODAG of that node). 319 (See Section 3.5.2.1). 321 DODAGID: The identifier of a DODAG root. The DODAGID must be unique 322 within the scope of a RPL Instance in the LLN. 324 DODAG Version: A specific sequence number iteration ("version") of a 325 DODAG with a given DODAGID. 327 RPL Instance: A set of possibly multiple DODAGs. A network may have 328 more than one RPL Instance, and a RPL node can participate in 329 multiple RPL Instances. Each RPL Instance operates 330 independently of other RPL Instances. This document describes 331 operation within a single RPL Instance. In RPL, a node can 332 belong to at most one DODAG per RPL Instance. The tuple 333 (RPLInstanceID, DODAGID) uniquely identifies a DODAG. 335 RPLInstanceID: Unique identifier of a RPL Instance. 337 DODAGVersionNumber: A sequential counter that is incremented by the 338 root to form a new Version of a DODAG. A DODAG Version is 339 identified uniquely by the (RPLInstanceID, DODAGID, 340 DODAGVersionNumber) tuple. 342 Up: Up refers to the direction from leaf nodes towards DODAG roots, 343 following the orientation of the edges within the DODAG. This 344 follows the common terminology used in graphs and depth-first- 345 search, where vertices further from the root are "deeper," or 346 "down," and vertices closer to the root are "shallower," or 347 "up." 349 Down: Down refers to the direction from DODAG roots towards leaf 350 nodes, going against the orientation of the edges within the 351 DODAG. This follows the common terminology used in graphs and 352 depth-first-search, where vertices further from the root are 353 "deeper," or "down," and vertices closer to the root are 354 "shallower," or "up." 356 Objective Code Point (OCP): An identifier, used to indicate which 357 Objective Function is in use for forming a DODAG. The 358 Objective Code Point is further described in 359 [I-D.ietf-roll-routing-metrics]. 361 Objective Function (OF): Defines which routing metrics, optimization 362 objectives, and related functions are in use in a DODAG. 364 Goal: The Goal is a host or set of hosts that satisfy a particular 365 application objective (OF). Whether or not a DODAG can provide 366 connectivity to a goal is a property of the DODAG. For 367 example, a goal might be a host serving as a data collection 368 point, or a gateway providing connectivity to an external 369 infrastructure. 371 Grounded: A DODAG is said to be grounded, when the root can reach 372 the Goal of the objective function. 374 Floating: A DODAG is floating if is not Grounded. A floating DODAG 375 is not expected to reach the Goal defined for the OF. 376 Typically, a DAG that is only intended to provide inner 377 connectivity is a Floating DAG. 379 As they form networks, LLN devices often mix the roles of 'host' and 380 'router' when compared to traditional IP networks. In this document, 381 'host' refers to an LLN device that can generate but does not forward 382 RPL traffic, 'router' refers to an LLN device that can forward as 383 well as generate RPL traffic, and 'node' refers to any RPL device, 384 either a host or a router. 386 3. Protocol Overview 388 The aim of this section is to describe RPL in the spirit of 389 [RFC4101]. Protocol details can be found in further sections. 391 3.1. Topology 393 This section describes how the basic RPL topologies, and the rules by 394 which these are constructed, i.e. the rules governing DODAG 395 formation. 397 3.1.1. Topology Identifiers 399 RPL uses four identifiers to maintain the topology: 401 o The first is a RPLInstanceID. A RPLInstanceID identifies a set of 402 one or more DODAGs. All DODAGs in the same RPL Instance use the 403 same OF. A network may have multiple RPLInstanceIDs, each of 404 which defines an independent set of DODAGs, which may be optimized 405 for different OFs and/or applications. The set of DODAGs 406 identified by a RPLInstanceID is called a RPL Instance. 408 o The second is a DODAGID. The scope of a DODAGID is a RPL 409 Instance. The combination of RPLInstanceID and DODAGID uniquely 410 identifies a single DODAG in the network. A RPL Instance may have 411 multiple DODAGs, each of which has an unique DODAGID. 413 o The third is a DODAGVersionNumber. The scope of a 414 DODAGVersionNumber is a DODAG. A DODAG is sometimes reconstructed 415 from the DODAG root, by incrementing the DODAGVersionNumber. The 416 combination of RPLInstanceID, DODAGID, and DODAGVersionNumber 417 uniquely identifies a DODAG Version. 419 o The fourth is rank. The scope of rank is a DODAG Version. Rank 420 establishes a partial order over a DODAG Version, defining 421 individual node positions with respect to the DODAG root. 423 3.2. Instances, DODAGs, and DODAG Versions 425 Each RPL Instance constructs a routing topology optimized for a 426 certain Objective Function (OF) and routing metrics 427 [I-D.ietf-roll-routing-metrics]. A RPL Instance may provide routes 428 to certain destination prefixes, reachable via the DODAG roots or 429 alternate paths within the DODAG. A single RPL Instance contains one 430 or more Destination Oriented DAG (DODAG) roots. These roots may 431 operate independently, or may coordinate over a non-LLN backchannel. 433 Each root has a unique identifier, the DODAGID. 435 A RPL Instance may comprise: 437 o a single DODAG with a single root 439 * For example, a DODAG optimized to minimize latency rooted at a 440 single centralized lighting controller in a home automation 441 application. 443 o multiple uncoordinated DODAGs with independent roots (differing 444 DODAGIDs) 446 * For example, multiple data collection points in an urban data 447 collection application that do not have an always-on backbone 448 suitable to coordinate to form a single DODAG, and further use 449 the formation of multiple DODAGs as a means to dynamically and 450 autonomously partition the network. 452 o a single DODAG with a single virtual root coordinating LLN sinks 453 (with the same DODAGID) over some non-LLN backbone 455 * For example, multiple border routers operating with a reliable 456 backbone, e.g. in support of a 6LowPAN application, that are 457 capable to act as logically equivalent sinks to the same DODAG. 459 o a combination of the above as suited to some application scenario. 461 Traffic is bound to a specific RPL Instance by meta-data that is 462 carried with the packet and associates the packet to a particular 463 RPLInstanceID (Section 8.2). The provisioning or automated discovery 464 of a mapping between a RPLInstanceID and a type or service of 465 application traffic is beyond the scope of this specification. 467 An example of a RPL Instance comprising a number of DODAGs is 468 depicted in Figure 1. Revision of a DODAG Version (two iterations of 469 the same DODAG) is depicted in Figure 2. 471 +----------------------------------------------------------------+ 472 | | 473 | +--------------+ | 474 | | | | 475 | | (R1) | (R2) (Rn) | 476 | | / \ | /| \ / | \ | 477 | | / \ | / | \ / | \ | 478 | | (A) (B) | (C) | (D) ... (F) (G) (H) | 479 | | /|\ |\ | / | |\ | | | | 480 | | : : : : : | : (E) : : : : : | 481 | | | / \ | 482 | +--------------+ : : | 483 | DODAG | 484 | | 485 +----------------------------------------------------------------+ 486 RPL Instance 488 Figure 1: RPL Instance 490 +----------------+ +----------------+ 491 | | | | 492 | (R1) | | (R1) | 493 | / \ | | / | 494 | / \ | | / | 495 | (A) (B) | \ | (A) | 496 | /|\ |\ | ------\ | /|\ | 497 | : : (C) : : | \ | : : (C) | 498 | | / | \ | 499 | | ------/ | \ | 500 | | / | (B) | 501 | | | |\ | 502 | | | : : | 503 | | | | 504 +----------------+ +----------------+ 505 Version N Version N+1 507 Figure 2: DODAG Version 509 3.3. Upward Routes and DODAG Construction 511 RPL provisions routes up towards DODAG roots, forming a DODAG 512 optimized according to the Objective Function (OF) and routing 513 metrics/constraints in use. RPL nodes construct and maintain these 514 DODAGs through exchange of DODAG Information Object (DIO) messages. 515 Undirected links between siblings are also identified during this 516 process, which can be used to provide additional diversity. 518 3.3.1. DAG Repair 520 RPL supports global repair over the DODAG. A DODAG Root may 521 increment the DODAG Version Number, thereby initiating a new DODAG 522 version. This institutes a global repair operation, revising the 523 DODAG and allowing nodes to choose an arbitrary new position within 524 the new DODAG version. Global repair can be seen as a global 525 reoptimization mechanism. 527 RPL also supports mechanisms which may be used for local repair 528 within the DODAG version. The DIO message specifies the necessary 529 parameters as configured from the DODAG root, as controlled by policy 530 at the root. 532 3.3.2. Grounded and Floating DODAGs 534 DODAGs can be grounded or floating. A grounded DODAG offers 535 connectivity to reach a goal. A floating DODAG offers no such 536 connectivity, and provides routes only to nodes within the DODAG. 538 Floating DODAGs may be used, for example, to preserve inner 539 connectivity during repair. 541 3.3.3. Administrative Preference 543 An implementation/deployment may specify that some DODAG roots should 544 be used over others through an administrative preference. 545 Administrative preference offers a way to control traffic and 546 engineer DODAG formation in order to better support application 547 requirements or needs. 549 3.3.4. Objective Function (OF) 551 The Objective Function (OF) implements the optimization objectives of 552 route selection within the RPL Instance. The OF is identified by an 553 Objective Code Point (OCP) within the DIO. The OF also specifies the 554 procedure used to select parents and compute rank within a DODAG 555 version along with potentially other DODAG characteristics. Further 556 details may be found in Section 11, [I-D.ietf-roll-routing-metrics], 557 [I-D.ietf-roll-of0], and related companion specifications. 559 3.3.5. Distributed Algorithm Operation 561 A high level overview of the distributed algorithm, which constructs 562 the DODAG, is as follows: 564 o Some nodes are configured to be DODAG roots, with associated DODAG 565 configuration. 567 o Nodes advertise their presence, affiliation with a DODAG, routing 568 cost, and related metrics by sending link-local multicast DIO 569 messages. 571 o Nodes may adjust the rate at which DIO messages are sent in 572 response to stability or detection of routing inconsistencies from 573 both control or data packets (see Section 8.2 for more). 575 o Nodes listen for DIOs and use their information to join a new 576 DODAG, or to maintain an existing DODAG, as according to the 577 specified Objective Function and rank-based loop avoidance rules. 579 o Nodes provision routing table entries, for the destinations 580 specified by the DIO, via their DODAG parents in the DODAG 581 version. Nodes MUST provision a DODAG parent as a default route 582 for the associated instance. It is up to the end-to-end 583 application to select the RPL instance to be associated to its 584 traffic (should there be more than one instance) and thus the 585 default route upwards when no longer-match exists. 587 o Nodes may identify DODAG siblings within the DODAG version to 588 increase path diversity and decrease convergence time during 589 repair. 591 3.4. Downward Routes and Destination Advertisement 593 RPL constructs and maintains DODAGs with DIO messages to establish 594 upward routes: it uses Destination Advertisement Object (DAO) 595 messages to establish downward routes along the DODAG as well as 596 other P2P routes. DAO messages are an optional feature for 597 applications that require P2MP or P2P traffic, in either storing 598 (fully stateful) or non-storing (fully source routed 599 [I-D.hui-6man-rpl-routing-header]) mode. 601 3.5. Routing Metrics and Constraints Used By RPL 603 Routing metrics are used by routing protocols to compute shortest 604 paths. Interior Gateway Protocols (IGPs) such as IS-IS ([RFC5120]) 605 and OSPF ([RFC4915]) use static link metrics. Such link metrics can 606 simply reflect the bandwidth or can also be computed according to a 607 polynomial function of several metrics defining different link 608 characteristics. Some routing protocols support more than one 609 metric: in the vast majority of the cases, one metric is used per 610 (sub)topology. Less often, a second metric may be used as a tie- 611 breaker in the presence of Equal Cost Multiple Paths (ECMP). The 612 optimization of multiple metrics is known as an NP complete problem 613 and is sometimes supported by some centralized path computation 614 engine. 616 In contrast, LLNs do require the support of both static and dynamic 617 metrics. Furthermore, both link and node metrics are required. In 618 the case of RPL, it is virtually impossible to define one metric, or 619 even a composite metric, that will satisfy all use cases. 621 In addition, RPL supports constrained-based routing where constraints 622 may be applied to both link and nodes. If a link or a node does not 623 satisfy a required constraint, it is 'pruned' from the candidate 624 list, thus leading to a constrained shortest path. 626 The set of supported link/node constraints and metrics is specified 627 in [I-D.ietf-roll-routing-metrics]. 629 An Objective Function specifies constraints in use, and how these are 630 used, in addition to the objectives used to compute the (constrained) 631 path. Upstream and Downstream metrics may be merged or advertised 632 separately depending on the OF and the metrics. When they are 633 advertised separately, it may happen that the set of DIO parents is 634 different from the set of DAO parents (a DAO parent is a node to 635 which unicast DAO messages are sent). Yet, all are DODAG parents 636 with regards to the rules for Rank computation. 638 Example 1: Shortest path: path offering the shortest end-to-end delay 640 Example 2: Constrained shortest path: the path that does not traverse 641 any battery-operated node and that optimizes the path 642 reliability 644 3.5.1. Loop Avoidance 646 RPL guarantees neither loop free path selection nor tight delay 647 convergence times. In order to reduce control overhead, however, 648 such as the cost of the count-to-infinity problem, RPL avoids 649 creating loops when undergoing topology changes. Furthermore, RPL 650 includes rank-based datapath validation mechanisms for detecting 651 loops when they do occur. RPL uses this loop detection to ensure 652 that packets make forward progress within the DODAG version and 653 trigger repairs when necessary. 655 3.5.1.1. Greediness and Rank-based Instabilities 657 A node is greedy if it attempts to move deeper in the DODAG version, 658 in order to increase the size of the parent set or improve some other 659 metric. Moving deeper in within a DODAG version in this manner could 660 result in instability and be detrimental to other nodes. 662 Once a node has joined a DODAG version, RPL disallows certain 663 behaviors, including greediness, in order to prevent resulting 664 instabilities in the DODAG version. 666 Suppose a node is willing to receive and process a DIO messages from 667 a node in its own sub-DODAG, and in general a node deeper than 668 itself. In this case, a possibility exists that a feedback loop is 669 created, wherein two or more nodes continue to try and move in the 670 DODAG version while attempting to optimize against each other. In 671 some cases, this will result in instability. It is for this reason 672 that RPL limits the cases where a node may process DIO messages from 673 deeper nodes to some forms of local repair. This approach creates an 674 'event horizon', whereby a node cannot be influenced beyond some 675 limit into an instability by the action of nodes that may be in its 676 own sub-DODAG. 678 3.5.1.2. DODAG Loops 680 A DODAG loop may occur when a node detaches from the DODAG and 681 reattaches to a device in its prior sub-DODAG. This may happen in 682 particular when DIO messages are missed. Strict use of the DODAG 683 Version Number can eliminate this type of loop, but this type of loop 684 may possibly be encountered when using some local repair mechanisms. 686 3.5.1.3. DAO Loops 688 A DAO loop may occur when the parent has a route installed upon 689 receiving and processing a DAO message from a child, but the child 690 has subsequently cleaned up the related DAO state. This loop happens 691 when a No-Path (a DAO message that invalidates a previously announced 692 prefix) was missed and persists until all state has been cleaned up. 693 RPL includes an optional mechanism to acknowledge DAO messages, which 694 may mitigate the impact of a single DAO message being missed. RPL 695 includes loop detection mechanisms that may mitigate the impact of 696 DAO loops and trigger their repair. 698 In the case where stateless DAO operation is used, i.e. source 699 routing specifies the down routes, then DAO Loops should not occur on 700 the stateless portions of the path. 702 3.5.1.4. Sibling Loops 704 Sibling loops could occur if a group of siblings kept choosing 705 amongst themselves as successors such that a packet does not make 706 forward progress. This specification limits the number of times that 707 sibling forwarding may be used at a given rank, in order to prevent 708 sibling loops. 710 3.5.2. Rank Properties 712 The rank of a node is a scalar representation of the location of that 713 node within a DODAG version. The rank is used to avoid and detect 714 loops, and as such must demonstrate certain properties. The exact 715 calculation of the rank is left to the Objective Function, and may 716 depend on parents, link metrics, and the node configuration and 717 policies. 719 The rank is not a cost metric, although its value can be derived from 720 and influenced by metrics. The rank has properties of its own that 721 are not necessarily those of all metrics: 723 Type: The rank is an abstract decimal value. 725 Function: The rank is the expression of a relative position within a 726 DODAG version with regard to neighbors and is not necessarily 727 a good indication or a proper expression of a distance or a 728 cost to the root. 730 Stability: The stability of the rank determines the stability of the 731 routing topology. Some dampening or filtering might be 732 applied to keep the topology stable, and thus the rank does 733 not necessarily change as fast as some physical metrics 734 would. A new DODAG version would be a good opportunity to 735 reconcile the discrepancies that might form over time between 736 metrics and ranks within a DODAG version. 738 Granularity: The portion of the rank that is used to define a node's 739 position in the DAG, DAGRank(node), is coarse grained. A 740 fine granularity would make the selection of siblings 741 difficult, since siblings must have the exact same rank 742 value. 744 Properties: The rank is strictly monotonic, and can be used to 745 validate a progression from or towards the root. A metric, 746 like bandwidth or jitter, does not necessarily exhibit this 747 property. 749 Abstract: The rank does not have a physical unit, but rather a range 750 of increment per hop, where the assignment of each increment 751 is to be determined by the Objective Function. 753 The rank value feeds into DODAG parent selection, according to the 754 RPL loop-avoidance strategy. Once a parent has been added, and a 755 rank value for the node within the DODAG has been advertised, the 756 nodes further options with regard to DODAG parent selection and 757 movement within the DODAG are restricted in favor of loop avoidance. 759 3.5.2.1. Rank Comparison (DAGRank()) 761 Rank may be thought of as a fixed point number, where the position of 762 the decimal point between the integer part and the fractional part is 763 determined by MinHopRankIncrease. MinHopRankIncrease is the minimum 764 increase in rank between a node and any of its DODAG parents. When 765 an objective function computes rank, the objective function operates 766 on the entire (i.e. 16-bit) rank quantity. When rank is compared, 767 e.g. for determination of parent/sibling relationships or loop 768 detection, the integer portion of the rank is to be used. The 769 integer portion of the Rank is computed by the DAGRank() macro as 770 follows: 772 DAGRank(rank) = floor(rank/MinHopRankIncrease) 774 MinHopRankIncrease is provisioned at the DODAG Root and propagated in 775 the DIO message. For efficient implementation the MinHopRankIncrease 776 MUST be a power of 2. An implementation may configure a value 777 MinHopRankIncrease as appropriate to balance between the loop 778 avoidance logic of RPL (i.e. selection of eligible parents and 779 siblings) and the metrics in use. 781 By convention in this document, using the macro DAGRank(node) may be 782 interpreted as DAGRank(node.rank), where node.rank is the rank value 783 as maintained by the node. 785 A node A has a rank less than the rank of a node B if DAGRank(A) is 786 less than DAGRank(B). 788 A node A has a rank equal to the rank of a node B if DAGRank(A) is 789 equal to DAGRank(B). 791 A node A has a rank greater than the rank of a node B if DAGRank(A) 792 is greater than DAGRank(B). 794 3.5.2.2. Rank Relationships 796 The computation of the rank MUST be done in such a way so as to 797 maintain the following properties for any nodes M and N that are 798 neighbors in the LLN: 800 DAGRank(M) is less than DAGRank(N): In this case, the position of M 801 is closer to the DODAG root than the position of N. Node M 802 may safely be a DODAG parent for Node N without risk of 803 creating a loop. Further, for a node N, all parents in the 804 DODAG parent set must be of rank less than DAGRank(N). In 805 other words, the rank presented by a node N MUST be greater 806 than that presented by any of its parents. 808 DAGRank(M) equals DAGRank(N): In this case the positions of M and N 809 within the DODAG and with respect to the DODAG root are 810 similar (identical). In some cases, Node M may be used as a 811 successor by Node N, which however entails the chance of 812 creating a loop (which must be detected and resolved by some 813 other means). 815 DAGRank(M) is greater than DAGRank(N): In this case, the position of 816 M is farther from the DODAG root than the position of N. 817 Further, Node M may in fact be in the sub-DODAG of Node N. If 818 node N selects node M as DODAG parent there is a risk to 819 create a loop. 821 As an example, the rank could be computed in such a way so as to 822 closely track ETX (Expected Transmission Count, a fairly common 823 routing metric used in LLN and defined in 825 [I-D.ietf-roll-routing-metrics]) when the objective function is to 826 minimize ETX, or latency when the objective function is to minimize 827 latency, or in a more complicated way as appropriate to the objective 828 function being used within the DODAG. 830 3.6. Traffic Flows Supported by RPL 832 3.6.1. Multipoint-to-Point Traffic 834 Multipoint-to-Point (MP2P) is a dominant traffic flow in many LLN 835 applications ([I-D.ietf-roll-building-routing-reqs], [RFC5826], 836 [RFC5673], [RFC5548]). The destinations of MP2P flows are designated 837 nodes that have some application significance, such as providing 838 connectivity to the larger Internet or core private IP network. RPL 839 supports MP2P traffic by allowing MP2P destinations to be reached via 840 DODAG roots. 842 3.6.2. Point-to-Multipoint Traffic 844 Point-to-multipoint (P2MP) is a traffic pattern required by several 845 LLN applications ([I-D.ietf-roll-building-routing-reqs], [RFC5826], 846 [RFC5673], [RFC5548]). RPL supports P2MP traffic by using a 847 destination advertisement mechanism that provisions routes toward 848 destination prefixes and away from roots. Destination advertisements 849 can update routing tables as the underlying DODAG topology changes. 851 3.6.3. Point-to-Point Traffic 853 RPL DODAGs provide a basic structure for point-to-point (P2P) 854 traffic. For a RPL network to support P2P traffic, a root must be 855 able to route packets to a destination. Nodes within the network may 856 also have routing tables to destinations. A packet flows towards a 857 root until it reaches an ancestor that has a known route to the 858 destination. As pointed out later in this document, in the most 859 constrained case (when nodes cannot store routes), that common 860 ancestor may be the DODAG root. In other cases it may be a node 861 closer to both the source and destination. 863 RPL also supports the case where a P2P destination is a 'one-hop' 864 neighbor. 866 RPL neither specifies nor precludes additional mechanisms for 867 computing and installing potentially more optimal routes to support 868 arbitrary P2P traffic. 870 4. RPL Instance 872 Within a given LLN, there may be multiple, logically independent RPL 873 instances. This document describes how a single instance behaves. 875 A node may belong to multiple RPL Instances. 877 An instance can be either local to a root or global. When the 878 instance is local, the DAG is a single DODAG that is rooted at the 879 node that owns the DODAGID. This is used in particular for the 880 construction of a temporary DODAG in support of P2P traffic 881 optimization between the root and some other nodes. 883 Control and Data Packets that traverse a RPL network MUST be tagged 884 in such a fashion that the instance is unambiguously identified (TBD 885 flow label or RPL Hop-by-hop option ([I-D.hui-6man-rpl-option])). 886 The identifiers include the RPLInstanceID and the DODAGID for local 887 instances. 889 4.1. RPL Instance ID 891 A global RPLInstanceID MUST be unique to the whole LLN. Mechanisms 892 for allocating and provisioning global RPLInstanceID are out of scope 893 for this document. There can be up to 128 global instance in the 894 whole network, and up 64 local instances per DODAGID. 896 A global RPLinstanceID is encoded in a RPLinstanceID field as 897 follows: 899 0 1 2 3 4 5 6 7 900 +-+-+-+-+-+-+-+-+ 901 |0| ID | Global RPLinstanceID in 0..127 902 +-+-+-+-+-+-+-+-+ 904 Figure 3: RPL Instance ID field format for global instances 906 A local RPLInstanceID is autoconfigured by the node that owns the 907 DODAGID and it MUST be unique for that DODAGID. In that case, the 908 DODAGID MUST be a valid address of the root that is used as an 909 endpoint of all communications within that instance. 911 A local RPLinstanceID is encoded in a RPLinstanceID field as follows: 913 0 1 2 3 4 5 6 7 914 +-+-+-+-+-+-+-+-+ 915 |1|D| ID | Local RPLInstanceID in 0..63 916 +-+-+-+-+-+-+-+-+ 918 Figure 4: RPL Instance ID field format for local instances 920 The D flag in a Local RPLInstanceID is always set to 0 in RPL control 921 messages. It is used in data packets to indicate whether the DODAGID 922 is the source or the destination of the packet. If the D flag is set 923 to 1 then the destination address of the IPv6 packet MUST be the 924 DODAGID. If the D flag is clear then the source address of the IPv6 925 packet MUST be the DODAGID. 927 5. ICMPv6 RPL Control Message 929 This document defines the RPL Control Message, a new ICMPv6 message. 930 A RPL Control Message is identified by a code, and composed of a base 931 that depends on the code, and a series of options. 933 A RPL Control Message has the scope of a link. The source address is 934 a link local address. The destination address is either all routers 935 multicast address (FF02::2) or a link local address. 937 In accordance with [RFC4443], the RPL Control Message consists of an 938 ICMPv6 header followed by a message body. The message body is 939 comprised of a message base and possibly a number of options as 940 illustrated in Figure 5. 942 0 1 2 3 943 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 1 944 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 945 | Type | Code | Checksum | 946 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 947 | | 948 . Base . 949 . . 950 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 951 | | 952 . Option(s) . 953 . . 954 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 956 Figure 5: RPL Control Message 958 The RPL Control message is an ICMPv6 information message with a 959 requested Type of 155 (to be confirmed by IANA). 961 The Code field identifies the type of RPL Control Message. This 962 document defines codes for the following RPL Control Message types 963 (all codes are to be confirmed by the IANA Section 15.2): 965 o 0x00: DODAG Information Solicitation (Section 5.2) 967 o 0x01: DODAG Information Object (Section 5.3) 969 o 0x02: Destination Advertisement Object (Section 5.4) 971 o 0x03: Destination Advertisement Object Acknowledgment 972 (Section 5.5) 974 o 0x80: Secure DODAG Information Solicitation (Section 5.2.2) 976 o 0x81: Secure DODAG Information Object (Section 5.3.2) 978 o 0x82: Secure Destination Advertisement Object (Section 5.4.2) 980 o 0x83: Secure Destination Advertisement Object Acknowledgment 981 (Section 5.5.2) 983 The high order bit (0x80) of the code denotes whether the RPL message 984 has security enabled. Secure versions of RPL messages have a 985 modified format to support confidentiality and integrity, illustrated 986 in Figure Figure 6. 988 0 1 2 3 989 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 1 990 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 991 | Type | Code | Checksum | 992 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 993 | | 994 . Security . 995 . . 996 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 997 | | 998 . Base . 999 . . 1000 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1001 | | 1002 . Option(s) . 1003 . . 1004 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1005 Figure 6: Secure RPL Control Message 1007 The remainder of this section describes the currently defined RPL 1008 Control Message Base formats followed by the currently defined RPL 1009 Control Message Options. 1011 5.1. RPL Security Fields 1013 Each RPL message has a secure version. The secure versions provide 1014 integrity and confidentiality. Because security covers the base 1015 message as well as options, in secured messages the security 1016 information lies between the checksum and base, as shown in Figure 1017 Figure 6. 1019 The format of the security section is as follows: 1021 0 1 2 3 1022 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 1023 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1024 |0|0|C|KIM| LVL | | 1025 +-+-+-+-+-+-+-+-+ + 1026 | Counter | 1027 . . 1028 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1029 | | 1030 . Key Identifier . 1031 . . 1032 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1034 Security 1036 All fields are considered as packet payload from a security 1037 processing perspective. The exact placement and format of message 1038 integrity/authentication codes has not yet been determined. 1040 Use of the Security section is further detailed in Section 14. 1042 Security Control Field: The Security Control Field has one flag and 1043 two fields: 1045 Counter Compression (C): If the Counter Compression flag is 1046 set then the Counter field is compressed from 4 bytes 1047 into 1 byte. If the Counter Compression flag is clear 1048 then the Counter field is 4 bytes and uncompressed. 1050 Key Identifier Mode (KIM): The Key Identifier Mode field 1051 indicates whether the key used for packet protection is 1052 determined implicitly or explicitly and indicates the 1053 particular representation of the Key Identifier field. 1054 The Key Identifier Mode is set one of the non-reserved 1055 values from the table below: 1057 +------+-----+-----------------------------+------------+ 1058 | Mode | KIM | Meaning | Key | 1059 | | | | Identifier | 1060 | | | | Length | 1061 | | | | (octets) | 1062 +------+-----+-----------------------------+------------+ 1063 | 0 | 00 | Peer-to-peer key determined | 0 | 1064 | | | implicitly from originator | | 1065 | | | and recipient of packet. | | 1066 | | | | | 1067 | | | Key Source is not present. | | 1068 | | | Key Index is not present. | | 1069 +------+-----+-----------------------------+------------+ 1070 | 1 | 01 | Group key determined | 1 | 1071 | | | implicitly from Key Index | | 1072 | | | and side information. | | 1073 | | | | | 1074 | | | Key Source is not present. | | 1075 | | | Key Index is present. | | 1076 +------+-----+-----------------------------+------------+ 1077 | 2 | 10 | Signature key used; group | 0/9 | 1078 | | | key determined explicitly | | 1079 | | | if encryption used. | | 1080 | | | | | 1081 | | | Key Source may be present. | | 1082 | | | Key Index may be present. | | 1083 +------+-----+-----------------------------+------------+ 1084 | 3 | 11 | Group key determined | 9 | 1085 | | | explicitly from Key Source | | 1086 | | | Identifier and Key Index. | | 1087 | | | | | 1088 | | | Key Source is present. | | 1089 | | | Key Index is present. | | 1090 +------+-----+-----------------------------+------------+ 1092 Key Identifier Mode (KIM) Encoding 1094 Security Level (LVL): The Security Level field indicates the 1095 provided packet protection. This value can be adapted on 1096 a per-packet basis and allows for varying levels of data 1097 authenticity and, optionally, for data confidentiality. 1098 When nontrivial protection is provided, replay protection 1099 is always provided. The Security Level is set to one of 1100 the non-reserved values in the table below: 1102 +--------------------+-------------------+ 1103 | Without Signatures | With Signatures | 1104 +----+-----+-------------+------+-------------+-----+ 1105 | ID | LVL | Attributes | Auth | Attributes | Sig | 1106 | | | | Len | | Len | 1107 +----+-----+-------------+------+-------------+-----+ 1108 | 0 | 000 | None | 0 | None | 37 | 1109 | 1 | 001 | MIC-32 | 4 | Sign-32 | 37 | 1110 | 2 | 010 | MIC-64 | 8 | Sign-64 | 45 | 1111 | 3 | 011 | Rsvd | N/A | Rsvd | N/A | 1112 | 4 | 100 | ENC | 0 | ENC | 37 | 1113 | 5 | 101 | ENC-MIC-32 | 4 | ENC-Sign-32 | 41 | 1114 | 6 | 110 | ENC-MIC-64 | 8 | ENC-Sign-64 | 45 | 1115 | 7 | 111 | Rsvd | N/A | Reserved | N/A | 1116 +----+-----+-------------+------+-------------+-----+ 1118 Security Level (LVL) Encoding 1120 Counter: The Counter field indicates the non-repeating value (nonce) 1121 used with the cryptographic mechanism that implements packet 1122 protection and allows for the provision of semantic security. 1123 This value is compressed from 4 octets to 1 octet if the 1124 Counter Compression field of the Security Control Field is set 1125 to one. 1127 Key Identifier: The Key Identifier field indicates which key was 1128 used to protect the packet. This field provides various levels 1129 of granularity of packet protection, including peer-to-peer 1130 keys, group keys, and signature keys. This field is 1131 represented as indicated by the Key Identifier Mode field and 1132 is formatted as follows: 1134 0 1 2 3 1135 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 1136 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1137 | | 1138 . Key Source . 1139 . . 1140 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1141 | | 1142 . Key Index . 1143 . . 1144 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1146 Key Identifier 1148 Key Source: The Key Source field, when present, indicates the 1149 logical identifier of the originator of a group key. 1150 When present this field is 8 bytes in length. 1152 Key Index: The Key Index field, when present, allows unique 1153 identification of different keys with the same 1154 originator. It is the responsibility of each key 1155 originator to make sure that actively used keys that it 1156 issues have distinct key indices and that all key indices 1157 have a value unequal to 0x00. When present this field is 1158 1 byte in length. 1160 Unassigned bits of the Security section are reserved. They MUST be 1161 set to zero on transmission and MUST be ignored on reception. 1163 5.2. DODAG Information Solicitation (DIS) 1165 The DODAG Information Solicitation (DIS) message may be used to 1166 solicit a DODAG Information Object from a RPL node. Its use is 1167 analogous to that of a Router Solicitation as specified in IPv6 1168 Neighbor Discovery; a node may use DIS to probe its neighborhood for 1169 nearby DODAGs. Section 6.3 describes how nodes respond to a DIS. 1171 5.2.1. Format of the DIS Base Object 1173 0 1 2 1174 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 1175 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1176 | Reserved | Option(s)... 1177 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1179 Figure 7: The DIS Base Object 1181 Unassigned bits of the DIS Base are reserved. They MUST be set to 1182 zero on transmission and MUST be ignored on reception. 1184 5.2.2. Secure DIS 1186 A Secure DIS message follows the format in Figure Figure 6, where the 1187 base format is the DIS message shown in Figure Figure 7. 1189 5.2.3. DIS Options 1191 The DIS message MAY carry valid options. 1193 This specification allows for the DIS message to carry the following 1194 options: 1195 0x00 Pad1 1196 0x01 PadN 1197 0x05 RPL Target 1198 0x07 Solicited Information 1200 5.3. DODAG Information Object (DIO) 1202 The DODAG Information Object carries information that allows a node 1203 to discover a RPL Instance, learn its configuration parameters, 1204 select a DODAG parent set, and maintain the upward routing topology. 1206 5.3.1. Format of the DIO Base Object 1208 0 1 2 3 1209 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 1210 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1211 | RPLInstanceID | Version | Rank | 1212 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1213 |G|A|T|MOP| Prf | DTSN | Reserved | 1214 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1215 | | 1216 + + 1217 | | 1218 + DODAGID + 1219 | | 1220 + + 1221 | | 1222 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1223 | Option(s)... 1224 +-+-+-+-+-+-+-+-+ 1226 Figure 8: The DIO Base Object 1228 Control Field: The DAG Control Field has three flags and two fields: 1230 Grounded (G): The Grounded (G) flag indicates whether the 1231 upward routes this node advertises provide connectivity 1232 to the set of addresses which are application-defined 1233 goals. If the flag is set, the DODAG is grounded and 1234 provides such connectivity. If the flag is cleared, the 1235 DODAG is floating and may not provide such connectivity. 1237 Destination Advertisement Supported (A): The Destination 1238 Advertisement Supported (A) flag indicates whether the 1239 root of this DODAG can collect and use downward route 1240 state. If the flag is set, nodes in the network are 1241 enabled to exchange destination advertisements messages 1242 to build downward routes (Section 7). If the flag is 1243 cleared, destination advertisement messages are disabled 1244 and the DODAG maintains only upward routes. 1246 Destination Advertisement Trigger (T): The Destination 1247 Advertisement Trigger (T) flag indicates a complete 1248 refresh of downward routes. If the flag is set, then a 1249 refresh of downward route state is to take place over the 1250 entire DODAG. If the flag is cleared, the downward route 1251 maintenance is in its normal mode of operation. The 1252 further details of this process are described in 1253 Section 7. 1255 Mode of Operation (MOP): The Mode of Operation (MOP) field 1256 identifies the mode of operation of the RPL Instance as 1257 administratively provisioned at and distributed by the 1258 DODAG Root. All nodes who join the DODAG must be able to 1259 honor the MOP in order to fully participate as a router, 1260 or else they must only join as a leaf. MOP is encoded as 1261 in the table below: 1263 +-----+-------------------------------------------------+ 1264 | MOP | Meaning | 1265 +-----+-------------------------------------------------+ 1266 | 00 | Non-storing | 1267 | 01 | Storing | 1268 | 10 | Reserved for future specification of mixed-mode | 1269 | 11 | Reserved | 1270 +-----+-------------------------------------------------+ 1272 Mode of Operation (MOP) Encoding 1274 DODAGPreference (Prf): A 3-bit unsigned integer that defines 1275 how preferable the root of this DODAG is compared to 1276 other DODAG roots within the instance. DAGPreference 1277 ranges from 0x00 (least preferred) to 0x07 (most 1278 preferred). The default is 0 (least preferred). 1279 Section 6.2 describes how DAGPreference affects DIO 1280 processing. 1282 Version Number: 8-bit unsigned integer set by the DODAG root. 1283 Section 6.2 describes the rules for version numbers and how 1284 they affect DIO processing. 1286 Rank: 16-bit unsigned integer indicating the DODAG rank of the node 1287 sending the DIO message. Section 6.2 describes how Rank is set 1288 and how it affects DIO processing. 1290 RPLInstanceID: 8-bit field set by the DODAG root that indicates 1291 which RPL Instance the DODAG is part of. 1293 Destination Advertisement Trigger Sequence Number (DTSN): 8-bit 1294 unsigned integer set by the node issuing the DIO message. The 1295 Destination Advertisement Trigger Sequence Number (DTSN) flag 1296 is used as part of the procedure to maintain downward routes. 1297 The details of this process are described in Section 7. 1299 DODAGID: 128-bit unsigned integer set by a DODAG root which uniquely 1300 identifies a DODAG. Possibly derived from the IPv6 address of 1301 the DODAG root. 1303 Unassigned bits of the DIO Base are reserved. They MUST be set to 1304 zero on transmission and MUST be ignored on reception. 1306 5.3.2. Secure DIO 1308 A Secure DIO message follows the format in Figure Figure 6, where the 1309 base format is the DIS message shown in Figure Figure 8. 1311 5.3.3. DIO Options 1313 The DIO message MAY carry valid options. 1315 This specification allows for the DIO message to carry the following 1316 options: 1317 0x00 Pad1 1318 0x01 PadN 1319 0x02 Metric Container 1320 0x03 Routing Information 1321 0x04 DODAG Configuration 1322 0x09 Prefix Information 1324 5.4. Destination Advertisement Object (DAO) 1326 The Destination Advertisement Object (DAO) is used to propagate 1327 destination information upwards along the DODAG. The DAO message is 1328 unicast by the child to the selected parent(s). The DAO message may 1329 optionally, upon explicit request or error, be acknowledged by the 1330 parent with a Destination Advertisement Acknowledgement (DAO-ACK) 1331 message back to the child. 1333 5.4.1. Format of the DAO Base Object 1335 0 1 2 3 1336 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 1337 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1338 | RPLInstanceID |K|D| Reserved | DAOSequence | 1339 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1340 | | 1341 + + 1342 | | 1343 + DODAGID* + 1344 | | 1345 + + 1346 | | 1347 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1348 | Option(s)... 1349 +-+-+-+-+-+-+-+-+ 1351 Figure 9: The DAO Base Object 1353 RPLInstanceID: 8-bit field indicating the topology instance 1354 associated with the DODAG, as learned from the DIO. 1356 K: The 'K' flag indicates that the parent is expected to send a 1357 DAO-ACK back. 1359 D: The 'D' flag indicates that the DODAGID field is present. This 1360 would typically only be set when a local RPLInstanceID is used. 1362 DAOSequence: Incremented at each unique DAO message, echoed in the 1363 DAO-ACK message. 1365 DODAGID*: 128-bit unsigned integer set by a DODAG root which 1366 uniquely identifies a DODAG. This field is only present when 1367 the 'D' flag is set. This field is typically only present when 1368 a local RPLInstanceID is in use, in order to identify the 1369 DODAGID that is associated with the RPLInstanceID. When a 1370 global RPLInstanceID is in use this field need not be present. 1372 Unassigned bits of the DAO Base are reserved. They MUST be set to 1373 zero on transmission and MUST be ignored on reception. 1375 5.4.2. Secure DAO 1377 A Secure DAO message follows the format in Figure Figure 6, where the 1378 base format is the DAO message shown in Figure Figure 9. 1380 5.4.3. DAO Options 1382 The DAO message MAY carry valid options. 1384 This specification allows for the DAO message to carry the following 1385 options: 1386 0x00 Pad1 1387 0x01 PadN 1388 0x05 RPL Target 1389 0x06 Transit Information 1391 A special case of the DAO message, termed a No-Path, is used to clear 1392 downward routing state that has been provisioned through DAO 1393 operation. The No-Path carries a RPL Transit Information option, 1394 which identifies the destination to which the DAO is associated, with 1395 a lifetime of 0x00000000 to indicate a loss of reachability. 1397 5.5. Destination Advertisement Object Acknowledgement (DAO-ACK) 1399 The DAO-ACK message is sent as a unicast packet by a DAO parent in 1400 response to a unicast DAO message from a child. 1402 5.5.1. Format of the DAO-ACK Base Object 1404 0 1 2 3 1405 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 1406 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1407 | RPLInstanceID | Reserved | DAOSequence | Status | 1408 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1409 | Option(s)... 1410 +-+-+-+-+-+-+-+-+ 1411 Figure 10: The DAO ACK Base Object 1413 RPLInstanceID: 8-bit field indicating the topology instance 1414 associated with the DODAG, as learned from the DIO. 1416 DAOSequence: Incremented at each DAO message from a given child, 1417 echoed in the DAO-ACK by the parent. The DAOSequence serves in 1418 the parent-child communication and is not to be confused with 1419 the Transit Information option Sequence that is associated to a 1420 given target down the DODAG. 1422 Status: Indicates the completion. 0 is unqualified acceptance, above 1423 128 are rejection code indicating that the node should select 1424 an alternate parent. 1426 Unassigned bits of the DAO-ACK Base are reserved. They MUST be set 1427 to zero on transmission and MUST be ignored on reception. 1429 5.5.2. Secure DAO-ACK 1431 A Secure DAO-ACK message follows the format in Figure Figure 6, where 1432 the base format is the DAO-ACK message shown in Figure Figure 10. 1434 5.5.3. DAO-ACK Options 1436 This specification does not define any options to be carried by the 1437 DAO-ACK message. 1439 5.6. RPL Control Message Options 1441 5.6.1. RPL Control Message Option Generic Format 1443 RPL Control Message Options all follow this format: 1445 0 1 2 1446 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 1447 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - - 1448 | Option Type | Option Length | Option Data 1449 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - - 1451 Figure 11: RPL Option Generic Format 1453 Option Type: 8-bit identifier of the type of option. The Option 1454 Type values are to be confirmed by the IANA Section 15.4. 1456 Option Length: 8-bit unsigned integer, representing the length in 1457 octets of the option, not including the Option Type and Length 1458 fields. 1460 Option Data: A variable length field that contains data specific to 1461 the option. 1463 When processing a RPL message containing an option for which the 1464 Option Type value is not recognized by the receiver, the receiver 1465 MUST silently ignore the unrecognized option and continue to process 1466 the following option, correctly handling any remaining options in the 1467 message. 1469 RPL message options may have alignment requirements. Following the 1470 convention in IPv6, options with alignment requirements are aligned 1471 in a packet such that multi-octet values within the Option Data field 1472 of each option fall on natural boundaries (i.e., fields of width n 1473 octets are placed at an integer multiple of n octets from the start 1474 of the header, for n = 1, 2, 4, or 8). 1476 5.6.2. Pad1 1478 The Pad1 option may be present in DIS, DIO, DAO, and DAO-ACK 1479 messages, and its format is as follows: 1481 0 1482 0 1 2 3 4 5 6 7 1483 +-+-+-+-+-+-+-+-+ 1484 | Type = 0 | 1485 +-+-+-+-+-+-+-+-+ 1487 Figure 12: Format of the Pad 1 Option 1489 The Pad1 option is used to insert one or two octets of padding into 1490 the message to enable options alignment. If more than one octet of 1491 padding is required, the PadN option should be used rather than 1492 multiple Pad1 options. 1494 NOTE! the format of the Pad1 option is a special case - it has 1495 neither Option Length nor Option Data fields. 1497 5.6.3. PadN 1499 The PadN option may be present in DIS, DIO, DAO, and DAO-ACK 1500 messages, and its format is as follows: 1502 0 1 2 1503 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 1504 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - - 1505 | Type = 1 | Option Length | 0x00 Padding... 1506 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - - 1508 Figure 13: Format of the Pad N Option 1510 The PadN option is used to insert two or more octets of padding into 1511 the message to enable options alignment. PadN Option data MUST be 1512 ignored by the receiver. 1514 Option Type: 0x01 (to be confirmed by IANA) 1516 Option Length: For N (N > 1) octets of padding, the Option Length 1517 field contains the value N-2. 1519 Option Data: For N (N > 1) octets of padding, the Option Data 1520 consists of N-2 zero-valued octets. 1522 5.6.4. Metric Container 1524 The Metric Container option may be present in DIO messages, and its 1525 format is as follows: 1527 0 1 2 1528 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 1529 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - - 1530 | Type = 2 | Option Length | Metric Data 1531 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- - - - - - - - 1533 Figure 14: Format of the Metric Container Option 1535 The Metric Container is used to report metrics along the DODAG. The 1536 Metric Container may contain a number of discrete node, link, and 1537 aggregate path metrics and constraints specified in 1538 [I-D.ietf-roll-routing-metrics] as chosen by the implementer. 1540 The processing and propagation of the Metric Container is governed by 1541 implementation specific policy functions. 1543 Option Type: 0x02 (to be confirmed by IANA) 1545 Option Length: The Option Length field contains the length in octets 1546 of the Metric Data. 1548 Metric Data: The order, content, and coding of the Metric Container 1549 data is as specified in [I-D.ietf-roll-routing-metrics]. 1551 5.6.5. Route Information 1553 The Route Information option may be present in DIO messages, and is 1554 equivalent in function to the IPv6 ND Route Information option as 1555 defined in [RFC4191]. The format of the option is modified slightly 1556 (Type, Length) in order to be carried as a RPL option as follows: 1558 0 1 2 3 1559 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 1560 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1561 | Type = 3 | Option Length | Prefix Length |Resvd|Prf|Resvd| 1562 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1563 | Route Lifetime | 1564 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1565 | | 1566 . Prefix (Variable Length) . 1567 . . 1568 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1570 Figure 15: Format of the Route Information Option 1572 The Route Information option is used to indicate that connectivity to 1573 the specified destination prefix is available from the DODAG root. 1575 In the event that a RPL Control Message may need to specify 1576 connectivity to more than one destination, the Route Information 1577 option may be repeated. 1579 [RFC4191] should be consulted as the authoritative reference with 1580 respect to the Route Information option. The field descriptions are 1581 transcribed here for convenience: 1583 Option Type: 0x03 (to be confirmed by IANA) 1585 Option Length: Variable, length of the option in octets excluding 1586 the Type and Length fields. Note that this length is expressed 1587 in units of single-octets, unlike in IPv6 ND. 1589 Prefix Length 8-bit unsigned integer. The number of leading bits in 1590 the Prefix that are valid. The value ranges from 0 to 128. 1591 The Prefix field is 0, 8, or 16 octets depending on Length. 1593 Prf: 2-bit signed integer. The Route Preference indicates whether 1594 to prefer the router associated with this prefix over others, 1595 when multiple identical prefixes (for different routers) have 1596 been received. If the Reserved (10) value is received, the 1597 Route Information Option MUST be ignored. 1599 Resvd: Two 3-bit unused fields. They MUST be initialized to zero by 1600 the sender and MUST be ignored by the receiver. 1602 Route Lifetime 32-bit unsigned integer. The length of time in 1603 seconds (relative to the time the packet is sent) that the 1604 prefix is valid for route determination. A value of all one 1605 bits (0xffffffff) represents infinity. 1607 Prefix Variable-length field containing an IP address or a prefix of 1608 an IP address. The Prefix Length field contains the number of 1609 valid leading bits in the prefix. The bits in the prefix after 1610 the prefix length (if any) are reserved and MUST be initialized 1611 to zero by the sender and ignored by the receiver. 1613 Unassigned bits of the Route Information option are reserved. They 1614 MUST be set to zero on transmission and MUST be ignored on reception. 1616 5.6.6. DODAG Configuration 1618 The DODAG Configuration option may be present in DIO messages, and 1619 its format is as follows: 1621 0 1 2 3 1622 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 1623 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1624 | Type = 4 | Option Length | Resvd | PCS | DIOIntDoubl. | 1625 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1626 | DIOIntMin. | DIORedun. | MaxRankIncrease | 1627 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1628 | MinHopRankIncrease | 1629 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1631 Figure 16: Format of the DODAG Configuration Option 1633 The DODAG Configuration option is used to distribute configuration 1634 information for DODAG Operation through the DODAG. 1636 The information communicated in this option is generally static and 1637 unchanging within the DODAG, therefore it is not necessary to include 1638 in every DIO. This information is configured at the DODAG Root and 1639 distributed throughout the DODAG with the DODAG Configuration Option. 1641 Nodes other than the DODAG Root MUST NOT modify this information when 1642 propagating the DODAG Configuration option. This option MAY be 1643 included occasionally by the DODAG Root (as determined by the DODAG 1644 Root), and MUST be included in response to a unicast request, e.g. a 1645 unicast DODAG Information Solicitation (DIS) message. 1647 Option Type: 0x04 (to be confirmed by IANA) 1649 Option Length: 8 bytes 1651 Path Control Size (PCS): 3-bit unsigned integer used to configure 1652 the number of bits that may be allocated to the Path Control 1653 field (see Section 7.1.4.2). 1655 DIOIntervalDoublings: 8-bit unsigned integer used to configure Imax 1656 of the DIO trickle timer (see Section 6.3.1). 1658 DIOIntervalMin: 8-bit unsigned integer used to configure Imin of the 1659 DIO trickle timer (see Section 6.3.1). 1661 DIORedundancyConstant: 8-bit unsigned integer used to configure k of 1662 the DIO trickle timer (see Section 6.3.1). 1664 MaxRankIncrease: 16-bit unsigned integer used to configure 1665 DAGMaxRankIncrease, the allowable increase in rank in support 1666 of local repair. If DAGMaxRankIncrease is 0 then this 1667 mechanism is disabled. 1669 MinHopRankInc 16-bit unsigned integer used to configure 1670 MinHopRankIncrease as described in Section 3.5.2.1. 1672 5.6.7. RPL Target 1674 The RPL Target option may be present in DAO messages, and its format 1675 is as follows: 1677 0 1 2 3 1678 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 1679 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1680 | Type = 5 | Option Length | Reserved | Prefix Length | 1681 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1682 | | 1683 + + 1684 | Target Prefix (Variable Length) | 1685 . . 1686 . . 1687 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1689 Figure 17: Format of the RPL Target Option 1691 The RPL Target Option is used to indicate a target IPv6 address, 1692 prefix, or multicast group that is reachable or queried along the 1693 DODAG. It is used in DIO to identify a resource that the root is 1694 trying to reach, and in a DAO to indicate reachability. It is used 1695 in a DAO message to indicate reachability. A set of one or more 1696 Transit Information options MAY directly follow the Target option in 1697 a DAO message in support of constructing source routes in a non- 1698 storing mode of operation [I-D.hui-6man-rpl-routing-header]. When 1699 the same set of Transit Information options apply equally to a set of 1700 DODAG Target options, the group of Target options MUST appear first, 1701 followed by the Transit Information options which apply to those 1702 Targets. 1704 The RPL Target option may be repeated as necessary to indicate 1705 multiple targets. 1707 Option Type: 0x05 (to be confirmed by IANA) 1709 Option Length: Variable, length of the option in octets excluding 1710 the Type and Length fields. 1712 Prefix Length: 8-bit unsigned integer. Number of valid leading bits 1713 in the IPv6 Prefix. 1715 Target Prefix: Variable-length field identifying an IPv6 destination 1716 address, prefix, or multicast group. The Prefix Length field 1717 contains the number of valid leading bits in the prefix. The 1718 bits in the prefix after the prefix length (if any) are 1719 reserved and MUST be set to zero on transmission and MUST be 1720 ignored on receipt. 1722 5.6.8. Transit Information 1724 The Transit Information option may be present in DAO messages, and 1725 its format is as follows: 1727 0 1 2 3 1728 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 1729 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1730 | Type = 6 | Option Length | Path Sequence | Path Control | 1731 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1732 | Path Lifetime | 1733 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1734 | | 1735 + + 1736 | | 1737 + Parent Address* + 1738 | | 1739 + + 1740 | | 1741 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1743 Figure 18: Format of the Transit Information option 1745 The Transit Information option is used for a node to indicate 1746 attributes for a path to one or more destinations. The destinations 1747 are indicated as by one or more Target options that immediately 1748 precede the Transit Information option(s). 1750 The Transit Information option can used for a node to indicate its 1751 DODAG parents to an ancestor that is collecting DODAG routing 1752 information, typically for the purpose of constructing source routes. 1753 In the non-storing mode of operation this ancestor will be the DODAG 1754 Root, and this option is carried by the DAO message. The option 1755 length is used to determine whether the Parent Address is present or 1756 not. 1758 A non-storing node that has more than one DAO parent MAY include a 1759 Transit Information option for each DAO parent as part of the non- 1760 storing Destination Advertisement operation. The node may code the 1761 Path Control field in order to signal a preference among parents. 1763 One or more Transit Information options MUST be preceded by one or 1764 more RPL Target options. In this manner the RPL Target option 1765 indicates the child node, and the Transit Information option(s) 1766 enumerate the DODAG parents. 1768 A typical non-storing node will use multiple Transit Information 1769 options, and it will send the DAO thus formed to only one parent that 1770 will forward it to the root. A typical storing node with use one 1771 Transit Information option with no parent field, and will send the 1772 DAO thus formed to multiple parents. 1774 Option Type: 0x06 (to be confirmed by IANA) 1776 Option Length: Variable, depending on whether or not Parent Address 1777 is present. 1779 Path-Sequence: 8-bit unsigned integer. When a RPL Target option is 1780 issued by the node that owns the Target Prefix (i.e. in a DAO 1781 message), that node sets the Path-Sequence and increments the 1782 Path-Sequence each time it issues a RPL Target option. 1784 Path Control: 8-bit bitfield. The Path Control field limits the 1785 number of DAO-Parents to which a DAO message advertising 1786 connectivity to a specific destination may be sent, as well as 1787 providing some indication of relative preference. The limit 1788 provides some bound on overall DAO fan-out in the LLN. The 1789 leftmost bit is associated with a path that contains a most- 1790 preferred link, and the subsequent bits are ordered down to the 1791 rightmost bit which is least preferred. 1793 Path Lifetime: 32-bit unsigned integer. The length of time in 1794 seconds (relative to the time the packet is sent) that the 1795 prefix is valid for route determination. A value of all one 1796 bits (0xFFFFFFFF) represents infinity. A value of all zero 1797 bits (0x00000000) indicates a loss of reachability. This is 1798 referred as a No-Path in this document. 1800 Parent Address (optional): IPv6 Address of the DODAG Parent of the 1801 node originally issuing the Transit Information Option. This 1802 field may not be present, as according to the DODAG Mode of 1803 Operation and indicated by the Transit Information option 1804 length. 1806 Unassigned bits of the Transit Information option are reserved. They 1807 MUST be set to zero on transmission and MUST be ignored on reception. 1809 5.6.9. Solicited Information 1811 The Solicited Information option may be present in DIS messages, and 1812 its format is as follows: 1814 0 1 2 3 1815 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 1816 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1817 | Type = 7 | Option Length | RPLInstanceID |V|I|D| Rsvd | 1818 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1819 | | 1820 + + 1821 | | 1822 + DODAGID + 1823 | | 1824 + + 1825 | | 1826 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1827 | Version | 1828 +-+-+-+-+-+-+-+-+ 1830 Figure 19: Format of the Solicited Information Option 1832 The Solicited Information option is used for a node to request a 1833 subset of neighboring nodes that meet the specified criteria to 1834 respond to a DIS message. 1836 The Solicited Information option may specify a number of predicate 1837 criteria to be matched by a receiving node. If a node receiving a 1838 multicast DIS message containing a Solicited Information option 1839 matches ALL of the predicates, then it MUST reset its trickle timer 1840 in order to trigger a DIO response to the DIS message. When a node 1841 receives a DIS message containing a Solicited information option, and 1842 the DIS message is unicast OR the node does not match ALL the 1843 predicates, then the node MUST NOT reset the trickle timer. 1845 Option Type: 0x07 (to be confirmed by IANA) 1847 Option Length: 19 bytes 1849 Control Field: The Solicited Information option Control Field has 1850 three flags: 1852 V: If the V flag is set then the Version field is valid and 1853 a node should only respond if its DODAGVersionNumber 1854 matches the requested version. If the V flag is clear 1855 then the Version field is not valid and the Version field 1856 MUST be set to zero on transmission and ignored upon 1857 receipt. 1859 I: If the I flag is set then the RPLInstanceID field is 1860 valid and a node should only respond if it matches the 1861 requested RPLInstanceID. If the I flag is clear then the 1862 RPLInstanceID field is not valid and the RPLInstanceID 1863 field MUST be set to zero on transmission and ignored 1864 upon receipt. 1866 D: If the D flag is set then the DODAGID field is valid and 1867 a node should only respond if it matches the requested 1868 DODAGID. If the D flag is clear then the DODAGID field 1869 is not valid and the DODAGID field MUST be set to zero on 1870 transmission and ignored upon receipt. 1872 Version: 8-bit unsigned integer containing the DODAG Version number 1873 that is being solicited when valid. 1875 RPLInstanceID: 8-bit unsigned integer containing the RPLInstanceID 1876 that is being solicited when valid. 1878 DODAGID: 128-bit unsigned integer containing the DODAGID that is 1879 being solicited when valid. 1881 Unassigned bits of the Solicited Information option are reserved. 1882 They MUST be set to zero on transmission and MUST be ignored on 1883 reception. 1885 5.6.10. Prefix Information 1887 The Prefix Information option may be present in DIO messages, and is 1888 equivalent in function to the IPv6 ND Prefix Information option as 1889 defined in [RFC4861]. The format of the option is modified slightly 1890 (Type, Length) in order to be carried as a RPL option as follows: 1892 0 1 2 3 1893 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 1894 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1895 | Type = 8 | Option Length | Prefix Length |L|A| Reserved1 | 1896 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1897 | Valid Lifetime | 1898 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1899 | Preferred Lifetime | 1900 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1901 | Reserved2 | 1902 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1903 | | 1904 + + 1905 | | 1906 + Prefix + 1907 | | 1908 + + 1909 | | 1910 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1912 Figure 20: Format of the Prefix Information Option 1914 The Prefix Information option may be used to distribute the prefix in 1915 use inside the DODAG, e.g. for address autoconfiguration. 1917 [RFC4861] should be consulted as the authoritative reference with 1918 respect to the Prefix Information option. The field descriptions are 1919 transcribed here for convenience: 1921 Option Type: 0x08 (to be confirmed by IANA) 1923 Option Length: 30. Note that this length is expressed in units of 1924 single-octets, unlike in IPv6 ND. 1926 Prefix Length 8-bit unsigned integer. The number of leading bits in 1927 the Prefix that are valid. The value ranges from 0 to 128. 1928 The prefix length field provides necessary information for on- 1929 link determination (when combined with the L flag in the prefix 1930 information option). It also assists with address 1931 autoconfiguration as specified in [RFC4862], for which there 1932 may be more restrictions on the prefix length. 1934 L 1-bit on-link flag. When set, indicates that this prefix can 1935 be used for on-link determination. When not set the 1936 advertisement makes no statement about on-link or off-link 1937 properties of the prefix. In other words, if the L flag is not 1938 set a host MUST NOT conclude that an address derived from the 1939 prefix is off-link. That is, it MUST NOT update a previous 1940 indication that the address is on-link. 1942 A 1-bit autonomous address-configuration flag. When set 1943 indicates that this prefix can be used for stateless address 1944 configuration as specified in [RFC4862]. 1946 Reserved1 6-bit unused field. It MUST be initialized to zero by the 1947 sender and MUST be ignored by the receiver. 1949 Valid Lifetime 32-bit unsigned integer. The length of time in 1950 seconds (relative to the time the packet is sent) that the 1951 prefix is valid for the purpose of on-link determination. A 1952 value of all one bits (0xffffffff) represents infinity. The 1953 Valid Lifetime is also used by [RFC4862]. 1955 Preferred Lifetime 32-bit unsigned integer. The length of time in 1956 seconds (relative to the time the packet is sent) that 1957 addresses generated from the prefix via stateless address 1958 autoconfiguration remain preferred [RFC4862]. A value of all 1959 one bits (0xffffffff) represents infinity. See [RFC4862]. 1960 Note that the value of this field MUST NOT exceed the Valid 1961 Lifetime field to avoid preferring addresses that are no longer 1962 valid. 1964 Reserved2 This field is unused. It MUST be initialized to zero by 1965 the sender and MUST be ignored by the receiver. 1967 Prefix An IP address or a prefix of an IP address. The Prefix 1968 Length field contains the number of valid leading bits in the 1969 prefix. The bits in the prefix after the prefix length are 1970 reserved and MUST be initialized to zero by the sender and 1971 ignored by the receiver. A router SHOULD NOT send a prefix 1972 option for the link-local prefix and a host SHOULD ignore such 1973 a prefix option. 1975 Unassigned bits of the Prefix Information option are reserved. They 1976 MUST be set to zero on transmission and MUST be ignored on reception. 1978 6. Upward Routes 1980 This section describes how RPL discovers and maintains upward routes. 1981 It describes the use of DODAG Information Objects (DIOs), the 1982 messages used to discover and maintain these routes. It specifies 1983 how RPL generates and responds to DIOs. It also describes DODAG 1984 Information Solicitation (DIS) messages, which are used to trigger 1985 DIO transmissions. 1987 6.1. DIO Base Rules 1989 1. If the 'A' flag of a DIO Base is cleared, the 'T' flag MUST also 1990 be cleared. 1992 2. For the following DIO Base fields, a node that is not a DODAG 1993 root MUST advertise the same values as its preferred DODAG parent 1994 (defined in Section 6.2.1). Therefore, if a DODAG root does not 1995 change these values, every node in a route to that DODAG root 1996 eventually advertises the same values for these fields. These 1997 fields are: 1998 1. Grounded (G) 1999 2. Destination Advertisement Supported (A) 2000 3. Destination Advertisement Trigger (T) 2001 4. DAGPreference (Prf) 2002 5. Version 2003 6. RPLInstanceID 2004 7. DODAGID 2006 3. A node MAY update the following fields at each hop: 2007 1. Destination Advertisements Stored (S) 2008 2. DAGRank 2009 3. DTSN 2011 4. The DODAGID field each root sets MUST be unique within the RPL 2012 Instance. 2014 6.2. Upward Route Discovery and Maintenance 2016 Upward route discovery allows a node to join a DODAG by discovering 2017 neighbors that are members of the DODAG of interest and identifying a 2018 set of parents. The exact policies for selecting neighbors and 2019 parents is implementation-dependent and driven by the OF. This 2020 section specifies the set of rules those policies must follow for 2021 interoperability. 2023 6.2.1. Neighbors and Parents within a DODAG Version 2025 RPL's upward route discovery algorithms and processing are in terms 2026 of three logical sets of link-local nodes. First, the candidate 2027 neighbor set is a subset of the nodes that can be reached via link- 2028 local multicast. The selection of this set is implementation- 2029 dependent and OF-dependent. Second, the parent set is a restricted 2030 subset of the candidate neighbor set. Finally, the preferred parent, 2031 a set of size one, is an element of the parent set that is the 2032 preferred next hop in upward routes. 2034 More precisely: 2036 1. The DODAG parent set MUST be a subset of the candidate neighbor 2037 set. 2039 2. A DODAG root MUST have a DODAG parent set of size zero. 2041 3. A node that is not a DODAG root MAY maintain a DODAG parent set 2042 of size greater than or equal to one. 2044 4. A node's preferred DODAG parent MUST be a member of its DODAG 2045 parent set. 2047 5. A node's rank MUST be greater than all elements of its DODAG 2048 parent set. 2050 6. When Neighbor Unreachability Detection (NUD), or an equivalent 2051 mechanism, determines that a neighbor is no longer reachable, a 2052 RPL node MUST NOT consider this node in the candidate neighbor 2053 set when calculating and advertising routes until it determines 2054 that it is again reachable. Routes through an unreachable 2055 neighbor MUST be removed from the routing table. 2057 These rules ensure that there is a consistent partial order on nodes 2058 within the DODAG. As long as node ranks do not change, following the 2059 above rules ensures that every node's route to a DODAG root is loop- 2060 free, as rank decreases on each hop to the root. The OF can guide 2061 candidate neighbor set and parent set selection, as discussed in 2062 [I-D.ietf-roll-routing-metrics] and [I-D.ietf-roll-of0]. 2064 6.2.2. Neighbors and Parents across DODAG Versions 2066 The above rules govern a single DODAG version. The rules in this 2067 section define how RPL operates when there are multiple DODAG 2068 versions: 2070 6.2.2.1. DODAG Version 2072 1. The tuple (RPLInstanceID, DODAGID, DODAGVersionNumber) uniquely 2073 defines a DODAG Version. Every element of a node's DODAG parent 2074 set, as conveyed by the last heard DIO message from each DODAG 2075 parent, MUST belong to the same DODAG version. Elements of a 2076 node's candidate neighbor set MAY belong to different DODAG 2077 Versions. 2079 2. A node is a member of a DODAG version if every element of its 2080 DODAG parent set belongs to that DODAG version, or if that node 2081 is the root of the corresponding DODAG. 2083 3. A node MUST NOT send DIOs for DODAG versions of which it is not a 2084 member. 2086 4. DODAG roots MAY increment the DODAGVersionNumber that they 2087 advertise and thus move to a new DODAG version. When a DODAG 2088 root increments its DODAGVersionNumber, it MUST follow the 2089 conventions of Serial Number Arithmetic as described in 2090 [RFC1982]. 2092 5. Within a given DODAG, a node that is a not a root MUST NOT 2093 advertise a DODAGVersionNumber higher than the highest 2094 DODAGVersionNumber it has heard. Higher is defined as the 2095 greater-than operator in [RFC1982]. 2097 6. Once a node has advertised a DODAG version by sending a DIO, it 2098 MUST NOT be member of a previous DODAG version of the same DODAG 2099 (i.e. with the same RPLInstanceID, the same DODAGID, and a lower 2100 DODAGVersionNumber). Lower is defined as the less-than operator 2101 in [RFC1982]. 2103 Within a particular implementation, a DODAG root may increment the 2104 DODAGVersionNumber periodically, at a rate that depends on the 2105 deployment, in order to trigger a global reoptimization of the DODAG. 2106 In other implementations, loop detection may be considered sufficient 2107 to solve routing issues by triggering local repair mechanisms, and 2108 the DODAG root may increment the DODAGVersionNumber only upon 2109 administrative intervention. Another possibility is that nodes 2110 within the LLN have some means by which they can signal detected 2111 routing inconsistencies or suboptimalities to the DODAG root, in 2112 order to request an on-demand DODAGVersionNumber increment (i.e. 2113 request a global repair of the DODAG). Note that such a mechanism is 2114 for further study and out of the scope of this document. 2116 When the DODAG parent set becomes empty on a node that is not a root, 2117 (i.e. the last parent has been removed, causing the node to no longer 2118 be associated with that DODAG), then the DODAG information should not 2119 be suppressed until after the expiration of an implementation- 2120 specific local timer in order to observe if the DODAGVersionNumber 2121 has been incremented, should any new parents appear for the DODAG. 2122 This will help protect against the possibility of loops that may 2123 occur of that node were to inadvertently rejoin the old DODAG version 2124 in its own prior sub-DODAG. 2126 As the DODAGVersionNumber is incremented, a new DODAG Version spreads 2127 outward from the DODAG root. Thus a parent that advertises the new 2128 DODAGVersionNumber cannot possibly belong to the sub-DODAG of a node 2129 that still advertises an older DODAGVersionNumber. A node may safely 2130 add such a parent, without risk of forming a loop, without regard to 2131 its relative rank in the prior DODAG Version. This is equivalent to 2132 jumping to a different DODAG. 2134 As a node transitions to new DODAG Versions as a consequence of 2135 following these rules, the node will be unable to advertise the 2136 previous DODAG Version (prior DODAGVersionNumber) once it has 2137 committed to advertising the new DODAG Version. 2139 During transition to a new DODAG Version, a node may decide to 2140 forward packets via 'future parents' that belong to the same DODAG 2141 (same RPLInstanceID and DODAGID), but are observed to advertise a 2142 more recent (incremented) DODAGVersionNumber. In that case, the node 2143 MUST act as a leaf with regard to the new version for the purpose of 2144 loop detection as specified in Section 8.2. 2146 6.2.2.2. DODAG Roots 2148 1. A DODAG root that does not have connectivity to the set of 2149 addresses described as application-level goals, MUST NOT set the 2150 Grounded bit. 2152 2. A DODAG root MUST advertise a rank of ROOT_RANK. 2154 3. A node whose DODAG parent set is empty MAY become the DODAG root 2155 of a floating DODAG. It MAY also set its DAGPreference such that 2156 it is less preferred. 2158 An LLN node that is a goal for the Objective Function is the root of 2159 its own grounded DODAG, at rank ROOT_RANK. 2161 In a deployment that uses a backbone link to federate a number of LLN 2162 roots, it is possible to run RPL over that backbone and use one 2163 router as a "backbone root". The backbone root is the virtual root 2164 of the DODAG, and exposes a rank of BASE_RANK over the backbone. All 2165 the LLN roots that are parented to that backbone root, including the 2166 backbone root if it also serves as LLN root itself, expose a rank of 2167 ROOT_RANK to the LLN, and are part of the same DODAG, coordinating 2168 DODAGVersionNumber and other DODAG root determined parameters with 2169 the virtual root over the backbone. 2171 6.2.2.3. DODAG Selection 2173 The DODAGPreference (Prf) provides an administrative mechanism to 2174 engineer the self-organization of the LLN, for example indicating the 2175 most preferred LBR. If a node has the option to join a more 2176 preferred DODAG while still meeting other optimization objectives, 2177 then the node will generally seek to join the more preferred DODAG as 2178 determined by the OF. All else being equal, it is left to the 2179 implementation to determine which DODAG is most preferred, possibly 2180 based on additional criteria beyond Prf and the OF. 2182 6.2.2.4. Rank and Movement within a DODAG Version 2184 1. A node MUST NOT advertise a rank less than or equal to any member 2185 of its parent set within the DODAG Version. 2187 2. A node MAY advertise a rank lower than its prior advertisement 2188 within the DODAG Version. 2190 3. Let L be the lowest rank within a DODAG version that a given node 2191 has advertised. Within the same DODAG Version, that node MUST 2192 NOT advertise an effective rank higher than L + 2193 DAGMaxRankIncrease. INFINITE_RANK is an exception to this rule: 2194 a node MAY advertise an INFINITE_RANK at any time. (This rule 2195 corresponds to a limited rank increase for the purpose of local 2196 repair within the DODAG Version.) 2198 4. A node MAY, at any time, choose to join a different DODAG within 2199 a RPL Instance. Such a join has no rank restrictions, unless 2200 that different DODAG is a DODAG Version of which this node has 2201 previously been a member, in which case the rule of the previous 2202 bullet (3) must be observed. Until a node transmits a DIO 2203 indicating its new DODAG membership, it MUST forward packets 2204 along the previous DODAG. 2206 5. A node MAY, at any time after hearing the next DODAGVersionNumber 2207 Version advertised from suitable DODAG parents, choose to migrate 2208 to the next DODAG Version within the DODAG. 2210 Conceptually, an implementation is maintaining a DODAG parent set 2211 within the DODAG Version. Movement entails changes to the DODAG 2212 parent set. Moving up does not present the risk to create a loop but 2213 moving down might, so that operation is subject to additional 2214 constraints. 2216 When a node migrates to the next DODAG Version, the DODAG parent and 2217 sibling sets need to be rebuilt for the new version. An 2218 implementation could defer to migrate for some reasonable amount of 2219 time, to see if some other neighbors with potentially better metrics 2220 but higher rank announce themselves. Similarly, when a node jumps 2221 into a new DODAG it needs to construct new DODAG parent/sibling sets 2222 for this new DODAG. 2224 When a node moves to improve its position, it must conceptually 2225 abandon all DODAG parents and siblings with a rank larger than 2226 itself. As a consequence of the movement it may also add new 2227 siblings. Such a movement may occur at any time to decrease the 2228 rank, as per the calculation indicated by the OF. Maintenance of the 2229 parent and sibling sets occurs as the rank of candidate neighbors is 2230 observed as reported in their DIOs. 2232 If a node needs to move down a DODAG that it is attached to, causing 2233 the rank to increase, then it MAY poison its routes and delay before 2234 moving as described in Section 6.2.2.5. 2236 6.2.2.5. Poisoning a Broken Path 2238 1. A node MAY poison, in order to avoid being used as an ancestor by 2239 the nodes in its sub-DODAG, by advertising an effective rank of 2240 INFINITE_RANK and resetting the associated DIO trickle timer to 2241 cause this INFINITE_RANK to be announced promptly. 2243 2. The node MAY advertise an effective rank of INFINITE_RANK for an 2244 arbitrary number of DIO timer events, before announcing a new 2245 rank. 2247 3. As per Section 6.2.2.4, the node MUST advertise INFINITE_RANK 2248 within the DODAG version in which it participates, if its 2249 revision in rank would exceed the maximum rank increase. 2251 An implementation may choose to employ this poisoning mechanism when 2252 a node loses all of its current parents, i.e. the set of DODAG 2253 parents becomes depleted, and it can not jump to an alternate DODAG. 2254 An alternate mechanism is to form a floating DODAG. 2256 The motivation for delaying announcement of the revised route through 2257 multiple DIO events is to (i) increase tolerance to DIO loss, (ii) 2258 allow time for the poisoning action to propagate, and (iii) to 2259 develop an accurate assessment of its new rank. Such gains are 2260 obtained at the expense of potentially increasing the delay before 2261 portions of the network are able to re-establish upwards routes. 2262 Path redundancy in the DODAG reduces the significance of either 2263 effect, since children with alternate parents should be able to 2264 utilize those alternates and retain their rank while the detached 2265 parent re-establishes its rank. 2267 Although an implementation may advertise INFINITE_RANK for the 2268 purposes of poisoning, it is not expected to be equivalent to setting 2269 the rank to INFINITE_RANK, and an implementation would likely retain 2270 its rank value prior to the poisoning in some form, for purpose of 2271 maintaining its effective position within (L + DAGMaxRankIncrease). 2273 6.2.2.6. Detaching 2275 1. A node unable to stay connected to a DODAG within a given DODAG 2276 version MAY detach from this DODAG version. A node that detaches 2277 becomes root of its own floating DODAG and SHOULD immediately 2278 advertise this new situation in a DIO as an alternate to 2279 poisoning. 2281 6.2.2.7. Following a Parent 2283 1. If a node receives a DIO from one of its DODAG parents, 2284 indicating that the parent has left the DODAG, that node SHOULD 2285 stay in its current DODAG through an alternative DODAG parent, if 2286 possible. It MAY follow the leaving parent. 2288 A DODAG parent may have moved, migrated to the next DODAG Version, or 2289 jumped to a different DODAG. A node should give some preference to 2290 remaining in the current DODAG, if possible via an alternate parent, 2291 but ought to follow the parent if there are no other options. 2293 6.2.3. DIO Message Communication 2295 When an DIO message is received, the receiving node must first 2296 determine whether or not the DIO message should be accepted for 2297 further processing, and subsequently present the DIO message for 2298 further processing if eligible. 2300 1. If the DIO message is malformed, then the DIO message is not 2301 eligible for further processing and MUST be silently discarded. 2302 A RPL implementation MAY log the reception of a malformed DIO 2303 message. 2305 2. If the sender of the DIO message is a member of the candidate 2306 neighbor set, then the DIO is eligible for further processing. 2308 6.2.3.1. DIO Message Processing 2310 As DIO messages are received from candidate neighbors, the neighbors 2311 may be promoted to DODAG parents by following the rules of DODAG 2312 discovery as described in Section 6.2. When a node places a neighbor 2313 into the DODAG parent set, the node becomes attached to the DODAG 2314 through the new DODAG parent node. 2316 The most preferred parent should be used to restrict which other 2317 nodes may become DODAG parents. Some nodes in the DODAG parent set 2318 may be of a rank less than or equal to the most preferred DODAG 2319 parent. (This case may occur, for example, if an energy constrained 2320 device is at a lesser rank but should be avoided as per an 2321 optimization objective, resulting in a more preferred parent at a 2322 greater rank). 2324 6.3. DIO Transmission 2326 RPL nodes transmit DIOs using a Trickle timer 2327 ([I-D.ietf-roll-trickle]). A DIO from a sender with a lower DAGRank 2328 that causes no changes to the recipient's parent set, preferred 2329 parent, or Rank SHOULD be considered consistent with respect to the 2330 Trickle timer. 2332 The following packets and events MUST be considered inconsistencies 2333 with respect to the Trickle timer, and cause the Trickle timer to 2334 reset: 2336 o When a node detects an inconsistency when forwarding a packet, as 2337 detailed in Section 8.2. 2339 o When a node receives a multicast DIS message whose constraints 2340 (Solicited Information) it satisfies. 2342 o When a node joins a new DODAG Version (e.g. by updating its 2343 DODAGVersionNumber, joining a new RPL Instance, etc.) 2345 Note that this list is not exhaustive, and an implementation MAY 2346 consider other messages or events to be inconsistencies. 2348 If a node receives a unicast DIS message whose constraints (Solicited 2349 Information) it satisfies, it MUST unicast a DIO in response, and 2350 this DIO MUST include the RPL instance's DODAG Configuration object. 2352 6.3.1. Trickle Parameters 2354 The configuration parameters of the trickle timer are specified as 2355 follows: 2357 Imin: learned from the DIO message as (2^DIOIntervalMin)ms. The 2358 default value of DIOIntervalMin is DEFAULT_DIO_INTERVAL_MIN. 2360 Imax: learned from the DIO message as DIOIntervalDoublings. The 2361 default value of DIOIntervalDoublings is 2362 DEFAULT_DIO_INTERVAL_DOUBLINGS. 2364 k: learned from the DIO message as DIORedundancyConstant. The 2365 default value of DIORedundancyConstant is 2366 DEFAULT_DIO_REDUNDANCY_CONSTANT. In RPL, when k has the value 2367 of 0x00 this is to be treated as a redundancy constant of 2368 infinity in RPL, i.e. Trickle never suppresses messages. 2370 6.4. DODAG Selection 2372 The DODAG selection is implementation and OF dependent. Nodes SHOULD 2373 prefer to join DODAGs for RPLInstanceIDs advertising OCPs and 2374 destinations compatible with their implementation specific 2375 objectives. In order to limit erratic movements, and all metrics 2376 being equal, nodes SHOULD keep their previous selection. Also, nodes 2377 SHOULD provide a means to filter out a parent whose availability is 2378 detected as fluctuating, at least when more stable choices are 2379 available. 2381 When connection to a grounded DODAG is not possible or preferable for 2382 security or other reasons, scattered DODAGs MAY aggregate as much as 2383 possible into larger DODAGs in order to allow connectivity within the 2384 LLN. 2386 A node SHOULD verify that bidirectional connectivity and adequate 2387 link quality is available with a candidate neighbor before it 2388 considers that candidate as a DODAG parent. 2390 6.5. Operation as a Leaf Node 2392 In some cases a RPL node may attach to a DODAG as a leaf node only. 2393 One example of such a case is when a node does not understand the RPL 2394 Instance's OF or advertised path metric. A leaf node does not extend 2395 DODAG connectivity but still needs to advertise its presence using 2396 DIOs. A node operating as a leaf node must obey the following rules: 2398 1. It MUST NOT transmit DIOs containing the DAG Metric Container. 2400 2. Its DIOs must advertise a DAGRank of INFINITE_RANK. 2402 3. It MAY transmit unicast DAOs as described in Section 7.1. 2404 4. It MAY transmit multicast DAOs to the '1 hop' neighborhood as 2405 described in Section 7.1.9. 2407 6.6. Administrative Rank 2409 In some cases it might be beneficial to adjust the rank advertised by 2410 a node beyond that computed by the OF based on some implementation 2411 specific policy and properties of the node. For example, a node that 2412 has limited battery should be a leaf unless there is no other choice, 2413 and may then augment the rank computation specified by the OF in 2414 order to expose an exaggerated rank. 2416 7. Downward Routes 2418 This section describes how RPL discovers and maintains downward 2419 routes. The use of messages containing the Destination Advertisement 2420 Object (DAO), used to construct downward routes, are described. The 2421 downward routes are necessary in support of P2MP flows, from the 2422 DODAG roots toward the leaves. It specifies non-storing and storing 2423 behavior of nodes with respect to DAO messaging and DAO routing table 2424 entries. Nodes, as according to their resources and the 2425 implementation, may selectively store routing table entries learned 2426 from DAO messages, or may instead propagate the DAO information 2427 upwards and independently source local topology information in a new 2428 DAO message. information. A further optimization is described 2429 whereby DAO messages may be used to populate routing table entries 2430 for the '1-hop' neighbors, which may be useful in some cases as a 2431 shortcut for P2P flows. 2433 7.1. Downward Route Discovery and Maintenance 2435 7.1.1. Overview 2437 Destination Advertisement operation produces DAO messages that flow 2438 up the DODAG, provisioning downward routing state for destination 2439 prefixes available in the sub-DODAG of the DODAG root, and possibly 2440 other nodes. The routing state provisioned with this mechanism is in 2441 the form of soft-state routing table entries. DAO operation is 2442 presently defined in two distinct modes of operation, non-storing and 2443 storing, and allowance is made for future expansion. 2445 Destination Advertisement may or may not be enabled over a DODAG 2446 rooted at a DODAG root. This is an a priori configuration determined 2447 by the implementation/deployment and not generally changed during the 2448 operation of the RPL LLN. 2450 Destination Advertisement may be configured to operate in either a 2451 storing or non-storing mode, as reported in the MOP in the DIO 2452 message. Every node in the network participating in Destination 2453 Advertisement must behave consistently with that configured mode of 2454 operation, or alternately behave only as a leaf node. Hybrid or 2455 mixed-mode operation is not currently specified. 2457 When Destination Advertisement is enabled: 2459 1. The RPL Instance will be configured a priori as appropriate to 2460 satisfy the application to operate in either non-storing or 2461 storing mode. 2463 2. All nodes who join the DODAG MUST abide with the MOP setting from 2464 the root. Nodes that would not have the capability to fully 2465 participate as a router (e.g. to operate as a storing node) can 2466 still join as a leaf (i.e. host). 2468 3. In storing mode operation, all non-root nodes are expected to 2469 either store routing table entries for ALL destinations learned 2470 from DAO operation, or to act as a leaf node only. 2472 4. In non-storing mode operation, no node other than the DODAG Root 2473 is expected to store routing table entries learned from DAO 2474 messages. Each node is only responsible to report its own set of 2475 parents to the DODAG Root. 2477 5. DODAG roots nodes SHOULD be capable to store routing table 2478 entries learned from DAO operation when the RPL Instance is 2479 operated in a non-storing mode. 2481 6. The mode of operation in the RPL Instance is signaled from the 2482 DODAG Root in the MOP control field of the DIO message. 2484 7.1.2. Mode of Operation 2486 o DAO Operation may not be required for all use cases. 2488 o Some applications may only need support for collection/upward/MP2P 2489 flow with no acknowledgement/reciprocal traffic. 2491 o Some DODAGs may not support DAO Operation, which could mean that 2492 DAO Operation is wasteful overhead. 2494 o As a special case, multicast DAO operation may be used to populate 2495 'one-hop' neighborhood routing table entries, and is distinct from 2496 the unicast DAO operation used to establish downward routes along 2497 the DODAG. This special case is an exception to the RPL Instance 2498 mode of operation as well. 2500 1. The 'A' flag in the DIO as conveyed from the DODAG root serves to 2501 enable/disable DAO operation over the entire DODAG. This flag 2502 should be administratively provisioned a priori at the DODAG root 2503 as a function of the implementation/deployment and not tend to 2504 change. 2506 2. When DAO Operation is disabled, a node MUST NOT emit DAO 2507 messages. 2509 3. When DAO Operation is disabled, a node MAY ignore the MOP field. 2511 4. When DAO Operation is disabled, a node MAY ignore received DAO 2512 messages. 2514 7.1.3. Destination Advertisement Parents 2516 o Nodes will select a subset of their DODAG Parents to whom DAO 2517 messages will be sent 2519 * This subset is the set of 'DAO Parents' 2521 * Each DAO parent MUST be a DODAG Parent. (Not all DODAG parents 2522 need to be DAO parents). 2524 o The selection of DAO parents is implementation specific and may be 2525 based on selecting the DODAG Parents that offer the best upwards 2526 cost (as opposed to downwards or mixed), as determined by the 2527 metrics in use and the Objective Function. 2529 o When DAO messages are unicast to the DAO Parent, the identity of 2530 the DAO Parent (DODAGID and DODAGVersionNumber) combined with the 2531 RPLInstanceID in the DAO message unambiguously associates the DAO 2532 message, and thus the particular destination prefix, with a DODAG 2533 Version. 2535 7.1.4. DAO Operation on Storing Nodes 2537 7.1.4.1. DAO Routing Table Entry 2539 A DAO Routing Table Entry conceptually contains the following 2540 elements: 2542 o Advertising Neighbor Information 2543 * IPv6 Address 2544 * Interface ID 2545 o To which DAO Parents has this entry been reported 2546 o Retry Counter 2547 o Logical equivalent of DAO Content: 2548 * DAO Sequence 2549 * DAO Lifetime 2550 * DAO Path Control (as learned from each child) 2551 * Destination Prefix (or Address or Mcast Group) 2553 The DAO Routing Table Entry is logically associated with the 2554 following states: 2556 CONNECTED This entry is 'owned' by the node - it is manually 2557 configured and is considered as a 'self' entry for DAO 2558 Operation 2560 REACHABLE This entry has been reported from a neighbor of the node. 2561 This state includes the following substates: 2563 CONFIRMED This entry is active, newly validated, and 2564 usable 2566 PENDING This entry is active, awaiting validation, and 2567 usable. A Retry Counter is associated with 2568 this substate 2570 UNREACHABLE This entry is being cleaned up. This entry may be 2571 suppressed when the cleanup process is complete. 2573 When an attempt is to be made to report the DAO entry to DAO Parents, 2574 the DAO Entry record is logically marked to indicate that an attempt 2575 has not yet been made for each parent. As the unicast attempts are 2576 completed for each parent, this mark may be cleared. This mechanism 2577 may serve to limit DAO entry updates for each parent to a subset that 2578 needs to be reported. 2580 7.1.4.1.1. DAO Routing Table Entry Management 2582 +---------------------------------+ 2583 | | 2584 | REACHABLE | +-------------+ 2585 | | | | 2586 | +-----------+ | | CONNECTED | 2587 (*)----------->| |-------+ | | | 2588 | | Confirmed | | | +-------------+ 2589 | +-->| |---+ | | 2590 | | +-----------+ | | | 2591 | | | | | 2592 | | | | | 2593 | | | | | 2594 | | +-----------+ | | | +-------------+ 2595 | | | |<--+ +-------->| | 2596 | +---| Pending | | | UNREACHABLE | 2597 | | |---------------->| |--->(*) 2598 | +-----------+ | +-------------+ 2599 | | 2600 +---------------------------------+ 2601 DAO Routing Table Entry FSM 2603 7.1.4.1.1.1. Operation in the CONNECTED state 2605 1. CONNECTED DAO entries are to be provisioned outside of the 2606 context of RPL, e.g. through a management API. An implementation 2607 SHOULD provide a means to provision/manage CONNECTED DAO entries, 2608 including whether they are to be redistributed in RPL. 2610 7.1.4.1.1.2. Operation in the REACHABLE state 2612 1. When a REACHABLE(*) entry times out, i.e. the DAO Lifetime has 2613 elapsed, the entry MUST be placed into the UNREACHABLE state and 2614 No-Path SHOULD be scheduled to send to the node's DAO Parents. 2616 2. When a No-Path for a REACHABLE(*) entry is received with a newer 2617 DAO Sequence Number, the entry MUST be placed into the 2618 UNREACHABLE state and No-Path SHOULD be scheduled to send to the 2619 node's DAO Parents. 2621 3. When a REACHABLE(*) entry is to be removed because NUD or 2622 equivalent has determined that the next-hop neighbor is no longer 2623 reachable, the entry MUST be placed into the UNREACHABLE state 2624 and No-Path SHOULD be scheduled to send to the node's DAO 2625 Parents. 2627 4. When a REACHABLE(*) entry is to be removed because an associated 2628 Forwarding Error has been returned by the next-hop neighbor, the 2629 entry MUST be placed into the UNREACHABLE state and No-Path 2630 SHOULD be scheduled to send to the node's DAO Parents. 2632 5. When a DAO (or No-Path) for a REACHABLE(*) entry is received with 2633 an older or unchanged DAO Sequence Number, then the DAO (or No- 2634 Path) SHOULD be ignored and the associated entry MUST NOT be 2635 updated with the stale information. 2637 7.1.4.1.1.2.1. REACHABLE(Confirmed) 2639 1. When a DAO for a previously unknown (or UNREACHABLE) destination 2640 is received and is to be stored, it MUST be entered into the 2641 routing table in the REACHABLE(Confirmed) state, and a DAO SHOULD 2642 be scheduled to send to the node's DAO Parents. 2644 2. When a DAO for a REACHABLE(Confirmed) entry is received with a 2645 newer DAO Sequence Number, the entry MUST be updated with the 2646 logical equivalent of the DAO contents and a DAO SHOULD be 2647 scheduled to send to the node's DAO Parents. 2649 3. When a DAO for a REACHABLE(Confirmed) entry is expected, e.g. 2650 because a DIO to request a DAO refresh is sent, then the DAO 2651 entry MUST be placed in the REACHABLE(Pending) state and the 2652 associated Retry Counter MUST be set to 0. 2654 7.1.4.1.1.2.2. REACHABLE(Pending) 2656 1. When a DAO for a REACHABLE(Pending) entry is received with a 2657 newer DAO Sequence Number, the entry MUST be updated with the 2658 logical equivalent of the DAO contents and the entry MUST be 2659 placed in the REACHABLE(Confirmed) state. 2661 2. When a DAO for a REACHABLE(Pending) entry is expected, e.g. 2662 because DAO has (again) been triggered with respect to that 2663 neighbor, then the associated Retry Counter MUST be incremented. 2665 3. When the associated Retry Counter for a REACHABLE(Pending) entry 2666 reaches a maximum threshold, the entry MUST be placed into the 2667 UNREACHABLE state and No-Path SHOULD be scheduled to send to the 2668 node's DAO Parents. 2670 7.1.4.1.1.3. Operation in the UNREACHABLE state 2672 1. An implementation SHOULD bound the time that the entry is 2673 allocated in the UNREACHABLE state. Upon the equivalent expiry 2674 of the related timer (RemoveTimer), the entry SHOULD be 2675 suppressed. 2677 2. While the entry is in the UNREACHABLE state a node SHOULD make a 2678 reasonable attempt to report a No-Path to each of the DAO 2679 parents. 2681 3. When the node has completed an attempt to report a No-Path to 2682 each of the DAO parents, the entry SHOULD be suppressed. 2684 7.1.4.2. Storing Mode DAO Message and Path Control 2686 In the storing mode of operation, a DAO message from a node will 2687 contain one or more Target Options, each Target Option specifying 2688 either a CONNECTED destination or a destination in the sub-DODAG of 2689 the node. 2691 For each attempt made to report the DAO entry to a set of DAO 2692 parents, the Path Control field will be constructed as follows: 2694 1. The size of the path control field will be specified by the PCS 2695 control field of the DODAG Configuration Option. The default 2696 value is DEFAULT_PATH_CONTROL_SIZE. 2698 2. For each unique destination to be reported that is CONNECTED, the 2699 logical equivalent of a path control bitmap that is the size of 2700 the path control field shall be initialized with the leftmost 2701 bits set, where the number of leftmost bits corresponds to the 2702 size of the path control field as specified by PCS. 2704 3. For each unique destination to be reported that is not CONNECTED, 2705 i.e. that destination is contained in the node's sub-DODAG, the 2706 logical equivalent of a path control bitmap that is the size of 2707 the path control field shall be initialized by ORing the content 2708 of all of the Path Control fields received in DAO messages from 2709 the node's children for that destination. 2711 4. For each DAO Parent that the node shall attempt an update to, the 2712 node shall exclusively allocate 1 or more set bits from the path 2713 control bitmap to that DAO Parent. The path control bits SHOULD 2714 be allocated in order of preference, such that the most 2715 significant bits, or groupings of bits, are allocated to the most 2716 preferred DAO parents as determined by the node. Once a bit from 2717 the path control bitmap has been allocated to a DAO Parent for 2718 this attempt, the corresponding bit MUST be set in the Path 2719 Control field in the DAO message sent to that DAO Parent, and 2720 that bit MUST NOT be allocated to any other DAO Parent. 2722 5. A unicast DAO message may be sent for DAO Parents that have a 2723 non-zero Path Control field. 2725 6. If any DAO Parent is left without any bits set in its Path 2726 Control field, then that a unicast DAO message MUST NOT be sent 2727 to that DAO parent for this attempt. 2729 7.1.5. Operation of DAO Non-storing Nodes 2731 1. In the non-storing mode of operation, each node sending a DAO 2732 message to its DODAG Parents will include a RPL Target option to 2733 describe itself, followed by RPL Transit Information option(s) to 2734 describe its parents. This information is sufficient for the 2735 DODAG Root to collect the DODAG topology and construct source 2736 routes in the downward direction. 2738 2. In the non-storing mode of operation, each node receiving a DAO 2739 message will arrange to pass the content of the DAO message along 2740 to the DODAG Root. When possible the content of DAO messages may 2741 be aggregated. 2743 3. When a DAO is received from a child by a node who will not store 2744 a routing table entry for the DAO, the node MUST schedule to pass 2745 the DAO contents along to its DAO parents. 2747 7.1.6. Scheduling to Send DAO (or No-Path) 2749 1. An implementation SHOULD arrange to rate-limit the sending of 2750 DAOs. 2752 2. When scheduling to send a DAO, an implementation SHOULD 2753 equivalently start a timer (DelayDAO) to delay sending the DAO. 2754 If the DelayDAO timer is already running then the DAO may be 2755 considered as already scheduled, and implementation SHOULD leave 2756 the timer running at its present duration. 2758 o When computing the delay before sending a DAO, in order to 2759 increase the effectiveness of aggregation, an implementation MAY 2760 allow time to receive DAOs from its sub-DODAG prior to emitting 2761 DAOs to its DAO Parents. 2763 * Suppose there is an implementation parameter DAO_LATENCY which 2764 represents the maximum expected time for a DAO operation to 2765 traverse the LLN from the farthest node to the root. The 2766 scheduled delay in such cases may be, for example, such that 2767 DAO_LATENCY/DAGRank(self_rank) <= DelayDAO < DAO_LATENCY/ 2768 DAGRank(parent_rank), where DAGRank() is defined as in 2769 Section 3.5.2, such that nodes deeper in the DODAG may tend to 2770 report DAO messages first before their parent nodes will report 2771 DAO messages. Note that this suggestion is intended as an 2772 optimization to allow efficient aggregation -- it is not 2773 required for correct operation in the general case. 2775 7.1.7. Triggering DAO Message from the Sub-DODAG 2777 Triggering DAO messages from the Sub-DODAG occurs by using the 2778 following control fields with the rules described below: 2780 The DTSN field from the DIO is a sequence number that is part of the 2781 mechanism to trigger DAO messages. The motivation to use a sequence 2782 number is to provide some means of reliable signaling to the sub- 2783 DODAG. Whereas a control flag that is activated for a short time may 2784 be unobserved by the sub-DODAG if the triggering DIO messages are 2785 lost, the DTSN increment may be observed later even if some 2786 intervening DIO messages have been lost. 2788 The 'T' flag provides a way to signal the refresh of DAO information 2789 over the entire DODAG version. Whereas a DTSN increment may only 2790 trigger a DAO refresh as far as the next storing node (because a 2791 storing node will not increment its own DTSN in response, as 2792 described in the rules below), the assertion of the 'T' flag in 2793 conjunction with an incremented DTSN will result in a DAO refresh 2794 from the entire DODAG. 2796 The control fields are used to trigger DAO messages as follows: 2798 1. A DAO Trigger Sequence Number (DTSN) MUST be maintained by each 2799 node per RPL Instance. The DTSN, in conjunction with the 'T' 2800 flag from the DIO message, provides a means by which DAO messages 2801 may be reliably triggered in the event of topology change. 2803 2. The DTSN MUST be advertised by the node in the DIO message. 2805 3. A node keeps track of the DTSN that it has heard from the last 2806 DIO from each of its DAO Parents. Note that there is one DTSN 2807 maintained per DAO Parent- each DAO Parent may independently 2808 increment it at will. 2810 4. DAO Transmission SHOULD be scheduled when a new parent is added 2811 to the DAO Parent set. 2813 5. A node that receives a newly incremented DTSN from a DAO Parent 2814 MUST schedule a DAO transmission. 2816 o In storing mode operation, when a node sees a DTSN increment, it 2817 is caused to reissue its entire set of routing table entries 2818 learned from DAO messages (or an aggregated subset thereof), but 2819 will not need to increment its own DTSN. 2821 o In either storing or non-storing modes of operation, when a node 2822 sees a DTSN increment AND the 'T' flag is set, it does increment 2823 its own DTSN as well. The 'T' flag 'punches through' all nodes, 2824 causing all routing state from the entire sub-DODAG to be 2825 refreshed. 2827 7.1.8. Sending DAO Messages to DAO Parents 2829 1. DAO Messages sent to DAO Parents MUST be unicast. 2831 * The IPv6 Source Address is a link local address of the node 2832 sending the DAO message. 2834 * The IPv6 Destination Address is a link local address of the 2835 DAO parent. 2837 2. A node MUST send the DAO with the same sequence to all its DAO 2838 parents that are to be used on the way back to the DAO target. 2840 3. When using source routing, a Destination that builds the DAO also 2841 indicates its parent in the DAO as a Transit Information option. 2842 If the node has multiple DAO parents, it MAY include one Transit 2843 Information Option per parent and pass the DAO to one or more 2844 parent. The Transit Information option indicates the preference 2845 for that parent encoded in the Path Control bitfield. 2847 4. When the appointed time arrives (DelayDAO) for the transmission 2848 of DAO messages (with jitter as appropriate) for the requested 2849 entries, the implementation MAY aggregate the the entries into a 2850 reduced numbers of DAOs to be reported to each parent, and 2851 perform compression if possible. 2853 5. Note: it is NOT RECOMMENDED that a DAO Transmission (No-Path) be 2854 scheduled when a DAO Parent is removed from the DAO Parent set. 2856 6. A node MAY set the K flag in a unicast DAO message to solicit a 2857 unicast DAO-ACK in response in order to confirm the attempt. A 2858 node receiving a unicast DAO message with the K flag set SHOULD 2859 respond with a DAO-ACK. A node receiving a DAO message without 2860 the K flag set MAY respond with a DAO-ACK, especially to report 2861 an error condition. 2863 7.1.9. Multicast Destination Advertisement Messages 2865 A special case of DAO operation, distinct from unicast DAO operation, 2866 is multicast DAO operation which may be used to populate '1-hop' 2867 routing table entries. 2869 1. A node MAY multicast a DAO message to the link-local scope all- 2870 nodes multicast address FF02::1. 2872 2. A multicast DAO message MUST be used only to advertise 2873 information about self, i.e. prefixes directly connected to or 2874 owned by this node, such as a multicast group that the node is 2875 subscribed to or a global address owned by the node. 2877 3. A multicast DAO message MUST NOT be used to relay connectivity 2878 information learned (e.g. through unicast DAO) from another node. 2880 4. Information obtained from a multicast DAO MAY be installed in the 2881 routing table and MAY be propagated by a node in unicast DAOs. 2883 5. A node MUST NOT perform any other DAO related processing on a 2884 received multicast DAO, in particular a node MUST NOT perform the 2885 actions of a DAO parent upon receipt of a multicast DAO. 2887 o The multicast DAO may be used to enable direct P2P communication, 2888 without needing the RPL routing structure to relay the packets. 2890 o The multicast DAO does not presume any DODAG relationship between 2891 the emitter and the receiver. 2893 8. Packet Forwarding and Loop Avoidance/Detection 2895 8.1. Suggestions for Packet Forwarding 2897 When forwarding a packet to a destination, precedence is given to 2898 selection of a next-hop successor as follows: 2900 1. This specification only covers how a successor is selected from 2901 the DODAG version that matches the RPLInstanceID marked in the 2902 IPv6 header of the packet being forwarded. Routing outside the 2903 instance can be done as long as additional rules are put in place 2904 such as strict ordering of instances and routing protocols to 2905 protect against loops. 2907 2. If a local administrative preference favors a route that has been 2908 learned from a different routing protocol than RPL, then use that 2909 successor. 2911 3. If there is an entry in the routing table matching the 2912 destination that has been learned from a multicast destination 2913 advertisement (e.g. the destination is a one-hop neighbor), then 2914 use that successor. 2916 4. If there is an entry in the routing table matching the 2917 destination that has been learned from a unicast destination 2918 advertisement (e.g. the destination is located down the sub- 2919 DODAG), then use that successor. If there are DAO Path Control 2920 bits associated with multiple successors, then consult the Path 2921 Control bits to order the successors by preference when choosing. 2923 5. If there is a DODAG version offering a route to a prefix matching 2924 the destination, then select one of those DODAG parents as a 2925 successor according to the OF and routing metrics. 2927 6. Any other as-yet-unattempted DODAG parent may be chosen for the 2928 next attempt to forward a unicast packet when no better match 2929 exists. 2931 7. If there is a DODAG version offering a route to a prefix matching 2932 the destination, but all DODAG parents have been tried and are 2933 temporarily unavailable (as determined by the forwarding 2934 procedure), then select a DODAG sibling as a successor (after 2935 appropriate packet marking for loop detection as described in 2936 Section 8.2. 2938 8. Finally, if no DODAG siblings are available, the packet is 2939 dropped. ICMP Destination Unreachable may be invoked (an 2940 inconsistency is detected). 2942 TTL must be decremented when forwarding. If the packet is being 2943 forwarded via a sibling, then the TTL may be decremented more 2944 aggressively (by more than one) to limit the impact of possible 2945 loops. 2947 Note that the chosen successor MUST NOT be the neighbor that was the 2948 predecessor of the packet (split horizon), except in the case where 2949 it is intended for the packet to change from an up to an down flow, 2950 such as switching from DIO routes to DAO routes as the destination is 2951 neared. 2953 8.2. Loop Avoidance and Detection 2955 RPL loop avoidance mechanisms are kept simple and designed to 2956 minimize churn and states. Loops may form for a number of reasons, 2957 from control packet loss to sibling forwarding. RPL includes a 2958 reactive loop detection technique that protects from meltdown and 2959 triggers repair of broken paths. 2961 RPL loop detection uses information that is placed into the packet. 2962 A future version of this specification will detail how this 2963 information is carried with the packet (e.g. a hop-by-hop option 2964 ([I-D.hui-6man-rpl-option]) or summarized somehow into the flow 2965 label). For the purpose of RPL operations, the information carried 2966 with a packet is constructed follows: 2968 0 1 2 3 2969 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 2970 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2971 |O|S|R|F|0|0|0|0| RPLInstanceID | SenderRank | 2972 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2974 RPL Packet Information 2976 Down 'O' bit: 1-bit flag indicating whether the packet is expected 2977 to progress up or down. A router sets the 'O' bit when the 2978 packet is expect to progress down (using DAO routes), and 2979 resets it when forwarding towards the root of the DODAG 2980 version. A host or RPL leaf node MUST set the bit to 0. 2982 Sibling 'S' bit: 1-bit flag indicating whether the packet has been 2983 forwarded via a sibling at the present rank, and denotes a risk 2984 of a sibling loop. A host or RPL leaf node MUST set the bit to 2985 0. 2987 Rank-Error 'R' bit: 1-bit flag indicating whether a rank error was 2988 detected. A rank error is detected when there is a mismatch in 2989 the relative ranks and the direction as indicated in the 'O' 2990 bit. A host or RPL leaf node MUST set the bit to 0. 2992 Forwarding-Error 'F' bit: 1-bit flag indicating that this node can 2993 not forward the packet further towards the destination. The 2994 'F' bit might be set by sibling that can not forward to a 2995 parent a packet with the Sibling 'S' bit set, or by a child 2996 node that does not have a route to destination for a packet 2997 with the down 'O' bit set. A host or RPL leaf node MUST set 2998 the bit to 0. 3000 RPLInstanceID: 8-bit field indicating the DODAG instance along which 3001 the packet is sent. 3003 SenderRank: 16-bit field set to zero by the source and to 3004 DAGRank(rank) by a router that forwards inside the RPL network. 3006 8.2.1. Source Node Operation 3008 If the source is aware of the RPLInstanceID that is preferred for the 3009 packet, then it MUST set the RPLInstanceID field associated with the 3010 packet accordingly, otherwise it MUST set it to the 3011 RPL_DEFAULT_INSTANCE. 3013 8.2.2. Router Operation 3015 8.2.2.1. Instance Forwarding 3017 Instance IDs are used to avoid loops between DODAGs from different 3018 origins. DODAGs that constructed for antagonistic constraints might 3019 contain paths that, if mixed together, would yield loops. Those 3020 loops are avoided by forwarding a packet along the DODAG that is 3021 associated to a given instance. 3023 The RPLInstanceID is associated by the source with the packet. This 3024 RPLInstanceID MUST match the RPL Instance onto which the packet is 3025 placed by any node, be it a host or router. For traffic originating 3026 outside of the RPL domain there may be a mapping occurring at the 3027 gateway into the RPL domain, possibly based on an encoding within the 3028 flow label. This aspect of RPL operation is to be clarified in a 3029 future version of this specification. 3031 When a router receives a packet that specifies a given RPLInstanceID 3032 and the node can forward the packet along the DODAG associated to 3033 that instance, then the router MUST do so and leave the RPLInstanceID 3034 value unchanged. 3036 If any node can not forward a packet along the DODAG associated to 3037 the RPLInstanceID, then the node SHOULD discard the packet and send 3038 an ICMP error message. 3040 8.2.2.2. DAG Inconsistency Loop Detection 3042 The DODAG is inconsistent if the direction of a packet does not match 3043 the rank relationship. A receiver detects an inconsistency if it 3044 receives a packet with either: 3046 the 'O' bit set (to down) from a node of a higher rank. 3048 the 'O' bit reset (for up) from a node of a lesser rank. 3050 the 'S' bit set (to sibling) from a node of a different rank. 3052 When the DODAG root increments the DODAGVersionNumber a temporary 3053 rank discontinuity may form between the next version and the prior 3054 version, in particular if nodes are adjusting their rank in the next 3055 version and deferring their migration into the next version. A 3056 router that is still a member of the prior version may choose to 3057 forward a packet to a (future) parent that is in the next version. 3058 In some cases this could cause the parent to detect an inconsistency 3059 because the rank-ordering in the prior version is not necessarily the 3060 same as in the next version and the packet may be judged to not be 3061 making forward progress. If the sending router is aware that the 3062 chosen successor has already joined the next version, then the 3063 sending router MUST update the SenderRank to INFINITE_RANK as it 3064 forwards the packets across the discontinuity into the next DODAG 3065 version in order to avoid a false detection of rank inconsistency. 3067 One inconsistency along the path is not considered as a critical 3068 error and the packet may continue. But a second detection along the 3069 path of a same packet should not occur and the packet is dropped. 3071 This process is controlled by the Rank-Error bit associated with the 3072 packet. When an inconsistency is detected on a packet, if the Rank- 3073 Error bit was not set then the Rank-Error bit is set. If it was set 3074 the packet is discarded and the trickle timer is reset. 3076 8.2.2.3. Sibling Loop Avoidance 3078 When a packet is forwarded along siblings, it cannot be checked for 3079 forward progress and may loop between siblings. Experimental 3080 evidence has shown that one sibling hop can be very useful and is 3081 generally sufficient to avoid loops. Based on that evidence, this 3082 specification enforces the simple rule that a packet may not make 2 3083 sibling hops in a row. 3085 When a host issues a packet or when a router forwards a packet to a 3086 non-sibling, the Sibling bit in the packet must be reset. When a 3087 router forwards to a sibling: if the Sibling bit was not set then the 3088 Sibling bit is set. If the Sibling bit was set then then the router 3089 SHOULD return the packet to the sibling that that passed it with the 3090 Forwarding-Error 'F' bit set and the 'S' bit left untouched. 3092 8.2.2.4. DAO Inconsistency Loop Detection and Recovery 3094 A DAO inconsistency happens when router that has an down DAO route 3095 via a child that is a remnant from an obsolete state that is not 3096 matched in the child. With DAO inconsistency loop recovery, a packet 3097 can be used to recursively explore and cleanup the obsolete DAO 3098 states along a sub-DODAG. 3100 In a general manner, a packet that goes down should never go up 3101 again. If DAO inconsistency loop recovery is applied, then the 3102 router SHOULD send the packet back to the parent that passed it with 3103 the Forwarding-Error 'F' bit set and the 'O' bit left untouched. 3104 Otherwise the router MUST silently discard the packet. 3106 8.2.2.5. Forward Path Recovery 3108 Upon receiving a packet with a Forwarding-Error bit set, the node 3109 MUST remove the routing states that caused forwarding to that 3110 neighbor, clear the Forwarding-Error bit and attempt to send the 3111 packet again. The packet may be sent to an alternate neighbor. If 3112 that alternate neighbor still has an inconsistent DAO state via this 3113 node, the process will recurse, this node will set the Forwarding- 3114 Error 'F' bit and the routing state in the alternate neighbor will be 3115 cleaned up as well. 3117 9. Multicast Operation 3119 This section describes further the multicast routing operations over 3120 an IPv6 RPL network, and specifically how unicast DAOs can be used to 3121 relay group registrations up. Wherever the following text mentions 3122 Multicast Listener Discovery (MLD), one can read MLDv1 ([RFC2710]) or 3123 MLDv2 ([RFC3810]). 3125 As is traditional, a listener uses a protocol such as MLD with a 3126 router to register to a multicast group. 3128 Along the path between the router and the DODAG root, MLD requests 3129 are mapped and transported as DAO messages within the RPL protocol; 3130 each hop coalesces the multiple requests for a same group as a single 3131 DAO message to the parent(s), in a fashion similar to proxy IGMP, but 3132 recursively between child router and parent up to the root. 3134 A router might select to pass a listener registration DAO message to 3135 its preferred parent only, in which case multicast packets coming 3136 back might be lost for all of its sub-DODAG if the transmission fails 3137 over that link. Alternatively the router might select to copy 3138 additional parents as it would do for DAO messages advertising 3139 unicast destinations, in which case there might be duplicates that 3140 the router will need to prune. 3142 As a result, multicast routing states are installed in each router on 3143 the way from the listeners to the root, enabling the root to copy a 3144 multicast packet to all its children routers that had issued a DAO 3145 message including a DAO for that multicast group, as well as all the 3146 attached nodes that registered over MLD. 3148 For unicast traffic, it is expected that the grounded root of an 3149 DODAG terminates RPL and MAY redistribute the RPL routes over the 3150 external infrastructure using whatever routing protocol is used in 3151 the other routing domain. For multicast traffic, the root MAY proxy 3152 MLD for all the nodes attached to the RPL domain (this would be 3153 needed if the multicast source is located in the external 3154 infrastructure). For such a source, the packet will be replicated as 3155 it flows down the DODAG based on the multicast routing table entries 3156 installed from the DAO message. 3158 For a source inside the DODAG, the packet is passed to the preferred 3159 parents, and if that fails then to the alternates in the DODAG. The 3160 packet is also copied to all the registered children, except for the 3161 one that passed the packet. Finally, if there is a listener in the 3162 external infrastructure then the DODAG root has to further propagate 3163 the packet into the external infrastructure. 3165 As a result, the DODAG Root acts as an automatic proxy Rendezvous 3166 Point for the RPL network, and as source towards the Internet for all 3167 multicast flows started in the RPL LLN. So regardless of whether the 3168 root is actually attached to the Internet, and regardless of whether 3169 the DODAG is grounded or floating, the root can serve inner multicast 3170 streams at all times. 3172 10. Maintenance of Routing Adjacency 3174 The selection of successors, along the default paths up along the 3175 DODAG, or along the paths learned from destination advertisements 3176 down along the DODAG, leads to the formation of routing adjacencies 3177 that require maintenance. 3179 In IGPs such as OSPF [RFC4915] or IS-IS [RFC5120], the maintenance of 3180 a routing adjacency involves the use of Keepalive mechanisms (Hellos) 3181 or other protocols such as BFD ([I-D.ietf-bfd-base]) and MANET 3182 Neighborhood Discovery Protocol (NHDP [I-D.ietf-manet-nhdp]). 3183 Unfortunately, such an approach is not desirable in constrained 3184 environments such as LLN and would lead to excessive control traffic 3185 in light of the data traffic with a negative impact on both link 3186 loads and nodes resources. Overhead to maintain the routing 3187 adjacency should be minimized. Furthermore, it is not always 3188 possible to rely on the link or transport layer to provide 3189 information of the associated link state. The network layer needs to 3190 fall back on its own mechanism. 3192 Thus RPL makes use of a different approach consisting of probing the 3193 neighbor using a Neighbor Solicitation message (see [RFC4861]). The 3194 reception of a Neighbor Advertisement (NA) message with the 3195 "Solicited Flag" set is used to verify the validity of the routing 3196 adjacency. Such mechanism MAY be used prior to sending a data 3197 packet. This allows for detecting whether or not the routing 3198 adjacency is still valid, and should it not be the case, select 3199 another feasible successor to forward the packet. 3201 11. Guidelines for Objective Functions 3203 An Objective Function (OF) allows for the selection of a DODAG to 3204 join, and a number of peers in that DODAG as parents. The OF is used 3205 to compute an ordered list of parents. The OF is also responsible to 3206 compute the rank of the device within the DODAG version. 3208 The Objective Function is indicated in the DIO message using an 3209 Objective Code Point (OCP), as specified in 3210 [I-D.ietf-roll-routing-metrics], and indicates the method that must 3211 be used to construct the DODAG. The Objective Code Points are 3212 specified in [I-D.ietf-roll-routing-metrics], [I-D.ietf-roll-of0], 3213 and related companion specifications. 3215 11.1. Objective Function Behavior 3217 Most Objective Functions are expected to follow the same abstract 3218 behavior: 3220 o The parent selection is triggered each time an event indicates 3221 that a potential next hop information is updated. This might 3222 happen upon the reception of a DIO message, a timer elapse, all 3223 DODAG parents are unavailable, or a trigger indicating that the 3224 state of a candidate neighbor has changed. 3226 o An OF scans all the interfaces on the device. Although there may 3227 typically be only one interface in most application scenarios, 3228 there might be multiple of them and an interface might be 3229 configured to be usable or not for RPL operation. An interface 3230 can also be configured with a preference or dynamically learned to 3231 be better than another by some heuristics that might be link-layer 3232 dependent and are out of scope. Finally an interface might or not 3233 match a required criterion for an Objective Function, for instance 3234 a degree of security. As a result some interfaces might be 3235 completely excluded from the computation, while others might be 3236 more or less preferred. 3238 o An OF scans all the candidate neighbors on the possible interfaces 3239 to check whether they can act as a router for a DODAG. There 3240 might be multiple of them and a candidate neighbor might need to 3241 pass some validation tests before it can be used. In particular, 3242 some link layers require experience on the activity with a router 3243 to enable the router as a next hop. 3245 o An OF computes self's rank by adding to the rank of the candidate 3246 a value representing the relative locations of self and the 3247 candidate in the DODAG version. 3249 * The increase in rank must be at least MinHopRankIncrease. 3251 * To keep loop avoidance and metric optimization in alignment, 3252 the increase in rank should reflect any increase in the metric 3253 value. For example, with a purely additive metric such as ETX, 3254 the increase in rank can be made proportional to the increase 3255 in the metric. 3257 * Candidate neighbors that would cause self's rank to increase 3258 are not considered for parent selection 3260 o Candidate neighbors that advertise an OF incompatible with the set 3261 of OF specified by the policy functions are ignored. 3263 o As it scans all the candidate neighbors, the OF keeps the current 3264 best parent and compares its capabilities with the current 3265 candidate neighbor. The OF defines a number of tests that are 3266 critical to reach the objective. A test between the routers 3267 determines an order relation. 3269 * If the routers are equal for that relation then the next test 3270 is attempted between the routers, 3272 * Else the best of the two routers becomes the current best 3273 parent and the scan continues with the next candidate neighbor 3275 * Some OFs may include a test to compare the ranks that would 3276 result if the node joined either router 3278 o When the scan is complete, the preferred parent is elected and 3279 self's rank is computed as the preferred parent rank plus the step 3280 in rank with that parent. 3282 o Other rounds of scans might be necessary to elect alternate 3283 parents and siblings. In the next rounds: 3285 * Candidate neighbors that are not in the same DODAG are ignored 3287 * Candidate neighbors that are of greater rank than self are 3288 ignored 3290 * Candidate neighbors of an equal rank to self (siblings) are 3291 ignored for parent selection 3293 * Candidate neighbors of a lesser rank than self (non-siblings) 3294 are preferred 3296 12. RPL Constants and Variables 3298 Following is a summary of RPL constants and variables. 3300 BASE_RANK This is the rank for a virtual root that might be used to 3301 coordinate multiple roots. BASE_RANK has a value of 0. 3303 ROOT_RANK This is the rank for a DODAG root. ROOT_RANK has a value 3304 of MinHopRankIncrease (as advertised by the DODAG root), such 3305 that DAGRank(ROOT_RANK) is 1. 3307 INFINITE_RANK This is the constant maximum for the rank. 3308 INFINITE_RANK has a value of 0xFFFF. 3310 RPL_DEFAULT_INSTANCE This is the RPLInstanceID that is used by this 3311 protocol by a node without any overriding policy. 3312 RPL_DEFAULT_INSTANCE has a value of 0. 3314 DEFAULT_PATH_CONTROL_SIZE TBD (To be determined) 3316 DEFAULT_DIO_INTERVAL_MIN TBD (To be determined) 3318 DEFAULT_DIO_INTERVAL_DOUBLINGS TBD (To be determined) 3319 DEFAULT_DIO_REDUNDANCY_CONSTANT TBD (To be determined) 3321 DEFAULT_MIN_HOP_RANK_INCREASE TBD a power of two (To be determined) 3323 DIO Timer One instance per DODAG that a node is a member of. Expiry 3324 triggers DIO message transmission. Trickle timer with variable 3325 interval in [0, DIOIntervalMin..2^DIOIntervalDoublings]. See 3326 Section 6.3.1 3328 DAG Version Increment Timer Up to one instance per DODAG that the 3329 node is acting as DODAG root of. May not be supported in all 3330 implementations. Expiry triggers increment of 3331 DODAGVersionNumber, causing a new series of updated DIO message 3332 to be sent. Interval should be chosen appropriate to 3333 propagation time of DODAG and as appropriate to application 3334 requirements (e.g. response time vs. overhead). 3336 DelayDAO Timer Up to one instance per DAO parent (the subset of 3337 DODAG parents chosen to receive destination advertisements) per 3338 DODAG. Expiry triggers sending of DAO message to the DAO 3339 parent. See Section 7.1.6 3341 RemoveTimer Up to one instance per DAO entry per neighbor (i.e. 3342 those neighbors that have given DAO messages to this node as a 3343 DODAG parent) Expiry triggers a change in state for the DAO 3344 entry, setting up to do unreachable (No-Path) advertisements or 3345 immediately deallocating the DAO entry if there are no DAO 3346 parents. See Section 7.1.4.1.1.3 3348 13. Manageability Considerations 3350 The aim of this section is to give consideration to the manageability 3351 of RPL, and how RPL will be operated in LLN beyond the use of a MIB 3352 module. The scope of this section is to consider the following 3353 aspects of manageability: fault management, configuration, accounting 3354 and performance. 3356 13.1. Control of Function and Policy 3358 13.1.1. Initialization Mode 3360 When a node is first powered up, it may either choose to stay silent 3361 and not send any multicast DIO message until it has joined a DODAG, 3362 or to immediately root a transient DODAG and start sending multicast 3363 DIO messages. A RPL implementation SHOULD allow configuring whether 3364 the node should stay silent or should start advertising DIO messages. 3366 Furthermore, the implementation SHOULD to allow configuring whether 3367 or not the node should start sending an DIS message as an initial 3368 probe for nearby DODAGs, or should simply wait until it received DIO 3369 messages from other nodes that are part of existing DODAGs. 3371 13.1.2. DIO Base option 3373 RPL specifies a number of protocol parameters. 3375 A RPL implementation SHOULD allow configuring the following routing 3376 protocol parameters, which are further described in Section 5.3: 3378 DAGPreference 3379 RPLInstanceID 3380 DAGObjectiveCodePoint 3381 DODAGID 3382 Routing Information 3383 Prefix Information 3384 DIOIntervalDoublings 3385 DIOIntervalMin 3386 DIORedundancyConstant 3388 DAG Root behavior: In some cases, a node may not want to permanently 3389 act as a DODAG root if it cannot join a grounded DODAG. For 3390 example a battery-operated node may not want to act as a DODAG 3391 root for a long period of time. Thus a RPL implementation MAY 3392 support the ability to configure whether or not a node could 3393 act as a DODAG root for a configured period of time. 3395 DODAG Table Entry Suppression A RPL implementation SHOULD provide 3396 the ability to configure a timer after the expiration of which 3397 logical equivalent of the DODAG table that contains all the 3398 records about a DODAG is suppressed, to be invoked if the DODAG 3399 parent set becomes empty. 3401 13.1.3. Trickle Timers 3403 A RPL implementation makes use of trickle timer to govern the sending 3404 of DIO message. Such an algorithm is determined a by a set of 3405 configurable parameters that are then advertised by the DODAG root 3406 along the DODAG in DIO messages. 3408 For each DODAG, a RPL implementation MUST allow for the monitoring of 3409 the following parameters, further described in Section 6.3.1: 3411 I 3412 T 3413 C 3414 I_min 3415 I_doublings 3417 A RPL implementation SHOULD provide a command (for example via API, 3418 CLI, or SNMP MIB) whereby any procedure that detects an inconsistency 3419 may cause the trickle timer to reset. 3421 13.1.4. DAG Version Number Increment 3423 A RPL implementation may allow by configuration at the DODAG root to 3424 refresh the DODAG states by updating the DODAGVersionNumber. A RPL 3425 implementation SHOULD allow configuring whether or not periodic or 3426 event triggered mechanism are used by the DODAG root to control 3427 DODAGVersionNumber change. 3429 13.1.5. Destination Advertisement Timers 3431 The following set of parameters of the DAO messages SHOULD be 3432 configurable: 3434 o The DelayDAO timer 3436 o The Remove timer 3438 13.1.6. Policy Control 3440 DAG discovery enables nodes to implement different policies for 3441 selecting their DODAG parents. 3443 A RPL implementation SHOULD allow configuring the set of acceptable 3444 or preferred Objective Functions (OF) referenced by their Objective 3445 Codepoints (OCPs) for a node to join a DODAG, and what action should 3446 be taken if none of a node's candidate neighbors advertise one of the 3447 configured allowable Objective Functions. 3449 A node in an LLN may learn routing information from different routing 3450 protocols including RPL. It is in this case desirable to control via 3451 administrative preference which route should be favored. An 3452 implementation SHOULD allow for specifying an administrative 3453 preference for the routing protocol from which the route was learned. 3455 13.1.7. Data Structures 3457 Some RPL implementation may limit the size of the candidate neighbor 3458 list in order to bound the memory usage, in which case some otherwise 3459 viable candidate neighbors may not be considered and simply dropped 3460 from the candidate neighbor list. 3462 A RPL implementation MAY provide an indicator on the size of the 3463 candidate neighbor list. 3465 13.2. Information and Data Models 3467 The information and data models necessary for the operation of RPL 3468 will be defined in a separate document specifying the RPL SNMP MIB. 3470 13.3. Liveness Detection and Monitoring 3472 The aim of this section is to describe the various RPL mechanisms 3473 specified to monitor the protocol. 3475 As specified in Section 3.1, an implementation is expected to 3476 maintain a set of data structures in support of DODAG discovery: 3478 o The candidate neighbors data structure 3480 o For each DODAG: 3482 * A set of DODAG parents 3484 13.3.1. Candidate Neighbor Data Structure 3486 A node in the candidate neighbor list is a node discovered by the 3487 some means and qualified to potentially become of neighbor or a 3488 sibling (with high enough local confidence). A RPL implementation 3489 SHOULD provide a way monitor the candidate neighbors list with some 3490 metric reflecting local confidence (the degree of stability of the 3491 neighbors) measured by some metrics. 3493 A RPL implementation MAY provide a counter reporting the number of 3494 times a candidate neighbor has been ignored, should the number of 3495 candidate neighbors exceeds the maximum authorized value. 3497 13.3.2. Directed Acyclic Graph (DAG) Table 3499 For each DAG, a RPL implementation is expected to keep track of the 3500 following DODAG table values: 3502 o DODAGID 3504 o DAGObjectiveCodePoint 3505 o A set of prefixes offered upwards along the DODAG 3507 o A set of DODAG Parents 3509 o timer to govern the sending of DIO messages for the DODAG 3511 o DODAGVersionNumber 3513 The set of DODAG parents structure is itself a table with the 3514 following entries: 3516 o A reference to the neighboring device which is the DAG parent 3518 o A record of most recent information taken from the DAG Information 3519 Object last processed from the DODAG Parent 3521 o A flag reporting if the Parent is a DAO Parent as described in 3522 Section 7 3524 13.3.3. Routing Table 3526 For each route provisioned by RPL operation, a RPL implementation 3527 MUST keep track of the following: 3529 o Routing Information (prefix, prefix length, ...) 3531 o Lifetime Timer 3533 o Next Hop 3535 o Next Hop Interface 3537 o Flag indicating that the route was provisioned from one of: 3539 * Unicast DAO message 3541 * DIO message 3543 * Multicast DAO message 3545 13.3.4. Other RPL Monitoring Parameters 3547 A RPL implementation SHOULD provide a counter reporting the number of 3548 a times the node has detected an inconsistency with respect to a 3549 DODAG parent, e.g. if the DODAGID has changed. 3551 A RPL implementation MAY log the reception of a malformed DIO message 3552 along with the neighbor identification if avialable. 3554 13.3.5. RPL Trickle Timers 3556 A RPL implementation operating on a DODAG root MUST allow for the 3557 configuration of the following trickle parameters: 3559 o The DIOIntervalMin expressed in ms 3561 o The DIOIntervalDoublings 3563 o The DIORedundancyConstant 3565 A RPL implementation MAY provide a counter reporting the number of 3566 times an inconsistency (and thus the trickle timer has been reset). 3568 13.4. Verifying Correct Operation 3570 This section has to be completed in further revision of this document 3571 to list potential Operations and Management (OAM) tools that could be 3572 used for verifying the correct operation of RPL. 3574 13.5. Requirements on Other Protocols and Functional Components 3576 RPL does not have any impact on the operation of existing protocols. 3578 13.6. Impact on Network Operation 3580 To be completed. 3582 14. Security Considerations 3584 +----------------------------------------------------------------+ 3585 | | 3586 | TBD | 3587 | Under Construction | 3588 | Deference given to Security Design Team | 3589 | | 3590 +----------------------------------------------------------------+ 3592 14.1. Overview 3594 From a security perspective, RPL networks are no different from any 3595 other network. They are vulnerable to passive eavesdropping attacks 3596 and potentially even active tampering when physical access to a wire 3597 is not required to participate in communications. The very nature of 3598 ad hoc networks and their cost objectives impose additional security 3599 constraints, which perhaps make these networks the most difficult 3600 environments to secure. Devices are low-cost and have limited 3601 capabilities in terms of computing power, available storage, and 3602 power drain; and it cannot always be assumed they have neither a 3603 trusted computing base nor a high-quality random number generator 3604 aboard. Communications cannot rely on the online availability of a 3605 fixed infrastructure and might involve short-term relationships 3606 between devices that may never have communicated before. These 3607 constraints might severely limit the choice of cryptographic 3608 algorithms and protocols and influence the design of the security 3609 architecture because the establishment and maintenance of trust 3610 relationships between devices need to be addressed with care. In 3611 addition, battery lifetime and cost constraints put severe limits on 3612 the security overhead these networks can tolerate, something that is 3613 of far less concern with higher bandwidth networks. Most of these 3614 security architectural elements can be implemented at higher layers 3615 and may, therefore, be considered to be outside the scope of this 3616 standard. Special care, however, needs to be exercised with respect 3617 to interfaces to these higher layers. 3619 The security mechanisms in this standard are based on symmetric-key 3620 and public-key cryptography and use keys that are to be provided by 3621 higher layer processes. The establishment and maintenance of these 3622 keys are outside the scope of this standard. The mechanisms assume a 3623 secure implementation of cryptographic operations and secure and 3624 authentic storage of keying material. 3626 The security mechanisms specified provide particular combinations of 3627 the following security services: 3629 Data confidentiality: Assurance that transmitted information is only 3630 disclosed to parties for which it is intended. 3632 Data authenticity: Assurance of the source of transmitted 3633 information (and, hereby, that information was not 3634 modified in transit). 3636 Replay protection: Assurance that a duplicate of transmitted 3637 information is detected. 3639 Timeliness (delay protection): Assurance that transmitted 3640 information was received in a timely manner. 3642 The actual protection provided can be adapted on a per-packet basis 3643 and allows for varying levels of data authenticity (to minimize 3644 security overhead in transmitted packets where required) and for 3645 optional data confidentiality. When nontrivial protection is 3646 required, replay protection is always provided. 3648 Replay protection is provided via the use of a non-repeating value 3649 (nonce) in the packet protection process and storage of some status 3650 information for each originating device on the receiving device, 3651 which allows detection of whether this particular nonce value was 3652 used previously by the originating device. In addition, so-called 3653 delay protection is provided amongst those devices that have a 3654 loosely synchronized clock on board. The acceptable time delay can 3655 be adapted on a per-packet basis and allows for varying latencies (to 3656 facilitate longer latencies in packets transmitted over a multi-hop 3657 communication path). 3659 Cryptographic protection may use a key shared between two peer 3660 devices (link key) or a key shared among a group of devices (group 3661 key), thus allowing some flexibility and application-specific 3662 tradeoffs between key storage and key maintenance costs versus the 3663 cryptographic protection provided. If a group key is used for peer- 3664 to-peer communication, protection is provided only against outsider 3665 devices and not against potential malicious devices in the key- 3666 sharing group. 3668 Data authenticity may be provided using symmetric-key based or 3669 public-key based techniques. With public-key based techniques (via 3670 signatures), one corroborates evidence as to the unique originator of 3671 transmitted information, whereas with symmetric-key based techniques 3672 data authenticity is only provided relative to devices in a key- 3673 sharing group. Thus, public-key based authentication may be useful 3674 in scenarios that require a more fine-grained authentication than can 3675 be provided with symmetric-key based authentication techniques alone, 3676 such as with group communications (broadcast, multicast), or in 3677 scenarios that require non-repudiation. 3679 14.2. Functional Description of Packet Protection 3681 14.2.1. Transmission of Outgoing Packets 3683 This section describes the transmission of secured RPL control 3684 packets. Give an outgoing RPL control packet and required security 3685 protection, this section describes how RPL generates the secured 3686 packet to transmit. It describes the order of cryptographic 3687 operations to provide the required protection. 3689 A RPL node MUST set the security section in the RPL packet to 3690 describes the required protection level. 3692 The Counter field of the security header MUST be an increment of the 3693 last Counter field transmitted. 3695 If the RPL packet is not a response to a Consistency Check message, 3696 the node MAY set the Counter Compression field of the security 3697 option. If the packet is a response to a Consistency Check message, 3698 the node MUST clear the Counter Compression field. 3700 A node sets the Key Identifier Mode (KIM) of the packet based on its 3701 understanding of what keys destinations have. 3703 A node MUST replaced the original packet payload with that payload 3704 encrypted using the security protection, key, and nonce specified in 3705 the security section. 3707 14.2.2. Reception of Incoming Packets 3709 This section describes the reception of a secured RPL packet. Given 3710 an incoming RPL packet, this section describes now RPL generates an 3711 unencrypted version of the packet and validates its integrity. 3713 The receiver uses the security control field of the security section 3714 to determine what processing to do. If the described level of 3715 security does not meet locally maintained security policies, a node 3716 MAY discard the packet without further processing. These policies 3717 can include security levels, keys used, or source identifiers. 3719 Using a nonce derived from the Counter field and other information 3720 (as described in Section Figure 21), the receiver checks the 3721 integrity of the packet by comparing the received MAC with the 3722 computed MAC. If this integrity check does not pass, a node MUST 3723 discard the packet. 3725 RPL uses the key information described in a RPL message to decrypt 3726 its contents as necessary. Once a message has passed its integrity 3727 checks and been successfully decrypted, the node can update its local 3728 security information, such as the source's expected counter value for 3729 counter compression. A node MUST NOT update security information on 3730 receipt of a message that fails security policy checks, integrity 3731 checks, or decryption. 3733 14.2.3. Cryptographic Mode of Operation 3735 The cryptographic mode of operation used is based on the CCM mode of 3736 operation specified with [TBDREF] and the block-cipher AES-128 3737 [TBDREF]. This mode of operation is widely supported by existing 3738 implementations and coincides with the CCM* mode of operation 3739 specified with [TBDREF]. 3741 14.2.3.1. Nonce 3743 The so-called nonce is constructed as follows: 3745 0 1 2 3 3746 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 3747 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3748 | | 3749 + Source Identifier + 3750 | | 3751 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3752 | Counter | 3753 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3754 |Reserved | LVL | 3755 +-+-+-+-+-+-+-+-+ 3757 Figure 21: CCM* Nonce 3759 Source Identifier: 8 bytes. Source Identifier is set to the logical 3760 identifier of the originator of the protected packet. 3762 Counter: 4 bytes. Counter is set to the (uncompressed) value of the 3763 corresponding field in the Security option of the RPL control 3764 message. 3766 Security Level (LVL): 3 bits. Security Level is set to the value of 3767 the corresponding field in the Security option of the RPL 3768 control message. 3770 Unassigned bits of the nonce are reserved. They MUST be set to zero 3771 when constructing the nonce. 3773 All fields of the nonce shall be represented is most-significant- 3774 octet and most-significant-bit first order. 3776 14.3. Protecting RPL ICMPv6 messages 3778 For a RPL ICMPv6 message, the entire packet is within the scope of 3779 RPL security. The message authentication code is calculated over the 3780 entire IPv6 packet. This calculation is done before any compression 3781 that lower layers may apply. The IPv6 and ICMPv6 headers are never 3782 encrypted. The body of the RPL ICMPv6 message MAY be encrypted, 3783 starting from the first byte after the security information and 3784 continuing to the end of the packet. 3786 14.4. Security State Machine 3788 A DAG root starting a DODAG sets the RPL routing security policy for 3789 the entire DODAG. 3791 A member of a secure DODAG MUST conform to the policy set by the DAG 3792 root. When starting a secure DODAG, the DAG root will send secure 3793 DIO messages. A node attempting to join the DODAG will send a secure 3794 Authentication Request (AREQ) to the DAG root. Nodes that are not 3795 authenticated in a secure DODAG will be unable to generate properly 3796 constructed secured RPL packets. These nodes are in state 3797 "unauthenticated". A member of a secure DODAG MUST forward an AREQ 3798 packet to the DAG root, and MUST NOT forward any other type of packet 3799 from an unauthenticated node. 3801 The DAG root may choose to respond to the AREQ with an ARSP packet. 3802 This packet will provide the authenticating node with the 3803 cryptographic materials necessary to participate in RPL routing. 3804 Some authentication flows may involve the exchange of more than one 3805 AREQ or ARSP packets. 3807 The simplest authentication flow will involve the use of a single 3808 pre-installed network-wide authentication key. The installation of 3809 this key is out of scope of this document. The authenticating node 3810 will use the pre-installed key to calculate a MIC for the AREQ 3811 packet. The DODAG root will verify the authenticity of the 3812 authenticating node using the same key. The DODAG root, having 3813 previously chosen a single random instance-wide shared key, will send 3814 this key, encrypted and authenticated with the pre-installed key, in 3815 the ARSP packet. The authenticating node, decoding this packet with 3816 the pre-installed key, will verify the authenticity of the DODAG 3817 root. 3819 It is assumed that additional authentication and key exchange 3820 mechanisms will be included in future drafts of the document. 3822 Periodic key updates will use the secure KU packet code. The 3823 responsibility for initiating key update will reside with the DODAG 3824 root, and is out of scope of this document. 3826 15. IANA Considerations 3828 15.1. RPL Control Message 3830 The RPL Control Message is an ICMP information message type that is 3831 to be used carry DODAG Information Objects, DODAG Information 3832 Solicitations, and Destination Advertisement Objects in support of 3833 RPL operation. 3835 IANA has defined an ICMPv6 Type Number Registry. The suggested type 3836 value for the RPL Control Message is 155, to be confirmed by IANA. 3838 15.2. New Registry for RPL Control Codes 3840 IANA is requested to create a registry, RPL Control Codes, for the 3841 Code field of the ICMPv6 RPL Control Message. 3843 New codes may be allocated only by an IETF Consensus action. Each 3844 code should be tracked with the following qualities: 3846 o Code 3848 o Description 3850 o Defining RFC 3852 Three codes are currently defined: 3854 +------+----------------------------------------------+-------------+ 3855 | Code | Description | Reference | 3856 +------+----------------------------------------------+-------------+ 3857 | 0x00 | DODAG Information Solicitation | This | 3858 | | | document | 3859 | 0x01 | DODAG Information Object | This | 3860 | | | document | 3861 | 0x02 | Destination Advertisement Object | This | 3862 | | | document | 3863 | 0x80 | Secure DODAG Information Solicitation | This | 3864 | | | document | 3865 | 0x81 | Secure DODAG Information Object | This | 3866 | | | document | 3867 | 0x82 | Secure Destination Advertisement Object | This | 3868 | | | document | 3869 | 0x83 | Secure Destination Advertisement Object | This | 3870 | | Acknowledgment | document | 3871 +------+----------------------------------------------+-------------+ 3873 RPL Control Codes 3875 15.3. New Registry for the Mode of Operation (MOP) DIO Control Field 3877 IANA is requested to create a registry for the Mode of Operation 3878 (MOP) DIO Control Field, which is contained in the DIO Base. 3880 New fields may be allocated only by an IETF Consensus action. Each 3881 field should be tracked with the following qualities: 3883 o Mode of Operation 3885 o Capability description 3887 o Defining RFC 3889 Two values are currently defined: 3891 +-----+-------------------------------+---------------+ 3892 | MOP | Description | Reference | 3893 +-----+-------------------------------+---------------+ 3894 | 00 | Non-Storing mode of operation | This document | 3895 | 01 | Storing mode of operation | This document | 3896 +-----+-------------------------------+---------------+ 3898 DIO Base Flags 3900 15.4. RPL Control Message Option 3902 IANA is requested to create a registry for the RPL Control Message 3903 Options 3905 +-------+-------------------------+---------------+ 3906 | Value | Meaning | Reference | 3907 +-------+-------------------------+---------------+ 3908 | 0 | Pad1 | This document | 3909 | 1 | PadN | This document | 3910 | 2 | DAG Metric Container | This Document | 3911 | 3 | Routing Information | This Document | 3912 | 4 | DAG Timer Configuration | This Document | 3913 | 5 | RPL Target | This Document | 3914 | 6 | Transit Information | This Document | 3915 | 7 | Solicited Information | This Document | 3916 | 8 | Prefix Information | This Document | 3917 +-------+-------------------------+---------------+ 3919 RPL Control Message Options 3921 16. Acknowledgements 3923 The authors would like to acknowledge the review, feedback, and 3924 comments from Roger Alexander, Emmanuel Baccelli, Dominique Barthel, 3925 Yusuf Bashir, Phoebus Chen, Mathilde Durvy, Manhar Goindi, Mukul 3926 Goyal, Anders Jagd, JeongGil (John) Ko, Quentin Lampin, Jerry 3927 Martocci, Matteo Paris, Alexandru Petrescu, Joseph Reddy, and Don 3928 Sturek. 3930 The authors would like to acknowledge the guidance and input provided 3931 by the ROLL Chairs, David Culler and JP Vasseur. 3933 The authors would like to acknowledge prior contributions of Robert 3934 Assimiti, Mischa Dohler, Julien Abeille, Ryuji Wakikawa, Teco Boot, 3935 Patrick Wetterwald, Bryan Mclaughlin, Carlos J. Bernardos, Thomas 3936 Watteyne, Zach Shelby, Caroline Bontoux, Marco Molteni, Billy Moon, 3937 and Arsalan Tavakoli, which have provided useful design 3938 considerations to RPL. 3940 17. Contributors 3942 RPL is the result of the contribution of the following members of the 3943 RPL Author Team, including the editors, and additional contributors 3944 as listed below: 3946 JP Vasseur 3947 Cisco Systems, Inc 3948 11, Rue Camille Desmoulins 3949 Issy Les Moulineaux, 92782 3950 France 3952 Email: jpv@cisco.com 3954 Thomas Heide Clausen 3955 LIX, Ecole Polytechnique, France 3957 Phone: +33 6 6058 9349 3958 EMail: T.Clausen@computer.org 3959 URI: http://www.ThomasClausen.org/ 3961 Philip Levis 3962 Stanford University 3963 358 Gates Hall, Stanford University 3964 Stanford, CA 94305-9030 3965 USA 3967 Email: pal@cs.stanford.edu 3969 Richard Kelsey 3970 Ember Corporation 3971 Boston, MA 3972 USA 3974 Phone: +1 617 951 1225 3975 Email: kelsey@ember.com 3977 Jonathan W. Hui 3978 Arch Rock Corporation 3979 501 2nd St. Ste. 410 3980 San Francisco, CA 94107 3981 USA 3983 Email: jhui@archrock.com 3985 Kris Pister 3986 Dust Networks 3987 30695 Huntwood Ave. 3988 Hayward, 94544 3989 USA 3991 Email: kpister@dustnetworks.com 3993 Anders Brandt 3994 Sigma Designs 3995 Emdrupvej 26A, 1. 3996 Copenhagen, DK-2100 3997 Denmark 3999 Email: abr@sdesigns.dk 4001 Stephen Dawson-Haggerty 4002 UC Berkeley 4003 Soda Hall, UC Berkeley 4004 Berkeley, CA 94720 4005 USA 4007 Email: stevedh@cs.berkeley.edu 4009 18. References 4010 18.1. Normative References 4012 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 4013 Requirement Levels", BCP 14, RFC 2119, March 1997. 4015 18.2. Informative References 4017 [I-D.hui-6man-rpl-option] 4018 Hui, J. and J. Vasseur, "RPL Option for Carrying RPL 4019 Information in Data-Plane Datagrams", 4020 draft-hui-6man-rpl-option-00 (work in progress), 4021 March 2010. 4023 [I-D.hui-6man-rpl-routing-header] 4024 Hui, J., Vasseur, J., and D. Culler, "A Source Routing 4025 Header for RPL", draft-hui-6man-rpl-routing-header-00 4026 (work in progress), May 2010. 4028 [I-D.ietf-bfd-base] 4029 Katz, D. and D. Ward, "Bidirectional Forwarding 4030 Detection", draft-ietf-bfd-base-11 (work in progress), 4031 January 2010. 4033 [I-D.ietf-manet-nhdp] 4034 Clausen, T., Dearlove, C., and J. Dean, "Mobile Ad Hoc 4035 Network (MANET) Neighborhood Discovery Protocol (NHDP)", 4036 draft-ietf-manet-nhdp-12 (work in progress), March 2010. 4038 [I-D.ietf-roll-building-routing-reqs] 4039 Martocci, J., Riou, N., Mil, P., and W. Vermeylen, 4040 "Building Automation Routing Requirements in Low Power and 4041 Lossy Networks", draft-ietf-roll-building-routing-reqs-09 4042 (work in progress), January 2010. 4044 [I-D.ietf-roll-of0] 4045 Thubert, P., "RPL Objective Function 0", 4046 draft-ietf-roll-of0-01 (work in progress), February 2010. 4048 [I-D.ietf-roll-routing-metrics] 4049 Vasseur, J., Kim, M., Networks, D., and H. Chong, "Routing 4050 Metrics used for Path Calculation in Low Power and Lossy 4051 Networks", draft-ietf-roll-routing-metrics-06 (work in 4052 progress), April 2010. 4054 [I-D.ietf-roll-terminology] 4055 Vasseur, J., "Terminology in Low power And Lossy 4056 Networks", draft-ietf-roll-terminology-03 (work in 4057 progress), March 2010. 4059 [I-D.ietf-roll-trickle] 4060 Levis, P., Clausen, T., Hui, J., and J. Ko, "The Trickle 4061 Algorithm", draft-ietf-roll-trickle-01 (work in progress), 4062 April 2010. 4064 [RFC1982] Elz, R. and R. Bush, "Serial Number Arithmetic", RFC 1982, 4065 August 1996. 4067 [RFC2710] Deering, S., Fenner, W., and B. Haberman, "Multicast 4068 Listener Discovery (MLD) for IPv6", RFC 2710, 4069 October 1999. 4071 [RFC3810] Vida, R. and L. Costa, "Multicast Listener Discovery 4072 Version 2 (MLDv2) for IPv6", RFC 3810, June 2004. 4074 [RFC3819] Karn, P., Bormann, C., Fairhurst, G., Grossman, D., 4075 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 4076 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 4077 RFC 3819, July 2004. 4079 [RFC4101] Rescorla, E. and IAB, "Writing Protocol Models", RFC 4101, 4080 June 2005. 4082 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 4083 More-Specific Routes", RFC 4191, November 2005. 4085 [RFC4443] Conta, A., Deering, S., and M. Gupta, "Internet Control 4086 Message Protocol (ICMPv6) for the Internet Protocol 4087 Version 6 (IPv6) Specification", RFC 4443, March 2006. 4089 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 4090 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 4091 September 2007. 4093 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 4094 Address Autoconfiguration", RFC 4862, September 2007. 4096 [RFC4915] Psenak, P., Mirtorabi, S., Roy, A., Nguyen, L., and P. 4097 Pillay-Esnault, "Multi-Topology (MT) Routing in OSPF", 4098 RFC 4915, June 2007. 4100 [RFC5120] Przygienda, T., Shen, N., and N. Sheth, "M-ISIS: Multi 4101 Topology (MT) Routing in Intermediate System to 4102 Intermediate Systems (IS-ISs)", RFC 5120, February 2008. 4104 [RFC5548] Dohler, M., Watteyne, T., Winter, T., and D. Barthel, 4105 "Routing Requirements for Urban Low-Power and Lossy 4106 Networks", RFC 5548, May 2009. 4108 [RFC5673] Pister, K., Thubert, P., Dwars, S., and T. Phinney, 4109 "Industrial Routing Requirements in Low-Power and Lossy 4110 Networks", RFC 5673, October 2009. 4112 [RFC5826] Brandt, A., Buron, J., and G. Porcu, "Home Automation 4113 Routing Requirements in Low-Power and Lossy Networks", 4114 RFC 5826, April 2010. 4116 Appendix A. Requirements 4118 A.1. Protocol Properties Overview 4120 RPL demonstrates the following properties, consistent with the 4121 requirements specified by the application-specific requirements 4122 documents. 4124 A.1.1. IPv6 Architecture 4126 RPL is strictly compliant with layered IPv6 architecture. 4128 Further, RPL is designed with consideration to the practical support 4129 and implementation of IPv6 architecture on devices which may operate 4130 under severe resource constraints, including but not limited to 4131 memory, processing power, energy, and communication. The RPL design 4132 does not presume high quality reliable links, and operates over lossy 4133 links (usually low bandwidth with low packet delivery success rate). 4135 A.1.2. Typical LLN Traffic Patterns 4137 Multipoint-to-Point (MP2P) and Point-to-multipoint (P2MP) traffic 4138 flows from nodes within the LLN from and to egress points are very 4139 common in LLNs. Low power and lossy network Border Router (LBR) 4140 nodes may typically be at the root of such flows, although such flows 4141 are not exclusively rooted at LBRs as determined on an application- 4142 specific basis. In particular, several applications such as building 4143 or home automation do require P2P (Point-to-Point) communication. 4145 As required by the aforementioned routing requirements documents, RPL 4146 supports the installation of multiple paths. The use of multiple 4147 paths include sending duplicated traffic along diverse paths, as well 4148 as to support advanced features such as Class of Service (CoS) based 4149 routing, or simple load balancing among a set of paths (which could 4150 be useful for the LLN to spread traffic load and avoid fast energy 4151 depletion on some, e.g. battery powered, nodes). Conceptually, 4152 multiple instances of RPL can be used to send traffic along different 4153 topology instances, the construction of which is governed by 4154 different Objective Functions (OF). Details of RPL operation in 4155 support of multiple instances are beyond the scope of the present 4156 specification. 4158 A.1.3. Constraint Based Routing 4160 The RPL design supports constraint based routing, based on a set of 4161 routing metrics and constraints. The routing metrics and constraints 4162 for links and nodes with capabilities supported by RPL are specified 4163 in a companion document to this specification, 4164 [I-D.ietf-roll-routing-metrics]. RPL signals the metrics, 4165 constraints, and related Objective Functions (OFs) in use in a 4166 particular implementation by means of an Objective Code Point (OCP). 4167 Both the routing metrics, constraints, and the OF help determine the 4168 construction of the Directed Acyclic Graphs (DAG) using a distributed 4169 path computation algorithm. 4171 A.2. Deferred Requirements 4173 NOTE: RPL is still a work in progress. At this time there remain 4174 several unsatisfied application requirements, but these are to be 4175 addressed as RPL is further specified. 4177 Appendix B. Outstanding Issues 4179 This section enumerates some outstanding issues that are to be 4180 addressed in future revisions of the RPL specification. 4182 B.1. Additional Support for P2P Routing 4184 In some situations the baseline mechanism to support arbitrary P2P 4185 traffic, by flowing upwards along the DODAG until a common ancestor 4186 is reached and then flowing down, may not be suitable for all 4187 application scenarios. A related scenario may occur when the down 4188 paths setup along the DODAG by the destination advertisement 4189 mechanism are not the most desirable downward paths for the specific 4190 application scenario (in part because the DODAG links may not be 4191 symmetric). It may be desired to support within RPL the discovery 4192 and installation of more direct routes 'across' the DAG. Such 4193 mechanisms need to be investigated. 4195 B.2. Address / Header Compression 4197 In order to minimize overhead within the LLN it is desirable to 4198 perform some sort of address and/or header compression, perhaps via 4199 labels, addresses aggregation, or some other means. This is still 4200 under investigation. 4202 B.3. Managing Multiple Instances 4204 A network may run multiple instances of RPL concurrently. Such a 4205 network will require methods for assigning and otherwise managing 4206 RPLInstanceIDs. This will likely be addressed in a separate 4207 document. 4209 Authors' Addresses 4211 Tim Winter (editor) 4213 Email: wintert@acm.org 4215 Pascal Thubert (editor) 4216 Cisco Systems 4217 Village d'Entreprises Green Side 4218 400, Avenue de Roumanille 4219 Batiment T3 4220 Biot - Sophia Antipolis 06410 4221 FRANCE 4223 Phone: +33 497 23 26 34 4224 Email: pthubert@cisco.com 4226 RPL Author Team 4227 IETF ROLL WG 4229 Email: rpl-authors@external.cisco.com