idnits 2.17.1 draft-ietf-anima-grasp-12.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. 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 19, 2017) is 2533 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) -- Looks like a reference, but probably isn't: '7' on line 3489 -- Looks like a reference, but probably isn't: '13767778' on line 3489 -- Looks like a reference, but probably isn't: '34965' on line 3489 == Outdated reference: A later version (-11) exists of draft-greevenbosch-appsawg-cbor-cddl-10 ** Downref: Normative reference to an Informational draft: draft-greevenbosch-appsawg-cbor-cddl (ref. 'I-D.greevenbosch-appsawg-cbor-cddl') == Outdated reference: A later version (-30) exists of draft-ietf-anima-autonomic-control-plane-06 ** Obsolete normative reference: RFC 5246 (Obsoleted by RFC 8446) ** Obsolete normative reference: RFC 6347 (Obsoleted by RFC 9147) ** Obsolete normative reference: RFC 7049 (Obsoleted by RFC 8949) == Outdated reference: A later version (-45) exists of draft-ietf-anima-bootstrapping-keyinfra-05 == Outdated reference: A later version (-10) exists of draft-ietf-anima-reference-model-03 == Outdated reference: A later version (-10) exists of draft-ietf-anima-stable-connectivity-02 == Outdated reference: A later version (-06) exists of draft-liu-anima-grasp-api-03 -- Obsolete informational reference (is this intentional?): RFC 3315 (Obsoleted by RFC 8415) -- Obsolete informational reference (is this intentional?): RFC 5226 (Obsoleted by RFC 8126) -- Obsolete informational reference (is this intentional?): RFC 7564 (Obsoleted by RFC 8264) Summary: 4 errors (**), 0 flaws (~~), 7 warnings (==), 7 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group C. Bormann 3 Internet-Draft Universitaet Bremen TZI 4 Intended status: Standards Track B. Carpenter, Ed. 5 Expires: November 20, 2017 Univ. of Auckland 6 B. Liu, Ed. 7 Huawei Technologies Co., Ltd 8 May 19, 2017 10 A Generic Autonomic Signaling Protocol (GRASP) 11 draft-ietf-anima-grasp-12 13 Abstract 15 This document establishes requirements for a signaling protocol that 16 enables autonomic nodes and autonomic service agents to dynamically 17 discover peers, to synchronize state with them, and to negotiate 18 parameter settings with them. The document then defines a general 19 protocol for discovery, synchronization and negotiation, while the 20 technical objectives for specific scenarios are to be described in 21 separate documents. An Appendix briefly discusses existing protocols 22 with comparable features. 24 Status of This Memo 26 This Internet-Draft is submitted in full conformance with the 27 provisions of BCP 78 and BCP 79. 29 Internet-Drafts are working documents of the Internet Engineering 30 Task Force (IETF). Note that other groups may also distribute 31 working documents as Internet-Drafts. The list of current Internet- 32 Drafts is at http://datatracker.ietf.org/drafts/current/. 34 Internet-Drafts are draft documents valid for a maximum of six months 35 and may be updated, replaced, or obsoleted by other documents at any 36 time. It is inappropriate to use Internet-Drafts as reference 37 material or to cite them other than as "work in progress." 39 This Internet-Draft will expire on November 20, 2017. 41 Copyright Notice 43 Copyright (c) 2017 IETF Trust and the persons identified as the 44 document authors. All rights reserved. 46 This document is subject to BCP 78 and the IETF Trust's Legal 47 Provisions Relating to IETF Documents 48 (http://trustee.ietf.org/license-info) in effect on the date of 49 publication of this document. Please review these documents 50 carefully, as they describe your rights and restrictions with respect 51 to this document. Code Components extracted from this document must 52 include Simplified BSD License text as described in Section 4.e of 53 the Trust Legal Provisions and are provided without warranty as 54 described in the Simplified BSD License. 56 Table of Contents 58 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 59 2. Requirement Analysis of Discovery, Synchronization and 60 Negotiation . . . . . . . . . . . . . . . . . . . . . . . . . 5 61 2.1. Requirements for Discovery . . . . . . . . . . . . . . . 5 62 2.2. Requirements for Synchronization and Negotiation 63 Capability . . . . . . . . . . . . . . . . . . . . . . . 6 64 2.3. Specific Technical Requirements . . . . . . . . . . . . . 9 65 3. GRASP Protocol Overview . . . . . . . . . . . . . . . . . . . 10 66 3.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . 10 67 3.2. High Level Deployment Model . . . . . . . . . . . . . . . 12 68 3.3. High Level Design Choices . . . . . . . . . . . . . . . . 13 69 3.4. Quick Operating Overview . . . . . . . . . . . . . . . . 16 70 3.5. GRASP Protocol Basic Properties and Mechanisms . . . . . 17 71 3.5.1. Required External Security Mechanism . . . . . . . . 17 72 3.5.2. Constrained Instances . . . . . . . . . . . . . . . . 17 73 3.5.3. Transport Layer Usage . . . . . . . . . . . . . . . . 19 74 3.5.4. Discovery Mechanism and Procedures . . . . . . . . . 20 75 3.5.5. Negotiation Procedures . . . . . . . . . . . . . . . 23 76 3.5.6. Synchronization and Flooding Procedures . . . . . . . 25 77 3.6. GRASP Constants . . . . . . . . . . . . . . . . . . . . . 27 78 3.7. Session Identifier (Session ID) . . . . . . . . . . . . . 28 79 3.8. GRASP Messages . . . . . . . . . . . . . . . . . . . . . 29 80 3.8.1. Message Overview . . . . . . . . . . . . . . . . . . 29 81 3.8.2. GRASP Message Format . . . . . . . . . . . . . . . . 29 82 3.8.3. Message Size . . . . . . . . . . . . . . . . . . . . 30 83 3.8.4. Discovery Message . . . . . . . . . . . . . . . . . . 30 84 3.8.5. Discovery Response Message . . . . . . . . . . . . . 31 85 3.8.6. Request Messages . . . . . . . . . . . . . . . . . . 32 86 3.8.7. Negotiation Message . . . . . . . . . . . . . . . . . 34 87 3.8.8. Negotiation End Message . . . . . . . . . . . . . . . 34 88 3.8.9. Confirm Waiting Message . . . . . . . . . . . . . 34 89 3.8.10. Synchronization Message . . . . . . . . . . . . . . . 35 90 3.8.11. Flood Synchronization Message . . . . . . . . . . . . 35 91 3.8.12. Invalid Message . . . . . . . . . . . . . . . . . . . 36 92 3.8.13. No Operation Message . . . . . . . . . . . . . . . . 36 93 3.9. GRASP Options . . . . . . . . . . . . . . . . . . . . . . 36 94 3.9.1. Format of GRASP Options . . . . . . . . . . . . . . . 37 95 3.9.2. Divert Option . . . . . . . . . . . . . . . . . . . . 37 96 3.9.3. Accept Option . . . . . . . . . . . . . . . . . . . . 37 97 3.9.4. Decline Option . . . . . . . . . . . . . . . . . . . 37 98 3.9.5. Locator Options . . . . . . . . . . . . . . . . . . . 38 99 3.10. Objective Options . . . . . . . . . . . . . . . . . . . . 40 100 3.10.1. Format of Objective Options . . . . . . . . . . . . 40 101 3.10.2. Objective flags . . . . . . . . . . . . . . . . . . 41 102 3.10.3. General Considerations for Objective Options . . . . 41 103 3.10.4. Organizing of Objective Options . . . . . . . . . . 42 104 3.10.5. Experimental and Example Objective Options . . . . . 44 105 4. Implementation Status [RFC Editor: please remove] . . . . . . 44 106 4.1. BUPT C++ Implementation . . . . . . . . . . . . . . . . . 44 107 4.2. Python Implementation . . . . . . . . . . . . . . . . . . 45 108 5. Security Considerations . . . . . . . . . . . . . . . . . . . 46 109 6. CDDL Specification of GRASP . . . . . . . . . . . . . . . . . 48 110 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 50 111 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 52 112 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 52 113 9.1. Normative References . . . . . . . . . . . . . . . . . . 52 114 9.2. Informative References . . . . . . . . . . . . . . . . . 53 115 Appendix A. Open Issues [RFC Editor: This section should be 116 empty. Please remove] . . . . . . . . . . . . . . . 57 117 Appendix B. Closed Issues [RFC Editor: Please remove] . . . . . 57 118 Appendix C. Change log [RFC Editor: Please remove] . . . . . . . 65 119 Appendix D. Example Message Formats . . . . . . . . . . . . . . 72 120 D.1. Discovery Example . . . . . . . . . . . . . . . . . . . . 72 121 D.2. Flood Example . . . . . . . . . . . . . . . . . . . . . . 73 122 D.3. Synchronization Example . . . . . . . . . . . . . . . . . 73 123 D.4. Simple Negotiation Example . . . . . . . . . . . . . . . 73 124 D.5. Complete Negotiation Example . . . . . . . . . . . . . . 74 125 Appendix E. Capability Analysis of Current Protocols . . . . . . 75 126 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 77 128 1. Introduction 130 The success of the Internet has made IP-based networks bigger and 131 more complicated. Large-scale ISP and enterprise networks have 132 become more and more problematic for human based management. Also, 133 operational costs are growing quickly. Consequently, there are 134 increased requirements for autonomic behavior in the networks. 135 General aspects of autonomic networks are discussed in [RFC7575] and 136 [RFC7576]. 138 One approach is to largely decentralize the logic of network 139 management by migrating it into network elements. A reference model 140 for autonomic networking on this basis is given in 141 [I-D.ietf-anima-reference-model]. The reader should consult this 142 document to understand how various autonomic components fit together. 143 In order to fulfill autonomy, devices that embody Autonomic Service 144 Agents (ASAs, [RFC7575]) have specific signaling requirements. In 145 particular they need to discover each other, to synchronize state 146 with each other, and to negotiate parameters and resources directly 147 with each other. There is no limitation on the types of parameters 148 and resources concerned, which can include very basic information 149 needed for addressing and routing, as well as anything else that 150 might be configured in a conventional non-autonomic network. The 151 atomic unit of discovery, synchronization or negotiation is referred 152 to as a technical objective, i.e, a configurable parameter or set of 153 parameters (defined more precisely in Section 3.1). 155 Following this Introduction, Section 2 describes the requirements for 156 discovery, synchronization and negotiation. Negotiation is an 157 iterative process, requiring multiple message exchanges forming a 158 closed loop between the negotiating entities. In fact, these 159 entities are ASAs, normally but not necessarily in different network 160 devices. State synchronization, when needed, can be regarded as a 161 special case of negotiation, without iteration. Section 3.3 162 describes a behavior model for a protocol intended to support 163 discovery, synchronization and negotiation. The design of GeneRic 164 Autonomic Signaling Protocol (GRASP) in Section 3 of this document is 165 based on this behavior model. The relevant capabilities of various 166 existing protocols are reviewed in Appendix E. 168 The proposed discovery mechanism is oriented towards synchronization 169 and negotiation objectives. It is based on a neighbor discovery 170 process on the local link, but also supports diversion to peers on 171 other links. There is no assumption of any particular form of 172 network topology. When a device starts up with no pre-configuration, 173 it has no knowledge of the topology. The protocol itself is capable 174 of being used in a small and/or flat network structure such as a 175 small office or home network as well as in a large professionally 176 managed network. Therefore, the discovery mechanism needs to be able 177 to allow a device to bootstrap itself without making any prior 178 assumptions about network structure. 180 Because GRASP can be used as part of a decision process among 181 distributed devices or between networks, it must run in a secure and 182 strongly authenticated environment. 184 In realistic deployments, not all devices will support GRASP. 185 Therefore, some autonomic service agents will directly manage a group 186 of non-autonomic nodes, and other non-autonomic nodes will be managed 187 traditionally. Such mixed scenarios are not discussed in this 188 specification. 190 2. Requirement Analysis of Discovery, Synchronization and Negotiation 192 This section discusses the requirements for discovery, negotiation 193 and synchronization capabilities. The primary user of the protocol 194 is an autonomic service agent (ASA), so the requirements are mainly 195 expressed as the features needed by an ASA. A single physical device 196 might contain several ASAs, and a single ASA might manage several 197 technical objectives. If a technical objective is managed by several 198 ASAs, any necessary coordination is outside the scope of the GRASP 199 signaling protocol. Furthermore, requirements for ASAs themselves, 200 such as the processing of Intent [RFC7575], are out of scope for the 201 present document. 203 2.1. Requirements for Discovery 205 D1. ASAs may be designed to manage any type of configurable device 206 or software, as required in Section 2.2. A basic requirement is 207 therefore that the protocol can represent and discover any kind of 208 technical objective among arbitrary subsets of participating nodes. 210 In an autonomic network we must assume that when a device starts up 211 it has no information about any peer devices, the network structure, 212 or what specific role it must play. The ASA(s) inside the device are 213 in the same situation. In some cases, when a new application session 214 starts up within a device, the device or ASA may again lack 215 information about relevant peers. For example, it might be necessary 216 to set up resources on multiple other devices, coordinated and 217 matched to each other so that there is no wasted resource. Security 218 settings might also need updating to allow for the new device or 219 user. The relevant peers may be different for different technical 220 objectives. Therefore discovery needs to be repeated as often as 221 necessary to find peers capable of acting as counterparts for each 222 objective that a discovery initiator needs to handle. From this 223 background we derive the next three requirements: 225 D2. When an ASA first starts up, it may have no knowledge of the 226 specific network to which it is attached. Therefore the discovery 227 process must be able to support any network scenario, assuming only 228 that the device concerned is bootstrapped from factory condition. 230 D3. When an ASA starts up, it must require no configured location 231 information about any peers in order to discover them. 233 D4. If an ASA supports multiple technical objectives, relevant peers 234 may be different for different discovery objectives, so discovery 235 needs to be performed separately to find counterparts for each 236 objective. Thus, there must be a mechanism by which an ASA can 237 separately discover peer ASAs for each of the technical objectives 238 that it needs to manage, whenever necessary. 240 D5. Following discovery, an ASA will normally perform negotiation or 241 synchronization for the corresponding objectives. The design should 242 allow for this by conveniently linking discovery to negotiation and 243 synchronization. It may provide an optional mechanism to combine 244 discovery and negotiation/synchronization in a single protocol 245 exchange. 247 D6. Some objectives may only be significant on the local link, but 248 others may be significant across the routed network and require off- 249 link operations. Thus, the relevant peers might be immediate 250 neighbors on the same layer 2 link, or they might be more distant and 251 only accessible via layer 3. The mechanism must therefore provide 252 both on-link and off-link discovery of ASAs supporting specific 253 technical objectives. 255 D7. The discovery process should be flexible enough to allow for 256 special cases, such as the following: 258 o During initialization, a device must be able to establish mutual 259 trust with the rest of the network and participate in an 260 authentication mechanism. Although this will inevitably start 261 with a discovery action, it is a special case precisely because 262 trust is not yet established. This topic is the subject of 263 [I-D.ietf-anima-bootstrapping-keyinfra]. We require that once 264 trust has been established for a device, all ASAs within the 265 device inherit the device's credentials and are also trusted. 266 This does not preclude the device having multiple credentials. 268 o Depending on the type of network involved, discovery of other 269 central functions might be needed, such as the Network Operations 270 Center (NOC) [I-D.ietf-anima-stable-connectivity]. The protocol 271 must be capable of supporting such discovery during 272 initialization, as well as discovery during ongoing operation. 274 D8. The discovery process must not generate excessive traffic and 275 must take account of sleeping nodes. 277 D9. There must be a mechanism for handling stale discovery results. 279 2.2. Requirements for Synchronization and Negotiation Capability 281 As background, consider the example of routing protocols, the closest 282 approximation to autonomic networking already in widespread use. 283 Routing protocols use a largely autonomic model based on distributed 284 devices that communicate repeatedly with each other. The focus is 285 reachability, so routing protocols primarily consider simple link 286 status and metrics, and an underlying assumption is that nodes need a 287 consistent, although partial, view of the network topology in order 288 for the routing algorithm to converge. Also, routing is mainly based 289 on simple information synchronization between peers, rather than on 290 bi-directional negotiation. 292 By contrast, autonomic networks need to be able to manage many 293 different types of parameter and consider many more dimensions, such 294 as latency, load, unused or limited resources, conflicting resource 295 requests, security settings, power saving, load balancing, etc. 296 Status information and resource metrics need to be shared between 297 nodes for dynamic adjustment of resources and for monitoring 298 purposes. While this might be achieved by existing protocols when 299 they are available, the new protocol needs to be able to support 300 parameter exchange, including mutual synchronization, even when no 301 negotiation as such is required. In general, these parameters do not 302 apply to all participating nodes, but only to a subset. 304 SN1. A basic requirement for the protocol is therefore the ability 305 to represent, discover, synchronize and negotiate almost any kind of 306 network parameter among selected subsets of participating nodes. 308 SN2. Negotiation is an iterative request/response process that must 309 be guaranteed to terminate (with success or failure). While tie- 310 breaking rules must be defined specifically for each use case, the 311 protocol should have some general mechanisms in support of loop and 312 deadlock prevention, such as hop count limits or timeouts. 314 SN3. Synchronization must be possible for groups of nodes ranging 315 from small to very large. 317 SN4. To avoid "reinventing the wheel", the protocol should be able 318 to encapsulate the data formats used by existing configuration 319 protocols (such as NETCONF/YANG) in cases where that is convenient. 321 SN5. Human intervention in complex situations is costly and error- 322 prone. Therefore, synchronization or negotiation of parameters 323 without human intervention is desirable whenever the coordination of 324 multiple devices can improve overall network performance. It follows 325 that the protocol's resource requirements must be appropriate for any 326 device that would otherwise need human intervention. The issue of 327 running in constrained nodes is discussed in 328 [I-D.ietf-anima-reference-model]. 330 SN6. Human intervention in large networks is often replaced by use 331 of a top-down network management system (NMS). It therefore follows 332 that the protocol, as part of the Autonomic Networking 333 Infrastructure, should be capable of running in any device that would 334 otherwise be managed by an NMS, and that it can co-exist with an NMS, 335 and with protocols such as SNMP and NETCONF. 337 SN7. Some features are expected to be implemented by individual 338 ASAs, but the protocol must be general enough to allow them: 340 o Dependencies and conflicts: In order to decide upon a 341 configuration for a given device, the device may need information 342 from neighbors. This can be established through the negotiation 343 procedure, or through synchronization if that is sufficient. 344 However, a given item in a neighbor may depend on other 345 information from its own neighbors, which may need another 346 negotiation or synchronization procedure to obtain or decide. 347 Therefore, there are potential dependencies and conflicts among 348 negotiation or synchronization procedures. Resolving dependencies 349 and conflicts is a matter for the individual ASAs involved. To 350 allow this, there need to be clear boundaries and convergence 351 mechanisms for negotiations. Also some mechanisms are needed to 352 avoid loop dependencies or uncontrolled growth in a tree of 353 dependencies. It is the ASA designer's responsibility to avoid or 354 detect looping dependencies or excessive growth of dependency 355 trees. The protocol's role is limited to bilateral signaling 356 between ASAs, and the avoidance of loops during bilateral 357 signaling. 359 o Recovery from faults and identification of faulty devices should 360 be as automatic as possible. The protocol's role is limited to 361 discovery, synchronization and negotiation. These processes can 362 occur at any time, and an ASA may need to repeat any of these 363 steps when the ASA detects an event such as a negotiation 364 counterpart failing. 366 o Since a major goal is to minimize human intervention, it is 367 necessary that the network can in effect "think ahead" before 368 changing its parameters. One aspect of this is an ASA that relies 369 on a knowledge base to predict network behavior. This is out of 370 scope for the signaling protocol. However, another aspect is 371 forecasting the effect of a change by a "dry run" negotiation 372 before actually installing the change. Signaling a dry run is 373 therefore a desirable feature of the protocol. 375 Note that management logging, monitoring, alerts and tools for 376 intervention are required. However, these can only be features of 377 individual ASAs, not of the protocol itself. Another document 378 [I-D.ietf-anima-stable-connectivity] discusses how such agents may be 379 linked into conventional OAM systems via an Autonomic Control Plane 380 [I-D.ietf-anima-autonomic-control-plane]. 382 SN8. The protocol will be able to deal with a wide variety of 383 technical objectives, covering any type of network parameter. 384 Therefore the protocol will need a flexible and easily extensible 385 format for describing objectives. At a later stage it may be 386 desirable to adopt an explicit information model. One consideration 387 is whether to adopt an existing information model or to design a new 388 one. 390 2.3. Specific Technical Requirements 392 T1. It should be convenient for ASA designers to define new 393 technical objectives and for programmers to express them, without 394 excessive impact on run-time efficiency and footprint. In 395 particular, it should be convenient for ASAs to be implemented 396 independently of each other as user space programs rather than as 397 kernel code, where such a programming model is possible. The classes 398 of device in which the protocol might run is discussed in 399 [I-D.ietf-anima-reference-model]. 401 T2. The protocol should be easily extensible in case the initially 402 defined discovery, synchronization and negotiation mechanisms prove 403 to be insufficient. 405 T3. To be a generic platform, the protocol payload format should be 406 independent of the transport protocol or IP version. In particular, 407 it should be able to run over IPv6 or IPv4. However, some functions, 408 such as multicasting on a link, might need to be IP version 409 dependent. By default, IPv6 should be preferred. 411 T4. The protocol must be able to access off-link counterparts via 412 routable addresses, i.e., must not be restricted to link-local 413 operation. 415 T5. It must also be possible for an external discovery mechanism to 416 be used, if appropriate for a given technical objective. In other 417 words, GRASP discovery must not be a prerequisite for GRASP 418 negotiation or synchronization. 420 T6. The protocol must be capable of distinguishing multiple 421 simultaneous operations with one or more peers, especially when wait 422 states occur. 424 T7. Intent: Although the distribution of Intent is out of scope for 425 this document, the protocol must not by design exclude its use for 426 Intent distribution. 428 T8. Management monitoring, alerts and intervention: Devices should 429 be able to report to a monitoring system. Some events must be able 430 to generate operator alerts and some provision for emergency 431 intervention must be possible (e.g. to freeze synchronization or 432 negotiation in a mis-behaving device). These features might not use 433 the signaling protocol itself, but its design should not exclude such 434 use. 436 T9. Because this protocol may directly cause changes to device 437 configurations and have significant impacts on a running network, all 438 protocol exchanges need to be fully secured against forged messages 439 and man-in-the middle attacks, and secured as much as reasonably 440 possible against denial of service attacks. There must also be an 441 encryption mechanism to resist unwanted monitoring. However, it is 442 not required that the protocol itself provides these security 443 features; it may depend on an existing secure environment. 445 3. GRASP Protocol Overview 447 3.1. Terminology 449 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 450 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 451 "OPTIONAL" in this document are to be interpreted as described in 452 [RFC2119] when they appear in ALL CAPS. When these words are not in 453 ALL CAPS (such as "should" or "Should"), they have their usual 454 English meanings, and are not to be interpreted as [RFC2119] key 455 words. 457 This document uses terminology defined in [RFC7575]. 459 The following additional terms are used throughout this document: 461 o Discovery: a process by which an ASA discovers peers according to 462 a specific discovery objective. The discovery results may be 463 different according to the different discovery objectives. The 464 discovered peers may later be used as negotiation counterparts or 465 as sources of synchronization data. 467 o Negotiation: a process by which two ASAs interact iteratively to 468 agree on parameter settings that best satisfy the objectives of 469 both ASAs. 471 o State Synchronization: a process by which ASAs interact to receive 472 the current state of parameter values stored in other ASAs. This 473 is a special case of negotiation in which information is sent but 474 the ASAs do not request their peers to change parameter settings. 475 All other definitions apply to both negotiation and 476 synchronization. 478 o Technical Objective (usually abbreviated as Objective): A 479 technical objective is a data structure, whose main contents are a 480 name and a value. The value consists of a single configurable 481 parameter or a set of parameters of some kind. The exact format 482 of an objective is defined in Section 3.10.1. An objective occurs 483 in three contexts: Discovery, Negotiation and Synchronization. 484 Normally, a given objective will not occur in negotiation and 485 synchronization contexts simultaneously. 487 * One ASA may support multiple independent objectives. 489 * The parameter(s) in the value of a given objective apply to a 490 specific service or function or action. They may in principle 491 be anything that can be set to a specific logical, numerical or 492 string value, or a more complex data structure, by a network 493 node. Each node is expected to contain one or more ASAs which 494 may each manage subsidiary non-autonomic nodes. 496 * Discovery Objective: an objective in the process of discovery. 497 Its value may be undefined. 499 * Synchronization Objective: an objective whose specific 500 technical content needs to be synchronized among two or more 501 ASAs. Thus, each ASA will maintain its own copy of the 502 objective. 504 * Negotiation Objective: an objective whose specific technical 505 content needs to be decided in coordination with another ASA. 506 Again, each ASA will maintain its own copy of the objective. 508 A detailed discussion of objectives, including their format, is 509 found in Section 3.10. 511 o Discovery Initiator: an ASA that starts discovery by sending a 512 discovery message referring to a specific discovery objective. 514 o Discovery Responder: a peer that either contains an ASA supporting 515 the discovery objective indicated by the discovery initiator, or 516 caches the locator(s) of the ASA(s) supporting the objective. It 517 sends a Discovery Response, as described later. 519 o Synchronization Initiator: an ASA that starts synchronization by 520 sending a request message referring to a specific synchronization 521 objective. 523 o Synchronization Responder: a peer ASA which responds with the 524 value of a synchronization objective. 526 o Negotiation Initiator: an ASA that starts negotiation by sending a 527 request message referring to a specific negotiation objective. 529 o Negotiation Counterpart: a peer with which the Negotiation 530 Initiator negotiates a specific negotiation objective. 532 o GRASP Instance: This refers to an instantiation of a GRASP 533 protocol engine, likely including multiple threads or processes as 534 well as dynamic data structures such as a discovery cache, running 535 in a given security environment on a single device. 537 o Interface or GRASP Interface: Unless otherwise stated, these refer 538 to a network interface - which might be physical or virtual - that 539 a specific instance of GRASP is currently using. A device might 540 have other interfaces that are not used by GRASP and which are 541 outside the scope of the autonomic network. 543 3.2. High Level Deployment Model 545 A GRASP implementation will be part of the Autonomic Networking 546 Infrastructure in an autonomic node, which must also provide an 547 appropriate security environment. In accordance with 548 [I-D.ietf-anima-reference-model], this SHOULD be the Autonomic 549 Control Plane (ACP) [I-D.ietf-anima-autonomic-control-plane]. It is 550 expected that GRASP will access the ACP by using a typical socket 551 programming interface and the ACP will make available only network 552 interfaces within the autonomic network. If there is no ACP, the 553 considerations described in Section 3.5.1 apply. 555 There will also be one or more Autonomic Service Agents (ASAs). In 556 the minimal case of a single-purpose device, these components might 557 be fully integrated with GRASP and the ACP. A more common model is 558 expected to be a multi-purpose device capable of containing several 559 ASAs. In this case it is expected that the ACP, GRASP and the ASAs 560 will be implemented as separate processes, which are probably multi- 561 threaded to support asynchronous and simultaneous operations. 563 In some scenarios, a limited negotiation model might be deployed 564 based on a limited trust relationship such as that between two 565 administrative domains. ASAs might then exchange limited information 566 and negotiate some particular configurations. 568 GRASP is explicitly designed to operate within a single addressing 569 realm. Its discovery and flooding mechanisms do not support 570 autonomic operations that cross any form of address translator or 571 upper layer proxy. 573 A suitable Application Programming Interface (API) will be needed 574 between GRASP and the ASAs. In some implementations, ASAs would run 575 in user space with a GRASP library providing the API, and this 576 library would in turn communicate via system calls with core GRASP 577 functions. Details of the API are out of scope for the present 578 document. For further details of possible deployment models, see 579 [I-D.ietf-anima-reference-model]. 581 An instance of GRASP must be aware of the network interfaces it will 582 use, and of the appropriate global-scope and link-local addresses. 583 In the presence of the ACP, such information will be available from 584 the adjacency table discussed in [I-D.ietf-anima-reference-model]. 585 In other cases, GRASP must determine such information for itself. 586 Details depend on the device and operating system. In the rest of 587 this document, the terms 'interfaces' or 'GRASP interfaces' refers 588 only to the set of network interfaces that a specific instance of 589 GRASP is currently using. 591 Because GRASP needs to work with very high reliability, especially 592 during bootstrapping and during fault conditions, it is essential 593 that every implementation is as robust as possible. For example, 594 discovery failures, or any kind of socket exception at any time, must 595 not cause irrecoverable failures in GRASP itself, and must return 596 suitable error codes through the API so that ASAs can also recover. 598 GRASP must not depend upon non-volatile data storage. All run time 599 error conditions, and events such as address renumbering, network 600 interface failures, and CPU sleep/wake cycles, must be handled in 601 such a way that GRASP will still operate correctly and securely 602 (Section 3.5.1) afterwards. 604 An autonomic node will normally run a single instance of GRASP, used 605 by multiple ASAs. Possible exceptions are mentioned below. 607 3.3. High Level Design Choices 609 This section describes a behavior model and design choices for GRASP, 610 supporting discovery, synchronization and negotiation, to act as a 611 platform for different technical objectives. 613 o A generic platform: 615 The protocol design is generic and independent of the 616 synchronization or negotiation contents. The technical contents 617 will vary according to the various technical objectives and the 618 different pairs of counterparts. 620 o Normally, a single main instance of the GRASP protocol engine will 621 exist in an autonomic node, and each ASA will run as an 622 independent asynchronous process. However, scenarios where 623 multiple instances of GRASP run in a single node, perhaps with 624 different security properties, are possible (Section 3.5.2). In 625 this case, each instance MUST listen independently for GRASP link- 626 local multicasts, and all instances MUST be woken by each such 627 multicast, in order for discovery and flooding to work correctly. 629 o Security infrastructure: 631 As noted above, the protocol itself has no built-in security 632 functionality, and relies on a separate secure infrastructure. 634 o Discovery, synchronization and negotiation are designed together: 636 The discovery method and the synchronization and negotiation 637 methods are designed in the same way and can be combined when this 638 is useful, allowing a rapid mode of operation described in 639 Section 3.5.4. These processes can also be performed 640 independently when appropriate. 642 * Thus, for some objectives, especially those concerned with 643 application layer services, another discovery mechanism such as 644 the future DNS Service Discovery [RFC7558] MAY be used. The 645 choice is left to the designers of individual ASAs. 647 o A uniform pattern for technical objectives: 649 The synchronization and negotiation objectives are defined 650 according to a uniform pattern. The values that they contain 651 could be carried either in a simple binary format or in a complex 652 object format. The basic protocol design uses the Concise Binary 653 Object Representation (CBOR) [RFC7049], which is readily 654 extensible for unknown future requirements. 656 o A flexible model for synchronization: 658 GRASP supports synchronization between two nodes, which could be 659 used repeatedly to perform synchronization among a small number of 660 nodes. It also supports an unsolicited flooding mode when large 661 groups of nodes, possibly including all autonomic nodes, need data 662 for the same technical objective. 664 * There may be some network parameters for which a more 665 traditional flooding mechanism such as DNCP [RFC7787] is 666 considered more appropriate. GRASP can coexist with DNCP. 668 o A simple initiator/responder model for negotiation: 670 Multi-party negotiations are very complicated to model and cannot 671 readily be guaranteed to converge. GRASP uses a simple bilateral 672 model and can support multi-party negotiations by indirect steps. 674 o Organizing of synchronization or negotiation content: 676 The technical content transmitted by GRASP will be organized 677 according to the relevant function or service. The objectives for 678 different functions or services are kept separate, because they 679 may be negotiated or synchronized with different counterparts or 680 have different response times. Thus a normal arrangement would be 681 a single ASA managing a small set of closely related objectives, 682 with a version of that ASA in each relevant autonomic node. 683 Further discussion of this aspect is out of scope for the current 684 document. 686 o Requests and responses in negotiation procedures: 688 The initiator can negotiate a specific negotiation objective with 689 relevant counterpart ASAs. It can request relevant information 690 from a counterpart so that it can coordinate its local 691 configuration. It can request the counterpart to make a matching 692 configuration. It can request simulation or forecast results by 693 sending some dry run conditions. 695 Beyond the traditional yes/no answer, the responder can reply with 696 a suggested alternative value for the objective concerned. This 697 would start a bi-directional negotiation ending in a compromise 698 between the two ASAs. 700 o Convergence of negotiation procedures: 702 To enable convergence, when a responder suggests a new value or 703 condition in a negotiation step reply, it should be as close as 704 possible to the original request or previous suggestion. The 705 suggested value of later negotiation steps should be chosen 706 between the suggested values from the previous two steps. GRASP 707 provides mechanisms to guarantee convergence (or failure) in a 708 small number of steps, namely a timeout and a maximum number of 709 iterations. 711 o Extensibility: 713 GRASP does not have a version number, and could be extended by 714 adding new message types and options. In normal use, new 715 semantics will be added by defining new synchronization or 716 negotiation objectives. 718 3.4. Quick Operating Overview 720 An instance of GRASP is expected to run as a separate core module, 721 providing an API (such as [I-D.liu-anima-grasp-api]) to interface to 722 various ASAs. These ASAs may operate without special privilege, 723 unless they need it for other reasons (such as configuring IP 724 addresses or manipulating routing tables). 726 The GRASP mechanisms used by the ASA are built around GRASP 727 objectives defined as data structures containing administrative 728 information such as the objective's unique name, and its current 729 value. The format and size of the value is not restricted by the 730 protocol, except that it must be possible to serialise it for 731 transmission in CBOR, which is no restriction at all in practice. 733 GRASP provides the following mechanisms: 735 o A discovery mechanism (M_DISCOVERY, M_RESPONSE), by which an ASA 736 can discover other ASAs supporting a given objective. 738 o A negotiation request mechanism (M_REQ_NEG), by which an ASA can 739 start negotiation of an objective with a counterpart ASA. Once a 740 negotiation has started, the process is symmetrical, and there is 741 a negotiation step message (M_NEGOTIATE) for each ASA to use in 742 turn. Two other functions support negotiating steps (M_WAIT, 743 M_END). 745 o A synchronization mechanism (M_REQ_SYN), by which an ASA can 746 request the current value of an objective from a counterpart ASA. 747 With this, there is a corresponding response function (M_SYNCH) 748 for an ASA that wishes to respond to synchronization requests. 750 o A flood mechanism (M_FLOOD), by which an ASA can cause the current 751 value of an objective to be flooded throughout the autonomic 752 network so that any ASA can receive it. One application of this 753 is to act as an announcement, avoiding the need for discovery of a 754 widely applicable objective. 756 Some example messages and simple message flows are provided in 757 Appendix D. 759 3.5. GRASP Protocol Basic Properties and Mechanisms 761 3.5.1. Required External Security Mechanism 763 The protocol SHOULD always run within a secure Autonomic Control 764 Plane (ACP) [I-D.ietf-anima-autonomic-control-plane]. The ACP is 765 assumed to carry all messages securely, including link-local 766 multicast when it is virtualized over the ACP. A GRASP instance MUST 767 verify whether the ACP is operational. 769 If there is no ACP, one of the following alternatives applies: 771 1. The protocol instance MUST use another form of strong 772 authentication and a form of strong encryption MUST be 773 implemented. An exception is that during initialization of nodes 774 there will be a transition period during which it might not be 775 practical to run with strong encryption. This period MUST be as 776 short as possible, changing to a fully secure setup as soon as 777 possible. See Section 3.5.2.1 for further discussion. 779 2. The protocol instance MUST operate as described in 780 Section 3.5.2.2 or Section 3.5.2.3. 782 Network interfaces could be at different security levels, for example 783 being part of the ACP or not. All the interfaces supported by a 784 given GRASP instance MUST be at the same security level. 786 The ACP, or in its absence another security mechanism, sets the 787 boundary within which nodes are trusted as GRASP peers. A GRASP 788 implementation MUST refuse to execute GRASP synchronization and 789 negotiation functions if there is neither an operational ACP nor 790 another secure environment. 792 Link-local multicast is used for discovery messages. Responses to 793 discovery messages MUST be secured, with one exception mentioned in 794 the next section. 796 3.5.2. Constrained Instances 798 This section describes some cases where additional instances of GRASP 799 subject to certain constraints are appropriate. 801 3.5.2.1. No ACP 803 As mentioned in Section 3.3, some GRASP operations might be performed 804 across an administrative domain boundary by mutual agreement, without 805 the benefit of an ACP. Such operations MUST be confined to a 806 separate instance of GRASP with its own copy of all GRASP data 807 structures. Messages MUST be authenticated and encryption MUST be 808 implemented. TLS [RFC5246] and DTLS [RFC6347] based on a Public Key 809 Infrastructure (PKI) [RFC5280] are RECOMMENDED for this purpose. 810 Further details are out of scope for this document. 812 3.5.2.2. Discovery Unsolicited Link-Local 814 Some services may need to use insecure GRASP discovery, response and 815 flood messages without being able to use pre-existing security 816 associations. Such operations being intrinsically insecure, they 817 need to be confined to link-local use to minimize the risk of 818 malicious actions. Possible examples include discovery of candidate 819 ACP neighbors [I-D.ietf-anima-autonomic-control-plane], discovery of 820 bootstrap proxies [I-D.ietf-anima-bootstrapping-keyinfra] or perhaps 821 initialization services in networks using GRASP without being fully 822 autonomic (e.g., no ACP). Such usage MUST be limited to link-local 823 operations and MUST be confined to a separate insecure instance of 824 GRASP with its own copy of all GRASP data structures. This instance 825 is nicknamed DULL - Discovery Unsolicited Link-Local. 827 The detailed rules for the DULL instance of GRASP are as follows: 829 o An initiator MUST only send Discovery or Flood Synchronization 830 link-local multicast messages with a loop count of 1. Other GRASP 831 message types MUST NOT be sent. 833 o A responder MUST silently discard any message whose loop count is 834 not 1. 836 o A responder MUST silently discard any message referring to a GRASP 837 Objective that is not directly part of a service that requires 838 this insecure mode. 840 o A responder MUST NOT relay any multicast messages. 842 o A Discovery Response MUST indicate a link-local address. 844 o A Discovery Response MUST NOT include a Divert option. 846 o A node MUST silently discard any message whose source address is 847 not link-local. 849 To minimize traffic possibly observed by third parties, GRASP traffic 850 SHOULD be minimized by using only Flood Synchronization to announce 851 objectives and their associated locators, rather than by using 852 Discovery and Response. Further details are out of scope for this 853 document 855 3.5.2.3. Secure Only Neighbor Negotiation 857 Some services might use insecure on-link operations as in DULL, but 858 also use unicast synchronization or negotiation operations protected 859 by TLS. A separate instance of GRASP is used, with its own copy of 860 all GRASP data structures. This instance is nicknamed SONN - Secure 861 Only Neighbor Negotiation. 863 The detailed rules for the SONN instance of GRASP are as follows: 865 o All types of GRASP message are permitted. 867 o An initiator MUST send any Discovery or Flood Synchronization 868 link-local multicast messages with a loop count of 1. 870 o A responder MUST silently discard any Discovery or Flood 871 Synchronization message whose loop count is not 1. 873 o A responder MUST silently discard any message referring to a GRASP 874 Objective that is not directly part of the service concerned. 876 o A responder MUST NOT relay any multicast messages. 878 o A Discovery Response MUST indicate a link-local address. 880 o A Discovery Response MUST NOT include a Divert option. 882 o A node MUST silently discard any message whose source address is 883 not link-local. 885 Further details are out of scope for this document. 887 3.5.3. Transport Layer Usage 889 GRASP discovery and flooding messages are designed for use over link- 890 local multicast UDP. They MUST NOT be fragmented, and therefore MUST 891 NOT exceed the link MTU size. 893 All other GRASP messages are unicast and could in principle run over 894 any transport protocol. An implementation MUST support use of TCP. 895 It MAY support use of another transport protocol but the details are 896 out of scope for this specification. However, GRASP itself does not 897 provide for error detection or retransmission. Use of an unreliable 898 transport protocol is therefore NOT RECOMMENDED. 900 For considerations when running without an ACP, see Section 3.5.2.1. 902 For link-local multicast, the GRASP protocol listens to the well- 903 known GRASP Listen Port (Section 3.6). For unicast transport 904 sessions used for discovery responses, synchronization and 905 negotiation, the ASA concerned normally listens on its own 906 dynamically assigned ports, which are communicated to its peers 907 during discovery. However, a minimal implementation MAY use the 908 GRASP Listen Port for this purpose. 910 3.5.4. Discovery Mechanism and Procedures 912 3.5.4.1. Separated discovery and negotiation mechanisms 914 Although discovery and negotiation or synchronization are defined 915 together in GRASP, they are separate mechanisms. The discovery 916 process could run independently from the negotiation or 917 synchronization process. Upon receiving a Discovery (Section 3.8.4) 918 message, the recipient node should return a response message in which 919 it either indicates itself as a discovery responder or diverts the 920 initiator towards another more suitable ASA. However, this response 921 may be delayed if the recipient needs to relay the discovery onwards, 922 as described below. 924 The discovery action (M_DISCOVERY) will normally be followed by a 925 negotiation (M_REQ_NEG) or synchronization (M_REQ_SYN) action. The 926 discovery results could be utilized by the negotiation protocol to 927 decide which ASA the initiator will negotiate with. 929 The initiator of a discovery action for a given objective need not be 930 capable of responding to that objective as a Negotiation Counterpart, 931 as a Synchronization Responder or as source for flooding. For 932 example, an ASA might perform discovery even if it only wishes to act 933 a Synchronization Initiator or Negotiation Initiator. Such an ASA 934 does not itself need to respond to discovery messages. 936 It is also entirely possible to use GRASP discovery without any 937 subsequent negotiation or synchronization action. In this case, the 938 discovered objective is simply used as a name during the discovery 939 process and any subsequent operations between the peers are outside 940 the scope of GRASP. 942 3.5.4.2. Discovery Overview 944 A complete discovery process will start with a multicast (of 945 M_DISCOVERY) on the local link. On-link neighbors supporting the 946 discovery objective will respond directly (with M_RESPONSE). A 947 neighbor with multiple interfaces will respond with a cached 948 discovery response if any. However, it SHOULD NOT respond with a 949 cached response on an interface if it learnt that information from 950 the same interface, because the peer in question will answer directly 951 if still operational. If it has no cached response, it will relay 952 the discovery on its other GRASP interfaces, for example reaching a 953 higher-level gateway in a hierarchical network. If a node receiving 954 the relayed discovery supports the discovery objective, it will 955 respond to the relayed discovery. If it has a cached response, it 956 will respond with that. If not, it will repeat the discovery 957 process, which thereby becomes iterative. The loop count and timeout 958 will ensure that the process ends. 960 A Discovery message MAY be sent unicast (via UDP or TCP) to a peer 961 node, which SHOULD then proceed exactly as if the message had been 962 multicast, except that when TCP is used, the response will be on the 963 same socket as the query. However, this mode does not guarantee 964 successful discovery in the general case. 966 3.5.4.3. Discovery Procedures 968 Discovery starts as an on-link operation. The Divert option can tell 969 the discovery initiator to contact an off-link ASA for that discovery 970 objective. A Discovery message is sent by a discovery initiator via 971 UDP to the ALL_GRASP_NEIGHBORS link-local multicast address 972 (Section 3.6). Every network device that supports GRASP always 973 listens to a well-known UDP port to capture the discovery messages. 974 Because this port is unique in a device, this is a function of the 975 GRASP instance and not of an individual ASA. As a result, each ASA 976 will need to register the objectives that it supports with the local 977 GRASP instance. 979 If an ASA in a neighbor device supports the requested discovery 980 objective, the device SHOULD respond to the link-local multicast with 981 a unicast Discovery Response message (Section 3.8.5) with locator 982 option(s), unless it is temporarily unavailable. Otherwise, if the 983 neighbor has cached information about an ASA that supports the 984 requested discovery objective (usually because it discovered the same 985 objective before), it SHOULD respond with a Discovery Response 986 message with a Divert option pointing to the appropriate Discovery 987 Responder. 989 If a device has no information about the requested discovery 990 objective, and is not acting as a discovery relay (see below) it MUST 991 silently discard the Discovery message. 993 If no discovery response is received within a reasonable timeout 994 (default GRASP_DEF_TIMEOUT milliseconds, Section 3.6), the Discovery 995 message MAY be repeated, with a newly generated Session ID 996 (Section 3.7). An exponential backoff SHOULD be used for subsequent 997 repetitions, to limit the load during busy periods. Frequent 998 repetition might be symptomatic of a denial of service attack. 1000 After a GRASP device successfully discovers a locator for a Discovery 1001 Responder supporting a specific objective, it MUST cache this 1002 information, including the interface index via which it was 1003 discovered. This cache record MAY be used for future negotiation or 1004 synchronization, and the locator SHOULD be passed on when appropriate 1005 as a Divert option to another Discovery Initiator. 1007 The cache mechanism MUST include a lifetime for each entry. The 1008 lifetime is derived from a time-to-live (ttl) parameter in each 1009 Discovery Response message. Cached entries MUST be ignored or 1010 deleted after their lifetime expires. In some environments, 1011 unplanned address renumbering might occur. In such cases, the 1012 lifetime SHOULD be short compared to the typical address lifetime and 1013 a mechanism to flush the discovery cache MUST be implemented. The 1014 discovery mechanism needs to track the node's current address to 1015 ensure that Discovery Responses always indicate the correct address. 1017 If multiple Discovery Responders are found for the same objective, 1018 they SHOULD all be cached, unless this creates a resource shortage. 1019 The method of choosing between multiple responders is an 1020 implementation choice. This choice MUST be available to each ASA but 1021 the GRASP implementation SHOULD provide a default choice. 1023 Because Discovery Responders will be cached in a finite cache, they 1024 might be deleted at any time. In this case, discovery will need to 1025 be repeated. If an ASA exits for any reason, its locator might still 1026 be cached for some time, and attempts to connect to it will fail. 1027 ASAs need to be robust in these circumstances. 1029 3.5.4.4. Discovery Relaying 1031 A GRASP instance with multiple link-layer interfaces (typically 1032 running in a router) MUST support discovery on all GRASP interfaces. 1033 We refer to this as a 'relaying instance'. 1035 Constrained Instances (Section 3.5.2) are always single-interface 1036 instances and therefore MUST NOT perform discovery relaying. 1038 If a relaying instance receives a Discovery message on a given 1039 interface for a specific objective that it does not support and for 1040 which it has not previously cached a Discovery Responder, it MUST 1041 relay the query by re-issuing a new Discovery message as a link-local 1042 multicast on its other GRASP interfaces. 1044 The relayed discovery message MUST have the same Session ID as the 1045 incoming discovery message and MUST be tagged with the IP address of 1046 its original initiator (see Section 3.8.4). Note that this initiator 1047 address is only used to allow for disambiguation of the Session ID 1048 and is never used to address Response packets, which are sent to the 1049 relaying instance, not the original initiator. 1051 Since the relay device is unaware of the timeout set by the original 1052 initiator it SHOULD set a timeout at least equal to GRASP_DEF_TIMEOUT 1053 milliseconds. 1055 The relaying instance MUST decrement the loop count within the 1056 objective, and MUST NOT relay the Discovery message if the result is 1057 zero. Also, it MUST limit the total rate at which it relays 1058 discovery messages to a reasonable value, in order to mitigate 1059 possible denial of service attacks. It MUST cache the Session ID 1060 value and initiator address of each relayed Discovery message until 1061 any Discovery Responses have arrived or the discovery process has 1062 timed out. To prevent loops, it MUST NOT relay a Discovery message 1063 which carries a given cached Session ID and initiator address more 1064 than once. These precautions avoid discovery loops and mitigate 1065 potential overload. 1067 The discovery results received by the relaying instance MUST in turn 1068 be sent as a Discovery Response message to the Discovery message that 1069 caused the relay action. 1071 3.5.4.5. Rapid Mode (Discovery/Negotiation binding) 1073 A Discovery message MAY include a Negotiation Objective option. This 1074 allows a rapid mode of negotiation described in Section 3.5.5. A 1075 similar mechanism is defined for synchronization in Section 3.5.6. 1077 Note that rapid mode is currently limited to a single objective for 1078 simplicity of design and implementation. A possible future extension 1079 is to allow multiple objectives in rapid mode for greater efficiency. 1081 3.5.5. Negotiation Procedures 1083 A negotiation initiator sends a negotiation request (using M_REQ_NEG) 1084 to a counterpart ASA, including a specific negotiation objective. It 1085 may request the negotiation counterpart to make a specific 1086 configuration. Alternatively, it may request a certain simulation or 1087 forecast result by sending a dry run configuration. The details, 1088 including the distinction between a dry run and a live configuration 1089 change, will be defined separately for each type of negotiation 1090 objective. Any state associated with a dry run operation, such as 1091 temporarily reserving a resource for subsequent use in a live run, is 1092 entirely a matter for the designer of the ASA concerned. 1094 Each negotiation session as a whole is subject to a timeout (default 1095 GRASP_DEF_TIMEOUT milliseconds, Section 3.6), initialised when the 1096 request is sent (see Section 3.8.6). If no reply message of any kind 1097 is received within a reasonable timeout, the negotiation request MAY 1098 be repeated, with a newly generated Session ID (Section 3.7). An 1099 exponential backoff SHOULD be used for subsequent repetitions. 1101 If the counterpart can immediately apply the requested configuration, 1102 it will give an immediate positive (O_ACCEPT) answer (using M_END). 1103 This will end the negotiation phase immediately. Otherwise, it will 1104 negotiate (using M_NEGOTIATE). It will reply with a proposed 1105 alternative configuration that it can apply (typically, a 1106 configuration that uses fewer resources than requested by the 1107 negotiation initiator). This will start a bi-directional negotiation 1108 (using M_NEGOTIATE) to reach a compromise between the two ASAs. 1110 The negotiation procedure is ended when one of the negotiation peers 1111 sends a Negotiation Ending (M_END) message, which contains an accept 1112 (O_ACCEPT) or decline (O_DECLINE) option and does not need a response 1113 from the negotiation peer. Negotiation may also end in failure 1114 (equivalent to a decline) if a timeout is exceeded or a loop count is 1115 exceeded. 1117 A negotiation procedure concerns one objective and one counterpart. 1118 Both the initiator and the counterpart may take part in simultaneous 1119 negotiations with various other ASAs, or in simultaneous negotiations 1120 about different objectives. Thus, GRASP is expected to be used in a 1121 multi-threaded mode. Certain negotiation objectives may have 1122 restrictions on multi-threading, for example to avoid over-allocating 1123 resources. 1125 Some configuration actions, for example wavelength switching in 1126 optical networks, might take considerable time to execute. The ASA 1127 concerned needs to allow for this by design, but GRASP does allow for 1128 a peer to insert latency in a negotiation process if necessary 1129 (Section 3.8.9, M_WAIT). 1131 3.5.5.1. Rapid Mode (Discovery/Negotiation Linkage) 1133 A Discovery message MAY include a Negotiation Objective option. In 1134 this case it is as if the initiator sent the sequence M_DISCOVERY, 1135 immediately followed by M_REQ_NEG. This has implications for the 1136 construction of the GRASP core, as it must carefully pass the 1137 contents of the Negotiation Objective option to the ASA so that it 1138 may evaluate the objective directly. When a Negotiation Objective 1139 option is present the ASA replies with an M_NEGOTIATE message (or 1140 M_END with O_ACCEPT if it is immediately satisfied with the 1141 proposal), rather than with an M_RESPONSE. However, if the recipient 1142 node does not support rapid mode, discovery will continue normally. 1144 It is possible that a Discovery Response will arrive from a responder 1145 that does not support rapid mode, before such a Negotiation message 1146 arrives. In this case, rapid mode will not occur. 1148 This rapid mode could reduce the interactions between nodes so that a 1149 higher efficiency could be achieved. However, a network in which 1150 some nodes support rapid mode and others do not will have complex 1151 timing-dependent behaviors. Therefore, the rapid negotiation 1152 function SHOULD be disabled by default. 1154 3.5.6. Synchronization and Flooding Procedures 1156 3.5.6.1. Unicast Synchronization 1158 A synchronization initiator sends a synchronization request to a 1159 counterpart, including a specific synchronization objective. The 1160 counterpart responds with a Synchronization message (Section 3.8.10) 1161 containing the current value of the requested synchronization 1162 objective. No further messages are needed. 1164 If no reply message of any kind is received within a reasonable 1165 timeout (default GRASP_DEF_TIMEOUT milliseconds, Section 3.6), the 1166 synchronization request MAY be repeated, with a newly generated 1167 Session ID (Section 3.7). An exponential backoff SHOULD be used for 1168 subsequent repetitions. 1170 3.5.6.2. Flooding 1172 In the case just described, the message exchange is unicast and 1173 concerns only one synchronization objective. For large groups of 1174 nodes requiring the same data, synchronization flooding is available. 1175 For this, a flooding initiator MAY send an unsolicited Flood 1176 Synchronization message containing one or more Synchronization 1177 Objective option(s), if and only if the specification of those 1178 objectives permits it. This is sent as a multicast message to the 1179 ALL_GRASP_NEIGHBORS multicast address (Section 3.6). 1181 Receiving flood multicasts is a function of the GRASP core, as in the 1182 case of discovery multicasts (Section 3.5.4.3). 1184 To ensure that flooding does not result in a loop, the originator of 1185 the Flood Synchronization message MUST set the loop count in the 1186 objectives to a suitable value (the default is GRASP_DEF_LOOPCT). 1187 Also, a suitable mechanism is needed to avoid excessive multicast 1188 traffic. This mechanism MUST be defined as part of the specification 1189 of the synchronization objective(s) concerned. It might be a simple 1190 rate limit or a more complex mechanism such as the Trickle algorithm 1191 [RFC6206]. 1193 A GRASP device with multiple link-layer interfaces (typically a 1194 router) MUST support synchronization flooding on all GRASP 1195 interfaces. If it receives a multicast Flood Synchronization message 1196 on a given interface, it MUST relay it by re-issuing a Flood 1197 Synchronization message as a link-local multicast on its other GRASP 1198 interfaces. The relayed message MUST have the same Session ID as the 1199 incoming message and MUST be tagged with the IP address of its 1200 original initiator. 1202 Link-layer Flooding is supported by GRASP by setting the loop count 1203 to 1, and sending with a link-local source address. Floods with 1204 link-local source addresses and a loop count other than 1 are 1205 invalid, and such messages MUST be discarded. 1207 The relaying device MUST decrement the loop count within the first 1208 objective, and MUST NOT relay the Flood Synchronization message if 1209 the result is zero. Also, it MUST limit the total rate at which it 1210 relays Flood Synchronization messages to a reasonable value, in order 1211 to mitigate possible denial of service attacks. It MUST cache the 1212 Session ID value and initiator address of each relayed Flood 1213 Synchronization message for a time not less than twice 1214 GRASP_DEF_TIMEOUT milliseconds. To prevent loops, it MUST NOT relay 1215 a Flood Synchronization message which carries a given cached Session 1216 ID and initiator address more than once. These precautions avoid 1217 synchronization loops and mitigate potential overload. 1219 Note that this mechanism is unreliable in the case of sleeping nodes, 1220 or new nodes that join the network, or nodes that rejoin the network 1221 after a fault. An ASA that initiates a flood SHOULD repeat the flood 1222 at a suitable frequency and SHOULD also act as a synchronization 1223 responder for the objective(s) concerned. Thus nodes that require an 1224 objective subject to flooding can either wait for the next flood or 1225 request unicast synchronization for that objective. 1227 The multicast messages for synchronization flooding are subject to 1228 the security rules in Section 3.5.1. In practice this means that 1229 they MUST NOT be transmitted and MUST be ignored on receipt unless 1230 there is an operational ACP or equivalent strong security in place. 1231 However, because of the security weakness of link-local multicast 1232 (Section 5), synchronization objectives that are flooded SHOULD NOT 1233 contain unencrypted private information and SHOULD be validated by 1234 the recipient ASA. 1236 3.5.6.3. Rapid Mode (Discovery/Synchronization Linkage) 1238 A Discovery message MAY include a Synchronization Objective option. 1239 In this case the Discovery message also acts as a Request 1240 Synchronization message to indicate to the Discovery Responder that 1241 it could directly reply to the Discovery Initiator with a 1242 Synchronization message Section 3.8.10 with synchronization data for 1243 rapid processing, if the discovery target supports the corresponding 1244 synchronization objective. The design implications are similar to 1245 those discussed in Section 3.5.5.1. 1247 It is possible that a Discovery Response will arrive from a responder 1248 that does not support rapid mode, before such a Synchronization 1249 message arrives. In this case, rapid mode will not occur. 1251 This rapid mode could reduce the interactions between nodes so that a 1252 higher efficiency could be achieved. However, a network in which 1253 some nodes support rapid mode and others do not will have complex 1254 timing-dependent behaviors. Therefore, the rapid synchronization 1255 function SHOULD be configured off by default and MAY be configured on 1256 or off by Intent. 1258 3.6. GRASP Constants 1260 o ALL_GRASP_NEIGHBORS 1262 A link-local scope multicast address used by a GRASP-enabled 1263 device to discover GRASP-enabled neighbor (i.e., on-link) devices. 1264 All devices that support GRASP are members of this multicast 1265 group. 1267 * IPv6 multicast address: TBD1 1269 * IPv4 multicast address: TBD2 1271 o GRASP_LISTEN_PORT (TBD3) 1273 A well-known UDP user port that every GRASP-enabled network device 1274 MUST always listen to for link-local multicasts. This user port 1275 MAY also be used to listen for TCP or UDP unicast messages in a 1276 simple implementation of GRASP (Section 3.5.3). 1278 o GRASP_DEF_TIMEOUT (60000 milliseconds) 1280 The default timeout used to determine that a discovery etc. has 1281 failed to complete. 1283 o GRASP_DEF_LOOPCT (6) 1285 The default loop count used to determine that a negotiation has 1286 failed to complete, and to avoid looping messages. 1288 o GRASP_DEF_MAX_SIZE (2048) 1290 The default maximum message size in bytes. 1292 3.7. Session Identifier (Session ID) 1294 This is an up to 32-bit opaque value used to distinguish multiple 1295 sessions between the same two devices. A new Session ID MUST be 1296 generated by the initiator for every new Discovery, Flood 1297 Synchronization or Request message. All responses and follow-up 1298 messages in the same discovery, synchronization or negotiation 1299 procedure MUST carry the same Session ID. 1301 The Session ID SHOULD have a very low collision rate locally. It 1302 MUST be generated by a pseudo-random algorithm using a locally 1303 generated seed which is unlikely to be used by any other device in 1304 the same network [RFC4086]. When allocating a new Session ID, GRASP 1305 MUST check that the value is not already in use and SHOULD check that 1306 it has not been used recently, by consulting a cache of current and 1307 recent sessions. In the unlikely event of a clash, GRASP MUST 1308 generate a new value. 1310 However, there is a finite probability that two nodes might generate 1311 the same Session ID value. For that reason, when a Session ID is 1312 communicated via GRASP, the receiving node MUST tag it with the 1313 initiator's IP address to allow disambiguation. In the highly 1314 unlikely event of two peers opening sessions with the same Session ID 1315 value, this tag will allow the two sessions to be distinguished. 1316 Multicast GRASP messages and their responses, which may be relayed 1317 between links, therefore include a field that carries the initiator's 1318 global IP address. 1320 There is a highly unlikely race condition in which two peers start 1321 simultaneous negotiation sessions with each other using the same 1322 Session ID value. Depending on various implementation choices, this 1323 might lead to the two sessions being confused. See Section 3.8.6 for 1324 details of how to avoid this. 1326 3.8. GRASP Messages 1328 3.8.1. Message Overview 1330 This section defines the GRASP message format and message types. 1331 Message types not listed here are reserved for future use. 1333 The messages currently defined are: 1335 Discovery and Discovery Response (M_DISCOVERY, M_RESPONSE). 1337 Request Negotiation, Negotiation, Confirm Waiting and Negotiation 1338 End (M_REQ_NEG, M_NEGOTIATE, M_WAIT, M_END). 1340 Request Synchronization, Synchronization, and Flood 1341 Synchronization (M_REQ_SYN, M_SYNCH, M_FLOOD. 1343 No Operation and Invalid (M_NOOP, M_INVALID). 1345 3.8.2. GRASP Message Format 1347 GRASP messages share an identical header format and a variable format 1348 area for options. GRASP message headers and options are transmitted 1349 in Concise Binary Object Representation (CBOR) [RFC7049]. In this 1350 specification, they are described using CBOR data definition language 1351 (CDDL) [I-D.greevenbosch-appsawg-cbor-cddl]. Fragmentary CDDL is 1352 used to describe each item in this section. A complete and normative 1353 CDDL specification of GRASP is given in Section 6, including 1354 constants such as message types. 1356 Every GRASP message, except the No Operation message, carries a 1357 Session ID (Section 3.7). Options are then presented serially in the 1358 options field. 1360 In fragmentary CDDL, every GRASP message follows the pattern: 1362 grasp-message = (message .within message-structure) / noop-message 1364 message-structure = [MESSAGE_TYPE, session-id, ?initiator, 1365 *grasp-option] 1367 MESSAGE_TYPE = 1..255 1368 session-id = 0..4294967295 ;up to 32 bits 1369 grasp-option = any 1371 The MESSAGE_TYPE indicates the type of the message and thus defines 1372 the expected options. Any options received that are not consistent 1373 with the MESSAGE_TYPE SHOULD be silently discarded. 1375 The No Operation (noop) message is described in Section 3.8.13. 1377 The various MESSAGE_TYPE values are defined in Section 6. 1379 All other message elements are described below and formally defined 1380 in Section 6. 1382 If an unrecognized MESSAGE_TYPE is received in a unicast message, an 1383 Invalid message (Section 3.8.12) MAY be returned. Otherwise the 1384 message MAY be logged and MUST be discarded. If an unrecognized 1385 MESSAGE_TYPE is received in a multicast message, it MAY be logged and 1386 MUST be silently discarded. 1388 3.8.3. Message Size 1390 GRASP nodes MUST be able to receive unicast messages of at least 1391 GRASP_DEF_MAX_SIZE bytes. GRASP nodes MUST NOT send unicast messages 1392 longer than GRASP_DEF_MAX_SIZE bytes unless a longer size is 1393 explicitly allowed for the objective concerned. For example, GRASP 1394 negotiation itself could be used to agree on a longer message size. 1396 The message parser used by GRASP should be configured to know about 1397 the GRASP_DEF_MAX_SIZE, or any larger negotiated message size, so 1398 that it may defend against overly long messages. 1400 The maximum size of multicast messages (M_DISCOVERY and M_FLOOD) 1401 depends on the link layer technology or link adaptation layer in use. 1403 3.8.4. Discovery Message 1405 In fragmentary CDDL, a Discovery message follows the pattern: 1407 discovery-message = [M_DISCOVERY, session-id, initiator, objective] 1409 A discovery initiator sends a Discovery message to initiate a 1410 discovery process for a particular objective option. 1412 The discovery initiator sends all Discovery messages via UDP to port 1413 GRASP_LISTEN_PORT at the link-local ALL_GRASP_NEIGHBORS multicast 1414 address on each link-layer interface in use by GRASP. It then 1415 listens for unicast TCP responses on a given port, and stores the 1416 discovery results (including responding discovery objectives and 1417 corresponding unicast locators). 1419 The listening port used for TCP MUST be the same port as used for 1420 sending the Discovery UDP multicast, on a given interface. In a low- 1421 end implementation this MAY be GRASP_LISTEN_PORT. In a more complex 1422 implementation, the GRASP discovery mechanism will find, for each 1423 interface, a dynamic port that it can bind to for both UDP and TCP 1424 before initiating any discovery. 1426 The 'initiator' field in the message is a globally unique IP address 1427 of the initiator, for the sole purpose of disambiguating the Session 1428 ID in other nodes. If for some reason the initiator does not have a 1429 globally unique IP address, it MUST use a link-local address for this 1430 purpose that is highly likely to be unique, for example using 1431 [RFC7217]. 1433 A Discovery message MUST include exactly one of the following: 1435 o a discovery objective option (Section 3.10.1). Its loop count 1436 MUST be set to a suitable value to prevent discovery loops 1437 (default value is GRASP_DEF_LOOPCT). If the discovery initiator 1438 requires only on-link responses, the loop count MUST be set to 1. 1440 o a negotiation objective option (Section 3.10.1). This is used 1441 both for the purpose of discovery and to indicate to the discovery 1442 target that it MAY directly reply to the discovery initiatior with 1443 a Negotiation message for rapid processing, if it could act as the 1444 corresponding negotiation counterpart. The sender of such a 1445 Discovery message MUST initialize a negotiation timer and loop 1446 count in the same way as a Request Negotiation message 1447 (Section 3.8.6). 1449 o a synchronization objective option (Section 3.10.1). This is used 1450 both for the purpose of discovery and to indicate to the discovery 1451 target that it MAY directly reply to the discovery initiator with 1452 a Synchronization message for rapid processing, if it could act as 1453 the corresponding synchronization counterpart. Its loop count 1454 MUST be set to a suitable value to prevent discovery loops 1455 (default value is GRASP_DEF_LOOPCT). 1457 As mentioned in Section 3.5.4.2, a Discovery message MAY be sent 1458 unicast to a peer node, which SHOULD then proceed exactly as if the 1459 message had been multicast. 1461 3.8.5. Discovery Response Message 1463 In fragmentary CDDL, a Discovery Response message follows the 1464 pattern: 1466 response-message = [M_RESPONSE, session-id, initiator, ttl, 1467 (+locator-option // divert-option), ?objective)] 1469 ttl = 0..4294967295 ; in milliseconds 1471 A node which receives a Discovery message SHOULD send a Discovery 1472 Response message if and only if it can respond to the discovery. 1474 It MUST contain the same Session ID and initiator as the Discovery 1475 message. 1477 It MUST contain a time-to-live (ttl) for the validity of the 1478 response, given as a positive integer value in milliseconds. Zero 1479 is treated as the default value GRASP_DEF_TIMEOUT (Section 3.6). 1481 It MAY include a copy of the discovery objective from the 1482 Discovery message. 1484 It is sent to the sender of the Discovery message via TCP at the port 1485 used to send the Discovery message (as explained in Section 3.8.4). 1486 In the case of a relayed Discovery message, the Discovery Response is 1487 thus sent to the relay, not the original initiator. 1489 If the responding node supports the discovery objective of the 1490 discovery, it MUST include at least one kind of locator option 1491 (Section 3.9.5) to indicate its own location. A sequence of multiple 1492 kinds of locator options (e.g. IP address option and FQDN option) is 1493 also valid. 1495 If the responding node itself does not support the discovery 1496 objective, but it knows the locator of the discovery objective, then 1497 it SHOULD respond to the discovery message with a divert option 1498 (Section 3.9.2) embedding a locator option or a combination of 1499 multiple kinds of locator options which indicate the locator(s) of 1500 the discovery objective. 1502 More details on the processing of Discovery Responses are given in 1503 Section 3.5.4. 1505 3.8.6. Request Messages 1507 In fragmentary CDDL, Request Negotiation and Request Synchronization 1508 messages follow the patterns: 1510 request-negotiation-message = [M_REQ_NEG, session-id, objective] 1512 request-synchronization-message = [M_REQ_SYN, session-id, objective] 1514 A negotiation or synchronization requesting node sends the 1515 appropriate Request message to the unicast address of the negotiation 1516 or synchronization counterpart, using the appropriate protocol and 1517 port numbers (selected from the discovery result). If the discovery 1518 result is an FQDN, it will be resolved first. 1520 A Request message MUST include the relevant objective option. In the 1521 case of Request Negotiation, the objective option MUST include the 1522 requested value. 1524 When an initiator sends a Request Negotiation message, it MUST 1525 initialize a negotiation timer for the new negotiation thread. The 1526 default is GRASP_DEF_TIMEOUT milliseconds. Unless this timeout is 1527 modified by a Confirm Waiting message (Section 3.8.9), the initiator 1528 will consider that the negotiation has failed when the timer expires. 1530 Similarly, when an initiator sends a Request Synchronization, it 1531 SHOULD initialize a synchronization timer. The default is 1532 GRASP_DEF_TIMEOUT milliseconds. The initiator will consider that 1533 synchronization has failed if there is no response before the timer 1534 expires. 1536 When an initiator sends a Request message, it MUST initialize the 1537 loop count of the objective option with a value defined in the 1538 specification of the option or, if no such value is specified, with 1539 GRASP_DEF_LOOPCT. 1541 If a node receives a Request message for an objective for which no 1542 ASA is currently listening, it MUST immediately close the relevant 1543 socket to indicate this to the initiator. This is to avoid 1544 unnecessary timeouts if, for example, an ASA exits prematurely but 1545 the GRASP core is listening on its behalf. 1547 To avoid the highly unlikely race condition in which two nodes 1548 simultaneously request sessions with each other using the same 1549 Session ID (Section 3.7), when a node receives a Request message, it 1550 MUST verify that the received Session ID is not already locally 1551 active. In case of a clash, it MUST discard the Request message, in 1552 which case the initiator will detect a timeout. 1554 3.8.7. Negotiation Message 1556 In fragmentary CDDL, a Negotiation message follows the pattern: 1558 negotiate-message = [M_NEGOTIATE, session-id, objective] 1560 A negotiation counterpart sends a Negotiation message in response to 1561 a Request Negotiation message, a Negotiation message, or a Discovery 1562 message in Rapid Mode. A negotiation process MAY include multiple 1563 steps. 1565 The Negotiation message MUST include the relevant Negotiation 1566 Objective option, with its value updated according to progress in the 1567 negotiation. The sender MUST decrement the loop count by 1. If the 1568 loop count becomes zero the message MUST NOT be sent. In this case 1569 the negotiation session has failed and will time out. 1571 3.8.8. Negotiation End Message 1573 In fragmentary CDDL, a Negotiation End message follows the pattern: 1575 end-message = [M_END, session-id, accept-option / decline-option] 1577 A negotiation counterpart sends an Negotiation End message to close 1578 the negotiation. It MUST contain either an accept or a decline 1579 option, defined in Section 3.9.3 and Section 3.9.4. It could be sent 1580 either by the requesting node or the responding node. 1582 3.8.9. Confirm Waiting Message 1584 In fragmentary CDDL, a Confirm Waiting message follows the pattern: 1586 wait-message = [M_WAIT, session-id, waiting-time] 1587 waiting-time = 0..4294967295 ; in milliseconds 1589 A responding node sends a Confirm Waiting message to ask the 1590 requesting node to wait for a further negotiation response. It might 1591 be that the local process needs more time or that the negotiation 1592 depends on another triggered negotiation. This message MUST NOT 1593 include any other options. When received, the waiting time value 1594 overwrites and restarts the current negotiation timer 1595 (Section 3.8.6). 1597 The responding node SHOULD send a Negotiation, Negotiation End or 1598 another Confirm Waiting message before the negotiation timer expires. 1599 If not, the initiator MUST abandon or restart the negotiation 1600 procedure, to avoid an indefinite wait. 1602 3.8.10. Synchronization Message 1604 In fragmentary CDDL, a Synchronization message follows the pattern: 1606 synch-message = [M_SYNCH, session-id, objective] 1608 A node which receives a Request Synchronization, or a Discovery 1609 message in Rapid Mode, sends back a unicast Synchronization message 1610 with the synchronization data, in the form of a GRASP Option for the 1611 specific synchronization objective present in the Request 1612 Synchronization. 1614 3.8.11. Flood Synchronization Message 1616 In fragmentary CDDL, a Flood Synchronization message follows the 1617 pattern: 1619 flood-message = [M_FLOOD, session-id, initiator, ttl, 1620 +[objective, (locator-option / [])]] 1622 ttl = 0..4294967295 ; in milliseconds 1624 A node MAY initiate flooding by sending an unsolicited Flood 1625 Synchronization Message with synchronization data. This MAY be sent 1626 to port GRASP_LISTEN_PORT at the link-local ALL_GRASP_NEIGHBORS 1627 multicast address, in accordance with the rules in Section 3.5.6. 1629 The initiator address is provided, as described for Discovery 1630 messages (Section 3.8.4), only to disambiguate the Session ID. 1632 The message MUST contain a time-to-live (ttl) for the validity of 1633 the contents, given as a positive integer value in milliseconds. 1634 There is no default; zero indicates an indefinite lifetime. 1636 The synchronization data are in the form of GRASP Option(s) for 1637 specific synchronization objective(s). The loop count(s) MUST be 1638 set to a suitable value to prevent flood loops (default value is 1639 GRASP_DEF_LOOPCT). 1641 Each objective option MAY be followed by a locator option 1642 associated with the flooded objective. In its absence, an empty 1643 option MUST be included to indicate a null locator. 1645 A node that receives a Flood Synchronization message MUST cache the 1646 received objectives for use by local ASAs. Each cached objective 1647 MUST be tagged with the locator option sent with it, or with a null 1648 tag if an empty locator option was sent. If a subsequent Flood 1649 Synchronization message carrying the same objective arrives with the 1650 same tag, the corresponding cached copy of the objective MUST be 1651 overwritten. If a subsequent Flood Synchronization message carrying 1652 the same objective arrives with a different tag, a new cached entry 1653 MUST be created. 1655 Note: the purpose of this mechanism is to allow the recipient of 1656 flooded values to distinguish between different senders of the same 1657 objective, and if necessary communicate with them using the locator, 1658 protocol and port included in the locator option. Many objectives 1659 will not need this mechanism, so they will be flooded with a null 1660 locator. 1662 Cached entries MUST be ignored or deleted after their lifetime 1663 expires. 1665 3.8.12. Invalid Message 1667 In fragmentary CDDL, an Invalid message follows the pattern: 1669 invalid-message = [M_INVALID, session-id, ?any] 1671 This message MAY be sent by an implementation in response to an 1672 incoming unicast message that it considers invalid. The session-id 1673 MUST be copied from the incoming message. The content SHOULD be 1674 diagnostic information such as a partial copy of the invalid message. 1675 An M_INVALID message MAY be silently ignored by a recipient. 1676 However, it could be used in support of extensibility, since it 1677 indicates that the remote node does not support a new or obsolete 1678 message or option. 1680 An M_INVALID message MUST NOT be sent in response to an M_INVALID 1681 message. 1683 3.8.13. No Operation Message 1685 In fragmentary CDDL, a No Operation message follows the pattern: 1687 noop-message = [M_NOOP] 1689 This message MAY be sent by an implementation that for practical 1690 reasons needs to initialize a socket. It MUST be silently ignored by 1691 a recipient. 1693 3.9. GRASP Options 1695 This section defines the GRASP options for the negotiation and 1696 synchronization protocol signaling. Additional options may be 1697 defined in the future. 1699 3.9.1. Format of GRASP Options 1701 GRASP options are CBOR objects that MUST start with an unsigned 1702 integer identifying the specific option type carried in this option. 1703 These option types are formally defined in Section 6. Apart from 1704 that the only format requirement is that each option MUST be a well- 1705 formed CBOR object. In general a CBOR array format is RECOMMENDED to 1706 limit overhead. 1708 GRASP options may be defined to include encapsulated GRASP options. 1710 3.9.2. Divert Option 1712 The Divert option is used to redirect a GRASP request to another 1713 node, which may be more appropriate for the intended negotiation or 1714 synchronization. It may redirect to an entity that is known as a 1715 specific negotiation or synchronization counterpart (on-link or off- 1716 link) or a default gateway. The divert option MUST only be 1717 encapsulated in Discovery Response messages. If found elsewhere, it 1718 SHOULD be silently ignored. 1720 A discovery initiator MAY ignore a Divert option if it only requires 1721 direct discovery responses. 1723 In fragmentary CDDL, the Divert option follows the pattern: 1725 divert-option = [O_DIVERT, +locator-option] 1727 The embedded Locator Option(s) (Section 3.9.5) point to diverted 1728 destination target(s) in response to a Discovery message. 1730 3.9.3. Accept Option 1732 The accept option is used to indicate to the negotiation counterpart 1733 that the proposed negotiation content is accepted. 1735 The accept option MUST only be encapsulated in Negotiation End 1736 messages. If found elsewhere, it SHOULD be silently ignored. 1738 In fragmentary CDDL, the Accept option follows the pattern: 1740 accept-option = [O_ACCEPT] 1742 3.9.4. Decline Option 1744 The decline option is used to indicate to the negotiation counterpart 1745 the proposed negotiation content is declined and end the negotiation 1746 process. 1748 The decline option MUST only be encapsulated in Negotiation End 1749 messages. If found elsewhere, it SHOULD be silently ignored. 1751 In fragmentary CDDL, the Decline option follows the pattern: 1753 decline-option = [O_DECLINE, ?reason] 1754 reason = text ;optional error message 1756 Note: there might be scenarios where an ASA wants to decline the 1757 proposed value and restart the negotiation process. In this case it 1758 is an implementation choice whether to send a Decline option or to 1759 continue with a Negotiate message, with an objective option that 1760 contains a null value, or one that contains a new value that might 1761 achieve convergence. 1763 3.9.5. Locator Options 1765 These locator options are used to present reachability information 1766 for an ASA, a device or an interface. They are Locator IPv6 Address 1767 Option, Locator IPv4 Address Option, Locator FQDN (Fully Qualified 1768 Domain Name) Option and URI (Uniform Resource Identifier) Option. 1770 Since ASAs will normally run as independent user programs, locator 1771 options need to indicate the network layer locator plus the transport 1772 protocol and port number for reaching the target. For this reason, 1773 the Locator Options for IP addresses and FQDNs include this 1774 information explicitly. In the case of the URI Option, this 1775 information can be encoded in the URI itself. 1777 Note: It is assumed that all locators are in scope throughout the 1778 GRASP domain. As stated in Section 3.2, GRASP is not intended to 1779 work across disjoint addressing or naming realms. 1781 3.9.5.1. Locator IPv6 address option 1783 In fragmentary CDDL, the IPv6 address option follows the pattern: 1785 ipv6-locator-option = [O_IPv6_LOCATOR, ipv6-address, 1786 transport-proto, port-number] 1787 ipv6-address = bytes .size 16 1789 transport-proto = IPPROTO_TCP / IPPROTO_UDP 1790 IPPROTO_TCP = 6 1791 IPPROTO_UDP = 17 1792 port-number = 0..65535 1794 The content of this option is a binary IPv6 address followed by the 1795 protocol number and port number to be used. 1797 Note 1: The IPv6 address MUST normally have global scope. However, 1798 during initialization, a link-local address MAY be used for specific 1799 objectives only (Section 3.5.2). In this case the corresponding 1800 Discovery Response message MUST be sent via the interface to which 1801 the link-local address applies. 1803 Note 2: A link-local IPv6 address MUST NOT be used when this option 1804 is included in a Divert option. 1806 3.9.5.2. Locator IPv4 address option 1808 In fragmentary CDDL, the IPv4 address option follows the pattern: 1810 ipv4-locator-option = [O_IPv4_LOCATOR, ipv4-address, 1811 transport-proto, port-number] 1812 ipv4-address = bytes .size 4 1814 The content of this option is a binary IPv4 address followed by the 1815 protocol number and port number to be used. 1817 Note: If an operator has internal network address translation for 1818 IPv4, this option MUST NOT be used within the Divert option. 1820 3.9.5.3. Locator FQDN option 1822 In fragmentary CDDL, the FQDN option follows the pattern: 1824 fqdn-locator-option = [O_FQDN_LOCATOR, text, 1825 transport-proto, port-number] 1827 The content of this option is the Fully Qualified Domain Name of the 1828 target followed by the protocol number and port number to be used. 1830 Note 1: Any FQDN which might not be valid throughout the network in 1831 question, such as a Multicast DNS name [RFC6762], MUST NOT be used 1832 when this option is used within the Divert option. 1834 Note 2: Normal GRASP operations are not expected to use this option. 1835 It is intended for special purposes such as discovering external 1836 services. 1838 3.9.5.4. Locator URI option 1840 In fragmentary CDDL, the URI option follows the pattern: 1842 uri-locator = [O_URI_LOCATOR, text] 1844 The content of this option is the Uniform Resource Identifier of the 1845 target [RFC3986]. 1847 Note 1: Any URI which might not be valid throughout the network in 1848 question, such as one based on a Multicast DNS name [RFC6762], MUST 1849 NOT be used when this option is used within the Divert option. 1851 Note 2: Normal GRASP operations are not expected to use this option. 1852 It is intended for special purposes such as discovering external 1853 services. Therefore its use is not further described in this 1854 specification. 1856 3.10. Objective Options 1858 3.10.1. Format of Objective Options 1860 An objective option is used to identify objectives for the purposes 1861 of discovery, negotiation or synchronization. All objectives MUST be 1862 in the following format, described in fragmentary CDDL: 1864 objective = [objective-name, objective-flags, loop-count, ?any] 1866 objective-name = text 1867 loop-count = 0..255 1869 All objectives are identified by a unique name which is a UTF-8 1870 string, to be compared byte by byte. 1872 The names of generic objectives MUST NOT include a colon (":") and 1873 MUST be registered with IANA (Section 7). 1875 The names of privately defined objectives MUST include at least one 1876 colon (":"). The string preceding the last colon in the name MUST be 1877 globally unique and in some way identify the entity or person 1878 defining the objective. The following three methods MAY be used to 1879 create such a globally unique string: 1881 1. The unique string is a decimal number representing a registered 1882 32 bit Private Enterprise Number (PEN) [I-D.liang-iana-pen] that 1883 uniquely identifies the enterprise defining the objective. 1885 2. The unique string is a fully qualified domain name that uniquely 1886 identifies the entity or person defining the objective. 1888 3. The unique string is an email address that uniquely identifies 1889 the entity or person defining the objective. 1891 The GRASP protocol treats the objective name as an opaque string. 1892 For example, "EX1", "411:EX1", "example.com:EX1", "example.org:EX1 1893 and "user@example.org:EX1" would be five different objectives. 1895 The 'objective-flags' field is described below. 1897 The 'loop-count' field is used for terminating negotiation as 1898 described in Section 3.8.7. It is also used for terminating 1899 discovery as described in Section 3.5.4, and for terminating flooding 1900 as described in Section 3.5.6.2. It is placed in the objective 1901 rather than in the GRASP message format because, as far as the ASA is 1902 concerned, it is a property of the objective itself. 1904 The 'any' field is to express the actual value of a negotiation or 1905 synchronization objective. Its format is defined in the 1906 specification of the objective and may be a simple value or a data 1907 structure of any kind. It is optional because it is optional in a 1908 Discovery or Discovery Response message. 1910 3.10.2. Objective flags 1912 An objective may be relevant for discovery only, for discovery and 1913 negotiation, or for discovery and synchronization. This is expressed 1914 in the objective by logical flag bits: 1916 objective-flags = uint .bits objective-flag 1917 objective-flag = &( 1918 F_DISC: 0 ; valid for discovery 1919 F_NEG: 1 ; valid for negotiation 1920 F_SYNCH: 2 ; valid for synchronization 1921 F_NEG_DRY: 3 ; negotiation is dry-run 1922 ) 1924 These bits are independent and may be combined appropriately, e.g. 1925 (F_DISC and F_SYNCH) or (F_DISC and F_NEG) or (F_DISC and F_NEG and 1926 F_NEG_DRY). 1928 Note that for a given negotiation session, an objective must be 1929 either used for negotiation, or for dry-run negotiation. Mixing the 1930 two modes in a single negotiation is not possible. 1932 3.10.3. General Considerations for Objective Options 1934 As mentioned above, Objective Options MUST be assigned a unique name. 1935 As long as privately defined Objective Options obey the rules above, 1936 this document does not restrict their choice of name, but the entity 1937 or person concerned SHOULD publish the names in use. 1939 Names are expressed as UTF-8 strings for convenience in designing 1940 Objective Options for localized use. For generic usage, names 1941 expressed in the ASCII subset of UTF-8 are RECOMMENDED. Designers 1942 planning to use non-ASCII names are strongly advised to consult 1943 [RFC7564] or its successor to understand the complexities involved. 1944 Since the GRASP protocol compares names byte by byte, all issues of 1945 Unicode profiling and canonicalization MUST be specified in the 1946 design of the Objective Option. 1948 All Objective Options MUST respect the CBOR patterns defined above as 1949 "objective" and MUST replace the "any" field with a valid CBOR data 1950 definition for the relevant use case and application. 1952 An Objective Option that contains no additional fields beyond its 1953 "loop-count" can only be a discovery objective and MUST only be used 1954 in Discovery and Discovery Response messages. 1956 The Negotiation Objective Options contain negotiation objectives, 1957 which vary according to different functions/services. They MUST be 1958 carried by Discovery, Request Negotiation or Negotiation messages 1959 only. The negotiation initiator MUST set the initial "loop-count" to 1960 a value specified in the specification of the objective or, if no 1961 such value is specified, to GRASP_DEF_LOOPCT. 1963 For most scenarios, there should be initial values in the negotiation 1964 requests. Consequently, the Negotiation Objective options MUST 1965 always be completely presented in a Request Negotiation message, or 1966 in a Discovery message in rapid mode. If there is no initial value, 1967 the value field SHOULD be set to the 'null' value defined by CBOR. 1969 Synchronization Objective Options are similar, but MUST be carried by 1970 Discovery, Discovery Response, Request Synchronization, or Flood 1971 Synchronization messages only. They include value fields only in 1972 Synchronization or Flood Synchronization messages. 1974 3.10.4. Organizing of Objective Options 1976 Generic objective options MUST be specified in documents available to 1977 the public and SHOULD be designed to use either the negotiation or 1978 the synchronization mechanism described above. 1980 As noted earlier, one negotiation objective is handled by each GRASP 1981 negotiation thread. Therefore, a negotiation objective, which is 1982 based on a specific function or action, SHOULD be organized as a 1983 single GRASP option. It is NOT RECOMMENDED to organize multiple 1984 negotiation objectives into a single option, nor to split a single 1985 function or action into multiple negotiation objectives. 1987 It is important to understand that GRASP negotiation does not support 1988 transactional integrity. If transactional integrity is needed for a 1989 specific objective, this must be ensured by the ASA. For example, an 1990 ASA might need to ensure that it only participates in one negotiation 1991 thread at the same time. Such an ASA would need to stop listening 1992 for incoming negotiation requests before generating an outgoing 1993 negotiation request. 1995 A synchronization objective SHOULD be organized as a single GRASP 1996 option. 1998 Some objectives will support more than one operational mode. An 1999 example is a negotiation objective with both a "dry run" mode (where 2000 the negotiation is to find out whether the other end can in fact make 2001 the requested change without problems) and a "live" mode. Such modes 2002 will be defined in the specification of such an objective. These 2003 objectives SHOULD include flags indicating the applicable mode(s). 2005 An issue requiring particular attention is that GRASP itself is a 2006 stateless protocol. Any state associated with a dry run operation, 2007 such as temporarily reserving a resource for subsequent use in a live 2008 run, is entirely a matter for the designer of the ASA concerned. 2010 As indicated in Section 3.1, an objective's value may include 2011 multiple parameters. Parameters might be categorized into two 2012 classes: the obligatory ones presented as fixed fields; and the 2013 optional ones presented in some other form of data structure embedded 2014 in CBOR. The format might be inherited from an existing management 2015 or configuration protocol, with the objective option acting as a 2016 carrier for that format. The data structure might be defined in a 2017 formal language, but that is a matter for the specifications of 2018 individual objectives. There are many candidates, according to the 2019 context, such as ABNF, RBNF, XML Schema, YANG, etc. The GRASP 2020 protocol itself is agnostic on these questions. The only restriction 2021 is that the format can be mapped into CBOR. 2023 It is NOT RECOMMENDED to mix parameters that have significantly 2024 different response time characteristics in a single objective. 2025 Separate objectives are more suitable for such a scenario. 2027 All objectives MUST support GRASP discovery. However, as mentioned 2028 in Section 3.3, it is acceptable for an ASA to use an alternative 2029 method of discovery. 2031 Normally, a GRASP objective will refer to specific technical 2032 parameters as explained in Section 3.1. However, it is acceptable to 2033 define an abstract objective for the purpose of managing or 2034 coordinating ASAs. It is also acceptable to define a special-purpose 2035 objective for purposes such as trust bootstrapping or formation of 2036 the ACP. 2038 To guarantee convergence, a limited number of rounds or a timeout is 2039 needed for each negotiation objective. Therefore, the definition of 2040 each negotiation objective SHOULD clearly specify this, for example a 2041 default loop count and timeout, so that the negotiation can always be 2042 terminated properly. If not, the GRASP defaults will apply. 2044 There must be a well-defined procedure for concluding that a 2045 negotiation cannot succeed, and if so deciding what happens next 2046 (e.g., deadlock resolution, tie-breaking, or revert to best-effort 2047 service). This MUST be specified for individual negotiation 2048 objectives. 2050 3.10.5. Experimental and Example Objective Options 2052 The names "EX0" through "EX9" have been reserved for experimental 2053 options. Multiple names have been assigned because a single 2054 experiment may use multiple options simultaneously. These 2055 experimental options are highly likely to have different meanings 2056 when used for different experiments. Therefore, they SHOULD NOT be 2057 used without an explicit human decision and SHOULD NOT be used in 2058 unmanaged networks such as home networks. 2060 These names are also RECOMMENDED for use in documentation examples. 2062 4. Implementation Status [RFC Editor: please remove] 2064 Two prototype implementations of GRASP have been made. 2066 4.1. BUPT C++ Implementation 2068 o Name: BaseNegotiator.cpp, msg.cpp, Client.cpp, Server.cpp 2070 o Description: C++ implementation of GRASP core and API 2072 o Maturity: Prototype code, interoperable between Ubuntu. 2074 o Coverage: Corresponds to draft-carpenter-anima-gdn-protocol-03. 2075 Since it was implemented based on the old version draft, the most 2076 significant limitations comparing to current protocol design 2077 include: 2079 * Not support CBOR 2081 * Not support Flooding 2082 * Not support loop avoidance 2084 * only coded for IPv6, any IPv4 is accidental 2086 o Licensing: Huawei License. 2088 o Experience: https://github.com/liubingpang/IETF-Anima-Signaling- 2089 Protocol/blob/master/README.md 2091 o Contact: https://github.com/liubingpang/IETF-Anima-Signaling- 2092 Protocol 2094 4.2. Python Implementation 2096 o Name: graspy 2098 o Description: Python 3 implementation of GRASP core and API. 2100 o Maturity: Prototype code, interoperable between Windows 7 and 2101 Linux. 2103 o Coverage: Corresponds to draft-ietf-anima-grasp-10. Limitations 2104 include: 2106 * insecure: uses a dummy ACP module and does not implement TLS 2108 * only coded for IPv6, any IPv4 is accidental 2110 * FQDN and URI locators incompletely supported 2112 * no code for rapid mode 2114 * relay code is lazy (no rate control) 2116 * all unicast transactions use TCP (no unicast UDP). 2117 Experimental code for unicast UDP proved to be complex and 2118 brittle. 2120 * optional Objective option in Response messages not implemented 2122 * workarounds for defects in Python socket module and Windows 2123 socket peculiarities 2125 o Licensing: Simplified BSD 2127 o Experience: https://www.cs.auckland.ac.nz/~brian/graspy/graspy.pdf 2129 o Contact: https://www.cs.auckland.ac.nz/~brian/graspy/ 2131 5. Security Considerations 2133 A successful attack on negotiation-enabled nodes would be extremely 2134 harmful, as such nodes might end up with a completely undesirable 2135 configuration that would also adversely affect their peers. GRASP 2136 nodes and messages therefore require full protection. As explained 2137 in Section 3.5.1, GRASP MUST run within a secure environment such as 2138 the Autonomic Control Plane [I-D.ietf-anima-autonomic-control-plane], 2139 except for the constrained instances described in Section 3.5.2. 2141 - Authentication 2143 A cryptographically authenticated identity for each device is 2144 needed in an autonomic network. It is not safe to assume that a 2145 large network is physically secured against interference or that 2146 all personnel are trustworthy. Each autonomic node MUST be 2147 capable of proving its identity and authenticating its messages. 2148 GRASP relies on a separate external certificate-based security 2149 mechanism to support authentication, data integrity protection, 2150 and anti-replay protection. 2152 Since GRASP must be deployed in an existing secure environment, 2153 the protocol itself specifies nothing concerning the trust anchor 2154 and certification authority. 2156 If GRASP is used temporarily without an external security 2157 mechanism, for example during system bootstrap (Section 3.5.1), 2158 the Session ID (Section 3.7) will act as a nonce to provide 2159 limited protection against third parties injecting responses. A 2160 full analysis of the secure bootstrap process is in 2161 [I-D.ietf-anima-bootstrapping-keyinfra]. 2163 - Authorization and Roles 2165 The GRASP protocol is agnostic about the roles and capabilities of 2166 individual ASAs and about which objectives a particular ASA is 2167 authorized to support. An implementation might support 2168 precautions such as allowing only one ASA in a given node to 2169 modify a given objective, but this may not be appropriate in all 2170 cases. For example, it might be operationally useful to allow an 2171 old and a new version of the same ASA to run simultaneously during 2172 an overlap period. These questions are out of scope for the 2173 present specification. 2175 - Privacy and confidentiality 2177 Generally speaking, no personal information is expected to be 2178 involved in the signaling protocol, so there should be no direct 2179 impact on personal privacy. Nevertheless, traffic flow paths, 2180 VPNs, etc. could be negotiated, which could be of interest for 2181 traffic analysis. Also, operators generally want to conceal 2182 details of their network topology and traffic density from 2183 outsiders. Therefore, since insider attacks cannot be excluded in 2184 a large network, the security mechanism for the protocol MUST 2185 provide message confidentiality. This is why Section 3.5.1 2186 requires either an ACP or an alternative security mechanism. 2188 - Link-local multicast security 2190 GRASP has no reasonable alternative to using link-local multicast 2191 for Discovery or Flood Synchronization messages and these messages 2192 are sent in clear and with no authentication. They are only sent 2193 on interfaces within the autonomic network (see Section 3.1 and 2194 Section 3.5.1). They are however available to on-link 2195 eavesdroppers, and could be forged by on-link attackers. In the 2196 case of Discovery, the Discovery Responses are unicast and will 2197 therefore be protected (Section 3.5.1), and an untrusted forger 2198 will not be able to receive responses. In the case of Flood 2199 Synchronization, an on-link eavesdropper will be able to receive 2200 the flooded objectives but there is no response message to 2201 consider. Some precautions for Flood Synchronization messages are 2202 suggested in Section 3.5.6.2. 2204 - DoS Attack Protection 2206 GRASP discovery partly relies on insecure link-local multicast. 2207 Since routers participating in GRASP sometimes relay discovery 2208 messages from one link to another, this could be a vector for 2209 denial of service attacks. Some mitigations are specified in 2210 Section 3.5.4. However, malicious code installed inside the 2211 Autonomic Control Plane could always launch DoS attacks consisting 2212 of spurious discovery messages, or of spurious discovery 2213 responses. It is important that firewalls prevent any GRASP 2214 messages from entering the domain from an unknown source. 2216 - Security during bootstrap and discovery 2218 A node cannot trust GRASP traffic from other nodes until the 2219 security environment (such as the ACP) has identified the trust 2220 anchor and can authenticate traffic by validating certificates for 2221 other nodes. Also, until it has succesfully enrolled 2222 [I-D.ietf-anima-bootstrapping-keyinfra] a node cannot assume that 2223 other nodes are able to authenticate its own traffic. Therefore, 2224 GRASP discovery during the bootstrap phase for a new device will 2225 inevitably be insecure. Secure synchronization and negotiation 2226 will be impossible until enrollment is complete. Further details 2227 are given in Section 3.5.2. 2229 - Security of discovered locators 2231 When GRASP discovery returns an IP address, it MUST be that of a 2232 node within the secure environment (Section 3.5.1). If it returns 2233 an FQDN or a URI, the ASA that receives it MUST NOT assume that 2234 the target of the locator is within the secure environment. 2236 6. CDDL Specification of GRASP 2238 2239 grasp-message = (message .within message-structure) / noop-message 2241 message-structure = [MESSAGE_TYPE, session-id, ?initiator, 2242 *grasp-option] 2244 MESSAGE_TYPE = 0..255 2245 session-id = 0..4294967295 ;up to 32 bits 2246 grasp-option = any 2248 message /= discovery-message 2249 discovery-message = [M_DISCOVERY, session-id, initiator, objective] 2251 message /= response-message ;response to Discovery 2252 response-message = [M_RESPONSE, session-id, initiator, ttl, 2253 (+locator-option // divert-option), ?objective] 2255 message /= synch-message ;response to Synchronization request 2256 synch-message = [M_SYNCH, session-id, objective] 2258 message /= flood-message 2259 flood-message = [M_FLOOD, session-id, initiator, ttl, 2260 +[objective, (locator-option / [])]] 2262 message /= request-negotiation-message 2263 request-negotiation-message = [M_REQ_NEG, session-id, objective] 2265 message /= request-synchronization-message 2266 request-synchronization-message = [M_REQ_SYN, session-id, objective] 2268 message /= negotiation-message 2269 negotiation-message = [M_NEGOTIATE, session-id, objective] 2271 message /= end-message 2272 end-message = [M_END, session-id, accept-option / decline-option ] 2273 message /= wait-message 2274 wait-message = [M_WAIT, session-id, waiting-time] 2276 message /= invalid-message 2277 invalid-message = [M_INVALID, session-id, ?any] 2279 noop-message = [M_NOOP] 2281 divert-option = [O_DIVERT, +locator-option] 2283 accept-option = [O_ACCEPT] 2285 decline-option = [O_DECLINE, ?reason] 2286 reason = text ;optional error message 2288 waiting-time = 0..4294967295 ; in milliseconds 2289 ttl = 0..4294967295 ; in milliseconds 2291 locator-option /= [O_IPv4_LOCATOR, ipv4-address, 2292 transport-proto, port-number] 2293 ipv4-address = bytes .size 4 2295 locator-option /= [O_IPv6_LOCATOR, ipv6-address, 2296 transport-proto, port-number] 2297 ipv6-address = bytes .size 16 2299 locator-option /= [O_FQDN_LOCATOR, text, transport-proto, port-number] 2301 transport-proto = IPPROTO_TCP / IPPROTO_UDP 2302 IPPROTO_TCP = 6 2303 IPPROTO_UDP = 17 2304 port-number = 0..65535 2306 locator-option /= [O_URI_LOCATOR, text] 2308 initiator = ipv4-address / ipv6-address 2310 objective-flags = uint .bits objective-flag 2312 objective-flag = &( 2313 F_DISC: 0 ; valid for discovery 2314 F_NEG: 1 ; valid for negotiation 2315 F_SYNCH: 2 ; valid for synchronization 2316 F_NEG_DRY: 3 ; negotiation is dry-run 2317 ) 2319 objective = [objective-name, objective-flags, loop-count, ?any] 2320 objective-name = text ;see specification for uniqueness rules 2322 loop-count = 0..255 2324 ; Constants for message types and option types 2326 M_NOOP = 0 2327 M_DISCOVERY = 1 2328 M_RESPONSE = 2 2329 M_REQ_NEG = 3 2330 M_REQ_SYN = 4 2331 M_NEGOTIATE = 5 2332 M_END = 6 2333 M_WAIT = 7 2334 M_SYNCH = 8 2335 M_FLOOD = 9 2336 M_INVALID = 99 2338 O_DIVERT = 100 2339 O_ACCEPT = 101 2340 O_DECLINE = 102 2341 O_IPv6_LOCATOR = 103 2342 O_IPv4_LOCATOR = 104 2343 O_FQDN_LOCATOR = 105 2344 O_URI_LOCATOR = 106 2345 2347 7. IANA Considerations 2349 This document defines the GeneRic Autonomic Signaling Protocol 2350 (GRASP). 2352 Section 3.6 explains the following link-local multicast addresses, 2353 which IANA is requested to assign for use by GRASP: 2355 ALL_GRASP_NEIGHBORS multicast address (IPv6): (TBD1). Assigned in 2356 the IPv6 Link-Local Scope Multicast Addresses registry. 2358 ALL_GRASP_NEIGHBORS multicast address (IPv4): (TBD2). Assigned in 2359 the IPv4 Multicast Local Network Control Block. 2361 Section 3.6 explains the following User Port, which IANA is requested 2362 to assign for use by GRASP for both UDP and TCP: 2364 GRASP_LISTEN_PORT: (TBD3) 2365 Service Name: Generic Autonomic Signaling Protocol (GRASP) 2366 Transport Protocols: UDP, TCP 2367 Assignee: iesg@ietf.org 2368 Contact: chair@ietf.org 2369 Description: See Section 3.6 2370 Reference: RFC XXXX (this document) 2372 The IANA is requested to create a GRASP Parameter Registry including 2373 two registry tables. These are the GRASP Messages and Options 2374 Table and the GRASP Objective Names Table. 2376 GRASP Messages and Options Table. The values in this table are names 2377 paired with decimal integers. Future values MUST be assigned using 2378 the Standards Action policy defined by [RFC5226]. The following 2379 initial values are assigned by this document: 2381 M_NOOP = 0 2382 M_DISCOVERY = 1 2383 M_RESPONSE = 2 2384 M_REQ_NEG = 3 2385 M_REQ_SYN = 4 2386 M_NEGOTIATE = 5 2387 M_END = 6 2388 M_WAIT = 7 2389 M_SYNCH = 8 2390 M_FLOOD = 9 2391 M_INVALID = 99 2393 O_DIVERT = 100 2394 O_ACCEPT = 101 2395 O_DECLINE = 102 2396 O_IPv6_LOCATOR = 103 2397 O_IPv4_LOCATOR = 104 2398 O_FQDN_LOCATOR = 105 2399 O_URI_LOCATOR = 106 2401 GRASP Objective Names Table. The values in this table are UTF-8 2402 strings. Future values MUST be assigned using the Specification 2403 Required policy defined by [RFC5226]. 2405 To assist expert review of a new objective, the specification should 2406 include a precise description of the format of the new objective, 2407 with sufficient explanation of its semantics to allow independent 2408 implementations. See Section 3.10.3 for more details. If the new 2409 objective is similar in name or purpose to a previously registered 2410 objective, the specification should explain why a new objective is 2411 justified. 2413 The following initial values are assigned by this document: 2415 EX0 2416 EX1 2417 EX2 2418 EX3 2419 EX4 2420 EX5 2421 EX6 2422 EX7 2423 EX8 2424 EX9 2426 8. Acknowledgements 2428 A major contribution to the original version of this document was 2429 made by Sheng Jiang. Significant review inputs were received from 2430 Toerless Eckert, Joel Halpern, Barry Leiba, Charles E. Perkins, and 2431 Michael Richardson. 2433 Valuable comments were received from Michael Behringer, Jeferson 2434 Campos Nobre, Laurent Ciavaglia, Zongpeng Du, Yu Fu, Joel Jaeggli, 2435 Zhenbin Li, Dimitri Papadimitriou, Pierre Peloso, Reshad Rahman, 2436 Markus Stenberg, Rene Struik, Martin Thomson, Dacheng Zhang, and 2437 other participants in the NMRG research group and the ANIMA working 2438 group. 2440 9. References 2442 9.1. Normative References 2444 [I-D.greevenbosch-appsawg-cbor-cddl] 2445 Birkholz, H., Vigano, C., and C. Bormann, "CBOR data 2446 definition language (CDDL): a notational convention to 2447 express CBOR data structures", draft-greevenbosch-appsawg- 2448 cbor-cddl-10 (work in progress), March 2017. 2450 [I-D.ietf-anima-autonomic-control-plane] 2451 Behringer, M., Eckert, T., and S. Bjarnason, "An Autonomic 2452 Control Plane", draft-ietf-anima-autonomic-control- 2453 plane-06 (work in progress), March 2017. 2455 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2456 Requirement Levels", BCP 14, RFC 2119, 2457 DOI 10.17487/RFC2119, March 1997, 2458 . 2460 [RFC3986] Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform 2461 Resource Identifier (URI): Generic Syntax", STD 66, 2462 RFC 3986, DOI 10.17487/RFC3986, January 2005, 2463 . 2465 [RFC4086] Eastlake 3rd, D., Schiller, J., and S. Crocker, 2466 "Randomness Requirements for Security", BCP 106, RFC 4086, 2467 DOI 10.17487/RFC4086, June 2005, 2468 . 2470 [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security 2471 (TLS) Protocol Version 1.2", RFC 5246, 2472 DOI 10.17487/RFC5246, August 2008, 2473 . 2475 [RFC5280] Cooper, D., Santesson, S., Farrell, S., Boeyen, S., 2476 Housley, R., and W. Polk, "Internet X.509 Public Key 2477 Infrastructure Certificate and Certificate Revocation List 2478 (CRL) Profile", RFC 5280, DOI 10.17487/RFC5280, May 2008, 2479 . 2481 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 2482 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 2483 January 2012, . 2485 [RFC7049] Bormann, C. and P. Hoffman, "Concise Binary Object 2486 Representation (CBOR)", RFC 7049, DOI 10.17487/RFC7049, 2487 October 2013, . 2489 [RFC7217] Gont, F., "A Method for Generating Semantically Opaque 2490 Interface Identifiers with IPv6 Stateless Address 2491 Autoconfiguration (SLAAC)", RFC 7217, 2492 DOI 10.17487/RFC7217, April 2014, 2493 . 2495 9.2. Informative References 2497 [I-D.chaparadza-intarea-igcp] 2498 Behringer, M., Chaparadza, R., Petre, R., Li, X., and H. 2499 Mahkonen, "IP based Generic Control Protocol (IGCP)", 2500 draft-chaparadza-intarea-igcp-00 (work in progress), July 2501 2011. 2503 [I-D.ietf-anima-bootstrapping-keyinfra] 2504 Pritikin, M., Richardson, M., Behringer, M., Bjarnason, 2505 S., and K. Watsen, "Bootstrapping Remote Secure Key 2506 Infrastructures (BRSKI)", draft-ietf-anima-bootstrapping- 2507 keyinfra-05 (work in progress), March 2017. 2509 [I-D.ietf-anima-reference-model] 2510 Behringer, M., Carpenter, B., Eckert, T., Ciavaglia, L., 2511 Pierre, P., Liu, B., Nobre, J., and J. Strassner, "A 2512 Reference Model for Autonomic Networking", draft-ietf- 2513 anima-reference-model-03 (work in progress), March 2017. 2515 [I-D.ietf-anima-stable-connectivity] 2516 Eckert, T. and M. Behringer, "Using Autonomic Control 2517 Plane for Stable Connectivity of Network OAM", draft-ietf- 2518 anima-stable-connectivity-02 (work in progress), February 2519 2017. 2521 [I-D.liang-iana-pen] 2522 Liang, P., Melnikov, A., and D. Conrad, "Private 2523 Enterprise Number (PEN) practices and Internet Assigned 2524 Numbers Authority (IANA) registration considerations", 2525 draft-liang-iana-pen-06 (work in progress), July 2015. 2527 [I-D.liu-anima-grasp-api] 2528 Carpenter, B., Liu, B., Wang, W., and X. Gong, "Generic 2529 Autonomic Signaling Protocol Application Program Interface 2530 (GRASP API)", draft-liu-anima-grasp-api-03 (work in 2531 progress), February 2017. 2533 [I-D.stenberg-anima-adncp] 2534 Stenberg, M., "Autonomic Distributed Node Consensus 2535 Protocol", draft-stenberg-anima-adncp-00 (work in 2536 progress), March 2015. 2538 [RFC2205] Braden, R., Ed., Zhang, L., Berson, S., Herzog, S., and S. 2539 Jamin, "Resource ReSerVation Protocol (RSVP) -- Version 1 2540 Functional Specification", RFC 2205, DOI 10.17487/RFC2205, 2541 September 1997, . 2543 [RFC2334] Luciani, J., Armitage, G., Halpern, J., and N. Doraswamy, 2544 "Server Cache Synchronization Protocol (SCSP)", RFC 2334, 2545 DOI 10.17487/RFC2334, April 1998, 2546 . 2548 [RFC2608] Guttman, E., Perkins, C., Veizades, J., and M. Day, 2549 "Service Location Protocol, Version 2", RFC 2608, 2550 DOI 10.17487/RFC2608, June 1999, 2551 . 2553 [RFC2865] Rigney, C., Willens, S., Rubens, A., and W. Simpson, 2554 "Remote Authentication Dial In User Service (RADIUS)", 2555 RFC 2865, DOI 10.17487/RFC2865, June 2000, 2556 . 2558 [RFC3209] Awduche, D., Berger, L., Gan, D., Li, T., Srinivasan, V., 2559 and G. Swallow, "RSVP-TE: Extensions to RSVP for LSP 2560 Tunnels", RFC 3209, DOI 10.17487/RFC3209, December 2001, 2561 . 2563 [RFC3315] Droms, R., Ed., Bound, J., Volz, B., Lemon, T., Perkins, 2564 C., and M. Carney, "Dynamic Host Configuration Protocol 2565 for IPv6 (DHCPv6)", RFC 3315, DOI 10.17487/RFC3315, July 2566 2003, . 2568 [RFC3416] Presuhn, R., Ed., "Version 2 of the Protocol Operations 2569 for the Simple Network Management Protocol (SNMP)", 2570 STD 62, RFC 3416, DOI 10.17487/RFC3416, December 2002, 2571 . 2573 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 2574 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 2575 DOI 10.17487/RFC4861, September 2007, 2576 . 2578 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 2579 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 2580 DOI 10.17487/RFC5226, May 2008, 2581 . 2583 [RFC5971] Schulzrinne, H. and R. Hancock, "GIST: General Internet 2584 Signalling Transport", RFC 5971, DOI 10.17487/RFC5971, 2585 October 2010, . 2587 [RFC6206] Levis, P., Clausen, T., Hui, J., Gnawali, O., and J. Ko, 2588 "The Trickle Algorithm", RFC 6206, DOI 10.17487/RFC6206, 2589 March 2011, . 2591 [RFC6241] Enns, R., Ed., Bjorklund, M., Ed., Schoenwaelder, J., Ed., 2592 and A. Bierman, Ed., "Network Configuration Protocol 2593 (NETCONF)", RFC 6241, DOI 10.17487/RFC6241, June 2011, 2594 . 2596 [RFC6733] Fajardo, V., Ed., Arkko, J., Loughney, J., and G. Zorn, 2597 Ed., "Diameter Base Protocol", RFC 6733, 2598 DOI 10.17487/RFC6733, October 2012, 2599 . 2601 [RFC6762] Cheshire, S. and M. Krochmal, "Multicast DNS", RFC 6762, 2602 DOI 10.17487/RFC6762, February 2013, 2603 . 2605 [RFC6763] Cheshire, S. and M. Krochmal, "DNS-Based Service 2606 Discovery", RFC 6763, DOI 10.17487/RFC6763, February 2013, 2607 . 2609 [RFC6887] Wing, D., Ed., Cheshire, S., Boucadair, M., Penno, R., and 2610 P. Selkirk, "Port Control Protocol (PCP)", RFC 6887, 2611 DOI 10.17487/RFC6887, April 2013, 2612 . 2614 [RFC7558] Lynn, K., Cheshire, S., Blanchet, M., and D. Migault, 2615 "Requirements for Scalable DNS-Based Service Discovery 2616 (DNS-SD) / Multicast DNS (mDNS) Extensions", RFC 7558, 2617 DOI 10.17487/RFC7558, July 2015, 2618 . 2620 [RFC7564] Saint-Andre, P. and M. Blanchet, "PRECIS Framework: 2621 Preparation, Enforcement, and Comparison of 2622 Internationalized Strings in Application Protocols", 2623 RFC 7564, DOI 10.17487/RFC7564, May 2015, 2624 . 2626 [RFC7575] Behringer, M., Pritikin, M., Bjarnason, S., Clemm, A., 2627 Carpenter, B., Jiang, S., and L. Ciavaglia, "Autonomic 2628 Networking: Definitions and Design Goals", RFC 7575, 2629 DOI 10.17487/RFC7575, June 2015, 2630 . 2632 [RFC7576] Jiang, S., Carpenter, B., and M. Behringer, "General Gap 2633 Analysis for Autonomic Networking", RFC 7576, 2634 DOI 10.17487/RFC7576, June 2015, 2635 . 2637 [RFC7787] Stenberg, M. and S. Barth, "Distributed Node Consensus 2638 Protocol", RFC 7787, DOI 10.17487/RFC7787, April 2016, 2639 . 2641 [RFC7788] Stenberg, M., Barth, S., and P. Pfister, "Home Networking 2642 Control Protocol", RFC 7788, DOI 10.17487/RFC7788, April 2643 2016, . 2645 [RFC8040] Bierman, A., Bjorklund, M., and K. Watsen, "RESTCONF 2646 Protocol", RFC 8040, DOI 10.17487/RFC8040, January 2017, 2647 . 2649 Appendix A. Open Issues [RFC Editor: This section should be empty. 2650 Please remove] 2652 o 68. (Placeholder) 2654 Appendix B. Closed Issues [RFC Editor: Please remove] 2656 o 1. UDP vs TCP: For now, this specification suggests UDP and TCP 2657 as message transport mechanisms. This is not clarified yet. UDP 2658 is good for short conversations, is necessary for multicast 2659 discovery, and generally fits the discovery and divert scenarios 2660 well. However, it will cause problems with large messages. TCP 2661 is good for stable and long sessions, with a little bit of time 2662 consumption during the session establishment stage. If messages 2663 exceed a reasonable MTU, a TCP mode will be required in any case. 2664 This question may be affected by the security discussion. 2666 RESOLVED by specifying UDP for short message and TCP for longer 2667 one. 2669 o 2. DTLS or TLS vs built-in security mechanism. For now, this 2670 specification has chosen a PKI based built-in security mechanism 2671 based on asymmetric cryptography. However, (D)TLS might be chosen 2672 as security solution to avoid duplication of effort. It also 2673 allows essentially similar security for short messages over UDP 2674 and longer ones over TCP. The implementation trade-offs are 2675 different. The current approach requires expensive asymmetric 2676 cryptographic calculations for every message. (D)TLS has startup 2677 overheads but cheaper crypto per message. DTLS is less mature 2678 than TLS. 2680 RESOLVED by specifying external security (ACP or (D)TLS). 2682 o The following open issues applied only if the original security 2683 model was retained: 2685 * 2.1. For replay protection, GRASP currently requires every 2686 participant to have an NTP-synchronized clock. Is this OK for 2687 low-end devices, and how does it work during device 2688 bootstrapping? We could take the Timestamp out of signature 2689 option, to become an independent and OPTIONAL (or RECOMMENDED) 2690 option. 2692 * 2.2. The Signature Option states that this option could be any 2693 place in a message. Wouldn't it be better to specify a 2694 position (such as the end)? That would be much simpler to 2695 implement. 2697 RESOLVED by changing security model. 2699 o 3. DoS Attack Protection needs work. 2701 RESOLVED by adding text. 2703 o 4. Should we consider preferring a text-based approach to 2704 discovery (after the initial discovery needed for bootstrapping)? 2705 This could be a complementary mechanism for multicast based 2706 discovery, especially for a very large autonomic network. 2707 Centralized registration could be automatically deployed 2708 incrementally. At the very first stage, the repository could be 2709 empty; then it could be filled in by the objectives discovered by 2710 different devices (for example using Dynamic DNS Update). The 2711 more records are stored in the repository, the less the multicast- 2712 based discovery is needed. However, if we adopt such a mechanism, 2713 there would be challenges: stateful solution, and security. 2715 RESOLVED for now by adding optional use of DNS-SD by ASAs. 2716 Subsequently removed by editors as irrelevant to GRASP istelf. 2718 o 5. Need to expand description of the minimum requirements for the 2719 specification of an individual discovery, synchronization or 2720 negotiation objective. 2722 RESOLVED for now by extra wording. 2724 o 6. Use case and protocol walkthrough. A description of how a 2725 node starts up, performs discovery, and conducts negotiation and 2726 synchronisation for a sample use case would help readers to 2727 understand the applicability of this specification. Maybe it 2728 should be an artificial use case or maybe a simple real one, based 2729 on a conceptual API. However, the authors have not yet decided 2730 whether to have a separate document or have it in the protocol 2731 document. 2733 RESOLVED: recommend a separate document. 2735 o 7. Cross-check against other ANIMA WG documents for consistency 2736 and gaps. 2738 RESOLVED: Satisfied by WGLC. 2740 o 8. Consideration of ADNCP proposal. 2742 RESOLVED by adding optional use of DNCP for flooding-type 2743 synchronization. 2745 o 9. Clarify how a GDNP instance knows whether it is running inside 2746 the ACP. (Sheng) 2748 RESOLVED by improved text. 2750 o 10. Clarify how a non-ACP GDNP instance initiates (D)TLS. 2751 (Sheng) 2753 RESOLVED by improved text and declaring DTLS out of scope for this 2754 draft. 2756 o 11. Clarify how UDP/TCP choice is made. (Sheng) [Like DNS? - 2757 Brian] 2759 RESOLVED by improved text. 2761 o 12. Justify that IP address within ACP or (D)TLS environment is 2762 sufficient to prove AN identity; or explain how Device Identity 2763 Option is used. (Sheng) 2765 RESOLVED for now: we assume that all ASAs in a device are trusted 2766 as soon as the device is trusted, so they share credentials. In 2767 that case the Device Identity Option is useless. This needs to be 2768 reviewed later. 2770 o 13. Emphasise that negotiation/synchronization are independent 2771 from discovery, although the rapid discovery mode includes the 2772 first step of a negotiation/synchronization. (Sheng) 2774 RESOLVED by improved text. 2776 o 14. Do we need an unsolicited flooding mechanism for discovery 2777 (for discovery results that everyone needs), to reduce scaling 2778 impact of flooding discovery messages? (Toerless) 2780 RESOLVED: Yes, added to requirements and solution. 2782 o 15. Do we need flag bits in Objective Options to distinguish 2783 distinguish Synchronization and Negotiation "Request" or rapid 2784 mode "Discovery" messages? (Bing) 2786 RESOLVED: yes, work on the API showed that these flags are 2787 essential. 2789 o 16. (Related to issue 14). Should we revive the "unsolicited 2790 Response" for flooding synchronisation data? This has to be done 2791 carefully due to the well-known issues with flooding, but it could 2792 be useful, e.g. for Intent distribution, where DNCP doesn't seem 2793 applicable. 2795 RESOLVED: Yes, see #14. 2797 o 17. Ensure that the discovery mechanism is completely proof 2798 against loops and protected against duplicate responses. 2800 RESOLVED: Added loop count mechanism. 2802 o 18. Discuss the handling of multiple valid discovery responses. 2804 RESOLVED: Stated that the choice must be available to the ASA but 2805 GRASP implementation should pick a default. 2807 o 19. Should we use a text-oriented format such as JSON/CBOR 2808 instead of native binary TLV format? 2810 RESOLVED: Yes, changed to CBOR. 2812 o 20. Is the Divert option needed? If a discovery response 2813 provides a valid IP address or FQDN, the recipient doesn't gain 2814 any extra knowledge from the Divert. On the other hand, the 2815 presence of Divert informs the receiver that the target is off- 2816 link, which might be useful sometimes. 2818 RESOLVED: Decided to keep Divert option. 2820 o 21. Rename the protocol as GRASP (GeneRic Autonomic Signaling 2821 Protocol)? 2823 RESOLVED: Yes, name changed. 2825 o 22. Does discovery mechanism scale robustly as needed? Need hop 2826 limit on relaying? 2828 RESOLVED: Added hop limit. 2830 o 23. Need more details on TTL for caching discovery responses. 2832 RESOLVED: Done. 2834 o 24. Do we need "fast withdrawal" of discovery responses? 2836 RESOLVED: This doesn't seem necessary. If an ASA exits or stops 2837 supporting a given objective, peers will fail to start future 2838 sessions and will simply repeat discovery. 2840 o 25. Does GDNP discovery meet the needs of multi-hop DNS-SD? 2842 RESOLVED: Decided not to consider this further as a GRASP protocol 2843 issue. GRASP objectives could embed DNS-SD formats if needed. 2845 o 26. Add a URL type to the locator options (for security bootstrap 2846 etc.) 2848 RESOLVED: Done, later renamed as URI. 2850 o 27. Security of Flood multicasts (Section 3.5.6.2). 2852 RESOLVED: added text. 2854 o 28. Does ACP support secure link-local multicast? 2856 RESOLVED by new text in the Security Considerations. 2858 o 29. PEN is used to distinguish vendor options. Would it be 2859 better to use a domain name? Anything unique will do. 2861 RESOLVED: Simplified this by removing PEN field and changing 2862 naming rules for objectives. 2864 o 30. Does response to discovery require randomized delays to 2865 mitigate amplification attacks? 2867 RESOLVED: WG feedback is that it's unnecessary. 2869 o 31. We have specified repeats for failed discovery etc. Is that 2870 sufficient to deal with sleeping nodes? 2872 RESOLVED: WG feedback is that it's unnecessary to say more. 2874 o 32. We have one-to-one synchronization and flooding 2875 synchronization. Do we also need selective flooding to a subset 2876 of nodes? 2878 RESOLVED: This will be discussed as a protocol extension in a 2879 separate draft (draft-liu-anima-grasp-distribution). 2881 o 33. Clarify if/when discovery needs to be repeated. 2883 RESOLVED: Done. 2885 o 34. Clarify what is mandatory for running in ACP, expand 2886 discussion of security boundary when running with no ACP - might 2887 rely on the local PKI infrastructure. 2889 RESOLVED: Done. 2891 o 35. State that role-based authorization of ASAs is out of scope 2892 for GRASP. GRASP doesn't recognize/handle any "roles". 2894 RESOLVED: Done. 2896 o 36. Reconsider CBOR definition for PEN syntax. ( objective-name 2897 = text / [pen, text] ; pen = uint ) 2899 RESOLVED: See issue 29. 2901 o 37. Are URI locators really needed? 2903 RESOLVED: Yes, e.g. for security bootstrap discovery, but added 2904 note that addresses are the normal case (same for FQDN locators). 2906 o 38. Is Session ID sufficient to identify relayed responses? 2907 Isn't the originator's address needed too? 2909 RESOLVED: Yes, this is needed for multicast messages and their 2910 responses. 2912 o 39. Clarify that a node will contain one GRASP instance 2913 supporting multiple ASAs. 2915 RESOLVED: Done. 2917 o 40. Add a "reason" code to the DECLINE option? 2919 RESOLVED: Done. 2921 o 41. What happens if an ASA cannot conveniently use one of the 2922 GRASP mechanisms? Do we (a) add a message type to GRASP, or (b) 2923 simply pass the discovery results to the ASA so that it can open 2924 its own socket? 2926 RESOLVED: Both would be possible, but (b) is preferred. 2928 o 42. Do we need a feature whereby an ASA can bypass the ACP and 2929 use the data plane for efficiency/throughput? This would require 2930 discovery to return non-ACP addresses and would evade ACP 2931 security. 2933 RESOLVED: This is considered out of scope for GRASP, but a comment 2934 has been added in security considerations. 2936 o 43. Rapid mode synchronization and negotiation is currently 2937 limited to a single objective for simplicity of design and 2938 implementation. A future consideration is to allow multiple 2939 objectives in rapid mode for greater efficiency. 2941 RESOLVED: This is considered out of scope for this version. 2943 o 44. In requirement T9, the words that encryption "may not be 2944 required in all deployments" were removed. Is that OK?. 2946 RESOLVED: No objections. 2948 o 45. Device Identity Option is unused. Can we remove it 2949 completely?. 2951 RESOLVED: No objections. Done. 2953 o 46. The 'initiator' field in DISCOVER, RESPONSE and FLOOD 2954 messages is intended to assist in loop prevention. However, we 2955 also have the loop count for that. Also, if we create a new 2956 Session ID each time a DISCOVER or FLOOD is relayed, that ID can 2957 be disambiguated by recipients. It would be simpler to remove the 2958 initiator from the messages, making parsing more uniform. Is that 2959 OK? 2961 RESOLVED: Yes. Done. 2963 o 47. REQUEST is a dual purpose message (request negotiation or 2964 request synchronization). Would it be better to split this into 2965 two different messages (and adjust various message names 2966 accordingly)? 2968 RESOLVED: Yes. Done. 2970 o 48. Should the Appendix "Capability Analysis of Current 2971 Protocols" be deleted before RFC publication? 2973 RESOLVED: No (per WG meeting at IETF 96). 2975 o 49. Section 3.5.1 Should say more about signaling between two 2976 autonomic networks/domains. 2978 RESOLVED: Description of separate GRASP instance added. 2980 o 50. Is Rapid mode limited to on-link only? What happens if first 2981 discovery responder does not support Rapid Mode? Section 3.5.5, 2982 Section 3.5.6) 2983 RESOLVED: Not limited to on-link. First responder wins. 2985 o 51. Should flooded objectives have a time-to-live before they are 2986 deleted from the flood cache? And should they be tagged in the 2987 cache with their source locator? 2989 RESOLVED: TTL added to Flood (and Discovery Response) messages. 2990 Cached flooded objectives must be tagged with their originating 2991 ASA locator, and multiple copies must be kept if necessary. 2993 o 52. Describe in detail what is allowed and disallowed in an 2994 insecure instance of GRASP. 2996 RESOLVED: Done. 2998 o 53. Tune IANA Considerations to support early assignment request. 3000 o 54. Is there a highly unlikely race condition if two peers 3001 simultaneously choose the same Session ID and send each other 3002 simultaneous M_REQ_NEG messages? 3004 RESOLVED: Yes. Enhanced text on Session ID generation, and added 3005 precaution when receiving a Request message. 3007 o 55. Could discovery be performed over TCP? 3009 RESOLVED: Unicast discovery added as an option. 3011 o 56. Change Session-ID to 32 bits? 3013 RESOLVED: Done. 3015 o 57. Add M_INVALID message? 3017 RESOLVED: Done. 3019 o 58. Maximum message size? 3021 RESOLVED by specifying default maximum message size (2048 bytes). 3023 o 59. Add F_NEG_DRY flag to specify a "dry run" objective?. 3025 RESOLVED: Done. 3027 o 60. Change M_FLOOD syntax to associate a locator with each 3028 objective? 3029 RESOLVED: Done. 3031 o 61. Is the SONN constrained instance really needed? 3033 RESOLVED: Retained but only as an option. 3035 o 62. Is it helpful to tag descriptive text with message names 3036 (M_DISCOVER etc.)? 3038 RESOLVED: Yes, done in various parts of the text. 3040 o 63. Should encryption be MUST instead of SHOULD in Section 3.5.1 3041 and Section 3.5.2.1? 3043 RESOLVED: Yes, MUST implement in both cases. 3045 o 64. Should more security text be moved from the main text into 3046 the Security Considerations? 3048 RESOLVED: No, on AD advice. 3050 o 65. Do we need to formally restrict Unicode characters allowed in 3051 objective names? 3053 RESOLVED: No, but need to point to guidance from PRECIS WG. 3055 o 66. Split requirements into separate document? 3057 RESOLVED: No, on AD advice. 3059 o 67. Remove normative dependency on draft-greevenbosch-appsawg- 3060 cbor-cddl? 3062 RESOLVED: No, on AD advice. In worst case, fix at AUTH48. 3064 Appendix C. Change log [RFC Editor: Please remove] 3066 draft-ietf-anima-grasp-12, 2017-05-19: 3068 Updates following IESG comments: 3070 Clarified that GRASP runs in a single addressing realm 3072 Improved wording about FQDN resolution, clarified that URI usage is 3073 out of scope. 3075 Clarified description of negotiation timeout. 3077 Noted that 'dry run' semantics are ASA-dependent 3079 Made the ACP a normative reference 3081 Clarified that LL multicasts are limited to GRASP interfaces 3083 Unicast UDP moved out of scope 3085 Editorial clarifications 3087 draft-ietf-anima-grasp-11, 2017-03-30: 3089 Updates following IETF 98 discussion: 3091 Encryption changed to a MUST implement. 3093 Pointed to guidance on UTF-8 names. 3095 draft-ietf-anima-grasp-10, 2017-03-10: 3097 Updates following IETF Last call: 3099 Protocol change: Specify that an objective with no initial value 3100 should have its value field set to CBOR 'null'. 3102 Protocol change: Specify behavior on receiving unrecognized message 3103 type. 3105 Noted that UTF-8 names are matched byte-for-byte. 3107 Added brief guidance for Expert Reviewer of new generic objectives. 3109 Numerous editorial improvements and clarifications and minor text 3110 rearrangements, none intended to change the meaning. 3112 draft-ietf-anima-grasp-09, 2016-12-15: 3114 Protocol change: Add F_NEG_DRY flag to specify a "dry run" objective. 3116 Protocol change: Change M_FLOOD syntax to associate a locator with 3117 each objective. 3119 Concentrated mentions of TLS in one section, with all details out of 3120 scope. 3122 Clarified text around constrained instances of GRASP. 3124 Strengthened text restricting LL addresses in locator options. 3126 Clarified description of rapid mode processsing. 3128 Specified that cached discovery results should not be returned on the 3129 same interface where they were learned. 3131 Shortened text in "High Level Design Choices" 3133 Dropped the word 'kernel' to avoid confusion with o/s kernel mode. 3135 Editorial improvements and clarifications. 3137 draft-ietf-anima-grasp-08, 2016-10-30: 3139 Protocol change: Added M_INVALID message. 3141 Protocol change: Increased Session ID space to 32 bits. 3143 Enhanced rules to avoid Session ID clashes. 3145 Corrected and completed description of timeouts for Request messages. 3147 Improved wording about exponential backoff and DoS. 3149 Clarified that discovery relaying is not done by limited security 3150 instances. 3152 Corrected and expanded explanation of port used for Discovery 3153 Response. 3155 Noted that Discovery message could be sent unicast in special cases. 3157 Added paragraph on extensibility. 3159 Specified default maximum message size. 3161 Added Appendix for sample messages. 3163 Added short protocol overview. 3165 Editorial fixes, including minor re-ordering for readability. 3167 draft-ietf-anima-grasp-07, 2016-09-13: 3169 Protocol change: Added TTL field to Flood message (issue 51). 3171 Protocol change: Added Locator option to Flood message (issue 51). 3173 Protocol change: Added TTL field to Discovery Response message 3174 (corrollary to issue 51). 3176 Clarified details of rapid mode (issues 43 and 50). 3178 Description of inter-domain GRASP instance added (issue 49). 3180 Description of limited security GRASP instances added (issue 52). 3182 Strengthened advice to use TCP rather than UDP. 3184 Updated IANA considerations and text about well-known port usage 3185 (issue 53). 3187 Amended text about ASA authorization and roles to allow for 3188 overlapping ASAs. 3190 Added text recommending that Flood should be repeated periodically. 3192 Editorial fixes. 3194 draft-ietf-anima-grasp-06, 2016-06-27: 3196 Added text on discovery cache timeouts. 3198 Noted that ASAs that are only initiators do not need to respond to 3199 discovery message. 3201 Added text on unexpected address changes. 3203 Added text on robust implementation. 3205 Clarifications and editorial fixes for numerous review comments 3207 Added open issues for some review comments. 3209 draft-ietf-anima-grasp-05, 2016-05-13: 3211 Noted in requirement T1 that it should be possible to implement ASAs 3212 independently as user space programs. 3214 Protocol change: Added protocol number and port to discovery 3215 response. Updated protocol description, CDDL and IANA considerations 3216 accordingly. 3218 Clarified that discovery and flood multicasts are handled by the 3219 GRASP core, not directly by ASAs. 3221 Clarified that a node may discover an objective without supporting it 3222 for synchronization or negotiation. 3224 Added Implementation Status section. 3226 Added reference to SCSP. 3228 Editorial fixes. 3230 draft-ietf-anima-grasp-04, 2016-03-11: 3232 Protocol change: Restored initiator field in certain messages and 3233 adjusted relaying rules to provide complete loop detection. 3235 Updated IANA Considerations. 3237 draft-ietf-anima-grasp-03, 2016-02-24: 3239 Protocol change: Removed initiator field from certain messages and 3240 adjusted relaying requirement to simplify loop detection. Also 3241 clarified narrative explanation of discovery relaying. 3243 Protocol change: Split Request message into two (Request Negotiation 3244 and Request Synchronization) and updated other message names for 3245 clarity. 3247 Protocol change: Dropped unused Device ID option. 3249 Further clarified text on transport layer usage. 3251 New text about multicast insecurity in Security Considerations. 3253 Various other clarifications and editorial fixes, including moving 3254 some material to Appendix. 3256 draft-ietf-anima-grasp-02, 2016-01-13: 3258 Resolved numerous issues according to WG discussions. 3260 Renumbered requirements, added D9. 3262 Protocol change: only allow one objective in rapid mode. 3264 Protocol change: added optional error string to DECLINE option. 3266 Protocol change: removed statement that seemed to say that a Request 3267 not preceded by a Discovery should cause a Discovery response. That 3268 made no sense, because there is no way the initiator would know where 3269 to send the Request. 3271 Protocol change: Removed PEN option from vendor objectives, changed 3272 naming rule accordingly. 3274 Protocol change: Added FLOOD message to simplify coding. 3276 Protocol change: Added SYNCH message to simplify coding. 3278 Protocol change: Added initiator id to DISCOVER, RESPONSE and FLOOD 3279 messages. But also allowed the relay process for DISCOVER and FLOOD 3280 to regenerate a Session ID. 3282 Protocol change: Require that discovered addresses must be global 3283 (except during bootstrap). 3285 Protocol change: Receiver of REQUEST message must close socket if no 3286 ASA is listening for the objective. 3288 Protocol change: Simplified Waiting message. 3290 Protocol change: Added No Operation message. 3292 Renamed URL locator type as URI locator type. 3294 Updated CDDL definition. 3296 Various other clarifications and editorial fixes. 3298 draft-ietf-anima-grasp-01, 2015-10-09: 3300 Updated requirements after list discussion. 3302 Changed from TLV to CBOR format - many detailed changes, added co- 3303 author. 3305 Tightened up loop count and timeouts for various cases. 3307 Noted that GRASP does not provide transactional integrity. 3309 Various other clarifications and editorial fixes. 3311 draft-ietf-anima-grasp-00, 2015-08-14: 3313 File name and protocol name changed following WG adoption. 3315 Added URL locator type. 3317 draft-carpenter-anima-gdn-protocol-04, 2015-06-21: 3319 Tuned wording around hierarchical structure. 3321 Changed "device" to "ASA" in many places. 3323 Reformulated requirements to be clear that the ASA is the main 3324 customer for signaling. 3326 Added requirement for flooding unsolicited synch, and added it to 3327 protocol spec. Recognized DNCP as alternative for flooding synch 3328 data. 3330 Requirements clarified, expanded and rearranged following design team 3331 discussion. 3333 Clarified that GDNP discovery must not be a prerequisite for GDNP 3334 negotiation or synchronization (resolved issue 13). 3336 Specified flag bits for objective options (resolved issue 15). 3338 Clarified usage of ACP vs TLS/DTLS and TCP vs UDP (resolved issues 3339 9,10,11). 3341 Updated DNCP description from latest DNCP draft. 3343 Editorial improvements. 3345 draft-carpenter-anima-gdn-protocol-03, 2015-04-20: 3347 Removed intrinsic security, required external security 3349 Format changes to allow DNCP co-existence 3351 Recognized DNS-SD as alternative discovery method. 3353 Editorial improvements 3355 draft-carpenter-anima-gdn-protocol-02, 2015-02-19: 3357 Tuned requirements to clarify scope, 3359 Clarified relationship between types of objective, 3361 Clarified that objectives may be simple values or complex data 3362 structures, 3364 Improved description of objective options, 3365 Added loop-avoidance mechanisms (loop count and default timeout, 3366 limitations on discovery relaying and on unsolicited responses), 3368 Allow multiple discovery objectives in one response, 3370 Provided for missing or multiple discovery responses, 3372 Indicated how modes such as "dry run" should be supported, 3374 Minor editorial and technical corrections and clarifications, 3376 Reorganized future work list. 3378 draft-carpenter-anima-gdn-protocol-01, restructured the logical flow 3379 of the document, updated to describe synchronization completely, add 3380 unsolicited responses, numerous corrections and clarifications, 3381 expanded future work list, 2015-01-06. 3383 draft-carpenter-anima-gdn-protocol-00, combination of draft-jiang- 3384 config-negotiation-ps-03 and draft-jiang-config-negotiation-protocol- 3385 02, 2014-10-08. 3387 Appendix D. Example Message Formats 3389 For readers unfamiliar with CBOR, this appendix shows a number of 3390 example GRASP messages conforming to the CDDL syntax given in 3391 Section 6. Each message is shown three times in the following 3392 formats: 3394 1. CBOR diagnostic notation. 3396 2. Similar, but showing the names of the constants. (Details of the 3397 flag bit encoding are omitted.) 3399 3. Hexadecimal version of the CBOR wire format. 3401 Long lines are split for display purposes only. 3403 D.1. Discovery Example 3405 The initiator (2001:db8:f000:baaa:28cc:dc4c:9703:6781) multicasts a 3406 discovery message looking for objective EX1: 3408 [1, 13948744, h'20010db8f000baaa28ccdc4c97036781', ["EX1", 5, 2, 0]] 3409 [M_DISCOVERY, 13948744, h'20010db8f000baaa28ccdc4c97036781', 3410 ["EX1", F_SYNCH_bits, 2, 0]] 3411 h'84011a00d4d7485020010db8f000baaa28ccdc4c970367818463455831050200' 3412 A peer (2001:0db8:f000:baaa:f000:baaa:f000:baaa) responds with a 3413 locator: 3415 [2, 13948744, h'20010db8f000baaa28ccdc4c97036781', 60000, 3416 [103, h'20010db8f000baaaf000baaaf000baaa', 6, 49443]] 3417 [M_RESPONSE, 13948744, h'20010db8f000baaa28ccdc4c97036781', 60000, 3418 [O_IPv6_LOCATOR, h'20010db8f000baaaf000baaaf000baaa', 3419 IPPROTO_TCP, 49443]] 3420 h'85021a00d4d7485020010db8f000baaa28ccdc4c9703678119ea6084186750 3421 20010db8f000baaaf000baaaf000baaa0619c123' 3423 D.2. Flood Example 3425 The initiator multicasts a flood message. The single objective has a 3426 null locator. There is no response: 3428 [9, 3504974, h'20010db8f000baaa28ccdc4c97036781', 10000, 3429 [["EX1", 5, 2, ["Example 1 value=", 100]],[] ] ] 3430 [M_FLOOD, 3504974, h'20010db8f000baaa28ccdc4c97036781', 10000, 3431 [["EX1", F_SYNCH_bits, 2, ["Example 1 value=", 100]],[] ] ] 3432 h'86091a00357b4e5020010db8f000baaa28ccdc4c97036781192710 3433 828463455831050282704578616d706c6520312076616c75653d186480' 3435 D.3. Synchronization Example 3437 Following successful discovery of objective EX2, the initiator 3438 unicasts a request: 3440 [4, 4038926, ["EX2", 5, 5, 0]] 3441 [M_REQ_SYN, 4038926, ["EX2", F_SYNCH_bits, 5, 0]] 3442 h'83041a003da10e8463455832050500' 3444 The peer responds with a value: 3446 [8, 4038926, ["EX2", 5, 5, ["Example 2 value=", 200]]] 3447 [M_SYNCH, 4038926, ["EX2", F_SYNCH_bits, 5, ["Example 2 value=", 200]]] 3448 h'83081a003da10e8463455832050582704578616d706c6520322076616c75653d18c8' 3450 D.4. Simple Negotiation Example 3452 Following successful discovery of objective EX3, the initiator 3453 unicasts a request: 3455 [3, 802813, ["EX3", 3, 6, ["NZD", 47]]] 3456 [M_REQ_NEG, 802813, ["EX3", F_NEG_bits, 6, ["NZD", 47]]] 3457 h'83031a000c3ffd8463455833030682634e5a44182f' 3458 The peer responds with immediate acceptance. Note that no objective 3459 is needed, because the initiator's request was accepted without 3460 change: 3462 [6, 802813, [101]] 3463 [M_END , 802813, [O_ACCEPT]] 3464 h'83061a000c3ffd811865' 3466 D.5. Complete Negotiation Example 3468 Again the initiator unicasts a request: 3470 [3, 13767778, ["EX3", 3, 6, ["NZD", 410]]] 3471 [M_REQ_NEG, 13767778, ["EX3", F_NEG_bits, 6, ["NZD", 410]]] 3472 h'83031a00d214628463455833030682634e5a4419019a' 3474 The responder starts to negotiate (making an offer): 3476 [5, 13767778, ["EX3", 3, 6, ["NZD", 80]]] 3477 [M_NEGOTIATE, 13767778, ["EX3", F_NEG_bits, 6, ["NZD", 80]]] 3478 h'83051a00d214628463455833030682634e5a441850' 3480 The initiator continues to negotiate (reducing its request, and note 3481 that the loop count is decremented): 3483 [5, 13767778, ["EX3", 3, 5, ["NZD", 307]]] 3484 [M_NEGOTIATE, 13767778, ["EX3", F_NEG_bits, 5, ["NZD", 307]]] 3485 h'83051a00d214628463455833030582634e5a44190133' 3487 The responder asks for more time: 3489 [7, 13767778, 34965] 3490 [M_WAIT, 13767778, 34965] 3491 h'83071a00d21462198895' 3493 The responder continues to negotiate (increasing its offer): 3495 [5, 13767778, ["EX3", 3, 4, ["NZD", 120]]] 3496 [M_NEGOTIATE, 13767778, ["EX3", F_NEG_bits, 4, ["NZD", 120]]] 3497 h'83051a00d214628463455833030482634e5a441878' 3499 The initiator continues to negotiate (reducing its request): 3501 [5, 13767778, ["EX3", 3, 3, ["NZD", 246]]] 3502 [M_NEGOTIATE, 13767778, ["EX3", F_NEG_bits, 3, ["NZD", 246]]] 3503 h'83051a00d214628463455833030382634e5a4418f6' 3505 The responder refuses to negotiate further: 3507 [6, 13767778, [102, "Insufficient funds"]] 3508 [M_END , 13767778, [O_DECLINE, "Insufficient funds"]] 3509 h'83061a00d2146282186672496e73756666696369656e742066756e6473' 3511 This negotiation has failed. If either side had sent [M_END, 3512 13767778, [O_ACCEPT]] it would have succeeded, converging on the 3513 objective value in the preceding M_NEGOTIATE. Note that apart from 3514 the initial M_REQ_NEG, the process is symmetrical. 3516 Appendix E. Capability Analysis of Current Protocols 3518 This appendix discusses various existing protocols with properties 3519 related to the requirements described in Section 2. The purpose is 3520 to evaluate whether any existing protocol, or a simple combination of 3521 existing protocols, can meet those requirements. 3523 Numerous protocols include some form of discovery, but these all 3524 appear to be very specific in their applicability. Service Location 3525 Protocol (SLP) [RFC2608] provides service discovery for managed 3526 networks, but requires configuration of its own servers. DNS-SD 3527 [RFC6763] combined with mDNS [RFC6762] provides service discovery for 3528 small networks with a single link layer. [RFC7558] aims to extend 3529 this to larger autonomous networks but this is not yet standardized. 3530 However, both SLP and DNS-SD appear to target primarily application 3531 layer services, not the layer 2 and 3 objectives relevant to basic 3532 network configuration. Both SLP and DNS-SD are text-based protocols. 3534 Routing protocols are mainly one-way information announcements. The 3535 receiver makes independent decisions based on the received 3536 information and there is no direct feedback information to the 3537 announcing peer. This remains true even though the protocol is used 3538 in both directions between peer routers; there is state 3539 synchronization, but no negotiation, and each peer runs its route 3540 calculations independently. 3542 Simple Network Management Protocol (SNMP) [RFC3416] uses a command/ 3543 response model not well suited for peer negotiation. Network 3544 Configuration Protocol (NETCONF) [RFC6241] uses an RPC model that 3545 does allow positive or negative responses from the target system, but 3546 this is still not adequate for negotiation. 3548 There are various existing protocols that have elementary negotiation 3549 abilities, such as Dynamic Host Configuration Protocol for IPv6 3550 (DHCPv6) [RFC3315], Neighbor Discovery (ND) [RFC4861], Port Control 3551 Protocol (PCP) [RFC6887], Remote Authentication Dial In User Service 3552 (RADIUS) [RFC2865], Diameter [RFC6733], etc. Most of them are 3553 configuration or management protocols. However, they either provide 3554 only a simple request/response model in a master/slave context or 3555 very limited negotiation abilities. 3557 There are some signaling protocols with an element of negotiation. 3558 For example Resource ReSerVation Protocol (RSVP) [RFC2205] was 3559 designed for negotiating quality of service parameters along the path 3560 of a unicast or multicast flow. RSVP is a very specialised protocol 3561 aimed at end-to-end flows. However, it has some flexibility, having 3562 been extended for MPLS label distribution [RFC3209]. A more generic 3563 design is General Internet Signalling Transport (GIST) [RFC5971], but 3564 it is complex, tries to solve many problems, and is also aimed at 3565 per-flow signaling across many hops rather than at device-to-device 3566 signaling. However, we cannot completely exclude extended RSVP or 3567 GIST as a synchronization and negotiation protocol. They do not 3568 appear to be directly useable for peer discovery. 3570 RESTCONF [RFC8040] is a protocol intended to convey NETCONF 3571 information expressed in the YANG language via HTTP, including the 3572 ability to transit HTML intermediaries. While this is a powerful 3573 approach in the context of centralised configuration of a complex 3574 network, it is not well adapted to efficient interactive negotiation 3575 between peer devices, especially simple ones that might not include 3576 YANG processing already. 3578 The Distributed Node Consensus Protocol (DNCP) [RFC7787] is defined 3579 as a generic form of state synchronization protocol, with a proposed 3580 usage profile being the Home Networking Control Protocol (HNCP) 3581 [RFC7788] for configuring Homenet routers. A specific application of 3582 DNCP for autonomic networking was proposed in 3583 [I-D.stenberg-anima-adncp]. 3585 DNCP "is designed to provide a way for each participating node to 3586 publish a set of TLV (Type-Length-Value) tuples, and to provide a 3587 shared and common view about the data published... DNCP is most 3588 suitable for data that changes only infrequently... If constant rapid 3589 state changes are needed, the preferable choice is to use an 3590 additional point-to-point channel..." 3592 Specific features of DNCP include: 3594 o Every participating node has a unique node identifier. 3596 o DNCP messages are encoded as a sequence of TLV objects, sent over 3597 unicast UDP or TCP, with or without (D)TLS security. 3599 o Multicast is used only for discovery of DNCP neighbors when lower 3600 security is acceptable. 3602 o Synchronization of state is maintained by a flooding process using 3603 the Trickle algorithm. There is no bilateral synchronization or 3604 negotiation capability. 3606 o The HNCP profile of DNCP is designed to operate between directly 3607 connected neighbors on a shared link using UDP and link-local IPv6 3608 addresses. 3610 DNCP does not meet the needs of a general negotiation protocol, 3611 because it is designed specifically for flooding synchronization. 3612 Also, in its HNCP profile it is limited to link-local messages and to 3613 IPv6. However, at the minimum it is a very interesting test case for 3614 this style of interaction between devices without needing a central 3615 authority, and it is a proven method of network-wide state 3616 synchronization by flooding. 3618 The Server Cache Synchronization Protocol (SCSP) [RFC2334] also 3619 describes a method for cache synchronization and cache replication 3620 among a group of nodes. 3622 A proposal was made some years ago for an IP based Generic Control 3623 Protocol (IGCP) [I-D.chaparadza-intarea-igcp]. This was aimed at 3624 information exchange and negotiation but not directly at peer 3625 discovery. However, it has many points in common with the present 3626 work. 3628 None of the above solutions appears to completely meet the needs of 3629 generic discovery, state synchronization and negotiation in a single 3630 solution. Many of the protocols assume that they are working in a 3631 traditional top-down or north-south scenario, rather than a fluid 3632 peer-to-peer scenario. Most of them are specialized in one way or 3633 another. As a result, we have not identified a combination of 3634 existing protocols that meets the requirements in Section 2. Also, 3635 we have not identified a path by which one of the existing protocols 3636 could be extended to meet the requirements. 3638 Authors' Addresses 3640 Carsten Bormann 3641 Universitaet Bremen TZI 3642 Postfach 330440 3643 D-28359 Bremen 3644 Germany 3646 Email: cabo@tzi.org 3647 Brian Carpenter (editor) 3648 Department of Computer Science 3649 University of Auckland 3650 PB 92019 3651 Auckland 1142 3652 New Zealand 3654 Email: brian.e.carpenter@gmail.com 3656 Bing Liu (editor) 3657 Huawei Technologies Co., Ltd 3658 Q14, Huawei Campus 3659 No.156 Beiqing Road 3660 Hai-Dian District, Beijing 100095 3661 P.R. China 3663 Email: leo.liubing@huawei.com