idnits 2.17.1 draft-ietf-dots-requirements-04.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 : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (March 13, 2017) is 2598 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 793 (Obsoleted by RFC 9293) ** Obsolete normative reference: RFC 2373 (Obsoleted by RFC 3513) ** Obsolete normative reference: RFC 5405 (Obsoleted by RFC 8085) == Outdated reference: A later version (-18) exists of draft-ietf-dots-architecture-01 == Outdated reference: A later version (-25) exists of draft-ietf-dots-use-cases-03 Summary: 4 errors (**), 0 flaws (~~), 3 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 DOTS A. Mortensen 3 Internet-Draft Arbor Networks, Inc. 4 Intended status: Informational R. Moskowitz 5 Expires: September 14, 2017 HTT Consulting 6 T. Reddy 7 Cisco Systems, Inc. 8 March 13, 2017 10 Distributed Denial of Service (DDoS) Open Threat Signaling Requirements 11 draft-ietf-dots-requirements-04 13 Abstract 15 This document defines the requirements for the Distributed Denial of 16 Service (DDoS) Open Threat Signaling (DOTS) protocols coordinating 17 attack response against DDoS attacks. 19 Status of This Memo 21 This Internet-Draft is submitted in full conformance with the 22 provisions of BCP 78 and BCP 79. 24 Internet-Drafts are working documents of the Internet Engineering 25 Task Force (IETF). Note that other groups may also distribute 26 working documents as Internet-Drafts. The list of current Internet- 27 Drafts is at http://datatracker.ietf.org/drafts/current/. 29 Internet-Drafts are draft documents valid for a maximum of six months 30 and may be updated, replaced, or obsoleted by other documents at any 31 time. It is inappropriate to use Internet-Drafts as reference 32 material or to cite them other than as "work in progress." 34 This Internet-Draft will expire on September 14, 2017. 36 Copyright Notice 38 Copyright (c) 2017 IETF Trust and the persons identified as the 39 document authors. All rights reserved. 41 This document is subject to BCP 78 and the IETF Trust's Legal 42 Provisions Relating to IETF Documents 43 (http://trustee.ietf.org/license-info) in effect on the date of 44 publication of this document. Please review these documents 45 carefully, as they describe your rights and restrictions with respect 46 to this document. Code Components extracted from this document must 47 include Simplified BSD License text as described in Section 4.e of 48 the Trust Legal Provisions and are provided without warranty as 49 described in the Simplified BSD License. 51 Table of Contents 53 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 54 1.1. Context and Motivation . . . . . . . . . . . . . . . . . 2 55 1.2. Terminology . . . . . . . . . . . . . . . . . . . . . . . 3 56 2. Requirements . . . . . . . . . . . . . . . . . . . . . . . . 5 57 2.1. General Requirements . . . . . . . . . . . . . . . . . . 7 58 2.2. Operational Requirements . . . . . . . . . . . . . . . . 8 59 2.3. Data Channel Requirements . . . . . . . . . . . . . . . . 11 60 2.4. Security requirements . . . . . . . . . . . . . . . . . . 12 61 2.5. Data Model Requirements . . . . . . . . . . . . . . . . . 13 62 3. Congestion Control Considerations . . . . . . . . . . . . . . 14 63 3.1. Signal Channel . . . . . . . . . . . . . . . . . . . . . 14 64 3.2. Data Channel . . . . . . . . . . . . . . . . . . . . . . 15 65 4. Security Considerations . . . . . . . . . . . . . . . . . . . 15 66 5. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 15 67 6. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 16 68 7. Change Log . . . . . . . . . . . . . . . . . . . . . . . . . 16 69 7.1. 04 revision . . . . . . . . . . . . . . . . . . . . . . . 16 70 7.2. 03 revision . . . . . . . . . . . . . . . . . . . . . . . 16 71 7.3. 02 revision . . . . . . . . . . . . . . . . . . . . . . . 16 72 7.4. 01 revision . . . . . . . . . . . . . . . . . . . . . . . 16 73 7.5. 00 revision . . . . . . . . . . . . . . . . . . . . . . . 17 74 7.6. Initial revision . . . . . . . . . . . . . . . . . . . . 17 75 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 17 76 8.1. Normative References . . . . . . . . . . . . . . . . . . 17 77 8.2. Informative References . . . . . . . . . . . . . . . . . 19 78 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 19 80 1. Introduction 82 1.1. Context and Motivation 84 Distributed Denial of Service (DDoS) attacks continue to plague 85 networks around the globe, from Tier-1 service providers on down to 86 enterprises and small businesses. Attack scale and frequency 87 similarly have continued to increase, in part as a result of software 88 vulnerabilities leading to reflection and amplification attacks. 89 Once-staggering attack traffic volume is now the norm, and the impact 90 of larger-scale attacks attract the attention of international press 91 agencies. 93 The greater impact of contemporary DDoS attacks has led to increased 94 focus on coordinated attack response. Many institutions and 95 enterprises lack the resources or expertise to operate on-premise 96 attack mitigation solutions themselves, or simply find themselves 97 constrained by local bandwidth limitations. To address such gaps, 98 security service providers have begun to offer on-demand traffic 99 scrubbing services, which aim to separate the DDoS traffic from 100 legitimate traffic and forward only the latter. Today each such 101 service offers its own interface for subscribers to request attack 102 mitigation, tying subscribers to proprietary implementations while 103 also limiting the subset of network elements capable of participating 104 in the attack response. As a result of incompatibility across 105 services, attack responses may be fragmentary or otherwise 106 incomplete, leaving key players in the attack path unable to assist 107 in the defense. 109 The lack of a common method to coordinate a real-time response among 110 involved actors and network domains inhibits the speed and 111 effectiveness of DDoS attack mitigation. This document describes the 112 required characteristics of a DOTS protocol enabling requests for 113 DDoS attack mitigation, reducing attack impact and leading to more 114 efficient defensive strategies. 116 DOTS communicates the need for defensive action in anticipation of or 117 in response to an attack, but does not dictate the form any defensive 118 action takes. DOTS supplements calls for help with pertinent details 119 about the detected attack, allowing entities participating in DOTS to 120 form ad hoc, adaptive alliances against DDoS attacks as described in 121 the DOTS use cases [I-D.ietf-dots-use-cases]. The requirements in 122 this document are derived from those use cases and 123 [I-D.ietf-dots-architecture]. 125 1.2. Terminology 127 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 128 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 129 document are to be interpreted as described in [RFC2119]. 131 This document adopts the following terms: 133 DDoS: A distributed denial-of-service attack, in which traffic 134 originating from multiple sources are directed at a target on a 135 network. DDoS attacks are intended to cause a negative impact on 136 the availability of servers, services, applications, and/or other 137 functionality of an attack target. Denial-of-service 138 considerations are discussed in detail in [RFC4732]. 140 DDoS attack target: A network connected entity with a finite set of 141 resources, such as network bandwidth, memory or CPU, that is the 142 focus of a DDoS attack. Potential targets include network 143 elements, network links, servers, and services. 145 DDoS attack telemetry: Collected measurements and behavioral 146 characteristics defining the nature of a DDoS attack. 148 Countermeasure: An action or set of actions taken to recognize and 149 filter out DDoS attack traffic while passing legitimate traffic to 150 the attack target. 152 Mitigation: A set of countermeasures enforced against traffic 153 destined for the target or targets of a detected or reported DDoS 154 attack, where countermeasure enforcement is managed by an entity 155 in the network path between attack sources and the attack target. 156 Mitigation methodology is out of scope for this document. 158 Mitigator: An entity, typically a network element, capable of 159 performing mitigation of a detected or reported DDoS attack. For 160 the purposes of this document, this entity is a black box capable 161 of mitigation, making no assumptions about availability or design 162 of countermeasures, nor about the programmable interface between 163 this entity and other network elements. The mitigator and DOTS 164 server are assumed to belong to the same administrative entity. 166 DOTS client: A DOTS-aware software module responsible for requesting 167 attack response coordination with other DOTS-aware elements. 169 DOTS server: A DOTS-aware software module handling and responding to 170 messages from DOTS clients. The DOTS server SHOULD enable 171 mitigation on behalf of the DOTS client, if requested, by 172 communicating the DOTS client's request to the mitigator and 173 returning selected mitigator feedback to the requesting DOTS 174 client. A DOTS server MAY also be a mitigator. 176 DOTS agent: Any DOTS-aware software module capable of participating 177 in a DOTS signaling session. 179 DOTS gateway: A logical DOTS agent resulting from the logical 180 concatenation of a DOTS server and a DOTS client, analogous to a 181 SIP Back-to-Back User Agent (B2BUA) [RFC3261]. DOTS gateways are 182 discussed in detail in [I-D.ietf-dots-architecture]. 184 Signal channel: A bidirectional, mutually authenticated 185 communication channel between DOTS agents characterized by 186 resilience even in conditions leading to severe packet loss, such 187 as a volumetric DDoS attack causing network congestion. 189 DOTS signal: A concise authenticated status/control message 190 transmitted between DOTS agents, used to indicate client's need 191 for mitigation, as well as to convey the status of any requested 192 mitigation. 194 Heartbeat: A message transmitted between DOTS agents over the signal 195 channel, used as a keep-alive and to measure peer health. 197 Client signal: A message sent from a DOTS client to a DOTS server 198 over the signal channel, indicating the DOTS client's need for 199 mitigation, as well as the scope of any requested mitigation, 200 optionally including additional attack details to supplement 201 server-initiated mitigation. 203 Server signal: A message sent from a DOTS server to a DOTS client 204 over the signal channel. Note that a server signal is not a 205 response to client signal, but a DOTS server-initiated status 206 message sent to DOTS clients with which the server has established 207 signaling sessions. 209 Data channel: A secure communication layer between DOTS clients and 210 DOTS servers used for infrequent bulk exchange of data not easily 211 or appropriately communicated through the signal channel under 212 attack conditions. 214 Filter: A policy matching a network traffic flow or set of flows and 215 rate-limiting or discarding matching traffic. 217 Blacklist: A filter list of addresses, prefixes and/or other 218 identifiers indicating sources from which traffic should be 219 blocked, regardless of traffic content. 221 Whitelist: A list of addresses, prefixes and/or other identifiers 222 from indicating sources from which traffic should always be 223 allowed, regardless of contradictory data gleaned in a detected 224 attack. 226 Multi-homed DOTS client: A DOTS client exchanging messages with 227 multiple DOTS servers, each in a separate administrative domain. 229 2. Requirements 231 This section describes the required features and characteristics of 232 the DOTS protocol. 234 DOTS is an advisory protocol. An active DDoS attack against the 235 entity controlling the DOTS client need not be present before 236 establishing DOTS communication between DOTS agents. Indeed, 237 establishing a relationship with peer DOTS agents during normal 238 network conditions provides the foundation for more rapid attack 239 response against future attacks, as all interactions setting up DOTS, 240 including any business or service level agreements, are already 241 complete. 243 DOTS must at a minimum make it possible for a DOTS client to request 244 a DOTS server's aid in mounting a coordinated defense against a 245 suspected attack, signaling within or between domains as requested by 246 local operators. DOTS clients should similarly be able to withdraw 247 aid requests. DOTS requires no justification from DOTS clients for 248 requests for help, nor do DOTS clients need to justify withdrawing 249 help requests: the decision is local to the DOTS clients' domain. 250 Regular feedback between DOTS clients and DOTS server supplement the 251 defensive alliance by maintaining a common understanding of DOTS peer 252 health and activity. Bidirectional communication between DOTS 253 clients and DOTS servers is therefore critical. 255 Yet DOTS must also work with a set of competing operational goals. 256 On the one hand, the protocol must be resilient under extremely 257 hostile network conditions, providing continued contact between DOTS 258 agents even as attack traffic saturates the link. Such resiliency 259 may be developed several ways, but characteristics such as small 260 message size, asynchronous, redundant message delivery and minimal 261 connection overhead (when possible given local network policy) will 262 tend to contribute to the robustness demanded by a viable DOTS 263 protocol. Operators of peer DOTS-enabled domains may enable quality- 264 or class-of-service traffic tagging to increase the probability of 265 successful DOTS signal delivery, but DOTS requires no such policies 266 be in place. The DOTS solution indeed must be viable especially in 267 their absence. 269 On the other hand, DOTS must include protections ensuring message 270 confidentiality, integrity and authenticity to keep the protocol from 271 becoming another vector for the very attacks it's meant to help fight 272 off. DOTS clients must be able to authenticate DOTS servers, and 273 vice versa, for DOTS to operate safely, meaning the DOTS agents must 274 have a way to negotiate and agree upon the terms of protocol 275 security. Attacks against the transport protocol should not offer a 276 means of attack against the message confidentiality, integrity and 277 authenticity. 279 The DOTS server and client must also have some common method of 280 defining the scope of any mitigation performed by the mitigator, as 281 well as making adjustments to other commonly configurable features, 282 such as listen ports, exchanging black- and white-lists, and so on. 284 Finally, DOTS should provide sufficient extensibility to meet local, 285 vendor or future needs in coordinated attack defense, although this 286 consideration is necessarily superseded by the other operational 287 requirements. 289 2.1. General Requirements 291 GEN-001 Extensibility: Protocols and data models developed as part 292 of DOTS MUST be extensible in order to keep DOTS adaptable to 293 operational and proprietary DDoS defenses. Future extensions MUST 294 be backward compatible. 296 GEN-002 Resilience and Robustness: The signaling protocol MUST be 297 designed to maximize the probability of signal delivery even under 298 the severely constrained network conditions imposed by particular 299 attack traffic. The protocol MUST be resilient, that is, continue 300 operating despite message loss and out-of-order or redundant 301 message delivery. In support signaling protocol robustness, DOTS 302 signals SHOULD be conveyed over a transport not susceptible to 303 Head of Line Blocking. 305 GEN-003 Bidirectionality: To support peer health detection, to 306 maintain an open signal channel, and to increase the probability 307 of signal delivery during attack, the signal channel MUST be 308 bidirectional, with client and server transmitting signals to each 309 other at regular intervals, regardless of any client request for 310 mitigation. Unidirectional messages MUST be supported within the 311 bidirectional signal channel to allow for unsolicited message 312 delivery, enabling asynchronous notifications between agents. 314 GEN-004 Sub-MTU Message Size: To avoid message fragmentation and the 315 consequently decreased probability of message delivery, signaling 316 protocol message size MUST be kept under signaling Path Maximum 317 Transmission Unit (PMTU), including the byte overhead of any 318 encapsulation, transport headers, and transport- or message-level 319 security. 321 DOTS agents SHOULD attempt to learn the PMTU through mechanisms 322 such as Path MTU Discovery [RFC1191] or Packetization Layer Path 323 MTU Discovery [RFC4821]. If the PMTU cannot be discovered, DOTS 324 agents SHOULD assume a PMTU of 1280 bytes. If IPv4 support on 325 legacy or otherwise unusual networks is a consideration and PMTU 326 is unknown, DOTS implementations MAY rely on a PMTU of 576 bytes, 327 as discussed in [RFC0791] and [RFC1122]. 329 GEN-005 Bulk Data Exchange: Infrequent bulk data exchange between 330 DOTS agents can also significantly augment attack response 331 coordination, permitting such tasks as population of black- or 332 white-listed source addresses; address or prefix group aliasing; 333 exchange of incident reports; and other hinting or configuration 334 supplementing attack response. 336 As the resilience requirements for the DOTS signal channel mandate 337 small signal message size, a separate, secure data channel 338 utilizing a reliable transport protocol MUST be used for bulk data 339 exchange. 341 2.2. Operational Requirements 343 OP-001 Use of Common Transport Protocols: DOTS MUST operate over 344 common widely deployed and standardized transport protocols. 345 While the User Datagram Protocol (UDP) [RFC0768] SHOULD be used 346 for the signal channel, the Transmission Control Protocol (TCP) 347 [RFC0793] MAY be used if necessary due to network policy or 348 middlebox capabilities or configurations. The data channel MUST 349 use a reliable transport; see Section 2.3 below. 351 OP-002 Session Health Monitoring: Peer DOTS agents MUST regularly 352 send heartbeats to each other after mutual authentication in order 353 to keep the DOTS session active. A session MUST be considered 354 active until a DOTS agent explicitly ends the session, or either 355 DOTS agent fails to receive heartbeats from the other after a 356 mutually agreed upon timeout period has elapsed. 358 OP-003 Session Redirection: In order to increase DOTS operational 359 flexibility and scalability, DOTS servers SHOULD be able to 360 redirect DOTS clients to another DOTS server at any time. DOTS 361 clients MUST NOT assume the redirection target DOTS server shares 362 security state with the redirecting DOTS server. DOTS clients MAY 363 attempt abbreviated security negotiation methods supported by the 364 protocol, such as DTLS session resumption, but MUST be prepared to 365 negotiate new security state with the redirection target DOTS 366 server. 368 Due to the increased likelihood of packet loss caused by link 369 congestion during an attack, it is RECOMMENDED DOTS servers avoid 370 redirecting while mitigation is enabled during an active attack 371 against a target in the DOTS client's domain. 373 OP-004 Mitigation Requests and Status: Authorized DOTS clients MUST 374 be able to request scoped mitigation from DOTS servers. DOTS 375 servers MUST send mitigation request status in response to DOTS 376 clients requests for mitigation, and SHOULD accept scoped 377 mitigation requests from authorized DOTS clients. DOTS servers 378 MAY reject authorized requests for mitigation, but MUST include a 379 reason for the rejection in the status message sent to the client. 381 Due to the higher likelihood of packet loss during a DDoS attack, 382 DOTS servers SHOULD regularly send mitigation status to authorized 383 DOTS clients which have requested and been granted mitigation, 384 regardless of client requests for mitigation status. 386 When DOTS client-requested mitigation is active, DOTS server 387 status messages SHOULD include the following mitigation metrics: 389 * Total number of packets blocked by the mitigation 391 * Current number of packets per second blocked 393 * Total number of bytes blocked 395 * Current number of bytes per second blocked 397 DOTS clients SHOULD take these metrics into account when 398 determining whether to ask the DOTS server to cease mitigation. 400 Once a DOTS client requests mitigation, the client MAY withdraw 401 that request at any time, regardless of whether mitigation is 402 currently active. The DOTS server MUST immediately acknowledge a 403 DOTS client's request to stop mitigation. 405 To protect against route or DNS flapping caused by a client 406 rapidly toggling mitigation, and to dampen the effect of 407 oscillating attacks, DOTS servers MAY continue mitigation for a 408 period of up to five minutes after acknowledging a DOTS client's 409 withdrawal of a mitigation request. During this period, DOTS 410 server status messages SHOULD indicate that mitigation is active 411 but terminating. After the five-minute period elapses, the DOTS 412 server MUST treat the mitigation as terminated, as the DOTS client 413 is no longer responsible for the mitigation. For example, if 414 there is a financial relationship between the DOTS client and 415 server domains, the DOTS client ceases incurring cost at this 416 point. 418 OP-005 Mitigation Lifetime: DOTS servers MUST support mitigation 419 lifetimes, and MUST terminate a mitigation when the lifetime 420 elapses. DOTS servers also MUST support renewal of mitigation 421 lifetimes in mitigation requests from DOTS clients, allowing 422 clients to extend mitigation as necessary for the duration of an 423 attack. 425 DOTS servers MUST treat a mitigation terminated due to lifetime 426 expiration exactly as if the DOTS client originating the 427 mitigation had asked to end the mitigation, including the five- 428 minute termination period, as described above in OP-004. 430 DOTS clients SHOULD include a mitigation lifetime in all 431 mitigation requests. If a DOTS client does not include a 432 mitigation lifetime in requests for help sent to the DOTS server, 433 the DOTS server will use a reasonable default as defined by the 434 protocol. 436 DOTS servers SHOULD support indefinite mitigation lifetimes, 437 enabling architectures in which the mitigator is always in the 438 traffic path to the resources for which the DOTS client is 439 requesting protection. DOTS servers MAY refuse mitigations with 440 indefinite lifetimes, for policy reasons. The reasons themselves 441 are out of scope for this document, but MUST be included in the 442 mitigation rejection message from the server, per OP-004. 444 OP-006 Mitigation Scope: DOTS clients MUST indicate desired 445 mitigation scope. The scope type will vary depending on the 446 resources requiring mitigation. All DOTS agent implementations 447 MUST support the following required scope types: 449 * IPv4 addresses in dotted quad format 451 * IPv4 address prefixes in CIDR notation [RFC4632] 453 * IPv6 addresses [RFC2373] 455 * IPv6 address prefixes [RFC2373] 457 * Domain names [RFC1035] 459 The following mitigation scope types are OPTIONAL: 461 * Uniform Resource Identifiers [RFC3986] 463 DOTS agents MUST support mitigation scope aliases, allowing DOTS 464 client and server to refer to collections of protected resources 465 by an opaque identifier created through the data channel, direct 466 configuration, or other means. 468 If there is additional information available narrowing the scope 469 of any requested attack response, such as targeted port range, 470 protocol, or service, DOTS clients SHOULD include that information 471 in client signals. DOTS clients MAY also include additional 472 attack details. Such supplemental information is OPTIONAL, and 473 DOTS servers MAY ignore it when enabling countermeasures on the 474 mitigator. 476 As an active attack evolves, clients MUST be able to adjust as 477 necessary the scope of requested mitigation by refining the scope 478 of resources requiring mitigation. 480 OP-007 Mitigation Efficacy: When a mitigation request by a DOTS 481 client is active, DOTS clients SHOULD transmit a metric of 482 perceived mitigation efficacy to the DOTS server, per "Automatic 483 or Operator-Assisted CPE or PE Mitigators Request Upstream DDoS 484 Mitigation Services" in [I-D.ietf-dots-use-cases]. DOTS servers 485 MAY use the efficacy metric to adjust countermeasures activated on 486 a mitigator on behalf of a DOTS client. 488 OP-008 Conflict Detection and Notification: Multiple DOTS clients 489 controlled by a single administrative entity may send conflicting 490 mitigation requests for pool of protected resources , as a result 491 of misconfiguration, operator error, or compromised DOTS clients. 492 DOTS servers attempting to honor conflicting requests may flap 493 network route or DNS information, degrading the networks 494 attempting to participate in attack response with the DOTS 495 clients. DOTS servers SHALL detect such conflicting requests, and 496 SHALL notify the DOTS clients in conflict. The notification 497 SHOULD indicate the nature and scope of the conflict, for example, 498 the overlapping prefix range in a conflicting mitigation request. 500 OP-009: Network Address Translator Traversal: The DOTS protocol MUST 501 operate over networks in which Network Address Translation (NAT) 502 is deployed. As UDP is the recommended transport for the DOTS 503 signal channel, all considerations in "Middlebox Traversal 504 Guidelines" in [RFC5405] apply to DOTS. Regardless of transport, 505 DOTS protocols MUST follow established best common practices 506 (BCPs) for NAT traversal. 508 2.3. Data Channel Requirements 510 The data channel is intended to be used for bulk data exchanges 511 between DOTS agents. Unlike the signal channel, which must operate 512 nominally even when confronted with signal degradation due to packet 513 loss, the data channel is not expected to be constructed to deal with 514 attack conditions. As the primary function of the data channel is 515 data exchange, a reliable transport is required in order for DOTS 516 agents to detect data delivery success or failure. 518 The data channel must be extensible. We anticipate the data channel 519 will be used for such purposes as configuration or resource 520 discovery. For example, a DOTS client may submit to the DOTS server 521 a collection of prefixes it wants to refer to by alias when 522 requesting mitigation, to which the server would respond with a 523 success status and the new prefix group alias, or an error status and 524 message in the event the DOTS client's data channel request failed. 525 The transactional nature of such data exchanges suggests a separate 526 set of requirements for the data channel, while the potentially 527 sensitive content sent between DOTS agents requires extra precautions 528 to ensure data privacy and authenticity. 530 DATA-001 Reliable transport: Messages sent over the data channel 531 MUST be delivered reliably, in order sent. 533 DATA-002 Data privacy and integrity: Transmissions over the data 534 channel are likely to contain operationally or privacy-sensitive 535 information or instructions from the remote DOTS agent. Theft or 536 modification of data channel transmissions could lead to 537 information leaks or malicious transactions on behalf of the 538 sending agent (see Section 4 below). Consequently data sent over 539 the data channel MUST be encrypted and authenticated using current 540 industry best practices. DOTS servers MUST enable means to 541 prevent leaking operationally or privacy-sensitive data. Although 542 administrative entities participating in DOTS may detail what data 543 may be revealed to third-party DOTS agents, such considerations 544 are not in scope for this document. 546 DATA-003 Resource Configuration: To help meet the general and 547 operational requirements in this document, DOTS server 548 implementations MUST provide an interface to configure resource 549 identifiers, as described in OP-007. DOTS server implementations 550 MAY expose additional configurability. Additional configurability 551 is implementation-specific. 553 DATA-004 Black- and whitelist management: DOTS servers SHOULD 554 provide methods for DOTS clients to manage black- and white-lists 555 of traffic destined for resources belonging to a client. 557 For example, a DOTS client should be able to create a black- or 558 whitelist entry; retrieve a list of current entries from either 559 list; update the content of either list; and delete entries as 560 necessary. 562 How the DOTS server determines client ownership of address space 563 is not in scope. 565 2.4. Security requirements 567 DOTS must operate within a particularly strict security context, as 568 an insufficiently protected signal or data channel may be subject to 569 abuse, enabling or supplementing the very attacks DOTS purports to 570 mitigate. 572 SEC-001 Peer Mutual Authentication: DOTS agents MUST authenticate 573 each other before a DOTS session is considered valid. The method 574 of authentication is not specified, but should follow current 575 industry best practices with respect to any cryptographic 576 mechanisms to authenticate the remote peer. 578 SEC-002 Message Confidentiality, Integrity and Authenticity: DOTS 579 protocols MUST take steps to protect the confidentiality, 580 integrity and authenticity of messages sent between client and 581 server. While specific transport- and message-level security 582 options are not specified, the protocols MUST follow current 583 industry best practices for encryption and message authentication. 585 In order for DOTS protocols to remain secure despite advancements 586 in cryptanalysis and traffic analysis, DOTS agents MUST be able to 587 negotiate the terms and mechanisms of protocol security, subject 588 to the interoperability and signal message size requirements 589 above. 591 While the interfaces between downstream DOTS server and upstream 592 DOTS client within a DOTS gateway are implementation-specific, 593 those interfaces nevertheless MUST provide security equivalent to 594 that of the signaling sessions bridged by gateways in the 595 signaling path. For example, when a DOTS gateway consisting of a 596 DOTS server and DOTS client is running on the same logical device, 597 they must be within the same process security boundary. 599 SEC-003 Message Replay Protection: In order to prevent a passive 600 attacker from capturing and replaying old messages, DOTS protocols 601 MUST provide a method for replay detection. 603 2.5. Data Model Requirements 605 The value of DOTS is in standardizing a mechanism to permit elements, 606 networks or domains under or under threat of DDoS attack to request 607 aid mitigating the effects of any such attack. A well-structured 608 DOTS data model is therefore critical to the development of a 609 successful DOTS protocol. 611 DM-001: Structure: The data model structure for the DOTS protocol 612 may be described by a single module, or be divided into related 613 collections of hierarchical modules and sub-modules. If the data 614 model structure is split across modules, those distinct modules 615 MUST allow references to describe the overall data model's 616 structural dependencies. 618 DM-002: Versioning: To ensure interoperability between DOTS protocol 619 implementations, data models MUST be versioned. The version 620 number of the initial data model SHALL be 1. Each published 621 change to the initial published DOTS data model SHALL increment 622 the data model version by 1. 624 How the protocol represents data model versions is not defined in 625 this document. 627 DM-003: Mitigation Status Representation: The data model MUST 628 provide the ability to represent a request for mitigation and the 629 withdrawal of such a request. The data model MUST also support a 630 representation of currently requested mitigation status, including 631 failures and their causes. 633 DM-004: Mitigation Scope Representation: The data model MUST support 634 representation of a requested mitigation's scope. As mitigation 635 scope may be represented in several different ways, per OP-006 636 above, the data model MUST be capable of flexible representation 637 of mitigation scope. 639 DM-005: Mitigation Lifetime Representation: The data model MUST 640 support representation of a mitigation request's lifetime, 641 including mitigations with no specified end time. 643 DM-006: Mitigation Efficacy Representation: The data model MUST 644 support representation of a DOTS client's understanding of the 645 efficacy of a mitigation enabled through a mitigation request. 647 DM-007: Acceptable Signal Loss Representation: The data model MUST 648 be able to represent the DOTS agent's preference for acceptable 649 signal loss when establishing a signaling session, as described in 650 GEN-002. 652 DM-008: Heartbeat Interval Representation: The data model MUST be 653 able to represent the DOTS agent's preferred heartbeat interval, 654 which the client may include when establishing the signal channel, 655 as described in OP-002. 657 DM-009: Relationship to Transport: The DOTS data model MUST NOT 658 depend on the specifics of any transport to represent fields in 659 the model. 661 3. Congestion Control Considerations 663 3.1. Signal Channel 665 As part of a protocol expected to operate over links affected by DDoS 666 attack traffic, the DOTS signal channel MUST NOT contribute 667 significantly to link congestion. To meet the operational 668 requirements above, DOTS signal channel implementations MUST support 669 UDP. However, UDP when deployed naively can be a source of network 670 congestion, as discussed in [RFC5405]. Signal channel 671 implementations using UDP MUST therefore include a congestion control 672 mechanism. 674 Signal channel implementations using TCP may rely on built-in TCP 675 congestion control support. 677 3.2. Data Channel 679 As specified in DATA-001, the data channel requires reliable, in- 680 order message delivery. Data channel implementations using TCP may 681 rely on the TCP implementation's built-in congestion control 682 mechanisms. 684 4. Security Considerations 686 DOTS is at risk from three primary attacks: 688 o DOTS agent impersonation 690 o Traffic injection 692 o Signaling blocking 694 The DOTS protocol MUST be designed for minimal data transfer to 695 address the blocking risk. Impersonation and traffic injection 696 mitigation can be managed through current secure communications best 697 practices. See Section 2.4 above for a detailed discussion. 699 5. Contributors 701 Mohamed Boucadair 702 Orange 704 mohamed.boucadair@orange.com 706 Flemming Andreasen 707 Cisco Systems, Inc. 709 fandreas@cisco.com 711 Dave Dolson 712 Sandvine 714 ddolson@sandvine.com 716 6. Acknowledgments 718 Thanks to Roman Danyliw and Matt Richardson for careful reading and 719 feedback. 721 7. Change Log 723 7.1. 04 revision 725 2017-03-13 727 o Establish required and optional mitigation scope types 729 o Specify message size for DOTS signal channel 731 o Recast mitigation lifetime as a DOTS server requirement 733 o Clarify DOTS server's responsibilities after client request to end 734 mitigation 736 o Specify security state handling on redirection 738 o Signal channel should use transport not susceptible to HOL 739 blocking 741 o Expanded list of DDoS types to include network links 743 7.2. 03 revision 745 2016-10-30 747 o Extended SEC-003 to require secure interfaces within DOTS 748 gateways. 750 o Changed DATA-003 to Resource Configuration, delegating control of 751 acceptable signal loss, heartbeat intervals, and mitigation 752 lifetime to DOTS client. 754 o Added data model requirements reflecting client control over the 755 above. 757 7.3. 02 revision 759 7.4. 01 revision 761 2016-03-21 762 o Reconciled terminology with -00 revision of 763 [I-D.ietf-dots-use-cases]. 765 o Terminology clarification based on working group feedback. 767 o Moved security-related requirements to separate section. 769 o Made resilience/robustness primary general requirement to align 770 with charter. 772 o Clarified support for unidirectional communication within the 773 bidirectional signal channel. 775 o Added proposed operational requirement to support session 776 redirection. 778 o Added proposed operational requirement to support conflict 779 notification. 781 o Added proposed operational requirement to support mitigation 782 lifetime in mitigation requests. 784 o Added proposed operational requirement to support mitigation 785 efficacy reporting from DOTS clients. 787 o Added proposed operational requirement to cache lookups of all 788 kinds. 790 o Added proposed operational requirement regarding NAT traversal. 792 o Removed redundant mutual authentication requirement from data 793 channel requirements. 795 7.5. 00 revision 797 2015-10-15 799 7.6. Initial revision 801 2015-09-24 Andrew Mortensen 803 8. References 805 8.1. Normative References 807 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 808 DOI 10.17487/RFC0768, August 1980, 809 . 811 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 812 DOI 10.17487/RFC0791, September 1981, 813 . 815 [RFC0793] Postel, J., "Transmission Control Protocol", STD 7, 816 RFC 793, DOI 10.17487/RFC0793, September 1981, 817 . 819 [RFC1035] Mockapetris, P., "Domain names - implementation and 820 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 821 November 1987, . 823 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 824 Communication Layers", STD 3, RFC 1122, 825 DOI 10.17487/RFC1122, October 1989, 826 . 828 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 829 DOI 10.17487/RFC1191, November 1990, 830 . 832 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 833 Requirement Levels", BCP 14, RFC 2119, 834 DOI 10.17487/RFC2119, March 1997, 835 . 837 [RFC2373] Hinden, R. and S. Deering, "IP Version 6 Addressing 838 Architecture", RFC 2373, DOI 10.17487/RFC2373, July 1998, 839 . 841 [RFC3986] Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform 842 Resource Identifier (URI): Generic Syntax", STD 66, 843 RFC 3986, DOI 10.17487/RFC3986, January 2005, 844 . 846 [RFC4632] Fuller, V. and T. Li, "Classless Inter-domain Routing 847 (CIDR): The Internet Address Assignment and Aggregation 848 Plan", BCP 122, RFC 4632, DOI 10.17487/RFC4632, August 849 2006, . 851 [RFC4821] Mathis, M. and J. Heffner, "Packetization Layer Path MTU 852 Discovery", RFC 4821, DOI 10.17487/RFC4821, March 2007, 853 . 855 [RFC5405] Eggert, L. and G. Fairhurst, "Unicast UDP Usage Guidelines 856 for Application Designers", RFC 5405, 857 DOI 10.17487/RFC5405, November 2008, 858 . 860 8.2. Informative References 862 [I-D.ietf-dots-architecture] 863 Mortensen, A., Andreasen, F., Reddy, T., 864 christopher_gray3@cable.comcast.com, c., Compton, R., and 865 N. Teague, "Distributed-Denial-of-Service Open Threat 866 Signaling (DOTS) Architecture", draft-ietf-dots- 867 architecture-01 (work in progress), October 2016. 869 [I-D.ietf-dots-use-cases] 870 Dobbins, R., Fouant, S., Migault, D., Moskowitz, R., 871 Teague, N., Xia, L., and K. Nishizuka, "Use cases for DDoS 872 Open Threat Signaling", draft-ietf-dots-use-cases-03 (work 873 in progress), November 2016. 875 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 876 A., Peterson, J., Sparks, R., Handley, M., and E. 877 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 878 DOI 10.17487/RFC3261, June 2002, 879 . 881 [RFC4732] Handley, M., Ed., Rescorla, E., Ed., and IAB, "Internet 882 Denial-of-Service Considerations", RFC 4732, 883 DOI 10.17487/RFC4732, December 2006, 884 . 886 Authors' Addresses 888 Andrew Mortensen 889 Arbor Networks, Inc. 890 2727 S. State St 891 Ann Arbor, MI 48104 892 United States 894 Email: amortensen@arbor.net 896 Robert Moskowitz 897 HTT Consulting 898 Oak Park, MI 42837 899 United States 901 Email: rgm@htt-consult.com 902 Tirumaleswar Reddy 903 Cisco Systems, Inc. 904 Cessna Business Park, Varthur Hobli 905 Sarjapur Marathalli Outer Ring Road 906 Bangalore, Karnataka 560103 907 India 909 Email: tireddy@cisco.com