idnits 2.17.1 draft-ietf-dots-architecture-13.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 : ---------------------------------------------------------------------------- ** There are 2 instances of too long lines in the document, the longest one being 2 characters in excess of 72. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == The document seems to lack the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords -- however, there's a paragraph with a matching beginning. Boilerplate error? (The document does seem to have the reference to RFC 2119 which the ID-Checklist requires). -- The document date (April 03, 2019) is 1821 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Outdated reference: A later version (-25) exists of draft-ietf-dots-use-cases-17 == Outdated reference: A later version (-43) exists of draft-ietf-tls-dtls13-31 -- Obsolete informational reference (is this intentional?): RFC 793 (Obsoleted by RFC 9293) -- Obsolete informational reference (is this intentional?): RFC 5246 (Obsoleted by RFC 8446) -- Obsolete informational reference (is this intentional?): RFC 5389 (Obsoleted by RFC 8489) -- Obsolete informational reference (is this intentional?): RFC 6347 (Obsoleted by RFC 9147) Summary: 1 error (**), 0 flaws (~~), 4 warnings (==), 5 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 DOTS A. Mortensen, Ed. 3 Internet-Draft Arbor Networks 4 Intended status: Informational T. Reddy, Ed. 5 Expires: October 5, 2019 McAfee, Inc. 6 F. Andreasen 7 Cisco 8 N. Teague 9 Verisign 10 R. Compton 11 Charter 12 April 03, 2019 14 Distributed-Denial-of-Service Open Threat Signaling (DOTS) Architecture 15 draft-ietf-dots-architecture-13 17 Abstract 19 This document describes an architecture for establishing and 20 maintaining Distributed Denial of Service (DDoS) Open Threat 21 Signaling (DOTS) within and between domains. The document does not 22 specify protocols or protocol extensions, instead focusing on 23 defining architectural relationships, components and concepts used in 24 a DOTS deployment. 26 Status of This Memo 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF). Note that other groups may also distribute 33 working documents as Internet-Drafts. The list of current Internet- 34 Drafts is at https://datatracker.ietf.org/drafts/current/. 36 Internet-Drafts are draft documents valid for a maximum of six months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on October 5, 2019. 43 Copyright Notice 45 Copyright (c) 2019 IETF Trust and the persons identified as the 46 document authors. All rights reserved. 48 This document is subject to BCP 78 and the IETF Trust's Legal 49 Provisions Relating to IETF Documents 50 (https://trustee.ietf.org/license-info) in effect on the date of 51 publication of this document. Please review these documents 52 carefully, as they describe your rights and restrictions with respect 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 Table of Contents 60 1. Context and Motivation . . . . . . . . . . . . . . . . . . . 3 61 1.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . 3 62 1.1.1. Key Words . . . . . . . . . . . . . . . . . . . . . . 3 63 1.1.2. Definition of Terms . . . . . . . . . . . . . . . . . 3 64 1.2. Scope . . . . . . . . . . . . . . . . . . . . . . . . . . 3 65 1.3. Assumptions . . . . . . . . . . . . . . . . . . . . . . . 4 66 2. DOTS Architecture . . . . . . . . . . . . . . . . . . . . . . 5 67 2.1. DOTS Operations . . . . . . . . . . . . . . . . . . . . . 8 68 2.2. Components . . . . . . . . . . . . . . . . . . . . . . . 9 69 2.2.1. DOTS Client . . . . . . . . . . . . . . . . . . . . . 9 70 2.2.2. DOTS Server . . . . . . . . . . . . . . . . . . . . . 10 71 2.2.3. DOTS Gateway . . . . . . . . . . . . . . . . . . . . 11 72 2.3. DOTS Agent Relationships . . . . . . . . . . . . . . . . 12 73 2.3.1. Gatewayed Signaling . . . . . . . . . . . . . . . . . 14 74 3. Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . 16 75 3.1. DOTS Sessions . . . . . . . . . . . . . . . . . . . . . . 16 76 3.1.1. Preconditions . . . . . . . . . . . . . . . . . . . . 16 77 3.1.2. Establishing the DOTS Session . . . . . . . . . . . . 17 78 3.1.3. Maintaining the DOTS Session . . . . . . . . . . . . 18 79 3.2. Modes of Signaling . . . . . . . . . . . . . . . . . . . 18 80 3.2.1. Direct Signaling . . . . . . . . . . . . . . . . . . 18 81 3.2.2. Redirected Signaling . . . . . . . . . . . . . . . . 18 82 3.2.3. Recursive Signaling . . . . . . . . . . . . . . . . . 20 83 3.2.4. Anycast Signaling . . . . . . . . . . . . . . . . . . 22 84 3.2.5. Signaling Considerations for Network Address 85 Translation . . . . . . . . . . . . . . . . . . . . . 23 86 3.3. Triggering Requests for Mitigation . . . . . . . . . . . 26 87 3.3.1. Manual Mitigation Request . . . . . . . . . . . . . . 26 88 3.3.2. Automated Conditional Mitigation Request . . . . . . 27 89 3.3.3. Automated Mitigation on Loss of Signal . . . . . . . 28 90 4. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 29 91 5. Security Considerations . . . . . . . . . . . . . . . . . . . 29 92 6. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 30 93 7. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 30 94 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 30 95 8.1. Normative References . . . . . . . . . . . . . . . . . . 30 96 8.2. Informative References . . . . . . . . . . . . . . . . . 30 97 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 33 99 1. Context and Motivation 101 Signaling the need for help defending against an active distributed 102 denial of service (DDoS) attack requires a common understanding of 103 mechanisms and roles among the parties coordinating defensive 104 response. The signaling layer and supplementary messaging is the 105 focus of DDoS Open Threat Signaling (DOTS). DOTS defines a method of 106 coordinating defensive measures among willing peers to mitigate 107 attacks quickly and efficiently, enabling hybrid attack responses 108 coordinated locally at or near the target of an active attack, or 109 anywhere in-path between attack sources and target. Sample DOTS use 110 cases are elaborated in [I-D.ietf-dots-use-cases]. 112 This document describes an architecture used in establishing, 113 maintaining or terminating a DOTS relationship within a domain or 114 between domains. 116 1.1. Terminology 118 1.1.1. Key Words 120 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 121 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 122 document are to be interpreted as described in BCP14 [RFC2119] 123 [RFC8174], when, and only when, they appear in all capitals. 125 1.1.2. Definition of Terms 127 This document uses the terms defined in [I-D.ietf-dots-requirements]. 129 1.2. Scope 131 In this architecture, DOTS clients and servers communicate using DOTS 132 signaling. As a result of signals from a DOTS client, the DOTS 133 server may modify the forwarding path of traffic destined for the 134 attack target(s), for example by diverting traffic to a mitigator or 135 pool of mitigators, where policy may be applied to distinguish and 136 discard attack traffic. Any such policy is deployment-specific. 138 The DOTS architecture presented here is applicable across network 139 administrative domains - for example, between an enterprise domain 140 and the domain of a third-party attack mitigation service - as well 141 as to a single administrative domain. DOTS is generally assumed to 142 be most effective when aiding coordination of attack response between 143 two or more participating networks, but single domain scenarios are 144 valuable in their own right, as when aggregating intra-domain DOTS 145 client signals for inter-domain coordinated attack response. 147 This document does not address any administrative or business 148 agreements that may be established between involved DOTS parties. 149 Those considerations are out of scope. Regardless, this document 150 assumes necessary authentication and authorization mechanisms are put 151 in place so that only authorized clients can invoke the DOTS service. 153 A detailed set of DOTS requirements are discussed in 154 [I-D.ietf-dots-requirements], and the DOTS architecture is designed 155 to follow those requirements. Only new behavioral requirements are 156 described in this document. 158 1.3. Assumptions 160 This document makes the following assumptions: 162 o All domains in which DOTS is deployed are assumed to offer the 163 required connectivity between DOTS agents and any intermediary 164 network elements, but the architecture imposes no additional 165 limitations on the form of connectivity. 167 o Congestion and resource exhaustion are intended outcomes of a DDoS 168 attack [RFC4732]. Some operators may utilize non-impacted paths 169 or networks for DOTS, but in general conditions should be assumed 170 to be hostile and DOTS must be able to function in all 171 circumstances, including when the signaling path is significantly 172 impaired. 174 o There is no universal DDoS attack scale threshold triggering a 175 coordinated response across administrative domains. A network 176 domain administrator, or service or application owner may 177 arbitrarily set attack scale threshold triggers, or manually send 178 requests for mitigation. 180 o Mitigation requests may be sent to one or more upstream DOTS 181 servers based on criteria determined by DOTS client administrators 182 and the underlying network configuration. The number of DOTS 183 servers with which a given DOTS client has established 184 communications is determined by local policy and is deployment- 185 specific. For example, a DOTS client of a multi-homed network may 186 support built-in policies to establish DOTS relationships with 187 DOTS servers located upstream of each interconnection link. 189 o The mitigation capacity and/or capability of domains receiving 190 requests for coordinated attack response is opaque to the domains 191 sending the request. The domain receiving the DOTS client signal 192 may or may not have sufficient capacity or capability to filter 193 any or all DDoS attack traffic directed at a target. In either 194 case, the upstream DOTS server may redirect a request to another 195 DOTS server. Redirection may be local to the redirecting DOTS 196 server's domain, or may involve a third-party domain. 198 o DOTS client and server signals, as well as messages sent through 199 the data channel, are sent across any transit networks with the 200 same probability of delivery as any other traffic between the DOTS 201 client domain and the DOTS server domain. Any encapsulation 202 required for successful delivery is left untouched by transit 203 network elements. DOTS server and DOTS client cannot assume any 204 preferential treatment of DOTS signals. Such preferential 205 treatment may be available in some deployments (e.g., intra-domain 206 scenarios), and the DOTS architecture does not preclude its use 207 when available. However, DOTS itself does not address how that 208 may be done. 210 o The architecture allows for, but does not assume, the presence of 211 Quality of Service (QoS) policy agreements between DOTS-enabled 212 peer networks or local QoS prioritization aimed at ensuring 213 delivery of DOTS messages between DOTS agents. QoS is an 214 operational consideration only, not a functional part of the DOTS 215 architecture. 217 o The signal and data channels are loosely coupled, and may not 218 terminate on the same DOTS server. 220 2. DOTS Architecture 222 The basic high-level DOTS architecture is illustrated in Figure 1: 224 +-----------+ +-------------+ 225 | Mitigator | ~~~~~~~~~~ | DOTS Server | 226 +-----------+ +-------------+ 227 | 228 | 229 | 230 +---------------+ +-------------+ 231 | Attack Target | ~~~~~~ | DOTS Client | 232 +---------------+ +-------------+ 234 Figure 1: Basic DOTS Architecture 236 A simple example instantiation of the DOTS architecture could be an 237 enterprise as the attack target for a volumetric DDoS attack, and an 238 upstream DDoS mitigation service as the mitigator. The enterprise 239 (attack target) is connected to the Internet via a link that is 240 getting saturated, and the enterprise suspects it is under DDoS 241 attack. The enterprise has a DOTS client, which obtains information 242 about the DDoS attack, and signals the DOTS server for help in 243 mitigating the attack. The DOTS server in turn invokes one or more 244 mitigators, which are tasked with mitigating the actual DDoS attack, 245 and hence aim to suppress the attack traffic while allowing valid 246 traffic to reach the attack target. 248 The scope of the DOTS specifications is the interfaces between the 249 DOTS client and DOTS server. The interfaces to the attack target and 250 the mitigator are out of scope of DOTS. Similarly, the operation of 251 both the attack target and the mitigator is out of scope of DOTS. 252 Thus, DOTS neither specifies how an attack target decides it is under 253 DDoS attack, nor does DOTS specify how a mitigator may actually 254 mitigate such an attack. A DOTS client's request for mitigation is 255 advisory in nature, and may not lead to any mitigation at all, 256 depending on the DOTS server domain's capacity and willingness to 257 mitigate on behalf of the DOTS client's domain. 259 The DOTS client may be provided with a list of DOTS servers, each 260 associated with one or more IP addresses. These addresses may or may 261 not be of the same address family. The DOTS client establishes one 262 or more sessions by connecting to the provided DOTS server addresses. 264 As illustrated in Figure 2, there are two interfaces between a DOTS 265 server and a DOTS client; a signal channel and (optionally) a data 266 channel. 268 +---------------+ +---------------+ 269 | | <------- Signal Channel ------> | | 270 | DOTS Client | | DOTS Server | 271 | | <======= Data Channel ======> | | 272 +---------------+ +---------------+ 274 Figure 2: DOTS Interfaces 276 The primary purpose of the signal channel is for a DOTS client to ask 277 a DOTS server for help in mitigating an attack, and for the DOTS 278 server to inform the DOTS client about the status of such mitigation. 279 The DOTS client does this by sending a client signal, which contains 280 information about the attack target(s). The client signal may also 281 include telemetry information about the attack, if the DOTS client 282 has such information available. The DOTS server in turn sends a 283 server signal to inform the DOTS client of whether it will honor the 284 mitigation request. Assuming it will, the DOTS server initiates 285 attack mitigation, and periodically informs the DOTS client about the 286 status of the mitigation. Similarly, the DOTS client periodically 287 informs the DOTS server about the client's status, which at a minimum 288 provides client (attack target) health information, but it should 289 also include efficacy information about the attack mitigation as it 290 is now seen by the client. At some point, the DOTS client may decide 291 to terminate the server-side attack mitigation, which it indicates to 292 the DOTS server over the signal channel. A mitigation may also be 293 terminated if a DOTS client-specified mitigation lifetime is 294 exceeded. Note that the signal channel may need to operate over a 295 link that is experiencing a DDoS attack and hence is subject to 296 severe packet loss and high latency. 298 While DOTS is able to request mitigation with just the signal 299 channel, the addition of the DOTS data channel provides for 300 additional and more efficient capabilities. The primary purpose of 301 the data channel is to support DOTS related configuration and policy 302 information exchange between the DOTS client and the DOTS server. 303 Examples of such information include, but are not limited to: 305 o Creating identifiers, such as names or aliases, for resources for 306 which mitigation may be requested. Such identifiers may then be 307 used in subsequent signal channel exchanges to refer more 308 efficiently to the resources under attack, as seen in Figure 3, 309 using JSON to serialize the data: 311 { 312 "https1": [ 313 "192.0.2.1:443", 314 "198.51.100.2:443", 315 ], 316 "proxies": [ 317 "203.0.113.3:3128", 318 "[2001:db8:ac10::1]:3128" 319 ], 320 "api_urls": "https://apiserver.example.com/api/v1", 321 } 323 Figure 3: Protected resource identifiers 325 o Drop-list management, which enables a DOTS client to inform the 326 DOTS server about sources to suppress. 328 o Accept-list management, which enables a DOTS client to inform the 329 DOTS server about sources from which traffic is always accepted. 331 o Filter management, which enables a DOTS client to install or 332 remove traffic filters dropping or rate-limiting unwanted traffic. 334 o DOTS client provisioning. 336 Note that while it is possible to exchange the above information 337 before, during or after a DDoS attack, DOTS requires reliable 338 delivery of this information and does not provide any special means 339 for ensuring timely delivery of it during an attack. In practice, 340 this means that DOTS deployments should not rely on such information 341 being exchanged during a DDoS attack. 343 2.1. DOTS Operations 345 DOTS does not prescribe any specific deployment models, however DOTS 346 is designed with some specific requirements around the different DOTS 347 agents and their relationships. 349 First of all, a DOTS agent belongs to a domain that has an identity 350 which can be authenticated and authorized. DOTS agents communicate 351 with each other over a mutually authenticated signal channel and 352 (optionally) data channel. However, before they can do so, a service 353 relationship needs to be established between them. The details and 354 means by which this is done is outside the scope of DOTS, however an 355 example would be for an enterprise A (DOTS client) to sign up for 356 DDoS service from provider B (DOTS server). This would establish a 357 (service) relationship between the two that enables enterprise A's 358 DOTS client to establish a signal channel with provider B's DOTS 359 server. A and B will authenticate each other, and B can verify that 360 A is authorized for its service. 362 From an operational and design point of view, DOTS assumes that the 363 above relationship is established prior to a request for DDoS attack 364 mitigation. In particular, it is assumed that bi-directional 365 communication is possible at this time between the DOTS client and 366 DOTS server. Furthermore, it is assumed that additional service 367 provisioning, configuration and information exchange can be performed 368 by use of the data channel, if operationally required. It is not 369 until this point that the mitigation service is available for use. 371 Once the mutually authenticated signal channel has been established, 372 it will remain active. This is done to increase the likelihood that 373 the DOTS client can signal the DOTS server for help when the attack 374 target is being flooded, and similarly raise the probability that 375 DOTS server signals reach the client regardless of inbound link 376 congestion. This does not necessarily imply that the attack target 377 and the DOTS client have to be co-located in the same administrative 378 domain, but it is expected to be a common scenario. 380 DDoS mitigation with the help of an upstream mitigator may involve 381 some form of traffic redirection whereby traffic destined for the 382 attack target is steered towards the mitigator. Common mechanisms to 383 achieve this redirection depend on BGP [RFC4271] and DNS [RFC1035]. 385 The mitigator in turn inspects and scrubs the traffic, and forwards 386 the resulting (hopefully non-attack) traffic to the attack target. 387 Thus, when a DOTS server receives an attack mitigation request from a 388 DOTS client, it can be viewed as a way of causing traffic redirection 389 for the attack target indicated. 391 DOTS relies on mutual authentication and the pre-established service 392 relationship between the DOTS client's domain and the DOTS server's 393 domain to provide basic authorization. The DOTS server should 394 enforce additional authorization mechanisms to restrict the 395 mitigation scope a DOTS client can request, but such authorization 396 mechanisms are deployment-specific. 398 Although co-location of DOTS server and mitigator within the same 399 domain is expected to be a common deployment model, it is assumed 400 that operators may require alternative models. Nothing in this 401 document precludes such alternatives. 403 2.2. Components 405 2.2.1. DOTS Client 407 A DOTS client is a DOTS agent from which requests for help 408 coordinating attack response originate. The requests may be in 409 response to an active, ongoing attack against a target in the DOTS 410 client's domain, but no active attack is required for a DOTS client 411 to request help. Operators may wish to have upstream mitigators in 412 the network path for an indefinite period, and are restricted only by 413 business relationships when it comes to duration and scope of 414 requested mitigation. 416 The DOTS client requests attack response coordination from a DOTS 417 server over the signal channel, including in the request the DOTS 418 client's desired mitigation scoping, as described in 419 [I-D.ietf-dots-requirements] (SIG-008). The actual mitigation scope 420 and countermeasures used in response to the attack are up to the DOTS 421 server and mitigator operators, as the DOTS client may have a narrow 422 perspective on the ongoing attack. As such, the DOTS client's 423 request for mitigation should be considered advisory: guarantees of 424 DOTS server availability or mitigation capacity constitute service 425 level agreements and are out of scope for this document. 427 The DOTS client adjusts mitigation scope and provides available 428 mitigation feedback (e.g., mitigation efficacy) at the direction of 429 its local administrator. Such direction may involve manual or 430 automated adjustments in response to updates from the DOTS server. 432 To provide a metric of signal health and distinguish an idle signal 433 channel from a disconnected or defunct session, the DOTS client sends 434 a heartbeat over the signal channel to maintain its half of the 435 channel. The DOTS client similarly expects a heartbeat from the DOTS 436 server, and may consider a session terminated in the extended absence 437 of a DOTS server heartbeat. 439 2.2.2. DOTS Server 441 A DOTS server is a DOTS agent capable of receiving, processing and 442 possibly acting on requests for help coordinating attack response 443 from DOTS clients. The DOTS server authenticates and authorizes DOTS 444 clients as described in Section 3.1, and maintains session state, 445 tracking requests for mitigation, reporting on the status of active 446 mitigations, and terminating sessions in the extended absence of a 447 client heartbeat or when a session times out. 449 Assuming the preconditions discussed below exist, a DOTS client 450 maintaining an active session with a DOTS server may reasonably 451 expect some level of mitigation in response to a request for 452 coordinated attack response. 454 For a given DOTS client (administrative) domain, the DOTS server 455 needs to be able to determine whether a given target resource is in 456 that domain. For example, this could take the form of associating a 457 set of IP addresses and/or prefixes per domain. The DOTS server 458 enforces authorization of DOTS clients' signals for mitigation. The 459 mechanism of enforcement is not in scope for this document, but is 460 expected to restrict requested mitigation scope to addresses, 461 prefixes, and/or services owned by the DOTS client domain, such that 462 a DOTS client from one domain is not able to influence the network 463 path to another domain. A DOTS server MUST reject requests for 464 mitigation of resources not owned by the requesting DOTS client's 465 administrative domain. A DOTS server MAY also refuse a DOTS client's 466 mitigation request for arbitrary reasons, within any limits imposed 467 by business or service level agreements between client and server 468 domains. If a DOTS server refuses a DOTS client's request for 469 mitigation, the DOTS server MUST include the refusal reason in the 470 server signal sent to the client. 472 A DOTS server is in regular contact with one or more mitigators. If 473 a DOTS server accepts a DOTS client's request for help, the DOTS 474 server forwards a translated form of that request to the mitigator(s) 475 responsible for scrubbing attack traffic. Note that the form of the 476 translated request passed from the DOTS server to the mitigator is 477 not in scope: it may be as simple as an alert to mitigator operators, 478 or highly automated using vendor or open application programming 479 interfaces supported by the mitigator. The DOTS server MUST report 480 the actual scope of any mitigation enabled on behalf of a client. 482 The DOTS server SHOULD retrieve available metrics for any mitigations 483 activated on behalf of a DOTS client, and SHOULD include them in 484 server signals sent to the DOTS client originating the request for 485 mitigation. 487 To provide a metric of signal health and distinguish an idle signal 488 channel from a disconnected or defunct channel, the DOTS server MUST 489 send a heartbeat over the signal channel to maintain its half of the 490 channel. The DOTS server similarly expects a heartbeat from the DOTS 491 client, and MAY consider a session terminated in the extended absence 492 of a DOTS client heartbeat. 494 2.2.3. DOTS Gateway 496 Traditional client/server relationships may be expanded by chaining 497 DOTS sessions. This chaining is enabled through "logical 498 concatenation" of a DOTS server and a DOTS client, resulting in an 499 application analogous to the Session Initiation Protocol (SIP) 500 [RFC3261] logical entity of a Back-to-Back User Agent (B2BUA) 501 [RFC7092]. The term DOTS gateway is used here in the descriptions of 502 selected scenarios involving this application. 504 A DOTS gateway may be deployed client-side, server-side or both. The 505 gateway may terminate multiple discrete client connections and may 506 aggregate these into a single or multiple DOTS sessions. 508 The DOTS gateway will appear as a server to its downstream agents and 509 as a client to its upstream agents, a functional concatenation of the 510 DOTS client and server roles, as depicted in Figure 4: 512 +-------------+ 513 | | D | | 514 +----+ | | O | | +----+ 515 | c1 |----------| s1 | T | c2 |---------| s2 | 516 +----+ | | S | | +----+ 517 | | G | | 518 +-------------+ 520 Figure 4: DOTS gateway 522 The DOTS gateway MUST perform full stack DOTS session termination and 523 reorigination between its client and server side. The details of how 524 this is achieved are implementation specific. The DOTS protocol does 525 not include any special features related to DOTS gateways, and hence 526 from a DOTS perspective, whenever a DOTS gateway is present, the DOTS 527 session simply terminates/originates there. 529 2.3. DOTS Agent Relationships 531 So far, we have only considered a relatively simple scenario of a 532 single DOTS client associated with a single DOTS server, however DOTS 533 supports more advanced relationships. 535 A DOTS server may be associated with one or more DOTS clients, and 536 those DOTS clients may belong to different domains. An example 537 scenario is a mitigation provider serving multiple attack targets 538 (Figure 5). 540 DOTS clients DOTS server 541 +---+ 542 | c |----------- 543 +---+ \ 544 c1.example.org \ 545 \ 546 +---+ \ +---+ 547 | c |----------------| S | 548 +---+ / +---+ 549 c1.example.com / dots1.example.net 550 / 551 +---+ / 552 | c |----------- 553 +---+ 554 c2.example.com 556 Figure 5: DOTS server with multiple clients 558 A DOTS client may be associated with one or more DOTS servers, and 559 those DOTS servers may belong to different domains. This may be to 560 ensure high availability or co-ordinate mitigation with more than one 561 directly connected ISP. An example scenario is for an enterprise to 562 have DDoS mitigation service from multiple providers, as shown in 563 Figure 6. 565 DOTS client DOTS servers 566 +---+ 567 -----------| S | 568 / +---+ 569 / dots1.example.net 570 / 571 +---+/ +---+ 572 | c |---------------| S | 573 +---+\ +---+ 574 \ dots.example.org 575 \ 576 \ +---+ 577 -----------| S | 578 +---+ 579 c.example.com dots2.example.net 581 Figure 6: Multi-Homed DOTS Client 583 Deploying a multi-homed client requires extra care and planning, as 584 the DOTS servers with which the multi-homed client communicates may 585 not be affiliated. Should the multi-homed client simultaneously 586 request for mitigation from all servers with which it has established 587 signal channels, the client may unintentionally inflict additional 588 network disruption on the resources it intends to protect. In one of 589 the worst cases, a multi-homed DOTS client could cause a permanent 590 routing loop of traffic destined for the client's protected services, 591 as the uncoordinated DOTS servers' mitigators all try to divert that 592 traffic to their own scrubbing centers. 594 The DOTS protocol itself provides no fool-proof method to prevent 595 such self-inflicted harms as a result deploying multi-homed DOTS 596 clients. If DOTS client implementations nevertheless include support 597 for multi-homing, they are expected to be aware of the risks, and 598 consequently to include measures aimed at reducing the likelihood of 599 negative outcomes. Simple measures might include: 601 o Requesting mitigation serially, ensuring only one mitigation 602 request for a given address space is active at any given time; 604 o Dividing the protected resources among the DOTS servers, such that 605 no two mitigators will be attempting to divert and scrub the same 606 traffic; 608 o Restricting multi-homing to deployments in which all DOTS servers 609 are coordinating management of a shared pool of mitigation 610 resources. 612 2.3.1. Gatewayed Signaling 614 As discussed in Section 2.2.3, a DOTS gateway is a logical function 615 chaining DOTS sessions through concatenation of a DOTS server and 616 DOTS client. 618 An example scenario, as shown in Figure 7 and Figure 8, is for an 619 enterprise to have deployed multiple DOTS capable devices which are 620 able to signal intra-domain using TCP [RFC0793] on un-congested links 621 to a DOTS gateway which may then transform these to a UDP [RFC0768] 622 transport inter-domain where connection oriented transports may 623 degrade; this applies to the signal channel only, as the data channel 624 requires a connection-oriented transport. The relationship between 625 the gateway and its upstream agents is opaque to the initial clients. 627 +---+ 628 | c |\ 629 +---+ \ +---+ 630 \-----TCP-----| D | +---+ 631 +---+ | O | | | 632 | c |--------TCP-----| T |------UDP------| S | 633 +---+ | S | | | 634 /-----TCP-----| G | +---+ 635 +---+ / +---+ 636 | c |/ 637 +---+ 638 example.com example.com example.net 639 DOTS clients DOTS gateway (DOTSG) DOTS server 641 Figure 7: Client-Side Gateway with Aggregation 643 +---+ 644 | c |\ 645 +---+ \ +---+ 646 \-----TCP-----| D |------UDP------+---+ 647 +---+ | O | | | 648 | c |--------TCP-----| T |------UDP------| S | 649 +---+ | S | | | 650 /-----TCP-----| G |------UDP------+---+ 651 +---+ / +---+ 652 | c |/ 653 +---+ 654 example.com example.com example.net 655 DOTS clients DOTS gateway (DOTSG) DOTS server 657 Figure 8: Client-Side Gateway without Aggregation 659 This may similarly be deployed in the inverse scenario where the 660 gateway resides in the server-side domain and may be used to 661 terminate and/or aggregate multiple clients to single transport as 662 shown in figures Figure 9 and Figure 10. 664 +---+ 665 | c |\ 666 +---+ \ +---+ 667 \-----UDP-----| D | +---+ 668 +---+ | O | | | 669 | c |--------TCP-----| T |------TCP------| S | 670 +---+ | S | | | 671 /-----TCP-----| G | +---+ 672 +---+ / +---+ 673 | c |/ 674 +---+ 675 example.com example.net example.net 676 DOTS clients DOTS gateway (DOTSG) DOTS server 678 Figure 9: Server-Side Gateway with Aggregation 680 +---+ 681 | c |\ 682 +---+ \ +---+ 683 \-----UDP-----| D |------TCP------+---+ 684 +---+ | O | | | 685 | c |--------TCP-----| T |------TCP------| S | 686 +---+ | S | | | 687 /-----UDP-----| G |------TCP------+---+ 688 +---+ / +---+ 689 | c |/ 690 +---+ 691 example.com example.net example.net 692 DOTS clients DOTS gateway (DOTSG) DOTS server 694 Figure 10: Server-Side Gateway without Aggregation 696 This document anticipates scenarios involving multiple DOTS gateways. 697 An example is a DOTS gateway at the network client's side, and 698 another one at the server side. The first gateway can be located at 699 a CPE to aggregate requests from multiple DOTS clients enabled in an 700 enterprise network. The second DOTS gateway is deployed on the 701 provider side. This scenario can be seen as a combination of the 702 client-side and server-side scenarios. 704 3. Concepts 706 3.1. DOTS Sessions 708 In order for DOTS to be effective as a vehicle for DDoS mitigation 709 requests, one or more DOTS clients must establish ongoing 710 communication with one or more DOTS servers. While the preconditions 711 for enabling DOTS in or among network domains may also involve 712 business relationships, service level agreements, or other formal or 713 informal understandings between network operators, such 714 considerations are out of scope for this document. 716 A DOTS session is established to support bilateral exchange of data 717 between an associated DOTS client and a DOTS server. In the DOTS 718 architecture, data is exchanged between DOTS agents over signal and 719 data channels. As such, a DOTS session can be a DOTS signal channel 720 session, a DOTS data channel session, or both. 722 A DOTS agent can maintain one or more DOTS sessions. 724 A DOTS signal channel session is associated with a single transport 725 connection (TCP or UDP session) and an ephemeral security association 726 (a TLS or DTLS session). Similarly, a DOTS data channel session is 727 associated with a single TCP connection and an ephemeral TLS security 728 association. 730 Mitigation requests created using DOTS signal channel are not bound 731 to the DOTS signal channel session. Instead, mitigation requests are 732 associated with a DOTS client and can be managed using different DOTS 733 signal channel sessions. 735 3.1.1. Preconditions 737 Prior to establishing a DOTS session between agents, the owners of 738 the networks, domains, services or applications involved are assumed 739 to have agreed upon the terms of the relationship involved. Such 740 agreements are out of scope for this document, but must be in place 741 for a functional DOTS architecture. 743 It is assumed that as part of any DOTS service agreement, the DOTS 744 client is provided with all data and metadata required to establish 745 communication with the DOTS server. Such data and metadata would 746 include any cryptographic information necessary to meet the message 747 confidentiality, integrity and authenticity requirement (SEC-002) in 748 [I-D.ietf-dots-requirements], and might also include the pool of DOTS 749 server addresses and ports the DOTS client should use for signal and 750 data channel messaging. 752 3.1.2. Establishing the DOTS Session 754 With the required business agreements in place, the DOTS client 755 initiates a DOTS session by contacting its DOTS server(s) over the 756 signal channel and (possibly) the data channel. To allow for DOTS 757 service flexibility, neither the order of contact nor the time 758 interval between channel creations is specified. A DOTS client MAY 759 establish signal channel first, and then data channel, or vice versa. 761 The methods by which a DOTS client receives the address and 762 associated service details of the DOTS server are not prescribed by 763 this document. For example, a DOTS client may be directly configured 764 to use a specific DOTS server IP address and port, and directly 765 provided with any data necessary to satisfy the Peer Mutual 766 Authentication requirement (SEC-001) in [I-D.ietf-dots-requirements], 767 such as symmetric or asymmetric keys, usernames and passwords, etc. 768 All configuration and authentication information in this scenario is 769 provided out-of-band by the domain operating the DOTS server. 771 At the other extreme, the architecture in this document allows for a 772 form of DOTS client auto-provisioning. For example, the domain 773 operating the DOTS server or servers might provide the client domain 774 only with symmetric or asymmetric keys to authenticate the 775 provisioned DOTS clients. Only the keys would then be directly 776 configured on DOTS clients, but the remaining configuration required 777 to provision the DOTS clients could be learned through mechanisms 778 similar to DNS SRV [RFC2782] or DNS Service Discovery [RFC6763]. 780 The DOTS client SHOULD successfully authenticate and exchange 781 messages with the DOTS server over both signal and (if used) data 782 channel as soon as possible to confirm that both channels are 783 operational. 785 As described in [I-D.ietf-dots-requirements] (DM-008), the DOTS 786 client can configure preferred values for acceptable signal loss, 787 mitigation lifetime, and heartbeat intervals when establishing the 788 DOTS signal channel session. A DOTS signal channel session is not 789 active until DOTS agents have agreed on the values for these DOTS 790 session parameters, a process defined by the protocol. 792 Once the DOTS client begins receiving DOTS server signals, the DOTS 793 session is active. At any time during the DOTS session, the DOTS 794 client may use the data channel to manage aliases, manage drop- and 795 accept-listed prefixes or addresses, leverage vendor-specific 796 extensions, and so on. Note that unlike the signal channel, there is 797 no requirement that the data channel remains operational in attack 798 conditions (See Data Channel Requirements, Section 2.3 of 799 [I-D.ietf-dots-requirements]). 801 3.1.3. Maintaining the DOTS Session 803 DOTS clients and servers periodically send heartbeats to each other 804 over the signal channel, discussed in [I-D.ietf-dots-requirements] 805 (SIG-004). DOTS agent operators SHOULD configure the heartbeat 806 interval such that the frequency does not lead to accidental denials 807 of service due to the overwhelming number of heartbeats a DOTS agent 808 must field. 810 Either DOTS agent may consider a DOTS signal channel session 811 terminated in the extended absence of a heartbeat from its peer 812 agent. The period of that absence will be established in the 813 protocol definition. 815 3.2. Modes of Signaling 817 This section examines the modes of signaling between agents in a DOTS 818 architecture. 820 3.2.1. Direct Signaling 822 A DOTS session may take the form of direct signaling between the DOTS 823 clients and servers, as shown in Figure 11. 825 +-------------+ +-------------+ 826 | DOTS client |<------signal session------>| DOTS server | 827 +-------------+ +-------------+ 829 Figure 11: Direct Signaling 831 In a direct DOTS session, the DOTS client and server are 832 communicating directly. Direct signaling may exist inter- or intra- 833 domain. The DOTS session is abstracted from the underlying networks 834 or network elements the signals traverse: in direct signaling, the 835 DOTS client and server are logically adjacent. 837 3.2.2. Redirected Signaling 839 In certain circumstances, a DOTS server may want to redirect a DOTS 840 client to an alternative DOTS server for a DOTS signal channel 841 session. Such circumstances include but are not limited to: 843 o Maximum number of DOTS signal channel sessions with clients has 844 been reached; 846 o Mitigation capacity exhaustion in the mitigator with which the 847 specific DOTS server is communicating; 849 o Mitigator outage or other downtime, such as scheduled maintenance; 851 o Scheduled DOTS server maintenance; 853 o Scheduled modifications to the network path between DOTS server 854 and DOTS client. 856 A basic redirected DOTS signal channel session resembles the 857 following, as shown in Figure 12. 859 +-------------+ +---------------+ 860 | |<-(1)--- DOTS signal ------>| | 861 | | channel session 1 | | 862 | |<=(2)== redirect to B ======| | 863 | DOTS client | | DOTS server A | 864 | |X-(4)--- DOTS signal ------X| | 865 | | channel session 1 | | 866 | | | | 867 +-------------+ +---------------+ 868 ^ 869 | 870 (3) DOTS signal channel 871 | session 2 872 v 873 +---------------+ 874 | DOTS server B | 875 +---------------+ 877 Figure 12: Redirected Signaling 879 1. Previously established DOTS signal channel session 1 exists 880 between a DOTS client and DOTS server A. 882 2. DOTS server A sends a server signal redirecting the client to 883 DOTS server B. 885 3. If the DOTS client does not already have a separate DOTS signal 886 channel session with the redirection target, the DOTS client 887 initiates and establishes DOTS signal channel session 2 with DOTS 888 server B. 890 4. Having redirected the DOTS client, DOTS server A ceases sending 891 server signals. The DOTS client likewise stops sending client 892 signals to DOTS server A. DOTS signal channel session 1 is 893 terminated. 895 3.2.3. Recursive Signaling 897 DOTS is centered around improving the speed and efficiency of 898 coordinated response to DDoS attacks. One scenario not yet discussed 899 involves coordination among federated domains operating DOTS servers 900 and mitigators. 902 In the course of normal DOTS operations, a DOTS client communicates 903 the need for mitigation to a DOTS server, and that server initiates 904 mitigation on a mitigator with which the server has an established 905 service relationship. The operator of the mitigator may in turn 906 monitor mitigation performance and capacity, as the attack being 907 mitigated may grow in severity beyond the mitigating domain's 908 capabilities. 910 The operator of the mitigator has limited options in the event a DOTS 911 client-requested mitigation is being overwhelmed by the severity of 912 the attack. Out-of-scope business or service level agreements may 913 permit the mitigating domain to drop the mitigation and let attack 914 traffic flow unchecked to the target, but this only encourages attack 915 escalation. In the case where the mitigating domain is the upstream 916 service provider for the attack target, this may mean the mitigating 917 domain and its other services and users continue to suffer the 918 incidental effects of the attack. 920 A recursive signaling model as shown in Figure 13 offers an 921 alternative. In a variation of the use case "Upstream DDoS 922 Mitigation by an Upstream Internet Transit Provider" described in 923 [I-D.ietf-dots-use-cases], a domain operating a DOTS server and 924 mitigator also operates a DOTS client. This DOTS client has an 925 established DOTS session with a DOTS server belonging to a separate 926 administrative domain. 928 With these preconditions in place, the operator of the mitigator 929 being overwhelmed or otherwise performing inadequately may request 930 mitigation for the attack target from this separate DOTS-aware 931 domain. Such a request recurses the originating mitigation request 932 to the secondary DOTS server, in the hope of building a cumulative 933 mitigation against the attack. 935 example.net domain 936 . . . . . . . . . . . . . . . . . 937 . Gn . 938 +----+ 1 . +----+ +-----------+ . 939 | Cc |<--------->| Sn |~~~~~~~| Mitigator | . 940 +----+ . +====+ | Mn | . 941 . | Cn | +-----------+ . 942 example.com . +----+ . 943 client . ^ . 944 . . .|. . . . . . . . . . . . . . 945 | 946 2 | 947 | 948 . . .|. . . . . . . . . . . . . . 949 . v . 950 . +----+ +-----------+ . 951 . | So |~~~~~~~| Mitigator | . 952 . +----+ | Mo | . 953 . +-----------+ . 954 . . 955 . . . . . . . . . . . . . . . . . 956 example.org domain 958 Figure 13: Recursive Signaling 960 In Figure 13, client Cc signals a request for mitigation across 961 inter-domain DOTS session 1 to the DOTS server Sn belonging to the 962 example.net domain. DOTS server Sn enables mitigation on mitigator 963 Mn. DOTS server Sn is half of DOTS gateway Gn, being deployed 964 logically back-to-back with DOTS client Cn, which has pre-existing 965 inter-domain DOTS session 2 with the DOTS server So belonging to the 966 example.org domain. At any point, DOTS server Sn MAY recurse an on- 967 going mitigation request through DOTS client Cn to DOTS server So, in 968 the expectation that mitigator Mo will be activated to aid in the 969 defense of the attack target. 971 Recursive signaling is opaque to the DOTS client. To maximize 972 mitigation visibility to the DOTS client, however, the recursing 973 domain SHOULD provide recursed mitigation feedback in signals 974 reporting on mitigation status to the DOTS client. For example, the 975 recursing domain's mitigator should incorporate into mitigation 976 status messages available metrics such as dropped packet or byte 977 counts from the recursed mitigation. 979 DOTS clients involved in recursive signaling must be able to withdraw 980 requests for mitigation without warning or justification, per SIG-006 981 in [I-D.ietf-dots-requirements]. 983 Operators recursing mitigation requests MAY maintain the recursed 984 mitigation for a brief, protocol-defined period in the event the DOTS 985 client originating the mitigation withdraws its request for help, as 986 per the discussion of managing mitigation toggling in SIG-006 of 987 [I-D.ietf-dots-requirements]. 989 Deployment of recursive signaling may result in traffic redirection, 990 examination and mitigation extending beyond the initial bilateral 991 relationship between DOTS client and DOTS server. As such, client 992 control over the network path of mitigated traffic may be reduced. 993 DOTS client operators should be aware of any privacy concerns, and 994 work with DOTS server operators employing recursive signaling to 995 ensure shared sensitive material is suitably protected. 997 3.2.4. Anycast Signaling 999 The DOTS architecture does not assume the availability of anycast 1000 within a DOTS deployment, but neither does the architecture exclude 1001 it. Domains operating DOTS servers MAY deploy DOTS servers with an 1002 anycast Service Address as described in BCP 126 [RFC4786]. In such a 1003 deployment, DOTS clients connecting to the DOTS Service Address may 1004 be communicating with distinct DOTS servers, depending on the network 1005 configuration at the time the DOTS clients connect. Among other 1006 benefits, anycast signaling potentially offers the following: 1008 o Simplified DOTS client configuration, including service discovery 1009 through the methods described in [RFC7094]. In this scenario, the 1010 "instance discovery" message would be a DOTS client initiating a 1011 DOTS session to the DOTS server anycast Service Address, to which 1012 the DOTS server would reply with a redirection to the DOTS server 1013 unicast address the client should use for DOTS. 1015 o Region- or customer-specific deployments, in which the DOTS 1016 Service Addresses route to distinct DOTS servers depending on the 1017 client region or the customer network in which a DOTS client 1018 resides. 1020 o Operational resiliency, spreading DOTS signaling traffic across 1021 the DOTS server domain's networks, and thereby also reducing the 1022 potential attack surface, as described in BCP 126 [RFC4786]. 1024 3.2.4.1. Anycast Signaling Considerations 1026 As long as network configuration remains stable, anycast DOTS 1027 signaling is to the individual DOTS client indistinct from direct 1028 signaling. However, the operational challenges inherent in anycast 1029 signaling are anything but negligible, and DOTS server operators must 1030 carefully weigh the risks against the benefits before deploying. 1032 While the DOTS signal channel primarily operates over UDP per SIG-001 1033 in [I-D.ietf-dots-requirements], the signal channel also requires 1034 mutual authentication between DOTS agents, with associated security 1035 state on both ends. 1037 Network instability is of particular concern with anycast signaling, 1038 as DOTS signal channels are expected to be long-lived, and 1039 potentially operating under congested network conditions caused by a 1040 volumetric DDoS attack. 1042 For example, a network configuration altering the route to the DOTS 1043 server during active anycast signaling may cause the DOTS client to 1044 send messages to a DOTS server other than the one with which it 1045 initially established a signaling session. That second DOTS server 1046 may not have the security state of the existing session, forcing the 1047 DOTS client to initialize a new DOTS session. This challenge might 1048 in part be mitigated by use of resumption via a PSK in TLS 1.3 1049 [RFC8446] and DTLS 1.3 [I-D.ietf-tls-dtls13] (session resumption in 1050 TLS 1.2 [RFC5246] and DTLS 1.2 [RFC6347]), but keying material must 1051 be available to all DOTS servers sharing the anycast Service Address 1052 in that case. 1054 While the DOTS client will try to establish a new DOTS session with 1055 the DOTS server now acting as the anycast DOTS Service Address, the 1056 link between DOTS client and server may be congested with attack 1057 traffic, making signal session establishment difficult. In such a 1058 scenario, anycast Service Address instability becomes a sort of 1059 signal session flapping, with obvious negative consequences for the 1060 DOTS deployment. 1062 Anycast signaling deployments similarly must also take into account 1063 active mitigations. Active mitigations initiated through a DOTS 1064 session may involve diverting traffic to a scrubbing center. If the 1065 DOTS session flaps due to anycast changes as described above, 1066 mitigation may also flap as the DOTS servers sharing the anycast DOTS 1067 service address toggles mitigation on detecting DOTS session loss, 1068 depending on whether the client has configured mitigation on loss of 1069 signal. 1071 3.2.5. Signaling Considerations for Network Address Translation 1073 Network address translators (NATs) are expected to be a common 1074 feature of DOTS deployments. The Middlebox Traversal Guidelines in 1075 [RFC8085] include general NAT considerations for DOTS deployements 1076 when the signal channel is established over UDP. 1078 Additional DOTS-specific considerations arise when NATs are part of 1079 the DOTS architecture. For example, DDoS attack detection behind a 1080 NAT will detect attacks against internal addresses. A DOTS client 1081 subsequently asked to request mitigation for the attacked scope of 1082 addresses cannot reasonably perform the task, due to the lack of 1083 externally routable addresses in the mitigation scope. 1085 The following considerations do not cover all possible scenarios, but 1086 are meant rather to highlight anticipated common issues when 1087 signaling through NATs. 1089 3.2.5.1. Direct Provisioning of Internal-to-External Address Mappings 1091 Operators may circumvent the problem of translating internal 1092 addresses or prefixes to externally routable mitigation scopes by 1093 directly provisioning the mappings of external addresses to internal 1094 protected resources on the DOTS client. When the operator requests 1095 mitigation scoped for internal addresses, directly or through 1096 automated means, the DOTS client looks up the matching external 1097 addresses or prefixes, and issues a mitigation request scoped to that 1098 externally routable information. 1100 When directly provisioning the address mappings, operators must 1101 ensure the mappings remain up to date, or risk losing the ability to 1102 request accurate mitigation scopes. To that aim, the DOTS client can 1103 rely on mechanisms, such as [RFC8512] to retrieve static explicit 1104 mappings. This document does not prescribe the method by which 1105 mappings are maintained once they are provisioned on the DOTS client. 1107 3.2.5.2. Resolving Public Mitigation Scope with Port Control Protocol 1108 (PCP) 1110 Port Control Protocol (PCP) [RFC6887] may be used to retrieve the 1111 external addresses/prefixes and/or port numbers if the NAT function 1112 embeds a PCP server. 1114 A DOTS client can use the information retrieved by means of PCP to 1115 feed the DOTS protocol(s) messages that will be sent to a DOTS 1116 server. These messages will convey the external addresses/prefixes 1117 as set by the NAT. 1119 PCP also enables discovery and configuration of the lifetime of port 1120 mappings instantiated in intermediate NAT devices. Discovery of port 1121 mapping lifetimes can reduce the dependency on heartbeat messages to 1122 maintain mappings, and therefore reduce the load on DOTS servers and 1123 the network. 1125 3.2.5.3. Resolving Public Mitigation Scope with Session Traversal 1126 Utilities (STUN) 1128 An internal resource, e.g., a Web server, can discover its reflexive 1129 transport address through a STUN Binding request/response 1130 transaction, as described in [RFC5389]. After learning its reflexive 1131 transport address from the STUN server, the internal resource can 1132 export its reflexive transport address and internal transport address 1133 to the DOTS client, thereby enabling the DOTS client to request 1134 mitigation with the correct external scope, as depicted in Figure 14. 1135 The mechanism for providing the DOTS client with the reflexive 1136 transport address and internal transport address is unspecified in 1137 this document. 1139 In order to prevent an attacker from modifying the STUN messages in 1140 transit, the STUN client and server MUST use the message-integrity 1141 mechanism discussed in Section 10 of [RFC5389] or use STUN over DTLS 1142 [RFC7350] or use STUN over TLS. If the STUN client is behind a NAT 1143 that performs Endpoint-Dependent Mapping [RFC5128], the internal 1144 service cannot provide the DOTS client with the reflexive transport 1145 address discovered using STUN. The behavior of a NAT between the 1146 STUN client and the STUN server could be discovered using the 1147 experimental techniques discussed in [RFC5780], but note that there 1148 is currently no standardized way for a STUN client to reliably 1149 determine if it is behind a NAT that performs Endpoint-Dependent 1150 Mapping. 1152 Binding Binding 1153 +--------+ request +---+ request +--------+ 1154 | STUN |<----------| N |<----------| STUN | 1155 | server | | A | | client | 1156 | |---------->| T |---------->| | 1157 +--------+ Binding +---+ Binding +--------+ 1158 response response | 1159 | reflexive transport address 1160 | & internal transport address 1161 v 1162 +--------+ 1163 | DOTS | 1164 | client | 1165 +--------+ 1167 Figure 14: Resolving mitigation scope with STUN 1169 3.2.5.4. Resolving Requested Mitigation Scope with DNS 1171 DOTS supports mitigation scoped to DNS names. As discussed in 1172 [RFC3235], using DNS names instead of IP addresses potentially avoids 1173 the address translation problem, as long as the name is internally 1174 and externally resolvable by the same name. For example, a detected 1175 attack's internal target address can be mapped to a DNS name through 1176 a reverse lookup. The DNS name returned by the reverse lookup can 1177 then be provided to the DOTS client as the external scope for 1178 mitigation. For the reverse DNS lookup, DNS Security Extensions 1179 (DNSSEC) [RFC4033] must be used where the authenticity of response is 1180 critical. 1182 3.3. Triggering Requests for Mitigation 1184 [I-D.ietf-dots-requirements] places no limitation on the 1185 circumstances in which a DOTS client operator may request mitigation, 1186 nor does it demand justification for any mitigation request, thereby 1187 reserving operational control over DDoS defense for the domain 1188 requesting mitigation. This architecture likewise does not prescribe 1189 the network conditions and mechanisms triggering a mitigation request 1190 from a DOTS client. 1192 However, considering selected possible mitigation triggers from an 1193 architectural perspective offers a model for alternative or 1194 unanticipated triggers for DOTS deployments. In all cases, what 1195 network conditions merit a mitigation request are at the discretion 1196 of the DOTS client operator. 1198 The mitigation request itself is defined by DOTS, however the 1199 interfaces required to trigger the mitigation request in the 1200 following scenarios are implementation-specific. 1202 3.3.1. Manual Mitigation Request 1204 A DOTS client operator may manually prepare a request for mitigation, 1205 including scope and duration, and manually instruct the DOTS client 1206 to send the mitigation request to the DOTS server. In context, a 1207 manual request is a request directly issued by the operator without 1208 automated decision-making performed by a device interacting with the 1209 DOTS client. Modes of manual mitigation requests include an operator 1210 entering a command into a text interface, or directly interacting 1211 with a graphical interface to send the request. 1213 An operator might do this, for example, in response to notice of an 1214 attack delivered by attack detection equipment or software, and the 1215 alerting detector lacks interfaces or is not configured to use 1216 available interfaces to translate the alert to a mitigation request 1217 automatically. 1219 In a variation of the above scenario, the operator may have 1220 preconfigured on the DOTS client mitigation requests for various 1221 resources in the operator's domain. When notified of an attack, the 1222 DOTS client operator manually instructs the DOTS client to send the 1223 relevant preconfigured mitigation request for the resources under 1224 attack. 1226 A further variant involves recursive signaling, as described in 1227 Section 3.2.3. The DOTS client in this case is the second half of a 1228 DOTS gateway (back-to-back DOTS server and client). As in the 1229 previous scenario, the scope and duration of the mitigation request 1230 are pre-existing, but in this case are derived from the mitigation 1231 request received from a downstream DOTS client by the DOTS server. 1232 Assuming the preconditions required by Section 3.2.3 are in place, 1233 the DOTS gateway operator may at any time manually request mitigation 1234 from an upstream DOTS server, sending a mitigation request derived 1235 from the downstream DOTS client's request. 1237 The motivations for a DOTS client operator to request mitigation 1238 manually are not prescribed by this architecture, but are expected to 1239 include some of the following: 1241 o Notice of an attack delivered via e-mail or alternative messaging 1243 o Notice of an attack delivered via phone call 1245 o Notice of an attack delivered through the interface(s) of 1246 networking monitoring software deployed in the operator's domain 1248 o Manual monitoring of network behavior through network monitoring 1249 software 1251 3.3.2. Automated Conditional Mitigation Request 1253 Unlike manual mitigation requests, which depend entirely on the DOTS 1254 client operator's capacity to react with speed and accuracy to every 1255 detected or detectable attack, mitigation requests triggered by 1256 detected attack conditions reduce the operational burden on the DOTS 1257 client operator, and minimize the latency between attack detection 1258 and the start of mitigation. 1260 Mitigation requests are triggered in this scenario by operator- 1261 specified network conditions. Attack detection is deployment- 1262 specific, and not constrained by this architecture. Similarly the 1263 specifics of a condition are left to the discretion of the operator, 1264 though common conditions meriting mitigation include the following: 1266 o Detected attack exceeding a rate in packets per second (pps). 1268 o Detected attack exceeding a rate in bytes per second (bps). 1270 o Detected resource exhaustion in an attack target. 1272 o Detected resource exhaustion in the local domain's mitigator. 1274 o Number of open connections to an attack target. 1276 o Number of attack sources in a given attack. 1278 o Number of active attacks against targets in the operator's domain. 1280 o Conditional detection developed through arbitrary statistical 1281 analysis or deep learning techniques. 1283 o Any combination of the above. 1285 When automated conditional mitigation requests are enabled, 1286 violations of any of the above conditions, or any additional 1287 operator-defined conditions, will trigger a mitigation request from 1288 the DOTS client to the DOTS server. The interfaces between the 1289 application detecting the condition violation and the DOTS client are 1290 implementation-specific. 1292 3.3.3. Automated Mitigation on Loss of Signal 1294 To maintain a DOTS signal channel session, the DOTS client and the 1295 DOTS server exchange regular but infrequent messages across the 1296 signal channel. In the absence of an attack, the probability of 1297 message loss in the signaling channel should be extremely low. Under 1298 attack conditions, however, some signal loss may be anticipated as 1299 attack traffic congests the link, depending on the attack type. 1301 While [I-D.ietf-dots-requirements] specifies the DOTS protocol be 1302 robust when signaling under attack conditions, there are nevertheless 1303 scenarios in which the DOTS signal is lost in spite of protocol best 1304 efforts. To handle such scenarios, a DOTS operator may request one 1305 or more mitigations which are triggered only when the DOTS server 1306 ceases receiving DOTS client heartbeats beyond the miss count or 1307 interval permitted by the protocol. 1309 The impact of mitigating due to loss of signal in either direction 1310 must be considered carefully before enabling it. Signal loss is not 1311 caused by links congested with attack traffic alone, and as such 1312 mitigation requests triggered by signal channel degradation in either 1313 direction may incur unnecessary costs, in network performance and 1314 operational expense alike. 1316 4. IANA Considerations 1318 This document has no actions for IANA. 1320 5. Security Considerations 1322 This section describes identified security considerations for the 1323 DOTS architecture. 1325 DOTS is at risk from three primary attack vectors: agent 1326 impersonation, traffic injection and signal blocking. These vectors 1327 may be exploited individually or in concert by an attacker to 1328 confuse, disable, take information from, or otherwise inhibit DOTS 1329 agents. 1331 Any attacker with the ability to impersonate a legitimate DOTS client 1332 or server or, indeed, inject false messages into the stream may 1333 potentially trigger/withdraw traffic redirection, trigger/cancel 1334 mitigation activities or subvert drop-/accept-lists. From an 1335 architectural standpoint, operators SHOULD ensure best current 1336 practices for secure communication are observed for data and signal 1337 channel confidentiality, integrity and authenticity. Care must be 1338 taken to ensure transmission is protected by appropriately secure 1339 means, reducing attack surface by exposing only the minimal required 1340 services or interfaces. Similarly, received data at rest SHOULD be 1341 stored with a satisfactory degree of security. 1343 As many mitigation systems employ diversion to scrub attack traffic, 1344 operators of DOTS agents SHOULD ensure DOTS sessions are resistant to 1345 Man-in-the-Middle (MitM) attacks. An attacker with control of a DOTS 1346 client may negatively influence network traffic by requesting and 1347 withdrawing requests for mitigation for particular prefixes, leading 1348 to route or DNS flapping. 1350 Any attack targeting the availability of DOTS servers may disrupt the 1351 ability of the system to receive and process DOTS signals resulting 1352 in failure to fulfill a mitigation request. DOTS agents SHOULD be 1353 given adequate protections, again in accordance with best current 1354 practices for network and host security. 1356 6. Contributors 1358 Mohamed Boucadair 1359 Orange 1361 mohamed.boucadair@orange.com 1363 Christopher Gray 1365 Christopher_Gray3@cable.comcast.com 1367 7. Acknowledgments 1369 Thanks to Matt Richardson, Roman Danyliw, Frank Xialiang, Roland 1370 Dobbins, Wei Pan, Kaname Nishizuka, Jon Shallow and Mohamed Boucadair 1371 for their comments and suggestions. 1373 8. References 1375 8.1. Normative References 1377 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1378 Requirement Levels", BCP 14, RFC 2119, 1379 DOI 10.17487/RFC2119, March 1997, 1380 . 1382 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 1383 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 1384 May 2017, . 1386 8.2. Informative References 1388 [I-D.ietf-dots-requirements] 1389 Mortensen, A., K, R., and R. Moskowitz, "Distributed 1390 Denial of Service (DDoS) Open Threat Signaling 1391 Requirements", draft-ietf-dots-requirements-22 (work in 1392 progress), March 2019. 1394 [I-D.ietf-dots-use-cases] 1395 Dobbins, R., Migault, D., Fouant, S., Moskowitz, R., 1396 Teague, N., Xia, L., and K. Nishizuka, "Use cases for DDoS 1397 Open Threat Signaling", draft-ietf-dots-use-cases-17 (work 1398 in progress), January 2019. 1400 [I-D.ietf-tls-dtls13] 1401 Rescorla, E., Tschofenig, H., and N. Modadugu, "The 1402 Datagram Transport Layer Security (DTLS) Protocol Version 1403 1.3", draft-ietf-tls-dtls13-31 (work in progress), March 1404 2019. 1406 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 1407 DOI 10.17487/RFC0768, August 1980, 1408 . 1410 [RFC0793] Postel, J., "Transmission Control Protocol", STD 7, 1411 RFC 793, DOI 10.17487/RFC0793, September 1981, 1412 . 1414 [RFC1035] Mockapetris, P., "Domain names - implementation and 1415 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 1416 November 1987, . 1418 [RFC2782] Gulbrandsen, A., Vixie, P., and L. Esibov, "A DNS RR for 1419 specifying the location of services (DNS SRV)", RFC 2782, 1420 DOI 10.17487/RFC2782, February 2000, 1421 . 1423 [RFC3235] Senie, D., "Network Address Translator (NAT)-Friendly 1424 Application Design Guidelines", RFC 3235, 1425 DOI 10.17487/RFC3235, January 2002, 1426 . 1428 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 1429 A., Peterson, J., Sparks, R., Handley, M., and E. 1430 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 1431 DOI 10.17487/RFC3261, June 2002, 1432 . 1434 [RFC4033] Arends, R., Austein, R., Larson, M., Massey, D., and S. 1435 Rose, "DNS Security Introduction and Requirements", 1436 RFC 4033, DOI 10.17487/RFC4033, March 2005, 1437 . 1439 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 1440 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 1441 DOI 10.17487/RFC4271, January 2006, 1442 . 1444 [RFC4732] Handley, M., Ed., Rescorla, E., Ed., and IAB, "Internet 1445 Denial-of-Service Considerations", RFC 4732, 1446 DOI 10.17487/RFC4732, December 2006, 1447 . 1449 [RFC4786] Abley, J. and K. Lindqvist, "Operation of Anycast 1450 Services", BCP 126, RFC 4786, DOI 10.17487/RFC4786, 1451 December 2006, . 1453 [RFC5128] Srisuresh, P., Ford, B., and D. Kegel, "State of Peer-to- 1454 Peer (P2P) Communication across Network Address 1455 Translators (NATs)", RFC 5128, DOI 10.17487/RFC5128, March 1456 2008, . 1458 [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security 1459 (TLS) Protocol Version 1.2", RFC 5246, 1460 DOI 10.17487/RFC5246, August 2008, 1461 . 1463 [RFC5389] Rosenberg, J., Mahy, R., Matthews, P., and D. Wing, 1464 "Session Traversal Utilities for NAT (STUN)", RFC 5389, 1465 DOI 10.17487/RFC5389, October 2008, 1466 . 1468 [RFC5780] MacDonald, D. and B. Lowekamp, "NAT Behavior Discovery 1469 Using Session Traversal Utilities for NAT (STUN)", 1470 RFC 5780, DOI 10.17487/RFC5780, May 2010, 1471 . 1473 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 1474 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 1475 January 2012, . 1477 [RFC6763] Cheshire, S. and M. Krochmal, "DNS-Based Service 1478 Discovery", RFC 6763, DOI 10.17487/RFC6763, February 2013, 1479 . 1481 [RFC6887] Wing, D., Ed., Cheshire, S., Boucadair, M., Penno, R., and 1482 P. Selkirk, "Port Control Protocol (PCP)", RFC 6887, 1483 DOI 10.17487/RFC6887, April 2013, 1484 . 1486 [RFC7092] Kaplan, H. and V. Pascual, "A Taxonomy of Session 1487 Initiation Protocol (SIP) Back-to-Back User Agents", 1488 RFC 7092, DOI 10.17487/RFC7092, December 2013, 1489 . 1491 [RFC7094] McPherson, D., Oran, D., Thaler, D., and E. Osterweil, 1492 "Architectural Considerations of IP Anycast", RFC 7094, 1493 DOI 10.17487/RFC7094, January 2014, 1494 . 1496 [RFC7350] Petit-Huguenin, M. and G. Salgueiro, "Datagram Transport 1497 Layer Security (DTLS) as Transport for Session Traversal 1498 Utilities for NAT (STUN)", RFC 7350, DOI 10.17487/RFC7350, 1499 August 2014, . 1501 [RFC8085] Eggert, L., Fairhurst, G., and G. Shepherd, "UDP Usage 1502 Guidelines", BCP 145, RFC 8085, DOI 10.17487/RFC8085, 1503 March 2017, . 1505 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 1506 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 1507 . 1509 [RFC8512] Boucadair, M., Ed., Sivakumar, S., Jacquenet, C., 1510 Vinapamula, S., and Q. Wu, "A YANG Module for Network 1511 Address Translation (NAT) and Network Prefix Translation 1512 (NPT)", RFC 8512, DOI 10.17487/RFC8512, January 2019, 1513 . 1515 Authors' Addresses 1517 Andrew Mortensen (editor) 1518 Arbor Networks 1519 2727 S. State St 1520 Ann Arbor, MI 48104 1521 United States 1523 EMail: andrewmortensen@gmail.com 1525 Tirumaleswar Reddy (editor) 1526 McAfee, Inc. 1527 Embassy Golf Link Business Park 1528 Bangalore, Karnataka 560071 1529 India 1531 EMail: kondtir@gmail.com 1533 Flemming Andreasen 1534 Cisco 1535 United States 1537 EMail: fandreas@cisco.com 1538 Nik Teague 1539 Verisign 1540 United States 1542 EMail: nteague@verisign.com 1544 Rich Compton 1545 Charter 1547 EMail: Rich.Compton@charter.com