idnits 2.17.1 draft-ietf-detnet-security-11.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (August 14, 2020) is 1351 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Outdated reference: A later version (-14) exists of draft-ietf-detnet-flow-information-model-10 == Outdated reference: A later version (-07) exists of draft-ietf-detnet-ip-over-tsn-03 == Outdated reference: A later version (-13) exists of draft-ietf-detnet-mpls-10 -- Obsolete informational reference (is this intentional?): RFC 4447 (Obsoleted by RFC 8077) Summary: 0 errors (**), 0 flaws (~~), 4 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Internet Engineering Task Force T. Mizrahi 3 Internet-Draft HUAWEI 4 Intended status: Informational E. Grossman, Ed. 5 Expires: February 15, 2021 DOLBY 6 August 14, 2020 8 Deterministic Networking (DetNet) Security Considerations 9 draft-ietf-detnet-security-11 11 Abstract 13 A DetNet (deterministic network) provides specific performance 14 guarantees to its data flows, such as extremely low data loss rates 15 and bounded latency. As a result, securing a DetNet requires that in 16 addition to the best practice security measures taken for any 17 mission-critical network, additional security measures may be needed 18 to secure the intended operation of these novel service properties. 20 This document addresses DetNet-specific security considerations from 21 the perspectives of both the DetNet system-level designer and 22 component designer. System considerations include a threat model, 23 taxonomy of relevant attacks, and associations of threats versus use 24 cases and service properties. Component-level considerations include 25 ingress filtering and packet arrival time violation detection. This 26 document also addresses DetNet security considerations specific to 27 the IP and MPLS data plane technologies thereby complementing the 28 Security Considerations sections of the various DetNet Data Plane 29 (and other) DetNet documents. 31 Status of This Memo 33 This Internet-Draft is submitted in full conformance with the 34 provisions of BCP 78 and BCP 79. 36 Internet-Drafts are working documents of the Internet Engineering 37 Task Force (IETF). Note that other groups may also distribute 38 working documents as Internet-Drafts. The list of current Internet- 39 Drafts is at https://datatracker.ietf.org/drafts/current/. 41 Internet-Drafts are draft documents valid for a maximum of six months 42 and may be updated, replaced, or obsoleted by other documents at any 43 time. It is inappropriate to use Internet-Drafts as reference 44 material or to cite them other than as "work in progress." 46 This Internet-Draft will expire on February 15, 2021. 48 Copyright Notice 50 Copyright (c) 2020 IETF Trust and the persons identified as the 51 document authors. All rights reserved. 53 This document is subject to BCP 78 and the IETF Trust's Legal 54 Provisions Relating to IETF Documents 55 (https://trustee.ietf.org/license-info) in effect on the date of 56 publication of this document. Please review these documents 57 carefully, as they describe your rights and restrictions with respect 58 to this document. Code Components extracted from this document must 59 include Simplified BSD License text as described in Section 4.e of 60 the Trust Legal Provisions and are provided without warranty as 61 described in the Simplified BSD License. 63 Table of Contents 65 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 66 2. Abbreviations and Terminology . . . . . . . . . . . . . . . . 6 67 3. Security Considerations for DetNet Component Design . . . . . 6 68 3.1. Resource Allocation . . . . . . . . . . . . . . . . . . . 7 69 3.2. Explicit Routes . . . . . . . . . . . . . . . . . . . . . 7 70 3.3. Redundant Path Support . . . . . . . . . . . . . . . . . 7 71 3.4. Timing (or other) Violation Reporting . . . . . . . . . . 8 72 4. DetNet Security Considerations Compared With DiffServ 73 Security Considerations . . . . . . . . . . . . . . . . . . . 9 74 5. Security Threats . . . . . . . . . . . . . . . . . . . . . . 10 75 5.1. Threat Model . . . . . . . . . . . . . . . . . . . . . . 10 76 5.2. Threat Analysis . . . . . . . . . . . . . . . . . . . . . 11 77 5.2.1. Delay . . . . . . . . . . . . . . . . . . . . . . . . 11 78 5.2.2. DetNet Flow Modification or Spoofing . . . . . . . . 11 79 5.2.3. Resource Segmentation (Inter-segment Attack) . . . . 12 80 5.2.4. Packet Replication and Elimination . . . . . . . . . 12 81 5.2.4.1. Replication: Increased Attack Surface . . . . . . 12 82 5.2.4.2. Replication-related Header Manipulation . . . . . 12 83 5.2.5. Path Choice . . . . . . . . . . . . . . . . . . . . . 13 84 5.2.5.1. Path Manipulation . . . . . . . . . . . . . . . . 13 85 5.2.5.2. Path Choice: Increased Attack Surface . . . . . . 13 86 5.2.6. Controller Plane . . . . . . . . . . . . . . . . . . 13 87 5.2.6.1. Control or Signaling Packet Modification . . . . 13 88 5.2.6.2. Control or Signaling Packet Injection . . . . . . 13 89 5.2.7. Scheduling or Shaping . . . . . . . . . . . . . . . . 13 90 5.2.7.1. Reconnaissance . . . . . . . . . . . . . . . . . 13 91 5.2.8. Time Synchronization Mechanisms . . . . . . . . . . . 13 92 5.3. Threat Summary . . . . . . . . . . . . . . . . . . . . . 14 93 6. Security Threat Impacts . . . . . . . . . . . . . . . . . . . 14 94 6.1. Delay-Attacks . . . . . . . . . . . . . . . . . . . . . . 17 95 6.1.1. Data Plane Delay Attacks . . . . . . . . . . . . . . 17 96 6.1.2. Controller Plane Delay Attacks . . . . . . . . . . . 18 97 6.2. Flow Modification and Spoofing . . . . . . . . . . . . . 18 98 6.2.1. Flow Modification . . . . . . . . . . . . . . . . . . 18 99 6.2.2. Spoofing . . . . . . . . . . . . . . . . . . . . . . 18 100 6.2.2.1. Dataplane Spoofing . . . . . . . . . . . . . . . 18 101 6.2.2.2. Controller Plane Spoofing . . . . . . . . . . . . 19 102 6.3. Segmentation Attacks (injection) . . . . . . . . . . . . 19 103 6.3.1. Data Plane Segmentation . . . . . . . . . . . . . . . 19 104 6.3.2. Controller Plane Segmentation . . . . . . . . . . . . 19 105 6.4. Replication and Elimination . . . . . . . . . . . . . . . 20 106 6.4.1. Increased Attack Surface . . . . . . . . . . . . . . 20 107 6.4.2. Header Manipulation at Elimination Routers . . . . . 20 108 6.5. Control or Signaling Packet Modification . . . . . . . . 20 109 6.6. Control or Signaling Packet Injection . . . . . . . . . . 20 110 6.7. Reconnaissance . . . . . . . . . . . . . . . . . . . . . 20 111 6.8. Attacks on Time Sync Mechanisms . . . . . . . . . . . . . 21 112 6.9. Attacks on Path Choice . . . . . . . . . . . . . . . . . 21 113 7. Security Threat Mitigation . . . . . . . . . . . . . . . . . 21 114 7.1. Path Redundancy . . . . . . . . . . . . . . . . . . . . . 21 115 7.2. Integrity Protection . . . . . . . . . . . . . . . . . . 22 116 7.3. DetNet Node Authentication . . . . . . . . . . . . . . . 22 117 7.4. Dummy Traffic Insertion . . . . . . . . . . . . . . . . . 23 118 7.5. Encryption . . . . . . . . . . . . . . . . . . . . . . . 23 119 7.5.1. Encryption Considerations for DetNet . . . . . . . . 24 120 7.6. Control and Signaling Message Protection . . . . . . . . 25 121 7.7. Dynamic Performance Analytics . . . . . . . . . . . . . . 25 122 7.8. Mitigation Summary . . . . . . . . . . . . . . . . . . . 26 123 8. Association of Attacks to Use Cases . . . . . . . . . . . . . 27 124 8.1. Association of Attacks to Use Case Common Themes . . . . 27 125 8.1.1. Sub-Network Layer . . . . . . . . . . . . . . . . . . 27 126 8.1.2. Central Administration . . . . . . . . . . . . . . . 28 127 8.1.3. Hot Swap . . . . . . . . . . . . . . . . . . . . . . 28 128 8.1.4. Data Flow Information Models . . . . . . . . . . . . 29 129 8.1.5. L2 and L3 Integration . . . . . . . . . . . . . . . . 29 130 8.1.6. End-to-End Delivery . . . . . . . . . . . . . . . . . 29 131 8.1.7. Replacement for Proprietary Fieldbuses and Ethernet- 132 based Networks . . . . . . . . . . . . . . . . . . . 30 133 8.1.8. Deterministic vs Best-Effort Traffic . . . . . . . . 30 134 8.1.9. Deterministic Flows . . . . . . . . . . . . . . . . . 31 135 8.1.10. Unused Reserved Bandwidth . . . . . . . . . . . . . . 31 136 8.1.11. Interoperability . . . . . . . . . . . . . . . . . . 31 137 8.1.12. Cost Reductions . . . . . . . . . . . . . . . . . . . 31 138 8.1.13. Insufficiently Secure Devices . . . . . . . . . . . . 32 139 8.1.14. DetNet Network Size . . . . . . . . . . . . . . . . . 32 140 8.1.15. Multiple Hops . . . . . . . . . . . . . . . . . . . . 33 141 8.1.16. Level of Service . . . . . . . . . . . . . . . . . . 33 142 8.1.17. Bounded Latency . . . . . . . . . . . . . . . . . . . 33 143 8.1.18. Low Latency . . . . . . . . . . . . . . . . . . . . . 34 144 8.1.19. Bounded Jitter (Latency Variation) . . . . . . . . . 34 145 8.1.20. Symmetrical Path Delays . . . . . . . . . . . . . . . 34 146 8.1.21. Reliability and Availability . . . . . . . . . . . . 34 147 8.1.22. Redundant Paths . . . . . . . . . . . . . . . . . . . 35 148 8.1.23. Security Measures . . . . . . . . . . . . . . . . . . 35 149 8.2. Summary of Attack Types per Use Case Common Theme . . . . 35 150 8.3. Security Considerations for OAM Traffic . . . . . . . . . 38 151 9. DetNet Technology-Specific Threats . . . . . . . . . . . . . 38 152 9.1. IP . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 153 9.2. MPLS . . . . . . . . . . . . . . . . . . . . . . . . . . 40 154 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 41 155 11. Security Considerations . . . . . . . . . . . . . . . . . . . 41 156 12. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 41 157 13. Informative References . . . . . . . . . . . . . . . . . . . 42 158 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 45 160 1. Introduction 162 A deterministic network is one that can carry data flows for real- 163 time applications with extremely low data loss rates and bounded 164 latency. Deterministic networks have been successfully deployed in 165 real-time Operational Technology (OT) applications for some years. 166 However, such networks are typically isolated from external access, 167 and thus the security threat from external attackers is low. IETF 168 Deterministic Networking (DetNet, [RFC8655]) specifies a set of 169 technologies that enable creation of deterministic networks on IP- 170 based networks of potentially wide area (on the scale of a corporate 171 network) potentially bringing the OT network into contact with 172 Information Technology (IT) traffic and security threats that lie 173 outside of a tightly controlled and bounded area (such as the 174 internals of an aircraft). 176 These DetNet technologies have not previously been deployed together 177 on a wide area IP-based network, and thus can present security 178 considerations that may be new to IP-based wide area network 179 designers; this document provides insight into such system-level 180 security considerations. In addition, designers of DetNet components 181 (such as routers) face new security-related challenges in providing 182 DetNet services, for example maintaining reliable isolation between 183 traffic flows in an environment where IT traffic co-mingles with 184 critical reserved-bandwidth OT traffic; this document also examines 185 security implications internal to DetNet components. 187 Security is of particularly high importance in DetNet networks 188 because many of the use cases which are enabled by DetNet [RFC8578] 189 include control of physical devices (power grid components, 190 industrial controls, building controls) which can have high 191 operational costs for failure, and present potentially attractive 192 targets for cyber-attackers. 194 This situation is even more acute given that one of the goals of 195 DetNet is to provide a "converged network", i.e. one that includes 196 both IT traffic and OT traffic, thus exposing potentially sensitive 197 OT devices to attack in ways that were not previously common (usually 198 because they were under a separate control system or otherwise 199 isolated from the IT network, for example [ARINC664P7]). Security 200 considerations for OT networks are not a new area, and there are many 201 OT networks today that are connected to wide area networks or the 202 Internet; this document focuses on the issues that are specific to 203 the DetNet technologies and use cases. 205 Given the above considerations, securing a DetNet starts with a 206 scrupulously well-designed and well-managed engineered network 207 following industry best practices for security at both the data plane 208 and controller plane; this is the assumed starting point for the 209 considerations discussed herein. Such assumptions also depend on the 210 network components themselves upholding the security-related 211 properties that are to be assumed by DetNet system-level designers; 212 for example, the assumption that network traffic associated with a 213 given flow can never affect traffic associated with a different flow 214 is only true if the underlying components make it so. Such 215 properties, which may represent new challenges to component 216 designers, are also considered herein. 218 In this context we view the network design and management aspects of 219 network security as being primarily concerned with denial-of service 220 prevention by ensuring that DetNet traffic goes where it's supposed 221 to and that an external attacker can't inject traffic that disrupts 222 the DetNet's delivery timing assurance. The time-specific aspects of 223 DetNet security presented here take up where the design and 224 management aspects leave off. 226 The exact security requirements for any given DetNet network are 227 necessarily specific to the use cases handled by that network. Thus 228 the reader is assumed to be familiar with the specific security 229 requirements of their use cases, for example those outlined in the 230 DetNet Use Cases [RFC8578] and the Security Considerations sections 231 of the DetNet documents applicable to the network technologies in 232 use, for example [I-D.ietf-detnet-ip]). A general introduction to 233 the DetNet architecture can be found in [RFC8655] and it is also 234 recommended to be familiar with the DetNet Data Plane 235 [I-D.ietf-detnet-data-plane-framework] and Flow Information Model 236 [I-D.ietf-detnet-flow-information-model]. 238 The DetNet technologies include ways to: 240 o Assign data plane resources for DetNet flows in some or all of the 241 intermediate nodes (routers) along the path of the flow 243 o Provide explicit routes for DetNet flows that do not dynamically 244 change with the network topology in ways that affect the quality 245 of service received by the affected flow(s) 247 o Distribute data from DetNet flow packets over time and/or space to 248 ensure delivery of each packet's data' in spite of the loss of a 249 path 251 This document includes sections considering DetNet component design 252 as well as system design. The latter includes threat modeling and 253 analysis, threat impact and mitigation, and the association of 254 attacks with use cases (based on the Use Case Common Themes section 255 of the DetNet Use Cases [RFC8578]). 257 The structure of the threat model and threat analysis sections were 258 originally derived from [RFC7384], which also considers time-related 259 security considerations in IP networks. 261 2. Abbreviations and Terminology 263 IT Information Technology (the application of computers to 264 store, study, retrieve, transmit, and manipulate data or information, 265 often in the context of a business or other enterprise - [IT_DEF]). 267 OT Operational Technology (the hardware and software 268 dedicated to detecting or causing changes in physical processes 269 through direct monitoring and/or control of physical devices such as 270 valves, pumps, etc. - [OT_DEF]) 272 MITM Man in the Middle 274 Component A component of a DetNet system - used here to refer 275 to any hardware or software element of a DetNet network which 276 implements DetNet-specific functionality, for example all or part of 277 a router, switch, or end system. 279 Resource Segmentation Used as a more general form for Network 280 Segmentation (the act or practice of splitting a computer network 281 into subnetworks, each being a network segment - [RS_DEF]) 283 3. Security Considerations for DetNet Component Design 285 As noted above, DetNet provides resource allocation, explicit routes 286 and redundant path support. Each of these has associated security 287 implications, which are discussed in this section, in the context of 288 component design. Detection, reporting and appropriate action in the 289 case of packet arrival time violations are also discussed. 291 3.1. Resource Allocation 293 A DetNet system security designer relies on the premise that any 294 resources allocated to a resource-reserved (OT-type) flow are 295 inviolable, in other words there is no physical possibility within a 296 DetNet component that resources allocated to a given flow can be 297 compromised by any type of traffic in the network; this includes both 298 malicious traffic as well as inadvertent traffic such as might be 299 produced by a malfunctioning component, for example one made by a 300 different manufacturer. From a security standpoint, this is a 301 critical assumption, for example when designing against DOS attacks. 302 It is the responsibility of the component designer to ensure that 303 this condition is met; this implies protection against excess traffic 304 from adjacent flows, and against compromises to the resource 305 allocation/deallocation process. 307 3.2. Explicit Routes 309 The DetNet-specific purpose for constraining the network's ability to 310 re-route OT traffic is to maintain the specified service parameters 311 (such as upper and lower latency boundaries) for a given flow. For 312 example if the network were to re-route a flow (or some part of a 313 flow) based exclusively on statistical path usage metrics, or due to 314 malicious activity, it is possible that the new path would have a 315 latency that is outside the required latency bounds which were 316 designed into the original TE-designed path, thereby violating the 317 quality of service for the affected flow (or part of that flow). 319 However, it is acceptable for the network to re-route OT traffic in 320 such a way as to maintain the specified latency bounds (and any other 321 specified service properties) for any reason, for example in response 322 to a runtime component or path failure. From a security standpoint, 323 the system designer relies on the premise that the packets will be 324 delivered with the specified latency boundaries; thus any component 325 that is involved in controlling or implementing any change of the 326 initially TE-configured flow routes needs to prevent malicious or 327 accidental re-routing of OT flows that might adversely affect 328 delivering the traffic within the specified service parameters. 330 3.3. Redundant Path Support 332 The DetNet provision for redundant paths (PREOF) (as defined in the 333 DetNet Architecture [RFC8655]) provides the foundation for high 334 reliablity of a DetNet, by virtually eliminating packet loss (i.e. to 335 a degree which is implementation-dependent) through hitless redundant 336 packet delivery. (Note that PREOF is not defined for a DetNet IP 337 data plane). 339 It is the responsibility of the system designer to determine the 340 level of reliability required by their use case, and to specify 341 redundant paths sufficient to provide the desired level of 342 reliability (in as much as that reliability can be provided through 343 the use of redundant paths). It is the responsibility of the 344 component designer to ensure that the relevant PREOF operations are 345 executed reliably and securely. (However, note that not all PREOF 346 operations are necessarily implemented in every network; for example 347 a packet re-ordering function may not be necessary if the packets are 348 either not required to be in order, or if the ordering is performed 349 in some other part of the network.) 351 As noted in Section 7.2, Integrity Protection, there is a trust 352 relationship between the pair of devices that replicate and remove 353 packets, so it is the responsibility of the system designer to define 354 these relationships with the appropriate security considerations, and 355 the components must each uphold the security rights implied by these 356 relationships. 358 Ideally a redundant path could be specified from end to end of the 359 flow's path, however given that this is not always possible (as 360 described in [RFC8655]) the system designer will need to consider the 361 resulting end-to-end reliability and security resulting from any 362 given arrangment of network segments along the path, each of which 363 provides its individual PREOF implementation and thus its individual 364 level of reliabiilty and security. 366 At the data plane the implementation of PREOF depends on the correct 367 assignment and interpretation of packet sequence numbers, as well as 368 the actions taken based on them, such as elimination. Thus the 369 integrity of these values must be maintained by the component as they 370 are assigned by the DetNet Data Plane's Service sub-layer, and 371 transported by the Forwarding sub-layer. 373 3.4. Timing (or other) Violation Reporting 375 Another fundamental assumption of a secure DetNet is that in any case 376 in which an incoming packet arrives with any timing or bandwidth 377 violation, something can be done about it which doesn't cause damage 378 to the system. For example having the network shut down a link if a 379 packet arrives outside of its prescribed time window may serve the 380 attacker better than it serves the network. That means that the 381 component's data plane must be able to detect and act on a variety of 382 such violations, at least alerting the controller plane. Any action 383 apart from that needs to be carefully considered in the context of 384 the specific system. Some possible violations that warrant detection 385 include cases where a packet arrives: 387 o Outside of its prescribed time window 389 o Within its time window but with a compromised time stamp that 390 makes it appear that it is not within its window 392 o Exceeding the reserved flow bandwidth 394 Logging of such issues is unlikely to be adequate, since a delay in 395 response to the situation could result in material damage, for 396 example to mechanical devices controlled by the network. Given that 397 the data plane component probably has no knowledge of the use case of 398 the network, or its applications and end systems, it would seem 399 useful for a data plane component to allow the system designer to 400 configure its actions in the face of such violations. 402 Possible direct actions that may be taken at the data plane include 403 dropping the packet and/or shutting down the link; however if any 404 such actions are configured to be taken, the system designer must 405 ensure that such actions do not compromise the continued safe 406 operation of the system. For example, the controller plane should 407 mitigate in a timely fashion any potential adverse effect on 408 mechanical devices controlled by the network. 410 4. DetNet Security Considerations Compared With DiffServ Security 411 Considerations 413 DetNet is designed to be compatible with DiffServ [RFC2474] as 414 applied to IT traffic in the DetNet. DetNet also incorporates the 415 use of the 6-bit value of the DSCP field of the TOS field of the IP 416 header for flow identification for OT traffic, however the DetNet 417 interpretation of the DSCP value for OT traffic is not equivalent to 418 the PHB selection behavior as defined by DiffServ. 420 Thus security consideration for DetNet have some aspects in common 421 with DiffServ, in fact overlapping 100% with respect to IP IT 422 traffic. Security considerations for these aspects are part of the 423 existing literature on IP network security, specifically the Security 424 sections of [RFC2474] and [RFC2475]. However DetNet also introduces 425 timing and other considerations which are not present in DiffServ, so 426 the DiffServ security considerations are necessary but not sufficient 427 for DetNet. 429 In the case of DetNet OT traffic, the DSCP value, although 430 interpreted differently than in DiffServ, does contribute to 431 determination of the service provided to the packet. Thus in DetNet 432 there are similar consequences to DiffServ for lack of detection of, 433 or incorrect handling of, packets with mismarked DSCP values, and 434 thus many of the points made in the DiffServ draft Security 435 discussions are also relevant to DetNet OT traffic, though perhaps in 436 modified form. For example, in DetNet the effect of an undetected or 437 incorrectly handled maliciously mismarked DSCP field in an OT packet 438 is not identical to affecting that packet's PHB, since DetNet does 439 not use the PHB concept for OT traffic, but nonetheless the service 440 provided to the packet could be affected, so mitigation measures 441 analogous to those prescribed by DiffServ would be appropriate for 442 DetNet. For example, mismarked DSCP values should not cause failure 443 of network nodes, and any internal link that cannot be adequately 444 secured against modification of DSCP values should be treated as a 445 boundary link (and hence any arriving traffic on that link is treated 446 as if it were entering the domain at an ingress node). The remarks 447 in [RFC2474] regarding IPsec and Tunnelling Interactions are also 448 relevant (though this is not to say that other sections are less 449 relevant). 451 5. Security Threats 453 This section presents a threat model, and analyzes the possible 454 threats in a DetNet-enabled network. The threats considered in this 455 section are independent of any specific technologies used to 456 implement the DetNet; Section 9) considers attacks that are 457 associated with the DetNet technologies encompassed by 458 [I-D.ietf-detnet-data-plane-framework]. 460 We distinguish controller plane threats from data plane threats. The 461 attack surface may be the same, but the types of attacks as well as 462 the motivation behind them, are different. For example, a delay 463 attack is more relevant to data plane than to controller plane. 464 There is also a difference in terms of security solutions: the way 465 you secure the data plane is often different than the way you secure 466 the controller plane. 468 5.1. Threat Model 470 The threat model used in this memo is based on the threat model of 471 Section 3.1 of [RFC7384]. This model classifies attackers based on 472 two criteria: 474 o Internal vs. external: internal attackers either have access to a 475 trusted segment of the network or possess the encryption or 476 authentication keys. External attackers, on the other hand, do 477 not have the keys and have access only to the encrypted or 478 authenticated traffic. 480 o Man in the Middle (MITM) vs. packet injector: MITM attackers are 481 located in a position that allows interception and modification of 482 in-flight protocol packets, whereas a traffic injector can only 483 attack by generating protocol packets. 485 Care has also been taken to adhere to Section 5 of [RFC3552], both 486 with respect to which attacks are considered out-of-scope for this 487 document, but also which are considered to be the most common threats 488 (explored further in Section 5.2, Threat Analysis). Most of the 489 direct threats to DetNet are active attacks, but it is highly 490 suggested that DetNet application developers take appropriate 491 measures to protect the content of the DetNet flows from passive 492 attacks. 494 DetNet-Service, one of the service scenarios described in 495 [I-D.varga-detnet-service-model], is the case where a service 496 connects DetNet networking islands, i.e. two or more otherwise 497 independent DetNet network domains are connected via a link that is 498 not intrinsically part of either network. This implies that there 499 could be DetNet traffic flowing over a non-DetNet link, which may 500 provide an attacker with an advantageous opportunity to tamper with 501 DetNet traffic. The security properties of non-DetNet links are 502 outside of the scope of DetNet Security, but it should be noted that 503 use of non-DetNet services to interconnect DetNet networks merits 504 security analysis to ensure the integrity of the DetNet networks 505 involved. 507 5.2. Threat Analysis 509 5.2.1. Delay 511 An attacker can maliciously delay DetNet data flow traffic. By 512 delaying the traffic, the attacker can compromise the service of 513 applications that are sensitive to high delays or to high delay 514 variation. The delay may be constant or modulated. 516 5.2.2. DetNet Flow Modification or Spoofing 518 An attacker can modify some header fields of en route packets in a 519 way that causes the DetNet flow identification mechanisms to 520 misclassify the flow. Alternatively, the attacker can inject traffic 521 that is tailored to appear as if it belongs to a legitimate DetNet 522 flow. The potential consequence is that the DetNet flow resource 523 allocation cannot guarantee the performance that is expected when the 524 flow identification works correctly. 526 5.2.3. Resource Segmentation (Inter-segment Attack) 528 An attacker can inject traffic that will consume network resources 529 such that it affects DetNet flows. This can be performed using non- 530 DetNet traffic that indirectly affects DetNet traffic (hardware 531 resource exhaustion), or by using DetNet traffic from one DetNet flow 532 that directly affects traffic from different DetNet flows. 534 5.2.4. Packet Replication and Elimination 536 5.2.4.1. Replication: Increased Attack Surface 538 Redundancy is intended to increase the robustness and survivability 539 of DetNet flows, and replication over multiple paths can potentially 540 mitigate an attack that is limited to a single path. However, the 541 fact that packets are replicated over multiple paths increases the 542 attack surface of the network, i.e., there are more points in the 543 network that may be subject to attacks. 545 5.2.4.2. Replication-related Header Manipulation 547 An attacker can manipulate the replication-related header fields. 548 This capability opens the door for various types of attacks. For 549 example: 551 o Forward both replicas - malicious change of a packet SN (Sequence 552 Number) can cause both replicas of the packet to be forwarded. 553 Note that this attack has a similar outcome to a replay attack. 555 o Eliminate both replicas - SN manipulation can be used to cause 556 both replicas to be eliminated. In this case an attacker that has 557 access to a single path can cause packets from other paths to be 558 dropped, thus compromising some of the advantage of path 559 redundancy. 561 o Flow hijacking - an attacker can hijack a DetNet flow with access 562 to a single path by systematically replacing the SNs on the given 563 path with higher SN values. For example, an attacker can replace 564 every SN value S with a higher value S+C, where C is a constant 565 integer. Thus, the attacker creates a false illusion that the 566 attacked path has the lowest delay, causing all packets from other 567 paths to be eliminated in favor of the attacked path. Once the 568 flow from the compromised path is favored by the elminating 569 bridge, the flow is hijacked by the attacker. It is now posible 570 to either replace en route packets with malicious packets, or 571 simply injecting errors, causing the packets to be dropped at 572 their destination. 574 5.2.5. Path Choice 576 5.2.5.1. Path Manipulation 578 An attacker can maliciously change, add, or remove a path, thereby 579 affecting the corresponding DetNet flows that use the path. 581 5.2.5.2. Path Choice: Increased Attack Surface 583 One of the possible consequences of a path manipulation attack is an 584 increased attack surface. Thus, when the attack described in the 585 previous subsection is implemented, it may increase the potential of 586 other attacks to be performed. 588 5.2.6. Controller Plane 590 5.2.6.1. Control or Signaling Packet Modification 592 An attacker can maliciously modify en route control packets in order 593 to disrupt or manipulate the DetNet path/resource allocation. 595 5.2.6.2. Control or Signaling Packet Injection 597 An attacker can maliciously inject control packets in order to 598 disrupt or manipulate the DetNet path/resource allocation. 600 5.2.7. Scheduling or Shaping 602 5.2.7.1. Reconnaissance 604 A passive eavesdropper can identify DetNet flows and then gather 605 information about en route DetNet flows, e.g., the number of DetNet 606 flows, their bandwidths, their schedules, or other temporal 607 properties. The gathered information can later be used to invoke 608 other attacks on some or all of the flows. 610 Note that in some cases DetNet flows may be identified based on an 611 explicit DetNet header, but in some cases the flow identification may 612 be based on fields from the L3/L4 headers. If L3/L4 headers are 613 involved, for the purposes of this document we assume they are 614 encrypted and/or integrity-protected from external attackers. 616 5.2.8. Time Synchronization Mechanisms 618 An attacker can use any of the attacks described in [RFC7384] to 619 attack the synchronization protocol, thus affecting the DetNet 620 service. 622 5.3. Threat Summary 624 A summary of the attacks that were discussed in this section is 625 presented in Figure 1. For each attack, the table specifies the type 626 of attackers that may invoke the attack. In the context of this 627 summary, the distinction between internal and external attacks is 628 under the assumption that a corresponding security mechanism is being 629 used, and that the corresponding network equipment takes part in this 630 mechanism. 632 +-----------------------------------------+----+----+----+----+ 633 | Attack | Attacker Type | 634 | +---------+---------+ 635 | |Internal |External | 636 | |MITM|Inj.|MITM|Inj.| 637 +-----------------------------------------+----+----+----+----+ 638 |Delay attack | + | + | + | + | 639 +-----------------------------------------+----+----+----+----+ 640 |DetNet Flow Modification or Spoofing | + | + | | | 641 +-----------------------------------------+----+----+----+----+ 642 |Inter-segment Attack | + | + | | | 643 +-----------------------------------------+----+----+----+----+ 644 |Replication: Increased Attack Surface | + | + | + | + | 645 +-----------------------------------------+----+----+----+----+ 646 |Replication-related Header Manipulation | + | | | | 647 +-----------------------------------------+----+----+----+----+ 648 |Path Manipulation | + | + | | | 649 +-----------------------------------------+----+----+----+----+ 650 |Path Choice: Increased Attack Surface | + | + | + | + | 651 +-----------------------------------------+----+----+----+----+ 652 |Control or Signaling Packet Modification | + | | | | 653 +-----------------------------------------+----+----+----+----+ 654 |Control or Signaling Packet Injection | | + | | | 655 +-----------------------------------------+----+----+----+----+ 656 |Reconnaissance | + | | + | | 657 +-----------------------------------------+----+----+----+----+ 658 |Attacks on Time Sync Mechanisms | + | + | + | + | 659 +-----------------------------------------+----+----+----+----+ 661 Figure 1: Threat Analysis Summary 663 6. Security Threat Impacts 665 This section describes and rates the impact of the attacks described 666 in Section 5, Security Threats. In this section, the impacts as 667 described assume that the associated mitigation is not present or has 668 failed. Mitigations are discussed in Section 7, Security Threat 669 Mitigation. 671 In computer security, the impact (or consequence) of an incident can 672 be measured in loss of confidentiality, integrity or availability of 673 information. In the case of time sensitive networks, the impact of a 674 network exploit can also include failure or malfunction of mechanical 675 and/or other OT systems. 677 DetNet raises these stakes significantly for OT applications, 678 particularly those which may have been designed to run in an OT-only 679 environment and thus may not have been designed for security in an IT 680 environment with its associated devices, services and protocols. 682 The severity of various components of the impact of a successful 683 vulnerability exploit to use cases by industry is available in more 684 detail in the DetNet Use Cases [RFC8578]. Each of these use cases is 685 represented in the table below, including Pro Audio, Electrical 686 Utilities, Industrial M2M (split into two areas, M2M Data Gathering 687 and M2M Control Loop), and others. 689 Components of Impact (left column) include Criticality of Failure, 690 Effects of Failure, Recovery, and DetNet Functional Dependence. 691 Criticality of failure summarizes the seriousness of the impact. The 692 impact of a resulting failure can affect many different metrics that 693 vary greatly in scope and severity. In order to reduce the number of 694 variables, only the following were included: Financial, Health and 695 Safety, People well being (People WB), Affect on a single 696 organization, and affect on multiple organizations. Recovery 697 outlines how long it would take for an affected use case to get back 698 to its pre-failure state (Recovery time objective, RTO), and how much 699 of the original service would be lost in between the time of service 700 failure and recovery to original state (Recovery Point Objective, 701 RPO). DetNet dependence maps how much the following DetNet service 702 objectives contribute to impact of failure: Time dependency, data 703 integrity, source node integrity, availability, latency/jitter. 705 The scale of the Impact mappings is low, medium, and high. In some 706 use cases there may be a multitude of specific applications in which 707 DetNet is used. For simplicity this section attempts to average the 708 varied impacts of different applications. This section does not 709 address the overall risk of a certain impact which would require the 710 likelihood of a failure happening. 712 In practice any such ratings will vary from case to case; the ratings 713 shown here are given as examples. 715 Table, Part One (of Two) 716 +------------------+-----------------------------------------+-----+ 717 | | Pro A | Util | Bldg |Wire- | Cell |M2M |M2M | 718 | | | | | less | |Data |Ctrl | 719 +------------------+-----------------------------------------+-----+ 720 | Criticality | Med | Hi | Low | Med | Med | Med | Med | 721 +------------------+-----------------------------------------+-----+ 722 | Effects 723 +------------------+-----------------------------------------+-----+ 724 | Financial | Med | Hi | Med | Med | Low | Med | Med | 725 +------------------+-----------------------------------------+-----+ 726 | Health/Safety | Med | Hi | Hi | Med | Med | Med | Med | 727 +------------------+-----------------------------------------+-----+ 728 | People WB | Med | Hi | Hi | Low | Hi | Low | Low | 729 +------------------+-----------------------------------------+-----+ 730 | Effect 1 org | Hi | Hi | Med | Hi | Med | Med | Med | 731 +------------------+-----------------------------------------+-----+ 732 | Effect >1 org | Med | Hi | Low | Med | Med | Med | Med | 733 +------------------+-----------------------------------------+-----+ 734 |Recovery 735 +------------------+-----------------------------------------+-----+ 736 | Recov Time Obj | Med | Hi | Med | Hi | Hi | Hi | Hi | 737 +------------------+-----------------------------------------+-----+ 738 | Recov Point Obj | Med | Hi | Low | Med | Low | Hi | Hi | 739 +------------------+-----------------------------------------+-----+ 740 |DetNet Dependence 741 +------------------+-----------------------------------------+-----+ 742 | Time Dependency | Hi | Hi | Low | Hi | Med | Low | Hi | 743 +------------------+-----------------------------------------+-----+ 744 | Latency/Jitter | Hi | Hi | Med | Med | Low | Low | Hi | 745 +------------------+-----------------------------------------+-----+ 746 | Data Integrity | Hi | Hi | Med | Hi | Low | Hi | Low | 747 +------------------+-----------------------------------------+-----+ 748 | Src Node Integ | Hi | Hi | Med | Hi | Med | Hi | Hi | 749 +------------------+-----------------------------------------+-----+ 750 | Availability | Hi | Hi | Med | Hi | Low | Hi | Hi | 751 +------------------+-----------------------------------------+-----+ 753 Table, Part Two (of Two) 754 +------------------+--------------------------+ 755 | | Mining | Block | Network | 756 | | | Chain | Slicing | 757 +------------------+--------------------------+ 758 | Criticality | Hi | Med | Hi | 759 +------------------+--------------------------+ 760 | Effects 761 +------------------+--------------------------+ 762 | Financial | Hi | Hi | Hi | 763 +------------------+--------------------------+ 764 | Health/Safety | Hi | Low | Med | 765 +------------------+--------------------------+ 766 | People WB | Hi | Low | Med | 767 +------------------+--------------------------+ 768 | Effect 1 org | Hi | Hi | Hi | 769 +------------------+--------------------------+ 770 | Effect >1 org | Hi | Low | Hi | 771 +------------------+--------------------------+ 772 |Recovery 773 +------------------+--------------------------+ 774 | Recov Time Obj | Hi | Low | Hi | 775 +------------------+--------------------------+ 776 | Recov Point Obj | Hi | Low | Hi | 777 +------------------+--------------------------+ 778 |DetNet Dependence 779 +------------------+--------------------------+ 780 | Time Dependency | Hi | Low | Hi | 781 +------------------+--------------------------+ 782 | Latency/Jitter | Hi | Low | Hi | 783 +------------------+--------------------------+ 784 | Data Integrity | Hi | Hi | Hi | 785 +------------------+--------------------------+ 786 | Src Node Integ | Hi | Hi | Hi | 787 +------------------+--------------------------+ 788 | Availability | Hi | Hi | Hi | 789 +------------------+--------------------------+ 791 Figure 2: Impact of Attacks by Use Case Industry 793 The rest of this section will cover impact of the different groups in 794 more detail. 796 6.1. Delay-Attacks 798 6.1.1. Data Plane Delay Attacks 800 Note that 'delay attack' also includes the possibility of a 'negative 801 delay' or early arrival of a packet, or possibly adversely changing 802 the timestamp value. 804 Delayed messages in a DetNet link can result in the same behavior as 805 dropped messages in ordinary networks as the services attached to the 806 DetNet flow have strict deterministic requirements. 808 For a single path scenario, disruption is a real possibility, whereas 809 in a multipath scenario, large delays or instabilities in one DetNet 810 flow can lead to increased buffer and processor resources at the 811 eliminating router. 813 A data-plane delay attack on a system controlling substantial moving 814 devices, for example in industrial automation, can cause physical 815 damage. For example, if the network promises a bounded latency of 816 2ms for a flow, yet the machine receives it with 5ms latency, the 817 machine's control loop can become unstable. 819 6.1.2. Controller Plane Delay Attacks 821 In and of itself, this is not directly a threat to the DetNet 822 service, but the effects of delaying control messages can have quite 823 adverse effects later. 825 o Delayed tear-down can lead to resource leakage, which in turn can 826 result in failure to allocate new DetNet flows, finally giving 827 rise to a denial of service attack. 829 o Failure to deliver, or severely delaying, controller plane 830 messages adding an endpoint to a multicast-group will prevent the 831 new endpoint from receiving expected frames thus disrupting 832 expected behavior. 834 o Delaying messages removing an endpoint from a group can lead to 835 loss of privacy as the endpoint will continue to receive messages 836 even after it is supposedly removed. 838 6.2. Flow Modification and Spoofing 840 6.2.1. Flow Modification 842 If the contents of a packet header or body can be modified by the 843 attacker, this can cause the packet to be routed incorrectly or 844 dropped, or the payload to be corrupted or subtly modified. 846 6.2.2. Spoofing 848 6.2.2.1. Dataplane Spoofing 850 Spoofing dataplane messages can result in increased resource 851 consumptions on the routers throughout the network as it will 852 increase buffer usage and processor utilization. This can lead to 853 resource exhaustion and/or increased delay. 855 If the attacker manages to create valid headers, the false messages 856 can be forwarded through the network, using part of the allocated 857 bandwidth. This in turn can cause legitimate messages to be dropped 858 when the resource budget has been exhausted. 860 Finally, the endpoint will have to deal with invalid messages being 861 delivered to the endpoint instead of (or in addition to) a valid 862 message. 864 6.2.2.2. Controller Plane Spoofing 866 A successful controller plane spoofing-attack will potentionally have 867 adverse effects. It can do virtually anything from: 869 o modifying existing DetNet flows by changing the available 870 bandwidth 872 o add or remove endpoints from a DetNet flow 874 o drop DetNet flows completely 876 o falsely create new DetNet flows (exhaust the systems resources, or 877 to enable DetNet flows that are outside the Network Engineer's 878 control) 880 6.3. Segmentation Attacks (injection) 882 6.3.1. Data Plane Segmentation 884 Injection of false messages in a DetNet flow could lead to exhaustion 885 of the available bandwidth for that flow if the routers attribute 886 these false messages to that flow's budget. 888 In a multipath scenario, injected messages will cause increased 889 processor utilization in elimination routers. If enough paths are 890 subject to malicious injection, the legitimate messages can be 891 dropped. Likewise it can cause an increase in buffer usage. In 892 total, it will consume more resources in the routers than normal, 893 giving rise to a resource exhaustion attack on the routers. 895 If a DetNet flow is interrupted, the end application will be affected 896 by what is now a non-deterministic flow. 898 6.3.2. Controller Plane Segmentation 900 In a successful controller plane segmentation attack, control 901 messages are acted on by nodes in the network, unbeknownst to the 902 central controller or the network engineer. This has the potential 903 to: 905 o create new DetNet flows (exhausting resources) 907 o drop existing DetNet flows (denial of service) 909 o add end-stations to a multicast group (loss of privacy) 911 o remove end-stations from a multicast group (reduction of service) 913 o modify the DetNet flow attributes (affecting available bandwidth) 915 6.4. Replication and Elimination 917 The Replication and Elimination is relevant only to data plane 918 messages as controller plane messages are not subject to multipath 919 routing. 921 6.4.1. Increased Attack Surface 923 Covered briefly in Section 6.3, Segmentation Attacks. 925 6.4.2. Header Manipulation at Elimination Routers 927 Covered briefly in Section 6.3, Segmentation Attacks. 929 6.5. Control or Signaling Packet Modification 931 If control packets are subject to manipulation undetected, the 932 network can be severely compromised. 934 6.6. Control or Signaling Packet Injection 936 If an attacker can inject control packets undetected, the network can 937 be severely compromised. 939 6.7. Reconnaissance 941 Of all the attacks, this is one of the most difficult to detect and 942 counter. Often, an attacker will start out by observing the traffic 943 going through the network and use the knowledge gathered in this 944 phase to mount future attacks. 946 The attacker can, at their leisure, observe over time all aspects of 947 the messaging and signalling, learning the intent and purpose of all 948 traffic flows. At some later date, possibly at an important time in 949 an operational context, the attacker can launch a multi-faceted 950 attack, possibly in conjunction with some demand for ransom. 952 The flow-id in the header of the data plane messages gives an 953 attacker a very reliable identifier for DetNet traffic, and this 954 traffic has a high probability of going to lucrative targets. 956 Applications which are ported from a private OT network to the higher 957 visibility DetNet environment may need to be adapted to limit 958 distinctive flow properties that could make them susceptible to 959 reconnaissance. 961 6.8. Attacks on Time Sync Mechanisms 963 Attacks on time sync mechanisms are addressed in [RFC7384]. 965 6.9. Attacks on Path Choice 967 This is covered in part in Section 6.3, Segmentation Attacks, and as 968 with Replication and Elimination (Section 6.4), this is relevant for 969 DataPlane messages. 971 7. Security Threat Mitigation 973 This section describes a set of measures that can be taken to 974 mitigate the attacks described in Section 5, Security Threats. These 975 mitigations should be viewed as a toolset that includes several 976 different and diverse tools. Each application or system will 977 typically use a subset of these tools, based on a system-specific 978 threat analysis. 980 7.1. Path Redundancy 982 Description 984 A DetNet flow that can be forwarded simultaneously over multiple 985 paths. Path replication and elimination [RFC8655] provides 986 resiliency to dropped or delayed packets. This redundancy 987 improves the robustness to failures and to man-in-the-middle 988 attacks. Note: At the time of this writing, PREOF is not defined 989 for the IP data plane. 991 Related attacks 993 Path redundancy can be used to mitigate various man-in-the-middle 994 attacks, including attacks described in Section 5.2.1, 995 Section 5.2.2, Section 5.2.3, and Section 5.2.8. However it is 996 also possible that multiple paths may make it more difficult to 997 locate the source of a MITM attacker. 999 A delay modulation attack could result in extensively exercising 1000 parts of the code that wouldn't normally be extensively exercised 1001 and thus might expose flaws in the system that might otherwise not 1002 be exposed. 1004 7.2. Integrity Protection 1006 Description 1008 An integrity protection mechanism, such as a Hash-based Message 1009 Authentication Code (HMAC) can be used to mitigate modification 1010 attacks on IP packets. Integrity protection in the controller 1011 plane is discussed in Section 7.6. 1013 Packet Sequence Number Integrity Considerations 1015 The use of PREOF in a DetNet implementation implies the use of a 1016 sequence number for each packet. There is a trust relationship 1017 between the device that adds the sequence number and the device 1018 that removes the sequence number. The sequence number may be end- 1019 to-end source to destination, or may be added/deleted by network 1020 edge devices. The adder and remover(s) have the trust 1021 relationship because they are the ones that ensure that the 1022 sequence numbers are not modifiable. Between those two points, 1023 there may or may not be replication and elimination functions. 1024 The elimination functions must be able to see the sequence 1025 numbers. Therefore any encryption that is done between adders and 1026 removers must not obscure the sequence number. If the sequence 1027 removers and the eliminators are in the same physical device, it 1028 may be possible to obscure the sequence number, however that is a 1029 layer violation, and is not recommended practice. Note: At the 1030 time of this writing, PREOF is not defined for the IP data plane. 1032 Related attacks 1034 Integrity protection mitigates attacks related to modification and 1035 tampering, including the attacks described in Section 5.2.2 and 1036 Section 5.2.4. 1038 7.3. DetNet Node Authentication 1040 Description 1042 Source authentication verifies the authenticity of DetNet sources, 1043 enabling mitigation of spoofing attacks. Note that while 1044 integrity protection (Section 7.2) prevents intermediate nodes 1045 from modifying information, authentication can provide traffic 1046 origin verification, i.e. to verify that each packet in a DetNet 1047 flow is from a trusted source. Authentication may be implemented 1048 as part of ingress filtering, for example. 1050 Related attacks 1052 DetNet node authentication is used to mitigate attacks related to 1053 spoofing, including the attacks of Section 5.2.2, and 1054 Section 5.2.4. 1056 7.4. Dummy Traffic Insertion 1058 Description 1060 With some queueing methods such as [IEEE802.1Qch-2017] it is 1061 possible to introduce dummy traffic in order to regularize the 1062 timing of packet transmission. 1064 Related attacks 1066 Removing distinctive temporal properties of individual packets or 1067 flows can be used to mitigate against reconnaissance attacks 1068 Section 5.2.7. 1070 7.5. Encryption 1072 Description 1074 DetNet flows can in principle be forwarded in encrypted form at 1075 the DetNet layer, however, regarding encryption of IP headers see 1076 Section 9. 1078 DetNet nodes do not have any need to inspect the payload of any 1079 DetNet packets, making them data-agnostic. This means that end- 1080 to- end encryption at the application layer is an acceptable way 1081 to protect user data. 1083 Encryption can also be applied at the subnet layer, for example 1084 for Ethernet using MACSec, as noted in Section 9. 1086 Related attacks 1088 Encryption can be used to mitigate recon attacks (Section 5.2.7). 1089 However, for a DetNet network to give differentiated quality of 1090 service on a flow-by-flow basis, the network must be able to 1091 identify the flows individually. This implies that in a recon 1092 attack the attacker may also be able to track individual flows to 1093 learn more about the system. 1095 7.5.1. Encryption Considerations for DetNet 1097 Any compute time which is required for encryption and decryption 1098 processing ('crypto') must be included in the flow latency 1099 calculations. Thus, crypto algorithms used in a DetNet must have 1100 bounded worst-case execution times, and these values must be used in 1101 the latency calculations. 1103 Some crypto algorithms are symmetric in encode/decode time (such as 1104 AES) and others are asymmetric (such as public key algorithms). 1105 There are advantages and disadvantages to the use of either type in a 1106 given DetNet context. The discussion in this document relates to the 1107 timing implications of crypto for DetNet; it is assumed that 1108 integrity considerations are covered elsewhere in the literature. 1110 Asymmetrical crypto is typically not used in networks on a packet-by- 1111 packet basis due to its computational cost. For example, if only 1112 endpoint checks or checks at a small number of intermediate points 1113 are required, asymmetric crypto can be used to authenticate 1114 distribution or exchange of a secret symmetric crypto key; a 1115 successful check based on that key will provide traffic origin 1116 verification, as long as the key is kept secret by the participants. 1117 TLS and IKE (for IPsec) are examples of this for endpoint checks. 1119 However, if secret symmetrical keys are used for this purpose the key 1120 must be given to all relays, which increases the probability of a 1121 secret key being leaked. Also, if any relay is compromised or 1122 misbehaving it may inject traffic into the flow. 1124 Alternatively, asymmetric crypto can provide traffic origin 1125 verification at every intermediate node. For example, a DetNet flow 1126 can be associated with an (asymmetric) keypair, such that the private 1127 key is available to the source of the flow and the public key is 1128 distributed with the flow information, allowing verification at every 1129 node for every packet. However, this is more computationally 1130 expensive. 1132 In either case, origin verification also requires replay detection as 1133 part of the security protocol to prevent an attacker from recording 1134 and resending traffic, e.g., as a denial of service attack on flow 1135 forwarding resources. 1137 If crypto keys are to be regenerated over the duration of the flow 1138 then the time required to accomplish this must be accounted for in 1139 the latency calculations. 1141 7.6. Control and Signaling Message Protection 1143 Description 1145 Control and sigaling messages can be protected using 1146 authentication and integrity protection mechanisms. 1148 Related attacks 1150 These mechanisms can be used to mitigate various attacks on the 1151 controller plane, as described in Section 5.2.6, Section 5.2.8 and 1152 Section 5.2.5. 1154 7.7. Dynamic Performance Analytics 1156 Description 1158 The expectation is that the network will have a way to monitor to 1159 detect if timing guarantees are not being met, and a way to alert 1160 the controller plane in that event. Information about the network 1161 performance can be gathered in real-time in order to detect 1162 anomalies and unusual behavior that may be the symptom of a 1163 security attack. The gathered information can be based, for 1164 example, on per-flow counters, bandwidth measurement, and 1165 monitoring of packet arrival times. Unusual behavior or 1166 potentially malicious nodes can be reported to a management 1167 system, or can be used as a trigger for taking corrective actions. 1168 The information can be tracked by DetNet end systems and transit 1169 nodes, and exported to a management system, for example using 1170 YANG. 1172 Related attacks 1174 Performance analytics can be used to mitigate various attacks, 1175 including the ones described in Section 5.2.1 (Delay Attack), 1176 Section 5.2.3 (Resource Segmentation Attack), and Section 5.2.8 1177 (Time Sync Attack). 1179 For example, in the case of data plane delay attacks, one possible 1180 mitigation is to timestamp the data at the source, and timestamp 1181 it again at the destination, and if the resulting latency exceeds 1182 the promised bound, discard that data and warn the operator (and/ 1183 or enter a fail-safe mode). Note that DetNet specifies packet 1184 sequence numbering, however it does not specify use of packet 1185 timestamps, although they may be used by the underlying transport 1186 (for example TSN) to provide the service. 1188 7.8. Mitigation Summary 1190 The following table maps the attacks of Section 5, Security Threats, 1191 to the impacts of Section 6, Security Threat Impacts, and to the 1192 mitigations of the current section. Each row specifies an attack, 1193 the impact of this attack if it is successfully implemented, and 1194 possible mitigation methods. 1196 +----------------------+---------------------+---------------------+ 1197 | Attack | Impact | Mitigations | 1198 +----------------------+---------------------+---------------------+ 1199 |Delay Attack |-Non-deterministic |-Path redundancy | 1200 | | delay |-Performance | 1201 | |-Data disruption | analytics | 1202 | |-Increased resource | | 1203 | | consumption | | 1204 +----------------------+---------------------+---------------------+ 1205 |Reconnaissance |-Enabler for other |-Encryption | 1206 | | attacks |-Dummy traffic | 1207 | | | insertion | 1208 +----------------------+---------------------+---------------------+ 1209 |DetNet Flow Modificat-|-Increased resource |-Path redundancy | 1210 |ion or Spoofing | consumption |-Integrity protection| 1211 | |-Data disruption |-DetNet Node | 1212 | | | authentication | 1213 +----------------------+---------------------+---------------------+ 1214 |Inter-Segment Attack |-Increased resource |-Path redundancy | 1215 | | consumption |-Performance | 1216 | |-Data disruption | analytics | 1217 +----------------------+---------------------+---------------------+ 1218 |Replication: Increased|-All impacts of other|-Integrity protection| 1219 |attack surface | attacks |-DetNet Node | 1220 | | | authentication | 1221 +----------------------+---------------------+---------------------+ 1222 |Replication-related |-Non-deterministic |-Integrity protection| 1223 |Header Manipulation | delay |-DetNet Node | 1224 | |-Data disruption | authentication | 1225 +----------------------+---------------------+---------------------+ 1226 |Path Manipulation |-Enabler for other |-Control message | 1227 | | attacks | protection | 1228 +----------------------+---------------------+---------------------+ 1229 |Path Choice: Increased|-All impacts of other|-Control message | 1230 |Attack Surface | attacks | protection | 1231 +----------------------+---------------------+---------------------+ 1232 |Control or Signaling |-Increased resource |-Control message | 1233 |Packet Modification | consumption | protection | 1234 | |-Non-deterministic | | 1235 | | delay | | 1236 | |-Data disruption | | 1237 +----------------------+---------------------+---------------------+ 1238 |Control or Signaling |-Increased resource |-Control message | 1239 |Packet Injection | consumption | protection | 1240 | |-Non-deterministic | | 1241 | | delay | | 1242 | |-Data disruption | | 1243 +----------------------+---------------------+---------------------+ 1244 |Attacks on Time Sync |-Non-deterministic |-Path redundancy | 1245 |Mechanisms | delay |-Control message | 1246 | |-Increased resource | protection | 1247 | | consumption |-Performance | 1248 | |-Data disruption | analytics | 1249 +----------------------+---------------------+---------------------+ 1251 Figure 3: Mapping Attacks to Impact and Mitigations 1253 8. Association of Attacks to Use Cases 1255 Different attacks can have different impact and/or mitigation 1256 depending on the use case, so we would like to make this association 1257 in our analysis. However since there is a potentially unbounded list 1258 of use cases, we categorize the attacks with respect to the common 1259 themes of the use cases as identified in the Use Case Common Themes 1260 section of the DetNet Use Cases [RFC8578]. 1262 See also Figure 2 for a mapping of the impact of attacks per use case 1263 by industry. 1265 8.1. Association of Attacks to Use Case Common Themes 1267 In this section we review each theme and discuss the attacks that are 1268 applicable to that theme, as well as anything specific about the 1269 impact and mitigations for that attack with respect to that theme. 1270 The table Figure 5, Mapping Between Themes and Attacks, then provides 1271 a summary of the attacks that are applicable to each theme. 1273 8.1.1. Sub-Network Layer 1275 DetNet is expected to run over various transmission mediums, with 1276 Ethernet being the first identified. Attacks such as Delay or 1277 Reconnaissance might be implemented differently on a different 1278 transmission medium, however the impact on the DetNet as a whole 1279 would be essentially the same. We thus conclude that all attacks and 1280 impacts that would be applicable to DetNet over Ethernet (i.e. all 1281 those named in this document) would also be applicable to DetNet over 1282 other transmission mediums. 1284 With respect to mitigations, some methods are specific to the 1285 Ethernet medium, for example time-aware scheduling using 802.1Qbv 1286 [IEEE802.1Qbv-2015] can protect against excessive use of bandwidth at 1287 the ingress - for other mediums, other mitigations would have to be 1288 implemented to provide analogous protection. 1290 8.1.2. Central Administration 1292 A DetNet network can be controlled by a centralized network 1293 configuration and control system. Such a system may be in a single 1294 central location, or it may be distributed across multiple control 1295 entities that function together as a unified control system for the 1296 network. 1298 In this document we distinguish between attacks on the DetNet 1299 Controller plane vs. Data Plane. But is an attack affecting 1300 controller plane packets synonymous with an attack on the controller 1301 plane itself? For the purposes of this document let us consider an 1302 attack on the control system itself to be out of scope, and consider 1303 all attacks named in this document which are relevant to controller 1304 plane packets to be relevant to this theme, including Path 1305 Manipulation, Path Choice, Control Packet Modification or Injection, 1306 Reconaissance and Attacks on Time Sync Mechanisms. 1308 8.1.3. Hot Swap 1310 A DetNet network is not expected to be "plug and play" - it is 1311 expected that there is some centralized network configuration and 1312 control system. However, the ability to "hot swap" components (e.g. 1313 due to malfunction) is similar enough to "plug and play" that this 1314 kind of behavior may be expected in DetNet networks, depending on the 1315 implementation. 1317 An attack surface related to Hot Swap is that the DetNet network must 1318 at least consider input at runtime from devices that were not part of 1319 the initial configuration of the network. Even a "perfect" (or 1320 "hitless") replacement of a device at runtime would not necessarily 1321 be ideal, since presumably one would want to distinguish it from the 1322 original for OAM purposes (e.g. to report hot swap of a failed 1323 device). 1325 This implies that an attack such as Flow Modification, Spoofing or 1326 Inter-segment (which could introduce packets from a "new" device 1327 (i.e. one heretofore unknown on the network) could be used to exploit 1328 the need to consider such packets (as opposed to rejecting them out 1329 of hand as one would do if one did not have to consider introduction 1330 of a new device). 1332 Similarly if the network was designed to support runtime replacement 1333 of a clock device, then presence (or apparent presence) and thus 1334 consideration of packets from a new such device could affect the 1335 network, or the time sync of the network, for example by initiating a 1336 new Best Master Clock selection process. Thus attacks on time sync 1337 should be considered when designing hot swap type functionality (see 1338 [RFC7384]). 1340 8.1.4. Data Flow Information Models 1342 Data Flow YANG models specific to DetNet networks are specified by 1343 DetNet, and thus are 'new' and thus potentially present a new attack 1344 surface. 1346 8.1.5. L2 and L3 Integration 1348 A DetNet network integrates Layer 2 (bridged) networks (e.g. AVB/TSN 1349 LAN) and Layer 3 (routed) networks via the use of well-known 1350 protocols such as IP, MPLS Pseudowire, and Ethernet. 1352 There are no specific entries in the mapping table Figure 4, however 1353 that does not imply that there could be no relevant attacks related 1354 to L2-L3 integration. 1356 8.1.6. End-to-End Delivery 1358 Packets sent over DetNet are not to be dropped by the network due to 1359 congestion. (Packets may however intentionally be dropped for 1360 intended reasons, e.g. per security measures). 1362 A data plane attack may force packets to be dropped, for example a 1363 "long" Delay or Replication/Elimination or Flow Modification attack. 1365 The same result might be obtained by a controller plane attack, e.g. 1366 Path Manipulation or Signaling Packet Modification. 1368 It may be that such attacks are limited to Internal MITM attackers, 1369 but other possibilities should be considered. 1371 An attack may also cause packets that should not be delivered to be 1372 delivered, such as by forcing packets from one (e.g. replicated) path 1373 to be preferred over another path when they should not be 1374 (Replication attack), or by Flow Modification, or by Path Choice or 1375 Packet Injection. A Time Sync attack could cause a system that was 1376 expecting certain packets at certain times to accept unintended 1377 packets based on compromised system time or time windowing in the 1378 scheduler. 1380 8.1.7. Replacement for Proprietary Fieldbuses and Ethernet-based 1381 Networks 1383 There are many proprietary "field buses" used in today's industrial 1384 and other industries, as well as proprietary non-interoperable 1385 deterministic Ethernet-based networks. DetNet is intended to provide 1386 an open-standards-based alternative to such buses/networks. In cases 1387 where a DetNet intersects with such fieldbuses/networks or their 1388 protocols, such as by protocol emulation or access via a gateway, new 1389 attack surfaces can be opened. 1391 For example an Inter-Segment or Controller plane attack such as Path 1392 Manipulation, Path Choice or Control Packet Modification/Injection 1393 could be used to exploit commands specific to such a protocol, or 1394 that are interpreted differently by the different protocols or 1395 gateway. 1397 8.1.8. Deterministic vs Best-Effort Traffic 1399 Most of the themes described in this document address OT (reserved) 1400 DetNet flows - this item is intended to address issues related to IT 1401 traffic on a DetNet. 1403 DetNet is intended to support coexistence of time-sensitive 1404 operational (OT, deterministic) traffic and information (IT, "best 1405 effort") traffic on the same ("unified") network. 1407 With DetNet, this coexistance will become more common, and 1408 mitigations will need to be established. The fact that the IT 1409 traffic on a DetNet is limited to a corporate controlled network 1410 makes this a less difficult problem compared to being exposed to the 1411 open Internet, however this aspect of DetNet security should not be 1412 underestimated. 1414 An Inter-segment attack can flood the network with IT-type traffic 1415 with the intent of disrupting handling of IT traffic, and/or the goal 1416 of interfering with OT traffic. Presumably if the DetNet flow 1417 reservation and isolation of the DetNet is well-designed (better- 1418 designed than the attack) then interference with OT traffic should 1419 not result from an attack that floods the network with IT traffic. 1421 However the DetNet's handling of IT traffic may not (by design) be as 1422 resilient to DOS attack, and thus designers must be otherwise 1423 prepared to mitigate DOS attacks on IT traffic in a DetNet. 1425 8.1.9. Deterministic Flows 1427 Reserved bandwidth data flows (deterministic flows) must provide the 1428 allocated bandwidth, and must be isolated from each other. 1430 A Spoofing or Inter-segment attack which adds packet traffic to a 1431 bandwidth-reserved DetNet flow could cause that flow to occupy more 1432 bandwidth than it was allocated, resulting in interference with other 1433 DetNet flows. 1435 A Flow Modification or Spoofing or Header Manipulation or Control 1436 Packet Modification attack could cause packets from one flow to be 1437 directed to another flow, thus breaching isolation between the flows. 1439 8.1.10. Unused Reserved Bandwidth 1441 If bandwidth reservations are made for a DetNet flow but the 1442 associated bandwidth is not used at any point in time, that bandwidth 1443 is made available on the network for best-effort traffic. However, 1444 note that security considerations for best-effort traffic on a DetNet 1445 network is out of scope of the present document, provided that such 1446 an attack does not affect performance for DetNet OT traffic. 1448 8.1.11. Interoperability 1450 The DetNet network specifications are intended to enable an ecosystem 1451 in which multiple vendors can create interoperable products, thus 1452 promoting device diversity and potentially higher numbers of each 1453 device manufactured. 1455 Given that the DetNet specifications are unambiguously written and 1456 that the implementations are accurate, then this should not in and of 1457 itself cause a security concern; however, in the real world, it could 1458 be. The network operator can mitigate this through sufficient 1459 interoperability testing. 1461 8.1.12. Cost Reductions 1463 The DetNet network specifications are intended to enable an ecosystem 1464 in which multiple vendors can create interoperable products, thus 1465 promoting higher numbers of each device manufactured, promoting cost 1466 reduction and cost competition among vendors. 1468 This envisioned breadth of DetNet-enabled products is in general a 1469 positive factor, however implementation flaws in any individual 1470 component can present an attack surface. In addition, implementation 1471 differences between components from different vendors can result in 1472 attack surfaces (resulting from their interaction) which may not 1473 exist in any individual component. 1475 Network operators can mitigate such concerns through sufficient 1476 product and interoperability testing. 1478 8.1.13. Insufficiently Secure Devices 1480 The DetNet network specifications are intended to enable an ecosystem 1481 in which multiple vendors can create interoperable products, thus 1482 promoting device diversity and potentially higher numbers of each 1483 device manufactured. However this raises the possibility that a 1484 vendor might repurpose for DetNet applications a hardware or software 1485 component that was originally designed for operation in an isolated 1486 OT network, and thus may not have been designed to be sufficiently 1487 secure, or secure at all. Deployment of such a device on a DetNet 1488 network that is intended to be highly secure may present an attack 1489 surface. 1491 The DetNet network operator may need to take specific actions to 1492 protect such devices, such as implementing a dedicated security layer 1493 around the device. 1495 8.1.14. DetNet Network Size 1497 DetNet networks range in size from very small, e.g. inside a single 1498 industrial machine, to very large, for example a Utility Grid network 1499 spanning a whole country. 1501 The size of the network might be related to how the attack is 1502 introduced into the network, for example if the entire network is 1503 local, there is a threat that power can be cut to the entire network. 1504 If the network is large, perhaps only a part of the network is 1505 attacked. 1507 A Delay attack might be as relevant to a small network as to a large 1508 network, although the amount of delay might be different. 1510 Attacks sourced from IT traffic might be more likely in large 1511 networks, since more people might have access to the network, 1512 presenting a larger attack surface. Similarly Path Manipulation, 1513 Path Choice and Time Sync attacks seem more likely relevant to large 1514 networks. 1516 8.1.15. Multiple Hops 1518 Large DetNet networks (e.g. a Utility Grid network) may involve many 1519 "hops" over various kinds of links for example radio repeaters, 1520 microwave links, fiber optic links, etc. 1522 An attack that takes advantage of flaws (or even normal operation) in 1523 the device drivers for the various links (through internal knowledge 1524 of how the individual driver or firmware operates) could take 1525 proportionately greater advantage of this topology. 1527 It is also possible that this DetNet topology will not be in as 1528 common use as other more homogeneous topologies so there may be more 1529 opportunity for attackers to exploit software and/or protocol flaws 1530 in the implementations which have not been tested through extensive 1531 use, particularly in the case of early adopters. 1533 Of the attacks we have defined, the ones identified in Section 8.1.14 1534 as germane to large networks are the most relevant. 1536 8.1.16. Level of Service 1538 A DetNet is expected to provide means to configure the network that 1539 include querying network path latency, requesting bounded latency for 1540 a given DetNet flow, requesting worst case maximum and/or minimum 1541 latency for a given path or DetNet flow, and so on. It is an 1542 expected case that the network cannot provide a given requested 1543 service level. In such cases the network control system should reply 1544 that the requested service level is not available (as opposed to 1545 accepting the parameter but then not delivering the desired 1546 behavior). 1548 Controller plane attacks such as Signaling Packet Modification and 1549 Injection could be used to modify or create control traffic that 1550 could interfere with the process of a user requesting a level of 1551 service and/or the network's reply. 1553 Reconnaissance could be used to characterize flows and perhaps target 1554 specific flows for attack via the controller plane as noted in 1555 Section 6.7. 1557 8.1.17. Bounded Latency 1559 DetNet provides the expectation of guaranteed bounded latency. 1561 Delay attacks can cause packets to miss their agreed-upon latency 1562 boundaries. 1564 Time Sync attacks can corrupt the system's time reference, resulting 1565 in missed latency deadlines (with respect to the "correct" time 1566 reference). 1568 8.1.18. Low Latency 1570 Applications may require "extremely low latency" however depending on 1571 the application these may mean very different latency values; for 1572 example "low latency" across a Utility grid network is on a different 1573 time scale than "low latency" in a motor control loop in a small 1574 machine. The intent is that the mechanisms for specifying desired 1575 latency include wide ranges, and that architecturally there is 1576 nothing to prevent arbitrarily low latencies from being implemented 1577 in a given network. 1579 Attacks on the controller plane (as described in the Level of Service 1580 theme Section 8.1.16) and Delay and Time attacks (as described in the 1581 Bounded Latency theme Section 8.1.17) both apply here. 1583 8.1.19. Bounded Jitter (Latency Variation) 1585 DetNet is expected to provide bounded jitter (packet to packet 1586 latency variation). 1588 Delay attacks can cause packets to vary in their arrival times, 1589 resulting in packet to packet latency variation, thereby violating 1590 the jitter specification. 1592 8.1.20. Symmetrical Path Delays 1594 Some applications would like to specify that the transit delay time 1595 values be equal for both the transmit and return paths. 1597 Delay attacks can cause path delays to materially differ between 1598 paths. 1600 Time Sync attacks can corrupt the system's time reference, resulting 1601 in path delays that may be perceived to be different (with respect to 1602 the "correct" time reference) even if they are not materially 1603 different. 1605 8.1.21. Reliability and Availability 1607 DetNet based systems are expected to be implemented with essentially 1608 arbitrarily high availability (for example 99.9999% up time, or even 1609 12 nines). The intent is that the DetNet designs should not make any 1610 assumptions about the level of reliability and availability that may 1611 be required of a given system, and should define parameters for 1612 communicating these kinds of metrics within the network. 1614 Any attack on the system, of any type, can affect its overall 1615 reliability and availability, thus in the mapping table Figure 4 we 1616 have marked every attack. Since every DetNet depends to a greater or 1617 lesser degree on reliability and availability, this essentially means 1618 that all networks have to mitigate all attacks, which to a greater or 1619 lesser degree defeats the purpose of associating attacks with use 1620 cases. It also underscores the difficulty of designing "extremely 1621 high reliability" networks. 1623 8.1.22. Redundant Paths 1625 DetNet based systems are expected to be implemented with essentially 1626 arbitrarily high reliability/availability. A strategy used by DetNet 1627 for providing such extraordinarily high levels of reliability is to 1628 provide redundant paths that can be seamlessly switched between, all 1629 the while maintaining the required performance of that system. 1631 Replication-related attacks are by definition applicable here. 1632 Controller plane attacks can also interfere with the configuration of 1633 redundant paths. 1635 8.1.23. Security Measures 1637 A DetNet network must be made secure against devices failures, 1638 attackers, misbehaving devices, and so on. Does the threat affect 1639 such security measures themselves, e.g. by attacking SW designed to 1640 protect against device failure? 1642 This is TBD, thus there are no specific entries in the mapping table 1643 Figure 4, however that does not imply that there could be no relevant 1644 attacks. 1646 8.2. Summary of Attack Types per Use Case Common Theme 1648 The List of Attacks table Figure 4 lists the attacks of Section 5, 1649 Security Threats, assigning a number to each type of attack. That 1650 number is then used as a short form identifier for the attack in 1651 Figure 5, Mapping Between Themes and Attacks. 1653 +--+----------------------------------------+----------------------+ 1654 | | Attack | Section | 1655 +--+----------------------------------------+----------------------+ 1656 | 1|Delay Attack | Section 3.2.1 | 1657 +--+----------------------------------------+----------------------+ 1658 | 2|DetNet Flow Modification or Spoofing | Section 3.2.2 | 1659 +--+----------------------------------------+----------------------+ 1660 | 3|Inter-Segment Attack | Section 3.2.3 | 1661 +--+----------------------------------------+----------------------+ 1662 | 4|Replication: Increased attack surface | Section 3.2.4.1 | 1663 +--+----------------------------------------+----------------------+ 1664 | 5|Replication-related Header Manipulation | Section 3.2.4.2 | 1665 +--+----------------------------------------+----------------------+ 1666 | 6|Path Manipulation | Section 3.2.5.1 | 1667 +--+----------------------------------------+----------------------+ 1668 | 7|Path Choice: Increased Attack Surface | Section 3.2.5.2 | 1669 +--+----------------------------------------+----------------------+ 1670 | 8|Control or Signaling Packet Modification| Section 3.2.6.1 | 1671 +--+----------------------------------------+----------------------+ 1672 | 9|Control or Signaling Packet Injection | Section 3.2.6.2 | 1673 +--+----------------------------------------+----------------------+ 1674 |10|Reconnaissance | Section 3.2.7 | 1675 +--+----------------------------------------+----------------------+ 1676 |11|Attacks on Time Sync Mechanisms | Section 3.2.8 | 1677 +--+----------------------------------------+----------------------+ 1679 Figure 4: List of Attacks 1681 The Mapping Between Themes and Attacks table Figure 5 maps the use 1682 case themes of [RFC8578] (as also enumerated in this document) to the 1683 attacks of Figure 4. Each row specifies a theme, and the attacks 1684 relevant to this theme are marked with a '+'. 1686 +----------------------------+--------------------------------+ 1687 | Theme | Attack | 1688 | +--+--+--+--+--+--+--+--+--+--+--+ 1689 | | 1| 2| 3| 4| 5| 6| 7| 8| 9|10|11| 1690 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1691 |Network Layer - AVB/TSN Eth.| +| +| +| +| +| +| +| +| +| +| +| 1692 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1693 |Central Administration | | | | | | +| +| +| +| +| +| 1694 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1695 |Hot Swap | | +| +| | | | | | | | +| 1696 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1697 |Data Flow Information Models| | | | | | | | | | | | 1698 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1699 |L2 and L3 Integration | | | | | | | | | | | | 1700 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1701 |End-to-end Delivery | +| +| +| +| +| +| +| +| +| | +| 1702 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1703 |Proprietary Deterministic | | | +| | | +| +| +| +| | | 1704 |Ethernet Networks | | | | | | | | | | | | 1705 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1706 |Replacement for Proprietary | | | +| | | +| +| +| +| | | 1707 |Fieldbuses | | | | | | | | | | | | 1708 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1709 |Deterministic vs. Best- | | | +| | | | | | | | | 1710 |Effort Traffic | | | | | | | | | | | | 1711 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1712 |Deterministic Flows | | +| +| | +| +| | +| | | | 1713 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1714 |Unused Reserved Bandwidth | | +| +| | | | | +| +| | | 1715 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1716 |Interoperability | | | | | | | | | | | | 1717 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1718 |Cost Reductions | | | | | | | | | | | | 1719 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1720 |Insufficiently Secure | | | | | | | | | | | | 1721 |Devices | | | | | | | | | | | | 1722 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1723 |DetNet Network Size | +| | | | | +| +| | | | +| 1724 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1725 |Multiple Hops | +| +| | | | +| +| | | | +| 1726 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1727 |Level of Service | | | | | | | | +| +| +| | 1728 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1729 |Bounded Latency | +| | | | | | | | | | +| 1730 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1731 |Low Latency | +| | | | | | | +| +| +| +| 1732 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1733 |Bounded Jitter | +| | | | | | | | | | | 1734 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1735 |Symmetric Path Delays | +| | | | | | | | | | +| 1736 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1737 |Reliability and Availability| +| +| +| +| +| +| +| +| +| +| +| 1738 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1739 |Redundant Paths | | | | +| +| | | +| +| | | 1740 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1741 |Security Measures | | | | | | | | | | | | 1742 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1744 Figure 5: Mapping Between Themes and Attacks 1746 8.3. Security Considerations for OAM Traffic 1748 This section considers DetNet-specific security considerations for 1749 packet traffic that is generated and transmitted over a DetNet as 1750 part of OAM (Operations, Administration, and Maintenance). For the 1751 purposes of this discussion, OAM traffic falls into one of two basic 1752 types: 1754 o OAM traffic generated by the network itself. The additional 1755 bandwidth required for such packets is added by the network 1756 administration, presumably transparent to the customer. Security 1757 considerations for such traffic are not DetNet-specific (apart 1758 from such traffic being subject to the same DetNet-specific 1759 security considerations as any other DetNet data flow) and are 1760 thus not covered in this document. 1762 o OAM traffic generated by the customer. From a DetNet security 1763 point of view, DetNet security considerations for such traffic are 1764 exactly the same as for any other customer data flows. 1766 Thus OAM traffic presents no additional (i.e. OAM-specific) DetNet 1767 security considerations. 1769 9. DetNet Technology-Specific Threats 1771 Section 5, Security Threats, described threats which are independent 1772 of a DetNet implementation. This section considers threats 1773 specifically related to the IP- and MPLS-specific aspects of DetNet 1774 implementations. 1776 The primary security considerations for the data plane specifically 1777 are to maintain the integrity of the data and the delivery of the 1778 associated DetNet service traversing the DetNet network. 1780 The primary relevant differences between IP and MPLS implementations 1781 are in flow identification and OAM methodologies. 1783 As noted in [RFC8655], DetNet operates at the IP layer 1784 ([I-D.ietf-detnet-ip]) and delivers service over sub-layer 1785 technologies such as MPLS ([I-D.ietf-detnet-mpls]) and IEEE 802.1 1786 Time-Sensitive Networking (TSN) ([I-D.ietf-detnet-ip-over-tsn]). 1787 Application flows can be protected through whatever means are 1788 provided by the layer and sub-layer technologies. For example, 1789 technology-specific encryption may be used, such as that provided by 1790 IPSec [RFC4301] for IP flows and/or by an underlying sub-net using 1791 MACSec [IEEE802.1AE-2018] for IP over Ethernet (Layer-2) flows. 1793 However, if the DetNet nodes cannot decrypt IPsec traffic, IPSec may 1794 not be a valid option; this is because the DetNet IP Data Plane 1795 identifies flows via a 6-tuple that consists of two IP addresses, the 1796 transport protocol ID, two transport protocol port numbers and the 1797 DSCP in the IP header. When IPsec is used, the transport header is 1798 encrypted and the next protocol ID is an IPsec protocol, usually ESP, 1799 and not a transport protocol (e.g., neither TCP nor UDP, etc.) 1800 leaving only three components of the 6-tuple, which are the two IP 1801 addresses and the DSCP, which are in general not sufficient to 1802 identify a DetNet flow. 1804 Sections below discuss threats specific to IP and MPLS in more 1805 detail. 1807 9.1. IP 1809 The IP protocol has a long history of security considerations and 1810 architectural protection mechanisms. From a data plane perspective 1811 DetNet does not add or modify any IP header information, so the 1812 carriage of DetNet traffic over an IP data plane does not introduce 1813 any new security issues that were not there before, apart from those 1814 already described in the data-plane-independent threats section 1815 Section 5, Security Threats. 1817 Thus the security considerations for a DetNet based on an IP data 1818 plane are purely inherited from the rich IP Security literature and 1819 code/application base, and the data-plane-independent section of this 1820 document. 1822 Maintaining security for IP segments of a DetNet may be more 1823 challenging than for the MPLS segments of the network, given that the 1824 IP segments of the network may reach the edges of the network, which 1825 are more likely to involve interaction with potentially malevolent 1826 outside actors. Conversely MPLS is inherently more secure than IP 1827 since it is internal to routers and it is well-known how to protect 1828 it from outside influence. 1830 Another way to look at DetNet IP security is to consider it in the 1831 light of VPN security; as an industry we have a lot of experience 1832 with VPNs running through networks with other VPNs, it is well known 1833 how to secure the network for that. However for a DetNet we have the 1834 additional subtlety that any possible interaction of one packet with 1835 another can have a potentially deleterious effect on the time 1836 properties of the flows. So the network must provide sufficient 1837 isolation between flows, for example by protecting the forwarding 1838 bandwidth and related resources so that they are available to detnet 1839 traffic, by whatever means are appropriate for that network's data 1840 plane. 1842 In a VPN, bandwidth is generally guaranteed over a period of time, 1843 whereas in DetNet it is not aggregated over time. This implies that 1844 any VPN-type protection mechanism must also maintain the DetNet 1845 timing constraints. 1847 9.2. MPLS 1849 An MPLS network carrying DetNet traffic is expected to be a "well- 1850 managed" network. Given that this is the case, it is difficult for 1851 an attacker to pass a raw MPLS encoded packet into a network because 1852 operators have considerable experience at excluding such packets at 1853 the network boundaries, as well as excluding MPLS packets being 1854 inserted through the use of a tunnel. 1856 MPLS security is discussed extensively in [RFC5920] ("Security 1857 Framework for MPLS and GMPLS Networks") to which the reader is 1858 referred. 1860 [RFC6941] builds on [RFC5920] by providing additional security 1861 considerations that are applicable to the MPLS-TP extensions 1862 appropriate to the MPLS Transport Profile [RFC5921], and thus to the 1863 operation of DetNet over some types of MPLS network. 1865 [RFC5921] introduces to MPLS new Operations, Administration, and 1866 Maintenance (OAM) capabilities, a transport-oriented path protection 1867 mechanism, and strong emphasis on static provisioning supported by 1868 network management systems. 1870 The operation of DetNet over an MPLS network is modeled on the 1871 operation of multi-segment pseudowires (MS-PW). Thus for guidance on 1872 securing the DetNet elements of DetNet over MPLS the reader is 1873 referred to the MS-PW security mechanisms as defined in [RFC4447], 1874 [RFC3931], [RFC3985], [RFC6073], and [RFC6478]. 1876 Having attended to the conventional aspects of network security it is 1877 necessary to attend to the dynamic aspects. The closest experience 1878 that the IETF has with securing protocols that are sensitive to 1879 manipulation of delay are the two way time transfer protocols (TWTT), 1880 which are NTP [RFC5905] and Precision Time Protocol [IEEE1588]. The 1881 security requirements for these are described in [RFC7384]. 1883 One particular problem that has been observed in operational tests of 1884 TWTT protocols is the ability for two closely but not completely 1885 synchronized flows to beat and cause a sudden phase hit to one of the 1886 flows. This can be mitigated by the careful use of a scheduling 1887 system in the underlying packet transport. 1889 Further consideration of protection against dynamic attacks is work 1890 in progress. 1892 10. IANA Considerations 1894 This memo includes no requests from IANA. 1896 11. Security Considerations 1898 The security considerations of DetNet networks are presented 1899 throughout this document. 1901 12. Contributors 1903 The Editor would like to recognize the contributions of the following 1904 individuals to this draft. 1906 Andrew J. Hacker (MistIQ Technologies, Inc) 1907 Harrisburg, PA, USA 1908 email ajhacker@mistiqtech.com, 1909 web http://www.mistiqtech.com 1911 Subir Das (Applied Communication Sciences) 1912 150 Mount Airy Road, Basking Ridge 1913 New Jersey, 07920, USA 1914 email sdas@appcomsci.com 1916 John Dowdell (Airbus Defence and Space) 1917 Celtic Springs, Newport, NP10 8FZ, United Kingdom 1918 email john.dowdell.ietf@gmail.com 1920 Henrik Austad (SINTEF Digital) 1921 Klaebuveien 153, Trondheim, 7037, Norway 1922 email henrik@austad.us 1924 Norman Finn 1925 email nfinn@nfinnconsulting.com 1927 Stewart Bryant 1928 Futurewei Technologies 1929 email: stewart.bryant@gmail.com 1931 David Black 1932 Dell EMC 1933 176 South Street, Hopkinton, MA 01748, USA 1934 email: david.black@dell.com 1936 Carsten Bormann 1938 13. Informative References 1940 [ARINC664P7] 1941 ARINC, "ARINC 664 Aircraft Data Network, Part 7, Avionics 1942 Full-Duplex Switched Ethernet Network", 2009. 1944 [I-D.ietf-detnet-data-plane-framework] 1945 Varga, B., Farkas, J., Berger, L., Malis, A., and S. 1946 Bryant, "DetNet Data Plane Framework", draft-ietf-detnet- 1947 data-plane-framework-06 (work in progress), May 2020. 1949 [I-D.ietf-detnet-flow-information-model] 1950 Varga, B., Farkas, J., Cummings, R., Jiang, Y., and D. 1951 Fedyk, "DetNet Flow Information Model", draft-ietf-detnet- 1952 flow-information-model-10 (work in progress), May 2020. 1954 [I-D.ietf-detnet-ip] 1955 Varga, B., Farkas, J., Berger, L., Fedyk, D., and S. 1956 Bryant, "DetNet Data Plane: IP", draft-ietf-detnet-ip-07 1957 (work in progress), July 2020. 1959 [I-D.ietf-detnet-ip-over-tsn] 1960 Varga, B., Farkas, J., Malis, A., and S. Bryant, "DetNet 1961 Data Plane: IP over IEEE 802.1 Time Sensitive Networking 1962 (TSN)", draft-ietf-detnet-ip-over-tsn-03 (work in 1963 progress), June 2020. 1965 [I-D.ietf-detnet-mpls] 1966 Varga, B., Farkas, J., Berger, L., Malis, A., Bryant, S., 1967 and J. Korhonen, "DetNet Data Plane: MPLS", draft-ietf- 1968 detnet-mpls-10 (work in progress), July 2020. 1970 [I-D.varga-detnet-service-model] 1971 Varga, B. and J. Farkas, "DetNet Service Model", draft- 1972 varga-detnet-service-model-02 (work in progress), May 1973 2017. 1975 [IEEE1588] 1976 IEEE, "IEEE 1588 Standard for a Precision Clock 1977 Synchronization Protocol for Networked Measurement and 1978 Control Systems Version 2", 2008. 1980 [IEEE802.1AE-2018] 1981 IEEE Standards Association, "IEEE Std 802.1AE-2018 MAC 1982 Security (MACsec)", 2018, 1983 . 1985 [IEEE802.1Qbv-2015] 1986 IEEE Standards Association, "IEEE Standard for Local and 1987 metropolitan area networks -- Bridges and Bridged Networks 1988 - Amendment 25: Enhancements for Scheduled Traffic", 2015, 1989 . 1991 [IEEE802.1Qch-2017] 1992 IEEE Standards Association, "IEEE Standard for Local and 1993 metropolitan area networks--Bridges and Bridged Networks-- 1994 Amendment 29: Cyclic Queuing and Forwarding", 2017, 1995 . 1997 [IT_DEF] Wikipedia, "IT Definition", 2020, 1998 . 2000 [OT_DEF] Wikipedia, "OT Definition", 2020, 2001 . 2003 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 2004 "Definition of the Differentiated Services Field (DS 2005 Field) in the IPv4 and IPv6 Headers", RFC 2474, 2006 DOI 10.17487/RFC2474, December 1998, 2007 . 2009 [RFC2475] Blake, S., Black, D., Carlson, M., Davies, E., Wang, Z., 2010 and W. Weiss, "An Architecture for Differentiated 2011 Services", RFC 2475, DOI 10.17487/RFC2475, December 1998, 2012 . 2014 [RFC3552] Rescorla, E. and B. Korver, "Guidelines for Writing RFC 2015 Text on Security Considerations", BCP 72, RFC 3552, 2016 DOI 10.17487/RFC3552, July 2003, 2017 . 2019 [RFC3931] Lau, J., Ed., Townsley, M., Ed., and I. Goyret, Ed., 2020 "Layer Two Tunneling Protocol - Version 3 (L2TPv3)", 2021 RFC 3931, DOI 10.17487/RFC3931, March 2005, 2022 . 2024 [RFC3985] Bryant, S., Ed. and P. Pate, Ed., "Pseudo Wire Emulation 2025 Edge-to-Edge (PWE3) Architecture", RFC 3985, 2026 DOI 10.17487/RFC3985, March 2005, 2027 . 2029 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 2030 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 2031 December 2005, . 2033 [RFC4447] Martini, L., Ed., Rosen, E., El-Aawar, N., Smith, T., and 2034 G. Heron, "Pseudowire Setup and Maintenance Using the 2035 Label Distribution Protocol (LDP)", RFC 4447, 2036 DOI 10.17487/RFC4447, April 2006, 2037 . 2039 [RFC5905] Mills, D., Martin, J., Ed., Burbank, J., and W. Kasch, 2040 "Network Time Protocol Version 4: Protocol and Algorithms 2041 Specification", RFC 5905, DOI 10.17487/RFC5905, June 2010, 2042 . 2044 [RFC5920] Fang, L., Ed., "Security Framework for MPLS and GMPLS 2045 Networks", RFC 5920, DOI 10.17487/RFC5920, July 2010, 2046 . 2048 [RFC5921] Bocci, M., Ed., Bryant, S., Ed., Frost, D., Ed., Levrau, 2049 L., and L. Berger, "A Framework for MPLS in Transport 2050 Networks", RFC 5921, DOI 10.17487/RFC5921, July 2010, 2051 . 2053 [RFC6073] Martini, L., Metz, C., Nadeau, T., Bocci, M., and M. 2054 Aissaoui, "Segmented Pseudowire", RFC 6073, 2055 DOI 10.17487/RFC6073, January 2011, 2056 . 2058 [RFC6478] Martini, L., Swallow, G., Heron, G., and M. Bocci, 2059 "Pseudowire Status for Static Pseudowires", RFC 6478, 2060 DOI 10.17487/RFC6478, May 2012, 2061 . 2063 [RFC6941] Fang, L., Ed., Niven-Jenkins, B., Ed., Mansfield, S., Ed., 2064 and R. Graveman, Ed., "MPLS Transport Profile (MPLS-TP) 2065 Security Framework", RFC 6941, DOI 10.17487/RFC6941, April 2066 2013, . 2068 [RFC7384] Mizrahi, T., "Security Requirements of Time Protocols in 2069 Packet Switched Networks", RFC 7384, DOI 10.17487/RFC7384, 2070 October 2014, . 2072 [RFC8578] Grossman, E., Ed., "Deterministic Networking Use Cases", 2073 RFC 8578, DOI 10.17487/RFC8578, May 2019, 2074 . 2076 [RFC8655] Finn, N., Thubert, P., Varga, B., and J. Farkas, 2077 "Deterministic Networking Architecture", RFC 8655, 2078 DOI 10.17487/RFC8655, October 2019, 2079 . 2081 [RS_DEF] Wikipedia, "RS Definition", 2020, 2082 . 2084 Authors' Addresses 2086 Tal Mizrahi 2087 Huawei Network.IO Innovation Lab 2089 Email: tal.mizrahi.phd@gmail.com 2090 Ethan Grossman (editor) 2091 Dolby Laboratories, Inc. 2092 1275 Market Street 2093 San Francisco, CA 94103 2094 USA 2096 Phone: +1 415 645 4726 2097 Email: ethan.grossman@dolby.com 2098 URI: http://www.dolby.com