idnits 2.17.1 draft-ietf-detnet-security-10.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (May 30, 2020) is 1426 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for intended status: Informational ---------------------------------------------------------------------------- == Outdated reference: A later version (-14) exists of draft-ietf-detnet-flow-information-model-10 == Outdated reference: A later version (-07) exists of draft-ietf-detnet-ip-06 == Outdated reference: A later version (-07) exists of draft-ietf-detnet-ip-over-tsn-02 == Outdated reference: A later version (-13) exists of draft-ietf-detnet-mpls-06 -- Obsolete informational reference (is this intentional?): RFC 4447 (Obsoleted by RFC 8077) Summary: 0 errors (**), 0 flaws (~~), 5 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: December 1, 2020 DOLBY 6 May 30, 2020 8 Deterministic Networking (DetNet) Security Considerations 9 draft-ietf-detnet-security-10 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 implies that in 16 addition to the best practice security measures taken for any 17 mission-critical network, additional security measures may be needed 18 whose purpose is exclusively to secure the intended operation of 19 these novel service properties. 21 Designers of DetNet components (such as routers) that provide these 22 unique DetNet properties have the responsibility to uphold certain 23 security-related properties that can be assumed by DetNet system- 24 level designers. For example, the assumption that network traffic 25 associated with a given flow can never affect traffic associated with 26 a different flow is only true if the underlying components make it 27 so. 29 This document addresses DetNet-specific security considerations from 30 the perspective of both the DetNet component designer and the DetNet 31 system-level designer. It is assumed that both classes of reader are 32 already familiar with network security best practices for the data 33 plane technologies underlying a given DetNet implementation. 34 Component-level considerations include isolation of data flows from 35 each other, ingress filtering, and detection and reporting of packet 36 arrival time violations. System-level considerations include a 37 threat model and a taxonomy of relevant attacks, including their 38 potential impacts and mitigations. 40 A given DetNet may require securing only certain aspects of DetNet 41 performance, for example extremely low data loss rates but not 42 necessarily bounded latency. Therefore this document provides an 43 association of threats against various use cases by industry, and 44 also against the individual service properties as defined in the 45 DetNet Use Cases. 47 This document also addresses common DetNet security considerations 48 related to the IP and MPLS data plane technologies (the first to be 49 identified as supported by DetNet), thereby complementing the 50 Security Considerations sections of the various DetNet Data Plane 51 (and other) DetNet documents. 53 Status of This Memo 55 This Internet-Draft is submitted in full conformance with the 56 provisions of BCP 78 and BCP 79. 58 Internet-Drafts are working documents of the Internet Engineering 59 Task Force (IETF). Note that other groups may also distribute 60 working documents as Internet-Drafts. The list of current Internet- 61 Drafts is at https://datatracker.ietf.org/drafts/current/. 63 Internet-Drafts are draft documents valid for a maximum of six months 64 and may be updated, replaced, or obsoleted by other documents at any 65 time. It is inappropriate to use Internet-Drafts as reference 66 material or to cite them other than as "work in progress." 68 This Internet-Draft will expire on December 1, 2020. 70 Copyright Notice 72 Copyright (c) 2020 IETF Trust and the persons identified as the 73 document authors. All rights reserved. 75 This document is subject to BCP 78 and the IETF Trust's Legal 76 Provisions Relating to IETF Documents 77 (https://trustee.ietf.org/license-info) in effect on the date of 78 publication of this document. Please review these documents 79 carefully, as they describe your rights and restrictions with respect 80 to this document. Code Components extracted from this document must 81 include Simplified BSD License text as described in Section 4.e of 82 the Trust Legal Provisions and are provided without warranty as 83 described in the Simplified BSD License. 85 Table of Contents 87 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 88 2. Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . 6 89 3. Security Considerations for DetNet Component Design . . . . . 7 90 3.1. Resource Allocation . . . . . . . . . . . . . . . . . . . 7 91 3.2. Explicit Routes . . . . . . . . . . . . . . . . . . . . . 7 92 3.3. Redundant Path Support . . . . . . . . . . . . . . . . . 8 93 3.4. Timing Violation Reporting . . . . . . . . . . . . . . . 9 94 4. DetNet Security Considerations Compared With DiffServ 95 Security Considerations . . . . . . . . . . . . . . . . . . . 9 96 5. Security Threats . . . . . . . . . . . . . . . . . . . . . . 10 97 5.1. Threat Model . . . . . . . . . . . . . . . . . . . . . . 10 98 5.2. Threat Analysis . . . . . . . . . . . . . . . . . . . . . 11 99 5.2.1. Delay . . . . . . . . . . . . . . . . . . . . . . . . 11 100 5.2.1.1. Delay Attack . . . . . . . . . . . . . . . . . . 11 101 5.2.2. DetNet Flow Modification or Spoofing . . . . . . . . 11 102 5.2.3. Resource Segmentation or Slicing . . . . . . . . . . 11 103 5.2.3.1. Inter-segment Attack . . . . . . . . . . . . . . 11 104 5.2.4. Packet Replication and Elimination . . . . . . . . . 12 105 5.2.4.1. Replication: Increased Attack Surface . . . . . . 12 106 5.2.4.2. Replication-related Header Manipulation . . . . . 12 107 5.2.5. Path Choice . . . . . . . . . . . . . . . . . . . . . 12 108 5.2.5.1. Path Manipulation . . . . . . . . . . . . . . . . 12 109 5.2.5.2. Path Choice: Increased Attack Surface . . . . . . 13 110 5.2.6. Controller Plane . . . . . . . . . . . . . . . . . . 13 111 5.2.6.1. Control or Signaling Packet Modification . . . . 13 112 5.2.6.2. Control or Signaling Packet Injection . . . . . . 13 113 5.2.7. Scheduling or Shaping . . . . . . . . . . . . . . . . 13 114 5.2.7.1. Reconnaissance . . . . . . . . . . . . . . . . . 13 115 5.2.8. Time Synchronization Mechanisms . . . . . . . . . . . 13 116 5.3. Threat Summary . . . . . . . . . . . . . . . . . . . . . 13 117 6. Security Threat Impacts . . . . . . . . . . . . . . . . . . . 14 118 6.1. Delay-Attacks . . . . . . . . . . . . . . . . . . . . . . 17 119 6.1.1. Data Plane Delay Attacks . . . . . . . . . . . . . . 17 120 6.1.2. Controller Plane Delay Attacks . . . . . . . . . . . 18 121 6.2. Flow Modification and Spoofing . . . . . . . . . . . . . 18 122 6.2.1. Flow Modification . . . . . . . . . . . . . . . . . . 18 123 6.2.2. Spoofing . . . . . . . . . . . . . . . . . . . . . . 18 124 6.2.2.1. Dataplane Spoofing . . . . . . . . . . . . . . . 18 125 6.2.2.2. Controller Plane Spoofing . . . . . . . . . . . . 19 126 6.3. Segmentation Attacks (injection) . . . . . . . . . . . . 19 127 6.3.1. Data Plane Segmentation . . . . . . . . . . . . . . . 19 128 6.3.2. Controller Plane Segmentation . . . . . . . . . . . . 19 129 6.4. Replication and Elimination . . . . . . . . . . . . . . . 20 130 6.4.1. Increased Attack Surface . . . . . . . . . . . . . . 20 131 6.4.2. Header Manipulation at Elimination Routers . . . . . 20 132 6.5. Control or Signaling Packet Modification . . . . . . . . 20 133 6.6. Control or Signaling Packet Injection . . . . . . . . . . 20 134 6.7. Reconnaissance . . . . . . . . . . . . . . . . . . . . . 20 135 6.8. Attacks on Time Sync Mechanisms . . . . . . . . . . . . . 21 136 6.9. Attacks on Path Choice . . . . . . . . . . . . . . . . . 21 137 7. Security Threat Mitigation . . . . . . . . . . . . . . . . . 21 138 7.1. Path Redundancy . . . . . . . . . . . . . . . . . . . . . 21 139 7.2. Integrity Protection . . . . . . . . . . . . . . . . . . 21 140 7.3. DetNet Node Authentication . . . . . . . . . . . . . . . 22 141 7.4. Dummy Traffic Insertion . . . . . . . . . . . . . . . . . 23 142 7.5. Encryption . . . . . . . . . . . . . . . . . . . . . . . 23 143 7.5.1. Encryption Considerations for DetNet . . . . . . . . 23 144 7.6. Control and Signaling Message Protection . . . . . . . . 24 145 7.7. Dynamic Performance Analytics . . . . . . . . . . . . . . 25 146 7.8. Mitigation Summary . . . . . . . . . . . . . . . . . . . 25 147 8. Association of Attacks to Use Cases . . . . . . . . . . . . . 27 148 8.1. Use Cases by Common Themes . . . . . . . . . . . . . . . 27 149 8.1.1. Sub-Network Layer . . . . . . . . . . . . . . . . . . 27 150 8.1.2. Central Administration . . . . . . . . . . . . . . . 28 151 8.1.3. Hot Swap . . . . . . . . . . . . . . . . . . . . . . 28 152 8.1.4. Data Flow Information Models . . . . . . . . . . . . 29 153 8.1.5. L2 and L3 Integration . . . . . . . . . . . . . . . . 29 154 8.1.6. End-to-End Delivery . . . . . . . . . . . . . . . . . 29 155 8.1.7. Proprietary Deterministic Ethernet Networks . . . . . 30 156 8.1.8. Replacement for Proprietary Fieldbuses . . . . . . . 30 157 8.1.9. Deterministic vs Best-Effort Traffic . . . . . . . . 30 158 8.1.10. Deterministic Flows . . . . . . . . . . . . . . . . . 31 159 8.1.11. Unused Reserved Bandwidth . . . . . . . . . . . . . . 31 160 8.1.12. Interoperability . . . . . . . . . . . . . . . . . . 31 161 8.1.13. Cost Reductions . . . . . . . . . . . . . . . . . . . 32 162 8.1.14. Insufficiently Secure Devices . . . . . . . . . . . . 32 163 8.1.15. DetNet Network Size . . . . . . . . . . . . . . . . . 32 164 8.1.16. Multiple Hops . . . . . . . . . . . . . . . . . . . . 33 165 8.1.17. Level of Service . . . . . . . . . . . . . . . . . . 33 166 8.1.18. Bounded Latency . . . . . . . . . . . . . . . . . . . 33 167 8.1.19. Low Latency . . . . . . . . . . . . . . . . . . . . . 34 168 8.1.20. Bounded Jitter (Latency Variation) . . . . . . . . . 34 169 8.1.21. Symmetrical Path Delays . . . . . . . . . . . . . . . 34 170 8.1.22. Reliability and Availability . . . . . . . . . . . . 34 171 8.1.23. Redundant Paths . . . . . . . . . . . . . . . . . . . 35 172 8.1.24. Security Measures . . . . . . . . . . . . . . . . . . 35 173 8.2. Attack Types by Use Case Common Theme . . . . . . . . . . 35 174 8.3. Security Considerations for OAM Traffic . . . . . . . . . 38 175 9. DetNet Technology-Specific Threats . . . . . . . . . . . . . 38 176 9.1. IP . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 177 9.2. MPLS . . . . . . . . . . . . . . . . . . . . . . . . . . 40 178 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 41 179 11. Security Considerations . . . . . . . . . . . . . . . . . . . 41 180 12. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 41 181 13. Informative References . . . . . . . . . . . . . . . . . . . 42 182 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 45 184 1. Introduction 186 A deterministic network is one that can carry data flows for real- 187 time applications with extremely low data loss rates and bounded 188 latency. Deterministic networks have been successfully deployed in 189 real-time operational technology (OT) applications for some years. 190 However, such networks are typically isolated from external access, 191 and thus the security threat from external attackers is low. IETF 192 Deterministic Networking (DetNet) specifies a set of technologies 193 that enable creation of deterministic networks on IP-based networks 194 of potentially wide area (on the scale of a corporate network) 195 potentially bringing the OT network into contact with Information 196 Technology (IT) traffic and security threats that lie outside of a 197 tightly controlled and bounded area (such as the internals of an 198 aircraft). 200 These DetNet technologies have not previously been deployed together 201 on a wide area IP-based network, and thus can present security 202 considerations that may be new to IP-based wide area network 203 designers; this document provides insight into such system-level 204 security considerations. In addition, designers of DetNet components 205 (such as routers) face new security-related challenges in providing 206 DetNet services, for example maintaining reliable isolation between 207 traffic flows in an environment where IT traffic co-mingles with 208 critical reserved-bandwidth OT traffic; this document also examines 209 security implications internal to DetNet components. 211 Security is of particularly high importance in DetNet networks 212 because many of the use cases which are enabled by DetNet [RFC8578] 213 include control of physical devices (power grid components, 214 industrial controls, building controls) which can have high 215 operational costs for failure, and present potentially attractive 216 targets for cyber-attackers. 218 This situation is even more acute given that one of the goals of 219 DetNet is to provide a "converged network", i.e. one that includes 220 both IT traffic and OT traffic, thus exposing potentially sensitive 221 OT devices to attack in ways that were not previously common (usually 222 because they were under a separate control system or otherwise 223 isolated from the IT network, for example [ARINC664P7]). Security 224 considerations for OT networks are not a new area, and there are many 225 OT networks today that are connected to wide area networks or the 226 Internet; this document focuses on the issues that are specific to 227 the DetNet technologies and use cases. 229 Given the above considerations, securing a DetNet starts with a 230 scrupulously well-designed and well-managed engineered network 231 following industry best practices for security at both the data plane 232 and controller plane; this is the assumed starting point for the 233 considerations discussed herein. Such assumptions also depend on the 234 network components themselves upholding the security-related 235 properties that are to be assumed by DetNet system-level designers; 236 for example, the assumption that network traffic associated with a 237 given flow can never affect traffic associated with a different flow 238 is only true if the underlying components make it so. Such 239 properties, which may represent new challenges to component 240 designers, are also considered herein. 242 In this context we view the network design and management aspects of 243 network security as being primarily concerned with denial-of service 244 prevention by ensuring that DetNet traffic goes where it's supposed 245 to and that an external attacker can't inject traffic that disrupts 246 the DetNet's delivery timing assurance. The time-specific aspects of 247 DetNet security presented here take up where the design and 248 management aspects leave off. 250 The exact security requirements for any given DetNet network are 251 necessarily specific to the use cases handled by that network. Thus 252 the reader is assumed to be familiar with the specific security 253 requirements of their use cases, for example those outlined in the 254 DetNet Use Cases [RFC8578] and the Security Considerations sections 255 of the DetNet documents applicable to the network technologies in 256 use, for example [I-D.ietf-detnet-ip]). A general introduction to 257 the DetNet architecture can be found in [RFC8655] and it is also 258 recommended to be familiar with the Data Plane model 259 [I-D.ietf-detnet-data-plane-framework] and Flow Information Model 260 [I-D.ietf-detnet-flow-information-model]. 262 The DetNet technologies include ways to: 264 o Assign data plane resources for DetNet flows in some or all of the 265 intermediate nodes (routers) along the path of the flow 267 o Provide explicit routes for DetNet flows that do not dynamically 268 change with the network topology in ways that affect the quality 269 of service received by the affected flow(s) 271 o Distribute data from DetNet flow packets over time and/or space to 272 ensure delivery of each packet's data' in spite of the loss of a 273 path 275 This document includes sections considering DetNet component design 276 as well as system design. The latter include threat modeling and 277 analysis, threat impact and mitigation, and the association of 278 attacks with use cases (based on the Use Case Common Themes section 279 of the DetNet Use Cases). 281 The structure of the threat model and threat analysis sections were 282 originally derived from [RFC7384], which also considers time-related 283 security considerations in IP networks. 285 2. Abbreviations 287 IT Information technology (the application of computers to 288 store, study, retrieve, transmit, and manipulate data or information, 289 often in the context of a business or other enterprise - Wikipedia). 291 OT Operational Technology (the hardware and software 292 dedicated to detecting or causing changes in physical processes 293 through direct monitoring and/or control of physical devices such as 294 valves, pumps, etc. - Wikipedia) 296 MITM Man in the Middle 298 3. Security Considerations for DetNet Component Design 300 As noted above, DetNet provides resource allocation, explicit routes 301 and redundant path support. Each of these have associated security 302 implications, which are discussed in this section, in the context of 303 component design. Detection, reporting and appropriate action in the 304 case of packet arrival time violations are also discussed. 306 3.1. Resource Allocation 308 A DetNet system security designer relies on the premise that any 309 resources allocated to a resource-reserved (OT-type) flow are 310 inviolable, in other words there is no physical possibility within a 311 DetNet component that resources allocated to a given flow can be 312 compromised by any type of traffic in the network; this includes both 313 malicious traffic as well as inadvertent traffic such as might be 314 produced by a malfunctioning component, for example one made by a 315 different manufacturer. From a security standpoint, this is a 316 critical assumption, for example when designing against DOS attacks. 317 It is the responsibility of the component designer to ensure that 318 this condition is met; this implies protection against excess traffic 319 from adjacent flows, and against compromises to the resource 320 allocation/deallocation process. 322 3.2. Explicit Routes 324 The DetNet-specific purpose for constraining the network's ability to 325 re-route OT traffic is to maintain the specified service parameters 326 (such as upper and lower latency boundaries) for a given flow. For 327 example if the network were to re-route a flow (or some part of a 328 flow) based exclusively on statistical path usage metrics, or due to 329 malicious activity, it is possible that the new path would have a 330 latency that is outside the required latency bounds which were 331 designed into the original TE-designed path, thereby violating the 332 quality of service for the affected flow (or part of that flow). 333 (However, is acceptable for the network to re-route OT traffic in 334 such a way as to maintain the specified latency bounds (and any other 335 specified service properties) for any reason, for example in response 336 to a runtime component or path failure). From a security standpoint, 337 the system designer relies on the premise that the packets will be 338 delivered with the specified latency boundaries; thus any component 339 that is involved in controlling or implementing any change of the 340 initially TE-configured flow routes needs to prevent malicious or 341 accidental re-routing of OT flows that might adversely affect 342 delivering the traffic within the specified service parameters. 344 3.3. Redundant Path Support 346 The DetNet provision for redundant paths (PREOF) (as defined in the 347 DetNet Architecture [RFC8655]) provides the foundation for high 348 reliablity of a DetNet, by virtually eliminating packet loss (i.e. to 349 a degree which is implementation-dependent) through hitless redundant 350 packet delivery. (Note that PREOF is not defined for a DetNet IP 351 data plane). 353 It is the responsibility of the system designer to determine the 354 level of reliability required by their use case, and to specify 355 redundant paths sufficient to provide the desired level of 356 reliability (in as much as that reliability can be provided through 357 the use of redundant paths). It is the responsibility of the 358 component designer to ensure that the relevant PREOF operations are 359 executed reliably and securely. (However, note that not all PREOF 360 operations are necessarily implemented in every network; for example 361 a packet re-ordering function may not be necessary if the packets are 362 either not required to be in order, or if the ordering is performed 363 in some other part of the network.) 365 As noted in Section 7.2, Packet Sequence Number Integrity 366 Considerations, there is a trust relationship between the pair of 367 devices that replicate and remove packets, so it is the 368 responsibility of the system designer to define these relationships 369 with the appropriate security considerations, and the components must 370 each uphold the security rights implied by these relationships. 372 Ideally a redundant path could be specified from end to end of the 373 flow's path, however given that this is not always possible (as 374 described in [RFC8655]) the system designer will need to consider the 375 resulting end-to-end reliability and security resulting from any 376 given arrangment of network segments along the path, each of which 377 provides its individual PREOF implementation and thus its individual 378 level of reliabiilty and security. 380 At the data plane the implementation of PREOF depends on the correct 381 assignment and interpretation of packet sequence numbers, as well as 382 the actions taken based on them, such as elimination. Thus the 383 integrity of these values must be maintained by the component as they 384 are assigned by the DetNet data plane's Service sub-layer, and 385 transported by the Forwarding sub-layer. 387 3.4. Timing Violation Reporting 389 Another fundamental assumption of a secure DetNet is that in any case 390 in which an incoming packet arrives outside of its prescribed time 391 window or exceeding the reserved flow bandwidth, something can be 392 done about it. That means that the component's data plane must be 393 able to detect such cases, then at least alert the control plane, 394 and/or drop the packet, and/or shut down the link if violations 395 persist. Logging of such issues may not be adequate, since a delay 396 in response to the situation could result in material damage, for 397 example to mechanical devices controlled by the network. 399 4. DetNet Security Considerations Compared With DiffServ Security 400 Considerations 402 DetNet is designed to be compatible with DiffServ [RFC2474] as 403 applied to IT traffic in the DetNet. DetNet also incorporates the 404 use of the 6-bit value of the DSCP field of the TOS field of the IP 405 header for flow identification for OT traffic, however the DetNet 406 interpretation of the DSCP value for OT traffic is not equivalent to 407 the PHB selection behavior as defined by DiffServ. 409 Thus security consideration for DetNet have some aspects in common 410 with DiffServ, in fact overlapping 100% with respect to IP IT 411 traffic. Security considerations for these aspects are part of the 412 existing literature on IP network security, specifically the Security 413 sections of [RFC2474] and [RFC2475]. However DetNet also introduce 414 timing and other considerations which are not present in DiffServ, so 415 the DiffServ security considerations are necessary but not sufficient 416 for DetNet. 418 In the case of DetNet OT traffic, the DSCP value, although 419 interpreted differently than in DiffServ, does contribute to 420 determination of the service provided to the packet. Thus in DetNet 421 there are similar consequences to DiffServ for lack of detection of, 422 or incorrect handling of, packets with mismarked DSCP values, and 423 thus many of the points made in the DiffServ draft Security 424 discussions are also relevant to DetNet OT traffic, though perhaps in 425 modified form. For example, in DetNet the effect of an undetected or 426 incorrectly handled maliciously mismarked DSCP field in an OT packet 427 is not identical to affecting that packet's PHB, since DetNet does 428 not use the PHB concept for OT traffic, but nonetheless the service 429 provided to the packet could be affected, so mitigation measures 430 analogous to those prescribed by DiffServ would be appropriate for 431 DetNet. For example, mismarked DSCP values should not cause failure 432 of network nodes, and any internal link that cannot be adequately 433 secured against modification of DSCP values should be treated as a 434 boundary link (and hence any arriving traffic on that link is treated 435 as if it were entering the domain at an ingress node). The remarks 436 in [RFC2474] regarding IPsec and Tunnelling Interactions are also 437 relevant (though this is not to say that other sections are less 438 relevant). 440 5. Security Threats 442 This section presents a threat model, and analyzes the possible 443 threats in a DetNet-enabled network. The threats considered in this 444 section are independent of any specific technologies used to 445 implement the DetNet; Section 9) considers attacks that are 446 associated with the DetNet technologies encompassed by 447 [I-D.ietf-detnet-data-plane-framework]. 449 We distinguish controller plane threats from data plane threats. The 450 attack surface may be the same, but the types of attacks as well as 451 the motivation behind them, are different. For example, a delay 452 attack is more relevant to data plane than to controller plane. 453 There is also a difference in terms of security solutions: the way 454 you secure the data plane is often different than the way you secure 455 the controller plane. 457 5.1. Threat Model 459 The threat model used in this memo is based on the threat model of 460 Section 3.1 of [RFC7384]. This model classifies attackers based on 461 two criteria: 463 o Internal vs. external: internal attackers either have access to a 464 trusted segment of the network or possess the encryption or 465 authentication keys. External attackers, on the other hand, do 466 not have the keys and have access only to the encrypted or 467 authenticated traffic. 469 o Man in the Middle (MITM) vs. packet injector: MITM attackers are 470 located in a position that allows interception and modification of 471 in-flight protocol packets, whereas a traffic injector can only 472 attack by generating protocol packets. 474 Care has also been taken to adhere to Section 5 of [RFC3552], both 475 with respect to which attacks are considered out-of-scope for this 476 document, but also which are considered to be the most common threats 477 (explored further in Section 5.2 (Threat Analysis). Most of the 478 direct threats to DetNet are active attacks, but it is highly 479 suggested that DetNet application developers take appropriate 480 measures to protect the content of the DetNet flows from passive 481 attacks. 483 DetNet-Service, one of the service scenarios described in 484 [I-D.varga-detnet-service-model], is the case where a service 485 connects DetNet networking islands, i.e. two or more otherwise 486 independent DetNet network domains are connected via a link that is 487 not intrinsically part of either network. This implies that there 488 could be DetNet traffic flowing over a non-DetNet link, which may 489 provide an attacker with an advantageous opportunity to tamper with 490 DetNet traffic. The security properties of non-DetNet links are 491 outside of the scope of DetNet Security, but it should be noted that 492 use of non-DetNet services to interconnect DetNet networks merits 493 security analysis to ensure the integrity of the DetNet networks 494 involved. 496 5.2. Threat Analysis 498 5.2.1. Delay 500 5.2.1.1. Delay Attack 502 An attacker can maliciously delay DetNet data flow traffic. By 503 delaying the traffic, the attacker can compromise the service of 504 applications that are sensitive to high delays or to high delay 505 variation. The delay may be constant or modulated. 507 5.2.2. DetNet Flow Modification or Spoofing 509 An attacker can modify some header fields of en route packets in a 510 way that causes the DetNet flow identification mechanisms to 511 misclassify the flow. Alternatively, the attacker can inject traffic 512 that is tailored to appear as if it belongs to a legitimate DetNet 513 flow. The potential consequence is that the DetNet flow resource 514 allocation cannot guarantee the performance that is expected when the 515 flow identification works correctly. 517 5.2.3. Resource Segmentation or Slicing 519 5.2.3.1. Inter-segment Attack 521 An attacker can inject traffic that will consume network resources 522 such that it affects DetNet flows. This can be performed using non- 523 DetNet traffic that indirectly affects DetNet traffic (hardware 524 resource exhaustion), or by using DetNet traffic from one DetNet flow 525 that directly affects traffic from different DetNet flows. 527 5.2.4. Packet Replication and Elimination 529 5.2.4.1. Replication: Increased Attack Surface 531 Redundancy is intended to increase the robustness and survivability 532 of DetNet flows, and replication over multiple paths can potentially 533 mitigate an attack that is limited to a single path. However, the 534 fact that packets are replicated over multiple paths increases the 535 attack surface of the network, i.e., there are more points in the 536 network that may be subject to attacks. 538 5.2.4.2. Replication-related Header Manipulation 540 An attacker can manipulate the replication-related header fields 541 (R-TAG). This capability opens the door for various types of 542 attacks. For example: 544 o Forward both replicas - malicious change of a packet SN (Sequence 545 Number) can cause both replicas of the packet to be forwarded. 546 Note that this attack has a similar outcome to a replay attack. 548 o Eliminate both replicas - SN manipulation can be used to cause 549 both replicas to be eliminated. In this case an attacker that has 550 access to a single path can cause packets from other paths to be 551 dropped, thus compromising some of the advantage of path 552 redundancy. 554 o Flow hijacking - an attacker can hijack a DetNet flow with access 555 to a single path by systematically replacing the SNs on the given 556 path with higher SN values. For example, an attacker can replace 557 every SN value S with a higher value S+C, where C is a constant 558 integer. Thus, the attacker creates a false illusion that the 559 attacked path has the lowest delay, causing all packets from other 560 paths to be eliminated in favor of the attacked path. Once the 561 flow from the compromised path is favored by the elminating 562 bridge, the flow is hijacked by the attacker. It is now posible 563 to either replace en route packets with malicious packets, or 564 simply injecting errors, causing the packets to be dropped at 565 their destination. 567 5.2.5. Path Choice 569 5.2.5.1. Path Manipulation 571 An attacker can maliciously change, add, or remove a path, thereby 572 affecting the corresponding DetNet flows that use the path. 574 5.2.5.2. Path Choice: Increased Attack Surface 576 One of the possible consequences of a path manipulation attack is an 577 increased attack surface. Thus, when the attack described in the 578 previous subsection is implemented, it may increase the potential of 579 other attacks to be performed. 581 5.2.6. Controller Plane 583 5.2.6.1. Control or Signaling Packet Modification 585 An attacker can maliciously modify en route control packets in order 586 to disrupt or manipulate the DetNet path/resource allocation. 588 5.2.6.2. Control or Signaling Packet Injection 590 An attacker can maliciously inject control packets in order to 591 disrupt or manipulate the DetNet path/resource allocation. 593 5.2.7. Scheduling or Shaping 595 5.2.7.1. Reconnaissance 597 A passive eavesdropper can identify DetNet flows and then gather 598 information about en route DetNet flows, e.g., the number of DetNet 599 flows, their bandwidths, their schedules, or other temporal 600 properties. The gathered information can later be used to invoke 601 other attacks on some or all of the flows. 603 Note that in some cases DetNet flows may be identified based on an 604 explicit DetNet header, but in some cases the flow identification may 605 be based on fields from the L3/L4 headers. If L3/L4 headers are 606 involved, for purposes of this document we assume they are encrypted 607 and/or integrity-protected from external attackers. 609 5.2.8. Time Synchronization Mechanisms 611 An attacker can use any of the attacks described in [RFC7384] to 612 attack the synchronization protocol, thus affecting the DetNet 613 service. 615 5.3. Threat Summary 617 A summary of the attacks that were discussed in this section is 618 presented in Figure 1. For each attack, the table specifies the type 619 of attackers that may invoke the attack. In the context of this 620 summary, the distinction between internal and external attacks is 621 under the assumption that a corresponding security mechanism is being 622 used, and that the corresponding network equipment takes part in this 623 mechanism. 625 +-----------------------------------------+----+----+----+----+ 626 | Attack | Attacker Type | 627 | +---------+---------+ 628 | |Internal |External | 629 | |MITM|Inj.|MITM|Inj.| 630 +-----------------------------------------+----+----+----+----+ 631 |Delay attack | + | + | + | + | 632 +-----------------------------------------+----+----+----+----+ 633 |DetNet Flow Modification or Spoofing | + | + | | | 634 +-----------------------------------------+----+----+----+----+ 635 |Inter-segment Attack | + | + | | | 636 +-----------------------------------------+----+----+----+----+ 637 |Replication: Increased Attack Surface | + | + | + | + | 638 +-----------------------------------------+----+----+----+----+ 639 |Replication-related Header Manipulation | + | | | | 640 +-----------------------------------------+----+----+----+----+ 641 |Path Manipulation | + | + | | | 642 +-----------------------------------------+----+----+----+----+ 643 |Path Choice: Increased Attack Surface | + | + | + | + | 644 +-----------------------------------------+----+----+----+----+ 645 |Control or Signaling Packet Modification | + | | | | 646 +-----------------------------------------+----+----+----+----+ 647 |Control or Signaling Packet Injection | | + | | | 648 +-----------------------------------------+----+----+----+----+ 649 |Reconnaissance | + | | + | | 650 +-----------------------------------------+----+----+----+----+ 651 |Attacks on Time Sync Mechanisms | + | + | + | + | 652 +-----------------------------------------+----+----+----+----+ 654 Figure 1: Threat Analysis Summary 656 6. Security Threat Impacts 658 This section describes and rates the impact of the attacks described 659 in Section 5, Security Threats. In this section, the impacts as 660 described assume that the associated mitigation is not present or has 661 failed. Mitigations are discussed in Section 7, Security Threat 662 Mitigation. 664 In computer security, the impact (or consequence) of an incident can 665 be measured in loss of confidentiality, integrity or availability of 666 information. In the case of time sensitive networks, the impact of a 667 network exploit can also include failure or malfunction of mechanical 668 and/or other OT systems. 670 DetNet raises these stakes significantly for OT applications, 671 particularly those which may have been designed to run in an OT-only 672 environment and thus may not have been designed for security in an IT 673 environment with its associated devices, services and protocols. 675 The severity of various components of the impact of a successful 676 vulnerability exploit to use cases by industry is available in more 677 detail in [RFC8578]. Each of the use cases in the DetNet Use Cases 678 is represented in the table below, including Pro Audio, Electrical 679 Utilities, Industrial M2M (split into two areas, M2M Data Gathering 680 and M2M Control Loop), and others. 682 Components of Impact (left column) include Criticality of Failure, 683 Effects of Failure, Recovery, and DetNet Functional Dependence. 684 Criticality of failure summarizes the seriousness of the impact. The 685 impact of a resulting failure can affect many different metrics that 686 vary greatly in scope and severity. In order to reduce the number of 687 variables, only the following were included: Financial, Health and 688 Safety, People well being (People WB), Affect on a single 689 organization, and affect on multiple organizations. Recovery 690 outlines how long it would take for an affected use case to get back 691 to its pre-failure state (Recovery time objective, RTO), and how much 692 of the original service would be lost in between the time of service 693 failure and recovery to original state (Recovery Point Objective, 694 RPO). DetNet dependence maps how much the following DetNet service 695 objectives contribute to impact of failure: Time dependency, data 696 integrity, source node integrity, availability, latency/jitter. 698 The scale of the Impact mappings is low, medium, and high. In some 699 use cases there may be a multitude of specific applications in which 700 DetNet is used. For simplicity this section attempts to average the 701 varied impacts of different applications. This section does not 702 address the overall risk of a certain impact which would require the 703 likelihood of a failure happening. 705 In practice any such ratings will vary from case to case; the ratings 706 shown here are given as examples. 708 Table, Part One (of Two) 709 +------------------+-----------------------------------------+-----+ 710 | | Pro A | Util | Bldg |Wire- | Cell |M2M |M2M | 711 | | | | | less | |Data |Ctrl | 712 +------------------+-----------------------------------------+-----+ 713 | Criticality | Med | Hi | Low | Med | Med | Med | Med | 714 +------------------+-----------------------------------------+-----+ 715 | Effects 716 +------------------+-----------------------------------------+-----+ 717 | Financial | Med | Hi | Med | Med | Low | Med | Med | 718 +------------------+-----------------------------------------+-----+ 719 | Health/Safety | Med | Hi | Hi | Med | Med | Med | Med | 720 +------------------+-----------------------------------------+-----+ 721 | People WB | Med | Hi | Hi | Low | Hi | Low | Low | 722 +------------------+-----------------------------------------+-----+ 723 | Effect 1 org | Hi | Hi | Med | Hi | Med | Med | Med | 724 +------------------+-----------------------------------------+-----+ 725 | Effect >1 org | Med | Hi | Low | Med | Med | Med | Med | 726 +------------------+-----------------------------------------+-----+ 727 |Recovery 728 +------------------+-----------------------------------------+-----+ 729 | Recov Time Obj | Med | Hi | Med | Hi | Hi | Hi | Hi | 730 +------------------+-----------------------------------------+-----+ 731 | Recov Point Obj | Med | Hi | Low | Med | Low | Hi | Hi | 732 +------------------+-----------------------------------------+-----+ 733 |DetNet Dependence 734 +------------------+-----------------------------------------+-----+ 735 | Time Dependency | Hi | Hi | Low | Hi | Med | Low | Hi | 736 +------------------+-----------------------------------------+-----+ 737 | Latency/Jitter | Hi | Hi | Med | Med | Low | Low | Hi | 738 +------------------+-----------------------------------------+-----+ 739 | Data Integrity | Hi | Hi | Med | Hi | Low | Hi | Low | 740 +------------------+-----------------------------------------+-----+ 741 | Src Node Integ | Hi | Hi | Med | Hi | Med | Hi | Hi | 742 +------------------+-----------------------------------------+-----+ 743 | Availability | Hi | Hi | Med | Hi | Low | Hi | Hi | 744 +------------------+-----------------------------------------+-----+ 746 Table, Part Two (of Two) 747 +------------------+--------------------------+ 748 | | Mining | Block | Network | 749 | | | Chain | Slicing | 750 +------------------+--------------------------+ 751 | Criticality | Hi | Med | Hi | 752 +------------------+--------------------------+ 753 | Effects 754 +------------------+--------------------------+ 755 | Financial | Hi | Hi | Hi | 756 +------------------+--------------------------+ 757 | Health/Safety | Hi | Low | Med | 758 +------------------+--------------------------+ 759 | People WB | Hi | Low | Med | 760 +------------------+--------------------------+ 761 | Effect 1 org | Hi | Hi | Hi | 762 +------------------+--------------------------+ 763 | Effect >1 org | Hi | Low | Hi | 764 +------------------+--------------------------+ 765 |Recovery 766 +------------------+--------------------------+ 767 | Recov Time Obj | Hi | Low | Hi | 768 +------------------+--------------------------+ 769 | Recov Point Obj | Hi | Low | Hi | 770 +------------------+--------------------------+ 771 |DetNet Dependence 772 +------------------+--------------------------+ 773 | Time Dependency | Hi | Low | Hi | 774 +------------------+--------------------------+ 775 | Latency/Jitter | Hi | Low | Hi | 776 +------------------+--------------------------+ 777 | Data Integrity | Hi | Hi | Hi | 778 +------------------+--------------------------+ 779 | Src Node Integ | Hi | Hi | Hi | 780 +------------------+--------------------------+ 781 | Availability | Hi | Hi | Hi | 782 +------------------+--------------------------+ 784 Figure 2: Impact of Attacks by Use Case Industry 786 The rest of this section will cover impact of the different groups in 787 more detail. 789 6.1. Delay-Attacks 791 6.1.1. Data Plane Delay Attacks 793 Note that 'delay attack' also includes the possibility of a 'negative 794 delay' or early arrival of a packet, or possibly adversely changing 795 the timestamp value. 797 Delayed messages in a DetNet link can result in the same behavior as 798 dropped messages in ordinary networks as the services attached to the 799 DetNet flow have strict deterministic requirements. 801 For a single path scenario, disruption is a real possibility, whereas 802 in a multipath scenario, large delays or instabilities in one DetNet 803 flow can lead to increased buffer and processor resources at the 804 eliminating router. 806 A data-plane delay attack on a system controlling substantial moving 807 devices, for example in industrial automation, can cause physical 808 damage. For example, if the network promises a bounded latency of 809 2ms for a flow, yet the machine receives it with 5ms latency, the 810 machine's control loop can become unstable. 812 6.1.2. Controller Plane Delay Attacks 814 In and of itself, this is not directly a threat to the DetNet 815 service, but the effects of delaying control messages can have quite 816 adverse effects later. 818 o Delayed tear-down can lead to resource leakage, which in turn can 819 result in failure to allocate new DetNet flows, finally giving 820 rise to a denial of service attack. 822 o Failure to deliver, or severely delaying, controller plane 823 messages adding an endpoint to a multicast-group will prevent the 824 new endpoint from receiving expected frames thus disrupting 825 expected behavior. 827 o Delaying messages removing an endpoint from a group can lead to 828 loss of privacy as the endpoint will continue to receive messages 829 even after it is supposedly removed. 831 6.2. Flow Modification and Spoofing 833 6.2.1. Flow Modification 835 If the contents of a packet header or body can be modified by the 836 attacker, this can cause the packet to be routed incorrectly or 837 dropped, or the payload to be corrupted or subtly modified. 839 6.2.2. Spoofing 841 6.2.2.1. Dataplane Spoofing 843 Spoofing dataplane messages can result in increased resource 844 consumptions on the routers throughout the network as it will 845 increase buffer usage and processor utilization. This can lead to 846 resource exhaustion and/or increased delay. 848 If the attacker manages to create valid headers, the false messages 849 can be forwarded through the network, using part of the allocated 850 bandwidth. This in turn can cause legitimate messages to be dropped 851 when the resource budget has been exhausted. 853 Finally, the endpoint will have to deal with invalid messages being 854 delivered to the endpoint instead of (or in addition to) a valid 855 message. 857 6.2.2.2. Controller Plane Spoofing 859 A successful controller plane spoofing-attack will potentionally have 860 adverse effects. It can do virtually anything from: 862 o modifying existing DetNet flows by changing the available 863 bandwidth 865 o add or remove endpoints from a DetNet flow 867 o drop DetNet flows completely 869 o falsely create new DetNet flows (exhaust the systems resources, or 870 to enable DetNet flows that are outside the Network Engineer's 871 control) 873 6.3. Segmentation Attacks (injection) 875 6.3.1. Data Plane Segmentation 877 Injection of false messages in a DetNet flow could lead to exhaustion 878 of the available bandwidth for that flow if the routers attribute 879 these false messages to that flow's budget. 881 In a multipath scenario, injected messages will cause increased 882 processor utilization in elimination routers. If enough paths are 883 subject to malicious injection, the legitimate messages can be 884 dropped. Likewise it can cause an increase in buffer usage. In 885 total, it will consume more resources in the routers than normal, 886 giving rise to a resource exhaustion attack on the routers. 888 If a DetNet flow is interrupted, the end application will be affected 889 by what is now a non-deterministic flow. 891 6.3.2. Controller Plane Segmentation 893 In a successful controller plane segmentation attack, control 894 messages are acted on by nodes in the network, unbeknownst to the 895 central controller or the network engineer. This has the potential 896 to: 898 o create new DetNet flows (exhausting resources) 900 o drop existing DetNet flows (denial of service) 902 o add/remove end-stations to a multicast group (loss of privacy) 904 o modify the DetNet flow attributes (affecting available bandwidth 906 6.4. Replication and Elimination 908 The Replication and Elimination is relevant only to Data Plane 909 messages as controller plane messages are not subject to multipath 910 routing. 912 6.4.1. Increased Attack Surface 914 Covered briefly in Section 6.3, Segmentation Attacks. 916 6.4.2. Header Manipulation at Elimination Routers 918 Covered briefly in Section 6.3, Segmentation Attacks. 920 6.5. Control or Signaling Packet Modification 922 If control packets are subject to manipulation undetected, the 923 network can be severely compromised. 925 6.6. Control or Signaling Packet Injection 927 If an attacker can inject control packets undetected, the network can 928 be severely compromised. 930 6.7. Reconnaissance 932 Of all the attacks, this is one of the most difficult to detect and 933 counter. Often, an attacker will start out by observing the traffic 934 going through the network and use the knowledge gathered in this 935 phase to mount future attacks. 937 The attacker can, at their leisure, observe over time all aspects of 938 the messaging and signalling, learning the intent and purpose of all 939 traffic flows. At some later date, possibly at an important time in 940 an operational context, the attacker can launch a multi-faceted 941 attack, possibly in conjunction with some demand for ransom. 943 The flow-id in the header of the data plane-messages gives an 944 attacker a very reliable identifier for DetNet traffic, and this 945 traffic has a high probability of going to lucrative targets. 947 Applications which are ported from a private OT network to the higher 948 visibility DetNet environment may need to be adapted to limit 949 distinctive flow properties that could make them susceptible to 950 reconnaissance. 952 6.8. Attacks on Time Sync Mechanisms 954 Attacks on time sync mechanisms are addressed in [RFC7384]. 956 6.9. Attacks on Path Choice 958 This is covered in part in Section 6.3, Segmentation Attacks, and as 959 with Replication and Elimination (Section 6.4), this is relevant for 960 DataPlane messages. 962 7. Security Threat Mitigation 964 This section describes a set of measures that can be taken to 965 mitigate the attacks described in Section 5, Security Threats. These 966 mitigations should be viewed as a toolset that includes several 967 different and diverse tools. Each application or system will 968 typically use a subset of these tools, based on a system-specific 969 threat analysis. 971 7.1. Path Redundancy 973 Description 975 A DetNet flow that can be forwarded simultaneously over multiple 976 paths. Path replication and elimination [RFC8655] provides 977 resiliency to dropped or delayed packets. This redundancy 978 improves the robustness to failures and to man-in-the-middle 979 attacks. Note: At the time of this writing, PREOF is not defined 980 for the IP data plane. 982 Related attacks 984 Path redundancy can be used to mitigate various man-in-the-middle 985 attacks, including attacks described in Section 5.2.1, 986 Section 5.2.2, Section 5.2.3, and Section 5.2.8. However it is 987 also possible that multiple paths may make it more difficult to 988 locate the source of a MITM attacker. 990 A delay modulation attack could result in extensively exercising 991 parts of the code that wouldn't normally be extensively exercised 992 and thus might expose flaws in the system that might otherwise not 993 be exposed. 995 7.2. Integrity Protection 997 Description 998 An integrity protection mechanism, such as a Hash-based Message 999 Authentication Code (HMAC) can be used to mitigate modification 1000 attacks on IP packets. Integrity protection in the controller 1001 plane is discussed in Section 7.6. 1003 Packet Sequence Number Integrity Considerations 1005 The use of PREOF in a DetNet implementation implies the use of a 1006 sequence number for each packet. There is a trust relationship 1007 between the device that adds the sequence number and the device 1008 that removes the sequence number. The sequence number may be end- 1009 to-end source to destination, or may be added/deleted by network 1010 edge devices. The adder and remover(s) have the trust 1011 relationship because they are the ones that ensure that the 1012 sequence numbers are not modifiable. Between those two points, 1013 there may or may not be replication and elimination functions. 1014 The elimination functions must be able to see the sequence 1015 numbers. Therefore any encryption that is done between adders and 1016 removers must not obscure the sequence number. If the sequence 1017 removers and the eliminators are in the same physical device, it 1018 may be possible to obscure the sequence number, however that is a 1019 layer violation, and is not recommended practice. Note: At the 1020 time of this writing, PREOF is not defined for the IP data plane. 1022 Related attacks 1024 Integrity protection mitigates attacks related to modification and 1025 tampering, including the attacks described in Section 5.2.2 and 1026 Section 5.2.4. 1028 7.3. DetNet Node Authentication 1030 Description 1032 Source authentication verifies the authenticity of DetNet sources, 1033 enabling mitigation of spoofing attacks. Note that while 1034 integrity protection (Section 7.2) prevents intermediate nodes 1035 from modifying information, authentication can provide traffic 1036 origin verification, i.e. to verify that each packet in a DetNet 1037 flow is from a trusted source. Authentication may be implemented 1038 as part of ingress filtering, for example. 1040 Related attacks 1042 DetNet node authentication is used to mitigate attacks related to 1043 spoofing, including the attacks of Section 5.2.2, and 1044 Section 5.2.4. 1046 7.4. Dummy Traffic Insertion 1048 Description 1050 With some queueing methods such as [IEEE802.1Qch-2017] it is 1051 possible to introduce dummy traffic in order to regularize the 1052 timing of packet transmission. 1054 Related attacks 1056 Removing distinctive temporal properties of individual packets or 1057 flows can be used to mitigate against reconnaissance attacks 1058 Section 5.2.7. 1060 7.5. Encryption 1062 Description 1064 DetNet flows can in principle be forwarded in encrypted form at 1065 the DetNet layer, however, regarding encryption of IP headers see 1066 Section 9. 1068 Alternatively, if the payload is end-to-end encrypted at the 1069 application layer, the DetNet nodes should not have any need to 1070 inspect the payload itself, and thus the DetNet implementation can 1071 be data-agnostic. 1073 Encryption can also be applied at the subnet layer, for example 1074 for Ethernet using MACSec, as noted in Section 9. 1076 Related attacks 1078 Encryption can be used to mitigate recon attacks (Section 5.2.7). 1079 However, for a DetNet network to give differentiated quality of 1080 service on a flow-by-flow basis, the network must be able to 1081 identify the flows individually. This implies that in a recon 1082 attack the attacker may also be able to track individual flows to 1083 learn more about the system. 1085 7.5.1. Encryption Considerations for DetNet 1087 Any compute time which is required for encryption and decryption 1088 processing ('crypto') must be included in the flow latency 1089 calculations. Thus, crypto algorithms used in a DetNet must have 1090 bounded worst-case execution times, and these values must be used in 1091 the latency calculations. 1093 Some crypto algorithms are symmetric in encode/decode time (such as 1094 AES) and others are asymmetric (such as public key algorithms). 1095 There are advantages and disadvantages to the use of either type in a 1096 given DetNet context. The discussion in this document relates to the 1097 timing implications of crypto for DetNet; it is assumed that 1098 integrity considerations are covered elsewhere in the literature. 1100 Asymmetrical crypto is typically not used in networks on a packet-by- 1101 packet basis due to its computational cost. For example, if only 1102 endpoint checks or checks at a small number of intermediate points 1103 are required, asymmetric crypto can be used to authenticate 1104 distribution or exchange of a secret symmetric crypto key; a 1105 successful check based on that key will provide traffic origin 1106 verification, as long as the key is kept secret by the participants. 1107 TLS and IKE (for IPsec) are examples of this for endpoint checks. 1109 However, if secret symmetrical keys are used for this purpose the key 1110 must be given to all relays, which increases the probability of a 1111 secret key being leaked. Also, if any relay is compromised or 1112 misbehaving it may inject traffic into the flow. 1114 Alternatively, asymmetric crypto can provide traffic origin 1115 verification at every intermediate node. For example, a DetNet flow 1116 can be associated with an (asymmetric) keypair, such that the private 1117 key is available to the source of the flow and the public key is 1118 distributed with the flow information, allowing verification at every 1119 node for every packet. However, this is more computationally 1120 expensive. 1122 In either case, origin verification also requires replay detection as 1123 part of the security protocol to prevent an attacker from recording 1124 and resending traffic, e.g., as a denial of service attack on flow 1125 forwarding resources. 1127 If crypto keys are to be regenerated over the duration of the flow 1128 then the time required to accomplish this must be accounted for in 1129 the latency calculations. 1131 7.6. Control and Signaling Message Protection 1133 Description 1135 Control and sigaling messages can be protected using 1136 authentication and integrity protection mechanisms. 1138 Related attacks 1139 These mechanisms can be used to mitigate various attacks on the 1140 controller plane, as described in Section 5.2.6, Section 5.2.8 and 1141 Section 5.2.5. 1143 7.7. Dynamic Performance Analytics 1145 Description 1147 The expectation is that the network will have a way to monitor to 1148 detect if timing guarantees are not being met, and a way to alert 1149 the controller plane in that event. Information about the network 1150 performance can be gathered in real-time in order to detect 1151 anomalies and unusual behavior that may be the symptom of a 1152 security attack. The gathered information can be based, for 1153 example, on per-flow counters, bandwidth measurement, and 1154 monitoring of packet arrival times. Unusual behavior or 1155 potentially malicious nodes can be reported to a management 1156 system, or can be used as a trigger for taking corrective actions. 1157 The information can be tracked by DetNet end systems and transit 1158 nodes, and exported to a management system, for example using 1159 YANG. 1161 Related attacks 1163 Performance analytics can be used to mitigate various attacks, 1164 including the ones described in Section 5.2.1 (Delay Attack), 1165 Section 5.2.3 (Resource Segmentation Attack), and Section 5.2.8 1166 (Time Sync Attack). 1168 For example, in the case of data plane delay attacks, one possible 1169 mitigation is to timestamp the data at the source, and timestamp 1170 it again at the destination, and if the resulting latency exceeds 1171 the promised bound, discard that data and warn the operator (and/ 1172 or enter a fail-safe mode). Note that DetNet specifies packet 1173 sequence numbering, however it does not specify use of packet 1174 timestamps, although they may be used by the underlying transport 1175 (for example TSN) to provide the service. 1177 7.8. Mitigation Summary 1179 The following table maps the attacks of Section 5, Security Threats, 1180 to the impacts of Section 6, Security Threat Impacts, and to the 1181 mitigations of the current section. Each row specifies an attack, 1182 the impact of this attack if it is successfully implemented, and 1183 possible mitigation methods. 1185 +----------------------+---------------------+---------------------+ 1186 | Attack | Impact | Mitigations | 1187 +----------------------+---------------------+---------------------+ 1188 |Delay Attack |-Non-deterministic |-Path redundancy | 1189 | | delay |-Performance | 1190 | |-Data disruption | analytics | 1191 | |-Increased resource | | 1192 | | consumption | | 1193 +----------------------+---------------------+---------------------+ 1194 |Reconnaissance |-Enabler for other |-Encryption | 1195 | | attacks |-Dummy traffic | 1196 | | | insertion | 1197 +----------------------+---------------------+---------------------+ 1198 |DetNet Flow Modificat-|-Increased resource |-Path redundancy | 1199 |ion or Spoofing | consumption |-Integrity protection| 1200 | |-Data disruption |-DetNet Node | 1201 | | | authentication | 1202 +----------------------+---------------------+---------------------+ 1203 |Inter-Segment Attack |-Increased resource |-Path redundancy | 1204 | | consumption |-Performance | 1205 | |-Data disruption | analytics | 1206 +----------------------+---------------------+---------------------+ 1207 |Replication: Increased|-All impacts of other|-Integrity protection| 1208 |attack surface | attacks |-DetNet Node | 1209 | | | authentication | 1210 +----------------------+---------------------+---------------------+ 1211 |Replication-related |-Non-deterministic |-Integrity protection| 1212 |Header Manipulation | delay |-DetNet Node | 1213 | |-Data disruption | authentication | 1214 +----------------------+---------------------+---------------------+ 1215 |Path Manipulation |-Enabler for other |-Control message | 1216 | | attacks | protection | 1217 +----------------------+---------------------+---------------------+ 1218 |Path Choice: Increased|-All impacts of other|-Control message | 1219 |Attack Surface | attacks | protection | 1220 +----------------------+---------------------+---------------------+ 1221 |Control or Signaling |-Increased resource |-Control message | 1222 |Packet Modification | consumption | protection | 1223 | |-Non-deterministic | | 1224 | | delay | | 1225 | |-Data disruption | | 1226 +----------------------+---------------------+---------------------+ 1227 |Control or Signaling |-Increased resource |-Control message | 1228 |Packet Injection | consumption | protection | 1229 | |-Non-deterministic | | 1230 | | delay | | 1231 | |-Data disruption | | 1232 +----------------------+---------------------+---------------------+ 1233 |Attacks on Time Sync |-Non-deterministic |-Path redundancy | 1234 |Mechanisms | delay |-Control message | 1235 | |-Increased resource | protection | 1236 | | consumption |-Performance | 1237 | |-Data disruption | analytics | 1238 +----------------------+---------------------+---------------------+ 1240 Figure 3: Mapping Attacks to Impact and Mitigations 1242 8. Association of Attacks to Use Cases 1244 Different attacks can have different impact and/or mitigation 1245 depending on the use case, so we would like to make this association 1246 in our analysis. However since there is a potentially unbounded list 1247 of use cases, we categorize the attacks with respect to the common 1248 themes of the use cases as identified in the Use Case Common Themes 1249 section of the DetNet Use Cases [RFC8578]. 1251 See also Figure 2 for a mapping of the impact of attacks per use case 1252 by industry. 1254 8.1. Use Cases by Common Themes 1256 In this section we review each theme and discuss the attacks that are 1257 applicable to that theme, as well as anything specific about the 1258 impact and mitigations for that attack with respect to that theme. 1259 The table Figure 5, Mapping Between Themes and Attacks, then provides 1260 a summary of the attacks that are applicable to each theme. 1262 8.1.1. Sub-Network Layer 1264 DetNet is expected to run over various transmission mediums, with 1265 Ethernet being the first identified. Attacks such as Delay or 1266 Reconnaissance might be implemented differently on a different 1267 transmission medium, however the impact on the DetNet as a whole 1268 would be essentially the same. We thus conclude that all attacks and 1269 impacts that would be applicable to DetNet over Ethernet (i.e. all 1270 those named in this document) would also be applicable to DetNet over 1271 other transmission mediums. 1273 With respect to mitigations, some methods are specific to the 1274 Ethernet medium, for example time-aware scheduling using 802.1Qbv can 1275 protect against excessive use of bandwidth at the ingress - for other 1276 mediums, other mitigations would have to be implemented to provide 1277 analogous protection. 1279 8.1.2. Central Administration 1281 A DetNet network can be controlled by a centralized network 1282 configuration and control system. Such a system may be in a single 1283 central location, or it may be distributed across multiple control 1284 entities that function together as a unified control system for the 1285 network. 1287 In this document we distinguish between attacks on the DetNet 1288 Controller plane vs. Data plane. But is an attack affecting control 1289 plane packets synonymous with an attack on the control plane itself? 1290 For purposes of this document let us consider an attack on the 1291 control system itself to be out of scope, and consider all attacks 1292 named in this document which are relevant to controller plane packets 1293 to be relevant to this theme, including Path Manipulation, Path 1294 Choice, Control Packet Modification or Injection, Reconaissance and 1295 Attacks on Time Sync Mechanisms. 1297 8.1.3. Hot Swap 1299 A DetNet network is not expected to be "plug and play" - it is 1300 expected that there is some centralized network configuration and 1301 control system. However, the ability to "hot swap" components (e.g. 1302 due to malfunction) is similar enough to "plug and play" that this 1303 kind of behavior may be expected in DetNet networks, depending on the 1304 implementation. 1306 An attack surface related to Hot Swap is that the DetNet network must 1307 at least consider input at runtime from devices that were not part of 1308 the initial configuration of the network. Even a "perfect" (or 1309 "hitless") replacement of a device at runtime would not necessarily 1310 be ideal, since presumably one would want to distinguish it from the 1311 original for OAM purposes (e.g. to report hot swap of a failed 1312 device). 1314 This implies that an attack such as Flow Modification, Spoofing or 1315 Inter-segment (which could introduce packets from a "new" device 1316 (i.e. one heretofore unknown on the network) could be used to exploit 1317 the need to consider such packets (as opposed to rejecting them out 1318 of hand as one would do if one did not have to consider introduction 1319 of a new device). 1321 Similarly if the network was designed to support runtime replacement 1322 of a clock device, then presence (or apparent presence) and thus 1323 consideration of packets from a new such device could affect the 1324 network, or the time sync of the network, for example by initiating a 1325 new Best Master Clock selection process. Thus attacks on time sync 1326 should be considered when designing hot swap type functionality (see 1327 [RFC7384]). 1329 8.1.4. Data Flow Information Models 1331 Data Flow YANG models specific to DetNet networks are specified by 1332 DetNet, and thus are 'new' and thus potentially present a new attack 1333 surface. 1335 8.1.5. L2 and L3 Integration 1337 A DetNet network integrates Layer 2 (bridged) networks (e.g. AVB/TSN 1338 LAN) and Layer 3 (routed) networks via the use of well-known 1339 protocols such as IP, MPLS-PW, and Ethernet. 1341 There are no specific entries in our table, however that does not 1342 imply that there could be no relevant attacks related to L2,L3 1343 integration. 1345 8.1.6. End-to-End Delivery 1347 Packets sent over DetNet are not to be dropped by the network due to 1348 congestion. (Packets may however intentionally be dropped for 1349 intended reasons, e.g. per security measures). 1351 A Data plane attack may force packets to be dropped, for example a 1352 "long" Delay or Replication/Elimination or Flow Modification attack. 1354 The same result might be obtained by a controller plane attack, e.g. 1355 Path Manipulation or Signaling Packet Modification. 1357 It may be that such attacks are limited to Internal MITM attackers, 1358 but other possibilities should be considered. 1360 An attack may also cause packets that should not be delivered to be 1361 delivered, such as by forcing packets from one (e.g. replicated) path 1362 to be preferred over another path when they should not be 1363 (Replication attack), or by Flow Modification, or by Path Choice or 1364 Packet Injection. A Time Sync attack could cause a system that was 1365 expecting certain packets at certain times to accept unintended 1366 packets based on compromised system time or time windowing in the 1367 scheduler. 1369 Packets may also be dropped due to malfunctioning software or 1370 hardware. 1372 8.1.7. Proprietary Deterministic Ethernet Networks 1374 There are many proprietary non-interoperable deterministic Ethernet- 1375 based networks currently available; DetNet is intended to provide an 1376 open-standards-based alternative to such networks. In cases where a 1377 DetNet intersects with remnants of such networks or their protocols, 1378 such as by protocol emulation or access to such a network via a 1379 gateway, new attack surfaces can be opened. 1381 For example an Inter-Segment or Controller plane attack such as Path 1382 Manipulation, Path Choice or Control Packet Modification/Injection 1383 could be used to exploit commands specific to such a protocol, or 1384 that are interpreted differently by the different protocols or 1385 gateway. 1387 8.1.8. Replacement for Proprietary Fieldbuses 1389 There are many proprietary "field buses" used in today's industrial 1390 and other industries; DetNet is intended to provide an open- 1391 standards-based alternative to such buses. In cases where a DetNet 1392 intersects with such fieldbuses or their protocols, such as by 1393 protocol emulation or access via a gateway, new attack surfaces can 1394 be opened. 1396 For example an Inter-Segment or Controller plane attack such as Path 1397 Manipulation, Path Choice or Control Packet Modification/Injection 1398 could be used to exploit commands specific to such a protocol, or 1399 that are interpreted differently by the different protocols or 1400 gateway. 1402 8.1.9. Deterministic vs Best-Effort Traffic 1404 Most of the themes described in this document address OT (reserved) 1405 DetNet flows - this item is intended to address issues related to IT 1406 traffic on a DetNet. 1408 DetNet is intended to support coexistence of time-sensitive 1409 operational (OT, deterministic) traffic and information (IT, "best 1410 effort") traffic on the same ("unified") network. 1412 With DetNet, this coexistance will become more common, and 1413 mitigations will need to be established. The fact that the IT 1414 traffic on a DetNet is limited to a corporate controlled network 1415 makes this a less difficult problem compared to being exposed to the 1416 open Internet, however this aspect of DetNet security should not be 1417 underestimated. 1419 An Inter-segment attack can flood the network with IT-type traffic 1420 with the intent of disrupting handling of IT traffic, and/or the goal 1421 of interfering with OT traffic. Presumably if the DetNet flow 1422 reservation and isolation of the DetNet is well-designed (better- 1423 designed than the attack) then interference with OT traffic should 1424 not result from an attack that floods the network with IT traffic. 1426 However the DetNet's handling of IT traffic may not (by design) be as 1427 resilient to DOS attack, and thus designers must be otherwise 1428 prepared to mitigate DOS attacks on IT traffic in a DetNet. 1430 8.1.10. Deterministic Flows 1432 Reserved bandwidth data flows (deterministic flows) must provide the 1433 allocated bandwidth, and must be isolated from each other. 1435 A Spoofing or Inter-segment attack which adds packet traffic to a 1436 bandwidth-reserved DetNet flow could cause that flow to occupy more 1437 bandwidth than it was allocated, resulting in interference with other 1438 DetNet flows. 1440 A Flow Modification or Spoofing or Header Manipulation or Control 1441 Packet Modification attack could cause packets from one flow to be 1442 directed to another flow, thus breaching isolation between the flows. 1444 8.1.11. Unused Reserved Bandwidth 1446 If bandwidth reservations are made for a DetNet flow but the 1447 associated bandwidth is not used at any point in time, that bandwidth 1448 is made available on the network for best-effort traffic. However, 1449 note that security considerations for best-effort traffic on a DetNet 1450 network is out of scope of the present document, provided that such 1451 an attack does not affect performance for DetNet OT traffic. 1453 8.1.12. Interoperability 1455 The DetNet network specifications are intended to enable an ecosystem 1456 in which multiple vendors can create interoperable products, thus 1457 promoting device diversity and potentially higher numbers of each 1458 device manufactured. 1460 Given that the DetNet specifications are unambiguously written and 1461 that the implementations are accurate, then this should not in and of 1462 itself cause a security concern; however, in the real world, it could 1463 be. The network operator can mitigate this through sufficient 1464 interoperability testing. 1466 8.1.13. Cost Reductions 1468 The DetNet network specifications are intended to enable an ecosystem 1469 in which multiple vendors can create interoperable products, thus 1470 promoting higher numbers of each device manufactured, promoting cost 1471 reduction and cost competition among vendors. Such "low cost" 1472 hardware or software components might present security concerns. 1474 Network operators can mitigate such concerns through sufficient 1475 product testing. 1477 8.1.14. Insufficiently Secure Devices 1479 The DetNet network specifications are intended to enable an ecosystem 1480 in which multiple vendors can create interoperable products, thus 1481 promoting device diversity and potentially higher numbers of each 1482 device manufactured. Software that was originally designed for 1483 operation in isolated OT networks (and thus may not have been 1484 designed to be sufficiently secure, or secure at all) but is then 1485 deployed on a DetNet network that is intended to be highly secure may 1486 present an attack surface. (For example IoT exploits like the Mirai 1487 video-camera botnet ([MIRAI]). 1489 The DetNet network operator may need to take specific actions to 1490 protect such devices. 1492 8.1.15. DetNet Network Size 1494 DetNet networks range in size from very small, e.g. inside a single 1495 industrial machine, to very large, for example a Utility Grid network 1496 spanning a whole country. 1498 The size of the network might be related to how the attack is 1499 introduced into the network, for example if the entire network is 1500 local, there is a threat that power can be cut to the entire network. 1501 If the network is large, perhaps only a part of the network is 1502 attacked. 1504 A Delay attack might be as relevant to a small network as to a large 1505 network, although the amount of delay might be different. 1507 Attacks sourced from IT traffic might be more likely in large 1508 networks, since more people might have access to the network, 1509 presenting a larger attack surface. Similarly Path Manipulation, 1510 Path Choice and Time Sync attacks seem more likely relevant to large 1511 networks. 1513 8.1.16. Multiple Hops 1515 Large DetNet networks (e.g. a Utility Grid network) may involve many 1516 "hops" over various kinds of links for example radio repeaters, 1517 microwave links, fiber optic links, etc. 1519 An attack that takes advantage of flaws (or even normal operation) in 1520 the device drivers for the various links (through internal knowledge 1521 of how the individual driver or firmware operates, perhaps like the 1522 Stuxnet attack) could take proportionately greater advantage of this 1523 topology. 1525 It is also possible that this DetNet topology will not be in as 1526 common use as other more homogeneous topologies so there may be more 1527 opportunity for attackers to exploit software and/or protocol flaws 1528 in the implementations which have not been wrung out by extensive 1529 use, particularly in the case of early adopters. 1531 Of the attacks we have defined, the ones identified above as relevant 1532 to "large" networks are the most relevant. 1534 8.1.17. Level of Service 1536 A DetNet is expected to provide means to configure the network that 1537 include querying network path latency, requesting bounded latency for 1538 a given DetNet flow, requesting worst case maximum and/or minimum 1539 latency for a given path or DetNet flow, and so on. It is an 1540 expected case that the network cannot provide a given requested 1541 service level. In such cases the network control system should reply 1542 that the requested service level is not available (as opposed to 1543 accepting the parameter but then not delivering the desired 1544 behavior). 1546 Controller plane attacks such as Signaling Packet Modification and 1547 Injection could be used to modify or create control traffic that 1548 could interfere with the process of a user requesting a level of 1549 service and/or the network's reply. 1551 Reconnaissance could be used to characterize flows and perhaps target 1552 specific flows for attack via the controller plane as noted above. 1554 8.1.18. Bounded Latency 1556 DetNet provides the expectation of guaranteed bounded latency. 1558 Delay attacks can cause packets to miss their agreed-upon latency 1559 boundaries. 1561 Time Sync attacks can corrupt the system's time reference, resulting 1562 in missed latency deadlines (with respect to the "correct" time 1563 reference). 1565 8.1.19. Low Latency 1567 Applications may require "extremely low latency" however depending on 1568 the application these may mean very different latency values; for 1569 example "low latency" across a Utility grid network is on a different 1570 time scale than "low latency" in a motor control loop in a small 1571 machine. The intent is that the mechanisms for specifying desired 1572 latency include wide ranges, and that architecturally there is 1573 nothing to prevent arbitrarily low latencies from being implemented 1574 in a given network. 1576 Attacks on the controller plane (as described in the Level of Service 1577 theme) and Delay and Time attacks (as described in the Bounded 1578 Latency theme) both apply here. 1580 8.1.20. Bounded Jitter (Latency Variation) 1582 DetNet is expected to provide bounded jitter (packet to packet 1583 latency variation). 1585 Delay attacks can cause packets to vary in their arrival times, 1586 resulting in packet to packet latency variation, thereby violating 1587 the jitter specification. 1589 8.1.21. Symmetrical Path Delays 1591 Some applications would like to specify that the transit delay time 1592 values be equal for both the transmit and return paths. 1594 Delay attacks can cause path delays to materially differ between 1595 paths. 1597 Time Sync attacks can corrupt the system's time reference, resulting 1598 in path delays that may be perceived to be different (with respect to 1599 the "correct" time reference) even if they are not materially 1600 different. 1602 8.1.22. Reliability and Availability 1604 DetNet based systems are expected to be implemented with essentially 1605 arbitrarily high availability (for example 99.9999% up time, or even 1606 12 nines). The intent is that the DetNet designs should not make any 1607 assumptions about the level of reliability and availability that may 1608 be required of a given system, and should define parameters for 1609 communicating these kinds of metrics within the network. 1611 Any attack on the system, of any type, can affect its overall 1612 reliability and availability, thus in our table we have marked every 1613 attack. Since every DetNet depends to a greater or lesser degree on 1614 reliability and availability, this essentially means that all 1615 networks have to mitigate all attacks, which to a greater or lesser 1616 degree defeats the purpose of associating attacks with use cases. It 1617 also underscores the difficulty of designing "extremely high 1618 reliability" networks. 1620 8.1.23. Redundant Paths 1622 DetNet based systems are expected to be implemented with essentially 1623 arbitrarily high reliability/availability. A strategy used by DetNet 1624 for providing such extraordinarily high levels of reliability is to 1625 provide redundant paths that can be seamlessly switched between, all 1626 the while maintaining the required performance of that system. 1628 Replication-related attacks are by definition applicable here. 1629 Controller plane attacks can also interfere with the configuration of 1630 redundant paths. 1632 8.1.24. Security Measures 1634 A DetNet network must be made secure against devices failures, 1635 attackers, misbehaving devices, and so on. Does the threat affect 1636 such security measures themselves, e.g. by attacking SW designed to 1637 protect against device failure? 1639 This is TBD, thus there are no specific entries in our table, however 1640 that does not imply that there could be no relevant attacks. 1642 8.2. Attack Types by Use Case Common Theme 1644 The following table lists the attacks of Section 5, Security Threats, 1645 assigning a number to each type of attack. That number is then used 1646 as a short form identifier for the attack in Figure 5, Mapping 1647 Between Themes and Attacks. 1649 +--+----------------------------------------+----------------------+ 1650 | | Attack | Section | 1651 +--+----------------------------------------+----------------------+ 1652 | 1|Delay Attack | Section 3.2.1 | 1653 +--+----------------------------------------+----------------------+ 1654 | 2|DetNet Flow Modification or Spoofing | Section 3.2.2 | 1655 +--+----------------------------------------+----------------------+ 1656 | 3|Inter-Segment Attack | Section 3.2.3 | 1657 +--+----------------------------------------+----------------------+ 1658 | 4|Replication: Increased attack surface | Section 3.2.4.1 | 1659 +--+----------------------------------------+----------------------+ 1660 | 5|Replication-related Header Manipulation | Section 3.2.4.2 | 1661 +--+----------------------------------------+----------------------+ 1662 | 6|Path Manipulation | Section 3.2.5.1 | 1663 +--+----------------------------------------+----------------------+ 1664 | 7|Path Choice: Increased Attack Surface | Section 3.2.5.2 | 1665 +--+----------------------------------------+----------------------+ 1666 | 8|Control or Signaling Packet Modification| Section 3.2.6.1 | 1667 +--+----------------------------------------+----------------------+ 1668 | 9|Control or Signaling Packet Injection | Section 3.2.6.2 | 1669 +--+----------------------------------------+----------------------+ 1670 |10|Reconnaissance | Section 3.2.7 | 1671 +--+----------------------------------------+----------------------+ 1672 |11|Attacks on Time Sync Mechanisms | Section 3.2.8 | 1673 +--+----------------------------------------+----------------------+ 1675 Figure 4: List of Attacks 1677 The following table maps the use case themes presented in this memo 1678 to the attacks of Figure 4. Each row specifies a theme, and the 1679 attacks relevant to this theme are marked with a '+'. 1681 +----------------------------+--------------------------------+ 1682 | Theme | Attack | 1683 | +--+--+--+--+--+--+--+--+--+--+--+ 1684 | | 1| 2| 3| 4| 5| 6| 7| 8| 9|10|11| 1685 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1686 |Network Layer - AVB/TSN Eth.| +| +| +| +| +| +| +| +| +| +| +| 1687 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1688 |Central Administration | | | | | | +| +| +| +| +| +| 1689 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1690 |Hot Swap | | +| +| | | | | | | | +| 1691 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1692 |Data Flow Information Models| | | | | | | | | | | | 1693 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1694 |L2 and L3 Integration | | | | | | | | | | | | 1695 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1696 |End-to-end Delivery | +| +| +| +| +| +| +| +| +| | +| 1697 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1698 |Proprietary Deterministic | | | +| | | +| +| +| +| | | 1699 |Ethernet Networks | | | | | | | | | | | | 1700 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1701 |Replacement for Proprietary | | | +| | | +| +| +| +| | | 1702 |Fieldbuses | | | | | | | | | | | | 1703 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1704 |Deterministic vs. Best- | | | +| | | | | | | | | 1705 |Effort Traffic | | | | | | | | | | | | 1706 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1707 |Deterministic Flows | | +| +| | +| +| | +| | | | 1708 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1709 |Unused Reserved Bandwidth | | +| +| | | | | +| +| | | 1710 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1711 |Interoperability | | | | | | | | | | | | 1712 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1713 |Cost Reductions | | | | | | | | | | | | 1714 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1715 |Insufficiently Secure | | | | | | | | | | | | 1716 |Devices | | | | | | | | | | | | 1717 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1718 |DetNet Network Size | +| | | | | +| +| | | | +| 1719 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1720 |Multiple Hops | +| +| | | | +| +| | | | +| 1721 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1722 |Level of Service | | | | | | | | +| +| +| | 1723 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1724 |Bounded Latency | +| | | | | | | | | | +| 1725 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1726 |Low Latency | +| | | | | | | +| +| +| +| 1727 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1728 |Bounded Jitter | +| | | | | | | | | | | 1729 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1730 |Symmetric Path Delays | +| | | | | | | | | | +| 1731 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1732 |Reliability and Availability| +| +| +| +| +| +| +| +| +| +| +| 1733 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1734 |Redundant Paths | | | | +| +| | | +| +| | | 1735 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1736 |Security Measures | | | | | | | | | | | | 1737 +----------------------------+--+--+--+--+--+--+--+--+--+--+--+ 1739 Figure 5: Mapping Between Themes and Attacks 1741 8.3. Security Considerations for OAM Traffic 1743 This section considers DetNet-specific security considerations for 1744 packet traffic that is generated and transmitted over a DetNet as 1745 part of OAM (Operations, Administration and Maintenance). For 1746 purposes of this discussion, OAM traffic falls into one of two basic 1747 types: 1749 o OAM traffic generated by the network itself. The additional 1750 bandwidth required for such packets is added by the network 1751 administration, presumably transparent to the customer. Security 1752 considerations for such traffic are not DetNet-specific (apart 1753 from such traffic being subject to the same DetNet-specific 1754 security considerations as any other DetNet data flow) and are 1755 thus not covered in this document. 1757 o OAM traffic generated by the customer. From a DetNet security 1758 point of view, DetNet security considerations for such traffic are 1759 exactly the same as for any other customer data flows. 1761 Thus OAM traffic presents no additional (i.e. OAM-specific) DetNet 1762 security considerations. 1764 9. DetNet Technology-Specific Threats 1766 Section 5, Security Threats, described threats which are independent 1767 of a DetNet implementation. This section considers threats 1768 specifically related to the IP- and MPLS-specific aspects of DetNet 1769 implementations. 1771 The primary security considerations for the data plane specifically 1772 are to maintain the integrity of the data and the delivery of the 1773 associated DetNet service traversing the DetNet network. 1775 The primary relevant differences between IP and MPLS implementations 1776 are in flow identification and OAM methodologies. 1778 As noted in [RFC8655], DetNet operates at the IP layer 1779 ([I-D.ietf-detnet-ip]) and delivers service over sub-layer 1780 technologies such as MPLS ([I-D.ietf-detnet-mpls]) and IEEE 802.1 1781 Time-Sensitive Networking (TSN) ([I-D.ietf-detnet-ip-over-tsn]). 1782 Application flows can be protected through whatever means are 1783 provided by the layer and sub-layer technologies. For example, 1784 technology-specific encryption may be used, such as that provided by 1785 IPSec [RFC4301] for IP flows and/or by an underlying sub-net using 1786 MACSec [IEEE802.1AE-2018] for IP over Ethernet (Layer-2) flows. 1788 However, if the DetNet nodes cannot decrypt IPsec traffic, IPSec may 1789 not be a valid option; this is because the DetNet IP data plane 1790 identifies flows via a 6-tuple that consists of two IP addresses, the 1791 transport protocol ID, two transport protocol port numbers and the 1792 DSCP in the IP header. When IPsec is used, the transport header is 1793 encrypted and the next protocol ID is an IPsec protocol, usually ESP, 1794 and not a transport protocol (e.g., neither TCP nor UDP, etc.) 1795 leaving only three components of the 6-tuple, which are the two IP 1796 addresses and the DSCP, which are in general not sufficient to 1797 identify a DetNet flow. 1799 Sections below discuss threats specific to IP and MPLS in more 1800 detail. 1802 9.1. IP 1804 The IP protocol has a long history of security considerations and 1805 architectural protection mechanisms. From a data plane perspective 1806 DetNet does not add or modify any IP header information, and its use 1807 as a DetNet Data Plane does not introduce any new security issues 1808 that were not there before, apart from those already described in the 1809 data-plane-independent threats section Section 5, Security Threats. 1811 Thus the security considerations for a DetNet based on an IP data 1812 plane are purely inherited from the rich IP Security literature and 1813 code/application base, and the data-plane-independent section of this 1814 document. 1816 Maintaining security for IP segments of a DetNet may be more 1817 challenging than for the MPLS segments of the network, given that the 1818 IP segments of the network may reach the edges of the network, which 1819 are more likely to involve interaction with potentially malevolent 1820 outside actors. Conversely MPLS is inherently more secure than IP 1821 since it is internal to routers and it is well-known how to protect 1822 it from outside influence. 1824 Another way to look at DetNet IP security is to consider it in the 1825 light of VPN security; as an industry we have a lot of experience 1826 with VPNs running through networks with other VPNs, it is well known 1827 how to secure the network for that. However for a DetNet we have the 1828 additional subtlety that any possible interaction of one packet with 1829 another can have a potentially deleterious effect on the time 1830 properties of the flows. So the network must provide sufficient 1831 isolation between flows, for example by protecting the forwarding 1832 bandwidth and related resources so that they are available to detnet 1833 traffic, by whatever means are appropriate for that network's data 1834 plane. 1836 In a VPN, bandwidth is generally guaranteed over a period of time, 1837 whereas in DetNet it is not aggregated over time. This implies that 1838 any VPN-type protection mechanism must also maintain the DetNet 1839 timing constraints. 1841 9.2. MPLS 1843 An MPLS network carrying DetNet traffic is expected to be a "well- 1844 managed" network. Given that this is the case, it is difficult for 1845 an attacker to pass a raw MPLS encoded packet into a network because 1846 operators have considerable experience at excluding such packets at 1847 the network boundaries, as well as excluding MPLS packets being 1848 inserted through the use of a tunnel. 1850 MPLS security is discussed extensively in [RFC5920] ("Security 1851 Framework for MPLS and GMPLS Networks") to which the reader is 1852 referred. 1854 [RFC6941] builds on [RFC5920] by providing additional security 1855 considerations that are applicable to the MPLS-TP extensions 1856 appropriate to the MPLS Transport Profile [RFC5921], and thus to the 1857 operation of DetNet over some types of MPLS network. 1859 [RFC5921] introduces to MPLS new Operations, Administration, and 1860 Maintenance (OAM) capabilities, a transport-oriented path protection 1861 mechanism, and strong emphasis on static provisioning supported by 1862 network management systems. 1864 The operation of DetNet over an MPLS network is modeled on the 1865 operation of multi-segment pseudowires (MS-PW). Thus for guidance on 1866 securing the DetNet elements of DetNet over MPLS the reader is 1867 referred to the MS-PW security mechanisms as defined in [RFC4447], 1868 [RFC3931], [RFC3985], [RFC6073], and [RFC6478]. 1870 Having attended to the conventional aspects of network security it is 1871 necessary to attend to the dynamic aspects. The closest experience 1872 that the IETF has with securing protocols that are sensitive to 1873 manipulation of delay are the two way time transfer protocols (TWTT), 1874 which are NTP [RFC5905] and Precision Time Protocol [IEEE1588]. The 1875 security requirements for these are described in [RFC7384]. 1877 One particular problem that has been observed in operational tests of 1878 TWTT protocols is the ability for two closely but not completely 1879 synchronized flows to beat and cause a sudden phase hit to one of the 1880 flows. This can be mitigated by the careful use of a scheduling 1881 system in the underlying packet transport. 1883 Further consideration of protection against dynamic attacks is work 1884 in progress. 1886 10. IANA Considerations 1888 This memo includes no requests from IANA. 1890 11. Security Considerations 1892 The security considerations of DetNet networks are presented 1893 throughout this document. 1895 12. Contributors 1897 The Editor would like to recognize the contributions of the following 1898 individuals to this draft. 1900 Andrew J. Hacker (MistIQ Technologies, Inc) 1901 Harrisburg, PA, USA 1902 email ajhacker@mistiqtech.com, 1903 web http://www.mistiqtech.com 1905 Subir Das (Applied Communication Sciences) 1906 150 Mount Airy Road, Basking Ridge 1907 New Jersey, 07920, USA 1908 email sdas@appcomsci.com 1910 John Dowdell (Airbus Defence and Space) 1911 Celtic Springs, Newport, NP10 8FZ, United Kingdom 1912 email john.dowdell.ietf@gmail.com 1914 Henrik Austad (SINTEF Digital) 1915 Klaebuveien 153, Trondheim, 7037, Norway 1916 email henrik@austad.us 1918 Norman Finn 1919 email nfinn@nfinnconsulting.com 1921 Stewart Bryant 1922 Futurewei Technologies 1923 email: stewart.bryant@gmail.com 1925 David Black 1926 Dell EMC 1927 176 South Street, Hopkinton, MA 01748, USA 1928 email: david.black@dell.com 1930 Carsten Bormann 1932 13. Informative References 1934 [ARINC664P7] 1935 ARINC, "ARINC 664 Aircraft Data Network, Part 7, Avionics 1936 Full-Duplex Switched Ethernet Network", 2009. 1938 [I-D.ietf-detnet-data-plane-framework] 1939 Varga, B., Farkas, J., Berger, L., Malis, A., and S. 1940 Bryant, "DetNet Data Plane Framework", draft-ietf-detnet- 1941 data-plane-framework-06 (work in progress), May 2020. 1943 [I-D.ietf-detnet-flow-information-model] 1944 Varga, B., Farkas, J., Cummings, R., Jiang, Y., and D. 1945 Fedyk, "DetNet Flow Information Model", draft-ietf-detnet- 1946 flow-information-model-10 (work in progress), May 2020. 1948 [I-D.ietf-detnet-ip] 1949 Varga, B., Farkas, J., Berger, L., Fedyk, D., and S. 1950 Bryant, "DetNet Data Plane: IP", draft-ietf-detnet-ip-06 1951 (work in progress), April 2020. 1953 [I-D.ietf-detnet-ip-over-tsn] 1954 Varga, B., Farkas, J., Malis, A., and S. Bryant, "DetNet 1955 Data Plane: IP over IEEE 802.1 Time Sensitive Networking 1956 (TSN)", draft-ietf-detnet-ip-over-tsn-02 (work in 1957 progress), March 2020. 1959 [I-D.ietf-detnet-mpls] 1960 Varga, B., Farkas, J., Berger, L., Malis, A., Bryant, S., 1961 and J. Korhonen, "DetNet Data Plane: MPLS", draft-ietf- 1962 detnet-mpls-06 (work in progress), April 2020. 1964 [I-D.varga-detnet-service-model] 1965 Varga, B. and J. Farkas, "DetNet Service Model", draft- 1966 varga-detnet-service-model-02 (work in progress), May 1967 2017. 1969 [IEEE1588] 1970 IEEE, "IEEE 1588 Standard for a Precision Clock 1971 Synchronization Protocol for Networked Measurement and 1972 Control Systems Version 2", 2008. 1974 [IEEE802.1AE-2018] 1975 IEEE Standards Association, "IEEE Std 802.1AE-2018 MAC 1976 Security (MACsec)", 2018, 1977 . 1979 [IEEE802.1Qch-2017] 1980 IEEE Standards Association, "IEEE Standard for Local and 1981 metropolitan area networks--Bridges and Bridged Networks-- 1982 Amendment 29: Cyclic Queuing and Forwarding", 2017, 1983 . 1985 [MIRAI] krebsonsecurity.com, "https://krebsonsecurity.com/2016/10/ 1986 hacked-cameras-dvrs-powered-todays-massive-internet- 1987 outage/", 2016. 1989 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 1990 "Definition of the Differentiated Services Field (DS 1991 Field) in the IPv4 and IPv6 Headers", RFC 2474, 1992 DOI 10.17487/RFC2474, December 1998, 1993 . 1995 [RFC2475] Blake, S., Black, D., Carlson, M., Davies, E., Wang, Z., 1996 and W. Weiss, "An Architecture for Differentiated 1997 Services", RFC 2475, DOI 10.17487/RFC2475, December 1998, 1998 . 2000 [RFC3552] Rescorla, E. and B. Korver, "Guidelines for Writing RFC 2001 Text on Security Considerations", BCP 72, RFC 3552, 2002 DOI 10.17487/RFC3552, July 2003, 2003 . 2005 [RFC3931] Lau, J., Ed., Townsley, M., Ed., and I. Goyret, Ed., 2006 "Layer Two Tunneling Protocol - Version 3 (L2TPv3)", 2007 RFC 3931, DOI 10.17487/RFC3931, March 2005, 2008 . 2010 [RFC3985] Bryant, S., Ed. and P. Pate, Ed., "Pseudo Wire Emulation 2011 Edge-to-Edge (PWE3) Architecture", RFC 3985, 2012 DOI 10.17487/RFC3985, March 2005, 2013 . 2015 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 2016 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 2017 December 2005, . 2019 [RFC4447] Martini, L., Ed., Rosen, E., El-Aawar, N., Smith, T., and 2020 G. Heron, "Pseudowire Setup and Maintenance Using the 2021 Label Distribution Protocol (LDP)", RFC 4447, 2022 DOI 10.17487/RFC4447, April 2006, 2023 . 2025 [RFC5905] Mills, D., Martin, J., Ed., Burbank, J., and W. Kasch, 2026 "Network Time Protocol Version 4: Protocol and Algorithms 2027 Specification", RFC 5905, DOI 10.17487/RFC5905, June 2010, 2028 . 2030 [RFC5920] Fang, L., Ed., "Security Framework for MPLS and GMPLS 2031 Networks", RFC 5920, DOI 10.17487/RFC5920, July 2010, 2032 . 2034 [RFC5921] Bocci, M., Ed., Bryant, S., Ed., Frost, D., Ed., Levrau, 2035 L., and L. Berger, "A Framework for MPLS in Transport 2036 Networks", RFC 5921, DOI 10.17487/RFC5921, July 2010, 2037 . 2039 [RFC6073] Martini, L., Metz, C., Nadeau, T., Bocci, M., and M. 2040 Aissaoui, "Segmented Pseudowire", RFC 6073, 2041 DOI 10.17487/RFC6073, January 2011, 2042 . 2044 [RFC6478] Martini, L., Swallow, G., Heron, G., and M. Bocci, 2045 "Pseudowire Status for Static Pseudowires", RFC 6478, 2046 DOI 10.17487/RFC6478, May 2012, 2047 . 2049 [RFC6941] Fang, L., Ed., Niven-Jenkins, B., Ed., Mansfield, S., Ed., 2050 and R. Graveman, Ed., "MPLS Transport Profile (MPLS-TP) 2051 Security Framework", RFC 6941, DOI 10.17487/RFC6941, April 2052 2013, . 2054 [RFC7384] Mizrahi, T., "Security Requirements of Time Protocols in 2055 Packet Switched Networks", RFC 7384, DOI 10.17487/RFC7384, 2056 October 2014, . 2058 [RFC8578] Grossman, E., Ed., "Deterministic Networking Use Cases", 2059 RFC 8578, DOI 10.17487/RFC8578, May 2019, 2060 . 2062 [RFC8655] Finn, N., Thubert, P., Varga, B., and J. Farkas, 2063 "Deterministic Networking Architecture", RFC 8655, 2064 DOI 10.17487/RFC8655, October 2019, 2065 . 2067 Authors' Addresses 2069 Tal Mizrahi 2070 Huawei Network.IO Innovation Lab 2072 Email: tal.mizrahi.phd@gmail.com 2074 Ethan Grossman (editor) 2075 Dolby Laboratories, Inc. 2076 1275 Market Street 2077 San Francisco, CA 94103 2078 USA 2080 Phone: +1 415 645 4726 2081 Email: ethan.grossman@dolby.com 2082 URI: http://www.dolby.com