idnits 2.17.1 draft-carpenter-limited-domains-12.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (November 30, 2019) is 1608 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Outdated reference: A later version (-09) exists of draft-fz-6man-ipv6-alt-mark-01 == Outdated reference: A later version (-07) exists of draft-herbert-fast-04 == Outdated reference: A later version (-03) exists of draft-herbert-ipv4-eh-01 == Outdated reference: A later version (-30) exists of draft-ietf-anima-autonomic-control-plane-21 == Outdated reference: A later version (-06) exists of draft-ietf-detnet-data-plane-framework-03 == Outdated reference: A later version (-04) exists of draft-ietf-dmm-5g-uplane-analysis-03 == Outdated reference: A later version (-30) exists of draft-ietf-ipwave-vehicular-networking-12 == Outdated reference: A later version (-10) exists of draft-ietf-opsec-ipv6-eh-filtering-06 == Outdated reference: A later version (-28) exists of draft-ietf-spring-srv6-network-programming-05 == Outdated reference: A later version (-17) exists of draft-ietf-teas-enhanced-vpn-03 == Outdated reference: A later version (-02) exists of draft-smith-6man-in-flight-eh-insertion-harmful-01 == Outdated reference: A later version (-10) exists of draft-voyer-6man-extension-header-insertion-08 Summary: 0 errors (**), 0 flaws (~~), 13 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group B. Carpenter 3 Internet-Draft Univ. of Auckland 4 Intended status: Informational B. Liu 5 Expires: June 2, 2020 Huawei Technologies 6 November 30, 2019 8 Limited Domains and Internet Protocols 9 draft-carpenter-limited-domains-12 11 Abstract 13 There is a noticeable trend towards network requirements, behaviours 14 and semantics that are specific to a particular set of requirements 15 applied within a limited region of the Internet. Policies, default 16 parameters, the options supported, the style of network management 17 and security requirements may vary between such limited regions. 18 This document reviews examples of such limited domains (also known as 19 controlled environments), notes emerging solutions, and includes a 20 related taxonomy. It then briefly discusses the standardization of 21 protocols for limited domains. Finally, it shows the needs for a 22 precise definition of "limited domain membership" and for mechanisms 23 to allow nodes to join a domain securely and to find other members, 24 including boundary nodes. 26 This document is the product of the research of the authors. It has 27 been produced through discussions and consultation within the IETF, 28 but is not the product of IETF conensus. 30 Status of This Memo 32 This Internet-Draft is submitted in full conformance with the 33 provisions of BCP 78 and BCP 79. 35 Internet-Drafts are working documents of the Internet Engineering 36 Task Force (IETF). Note that other groups may also distribute 37 working documents as Internet-Drafts. The list of current Internet- 38 Drafts is at https://datatracker.ietf.org/drafts/current/. 40 Internet-Drafts are draft documents valid for a maximum of six months 41 and may be updated, replaced, or obsoleted by other documents at any 42 time. It is inappropriate to use Internet-Drafts as reference 43 material or to cite them other than as "work in progress." 45 This Internet-Draft will expire on June 2, 2020. 47 Copyright Notice 49 Copyright (c) 2019 IETF Trust and the persons identified as the 50 document authors. All rights reserved. 52 This document is subject to BCP 78 and the IETF Trust's Legal 53 Provisions Relating to IETF Documents 54 (https://trustee.ietf.org/license-info) in effect on the date of 55 publication of this document. Please review these documents 56 carefully, as they describe your rights and restrictions with respect 57 to this document. Code Components extracted from this document must 58 include Simplified BSD License text as described in Section 4.e of 59 the Trust Legal Provisions and are provided without warranty as 60 described in the Simplified BSD License. 62 Table of Contents 64 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 65 2. Failure Modes in Today's Internet . . . . . . . . . . . . . . 4 66 3. Examples of Limited Domain Requirements . . . . . . . . . . . 5 67 4. Examples of Limited Domain Solutions . . . . . . . . . . . . 9 68 5. The Scope of Protocols in Limited Domains . . . . . . . . . . 12 69 6. Functional Requirements of Limited Domains . . . . . . . . . 14 70 7. Security Considerations . . . . . . . . . . . . . . . . . . . 16 71 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 17 72 9. Contributor . . . . . . . . . . . . . . . . . . . . . . . . . 17 73 10. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 17 74 11. Informative References . . . . . . . . . . . . . . . . . . . 17 75 Appendix A. Change log [RFC Editor: Please remove] . . . . . . . 24 76 Appendix B. Taxonomy of Limited Domains . . . . . . . . . . . . 26 77 B.1. The Domain as a Whole . . . . . . . . . . . . . . . . . . 27 78 B.2. Individual Nodes . . . . . . . . . . . . . . . . . . . . 27 79 B.3. The Domain Boundary . . . . . . . . . . . . . . . . . . . 27 80 B.4. Topology . . . . . . . . . . . . . . . . . . . . . . . . 28 81 B.5. Technology . . . . . . . . . . . . . . . . . . . . . . . 28 82 B.6. Connection to the Internet . . . . . . . . . . . . . . . 28 83 B.7. Security, Trust and Privacy Model . . . . . . . . . . . . 29 84 B.8. Operations . . . . . . . . . . . . . . . . . . . . . . . 29 85 B.9. Making use of this taxonomy . . . . . . . . . . . . . . . 29 86 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 29 88 1. Introduction 90 As the Internet continues to grow and diversify, with a realistic 91 prospect of tens of billions of nodes being connected directly and 92 indirectly, there is a noticeable trend towards network-specific and 93 local requirements, behaviours and semantics. The word "local" 94 should be understood in a special sense, however. In some cases it 95 may refer to geographical and physical locality - all the nodes in a 96 single building, on a single campus, or in a given vehicle. In other 97 cases it may refer to a defined set of users or nodes distributed 98 over a much wider area, but drawn together by a single virtual 99 network over the Internet, or a single physical network running in 100 parallel with the Internet. We expand on these possibilities below. 101 To capture the topic, this document refers to such networks as 102 "limited domains". Of course a similar situation may arise for a 103 network that is completely disconnected from the Internet, but that 104 is not our direct concern here. However, it should not be forgotten 105 that interoperability is needed even within a disconnected network. 107 Some people have concerns about splintering of the Internet along 108 political or linguistic boundaries by mechanisms that block the free 109 flow of information. That is not the topic of this document, which 110 does not discuss filtering mechanisms and does not apply to protocols 111 that are designed for use across the whole Internet. It is only 112 concerned with domains that have specific technical requirements. 114 The word "domain" in this document does not refer to naming domains 115 in the DNS, although in some cases a limited domain might 116 incidentally be congruent with a DNS domain. In particular, with a 117 "split horizon" DNS configuration [RFC6950], the split might be at 118 the edge of a limited domain. A recent proposal for defining 119 definite perimeters within the DNS namespace 120 [I-D.dcrocker-dns-perimeter] might also be considered to be a limited 121 domain mechanism. 123 Another term that has been used in some contexts is "controlled 124 environment". For example, [RFC8085] uses this to delimit the 125 operational scope within which a particular tunnel encapsulation 126 might be used. A specific example is GRE-in-UDP encapsulation 127 [RFC8086] which explicitly states that "The controlled environment 128 has less restrictive requirements than the general Internet." For 129 example, non-congestion-controlled traffic might be acceptable within 130 the controlled environment. The same phrase has been used to delimit 131 the useful scope of quality of service or security protocols, e.g. 132 [RFC6398], [RFC6455]. It is not necessarily the case that protocols 133 will fail to operate outside the controlled environment, but rather 134 that they might not operate optimally. In this document, we assume 135 that "limited domain" and "controlled environment" mean the same 136 thing in practice. The term "managed network" has been used in a 137 similar way, e.g. [RFC6947]. In the context of secure multicast, a 138 "group domain of interpretation" is defined by [RFC6407]. 140 Yet more definitions of types of domain are to be found in the 141 routing area, such as [RFC4397], [RFC4427], and [RFC4655]. We 142 conclude that the notion of a limited domain is very widespread in 143 many aspects of Internet technology. 145 The requirements of limited domains will depend on the deployment 146 scenario. Policies, default parameters, and the options supported 147 may vary. Also, the style of network management may vary, between a 148 completely unmanaged network, one with fully autonomic management, 149 one with traditional central management, and mixtures of the above. 150 Finally, the requirements and solutions for security and privacy may 151 vary. 153 This document analyses and discusses some of the consequences of this 154 trend, and how it may impact the idea of universal interoperability 155 in the Internet. Firstly we list examples of limited domain 156 scenarios and of technical solutions for limited domains, with the 157 main focus being the Internet layer of the protocol stack. An 158 appendix provides a taxonomy of the features to be found in limited 159 domains. With this background, we discuss the resulting challenge to 160 the idea that all Internet standards must be universal in scope and 161 applicability. To the contrary, we assert that some protocols, 162 although needing to be standardized and interoperable, also need to 163 be specifically limited in their applicability. This implies that 164 the concepts of a limited domain, and of its membership, need to be 165 formalised and supported by secure mechanisms. While this document 166 does not propose a design for such mechanisms, it does outline some 167 functional requirements. 169 This document is the product of the research of the authors. It has 170 been produced through discussions and consultation within the IETF, 171 but is not the product of IETF conensus. 173 2. Failure Modes in Today's Internet 175 Today, the Internet does not have a well-defined concept of limited 176 domains. One result of this is that certain protocols and features 177 fail on certain paths. Earlier analyses of this topic have focused 178 either on the loss of transparency of the Internet [RFC2775], 179 [RFC4924] or on the middleboxes responsible for that loss [RFC3234], 180 [RFC7663], [RFC8517]. Unfortunately the problems persist, both in 181 application protocols, and even in very fundamental mechanisms. For 182 example, the Internet is not transparent to IPv6 extension headers 183 [RFC7872], and Path MTU Discovery has been unreliable for many years 184 [RFC2923], [RFC4821]. IP fragmentation is also unreliable 185 [I-D.ietf-intarea-frag-fragile], and problems in TCP MSS negotiation 186 have been reported [I-D.andrews-tcp-and-ipv6-use-minmtu]. 188 On the security side, the widespread insertion of firewalls at domain 189 boundaries that are perceived by humans but unknown to protocols 190 results in arbitrary failure modes as far as the application layer is 191 concerned. There are operational recommendations and practices that 192 effectively guarantee arbitrary failures in realistic scenarios 193 [I-D.ietf-opsec-ipv6-eh-filtering]. 195 Investigations of the unreliability of IP fragmentation 196 [I-D.ietf-intarea-frag-fragile] and the filtering of IPv6 extension 197 headers [RFC7872] strongly suggest that at least for some protocol 198 elements, transparency is a lost cause and middleboxes are here to 199 stay. In the following two sections, we show that some application 200 environments require protocol features that cannot, or should not, 201 cross the whole Internet. 203 3. Examples of Limited Domain Requirements 205 This section describes various examples where limited domain 206 requirements can easily be identified, either based on an application 207 scenario or on a technical imperative. It is of course not a 208 complete list, and it is presented in an arbitrary order, loosely 209 from smaller to bigger. 211 1. A home network. It will be mainly unmanaged, constructed by a 212 non-specialist. It must work with devices "out of the box" as 213 shipped by their manufacturers and must create adequate security 214 by default. Remote access may be required. The requirements 215 and applicable principles are summarised in [RFC7368]. 217 2. A small office network. This is sometimes very similar to a 218 home network, if whoever is in charge has little or no 219 specialist knowledge, but may have differing security and 220 privacy requirements. In other cases it may be professionally 221 constructed using recommended products and configurations, but 222 operate unmanaged. Remote access may be required. 224 3. A vehicle network. This will be designed by the vehicle 225 manufacturer but may include devices added by the vehicle's 226 owner or operator. Parts of the network will have demanding 227 performance and reliability requirements with implications for 228 human safety. Remote access may be required to certain 229 functions, but absolutely forbidden for others. Communication 230 with other vehicles, roadside infrastructure, and external data 231 sources will be required. See 232 [I-D.ietf-ipwave-vehicular-networking] for a survey of use 233 cases. 235 4. Supervisory Control And Data Acquisition (SCADA) networks, and 236 other hard real time networks. These will exhibit specific 237 technical requirements, including tough real-time performance 238 targets. See for example [RFC8578] for numerous use cases. An 239 example is a building services network. This will be designed 240 specifically for a particular building, but using standard 241 components. Additional devices may need to be added at any 242 time. Parts of the network may have demanding reliability 243 requirements with implications for human safety. Remote access 244 may be required to certain functions, but absolutely forbidden 245 for others. An extreme example is a network used for Virtual 246 Reality or Augmented Reality applications, where the latency 247 requirements are very stringent. 249 5. Sensor networks. The two preceding cases will all include 250 sensors, but some networks may be specifically limited to 251 sensors and the collection and processing of sensor data. They 252 may be in remote or technically challenging locations and 253 installed by non-specialists. 255 6. Internet of Things (IoT) networks. While this term is very 256 flexible and covers many innovative types of network, including 257 ad hoc networks that are formed spontaneously, and some 258 applications of 5G technology, it seems reasonable to expect 259 that IoT edge networks will have special requirements and 260 protocols that are useful only within a specific domain, and 261 that these protocols cannot, and for security reasons should 262 not, run over the Internet as a whole. 264 7. An important subclass of IoT networks consists of constrained 265 networks [RFC7228] in which the nodes are limited in power 266 consumption and communications bandwidth, and are therefore 267 limited to using very frugal protocols. 269 8. Delay tolerant networks may consist of domains that are 270 relatively isolated and constrained in power (e.g. deep space 271 networks) and are connected only intermittently to the outside, 272 with a very long latency on such connections [RFC4838]. Clearly 273 the protocol requirements and possibilities are very specialised 274 in such networks. 276 9. "Traditional" enterprise and campus networks, which may be 277 spread over many kilometres and over multiple separate sites, 278 with multiple connections to the Internet. Interestingly, the 279 IETF appears never to have analysed this long-established class 280 of networks in a general way, except in connection with IPv6 281 deployment (e.g. [RFC7381]). 283 10. Inappropriate standards. A situation that can arise in an 284 enterprise network is that the Internet-wide solution for a 285 particular requirement may either fail locally, or be much more 286 complicated than is necessary. An example is that the 287 complexity induced by a mechanism such as ICE [RFC8445] is not 288 justified within such a network. Furthermore, ICE cannot be 289 used in some cases because candidate addresses are not known 290 before a call is established, so a different local solution is 291 essential [RFC6947]. 293 11. Managed wide area networks run by service providers for 294 enterprise services such as layer 2 (Ethernet, etc.) point-to- 295 point pseudowires, multipoint layer 2 Ethernet VPNs using VPLS 296 or EVPN, and layer 3 IP VPNs. These are generally characterized 297 by service level agreements for availability and packet loss, 298 and possibly for multicast service. These are different from 299 the previous case in that they mostly run over MPLS 300 infrastructures and the requirements for these services are 301 well-defined by the IETF. 303 12. Data centres and hosting centres, or distributed services acting 304 as such centres. These will have high performance, security and 305 privacy requirements and will typically include large numbers of 306 independent "tenant" networks overlaid on shared infrastructure. 308 13. Content Delivery Networks (CDNs), comprising distributed data 309 centres and the paths between them, spanning thousands of 310 kilometres, with numerous connections to the Internet. 312 14. Massive Web Service Provider Networks. This is a small class of 313 networks with well known trademarked names, combining aspects of 314 distributed enterprise networks, data centres and CDNs. They 315 have their own international networks bypassing the generic 316 carriers. Like CDNs, they have numerous connections to the 317 Internet, typically offering a tailored service in each economy. 319 Three other aspects, while not tied to specific network types, also 320 strongly depend on the concept of limited domains: 322 1. Many of the above types of network may be extended throughout the 323 Internet by a variety of virtual private network (VPN) 324 techniques. Therefore we argue that limited domains may overlap 325 each other in an arbitrary fashion by use of virtualization 326 techniques. As noted above in the discussion of controlled 327 environments, specific tunneling and encapsulation techniques may 328 be tailored for use within a given domain. 330 2. Intent Based Networking. In this concept, a network domain is 331 configured and managed in accordance with an abstract policy 332 known as "Intent", to ensure that the network performs as 333 required [I-D.clemm-nmrg-dist-intent]. Whatever technologies are 334 used to support this, they will be applied within the domain 335 boundary, even if the services supported in the domain are 336 globally accessible. 338 3. Network Slicing. A network slice is a form of virtual network 339 that consists of a managed set of resources carved off from a 340 larger network [I-D.ietf-teas-enhanced-vpn]. This is expected to 341 be significant in 5G deployments 342 [I-D.ietf-dmm-5g-uplane-analysis]. Whatever technologies are 343 used to support slicing, they will require a clear definition of 344 the boundary of a given slice within a larger domain. 346 While it is clearly desirable to use common solutions, and therefore 347 common standards, wherever possible, it is increasingly difficult to 348 do so while satisfying the widely varying requirements outlined 349 above. However, there is a tendency when new protocols and protocol 350 extensions are proposed to always ask the question "How will this 351 work across the open Internet?" This document suggests that this is 352 not always the best question. There are protocols and extensions 353 that are not intended to work across the open Internet. On the 354 contrary, their requirements and semantics are specifically limited 355 (in the sense defined above). 357 A common argument is that if a protocol is intended for limited use, 358 the chances are very high that it will in fact be used (or misused) 359 in other scenarios including the so-called open Internet. This is 360 undoubtedly true and means that limited use is not an excuse for bad 361 design or poor security. In fact, a limited use requirement 362 potentially adds complexity to both the protocol and its security 363 design, as discussed later. 365 Nevertheless, because of the diversity of limited domains with 366 specific requirements that is now emerging, specific standards (and 367 ad hoc standards) will probably emerge for different types of domain. 368 There will be attempts to capture each market sector, but the market 369 will demand standardized solutions within each sector. In addition, 370 operational choices will be made that can in fact only work within a 371 limited domain. The history of RSVP [RFC2205] illustrates that a 372 standard defined as if it could work over the open Internet might not 373 in fact do so. In general we can no longer assume that a protocol 374 designed according to classical Internet guidelines will in fact work 375 reliably across the network as a whole. However, the "open Internet" 376 must remain as the universal method of interconnection. Reconciling 377 these two aspects is a major challenge. 379 4. Examples of Limited Domain Solutions 381 This section lists various examples of specific limited domain 382 solutions that have been proposed or defined. It intentionally does 383 not include Layer 2 technology solutions, which by definition apply 384 to limited domains. It is worth noting, however, that with recent 385 developments such as TRILL [RFC6325] or Shortest Path Bridging [SPB], 386 Layer 2 domains may become very large. 388 1. Differentiated Services. This mechanism [RFC2474] allows a 389 network to assign locally significant values to the 6-bit 390 Differentiated Services Code Point field in any IP packet. 391 Although there are some recommended codepoint values for 392 specific per-hop queue management behaviours, these are 393 specifically intended to be domain-specific codepoints with 394 traffic being classified, conditioned and mapped or re-marked at 395 domain boundaries (unless there is an inter-domain agreement 396 that makes mapping or re-marking unnecessary). 398 2. Integrated Services. Although it is not intrinsic in the design 399 of RSVP [RFC2205], it is clear from many years' experience that 400 Integrated Services can only be deployed successfully within a 401 limited domain that is configured with adequate equipment and 402 resources. 404 3. Network function virtualisation. As described in 405 [I-D.irtf-nfvrg-gaps-network-virtualization], this general 406 concept is an open research topic, in which virtual network 407 functions are orchestrated as part of a distributed system. 408 Inevitably such orchestration applies to an administrative 409 domain of some kind, even though cross-domain orchestration is 410 also a research area. 412 4. Service Function Chaining (SFC). This technique [RFC7665] 413 assumes that services within a network are constructed as 414 sequences of individual service functions within a specific SFC- 415 enabled domain such as a 5G domain. As that RFC states: 416 "Specific features may need to be enforced at the boundaries of 417 an SFC-enabled domain, for example to avoid leaking SFC 418 information". A Network Service Header (NSH) [RFC8300] is used 419 to encapsulate packets flowing through the service function 420 chain: "The intended scope of the NSH is for use within a single 421 provider's operational domain." 423 5. Firewall and Service Tickets (FAST). Such tickets would 424 accompany a packet to claim the right to traverse a network or 425 request a specific network service [I-D.herbert-fast]. They 426 would only be meaningful within a particular domain. 428 6. Data Centre Network Virtualization Overlays. A common 429 requirement in data centres that host many tenants (clients) is 430 to provide each one with a secure private network, all running 431 over the same physical infrastructure. [RFC8151] describes 432 various use cases for this, and specifications are under 433 development. These include use cases in which the tenant 434 network is physically split over several data centres, but which 435 must appear to the user as a single secure domain. 437 7. Segment Routing. This is a technique which "steers a packet 438 through an ordered list of instructions, called segments" 439 [RFC8402]. The semantics of these instructions are explicitly 440 local to a segment routing domain or even to a single node. 441 Technically, these segments or instructions are represented as 442 an MPLS label or an IPv6 address, which clearly adds a semantic 443 interpretation to them within the domain. 445 8. Autonomic Networking. As explained in 446 [I-D.ietf-anima-reference-model], an autonomic network is also a 447 security domain within which an autonomic control plane 448 [I-D.ietf-anima-autonomic-control-plane] is used by autonomic 449 service agents. These agents manage technical objectives, which 450 may be locally defined, subject to domain-wide policy. Thus the 451 domain boundary is important for both security and protocol 452 purposes. 454 9. Homenet. As shown in [RFC7368], a home networking domain has 455 specific protocol needs that differ from those in an enterprise 456 network or the Internet as a whole. These include the Home 457 Network Control Protocol (HNCP) [RFC7788] and a naming and 458 discovery solution [I-D.ietf-homenet-simple-naming]. 460 10. Creative uses of IPv6 features. As IPv6 enters more general 461 use, engineers notice that it has much more flexibility than 462 IPv4. Innovative suggestions have been made for: 464 * The flow label, e.g. [RFC6294]. 466 * Extension headers, e.g. for segment routing 467 [I-D.ietf-6man-segment-routing-header] or OAM marking 468 [I-D.fz-6man-ipv6-alt-mark]. 470 * Meaningful address bits, e.g. [I-D.jiang-semantic-prefix]. 471 Also, segment routing uses IPv6 addresses as segment 472 identifiers with specific local meanings [RFC8402]. 474 * If segment routing is used for network programming 475 [I-D.ietf-spring-srv6-network-programming], IPv6 extension 476 headers can support rather complex local functionality. 478 The case of the extension header is particularly interesting, 479 since its existence has been a major "selling point" for IPv6, 480 but it is notorious that new extension headers are virtually 481 impossible to deploy across the whole Internet [RFC7045], 482 [RFC7872]. It is worth noting that extension header filtering 483 is considered as an important security issue 484 [I-D.ietf-opsec-ipv6-eh-filtering]. There is considerable 485 appetite among vendors or operators to have flexibility in 486 defining extension headers for use in limited or specialised 487 domains, e.g. [I-D.voyer-6man-extension-header-insertion], 488 [BIGIP], and [I-D.li-6man-service-aware-ipv6-network]. Locally 489 significant hop-by-hop options are also envisaged, that would be 490 understood by routers inside a domain but not elsewhere, e.g., 491 [I-D.ioametal-ippm-6man-ioam-ipv6-options]. 493 11. Deterministic Networking (DetNet). The Deterministic Networking 494 Architecture [RFC8655] and encapsulation 495 [I-D.ietf-detnet-data-plane-framework] aim to support flows with 496 extremely low data loss rates and bounded latency, but only 497 within a part of the network that is "DetNet aware". Thus, as 498 for differentiated services above, the concept of a domain is 499 fundamental. 501 12. Provisioning Domains (PvDs). An architecture for Multiple 502 Provisioning Domains has been defined [RFC7556] to allow hosts 503 attached to multiple networks to learn explicit details about 504 the services provided by each of those networks. 506 13. Address Scopes. For completeness, we mention that, particularly 507 in IPv6, some addresses have explicitly limited scope. In 508 particular, link-local addresses are limited to a single 509 physical link [RFC4291], and Unique Local Addresses [RFC4193] 510 are limited to a somewhat loosely defined local site scope. 511 Previously, site-local addresses were defined, but they were 512 obsoleted precisely because of "the fuzzy nature of the site 513 concept" [RFC3879]. Multicast addresses also have explicit 514 scoping [RFC4291]. 516 14. As an application layer example, consider streaming services 517 such as IPTV infrastructures that rely on standard protocols, 518 but for which access is not globally available. 520 All of these suggestions are only viable within a specified domain. 521 Neverthless, all of them are clearly intended for multivendor 522 implementation on thousands or millions of network domains, so 523 interoperable standardization would be beneficial. This argument 524 might seem irrelevant to private or proprietary implementations, but 525 these have a strong tendency to become de facto standards if they 526 succeed, so the arguments of this document still apply. 528 5. The Scope of Protocols in Limited Domains 530 One consequence of the deployment of limited domains in the Internet 531 is that some protocols will be designed, extended or configured so 532 that they only work correctly between end systems in such domains. 533 This is to some extent encouraged by some existing standards and by 534 the assignment of code points for local or experimental use. In any 535 case it cannot be prevented. Also, by endorsing efforts such as 536 Service Function Chaining, Segment Routing and Deterministic 537 Networking, the IETF is in effect encouraging such deployments. 538 Furthermore, it seems inevitable, if the "Internet of Things" becomes 539 reality, that millions of edge networks containing completely novel 540 types of node will be connected to the Internet; each one of these 541 edge networks will be a limited domain. 543 It is therefore appropriate to discuss whether protocols or protocol 544 extensions should sometimes be standardized to interoperate only 545 within a Limited Domain boundary. Such protocols would not be 546 required to interoperate across the Internet as a whole. Various 547 scenarios could then arise if there are multiple domains using the 548 limited-domain protocol in question: 550 A. If a domain is split into two parts connected over the 551 Internet directly at the IP layer (i.e. with no tunnel 552 encapsulating the packets), a limited-domain protocol could be 553 operated between those two parts regardless of its special nature, 554 as long as it respects standard IP formats and is not arbitrarily 555 blocked by firewalls. A simple example is any protocol using a 556 port number assigned to a specific non-IETF protocol. 558 Such a protocol could reasonably be described as an "inter-domain" 559 protocol because the Internet is transparent to it, even if it is 560 meaningless except in the two limited domains. This is of course 561 nothing new in the Internet architecture. 563 B. If a limited-domain protocol does not respect standard IP 564 formats (for example, if it includes a non-standard IPv6 extension 565 header), it could not be operated between two domains connected 566 over the Internet directly at the IP layer. 568 Such a protocol could reasonably be described as an "intra-domain" 569 protocol, and the Internet is opaque to it. 571 C. If a limited-domain protocol is clearly specified to be 572 invalid outside its domain of origin, neither scenario A nor B 573 applies. The only solution would be a single virtual domain. For 574 example, an encapsulating tunnel between two domains could be used 575 to create the virtual domain. Also, nodes at the domain boundary 576 must drop all packets using the limited-domain protocol. 578 D. If a limited-domain protocol has domain-specific variants, 579 such that implementations in different domains could not 580 interoperate if those domains were unified by some mechanism as in 581 scenario C, the protocol is not interoperable in the normal sense. 582 If two domains using it were merged, the protocol might fail 583 unpredictably. A simple example is any protocol using a port 584 number assigned for experimental use. Related issues are 585 discussed in [RFC5704], including the complex example of Transport 586 MPLS. 588 To provide a widespread example, consider Differentiated Services 589 [RFC2474]. A packet containing any value whatever in the 6 bits of 590 the Differentiated Service Code Point (DSCP) is well-formed and falls 591 into scenario A. However, because the semantics of DSCP values are 592 locally significant, the packet also falls into scenario D. In fact, 593 differentiated services are only interoperable across domain 594 boundaries if there is a corresponding agreement between the 595 operators; otherwise a specific gateway function is required for 596 meaningful interoperability. Much more detailed discussion is to be 597 found in [RFC2474] and [RFC8100]. 599 To provide a provocative example, consider the proposal in 600 [I-D.voyer-6man-extension-header-insertion] that the restrictions in 601 [RFC8200] should be relaxed to allow IPv6 extension headers to be 602 inserted on the fly in IPv6 packets. If this is done in such a way 603 that the affected packets can never leave the specific limited domain 604 in which they were modified, scenario C applies. If the semantic 605 content of the inserted headers is locally defined, scenario D also 606 applies. In neither case is the Internet outside the limited domain 607 disturbed. However, inside the domain nodes must understand the 608 variant protocol. Unless it is standardized as a formal version, 609 with all the complexity that implies [RFC6709], the nodes must all be 610 non-standard to the extent of understanding the variant protocol. 611 For the example of IPv6 header insertion, that means non-compliance 612 with [RFC8200] within the domain, even if the inserted headers are 613 themselves fully compliant. Apart from the issue of formal 614 compliance, such deviations from documented standard behaviour might 615 lead to significant debugging issues. The possible practical impact 616 of the header insertion example is explored in 617 [I-D.smith-6man-in-flight-eh-insertion-harmful]. 619 The FAST proposal mentioned in Paragraph 5 of Section 4 is also an 620 interesting case study. The semantics of FAST tickets 621 [I-D.herbert-fast] have limited scope. However, they are designed in 622 a way that in principle allows them to traverse the open Internet, as 623 standardized IPv6 hop-by-hop options or even as a proposed form of 624 IPv4 extension header [I-D.herbert-ipv4-eh]. Whether such options 625 can be used reliably across the open Internet remains unclear 626 [I-D.ietf-opsec-ipv6-eh-filtering]. 628 We conclude that it is reasonable to explicitly define limited-domain 629 protocols, either as standards or as proprietary mechanisms, as long 630 as they describe which of the above scenarios apply and they clarify 631 how the domain is defined. As long as all relevant standards are 632 respected outside the domain boundary, a well-specified limited- 633 domain protocol need not damage the rest of the Internet. However, 634 as described in the next section, mechanisms are needed to support 635 domain membership operations. 637 Note that this conclusion is not a recommendation to abandon the 638 normal goal that a standardized protocol should be global in scope 639 and able to interoperate across the open Internet. It is simply a 640 recognition that this will not always be the case. 642 6. Functional Requirements of Limited Domains 644 Noting that limited-domain protocols have been defined in the past, 645 and that others will undoubtedly be defined in the future, it is 646 useful to consider how a protocol can be made aware of the domain 647 within which it operates, and how the domain boundary nodes can be 648 identified. As the taxonomy in Appendix B shows, there are numerous 649 aspects to a domain. However, we can identify some generally 650 required features and functions that would apply partially or 651 completely to many cases. 653 Today, where limited domains exist, they are essentially created by 654 careful configuration of boundary routers and firewalls. If a domain 655 is characterized by one or more address prefixes, address assignment 656 to hosts must also be carefully managed. This is an error-prone 657 method and a combination of configuration errors and default routing 658 can lead to unwanted traffic escaping the domain. Our basic 659 assumption is therefore that it should be possible for domains to be 660 created and managed automatically, with minimal human configuration. 661 We now discuss requirements for automating domain creation and 662 management. 664 Firstly, if we drew a topology map, any domain -- virtual or physical 665 -- will have a well defined boundary between "inside" and "outside". 666 However, that boundary in itself has no technical meaning. What 667 matters in reality is whether a node is a member of the domain, and 668 whether it is at the boundary between the domain and the rest of the 669 Internet. Thus the boundary in itself does not need to be 670 identified, but boundary nodes face both inwards and outwards. 671 Inside the domain, a sending node needs to know whether it is sending 672 to an inside or outside destination; and a receiving node needs to 673 know whether a packet originated inside or outside. Also, a boundary 674 node needs to know which of its interfaces are inward-facing or 675 outward-facing. It is irrelevant whether the interfaces involved are 676 physical or virtual. 678 To underline that domain boundaries need to be identifiable, consider 679 the statement from the Deterministic Networking Problem Statement 680 [RFC8557] that "there is still a lack of clarity regarding the limits 681 of a domain where a deterministic path can be set up". This remark 682 can certainly be generalised. 684 With this perspective, we can list some general functional 685 requirements. An underlying assumption here is that domain 686 membership operations should be cryptographically secured; a domain 687 without such security cannot be reliably protected from attack. 689 1. Domain Identity. A domain must have a unique and verifiable 690 identifier; effectively this should be a public key for the 691 domain. Without this, there is no way to secure domain 692 operations and domain membership. The holder of the 693 corresponding private key becomes the trust anchor for the 694 domain. 696 2. Nesting. It must be possible for domains to be nested (see, for 697 example, the network slicing example mentioned above). 699 3. Overlapping. It must be possible for nodes and links to be in 700 more than one domain (see, for example, the case of PVDs 701 mentioned above). 703 4. Node Eligibility. It must be possible for a node to determine 704 which domain(s) it can potentially join, and on which 705 interface(s). 707 5. Secure Enrolment. A node must be able to enrol in a given 708 domain via secure node identfication and to acquire relevant 709 security credentials (authorization) for operations within the 710 domain. If a node has multiple physical or virtual interfaces, 711 they may require to be individually enrolled. 713 6. Withdrawal. A node must be able to cancel enrolment in a given 714 domain. 716 7. Dynamic Membership. Optionally, a node should be able 717 temporarily leave or rejoin a domain (i.e. enrolment is 718 persistent but membership is intermittent). 720 8. Role, implying authorization to perform a certain set of 721 actions. A node must have a verifiable role. In the simplest 722 case, the choices of role are "interior node" and "boundary 723 node". In a boundary node, individual interfaces may have 724 different roles, e.g. "inward facing" and "outward facing". 726 9. Verify Peer. A node must be able to verify whether another node 727 is a member of the domain. 729 10. Verify Role. A node must be able to learn the verified role of 730 another node. In particular, it must be possible for a node to 731 find boundary nodes (interfacing to the Internet). 733 11. Domain Data. In a domain with management requirements, it must 734 be possible for a node to acquire domain policy and/or domain 735 configuration data. This would include, for example, filtering 736 policy to ensure that inappropriate packets do not leave the 737 domain. 739 These requirements could form the basis for further analysis and 740 solution design. 742 Another aspect is whether individual packets within a limited domain 743 need to carry any sort of indicator that they belong to that domain, 744 or whether this information will be implicit in the IP addresses of 745 the packet. A related question is whether individual packets need 746 cryptographic authentication. This topic is for further study. 748 7. Security Considerations 750 Often, the boundary of a limited domain will also act as a security 751 boundary. In particular, it will serve as a trust boundary, and as a 752 boundary of authority for defining capabilities. For example, 753 segment routing [RFC8402] explicitly uses the concept of a "trusted 754 domain" in this way. Within the boundary, limited-domain protocols 755 or protocol features will be useful, but they will in many cases be 756 meaningless or harmful if they enter or leave the domain. 758 The boundary also serves to provide confidentiality and privacy of 759 operational parameters that the operator does not wish to reveal. 760 Note that this is distinct from privacy protection for individual 761 users within the domain. 763 The security model for a limited-scope protocol must allow for the 764 boundary, and in particular for a trust model that changes at the 765 boundary. Typically, credentials will need to be signed by a domain- 766 specific authority. 768 8. IANA Considerations 770 This document makes no request of the IANA. 772 9. Contributor 774 Sheng Jiang 775 Huawei Technologies 776 Q14, Huawei Campus 777 No.156 Beiqing Road 778 Hai-Dian District, Beijing 100095 779 P.R. China 781 Email: jiangsheng@huawei.com 783 10. Acknowledgements 785 Useful comments were received from Amelia Andersdotter, Edward 786 Birrane, David Black, Ron Bonica, Mohamed Boucadair, Tim Chown, 787 Darren Dukes, Donald Eastlake, Adrian Farrel, Tom Herbert, John 788 Klensin, Andy Malis, Michael Richardson, Mark Smith, Rick Taylor, 789 Niels ten Oever, and other members of the ANIMA and INTAREA WGs. 791 11. Informative References 793 [BIGIP] Li, R., "HUAWEI - Big IP Initiative.", 2018, 794 . 796 [I-D.andrews-tcp-and-ipv6-use-minmtu] 797 Andrews, M., "TCP Fails To Respect IPV6_USE_MIN_MTU", 798 draft-andrews-tcp-and-ipv6-use-minmtu-04 (work in 799 progress), October 2015. 801 [I-D.clemm-nmrg-dist-intent] 802 Clemm, A., Ciavaglia, L., Granville, L., and J. Tantsura, 803 "Intent-Based Networking - Concepts and Overview", draft- 804 clemm-nmrg-dist-intent-03 (work in progress), November 805 2019. 807 [I-D.dcrocker-dns-perimeter] 808 Crocker, D. and T. Adams, "DNS Perimeter Overlay", draft- 809 dcrocker-dns-perimeter-01 (work in progress), June 2019. 811 [I-D.fz-6man-ipv6-alt-mark] 812 Fioccola, G., Zhou, T., and M. Cociglio, "IPv6 Application 813 of the Alternate Marking Method", draft-fz-6man-ipv6-alt- 814 mark-01 (work in progress), October 2019. 816 [I-D.herbert-fast] 817 Herbert, T., "Firewall and Service Tickets", draft- 818 herbert-fast-04 (work in progress), April 2019. 820 [I-D.herbert-ipv4-eh] 821 Herbert, T., "IPv4 Extension Headers and Flow Label", 822 draft-herbert-ipv4-eh-01 (work in progress), May 2019. 824 [I-D.ietf-6man-segment-routing-header] 825 Filsfils, C., Dukes, D., Previdi, S., Leddy, J., 826 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 827 (SRH)", draft-ietf-6man-segment-routing-header-26 (work in 828 progress), October 2019. 830 [I-D.ietf-anima-autonomic-control-plane] 831 Eckert, T., Behringer, M., and S. Bjarnason, "An Autonomic 832 Control Plane (ACP)", draft-ietf-anima-autonomic-control- 833 plane-21 (work in progress), November 2019. 835 [I-D.ietf-anima-reference-model] 836 Behringer, M., Carpenter, B., Eckert, T., Ciavaglia, L., 837 and J. Nobre, "A Reference Model for Autonomic 838 Networking", draft-ietf-anima-reference-model-10 (work in 839 progress), November 2018. 841 [I-D.ietf-detnet-data-plane-framework] 842 Varga, B., Farkas, J., Berger, L., Fedyk, D., Malis, A., 843 Bryant, S., and J. Korhonen, "DetNet Data Plane 844 Framework", draft-ietf-detnet-data-plane-framework-03 845 (work in progress), October 2019. 847 [I-D.ietf-dmm-5g-uplane-analysis] 848 Homma, S., Miyasaka, T., Matsushima, S., and D. Voyer, 849 "User Plane Protocol and Architectural Analysis on 3GPP 5G 850 System", draft-ietf-dmm-5g-uplane-analysis-03 (work in 851 progress), November 2019. 853 [I-D.ietf-homenet-simple-naming] 854 Lemon, T., Migault, D., and S. Cheshire, "Homenet Naming 855 and Service Discovery Architecture", draft-ietf-homenet- 856 simple-naming-03 (work in progress), October 2018. 858 [I-D.ietf-intarea-frag-fragile] 859 Bonica, R., Baker, F., Huston, G., Hinden, R., Troan, O., 860 and F. Gont, "IP Fragmentation Considered Fragile", draft- 861 ietf-intarea-frag-fragile-17 (work in progress), September 862 2019. 864 [I-D.ietf-ipwave-vehicular-networking] 865 Jeong, J., "IP Wireless Access in Vehicular Environments 866 (IPWAVE): Problem Statement and Use Cases", draft-ietf- 867 ipwave-vehicular-networking-12 (work in progress), October 868 2019. 870 [I-D.ietf-opsec-ipv6-eh-filtering] 871 Gont, F. and W. LIU, "Recommendations on the Filtering of 872 IPv6 Packets Containing IPv6 Extension Headers", draft- 873 ietf-opsec-ipv6-eh-filtering-06 (work in progress), July 874 2018. 876 [I-D.ietf-spring-srv6-network-programming] 877 Filsfils, C., Camarillo, P., Leddy, J., Voyer, D., 878 Matsushima, S., and Z. Li, "SRv6 Network Programming", 879 draft-ietf-spring-srv6-network-programming-05 (work in 880 progress), October 2019. 882 [I-D.ietf-teas-enhanced-vpn] 883 Dong, J., Bryant, S., Li, Z., Miyasaka, T., and Y. Lee, "A 884 Framework for Enhanced Virtual Private Networks (VPN+) 885 Service", draft-ietf-teas-enhanced-vpn-03 (work in 886 progress), September 2019. 888 [I-D.ioametal-ippm-6man-ioam-ipv6-options] 889 Bhandari, S., Brockners, F., Pignataro, C., Gredler, H., 890 Leddy, J., Youell, S., Mizrahi, T., Kfir, A., Gafni, B., 891 Lapukhov, P., Spiegel, M., Krishnan, S., and R. Asati, 892 "In-situ OAM IPv6 Options", draft-ioametal-ippm-6man-ioam- 893 ipv6-options-02 (work in progress), March 2019. 895 [I-D.irtf-nfvrg-gaps-network-virtualization] 896 Bernardos, C., Rahman, A., Zuniga, J., Contreras, L., 897 Aranda, P., and P. Lynch, "Network Virtualization Research 898 Challenges", draft-irtf-nfvrg-gaps-network- 899 virtualization-10 (work in progress), September 2018. 901 [I-D.jiang-semantic-prefix] 902 Jiang, S., Qiong, Q., Farrer, I., Bo, Y., and T. Yang, 903 "Analysis of Semantic Embedded IPv6 Address Schemas", 904 draft-jiang-semantic-prefix-06 (work in progress), July 905 2013. 907 [I-D.li-6man-service-aware-ipv6-network] 908 Li, Z. and S. Peng, "Service-aware IPv6 Network", draft- 909 li-6man-service-aware-ipv6-network-00 (work in progress), 910 March 2019. 912 [I-D.smith-6man-in-flight-eh-insertion-harmful] 913 Smith, M., Kottapalli, N., Bonica, R., Gont, F., and T. 914 Herbert, "In-Flight IPv6 Extension Header Insertion 915 Considered Harmful", draft-smith-6man-in-flight-eh- 916 insertion-harmful-01 (work in progress), November 2019. 918 [I-D.voyer-6man-extension-header-insertion] 919 Voyer, D., Filsfils, C., Dukes, D., Matsushima, S., Leddy, 920 J., Li, Z., and J. Guichard, "Deployments With Insertion 921 of IPv6 Segment Routing Headers", draft-voyer-6man- 922 extension-header-insertion-08 (work in progress), November 923 2019. 925 [RFC2205] Braden, R., Ed., Zhang, L., Berson, S., Herzog, S., and S. 926 Jamin, "Resource ReSerVation Protocol (RSVP) -- Version 1 927 Functional Specification", RFC 2205, DOI 10.17487/RFC2205, 928 September 1997, . 930 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 931 "Definition of the Differentiated Services Field (DS 932 Field) in the IPv4 and IPv6 Headers", RFC 2474, 933 DOI 10.17487/RFC2474, December 1998, 934 . 936 [RFC2775] Carpenter, B., "Internet Transparency", RFC 2775, 937 DOI 10.17487/RFC2775, February 2000, 938 . 940 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", 941 RFC 2923, DOI 10.17487/RFC2923, September 2000, 942 . 944 [RFC3234] Carpenter, B. and S. Brim, "Middleboxes: Taxonomy and 945 Issues", RFC 3234, DOI 10.17487/RFC3234, February 2002, 946 . 948 [RFC3879] Huitema, C. and B. Carpenter, "Deprecating Site Local 949 Addresses", RFC 3879, DOI 10.17487/RFC3879, September 950 2004, . 952 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 953 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 954 . 956 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 957 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 958 2006, . 960 [RFC4397] Bryskin, I. and A. Farrel, "A Lexicography for the 961 Interpretation of Generalized Multiprotocol Label 962 Switching (GMPLS) Terminology within the Context of the 963 ITU-T's Automatically Switched Optical Network (ASON) 964 Architecture", RFC 4397, DOI 10.17487/RFC4397, February 965 2006, . 967 [RFC4427] Mannie, E., Ed. and D. Papadimitriou, Ed., "Recovery 968 (Protection and Restoration) Terminology for Generalized 969 Multi-Protocol Label Switching (GMPLS)", RFC 4427, 970 DOI 10.17487/RFC4427, March 2006, 971 . 973 [RFC4655] Farrel, A., Vasseur, J., and J. Ash, "A Path Computation 974 Element (PCE)-Based Architecture", RFC 4655, 975 DOI 10.17487/RFC4655, August 2006, 976 . 978 [RFC4821] Mathis, M. and J. Heffner, "Packetization Layer Path MTU 979 Discovery", RFC 4821, DOI 10.17487/RFC4821, March 2007, 980 . 982 [RFC4838] Cerf, V., Burleigh, S., Hooke, A., Torgerson, L., Durst, 983 R., Scott, K., Fall, K., and H. Weiss, "Delay-Tolerant 984 Networking Architecture", RFC 4838, DOI 10.17487/RFC4838, 985 April 2007, . 987 [RFC4924] Aboba, B., Ed. and E. Davies, "Reflections on Internet 988 Transparency", RFC 4924, DOI 10.17487/RFC4924, July 2007, 989 . 991 [RFC5704] Bryant, S., Ed., Morrow, M., Ed., and IAB, "Uncoordinated 992 Protocol Development Considered Harmful", RFC 5704, 993 DOI 10.17487/RFC5704, November 2009, 994 . 996 [RFC6294] Hu, Q. and B. Carpenter, "Survey of Proposed Use Cases for 997 the IPv6 Flow Label", RFC 6294, DOI 10.17487/RFC6294, June 998 2011, . 1000 [RFC6325] Perlman, R., Eastlake 3rd, D., Dutt, D., Gai, S., and A. 1001 Ghanwani, "Routing Bridges (RBridges): Base Protocol 1002 Specification", RFC 6325, DOI 10.17487/RFC6325, July 2011, 1003 . 1005 [RFC6398] Le Faucheur, F., Ed., "IP Router Alert Considerations and 1006 Usage", BCP 168, RFC 6398, DOI 10.17487/RFC6398, October 1007 2011, . 1009 [RFC6407] Weis, B., Rowles, S., and T. Hardjono, "The Group Domain 1010 of Interpretation", RFC 6407, DOI 10.17487/RFC6407, 1011 October 2011, . 1013 [RFC6455] Fette, I. and A. Melnikov, "The WebSocket Protocol", 1014 RFC 6455, DOI 10.17487/RFC6455, December 2011, 1015 . 1017 [RFC6709] Carpenter, B., Aboba, B., Ed., and S. Cheshire, "Design 1018 Considerations for Protocol Extensions", RFC 6709, 1019 DOI 10.17487/RFC6709, September 2012, 1020 . 1022 [RFC6947] Boucadair, M., Kaplan, H., Gilman, R., and S. 1023 Veikkolainen, "The Session Description Protocol (SDP) 1024 Alternate Connectivity (ALTC) Attribute", RFC 6947, 1025 DOI 10.17487/RFC6947, May 2013, 1026 . 1028 [RFC6950] Peterson, J., Kolkman, O., Tschofenig, H., and B. Aboba, 1029 "Architectural Considerations on Application Features in 1030 the DNS", RFC 6950, DOI 10.17487/RFC6950, October 2013, 1031 . 1033 [RFC7045] Carpenter, B. and S. Jiang, "Transmission and Processing 1034 of IPv6 Extension Headers", RFC 7045, 1035 DOI 10.17487/RFC7045, December 2013, 1036 . 1038 [RFC7228] Bormann, C., Ersue, M., and A. Keranen, "Terminology for 1039 Constrained-Node Networks", RFC 7228, 1040 DOI 10.17487/RFC7228, May 2014, 1041 . 1043 [RFC7368] Chown, T., Ed., Arkko, J., Brandt, A., Troan, O., and J. 1044 Weil, "IPv6 Home Networking Architecture Principles", 1045 RFC 7368, DOI 10.17487/RFC7368, October 2014, 1046 . 1048 [RFC7381] Chittimaneni, K., Chown, T., Howard, L., Kuarsingh, V., 1049 Pouffary, Y., and E. Vyncke, "Enterprise IPv6 Deployment 1050 Guidelines", RFC 7381, DOI 10.17487/RFC7381, October 2014, 1051 . 1053 [RFC7556] Anipko, D., Ed., "Multiple Provisioning Domain 1054 Architecture", RFC 7556, DOI 10.17487/RFC7556, June 2015, 1055 . 1057 [RFC7663] Trammell, B., Ed. and M. Kuehlewind, Ed., "Report from the 1058 IAB Workshop on Stack Evolution in a Middlebox Internet 1059 (SEMI)", RFC 7663, DOI 10.17487/RFC7663, October 2015, 1060 . 1062 [RFC7665] Halpern, J., Ed. and C. Pignataro, Ed., "Service Function 1063 Chaining (SFC) Architecture", RFC 7665, 1064 DOI 10.17487/RFC7665, October 2015, 1065 . 1067 [RFC7788] Stenberg, M., Barth, S., and P. Pfister, "Home Networking 1068 Control Protocol", RFC 7788, DOI 10.17487/RFC7788, April 1069 2016, . 1071 [RFC7872] Gont, F., Linkova, J., Chown, T., and W. Liu, 1072 "Observations on the Dropping of Packets with IPv6 1073 Extension Headers in the Real World", RFC 7872, 1074 DOI 10.17487/RFC7872, June 2016, 1075 . 1077 [RFC8085] Eggert, L., Fairhurst, G., and G. Shepherd, "UDP Usage 1078 Guidelines", BCP 145, RFC 8085, DOI 10.17487/RFC8085, 1079 March 2017, . 1081 [RFC8086] Yong, L., Ed., Crabbe, E., Xu, X., and T. Herbert, "GRE- 1082 in-UDP Encapsulation", RFC 8086, DOI 10.17487/RFC8086, 1083 March 2017, . 1085 [RFC8100] Geib, R., Ed. and D. Black, "Diffserv-Interconnection 1086 Classes and Practice", RFC 8100, DOI 10.17487/RFC8100, 1087 March 2017, . 1089 [RFC8151] Yong, L., Dunbar, L., Toy, M., Isaac, A., and V. Manral, 1090 "Use Cases for Data Center Network Virtualization Overlay 1091 Networks", RFC 8151, DOI 10.17487/RFC8151, May 2017, 1092 . 1094 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 1095 (IPv6) Specification", STD 86, RFC 8200, 1096 DOI 10.17487/RFC8200, July 2017, 1097 . 1099 [RFC8300] Quinn, P., Ed., Elzur, U., Ed., and C. Pignataro, Ed., 1100 "Network Service Header (NSH)", RFC 8300, 1101 DOI 10.17487/RFC8300, January 2018, 1102 . 1104 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 1105 Decraene, B., Litkowski, S., and R. Shakir, "Segment 1106 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 1107 July 2018, . 1109 [RFC8445] Keranen, A., Holmberg, C., and J. Rosenberg, "Interactive 1110 Connectivity Establishment (ICE): A Protocol for Network 1111 Address Translator (NAT) Traversal", RFC 8445, 1112 DOI 10.17487/RFC8445, July 2018, 1113 . 1115 [RFC8517] Dolson, D., Ed., Snellman, J., Boucadair, M., Ed., and C. 1116 Jacquenet, "An Inventory of Transport-Centric Functions 1117 Provided by Middleboxes: An Operator Perspective", 1118 RFC 8517, DOI 10.17487/RFC8517, February 2019, 1119 . 1121 [RFC8557] Finn, N. and P. Thubert, "Deterministic Networking Problem 1122 Statement", RFC 8557, DOI 10.17487/RFC8557, May 2019, 1123 . 1125 [RFC8578] Grossman, E., Ed., "Deterministic Networking Use Cases", 1126 RFC 8578, DOI 10.17487/RFC8578, May 2019, 1127 . 1129 [RFC8655] Finn, N., Thubert, P., Varga, B., and J. Farkas, 1130 "Deterministic Networking Architecture", RFC 8655, 1131 DOI 10.17487/RFC8655, October 2019, 1132 . 1134 [SPB] "IEEE Standard for Local and metropolitan area networks - 1135 Bridges and Bridged Networks", IEEE Standard 802.1Q-2018, 1136 2018, . 1139 Appendix A. Change log [RFC Editor: Please remove] 1141 draft-carpenter-limited-domains-00, 2018-06-11: 1143 Initial version 1145 draft-carpenter-limited-domains-01, 2018-07-01: 1147 Minor terminology clarifications 1149 draft-carpenter-limited-domains-02, 2018-08-03: 1151 Additions following IETF102 discussions 1153 Updated authorship/contributors 1155 draft-carpenter-limited-domains-03, 2018-09-12: 1157 First draft of taxonomy 1159 Editorial improvements 1161 draft-carpenter-limited-domains-04, 2018-10-14: 1163 Reorganized section 3 1165 Newly written sections 6 and 7 1167 Editorial improvements 1169 draft-carpenter-limited-domains-05, 2018-12-12: 1171 Added discussion of transparency/filtering debates 1173 Added discussion of "controlled environment" 1175 Modified assertion about localized standards 1177 Editorial improvements 1179 draft-carpenter-limited-domains-06, 2019-03-02: 1181 Minor updates, fixed reference nits 1183 draft-carpenter-limited-domains-07, 2019-04-15: 1185 Moved taxonomy to an appendix. 1187 Added examples and references. 1189 Editorial improvements 1191 draft-carpenter-limited-domains-08, 2019-06-12: 1193 Added short discussion of address scopes. 1195 Added possibility of nested or overlapped domains. 1197 Integrated other comments received. 1199 Editorial improvements 1201 draft-carpenter-limited-domains-09, 2019-06-21: 1203 Additional 5G citations. 1205 draft-carpenter-limited-domains-10, 2019-08-02: 1207 ISE comments. 1209 draft-carpenter-limited-domains-11, 2019-10-31: 1211 Incorporate review comments. 1213 Editorial improvements. 1215 draft-carpenter-limited-domains-12, 2019-11-30: 1217 Incorporate ISE comments. 1219 Appendix B. Taxonomy of Limited Domains 1221 This appendix develops a taxonomy for describing limited domains. 1222 Several major aspects are considered in this taxonomy: 1224 o The domain as a whole. 1226 o The individual nodes. 1228 o The domain boundary. 1230 o The domain's topology. 1232 o The domain's technology. 1234 o How the domain connects to the Internet. 1236 o The security, trust, and privacy model. 1238 o Operations. 1240 The following sub-sections analyse each of these aspects. 1242 B.1. The Domain as a Whole 1244 o Why does the domain exist? (e.g., human choice, administrative 1245 policy, orchestration requirements, technical requirements such as 1246 operational partitioning for scaling reasons) 1248 o If there are special requirements, are they at Layer 2, Layer 3 or 1249 an upper layer? 1251 o Where does the domain lie on the spectrum between completely 1252 managed by humans and completely autonomic? 1254 o If managed, what style of management applies? (Manual 1255 configuration, automated configuration, orchestration?) 1257 o Is there a policy model? (Intent, configuration policies?) 1259 o Does the domain provide controlled or paid service or open access? 1261 B.2. Individual Nodes 1263 o Is a domain member a complete node, or only one interface of a 1264 node? 1266 o Are nodes permanent members of a given domain, or are join and 1267 leave operations possible? 1269 o Are nodes physical or virtual devices? 1271 o Are virtual nodes general-purpose, or limited to specific 1272 functions, applications or users? 1274 o Are nodes constrained (by battery etc)? 1276 o Are devices installed "out of the box" or pre-configured? 1278 B.3. The Domain Boundary 1280 o How is the domain boundary identified or defined? 1282 o Is the domain boundary fixed or dynamic? 1284 o Are boundary nodes special? Or can any node be at the boundary? 1286 B.4. Topology 1288 o Is the domain a subset of a layer 2 or 3 connectivity domain? 1290 o Does the domain overlap other domains? (In other words, a node 1291 may or may not be allowed to be a member of multiple domains.) 1293 o Does the domain match physical topology, or does it have a virtual 1294 (overlay) topology? 1296 o Is the domain in a single building, vehicle or campus? Or is it 1297 distributed? 1299 o If distributed, are the interconnections private or over the 1300 Internet? 1302 o In IP addressing terms, is the domain Link-local, Site-local, or 1303 Global? 1305 o Does the scope of IP unicast or multicast addresses map to the 1306 domain boundary? 1308 B.5. Technology 1310 o What routing protocol(s) are used, or even different forwarding 1311 mechanisms (MPLS or other non-IP mechanism)? 1313 o In an overlay domain, what overlay technique is used (L2VPN, 1314 L3VPN,...)? 1316 o Are there specific QoS requirements? 1318 o Link latency - normal or long latency links? 1320 o Mobility - are nodes mobile? Is the whole network mobile? 1322 o Which specific technologies, such as those in Section 4, are 1323 applicable? 1325 B.6. Connection to the Internet 1327 o Is the Internet connection permanent or intermittent? (Never 1328 connected is out of scope.) 1330 o What traffic is blocked, in and out? 1332 o What traffic is allowed, in and out? 1333 o What traffic is transformed, in and out? 1335 o Is secure and privileged remote access needed? 1337 o Does the domain allow unprivileged remote sessions? 1339 B.7. Security, Trust and Privacy Model 1341 o Must domain members be authorized? 1343 o Are all nodes in the domain at the same trust level? 1345 o Is traffic authenticated? 1347 o Is traffic encrypted? 1349 o What is hidden from the outside? 1351 B.8. Operations 1353 o Safety level - does the domain have a critical (human) safety 1354 role? 1356 o Reliability requirement - normal or 99.999% ? 1358 o Environment - hazardous conditions? 1360 o Installation - are specialists needed? 1362 o Service visits - easy, difficult, impossible? 1364 o Software/firmware updates - possible or impossible? 1366 B.9. Making use of this taxonomy 1368 This taxonomy could be used to design or analyse a specific type of 1369 limited domain. For the present document, it is intended only to 1370 form a background to the scope of protocols used in limited domains, 1371 and the mechanisms required to securely define domain membership and 1372 properties. 1374 Authors' Addresses 1375 Brian Carpenter 1376 The University of Auckland 1377 School of Computer Science 1378 University of Auckland 1379 PB 92019 1380 Auckland 1142 1381 New Zealand 1383 Email: brian.e.carpenter@gmail.com 1385 Bing Liu 1386 Huawei Technologies 1387 Q14, Huawei Campus 1388 No.156 Beiqing Road 1389 Hai-Dian District, Beijing 100095 1390 P.R. China 1392 Email: leo.liubing@huawei.com