idnits 2.17.1 draft-ietf-homenet-arch-15.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 (June 9, 2014) is 3610 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 2460 (Obsoleted by RFC 8200) ** Obsolete normative reference: RFC 3633 (Obsoleted by RFC 8415) -- Obsolete informational reference (is this intentional?): RFC 4941 (Obsoleted by RFC 8981) -- Obsolete informational reference (is this intentional?): RFC 6145 (Obsoleted by RFC 7915) -- Obsolete informational reference (is this intentional?): RFC 6204 (Obsoleted by RFC 7084) -- Obsolete informational reference (is this intentional?): RFC 6555 (Obsoleted by RFC 8305) -- Obsolete informational reference (is this intentional?): RFC 6824 (Obsoleted by RFC 8684) Summary: 2 errors (**), 0 flaws (~~), 1 warning (==), 6 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group T. Chown, Ed. 3 Internet-Draft University of Southampton 4 Intended status: Informational J. Arkko 5 Expires: December 11, 2014 Ericsson 6 A. Brandt 7 Sigma Designs 8 O. Troan 9 Cisco Systems, Inc. 10 J. Weil 11 Time Warner Cable 12 June 9, 2014 14 IPv6 Home Networking Architecture Principles 15 draft-ietf-homenet-arch-15 17 Abstract 19 This text describes evolving networking technology within residential 20 home networks with increasing numbers of devices and a trend towards 21 increased internal routing. The goal of this document is to define a 22 general architecture for IPv6-based home networking, describing the 23 associated principles, considerations and requirements. The text 24 briefly highlights specific implications of the introduction of IPv6 25 for home networking, discusses the elements of the architecture, and 26 suggests how standard IPv6 mechanisms and addressing can be employed 27 in home networking. The architecture describes the need for specific 28 protocol extensions for certain additional functionality. It is 29 assumed that the IPv6 home network is not actively managed, and runs 30 as an IPv6-only or dual-stack network. There are no recommendations 31 in this text for the IPv4 part of the network. 33 Status of this Memo 35 This Internet-Draft is submitted in full conformance with the 36 provisions of BCP 78 and BCP 79. 38 Internet-Drafts are working documents of the Internet Engineering 39 Task Force (IETF). Note that other groups may also distribute 40 working documents as Internet-Drafts. The list of current Internet- 41 Drafts is at http://datatracker.ietf.org/drafts/current/. 43 Internet-Drafts are draft documents valid for a maximum of six months 44 and may be updated, replaced, or obsoleted by other documents at any 45 time. It is inappropriate to use Internet-Drafts as reference 46 material or to cite them other than as "work in progress." 48 This Internet-Draft will expire on December 11, 2014. 50 Copyright Notice 52 Copyright (c) 2014 IETF Trust and the persons identified as the 53 document authors. All rights reserved. 55 This document is subject to BCP 78 and the IETF Trust's Legal 56 Provisions Relating to IETF Documents 57 (http://trustee.ietf.org/license-info) in effect on the date of 58 publication of this document. Please review these documents 59 carefully, as they describe your rights and restrictions with respect 60 to this document. Code Components extracted from this document must 61 include Simplified BSD License text as described in Section 4.e of 62 the Trust Legal Provisions and are provided without warranty as 63 described in the Simplified BSD License. 65 Table of Contents 67 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 68 1.1. Terminology and Abbreviations . . . . . . . . . . . . . . 5 69 2. Effects of IPv6 on Home Networking . . . . . . . . . . . . . . 6 70 2.1. Multiple subnets and routers . . . . . . . . . . . . . . . 7 71 2.2. Global addressability and elimination of NAT . . . . . . . 8 72 2.3. Multi-Addressing of devices . . . . . . . . . . . . . . . 8 73 2.4. Unique Local Addresses (ULAs) . . . . . . . . . . . . . . 9 74 2.5. Avoiding manual configuration of IP addresses . . . . . . 10 75 2.6. IPv6-only operation . . . . . . . . . . . . . . . . . . . 11 76 3. Homenet Architecture Principles . . . . . . . . . . . . . . . 11 77 3.1. General Principles . . . . . . . . . . . . . . . . . . . . 12 78 3.1.1. Reuse existing protocols . . . . . . . . . . . . . . . 12 79 3.1.2. Minimise changes to hosts and routers . . . . . . . . 12 80 3.2. Homenet Topology . . . . . . . . . . . . . . . . . . . . . 13 81 3.2.1. Supporting arbitrary topologies . . . . . . . . . . . 13 82 3.2.2. Network topology models . . . . . . . . . . . . . . . 13 83 3.2.3. Dual-stack topologies . . . . . . . . . . . . . . . . 18 84 3.2.4. Multihoming . . . . . . . . . . . . . . . . . . . . . 19 85 3.2.5. Mobility support . . . . . . . . . . . . . . . . . . . 20 86 3.3. A Self-Organising Network . . . . . . . . . . . . . . . . 20 87 3.3.1. Differentiating neighbouring homenets . . . . . . . . 21 88 3.3.2. Largest practical subnets . . . . . . . . . . . . . . 21 89 3.3.3. Handling varying link technologies . . . . . . . . . . 22 90 3.3.4. Homenet realms and borders . . . . . . . . . . . . . . 22 91 3.3.5. Configuration information from the ISP . . . . . . . . 23 92 3.4. Homenet Addressing . . . . . . . . . . . . . . . . . . . . 23 93 3.4.1. Use of ISP-delegated IPv6 prefixes . . . . . . . . . . 24 94 3.4.2. Stable internal IP addresses . . . . . . . . . . . . . 26 95 3.4.3. Internal prefix delegation . . . . . . . . . . . . . . 27 96 3.4.4. Coordination of configuration information . . . . . . 28 97 3.4.5. Privacy . . . . . . . . . . . . . . . . . . . . . . . 28 98 3.5. Routing functionality . . . . . . . . . . . . . . . . . . 28 99 3.5.1. Multicast support . . . . . . . . . . . . . . . . . . 30 100 3.6. Security . . . . . . . . . . . . . . . . . . . . . . . . . 31 101 3.6.1. Addressability vs reachability . . . . . . . . . . . . 31 102 3.6.2. Filtering at borders . . . . . . . . . . . . . . . . . 32 103 3.6.3. Partial Effectiveness of NAT and Firewalls . . . . . . 32 104 3.6.4. Exfiltration concerns . . . . . . . . . . . . . . . . 33 105 3.6.5. Device capabilities . . . . . . . . . . . . . . . . . 33 106 3.6.6. ULAs as a hint of connection origin . . . . . . . . . 33 107 3.7. Naming and Service Discovery . . . . . . . . . . . . . . . 33 108 3.7.1. Discovering services . . . . . . . . . . . . . . . . . 34 109 3.7.2. Assigning names to devices . . . . . . . . . . . . . . 35 110 3.7.3. The homenet name service . . . . . . . . . . . . . . . 35 111 3.7.4. Name spaces . . . . . . . . . . . . . . . . . . . . . 36 112 3.7.5. Independent operation . . . . . . . . . . . . . . . . 38 113 3.7.6. Considerations for LLNs . . . . . . . . . . . . . . . 39 114 3.7.7. DNS resolver discovery . . . . . . . . . . . . . . . . 39 115 3.7.8. Devices roaming to/from the homenet . . . . . . . . . 39 116 3.8. Other Considerations . . . . . . . . . . . . . . . . . . . 40 117 3.8.1. Quality of Service . . . . . . . . . . . . . . . . . . 40 118 3.8.2. Operations and Management . . . . . . . . . . . . . . 40 119 3.9. Implementing the Architecture on IPv6 . . . . . . . . . . 41 120 4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 42 121 5. Security Considerations . . . . . . . . . . . . . . . . . . . 42 122 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 42 123 7. References . . . . . . . . . . . . . . . . . . . . . . . . . . 42 124 7.1. Normative References . . . . . . . . . . . . . . . . . . . 42 125 7.2. Informative References . . . . . . . . . . . . . . . . . . 43 126 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . . 45 127 Appendix B. Changes . . . . . . . . . . . . . . . . . . . . . . . 46 128 B.1. Version 15 . . . . . . . . . . . . . . . . . . . . . . . . 46 129 B.2. Version 14 . . . . . . . . . . . . . . . . . . . . . . . . 46 130 B.3. Version 13 . . . . . . . . . . . . . . . . . . . . . . . . 46 131 B.4. Version 12 . . . . . . . . . . . . . . . . . . . . . . . . 47 132 B.5. Version 11 (after IESG review) . . . . . . . . . . . . . . 47 133 B.6. Version 10 (after AD review) . . . . . . . . . . . . . . . 47 134 B.7. Version 09 (after WGLC) . . . . . . . . . . . . . . . . . 47 135 B.8. Version 08 . . . . . . . . . . . . . . . . . . . . . . . . 48 136 B.9. Version 07 . . . . . . . . . . . . . . . . . . . . . . . . 48 137 B.10. Version 06 . . . . . . . . . . . . . . . . . . . . . . . . 49 138 B.11. Version 05 . . . . . . . . . . . . . . . . . . . . . . . . 49 139 B.12. Version 04 . . . . . . . . . . . . . . . . . . . . . . . . 50 140 B.13. Version 03 . . . . . . . . . . . . . . . . . . . . . . . . 50 141 B.14. Version 02 . . . . . . . . . . . . . . . . . . . . . . . . 51 142 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 52 144 1. Introduction 146 This document focuses on evolving networking technology within 147 residential home networks with increasing numbers of devices and a 148 trend towards increased internal routing, and the associated 149 challenges with their deployment and operation. There is a growing 150 trend in home networking for the proliferation of networking 151 technology through an increasingly broad range of devices and media. 152 This evolution in scale and diversity sets requirements on IETF 153 protocols. Some of these requirements relate to the introduction of 154 IPv6, others to the introduction of specialised networks for home 155 automation and sensors. 157 While at the time of writing some complex home network topologies 158 exist, most are relatively simple single subnet networks, and 159 ostensibly operate using just IPv4. While there may be IPv6 traffic 160 within the network, e.g., for service discovery, the homenet is 161 provisioned by the ISP as an IPv4 network. Such networks also 162 typically employ solutions that should be avoided, such as private 163 [RFC1918] addressing with (cascaded) network address translation 164 (NAT) [RFC3022], or they may require expert assistance to set up. 166 In contrast, emerging IPv6-capable home networks are very likely to 167 have multiple internal subnets, e.g., to facilitate private and guest 168 networks, heterogeneous link layers, and smart grid components, and 169 have enough address space available to allow every device to have a 170 globally unique address. This implies that internal routing 171 functionality is required, and that the homenet's ISP both provides a 172 large enough prefix to allocate a prefix to each subnet, and that a 173 method is supported for such prefixes to be delegated efficiently to 174 those subnets. 176 It is not practical to expect home users to configure their networks. 177 Thus the assumption of this document is that the homenet is as far as 178 possible self-organising and self-configuring, i.e., it should 179 function without pro-active management by the residential user. 181 The architectural constructs in this document are focused on the 182 problems to be solved when introducing IPv6, with an eye towards a 183 better result than what we have today with IPv4, as well as aiming at 184 a more consistent solution that addresses as many of the identified 185 requirements as possible. The document aims to provide the basis and 186 guiding principles for how standard IPv6 mechanisms and addressing 187 [RFC2460] [RFC4291] can be employed in home networking, while 188 coexisting with existing IPv4 mechanisms. In emerging dual-stack 189 home networks it is vital that introducing IPv6 does not adversely 190 affect IPv4 operation. We assume that the IPv4 network architecture 191 in home networks is what it is, and can not be modified by new 192 recommendations. This document does not discuss how IPv4 home 193 networks provision or deliver support for multiple subnets. It 194 should not be assumed that any future new functionality created with 195 IPv6 in mind will be backward-compatible to include IPv4 support. 196 Further, future deployments, or specific subnets within an otherwise 197 dual-stack home network, may be IPv6-only, in which case 198 considerations for IPv4 impact would not apply. 200 This document proposes a baseline homenet architecture, using 201 protocols and implementations that are as far as possible proven and 202 robust. The scope of the document is primarily the network layer 203 technologies that provide the basic functionality to enable 204 addressing, connectivity, routing, naming and service discovery. 205 While it may, for example, state that homenet components must be 206 simple to deploy and use, it does not discuss specific user 207 interfaces, nor does it discuss specific physical, wireless or data- 208 link layer considerations. Likewise, we also do not specify the 209 whole design of a homenet router from top to bottom, rather we focus 210 on the Layer 3 aspects. This means that Layer 2 is largely out of 211 scope, we're assuming a data link layer that supports IPv6 is 212 present, and that we react accordingly. Any IPv6-over-Foo 213 definitions occur elsewhere. 215 [RFC6204] defines basic requirements for customer edge routers 216 (CERs). This document has recently been updated with the definition 217 of requirements for specific transition tools on the CER in 218 [RFC7084], specifically DS-Lite [RFC6333] and 6rd [RFC5969]. Such 219 detailed specification of CER devices is considered out of scope of 220 this architecture document, and we assume that any required update of 221 the CER device specification as a result of adopting this 222 architecture will be handled as separate and specific updates to 223 these existing documents. Further, the scope of this text is the 224 internal homenet, and thus specific features on the WAN side of the 225 CER are out of scope for this text. 227 1.1. Terminology and Abbreviations 229 In this section we define terminology and abbreviations used 230 throughout the text. 232 o Border: a point, typically resident on a router, between two 233 networks, e.g., between the main internal homenet and a guest 234 network. This defines point(s) at which filtering and forwarding 235 policies for different types of traffic may be applied. 237 o CER: Customer Edge Router: A border router intended for use in a 238 homenet, which connects the homenet to a service provider network. 240 o FQDN: Fully Qualified Domain Name. A globally unique name. 242 o Guest network: A part of the home network intended for use by 243 visitors or guests to the home(net). Devices on the guest network 244 may typically not see or be able to use all services in the 245 home(net). 247 o Homenet: A home network, comprising host and router equipment, 248 with one or more CERs providing connectivity to service provider 249 network(s). 251 o Internet Service Provider (ISP): an entity that provides access to 252 the Internet. In this document, a service provider specifically 253 offers Internet access using IPv6, and may also offer IPv4 254 Internet access. The service provider can provide such access 255 over a variety of different transport methods such as DSL, cable, 256 wireless, and others. 258 o LLN: Low-power and lossy network. 260 o LQDN: Locally Qualified Domain Name. A name local to the homenet. 262 o NAT: Network Address Translation. Typically referring to IPv4 263 Network Address and Port Translation (NAPT) [RFC3022]. 265 o NPTv6: Network Prefix Translation for IPv6 [RFC6296]. 267 o PCP: Port Control Protocol [RFC6887]. 269 o Realm: a network delimited by a defined border. A guest network 270 within a homenet may form one realm. 272 o 'Simple Security'. Defined in [RFC4864] and expanded further in 273 [RFC6092]; describes recommended perimeter security capabilities 274 for IPv6 networks. 276 o ULA: IPv6 Unique Local Address [RFC4193]. 278 o VM: Virtual machine. 280 2. Effects of IPv6 on Home Networking 282 While IPv6 resembles IPv4 in many ways, there are some notable 283 differences in the way it may typically be deployed. It changes 284 address allocation principles, making multi-addressing the norm, and, 285 through the vastly increased address space, allows globally unique IP 286 addresses to be used for all devices in a home network. This section 287 presents an overview of some of the key implications of the 288 introduction of IPv6 for home networking, that are simultaneously 289 both promising and problematic. 291 2.1. Multiple subnets and routers 293 While simple layer 3 topologies involving as few subnets as possible 294 are preferred in home networks, the incorporation of dedicated 295 (routed) subnets remains necessary for a variety of reasons. For 296 instance, an increasingly common feature in modern home routers is 297 the ability to support both guest and private network subnets. 298 Likewise, there may be a need to separate home automation or 299 corporate extension LANs (whereby a home worker can have their 300 corporate network extended into the home using a virtual private 301 network, commonly presented as one port on an Ethernet device) from 302 the main Internet access network, or different subnets may in general 303 be associated with parts of the homenet that have different routing 304 and security policies. Further, link layer networking technology is 305 poised to become more heterogeneous, as networks begin to employ both 306 traditional Ethernet technology and link layers designed for low- 307 power and lossy networks (LLNs), such as those used for certain types 308 of sensor devices. Constraining the flow of certain traffic from 309 Ethernet links to much lower capacity links thus becomes an important 310 topic. 312 The introduction of IPv6 for home networking makes it possible for 313 every home network to be delegated enough address space from its ISP 314 to provision globally unique prefixes for each such subnet in the 315 home. While the number of addresses in a standard /64 IPv6 prefix is 316 practically unlimited, the number of prefixes available for 317 assignment to the home network is not. As a result the growth 318 inhibitor for the home network shifts from the number of addresses to 319 the number of prefixes offered by the provider; this topic is 320 discussed in [RFC6177] (BCP 157), which recommends that "end sites 321 always be able to obtain a reasonable amount of address space for 322 their actual and planned usage". 324 The addition of routing between subnets raises a number of issues. 325 One is a method by which prefixes can be efficiently allocated to 326 each subnet, without user intervention. Another is the issue of how 327 to extend mechanisms such as zero configuration service discovery 328 which currently only operate within a single subnet using link-local 329 traffic. In a typical IPv4 home network, there is only one subnet, 330 so such mechanisms would normally operate as expected. For multi- 331 subnet IPv6 home networks there are two broad choices to enable such 332 protocols to work across the scope of the entire homenet; extend 333 existing protocols to work across that scope, or introduce proxies 334 for existing link layer protocols. This topic is discussed in 335 Section 3.7. 337 2.2. Global addressability and elimination of NAT 339 The possibility for direct end-to-end communication on the Internet 340 to be restored by the introduction of IPv6 is on the one hand an 341 incredible opportunity for innovation and simpler network operation, 342 but on the other hand it is also a concern as it potentially exposes 343 nodes in the internal networks to receipt of unwanted and possibly 344 malicious traffic from the Internet. 346 With devices and applications able to talk directly to each other 347 when they have globally unique addresses, there may be an expectation 348 of improved host security to compensate for this. It should be noted 349 that many devices may (for example) ship with default settings that 350 make them readily vulnerable to compromise by external attackers if 351 globally accessible, or may simply not have robustness designed-in 352 because it was either assumed such devices would only be used on 353 private networks or the device itself doesn't have the computing 354 power to apply the necessary security methods. In addition, the 355 upgrade cycle for devices (or their firmware) may be slow, and/or 356 lack auto-update mechanisms. 358 It is thus important to distinguish between addressability and 359 reachability. While IPv6 offers global addressability through use of 360 globally unique addresses in the home, whether devices are globally 361 reachable or not would depend on any firewall or filtering 362 configuration, and not, as is commonly the case with IPv4, the 363 presence or use of NAT. In this respect, IPv6 networks may or may 364 not have filters applied at their borders to control such traffic, 365 i.e., at the homenet CER. [RFC4864] and [RFC6092] discuss such 366 filtering, and the merits of 'default allow' against 'default deny' 367 policies for external traffic initiated into a homenet. This topic 368 is discussed further in Section 3.6.1. 370 2.3. Multi-Addressing of devices 372 In an IPv6 network, devices will often acquire multiple addresses, 373 typically at least a link-local address and one or more globally 374 unique addresses. Where a homenet is multihomed, a device would 375 typically receive a globally unique address (GUA) from within the 376 delegated prefix from each upstream ISP. Devices may also have an 377 IPv4 address if the network is dual-stack, an IPv6 Unique Local 378 Address (ULA) [RFC4193] (see below), and one or more IPv6 Privacy 379 Addresses [RFC4941]. 381 It should thus be considered the norm for devices on IPv6 home 382 networks to be multi-addressed, and to need to make appropriate 383 address selection decisions for the candidate source and destination 384 address pairs for any given connection. In multihoming scenarios 385 nodes will be configured with one address from each upstream ISP 386 prefix. In such cases the presence of upstream BCP 38 [RFC2827] 387 ingress filtering requires such multi-addressed nodes to select the 388 correct source address to be used for the corresponding uplink. 389 Default Address Selection for IPv6 [RFC6724] provides a solution for 390 this, but a challenge here is that the node may not have the 391 information it needs to make that decision based on addresses alone. 392 We discuss this challenge in Section 3.2.4. 394 2.4. Unique Local Addresses (ULAs) 396 [RFC4193] defines Unique Local Addresses (ULAs) for IPv6 that may be 397 used to address devices within the scope of a single site. Support 398 for ULAs for IPv6 CERs is described in [RFC6204]. A home network 399 running IPv6 should deploy ULAs alongside its globally unique 400 prefix(es) to allow stable communication between devices (on 401 different subnets) within the homenet where that externally allocated 402 globally unique prefix may change over time, e.g., due to renumbering 403 within the subscriber's ISP, or where external connectivity may be 404 temporarily unavailable. A homenet using provider-assigned global 405 addresses is exposed to its ISP renumbering the network to a much 406 larger degree than before whereas, for IPv4, NAT isolated the user 407 against ISP renumbering to some extent. 409 While setting up a network there may be a period where it has no 410 external connectivity, in which case ULAs would be required for 411 inter-subnet communication. In the case where home automation 412 networks are being set up in a new home/deployment (as early as 413 during construction of the home), such networks will likely need to 414 use their own /48 ULA prefix. Depending upon circumstances beyond 415 the control of the owner of the homenet, it may be impossible to 416 renumber the ULA used by the home automation network so routing 417 between ULA /48s may be required. Also, some devices, particularly 418 constrained devices, may have only a ULA (in addition to a link- 419 local), while others may have both a GUA and a ULA. 421 Note that unlike private IPv4 RFC 1918 space, the use of ULAs does 422 not imply use of an IPv6 equivalent of a traditional IPv4 NAT 423 [RFC3022], or of NPTv6 prefix-based NAT [RFC6296]. When an IPv6 node 424 in a homenet has both a ULA and a globally unique IPv6 address, it 425 should only use its ULA address internally, and use its additional 426 globally unique IPv6 address as a source address for external 427 communications. This should be the natural behaviour given support 428 for Default Address Selection for IPv6 [RFC6724]. By using such 429 globally unique addresses between hosts and devices in remote 430 networks, the architectural cost and complexity, particularly to 431 applications, of NAT or NPTv6 translation is avoided. As such, 432 neither IPv6 NAT or NPTv6 is recommended for use in the homenet 433 architecture. Further, the homenet border router(s) should filter 434 packets with ULA source/destination addresses as discussed in 435 Section 3.4.2. 437 Devices in a homenet may be given only a ULA as a means to restrict 438 reachability from outside the homenet. ULAs can be used by default 439 for devices that, without additional configuration (e.g., via a web 440 interface), would only offer services to the internal network. For 441 example, a printer might only accept incoming connections on a ULA 442 until configured to be globally reachable, at which point it acquires 443 a global IPv6 address and may be advertised via a global name space. 445 Where both a ULA and a global prefix are in use, the ULA source 446 address is used to communicate with ULA destination addresses when 447 appropriate, i.e., when the ULA source and destination lie within the 448 /48 ULA prefix(es) known to be used within the same homenet. In 449 cases where multiple /48 ULA prefixes are in use within a single 450 homenet (perhaps because multiple homenet routers each independently 451 auto-generate a /48 ULA prefix and then share prefix/routing 452 information), utilising a ULA source address and a ULA destination 453 address from two disjoint internal ULA prefixes is preferable to 454 using GUAs. 456 While a homenet should operate correctly with two or more /48 ULAs 457 enabled, a mechanism for the creation and use of a single /48 ULA 458 prefix is desirable for addressing consistency and policy 459 enforcement. 461 A counter-argument to using ULAs is that it is undesirable to 462 aggressively deprecate global prefixes for temporary loss of 463 connectivity, so for a host to lose its global address there would 464 have to be a connection breakage longer than the lease period, and 465 even then, deprecating prefixes when there is no connectivity may not 466 be advisable. However, it is assumed in this architecture that 467 homenets should support and use ULAs. 469 2.5. Avoiding manual configuration of IP addresses 471 Some IPv4 home networking devices expose IPv4 addresses to users, 472 e.g., the IPv4 address of a home IPv4 CER that may be configured via 473 a web interface. In potentially complex future IPv6 homenets, users 474 should not be expected to enter IPv6 literal addresses in devices or 475 applications, given their much greater length and the apparent 476 randomness of such addresses to a typical home user. Thus, even for 477 the simplest of functions, simple naming and the associated (minimal, 478 and ideally zero configuration) discovery of services is imperative 479 for the easy deployment and use of homenet devices and applications. 481 2.6. IPv6-only operation 483 It is likely that IPv6-only networking will be deployed first in new 484 home network deployments, often referred to as 'greenfield' 485 scenarios, where there is no existing IPv4 capability, or perhaps as 486 one element of an otherwise dual-stack network. Running IPv6-only 487 adds additional requirements, e.g., for devices to get configuration 488 information via IPv6 transport (not relying on an IPv4 protocol such 489 as IPv4 DHCP), and for devices to be able to initiate communications 490 to external devices that are IPv4-only. 492 Some specific transition technologies which may be deployed by the 493 homenet's ISP are discussed in [RFC7084]. In addition, certain other 494 functions may be desirable on the CER, e.g., to access content in the 495 IPv4 Internet, NAT64 [RFC6144] and DNS64 [RFC6145] may be applicable. 497 The widespread availability of robust solutions to these types of 498 requirements will help accelerate the uptake of IPv6-only homenets. 499 The specifics of these are however beyond the scope of this document, 500 especially those functions that reside on the CER. 502 3. Homenet Architecture Principles 504 The aim of this text is to outline how to construct advanced IPv6- 505 based home networks involving multiple routers and subnets using 506 standard IPv6 addressing and protocols [RFC2460] [RFC4291] as the 507 basis. As described in Section 3.1, solutions should as far as 508 possible re-use existing protocols, and minimise changes to hosts and 509 routers, but some new protocols, or extensions, are likely to be 510 required. In this section, we present the elements of the proposed 511 home networking architecture, with discussion of the associated 512 design principles. 514 In general, home network equipment needs to be able to operate in 515 networks with a range of different properties and topologies, where 516 home users may plug components together in arbitrary ways and expect 517 the resulting network to operate. Significant manual configuration 518 is rarely, if at all, possible, or even desirable given the knowledge 519 level of typical home users. Thus the network should, as far as 520 possible, be self-configuring, though configuration by advanced users 521 should not be precluded. 523 The homenet needs to be able to handle or provision at least 524 o Routing 526 o Prefix configuration for routers 528 o Name resolution 530 o Service discovery 532 o Network security 534 The remainder of this document describes the principles by which the 535 homenet architecture may deliver these properties. 537 3.1. General Principles 539 There is little that the Internet standards community can do about 540 the physical topologies or the need for some networks to be separated 541 at the network layer for policy or link layer compatibility reasons. 542 However, there is a lot of flexibility in using IP addressing and 543 inter-networking mechanisms. This text discusses how such 544 flexibility should be used to provide the best user experience and 545 ensure that the network can evolve with new applications in the 546 future. The principles described in this text should be followed 547 when designing homenet protocol solutions. 549 3.1.1. Reuse existing protocols 551 Existing protocols will be used to meet the requirements of home 552 networks. Where necessary, extensions will be made to those 553 protocols. When no existing protocol is found to be suitable, a new 554 or emerging protocol may be used. Therefore, it is important that no 555 design or architectural decisions are made that would preclude the 556 use of new or emerging protocols. 558 A generally conservative approach, giving weight to running (and 559 available) code, is preferable. Where new protocols are required, 560 evidence of commitment to implementation by appropriate vendors or 561 development communities is highly desirable. Protocols used should 562 be backwardly compatible, and forward compatible where changes are 563 made. 565 3.1.2. Minimise changes to hosts and routers 567 In order to maximise deployability of new homenets, where possible 568 any requirement for changes to hosts and routers should be minimised, 569 though solutions which, for example, incrementally improve capability 570 with host or router changes may be acceptable. There may be cases 571 where changes are unavoidable, e.g., to allow a given homenet routing 572 protocol to be self-configuring, or to support routing based on 573 sources addresses in addition to destination addresses (to improve 574 multihoming support, as discussed in Section 3.2.4). 576 3.2. Homenet Topology 578 This section considers homenet topologies, and the principles that 579 may be applied in designing an architecture to support as wide a 580 range of such topologies as possible. 582 3.2.1. Supporting arbitrary topologies 584 There should ideally be no built-in assumptions about the topology in 585 home networks, as users are capable of connecting their devices in 586 'ingenious' ways. Thus arbitrary topologies and arbitrary routing 587 will need to be supported, or at least the failure mode for when the 588 user makes a mistake should be as robust as possible, e.g., de- 589 activating a certain part of the infrastructure to allow the rest to 590 operate. In such cases, the user should ideally have some useful 591 indication of the failure mode encountered. 593 There should be no topology scenarios which cause loss of 594 connectivity, except when the user creates a physical island within 595 the topology. Some potentially pathological cases that can be 596 created include bridging ports of a router together, however this 597 case can be detected and dealt with by the router. Loops within a 598 routed topology are in a sense good in that they offer redundancy. 599 Topologies that include potential bridging loops can be dangerous but 600 are also detectable when a switch learns the MAC of one of its 601 interfaces on another or runs a spanning tree or link state protocol. 602 It is only topologies with such potential loops using simple 603 repeaters that are truly pathological. 605 The topology of the homenet may change over time, due to the addition 606 or removal of equipment, but also due to temporary failures or 607 connectivity problems. In some cases this may lead to, for example, 608 a multihomed homenet being split into two isolated homenets, or, 609 after such a fault is remedied, two isolated parts reconfiguring back 610 to a single network. 612 3.2.2. Network topology models 614 As hinted above, while the architecture may focus on likely common 615 topologies, it should not preclude any arbitrary topology from being 616 constructed. 618 Most IPv4 home network models at the time of writing tend to be 619 relatively simple, typically a single NAT router to the ISP and a 620 single internal subnet but, as discussed earlier, evolution in 621 network architectures is driving more complex topologies, such as the 622 separation of guest and private networks. There may also be some 623 cascaded IPv4 NAT scenarios, which we mention in the next section. 624 For IPv6 homenets, the Network Architectures described in [RFC6204] 625 and its successor [RFC7084] should, as a minimum, be supported. 627 There are a number of properties or attributes of a home network that 628 we can use to describe its topology and operation. The following 629 properties apply to any IPv6 home network: 631 o Presence of internal routers. The homenet may have one or more 632 internal routers, or may only provide subnetting from interfaces 633 on the CER. 635 o Presence of isolated internal subnets. There may be isolated 636 internal subnets, with no direct connectivity between them within 637 the homenet (with each having its own external connectivity). 638 Isolation may be physical, or implemented via IEEE 802.1q VLANs. 639 The latter is however not something a typical user would be 640 expected to configure. 642 o Demarcation of the CER. The CER(s) may or may not be managed by 643 the ISP. If the demarcation point is such that the customer can 644 provide or manage the CER, its configuration must be simple. Both 645 models must be supported. 647 Various forms of multihoming are likely to become more prevalent with 648 IPv6 home networks, where the homenet may have two or more external 649 ISP connections, as discussed further below. Thus the following 650 properties should also be considered for such networks: 652 o Number of upstream providers. The majority of home networks today 653 consist of a single upstream ISP, but it may become more common in 654 the future for there to be multiple ISPs, whether for resilience 655 or provision of additional services. Each would offer its own 656 prefix. Some may or may not provide a default route to the public 657 Internet. 659 o Number of CERs. The homenet may have a single CER, which might be 660 used for one or more providers, or multiple CERs. The presence of 661 multiple CERs adds additional complexity for multihoming 662 scenarios, and protocols like PCP that may need to manage 663 connection-oriented state mappings on the same CER as used for 664 subsequent traffic flows. 666 In the following sections we give some examples of the types of 667 homenet topologies we may see in the future. This is not intended to 668 be an exhaustive or complete list, rather an indicative one to 669 facilitate the discussion in this text. 671 3.2.2.1. A: Single ISP, Single CER, Internal routers 673 Figure 1 shows a home network with multiple local area networks. 674 These may be needed for reasons relating to different link layer 675 technologies in use or for policy reasons, e.g., classic Ethernet in 676 one subnet and a LLN link layer technology in another. In this 677 example there is no single router that a priori understands the 678 entire topology. The topology itself may also be complex, and it may 679 not be possible to assume a pure tree form, for instance (because 680 home users may plug routers together to form arbitrary topologies 681 including those with potential loops in them). 683 +-------+-------+ \ 684 | Service | \ 685 | Provider | | Service 686 | Router | | Provider 687 +-------+-------+ | network 688 | / 689 | Customer / 690 | Internet connection 691 | 692 +------+--------+ \ 693 | IPv6 | \ 694 | Customer Edge | \ 695 | Router | | 696 +----+-+---+----+ | 697 Network A | | | Network B(E) | 698 ----+-------------+----+ | +---+-------------+------+ | 699 | | | | | | | 700 +----+-----+ +-----+----+ | +----+-----+ +-----+----+ | | 701 |IPv6 Host | |IPv6 Host | | | IPv6 Host| |IPv6 Host | | | 702 | H1 | | H2 | | | H3 | | H4 | | | 703 +----------+ +----------+ | +----------+ +----------+ | | 704 | | | | | 705 Link F | ---+------+------+-----+ | 706 | | Network E(B) | 707 +------+--------+ | | End-User 708 | IPv6 | | | networks 709 | Interior +------+ | 710 | Router | | 711 +---+-------+-+-+ | 712 Network C | | Network D | 713 ----+-------------+---+ +---+-------------+--- | 714 | | | | | 715 +----+-----+ +-----+----+ +----+-----+ +-----+----+ | 716 |IPv6 Host | |IPv6 Host | | IPv6 Host| |IPv6 Host | | 717 | H5 | | H6 | | H7 | | H8 | / 718 +----------+ +----------+ +----------+ +----------+ / 720 Figure 1 722 In this diagram there is one CER. It has a single uplink interface. 723 It has three additional interfaces connected to Network A, Link F, 724 and Network B. IPv6 Internal Router (IR) has four interfaces 725 connected to Link F, Network C, Network D and Network E. Network B 726 and Network E have been bridged, likely inadvertently. This could be 727 as a result of connecting a wire between a switch for Network B and a 728 switch for Network E. 730 Any of logical Networks A through F might be wired or wireless. 732 Where multiple hosts are shown, this might be through one or more 733 physical ports on the CER or IPv6 (IR), wireless networks, or through 734 one or more layer-2 only Ethernet switches. 736 3.2.2.2. B: Two ISPs, Two CERs, Shared subnet 738 +-------+-------+ +-------+-------+ \ 739 | Service | | Service | \ 740 | Provider A | | Provider B | | Service 741 | Router | | Router | | Provider 742 +------+--------+ +-------+-------+ | network 743 | | / 744 | Customer | / 745 | Internet connections | / 746 | | 747 +------+--------+ +-------+-------+ \ 748 | IPv6 | | IPv6 | \ 749 | Customer Edge | | Customer Edge | \ 750 | Router 1 | | Router 2 | / 751 +------+--------+ +-------+-------+ / 752 | | / 753 | | | End-User 754 ---+---------+---+---------------+--+----------+--- | network(s) 755 | | | | \ 756 +----+-----+ +-----+----+ +----+-----+ +-----+----+ \ 757 |IPv6 Host | |IPv6 Host | | IPv6 Host| |IPv6 Host | / 758 | H1 | | H2 | | H3 | | H4 | / 759 +----------+ +----------+ +----------+ +----------+ 761 Figure 2 763 Figure 2 illustrates a multihomed homenet model, where the customer 764 has connectivity via CER1 to ISP A and via CER2 to ISP B. This 765 example shows one shared subnet where IPv6 nodes would potentially be 766 multihomed and receive multiple IPv6 global prefixes, one per ISP. 767 This model may also be combined with that shown in Figure 1 to create 768 a more complex scenario with multiple internal routers. Or the above 769 shared subnet may be split in two, such that each CER serves a 770 separate isolated subnet, which is a scenario seen with some IPv4 771 networks today. 773 3.2.2.3. C: Two ISPs, One CER, Shared subnet 775 +-------+-------+ +-------+-------+ \ 776 | Service | | Service | \ 777 | Provider A | | Provider B | | Service 778 | Router | | Router | | Provider 779 +-------+-------+ +-------+-------+ | network 780 | | / 781 | Customer | / 782 | Internet | / 783 | connections | 784 +---------+---------+ \ 785 | IPv6 | \ 786 | Customer Edge | \ 787 | Router | / 788 +---------+---------+ / 789 | / 790 | | End-User 791 ---+------------+-------+--------+-------------+--- | network(s) 792 | | | | \ 793 +----+-----+ +----+-----+ +----+-----+ +-----+----+ \ 794 |IPv6 Host | |IPv6 Host | | IPv6 Host| |IPv6 Host | / 795 | H1 | | H2 | | H3 | | H4 | / 796 +----------+ +----------+ +----------+ +----------+ 798 Figure 3 800 Figure 3 illustrates a model where a home network may have multiple 801 connections to multiple providers or multiple logical connections to 802 the same provider, with shared internal subnets. 804 3.2.3. Dual-stack topologies 806 It is expected that most homenet deployments will for the immediate 807 future be dual-stack IPv4/IPv6. In such networks it is important not 808 to introduce new IPv6 capabilities that would cause a failure if used 809 alongside IPv4+NAT, given that such dual-stack homenets will be 810 commonplace for some time. That said, it is desirable that IPv6 811 works better than IPv4 in as many scenarios as possible. Further, 812 the homenet architecture must operate in the absence of IPv4. 814 A general recommendation is to follow the same topology for IPv6 as 815 is used for IPv4, but not to use NAT. Thus there should be routed 816 IPv6 where an IPv4 NAT is used and, where there is no NAT, routing or 817 bridging may be used. Routing may have advantages when compared to 818 bridging together high speed and lower speed shared media, and in 819 addition bridging may not be suitable for some networks, such as ad- 820 hoc mobile networks. 822 In some cases IPv4 home networks may feature cascaded NATs. End 823 users are frequently unaware that they have created such networks as 824 'home routers' and 'home switches' are frequently confused. In 825 addition, there are cases where NAT routers are included within 826 Virtual Machine Hypervisors, or where Internet connection sharing 827 services have been enabled. This document applies equally to such 828 hidden NAT 'routers'. IPv6 routed versions of such cases will be 829 required. We should thus also note that routers in the homenet may 830 not be separate physical devices; they may be embedded within other 831 devices. 833 3.2.4. Multihoming 835 A homenet may be multihomed to multiple providers, as the network 836 models above illustrate. This may either take a form where there are 837 multiple isolated networks within the home or a more integrated 838 network where the connectivity selection needs to be dynamic. 839 Current practice is typically of the former kind, but the latter is 840 expected to become more commonplace. 842 In the general homenet architecture, multihomed hosts should be 843 multi-addressed with a global IPv6 address from the global prefix 844 delegated from each ISP they communicate with or through. When such 845 multi-addressing is in use, hosts need some way to pick source and 846 destination address pairs for connections. A host may choose a 847 source address to use by various methods, most commonly [RFC6724]. 848 Applications may of course do different things, and this should not 849 be precluded. 851 For the single CER Network Model C illustrated above, multihoming may 852 be offered by source-based routing at the CER. With multiple exit 853 routers, as in CER Network Model B, the complexity rises. Given a 854 packet with a source address on the home network, the packet must be 855 routed to the proper egress to avoid BCP 38 ingress filtering if 856 exiting through the wrong ISP. It is highly desirable that the 857 packet is routed in the most efficient manner to the correct exit, 858 though as a minimum requirement the packet should not be dropped. 860 The homenet architecture should support both the above models, i.e., 861 one or more CERs. However, the general multihoming problem is broad, 862 and solutions suggested to date within the IETF have included complex 863 architectures for monitoring connectivity, traffic engineering, 864 identifier-locator separation, connection survivability across 865 multihoming events, and so on. It is thus important that the homenet 866 architecture should as far as possible minimise the complexity of any 867 multihoming support. 869 An example of such a 'simpler' approach has been documented in 870 [RFC7157]. Alternatively a flooding/routing protocol could 871 potentially be used to pass information through the homenet, such 872 that internal routers and ultimately end hosts could learn per-prefix 873 configuration information, allowing better address selection 874 decisions to be made. However, this would imply router and, most 875 likely, host changes. Another avenue is to introduce support 876 throughout the homenet for routing which is based on the source as 877 well as the destination address of each packet. While greatly 878 improving the 'intelligence' of routing decisions within the homenet, 879 such an approach would require relatively significant router changes 880 but avoid host changes. 882 As explained previously, while NPTv6 has been proposed for providing 883 multi-homing support in networks, its use is not recommended in the 884 homenet architecture. 886 It should be noted that some multihoming scenarios may see one 887 upstream being a "walled garden", and thus only appropriate for 888 connectivity to the services of that provider; an example may be a 889 VPN service that only routes back to the enterprise business network 890 of a user in the homenet. As per Section 4.2.1 of [RFC3002] we do 891 not specifically target walled garden multihoming as a goal of this 892 document. 894 The homenet architecture should also not preclude use of host or 895 application-oriented tools, e.g., Shim6 [RFC5533], MPTCP [RFC6824] or 896 Happy Eyeballs [RFC6555]. In general, any incremental improvements 897 obtained by host changes should give benefit for the hosts 898 introducing them, but not be required. 900 3.2.5. Mobility support 902 Devices may be mobile within the homenet. While resident on the same 903 subnet, their address will remain persistent, but should devices move 904 to a different (wireless) subnet, they will acquire a new address in 905 that subnet. It is desirable that the homenet supports internal 906 device mobility. To do so, the homenet may either extend the reach 907 of specific wireless subnets to enable wireless roaming across the 908 home (availability of a specific subnet across the home), or it may 909 support mobility protocols to facilitate such roaming where multiple 910 subnets are used. 912 3.3. A Self-Organising Network 914 The home network infrastructure should be naturally self-organising 915 and self-configuring under different circumstances relating to the 916 connectivity status to the Internet, number of devices, and physical 917 topology. At the same time, it should be possible for advanced users 918 to manually adjust (override) the current configuration. 920 While a goal of the homenet architecture is for the network to be as 921 self-organising as possible, there may be instances where some manual 922 configuration is required, e.g., the entry of a cryptographic key to 923 apply wireless security, or to configure a shared routing secret. 924 The latter may be relevant when considering how to bootstrap a 925 routing configuration. It is highly desirable that the number of 926 such configurations is minimised. 928 3.3.1. Differentiating neighbouring homenets 930 It is important that self-configuration with 'unintended' devices is 931 avoided. There should be a way for a user to administratively assert 932 in a simple way whether or not a device belongs to a given homenet. 933 The goal is to allow the establishment of borders, particularly 934 between two adjacent homenets, and to avoid unauthorised devices from 935 participating in the homenet. Such an authorisation capability may 936 need to operate through multiple hops in the homenet. 938 The homenet should thus support a way for a homenet owner to claim 939 ownership of their devices in a reasonably secure way. This could be 940 achieved by a pairing mechanism, by for example pressing buttons 941 simultaneously on an authenticated and a new homenet device, or by an 942 enrolment process as part of an autonomic networking environment. 944 While there may be scenarios where one homenet may wish to 945 intentionally gain access through another, e.g. to share external 946 connectivity costs, such scenarios are not discussed in this 947 document. 949 3.3.2. Largest practical subnets 951 Today's IPv4 home networks generally have a single subnet, and early 952 dual-stack deployments have a single congruent IPv6 subnet, possibly 953 with some bridging functionality. More recently, some vendors have 954 started to introduce 'home' and 'guest' functions, which in IPv6 955 would be implemented as two subnets. 957 Future home networks are highly likely to have one or more internal 958 routers and thus need multiple subnets, for the reasons described 959 earlier. As part of the self-organisation of the network, the 960 homenet should subdivide itself into the largest practical subnets 961 that can be constructed within the constraints of link layer 962 mechanisms, bridging, physical connectivity, and policy, and where 963 applicable performance or other criteria. In such subdivisions the 964 logical topology may not necessarily match the physical topology. 966 This text does not, however, make recommendations on how such 967 subdivision should occur. It is expected that subsequent documents 968 will address this problem. 970 While it may be desirable to maximise the chance of link-local 971 protocols operating across a homenet by maximising the size of a 972 subnet, multi-subnet home networks are inevitable, so their support 973 must be included. 975 3.3.3. Handling varying link technologies 977 Homenets tend to grow organically over many years, and a homenet will 978 typically be built over link-layer technologies from different 979 generations. Current homenets typically use links ranging from 980 1Mbit/s up to 1Gbit/s, which is a three orders of magnitude 981 throughput discrepancy. We expect this discrepancy to widen further 982 as both high-speed and low-power technologies are deployed. 984 Homenet protocols should be designed to deal well with 985 interconnecting links of very different throughputs. In particular, 986 flows local to a link should not be flooded throughout the homenet, 987 even when sent over multicast, and, whenever possible, the homenet 988 protocols should be able to choose the faster links and avoid the 989 slower ones. 991 Links (particularly wireless links) may also have limited numbers of 992 transmit opportunities (txops), and there is a clear trend driven by 993 both power and downward compatibility constraints toward aggregation 994 of packets into these limited txops while increasing throughput. 995 Transmit opportunities may be a system's scarcest resource and 996 therefore also strongly limit actual throughput available. 998 3.3.4. Homenet realms and borders 1000 The homenet will need to be aware of the extent of its own 'site', 1001 which will, for example, define the borders for ULA and site scope 1002 multicast traffic, and may require specific security policies to be 1003 applied. The homenet will have one or more such borders with 1004 external connectivity providers. 1006 A homenet will most likely also have internal borders between 1007 internal realms, e.g., a guest realm or a corporate network extension 1008 realm. It is desirable that appropriate borders can be configured to 1009 determine, for example, the scope of where network prefixes, routing 1010 information, network traffic, service discovery and naming may be 1011 shared. The default mode internally should be to share everything. 1013 It is expected that a realm would span at least an entire subnet, and 1014 thus the borders lie at routers which receive delegated prefixes 1015 within the homenet. It is also desirable, for a richer security 1016 model, that hosts are able to make communication decisions based on 1017 available realm and associated prefix information in the same way 1018 that routers at realm borders can. 1020 A simple homenet model may just consider three types of realm and the 1021 borders between them, namely the internal homenet, the ISP and a 1022 guest network. In this case the borders will include that from the 1023 homenet to the ISP, that from the guest network to the ISP, and that 1024 from the homenet to the guest network. Regardless, it should be 1025 possible for additional types of realms and borders to be defined, 1026 e.g., for some specific LLN-based network, such as Smart Grid, and 1027 for these to be detected automatically, and for an appropriate 1028 default policy to be applied as to what type of traffic/data can flow 1029 across such borders. 1031 It is desirable to classify the external border of the home network 1032 as a unique logical interface separating the home network from 1033 service provider network/s. This border interface may be a single 1034 physical interface to a single service provider, multiple layer 2 1035 sub-interfaces to a single service provider, or multiple connections 1036 to a single or multiple providers. This border makes it possible to 1037 describe edge operations and interface requirements across multiple 1038 functional areas including security, routing, service discovery, and 1039 router discovery. 1041 It should be possible for the homenet user to override any 1042 automatically determined borders and the default policies applied 1043 between them, the exception being that it may not be possible to 1044 override policies defined by the ISP at the external border. 1046 3.3.5. Configuration information from the ISP 1048 In certain cases, it may be useful for the homenet to get certain 1049 configuration information from its ISP. For example, the homenet 1050 DHCP server may request and forward some options that it gets from 1051 its upstream DHCP server, though the specifics of the options may 1052 vary across deployments. There is potential complexity here of 1053 course should the homenet be multihomed. 1055 3.4. Homenet Addressing 1057 The IPv6 addressing scheme used within a homenet must conform to the 1058 IPv6 addressing architecture [RFC4291]. In this section we discuss 1059 how the homenet needs to adapt to the prefixes made available to it 1060 by its upstream ISP, such that internal subnets, hosts and devices 1061 can obtain the and configure the necessary addressing information to 1062 operate. 1064 3.4.1. Use of ISP-delegated IPv6 prefixes 1066 Discussion of IPv6 prefix allocation policies is included in 1067 [RFC6177]. In practice, a homenet may receive an arbitrary length 1068 IPv6 prefix from its provider, e.g., /60, /56 or /48. The offered 1069 prefix may be stable or change from time to time; it is generally 1070 expected that ISPs will offer relatively stable prefixes to their 1071 residential customers. Regardless, the home network needs to be 1072 adaptable as far as possible to ISP prefix allocation policies, and 1073 thus make no assumptions about the stability of the prefix received 1074 from an ISP, or the length of the prefix that may be offered. 1076 However, if, for example, only a /64 is offered by the ISP, the 1077 homenet may be severely constrained or even unable to function. 1078 [RFC6177] (BCP 157) states that "a key principle for address 1079 management is that end sites always be able to obtain a reasonable 1080 amount of address space for their actual and planned usage, and over 1081 time ranges specified in years rather than just months. In practice, 1082 that means at least one /64, and in most cases significantly more. 1083 One particular situation that must be avoided is having an end site 1084 feel compelled to use IPv6-to-IPv6 Network Address Translation or 1085 other burdensome address conservation techniques because it could not 1086 get sufficient address space." This architecture document assumes 1087 that the guidance in the quoted text is being followed by ISPs. 1089 There are many problems that would arise from a homenet not being 1090 offered a sufficient prefix size for its needs. Rather than attempt 1091 to contrive a method for a homenet to operate in a constrained manner 1092 when faced with insufficient prefixes, such as the use of subnet 1093 prefixes longer than /64 (which would break stateless address 1094 autoconfiguration [RFC4862]), use of NPTv6, or falling back to 1095 bridging across potentially very different media, it is recommended 1096 that the receiving router instead enters an error state and issues 1097 appropriate warnings. Some consideration may need to be given to how 1098 such a warning or error state should best be presented to a typical 1099 home user. 1101 Thus a homenet CER should request, for example via DHCP Prefix 1102 Delegation (DHCP PD) [RFC3633], that it would like a /48 prefix from 1103 its ISP, i.e., it asks the ISP for the maximum size prefix it might 1104 expect to be offered, even if in practice it may only be offered a 1105 /56 or /60. For a typical IPv6 homenet, it is not recommended that 1106 an ISP offer less than a /60 prefix, and it is highly preferable that 1107 the ISP offers at least a /56. It is expected that the allocated 1108 prefix to the homenet from any single ISP is a contiguous, aggregated 1109 one. While it may be possible for a homenet CER to issue multiple 1110 prefix requests to attempt to obtain multiple delegations, such 1111 behaviour is out of scope of this document. 1113 The norm for residential customers of large ISPs may be similar to 1114 their single IPv4 address provision; by default it is likely to 1115 remain persistent for some time, but changes in the ISP's own 1116 provisioning systems may lead to the customer's IP (and in the IPv6 1117 case their prefix pool) changing. It is not expected that ISPs will 1118 generally support Provider Independent (PI) addressing for 1119 residential homenets. 1121 When an ISP does need to restructure, and in doing so renumber its 1122 customer homenets, 'flash' renumbering is likely to be imposed. This 1123 implies a need for the homenet to be able to handle a sudden 1124 renumbering event which, unlike the process described in [RFC4192], 1125 would be a 'flag day" event, which means that a graceful renumbering 1126 process moving through a state with two active prefixes in use would 1127 not be possible. While renumbering can be viewed as an extended 1128 version of an initial numbering process, the difference between flash 1129 renumbering and an initial 'cold start' is the need to provide 1130 service continuity. 1132 There may be cases where local law means some ISPs are required to 1133 change IPv6 prefixes (current IPv4 addresses) for privacy reasons for 1134 their customers. In such cases it may be possible to avoid an 1135 instant 'flash' renumbering and plan a non-flag day renumbering as 1136 per RFC 4192. Similarly, if an ISP has a planned renumbering 1137 process, it may be able to adjust lease timers, etc appropriately. 1139 The customer may of course also choose to move to a new ISP, and thus 1140 begin using a new prefix. In such cases the customer should expect a 1141 discontinuity, and not only may the prefix change, but potentially 1142 also the prefix length if the new ISP offers a different default size 1143 prefix. The homenet may also be forced to renumber itself if 1144 significant internal 'replumbing' is undertaken by the user. 1145 Regardless, it's desirable that homenet protocols support rapid 1146 renumbering and that operational processes don't add unnecessary 1147 complexity for the renumbering process. Further, the introduction of 1148 any new homenet protocols should not make any form of renumbering any 1149 more complex than it already is. 1151 Finally, the internal operation of the home network should also not 1152 depend on the availability of the ISP network at any given time, 1153 other than of course for connectivity to services or systems off the 1154 home network. This reinforces the use of ULAs for stable internal 1155 communication, and the need for a naming and service discovery 1156 mechanism that can operate independently within the homenet. 1158 3.4.2. Stable internal IP addresses 1160 The network should by default attempt to provide IP-layer 1161 connectivity between all internal parts of the homenet as well as to 1162 and from the external Internet, subject to the filtering policies or 1163 other policy constraints discussed later in the security section. 1165 ULAs should be used within the scope of a homenet to support stable 1166 routing and connectivity between subnets and hosts regardless of 1167 whether a globally unique ISP-provided prefix is available. In the 1168 case of a prolonged external connectivity outage, ULAs allow internal 1169 operations across routed subnets to continue. ULA addresses also 1170 allow constrained devices to create permanent relationships between 1171 IPv6 addresses, e.g., from a wall controller to a lamp, where 1172 symbolic host names would require additional non-volatile memory and 1173 updating global prefixes in sleeping devices might also be 1174 problematic. 1176 As discussed previously, it would be expected that ULAs would 1177 normally be used alongside one or more global prefixes in a homenet, 1178 such that hosts become multi-addressed with both globally unique and 1179 ULA prefixes. ULAs should be used for all devices, not just those 1180 intended to only have internal connectivity. Default address 1181 selection would then enable ULAs to be preferred for internal 1182 communications between devices that are using ULA prefixes generated 1183 within the same homenet. 1185 In cases where ULA prefixes are in use within a homenet but there is 1186 no external IPv6 connectivity (and thus no GUAs in use), 1187 recommendations ULA-5, L-3 and L-4 in RFC 6204 should be followed to 1188 ensure correct operation, in particular where the homenet may be 1189 dual-stack with IPv4 external connectivity. The use of the Route 1190 Information Option described in [RFC4191] provides a mechanism to 1191 advertise such more-specific ULA routes. 1193 The use of ULAs should be restricted to the homenet scope through 1194 filtering at the border(s) of the homenet, as mandated by RFC 6204 1195 requirement S-2. 1197 Note that it is possible that in some cases multiple /48 ULA prefixes 1198 may be in use within the same homenet, e.g., when the network is 1199 being deployed, perhaps also without external connectivity. In cases 1200 where multiple ULA /48's are in use, hosts need to know that each /48 1201 is local to the homenet, e.g., by inclusion in their local address 1202 selection policy table. 1204 3.4.3. Internal prefix delegation 1206 As mentioned above, there are various sources of prefixes. From the 1207 homenet perspective, a single global prefix from each ISP should be 1208 received on the border CER [RFC3633]. Where multiple CERs exist with 1209 multiple ISP prefix pools, it is expected that routers within the 1210 homenet would assign themselves prefixes from each ISP they 1211 communicate with/through. As discussed above, a ULA prefix should be 1212 provisioned for stable internal communications or for use on 1213 constrained/LLN networks. 1215 The delegation or availability of a prefix pool to the homenet should 1216 allow subsequent internal autonomous delegation of prefixes for use 1217 within the homenet. Such internal delegation should not assume a 1218 flat or hierarchical model, nor should it make an assumption about 1219 whether the delegation of internal prefixes is distributed or 1220 centralised. The assignment mechanism should provide reasonable 1221 efficiency, so that typical home network prefix allocation sizes can 1222 accommodate all the necessary /64 allocations in most cases, and not 1223 waste prefixes. Further, duplicate assignment of multiple /64s to 1224 the same network should be avoided, and the network should behave as 1225 gracefully as possible in the event of prefix exhaustion (though the 1226 options in such cases may be limited). 1228 Where the home network has multiple CERs and these are delegated 1229 prefix pools from their attached ISPs, the internal prefix delegation 1230 would be expected to be served by each CER for each prefix associated 1231 with it. Where ULAs are used, it is preferable that only one /48 ULA 1232 covers the whole homenet, from which /64's can be delegated to the 1233 subnets. In cases where two /48 ULAs are generated within a homenet, 1234 the network should still continue to function, meaning that hosts 1235 will need to determine that each ULA is local to the homenet. 1237 Delegation within the homenet should result in each link being 1238 assigned a stable prefix that is persistent across reboots, power 1239 outages and similar short-term outages. The availability of 1240 persistent prefixes should not depend on the router boot order. The 1241 addition of a new routing device should not affect existing 1242 persistent prefixes, but persistence may not be expected in the face 1243 of significant 'replumbing' of the homenet. However, delegated ULA 1244 prefixes within the homenet should remain persistent through an ISP- 1245 driven renumbering event. 1247 Provisioning such persistent prefixes may imply the need for stable 1248 storage on routing devices, and also a method for a home user to 1249 'reset' the stored prefix should a significant reconfiguration be 1250 required (though ideally the home user should not be involved at 1251 all). 1253 This document makes no specific recommendation towards solutions, but 1254 notes that it is very likely that all routing devices participating 1255 in a homenet must use the same internal prefix delegation method. 1256 This implies that only one delegation method should be in use. 1258 3.4.4. Coordination of configuration information 1260 The network elements will need to be integrated in a way that takes 1261 account of the various lifetimes on timers that are used on different 1262 elements, e.g., DHCPv6 PD, router, valid prefix and preferred prefix 1263 timers. 1265 3.4.5. Privacy 1267 If ISPs offer relatively stable IPv6 prefixes to customers, the 1268 network prefix part of addresses associated with the homenet may not 1269 change over a reasonably long period of time. 1271 The exposure of which traffic is sourced from the same homenet is 1272 thus similar to IPv4; the single IPv4 global address seen through use 1273 of IPv4 NAT gives the same hint as the global IPv6 prefix seen for 1274 IPv6 traffic. 1276 While IPv4 NAT may obfuscate to an external observer which internal 1277 devices traffic is sourced from, IPv6, even with use of Privacy 1278 Addresses [RFC4941], adds additional exposure of which traffic is 1279 sourced from the same internal device, through use of the same IPv6 1280 source address for a period of time. 1282 3.5. Routing functionality 1284 Routing functionality is required when there are multiple routers 1285 deployed within the internal home network. This functionality could 1286 be as simple as the current 'default route is up' model of IPv4 NAT, 1287 or, more likely, it would involve running an appropriate routing 1288 protocol. Regardless of the solution method, the functionality 1289 discussed below should be met. 1291 The homenet unicast routing protocol should be based on a previously 1292 deployed protocol that has been shown to be reliable and robust, and 1293 that allows lightweight implementations, but that does not preclude 1294 the selection of a newer protocol for which a high quality open 1295 source implementation becomes available. Using information 1296 distributed through the routing protocol, each node in the homenet 1297 should be able to build a graph of the topology of the whole homenet 1298 including attributes such as links, nodes, connectivity, and (if 1299 supported by the protocol in use) link metrics. In the latter case, 1300 link metrics may be configured or automatically derived per-link 1301 based on consideration of factors such as worst-case queue depth and 1302 router processing capabilities. 1304 The routing protocol should support the generic use of multiple 1305 customer Internet connections, and the concurrent use of multiple 1306 delegated prefixes. A routing protocol that can make routing 1307 decisions based on source and destination addresses is thus 1308 desirable, to avoid upstream ISP BCP 38 ingress filtering problems. 1309 Multihoming support should also include load-balancing to multiple 1310 providers, and failover from a primary to a backup link when 1311 available. The protocol however should not require upstream ISP 1312 connectivity to be established to continue routing within the 1313 homenet. 1315 Multiple types of physical interfaces must be accounted for in the 1316 homenet routed topology. Technologies such as Ethernet, WiFi, 1317 Multimedia over Coax Alliance (MoCA), etc. must be capable of 1318 coexisting in the same environment and should be treated as part of 1319 any routed deployment. The inclusion of physical layer 1320 characteristics including bandwidth, loss, and latency in path 1321 computation should be considered for optimising communication in the 1322 homenet. 1324 The routing environment should be self-configuring, as discussed 1325 previously. Minimising convergence time should be a goal in any 1326 routed environment, but as a guideline a maximum convergence time at 1327 most 30 seconds should be the target (this target is somewhat 1328 arbitrary, and was chosen based on how long a typical home user might 1329 wait before attempting another reset; ideally the routers might have 1330 some status light indicating they are converging, similar to an ADSL 1331 router light indicating it is establishing a connection to its ISP). 1333 Homenets may use a variety of underlying link layer technologies, and 1334 may therefore benefit from being able to use link metrics if 1335 available. It may be beneficial for traffic to use multiple paths to 1336 a given destination within the homenet where available, rather than a 1337 single best path. 1339 At most one routing protocol should be in use at a given time in a 1340 given homenet. In some simple topologies, no routing protocol may be 1341 needed. If more than one routing protocol is supported by routers in 1342 a given homenet, then a mechanism is required to ensure that all 1343 routers in that homenet use the same protocol. 1345 An appropriate mechanism is required to discover which router(s) in 1346 the homenet are providing the CER function. Borders may include but 1347 are not limited to the interface to the upstream ISP, a gateway 1348 device to a separate home network such as a LLN network, or a gateway 1349 to a guest or private corporate extension network. In some cases 1350 there may be no border present, which may for example occur before an 1351 upstream connection has been established. The border discovery 1352 functionality may be integrated into the routing protocol itself, but 1353 may also be imported via a separate discovery mechanism. 1355 Ideally, LLN or other logically separate networks should be able 1356 exchange routes such that IP traffic may be forwarded among the 1357 networks via gateway routers which interoperate with both the homenet 1358 and the LLN. Current home deployments use largely different 1359 mechanisms in sensor and basic Internet connectivity networks. IPv6 1360 virtual machine (VM) solutions may also add additional routing 1361 requirements. 1363 3.5.1. Multicast support 1365 It is desirable that, subject to the capacities of devices on certain 1366 media types, multicast routing is supported across the homenet. 1368 [RFC4291] requires that any boundary of scope 4 or higher (i.e., 1369 admin-local or higher) be administratively configured. Thus the 1370 boundary at the homenet-ISP border must be administratively 1371 configured, though that may be triggered by an administrative 1372 function such as DHCP-PD. Other multicast forwarding policy borders 1373 may also exist within the homenet, e.g., to/from a guest subnet, 1374 whilst the use of certain link media types may also affect where 1375 specific multicast traffic is forwarded or routed. 1377 There may be different drivers for multicast to be supported across 1378 the homenet, e.g., for homenet-wide service discovery should a 1379 multicast service discovery protocol of scope greater than link-local 1380 be defined, or potentially for multicast-based streaming or 1381 filesharing applications. Where multicast is routed across a homenet 1382 an appropriate multicast routing protocol is required, one that as 1383 per the unicast routing protocol should be self-configuring. As 1384 hinted above, it must be possible to scope or filter multicast 1385 traffic to avoid it being flooded to network media where devices 1386 cannot reasonably support it. 1388 A homenet may not only use multicast internally, it may also be a 1389 consumer or provider of external multicast traffic, where the 1390 homenet's ISP supports such multicast operation. This may be 1391 valuable for example where live video applications are being sourced 1392 to/from the homenet. 1394 The multicast environment should support the ability for applications 1395 to pick a unique multicast group to use. 1397 3.6. Security 1399 The security of an IPv6 homenet is an important consideration. The 1400 most notable difference to the IPv4 operational model is the removal 1401 of NAT, the introduction of global addressability of devices, and 1402 thus a need to consider whether devices should have global 1403 reachability. Regardless, hosts need to be able to operate securely, 1404 end-to-end where required, and also be robust against malicious 1405 traffic directed towards them. However, there are other challenges 1406 introduced, e.g., default filtering policies at the borders between 1407 various homenet realms. 1409 3.6.1. Addressability vs reachability 1411 An IPv6-based home network architecture should embrace the 1412 transparent end-to-end communications model as described in 1413 [RFC2775]. Each device should be globally addressable, and those 1414 addresses must not be altered in transit. However, security 1415 perimeters can be applied to restrict end-to-end communications, and 1416 thus while a host may be globally addressable it may not be globally 1417 reachable. 1419 [RFC4864] describes a 'Simple Security' model for IPv6 networks, 1420 whereby stateful perimeter filtering can be applied to control the 1421 reachability of devices in a homenet. RFC 4864 states in Section 4.2 1422 that "the use of firewalls ... is recommended for those that want 1423 boundary protection in addition to host defences". It should be 1424 noted that a 'default deny' filtering approach would effectively 1425 replace the need for IPv4 NAT traversal protocols with a need to use 1426 a signalling protocol to request a firewall hole be opened, e.g., a 1427 protocol such as PCP [RFC6887]. In networks with multiple CERs, the 1428 signalling would need to handle the cases of flows that may use one 1429 or more exit routers. CERs would need to be able to advertise their 1430 existence for such protocols. 1432 [RFC6092] expands on RFC 4864, giving a more detailed discussion of 1433 IPv6 perimeter security recommendations, without mandating a 'default 1434 deny' approach. Indeed, RFC 6092 does not enforce a particular mode 1435 of operation, instead stating that CERs must provide an easily 1436 selected configuration option that permits a 'transparent' mode, thus 1437 ensuring a 'default allow' model is available. 1439 The topic of whether future home networks as described in this 1440 document should have have a 'default deny' or 'default allow' 1441 position has been discussed at length in various IETF meetings 1442 without any consensus being reached on which approach is more 1443 appropriate. Further, the choice of which default to apply may be 1444 situational, and thus this text makes no recommendation on the 1445 default setting beyond what is written on this topic in RFC 6092. We 1446 note in Section 3.6.3 below that the implicit firewall function of an 1447 IPv4 NAT is commonplace today, and thus future CERs targeted at home 1448 networks should continue to support the option of running in 'default 1449 deny mode', whether or not that is the default setting 1451 3.6.2. Filtering at borders 1453 It is desirable that there are mechanisms to detect different types 1454 of borders within the homenet, as discussed previously, and further 1455 mechanisms to then apply different types of filtering policies at 1456 those borders, e.g., whether naming and service discovery should pass 1457 a given border. Any such policies should be able to be easily 1458 applied by typical home users, e.g., to give a user in a guest 1459 network access to media services in the home, or access to a printer. 1460 Simple mechanisms to apply policy changes, or associations between 1461 devices, will be required. 1463 There are cases where full internal connectivity may not be 1464 desirable, e.g., in certain utility networking scenarios, or where 1465 filtering is required for policy reasons against guest network 1466 subnet(s). Some scenarios/models may as a result involve running 1467 isolated subnet(s) with their own CERs. In such cases connectivity 1468 would only be expected within each isolated network (though traffic 1469 may potentially pass between them via external providers). 1471 LLNs provide an another example of where there may be secure 1472 perimeters inside the homenet. Constrained LLN nodes may implement 1473 network key security but may depend on access policies enforced by 1474 the LLN border router. 1476 Considerations for differentiating neighbouring homenets are 1477 discussed in Section 3.3.1. 1479 3.6.3. Partial Effectiveness of NAT and Firewalls 1481 Security by way of obscurity (address translation) or through 1482 firewalls (filtering) is at best only partially effective. The very 1483 poor security track record of home computer, home networking and 1484 business PC computers and networking is testimony to this. A 1485 security compromise behind the firewall of any device exposes all 1486 others, making an entire network that relies on obscurity or a 1487 firewall as vulnerable as the most insecure device on the private 1488 side of the network. 1490 However, given current evidence of home network products with very 1491 poor default device security, putting a firewall in place does 1492 provide some level of protection. The use of firewalls today, 1493 whether a good practice or not, is common practice and the capability 1494 to afford protection via a 'default deny' setting, even if marginally 1495 effective, should not be lost. Thus, while it is highly desirable 1496 that all hosts in a homenet be adequately protected by built-in 1497 security functions, it should also be assumed that all CERs will 1498 continue to support appropriate perimeter defence functions, as per 1499 [RFC7084]. 1501 3.6.4. Exfiltration concerns 1503 As homenets become more complex, with more devices, and with service 1504 discovery potentially enabled across the whole home, there are 1505 potential concerns over the leakage of information should devices use 1506 discovery protocols to gather information and report it to equipment 1507 vendors or application service providers. 1509 While it is not clear how such exfiltration could be easily avoided, 1510 the threat should be recognised, be it from a new piece of hardware 1511 or some 'app' installed on a personal device. 1513 3.6.5. Device capabilities 1515 In terms of the devices, homenet hosts should implement their own 1516 security policies in accordance to their computing capabilities. 1517 They should have the means to request transparent communications to 1518 be able to be initiated to them through security filters in the 1519 homenet, either for all ports or for specific services. Users should 1520 have simple methods to associate devices to services that they wish 1521 to operate transparently through (CER) borders. 1523 3.6.6. ULAs as a hint of connection origin 1525 As noted in Section 3.6, if appropriate filtering is in place on the 1526 CER(s), as mandated by RFC 6204 requirement S-2, a ULA source address 1527 may be taken as an indication of locally sourced traffic. This 1528 indication could then be used with security settings to designate 1529 between which nodes a particular application is allowed to 1530 communicate, provided ULA address space is filtered appropriately at 1531 the boundary of the realm. 1533 3.7. Naming and Service Discovery 1535 The homenet requires devices to be able to determine and use unique 1536 names by which they can be accessed on the network, and which are not 1537 used by other devices on the network. Users and devices will need to 1538 be able to discover devices and services available on the network, 1539 e.g., media servers, printers, displays or specific home automation 1540 devices. Thus naming and service discovery must be supported in the 1541 homenet, and, given the nature of typical home network users, the 1542 service(s) providing this function must as far as possible support 1543 unmanaged operation. 1545 The naming system will be required to work internally or externally, 1546 be the user within the homenet or outside it, i.e., the user should 1547 be able to refer to devices by name, and potentially connect to them, 1548 wherever they may be. The most natural way to think about such 1549 naming and service discovery is to enable it to work across the 1550 entire homenet residence (site), disregarding technical borders such 1551 as subnets but respecting policy borders such as those between guest 1552 and other internal network realms. Remote access may be desired by 1553 the homenet residents while travelling, but also potentially by 1554 manufacturers or other 'benevolent' third parties. 1556 3.7.1. Discovering services 1558 Users will typically perform service discovery through graphical user 1559 interfaces (GUIs) that allow them to browse services on their network 1560 in an appropriate and intuitive way. Devices may also need to 1561 discover other devices, without any user intervention or choice. 1562 Either way, such interfaces are beyond the scope of this document, 1563 but the interface should have an appropriate application programming 1564 interface (API) for the discovery to be performed. 1566 Such interfaces may also typically hide the local domain name element 1567 from users, especially where only one name space is available. 1568 However, as we discuss below, in some cases the ability to discover 1569 available domains may be useful. 1571 We note that current zero-configuration service discovery protocols 1572 are generally aimed at single subnets. There is thus a choice to 1573 make for multi-subnet homenets as to whether such protocols should be 1574 proxied or extended to operate across a whole homenet. In this 1575 context, that may mean bridging a link-local method, taking care to 1576 avoid packets entering looping paths, or extending the scope of 1577 multicast traffic used for the purpose. It may mean that some proxy 1578 or hybrid service is utilised, perhaps co-resident on the CER. Or it 1579 may be that a new approach is preferable, e.g., flooding information 1580 around the homenet as attributes within the routing protocol (which 1581 could allow per-prefix configuration). However, we should prefer 1582 approaches that are backwardly compatible, and allow current 1583 implementations to continue to be used. Note that this document does 1584 not mandate a particular solution, rather it expresses the principles 1585 that should be used for a homenet naming and service discovery 1586 environment. 1588 One of the primary challenges facing service discovery today is lack 1589 of interoperability due to the ever increasing number of service 1590 discovery protocols available. While it is conceivable for consumer 1591 devices to support multiple discovery protocols, this is clearly not 1592 the most efficient use of network and computational resources. One 1593 goal of the homenet architecture should be a path to service 1594 discovery protocol interoperability either through a standards based 1595 translation scheme, hooks into current protocols to allow some for of 1596 communication among discovery protocols, extensions to support a 1597 central service repository in the homenet, or simply convergence 1598 towards a unified protocol suite. 1600 3.7.2. Assigning names to devices 1602 Given the large number of devices that may be networked in the 1603 future, devices should have a means to generate their own unique 1604 names within a homenet, and to detect clashes should they arise, 1605 e.g., where a second device of the same type/vendor as an existing 1606 device with the same default name is deployed, or where a new subnet 1607 is added to the homenet which already has a device of the same name. 1608 It is expected that a device should have a fixed name while within 1609 the scope of the homenet. 1611 Users will also want simple ways to (re)name devices, again most 1612 likely through an appropriate and intuitive interface that is beyond 1613 the scope of this document. Note the name a user assigns to a device 1614 may be a label that is stored on the device as an attribute of the 1615 device, and may be distinct from the name used in a name service, 1616 e.g., 'Study Laser Printer' as opposed to printer2.. 1618 3.7.3. The homenet name service 1620 The homenet name service should support both lookups and discovery. 1621 A lookup would operate via a direct query to a known service, while 1622 discovery may use multicast messages or a service where applications 1623 register in order to be found. 1625 It is highly desirable that the homenet name service must at the very 1626 least co-exist with the Internet name service. There should also be 1627 a bias towards proven, existing solutions. The strong implication is 1628 thus that the homenet service is DNS-based, or DNS-compatible. There 1629 are naming protocols that are designed to be configured and operate 1630 Internet-wide, like unicast-based DNS, but also protocols that are 1631 designed for zero-configuration local environments, like mDNS 1632 [RFC6762]. 1634 When DNS is used as the homenet name service, it typically includes 1635 both a resolving service and an authoritative service. The 1636 authoritative service hosts the homenet related zone. One approach 1637 when provisioning such a name service, which is designed to 1638 facilitate name resolution from the global Internet, is to run an 1639 authoritative name service on the CER and a secondary authoritative 1640 name service provided by the ISP or perhaps an external third party. 1642 Where zero configuration name services are used, it is desirable that 1643 these can also coexist with the Internet name service. In 1644 particular, where the homenet is using a global name space, it is 1645 desirable that devices have the ability, where desired, to add 1646 entries to that name space. There should also be a mechanism for 1647 such entries to be removed or expired from the global name space. 1649 To protect against attacks such as cache poisoning, where an attacker 1650 is able to insert a bogus DNS entry in the local cache, it is 1651 desirable to support appropriate name service security methods, 1652 including DNS Security Extensions (DNSSEC) [RFC4033], on both the 1653 authoritative server and the resolver sides. Where DNS is used, the 1654 homenet router or naming service must not prevent DNSSEC from 1655 operating. 1657 While this document does not specify hardware requirements, it is 1658 worth noting briefly here that e.g., in support of DNSSEC, 1659 appropriate homenet devices should have good random number generation 1660 capability, and future homenet specifications should indicate where 1661 high quality random number generators, i.e., with decent entropy, are 1662 needed. 1664 Finally, the impact of a change in CER must be considered. It would 1665 be desirable to retain any relevant state (configuration) that was 1666 held in the old CER. This might imply that state information should 1667 be distributed in the homenet, to be recoverable by/to the new CER, 1668 or to the homenet's ISP or a third party externally provided service 1669 by some means. 1671 3.7.4. Name spaces 1673 If access to homenet devices is required remotely from anywhere on 1674 the Internet, then at least one globally unique name space is 1675 required, though the use of multiple name spaces should not be 1676 precluded. One approach is that the name space(s) used for the 1677 homenet would be served authoritatively by the homenet, most likely 1678 by a server resident on the CER. Such name spaces may be acquired by 1679 the user or provided/generated by their ISP or an alternative 1680 externally provided service. It is likely that the default case is 1681 that a homenet will use a global domain provided by the ISP, but 1682 advanced users wishing to use a name space that is independent of 1683 their provider in the longer term should be able to acquire and use 1684 their own domain name. For users wanting to use their own 1685 independent domain names, such services are already available. 1687 Devices may also be assigned different names in different name 1688 spaces, e.g., by third parties who may manage systems or devices in 1689 the homenet on behalf of the resident(s). Remote management of the 1690 homenet is out of scope of this document. 1692 If however a global name space is not available, the homenet will 1693 need to pick and use a local name space which would only have meaning 1694 within the local homenet (i.e., it would not be used for remote 1695 access to the homenet). The .local name space currently has a 1696 special meaning for certain existing protocols which have link-local 1697 scope, and is thus not appropriate for multi-subnet home networks. A 1698 different name space is thus required for the homenet. 1700 One approach for picking a local name space is to use an Ambiguous 1701 Local Qualified Domain Name (ALQDN) space, such as .sitelocal (or an 1702 appropriate name reserved for the purpose). While this is a simple 1703 approach, there is the potential in principle for devices that are 1704 bookmarked somehow by name by an application in one homenet to be 1705 confused with a device with the same name in another homenet. In 1706 practice however the underlying service discovery protocols should be 1707 capable of handling moving to a network where a new device is using 1708 the same name as a device used previously in another homenet. 1710 An alternative approach for a local name space would be to use a 1711 Unique Locally Qualified Domain Name (ULQDN) space such as 1712 ..sitelocal. The could be generated in 1713 a variety of ways, one potentially being based on the local /48 ULA 1714 prefix being used across the homenet. Such a should 1715 survive a cold restart, i.e., be consistent after a network power- 1716 down, or, if a value is not set on startup, the CER or device running 1717 the name service should generate a default value. It would be 1718 desirable for the homenet user to be able to override the 1719 with a value of their choice, but that would increase 1720 the likelihood of a name conflict. Any generated 1721 should not be predictable; thus adding a salt/hash function would be 1722 desirable. 1724 In the (likely) event that the homenet is accessible from outside the 1725 homenet (using the global name space), it is vital that the homenet 1726 name space follow the rules and conventions of the global name space. 1727 In this mode of operation, names in the homenet (including those 1728 automatically generated by devices) must be usable as labels in the 1729 global name space. [RFC5890] describes considerations for 1730 Internationalizing Domain Names in Applications (IDNA). 1732 Also, with the introduction of new 'dotless' top level domains, there 1733 is also potential for ambiguity between, for example, a local host 1734 called 'computer' and (if it is registered) a .computer gTLD. Thus 1735 qualified names should always be used, whether these are exposed to 1736 the user or not. The IAB has issued a statement which explains why 1737 dotless domains should be considered harmful [IABdotless]. 1739 There may be use cases where either different name spaces may be 1740 desired for different realms in the homenet, or for segmentation of a 1741 single name space within the homenet. Thus hierarchical name space 1742 management is likely to be required. There should also be nothing to 1743 prevent individual device(s) being independently registered in 1744 external name spaces. 1746 It may be the case that if there are two or more CERs serving the 1747 home network, that if each has name space delegated from a different 1748 ISP there is the potential for devices in the home to have multiple 1749 fully qualified names under multiple domains. 1751 Where a user is in a remote network wishing to access devices in 1752 their home network, there may be a requirement to consider the domain 1753 search order presented where multiple associated name spaces exist. 1754 This also implies that a domain discovery function is desirable. 1756 It may be the case that not all devices in the homenet are made 1757 available by name via an Internet name space, and that a 'split view' 1758 (as described in [RFC6950] Section 4) is preferred for certain 1759 devices, whereby devices inside the homenet see different DNS 1760 responses to those outside. 1762 Finally, this document makes no assumption about the presence or 1763 omission of a reverse lookup service. There is an argument that it 1764 may be useful for presenting logging information to users with 1765 meaningful device names rather than literal addresses. There are 1766 also some services, most notably email mail exchangers, where some 1767 operators have chosen to require a valid reverse lookup before 1768 accepting connections. 1770 3.7.5. Independent operation 1772 Name resolution and service discovery for reachable devices must 1773 continue to function if the local network is disconnected from the 1774 global Internet, e.g., a local media server should still be available 1775 even if the Internet link is down for an extended period. This 1776 implies the local network should also be able to perform a complete 1777 restart in the absence of external connectivity, and have local 1778 naming and service discovery operate correctly. 1780 The approach described above of a local authoritative name service 1781 with a cache would allow local operation for sustained ISP outages. 1783 Having an independent local trust anchor is desirable, to support 1784 secure exchanges should external connectivity be unavailable. 1786 A change in ISP should not affect local naming and service discovery. 1787 However, if the homenet uses a global name space provided by the ISP, 1788 then this will obviously have an impact if the user changes their 1789 network provider. 1791 3.7.6. Considerations for LLNs 1793 In some parts of the homenet, in particular LLNs or any devices where 1794 battery power is used, devices may be sleeping, in which case a proxy 1795 for such nodes may be required, that could respond (for example) to 1796 multicast service discovery requests. Those same devices or parts of 1797 the network may have less capacity for multicast traffic that may be 1798 flooded from other parts of the network. In general, message 1799 utilisation should be efficient considering the network technologies 1800 and constrained devices that the service may need to operate over. 1802 There are efforts underway to determine naming and discovery 1803 solutions for use by the Constrained Application Protocol (CoAP) 1804 [I-D.ietf-core-coap] in LLN networks. These are outside the scope of 1805 this document. 1807 3.7.7. DNS resolver discovery 1809 Automatic discovery of a name service to allow client devices in the 1810 homenet to resolve external domains on the Internet is required, and 1811 such discovery must support clients that may be a number of router 1812 hops away from the name service. Similarly it may be desirable to 1813 convey any DNS domain search list that may be in effect for the 1814 homenet. 1816 3.7.8. Devices roaming to/from the homenet 1818 It is likely that some devices which have registered names within the 1819 homenet Internet name space and that are mobile will attach to the 1820 Internet at other locations and acquire an IP address at those 1821 locations. Devices may move between different homenets. In such 1822 cases it is desirable that devices may be accessed by the same name 1823 as is used in their home network. 1825 Solutions to this problem are not discussed in this document. They 1826 may include use of Mobile IPv6 or Dynamic DNS, either of which would 1827 put additional requirements on to the homenet, or establishment of a 1828 (VPN) tunnel to a server in the home network. 1830 3.8. Other Considerations 1832 This section discusses two other considerations for home networking 1833 that the architecture should not preclude, but that this text is 1834 neutral towards. 1836 3.8.1. Quality of Service 1838 Support for Quality of Service in a multi-service homenet may be a 1839 requirement, e.g., for a critical system (perhaps healthcare 1840 related), or for differentiation between different types of traffic 1841 (file sharing, cloud storage, live streaming, VoIP, etc). Different 1842 link media types may have different such properties or capabilities. 1844 However, homenet scenarios should require no new Quality of Service 1845 protocols. A DiffServ [RFC2475] approach with a small number of 1846 predefined traffic classes may generally be sufficient, though at 1847 present there is little experience of Quality of Service deployment 1848 in home networks. It is likely that QoS, or traffic prioritisation, 1849 methods will be required at the CER, and potentially around 1850 boundaries between different link media types (where for example some 1851 traffic may simply not be appropriate for some media, and need to be 1852 dropped to avoid overloading the constrained media). 1854 There may also be complementary mechanisms that could be beneficial 1855 to application performance and behaviour in the homenet domain, such 1856 as ensuring proper buffering algorithms are used as described in 1857 [Gettys11]. 1859 3.8.2. Operations and Management 1861 In this section we briefly review some initial considerations for 1862 operations and management in the type of homenet described in this 1863 document. It is expected that a separate document will define an 1864 appropriate operations and management framework for such homenets. 1866 As described in this document, the homenet should have the general 1867 goal of being self-organising and configuring from the network layer 1868 perspective, e.g. prefixes should be able to be assigned to router 1869 interfaces. Further, applications running on devices should be able 1870 to use zero configuration service discovery protocols to discover 1871 services of interest to the home user. In contrast, a home user 1872 would not be expected, for example, to have to assign prefixes to 1873 links, or manage the DNS entries for the home network. Such expert 1874 operation should not be precluded, but it is not the norm. 1876 The user may still be required to, or wish to, perform some 1877 configuration of the network and the devices on it. Examples might 1878 include entering a security key to enable access to their wireless 1879 network, or choosing to give a 'friendly name' to a device presented 1880 to them through service discovery. Configuration of link layer and 1881 application layer services is out of scope of this architectural 1882 principles document, but are likely to be required in an operational 1883 homenet. 1885 While not being expected to actively configure the networking 1886 elements of their homenet, users may be interested in being able to 1887 view the status of their networks and the devices connected to it, in 1888 which case appropriate network monitoring protocols will be required 1889 to allow them to view their network, and its status, e.g. via a web 1890 interface or equivalent. While the user may not understand how the 1891 network operates, it is reasonable to assume they are interested in 1892 understanding what faults or problems may exist on it. Such 1893 monitoring may extend to other devices on the network, e.g. storage 1894 devices, or web cameras, but such devices are beyond the scope of 1895 this document. 1897 It may also be the case that an ISP, or a third party, might wish to 1898 offer a remote management service for the homenet on behalf of the 1899 user, or to be able to assist the user in event of some problem they 1900 are experiencing, in which case appropriate management and monitoring 1901 protocols would be required. 1903 Specifying the required protocols to facilitate homenet management 1904 and monitoring is out of scope of this document. As stated above, it 1905 is expected that a separate document will be produced to describe the 1906 operations and management framework for the types of home network 1907 presented in this document. 1909 As a final point, we note that it is desirable that all network 1910 management and monitoring functions should be available over IPv6 1911 transport, even where the homenet is dual-stack. 1913 3.9. Implementing the Architecture on IPv6 1915 This architecture text encourages re-use of existing protocols. Thus 1916 the necessary mechanisms are largely already part of the IPv6 1917 protocol set and common implementations, though there are some 1918 exceptions. 1920 For automatic routing, it is expected that solutions can be found 1921 based on existing protocols. Some relatively smaller updates are 1922 likely to be required, e.g., a new mechanism may be needed in order 1923 to turn a selected protocol on by default, a mechanism may be 1924 required to automatically assign prefixes to links within the 1925 homenet. 1927 Some functionality, if required by the architecture, may need more 1928 significant changes or require development of new protocols, e.g., 1929 support for multihoming with multiple exit routers would likely 1930 require extensions to support source and destination address based 1931 routing within the homenet. 1933 Some protocol changes are however required in the architecture, e.g., 1934 for name resolution and service discovery, extensions to existing 1935 zero configuration link-local name resolution protocols are needed to 1936 enable them to work across subnets, within the scope of the home 1937 network site. 1939 Some of the hardest problems in developing solutions for home 1940 networking IPv6 architectures include discovering the right borders 1941 where the 'home' domain ends and the service provider domain begins, 1942 deciding whether some of the necessary discovery mechanism extensions 1943 should affect only the network infrastructure or also hosts, and the 1944 ability to turn on routing, prefix delegation and other functions in 1945 a backwards compatible manner. 1947 4. Conclusions 1949 This text defines principles and requirements for a homenet 1950 architecture. The principles and requirements documented here should 1951 be observed by any future texts describing homenet protocols for 1952 routing, prefix management, security, naming or service discovery. 1954 5. Security Considerations 1956 Security considerations for the homenet architecture are discussed in 1957 Section 3.6 above. 1959 6. IANA Considerations 1961 This document has no actions for IANA. 1963 7. References 1965 7.1. Normative References 1967 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 1968 (IPv6) Specification", RFC 2460, December 1998. 1970 [RFC3633] Troan, O. and R. Droms, "IPv6 Prefix Options for Dynamic 1971 Host Configuration Protocol (DHCP) version 6", RFC 3633, 1972 December 2003. 1974 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 1975 Addresses", RFC 4193, October 2005. 1977 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 1978 Architecture", RFC 4291, February 2006. 1980 7.2. Informative References 1982 [RFC1918] Rekhter, Y., Moskowitz, R., Karrenberg, D., Groot, G., and 1983 E. Lear, "Address Allocation for Private Internets", 1984 BCP 5, RFC 1918, February 1996. 1986 [RFC2475] Blake, S., Black, D., Carlson, M., Davies, E., Wang, Z., 1987 and W. Weiss, "An Architecture for Differentiated 1988 Services", RFC 2475, December 1998. 1990 [RFC2775] Carpenter, B., "Internet Transparency", RFC 2775, 1991 February 2000. 1993 [RFC2827] Ferguson, P. and D. Senie, "Network Ingress Filtering: 1994 Defeating Denial of Service Attacks which employ IP Source 1995 Address Spoofing", BCP 38, RFC 2827, May 2000. 1997 [RFC3002] Mitzel, D., "Overview of 2000 IAB Wireless Internetworking 1998 Workshop", RFC 3002, December 2000. 2000 [RFC3022] Srisuresh, P. and K. Egevang, "Traditional IP Network 2001 Address Translator (Traditional NAT)", RFC 3022, 2002 January 2001. 2004 [RFC4033] Arends, R., Austein, R., Larson, M., Massey, D., and S. 2005 Rose, "DNS Security Introduction and Requirements", 2006 RFC 4033, March 2005. 2008 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 2009 More-Specific Routes", RFC 4191, November 2005. 2011 [RFC4192] Baker, F., Lear, E., and R. Droms, "Procedures for 2012 Renumbering an IPv6 Network without a Flag Day", RFC 4192, 2013 September 2005. 2015 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 2016 Address Autoconfiguration", RFC 4862, September 2007. 2018 [RFC4864] Van de Velde, G., Hain, T., Droms, R., Carpenter, B., and 2019 E. Klein, "Local Network Protection for IPv6", RFC 4864, 2020 May 2007. 2022 [RFC4941] Narten, T., Draves, R., and S. Krishnan, "Privacy 2023 Extensions for Stateless Address Autoconfiguration in 2024 IPv6", RFC 4941, September 2007. 2026 [RFC5533] Nordmark, E. and M. Bagnulo, "Shim6: Level 3 Multihoming 2027 Shim Protocol for IPv6", RFC 5533, June 2009. 2029 [RFC5890] Klensin, J., "Internationalized Domain Names for 2030 Applications (IDNA): Definitions and Document Framework", 2031 RFC 5890, August 2010. 2033 [RFC5969] Townsley, W. and O. Troan, "IPv6 Rapid Deployment on IPv4 2034 Infrastructures (6rd) -- Protocol Specification", 2035 RFC 5969, August 2010. 2037 [RFC6092] Woodyatt, J., "Recommended Simple Security Capabilities in 2038 Customer Premises Equipment (CPE) for Providing 2039 Residential IPv6 Internet Service", RFC 6092, 2040 January 2011. 2042 [RFC6144] Baker, F., Li, X., Bao, C., and K. Yin, "Framework for 2043 IPv4/IPv6 Translation", RFC 6144, April 2011. 2045 [RFC6145] Li, X., Bao, C., and F. Baker, "IP/ICMP Translation 2046 Algorithm", RFC 6145, April 2011. 2048 [RFC6177] Narten, T., Huston, G., and L. Roberts, "IPv6 Address 2049 Assignment to End Sites", BCP 157, RFC 6177, March 2011. 2051 [RFC6204] Singh, H., Beebee, W., Donley, C., Stark, B., and O. 2052 Troan, "Basic Requirements for IPv6 Customer Edge 2053 Routers", RFC 6204, April 2011. 2055 [RFC6296] Wasserman, M. and F. Baker, "IPv6-to-IPv6 Network Prefix 2056 Translation", RFC 6296, June 2011. 2058 [RFC6333] Durand, A., Droms, R., Woodyatt, J., and Y. Lee, "Dual- 2059 Stack Lite Broadband Deployments Following IPv4 2060 Exhaustion", RFC 6333, August 2011. 2062 [RFC6555] Wing, D. and A. Yourtchenko, "Happy Eyeballs: Success with 2063 Dual-Stack Hosts", RFC 6555, April 2012. 2065 [RFC6724] Thaler, D., Draves, R., Matsumoto, A., and T. Chown, 2066 "Default Address Selection for Internet Protocol Version 6 2067 (IPv6)", RFC 6724, September 2012. 2069 [RFC6762] Cheshire, S. and M. Krochmal, "Multicast DNS", RFC 6762, 2070 February 2013. 2072 [RFC6824] Ford, A., Raiciu, C., Handley, M., and O. Bonaventure, 2073 "TCP Extensions for Multipath Operation with Multiple 2074 Addresses", RFC 6824, January 2013. 2076 [RFC6887] Wing, D., Cheshire, S., Boucadair, M., Penno, R., and P. 2077 Selkirk, "Port Control Protocol (PCP)", RFC 6887, 2078 April 2013. 2080 [RFC6950] Peterson, J., Kolkman, O., Tschofenig, H., and B. Aboba, 2081 "Architectural Considerations on Application Features in 2082 the DNS", RFC 6950, October 2013. 2084 [RFC7084] Singh, H., Beebee, W., Donley, C., and B. Stark, "Basic 2085 Requirements for IPv6 Customer Edge Routers", RFC 7084, 2086 November 2013. 2088 [RFC7157] Troan, O., Miles, D., Matsushima, S., Okimoto, T., and D. 2089 Wing, "IPv6 Multihoming without Network Address 2090 Translation", RFC 7157, March 2014. 2092 [I-D.ietf-core-coap] 2093 Shelby, Z., Hartke, K., and C. Bormann, "Constrained 2094 Application Protocol (CoAP)", draft-ietf-core-coap-18 2095 (work in progress), June 2013. 2097 [IABdotless] 2098 "IAB Statement: Dotless Domains Considered Harmful", 2099 February 2013, . 2103 [Gettys11] 2104 Gettys, J., "Bufferbloat: Dark Buffers in the Internet", 2105 March 2011, 2106 . 2108 Appendix A. Acknowledgments 2110 The authors would like to thank Aamer Akhter, Mikael Abrahamsson, 2111 Mark Andrews, Dmitry Anipko, Ran Atkinson, Fred Baker, Ray Bellis, 2112 Teco Boot, John Brzozowski, Cameron Byrne, Brian Carpenter, Stuart 2113 Cheshire, Julius Chroboczek, Lorenzo Colitti, Robert Cragie, Elwyn 2114 Davies, Ralph Droms, Lars Eggert, Jim Gettys, Olafur Gudmundsson, 2115 Wassim Haddad, Joel M. Halpern, David Harrington, Lee Howard, Ray 2116 Hunter, Joel Jaeggli, Heather Kirksey, Ted Lemon, Acee Lindem, Kerry 2117 Lynn, Daniel Migault, Erik Nordmark, Michael Richardson, Mattia 2118 Rossi, Barbara Stark, Markus Stenberg, Sander Steffann, Don Sturek, 2119 Andrew Sullivan, Dave Taht, Dave Thaler, Michael Thomas, Mark 2120 Townsley, JP Vasseur, Curtis Villamizar, Dan Wing, Russ White, and 2121 James Woodyatt for their comments and contributions within homenet WG 2122 meetings and on the WG mailing list. An acknowledgement generally 2123 means that person's text made it in to the document, or was helpful 2124 in clarifying or reinforcing an aspect of the document. It does not 2125 imply that each contributor agrees with every point in the document. 2127 Appendix B. Changes 2129 This section will be removed in the final version of the text. 2131 B.1. Version 15 2133 Changes made include: 2135 o Removed spurious paragraph, and spurious sentence. 2137 B.2. Version 14 2139 Changes made include: 2141 o Changes for Adrian Farrell discuss/comment. 2143 o Very minor wordsmithing requested by Benoit for OAM text. 2145 o Very minor wordsmithing from IETF89 session. 2147 o Added note to support SSM. 2149 o Emphasised at most one routing protocol in use, possibly none. 2151 B.3. Version 13 2153 Changes made include: 2155 o Changes to address last outstanding IESG DISCUSSes/COMMENTs. 2157 B.4. Version 12 2159 Changes made include: 2161 o Fixed minor typo nits introduced in -11. 2163 o Elwyn Davies' gen-art review comments addressed. 2165 o Some further IESG DISCUSSes/COMMENTs addressed. 2167 B.5. Version 11 (after IESG review) 2169 Changes made include: 2171 o Jouni Korhonen's OPSDIR review comments addressed. 2173 o Elwyn Davies' gen-art review comments addressed. 2175 o Considered secdir review by Samiel Weiler; many points addressed. 2177 o Considered APPSDIR review. 2179 o Addressed a large number of IESG comments and discusses. 2181 B.6. Version 10 (after AD review) 2183 Changes made include: 2185 o Minor changes/clarifications resulting from AD review 2187 B.7. Version 09 (after WGLC) 2189 Changes made include: 2191 o Added note about multicast into or out of site 2193 o Removed further personal draft references, replaced with covering 2194 text 2196 o Routing functionality text updated to avoid ambiguity 2198 o Added note that devices away from homenet may tunnel home (via 2199 VPN) 2201 o Added note that homenets more exposed to provider renumbering than 2202 with IPv4 and NAT 2204 o Added note about devices that may be ULA-only until configured to 2205 be globally addressable 2207 o Removed paragraph about broken CERs that do not work with prefixes 2208 other than /64 2210 o Noted no recommendation on methods to convey prefix information is 2211 made in this text 2213 o Stated that this text does not recommend how to form largest 2214 possible subnets 2216 o Added text about homenet evolution and handling disparate media 2217 types 2219 o Rephrased NAT/firewall text on marginal effectiveness 2221 o Emphasised that multihoming may be to any number of ISPs 2223 B.8. Version 08 2225 Changes made include: 2227 o Various clarifications made in response to list comments 2229 o Added note on ULAs with IPv4, where no GUAs in use 2231 o Added note on naming and internationalisation (IDNA) 2233 o Added note on trust relationships when adding devices 2235 o Added note for MPTCP 2237 o Added various naming and SD notes 2239 o Added various notes on delegated ISP prefixes 2241 B.9. Version 07 2243 Changes made include: 2245 o Removed reference to NPTv6 in section 3.2.4. Instead now say it 2246 has an architectural cost to use in the earlier section, and thus 2247 it is not recommended for use in the homenet architecture. 2249 o Removed 'proxy or extend?' section. Included shorter text in main 2250 body, without mandating either approach for service discovery. 2252 o Made it clearer that ULAs are expected to be used alongside 2253 globals. 2255 o Removed reference to 'advanced security' as described in 2256 draft-vyncke-advanced-ipv6-security. 2258 o Balanced the text between ULQDN and ALQDN. 2260 o Clarify text does not assume default deny or allow on CER, but 2261 that either mode may be enabled. 2263 o Removed ULA-C reference for 'simple' addresses. Instead only 2264 suggested service discovery to find such devices. 2266 o Reiterated that single/multiple CER models to be supported for 2267 multihoming. 2269 o Reordered section 3.3 to improve flow. 2271 o Added recommendation that homenet is not allocated less than /60, 2272 and a /56 is preferable. 2274 o Tidied up first few intro sections. 2276 o Other minor edits from list feedback. 2278 B.10. Version 06 2280 Changes made include: 2282 o Stated that unmanaged goal is 'as far as possible'. 2284 o Added note about multiple /48 ULAs potentially being in use. 2286 o Minor edits from list feedback. 2288 B.11. Version 05 2290 Changes made include: 2292 o Some significant changes to naming and SD section. 2294 o Removed some expired drafts. 2296 o Added notes about issues caused by ISP only delegating a /64. 2298 o Recommended against using prefixes longer than /64. 2300 o Suggested CER asks for /48 by DHCP PD, even if it only receives 2301 less. 2303 o Added note about DS-Lite but emphasised transition is out of 2304 scope. 2306 o Added text about multicast routing. 2308 B.12. Version 04 2310 Changes made include: 2312 o Moved border section from IPv6 differences to principles section. 2314 o Restructured principles into areas. 2316 o Added summary of naming and service discovery discussion from WG 2317 list. 2319 B.13. Version 03 2321 Changes made include: 2323 o Various improvements to the readability. 2325 o Removed bullet lists of requirements, as requested by chair. 2327 o Noted 6204bis has replaced advanced-cpe draft. 2329 o Clarified the topology examples are just that. 2331 o Emphasised we are not targetting walled gardens, but they should 2332 not be precluded. 2334 o Also changed text about requiring support for walled gardens. 2336 o Noted that avoiding falling foul of ingress filtering when 2337 multihomed is desirable. 2339 o Improved text about realms, detecting borders and policies at 2340 borders. 2342 o Stated this text makes no recommendation about default security 2343 model. 2345 o Added some text about failure modes for users plugging things 2346 arbitrarily. 2348 o Expanded naming and service discovery text. 2350 o Added more text about ULAs. 2352 o Removed reference to version 1 on chair feedback. 2354 o Stated that NPTv6 adds architectural cost but is not a homenet 2355 matter if deployed at the CER. This text only considers the 2356 internal homenet. 2358 o Noted multihoming is supported. 2360 o Noted routers may not by separate devices, they may be embedded in 2361 devices. 2363 o Clarified simple and advanced security some more, and RFC 4864 and 2364 6092. 2366 o Stated that there should be just one secret key, if any are used 2367 at all. 2369 o For multihoming, support multiple CERs but note that routing to 2370 the correct CER to avoid ISP filtering may not be optimal within 2371 the homenet. 2373 o Added some ISPs renumber due to privacy laws. 2375 o Removed extra repeated references to Simple Security. 2377 o Removed some solution creep on RIOs/RAs. 2379 o Load-balancing scenario added as to be supported. 2381 B.14. Version 02 2383 Changes made include: 2385 o Made the IPv6 implications section briefer. 2387 o Changed Network Models section to describe properties of the 2388 homenet with illustrative examples, rather than implying the 2389 number of models was fixed to the six shown in 01. 2391 o Text to state multihoming support focused on single CER model. 2392 Multiple CER support is desirable, but not required. 2394 o Stated that NPTv6 not supported. 2396 o Added considerations section for operations and management. 2398 o Added bullet point principles/requirements to Section 3.4. 2400 o Changed IPv6 solutions must not adversely affect IPv4 to should 2401 not. 2403 o End-to-end section expanded to talk about "Simple Security" and 2404 borders. 2406 o Extended text on naming and service discovery. 2408 o Added reference to RFC 2775, RFC 6177. 2410 o Added reference to the new xmDNS draft. 2412 o Added naming/SD requirements from Ralph Droms. 2414 Authors' Addresses 2416 Tim Chown (editor) 2417 University of Southampton 2418 Highfield 2419 Southampton, Hampshire SO17 1BJ 2420 United Kingdom 2422 Email: tjc@ecs.soton.ac.uk 2424 Jari Arkko 2425 Ericsson 2426 Jorvas 02420 2427 Finland 2429 Email: jari.arkko@piuha.net 2431 Anders Brandt 2432 Sigma Designs 2433 Emdrupvej 26A, 1 2434 Copenhagen DK-2100 2435 Denmark 2437 Email: abr@sdesigns.dk 2438 Ole Troan 2439 Cisco Systems, Inc. 2440 Drammensveien 145A 2441 Oslo N-0212 2442 Norway 2444 Email: ot@cisco.com 2446 Jason Weil 2447 Time Warner Cable 2448 13820 Sunrise Valley Drive 2449 Herndon, VA 20171 2450 USA 2452 Email: jason.weil@twcable.com