idnits 2.17.1 draft-ietf-dhc-v6only-08.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 (Using the creation date from RFC2563, updated by this document, for RFC5378 checks: 1998-08-14) -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (August 13, 2020) is 1345 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) No issues found here. Summary: 0 errors (**), 0 flaws (~~), 1 warning (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Dynamic Host Configuration L. Colitti 3 Internet-Draft J. Linkova 4 Updates: 2563 (if approved) Google 5 Intended status: Standards Track M. Richardson 6 Expires: February 14, 2021 Sandelman 7 T. Mrugalski 8 ISC 9 August 13, 2020 11 IPv6-Only-Preferred Option for DHCPv4 12 draft-ietf-dhc-v6only-08 14 Abstract 16 This document specifies a DHCPv4 option to indicate that a host 17 supports an IPv6-only mode and is willing to forgo obtaining an IPv4 18 address if the network provides IPv6 connectivity. It also updates 19 RFC2563 to specify the DHCPv4 server behavior when the server 20 receives a DHCPDISCOVER not containing the Auto-Configure option but 21 containing the new option defined in this document. 23 Status of This Memo 25 This Internet-Draft is submitted in full conformance with the 26 provisions of BCP 78 and BCP 79. 28 Internet-Drafts are working documents of the Internet Engineering 29 Task Force (IETF). Note that other groups may also distribute 30 working documents as Internet-Drafts. The list of current Internet- 31 Drafts is at https://datatracker.ietf.org/drafts/current/. 33 Internet-Drafts are draft documents valid for a maximum of six months 34 and may be updated, replaced, or obsoleted by other documents at any 35 time. It is inappropriate to use Internet-Drafts as reference 36 material or to cite them other than as "work in progress." 38 This Internet-Draft will expire on February 14, 2021. 40 Copyright Notice 42 Copyright (c) 2020 IETF Trust and the persons identified as the 43 document authors. All rights reserved. 45 This document is subject to BCP 78 and the IETF Trust's Legal 46 Provisions Relating to IETF Documents 47 (https://trustee.ietf.org/license-info) in effect on the date of 48 publication of this document. Please review these documents 49 carefully, as they describe your rights and restrictions with respect 50 to this document. Code Components extracted from this document must 51 include Simplified BSD License text as described in Section 4.e of 52 the Trust Legal Provisions and are provided without warranty as 53 described in the Simplified BSD License. 55 Table of Contents 57 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 58 1.1. Requirements Language . . . . . . . . . . . . . . . . . . 3 59 1.2. Terminology . . . . . . . . . . . . . . . . . . . . . . . 4 60 2. Reasons to Signal IPv6-Only Support in DHCPv4 Packets . . . . 5 61 3. IPv6-Only Preferred Option . . . . . . . . . . . . . . . . . 5 62 3.1. Option format . . . . . . . . . . . . . . . . . . . . . . 6 63 3.2. DHCPv4 Client Behavior . . . . . . . . . . . . . . . . . 6 64 3.3. DHCPv4 Server Behavior . . . . . . . . . . . . . . . . . 8 65 3.3.1. Interaction with RFC2563 . . . . . . . . . . . . . . 9 66 3.4. Constants and Configuration Variables . . . . . . . . . . 10 67 4. IPv6-Only Transition Technologies Considerations . . . . . . 10 68 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 11 69 6. Security Considerations . . . . . . . . . . . . . . . . . . . 12 70 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 12 71 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 12 72 8.1. Normative References . . . . . . . . . . . . . . . . . . 12 73 8.2. Informative References . . . . . . . . . . . . . . . . . 13 74 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 14 76 1. Introduction 78 One of the biggest challenges of deploying IPv6-only LANs is that 79 such networks might contain rather heterogeneous collection of hosts. 80 While some hosts are capable of operating in IPv6-only mode (either 81 because the OS and all applications are IPv6-only capable or because 82 the host has some form of 464XLAT [RFC6877] deployed), others might 83 still have IPv4 dependencies and need IPv4 addresses to operate 84 properly. To incrementally rollout IPv6-only, network operators 85 might need to provide IPv4 on demand whereby a host receives an IPv4 86 address if it needs it, while IPv6-only capable hosts (such as modern 87 mobile devices) are not allocated IPv4 addresses. Traditionally that 88 goal is achieved by placing IPv6-only capable devices into a 89 dedicated IPv6-only network segment or WiFi SSID, while dual-stack 90 devices reside in another network with IPv4 and DHCPv4 enabled. 91 However such an approach has a number of drawbacks, including but not 92 limited to: 94 o Doubling the number of network segments leads to operational 95 complexity and performance impact, for instance due to high memory 96 utilization caused by an increased number of ACL entries. 98 o Placing a host into the correct network segment is problematic. 99 For example, in the case of 802.11 Wi-Fi the user might select the 100 wrong SSID. In the case of wired 802.1x authentication the 101 authentication server might not have all the information required 102 to make the correct decision and choose between an IPv6-only and a 103 dual-stack VLAN. 105 It would be beneficial for IPv6 deployment if operators could 106 implement IPv6-mostly (or IPv4-on-demand) segments where IPv6-only 107 hosts co-exist with legacy dual-stack devices. The trivial solution 108 of disabling IPv4 stack on IPv6-only capable hosts is not feasible as 109 those clients must be able to operate on IPv4-only networks as well. 110 While IPv6-only capable devices might use a heuristic approach to 111 learning if the network provides IPv6-only functionality and stop 112 using IPv4 if it does, such an approach might be practically 113 undesirable. One important reason is that when a host connects to a 114 network, it does not know if the network is IPv4-only, dual-stack or 115 IPv6-only. To ensure that the connectivity over whatever protocol is 116 present becomes available as soon as possible the host usually starts 117 configuring both IPv4 and IPv6 immediately. If hosts were to delay 118 requesting IPv4 until IPv6 reachability is confirmed, that would 119 penalize IPv4-only and dual-stack networks, which does not seem 120 practical. Requesting IPv4 and then releasing it later, after IPv6 121 reachability is confirmed, might cause user-visible errors as it 122 would be disruptive for applications which have started using the 123 assigned IPv4 address already. Instead it would be useful to have a 124 mechanism which would allow a host to indicate that its request for 125 an IPv4 address is optional and a network to signal that IPv6-only 126 functionality (such as NAT64, [RFC6146]) is available. The proposed 127 solution is to introduce a new DHCPv4 option which a client uses to 128 indicate that it does not need an IPv4 address if the network 129 provides IPv6-only connectivity (as NAT64 and DNS64). If the 130 particular network segment provides IPv4-on-demand such clients would 131 not be supplied with IPv4 addresses, while on IPv4-only or dual-stack 132 segments without NAT64 services IPv4 addresses will be provided. 134 [RFC2563] introduces the Auto-Configure DHCPv4 option and describes 135 DHCPv4 servers behavior if no address is chosen for a host. This 136 document updates [RFC2563] to modify the server behavior if the 137 DHCPOFFER contains the IPv6-only Preferred option. 139 1.1. Requirements Language 141 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 142 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 143 "OPTIONAL" in this document are to be interpreted as described in BCP 144 14 [RFC2119] [RFC8174] when, and only when, they appear in all 145 capitals, as shown here. 147 1.2. Terminology 149 Dual-stack network or device: a network or device which has both 150 versions of the Internet Protocol (IPv4 and IPv6) enabled and 151 operational. 153 IPv6-only capable host: a host which does not require an IPv4 address 154 and can operate on IPv6-only networks. More precisely, IPv6-only 155 capability is specific to a given interface of the host: if some 156 application on a host require IPv4 and 464XLAT CLAT [RFC6877] is only 157 enabled on one interface, the host is IPv6-only capable if connected 158 to a NAT64 network via that interface. This document implies that 159 IPv6-only capable hosts reach IPv4-only destinations via a NAT64 160 service provided by the network. Section 4 discusses hypothetical 161 scenarios of other transition technologies being used. 163 IPv4-requiring host: a host which is not IPv6-only capable and can 164 not operate in an IPv6-only network providing NAT64 service. 166 IPv4-on-demand: a deployment scenario where end hosts are expected to 167 operate in IPv6-only mode by default and IPv4 addresses can be 168 assigned to some hosts if those hosts explicitly opt-in to receiving 169 IPv4 addresses. 171 IPv6-mostly network: a network which provides NAT64 (possibly with 172 DNS64) service as well as IPv4 connectivity and allows coexistence of 173 IPv6-only, dual-stack and IPv4-only hosts on the same segment. Such 174 deployment scenario allows operators to incrementally turn off IPv4 175 on end hosts, while still providing IPv4 to devices which require 176 IPv4 to operate. But, IPv6-only capable devices need not be assigned 177 IPv4 addresses. 179 IPv6-only mode: a mode of operation when a host acts as an IPv6-only 180 capable host and does not have IPv4 addresses assigned (except that 181 IPv4 link-local addresses [RFC3927] may have been configured). 183 IPv6-only network: a network which does not provide routing 184 functionality for IPv4 packets. Such networks may or may not allow 185 intra-LAN IPv4 connectivity. IPv6-only network usually provides 186 access to IPv4-only resources via NAT64 [RFC6146]. 188 NAT64: Network Address and Protocol Translation from IPv6 Clients to 189 IPv4 Servers [RFC6146]. 191 RA: Router Advertisement, a message used by IPv6 routers to advertise 192 their presence together with various link and Internet parameters 193 [RFC4861]. 195 DNS64: a mechanism for synthesizing AAAA records from A records 196 [RFC6147]. 198 Network attachment event: A Link Up event, as described by [RFC4957] 199 which results in a host detecting an available network. 201 Disabling IPv4 stack on the host interface: the host behavior when 202 the host: 204 o does not send any IPv4 packets from that interface, 206 o drops all IPv4 packets received on that interface and 208 o does not forward any IPv4 packets to that interface. 210 2. Reasons to Signal IPv6-Only Support in DHCPv4 Packets 212 For networks which contain a mix of both IPv6-only capable hosts and 213 IPv4-requiring hosts, and which utilize DHCPv4 for configuring the 214 IPv4 network stack on hosts, it seems natural to leverage the same 215 protocol to signal that IPv4 is discretional on a given segment. An 216 ability to remotely disable IPv4 on a host can be seen as a new 217 denial-of-service attack vector. The proposed approach limits the 218 attack surface to DHCPv4-related attacks without introducing new 219 vulnerable elements. 221 Another benefit of using DHCPv4 for signaling is that IPv4 will be 222 disabled only if both the client and the server indicate IPv6-only 223 capability. It allows IPv6-only capable hosts to turn off IPv4 only 224 upon receiving an explicit signal from the network and operate in 225 dual-stack or IPv4-only mode otherwise. In addition, the proposed 226 mechanism does not introduce any additional delays to the process of 227 configuring IP stack on hosts. If the network does not support IPv6- 228 only/IPv4-on-demand mode, an IPv6-only capable host would configure 229 an IPv4 address as quickly as on any other host. 231 Being a client/server protocol, DHCPv4 allows IPv4 to be selectively 232 disabled on a per-host basis on a given network segment. Coexistence 233 of IPv6-only, dual-stack and even IPv4-only hosts on the same LAN 234 would not only allow network administrators to preserve scarce IPv4 235 addresses but would also drastically simplify incremental deployment 236 of IPv6-only networks, positively impacting IPv6 adoption. 238 3. IPv6-Only Preferred Option 239 3.1. Option format 241 0 1 2 3 242 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 243 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 244 | Code | Length | Value | 245 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 246 | Value (contd) | 247 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 249 Figure 1: IPv6-Only Preferred Option Format 251 Fields: 253 Code: 8-bit identifier of the IPv6-Only Preferred option code as 254 assigned by IANA: TBD. 255 The client includes the Code in the Parameter Request List in 256 DHCPDISCOVER and DHCPREQUEST messages as described in 257 Section 3.2. 258 Length: 8-bit unsigned integer. The length of the option excluding 259 the Code and Length Fields. The server MUST set the length 260 field to 4. The client MUST ignore the IPv6-Only Preferred 261 option if the length field value is not 4. 262 Value: 32-bit unsigned integer. 263 The number of seconds the client should disable DHCPv4 for 264 (V6ONLY_WAIT configuration variable). 265 If the server pool is explicitly configured with a 266 V6ONLY_WAIT timer the server MUST set the field to that 267 configured value. Otherwise the server MUST set it to zero. 268 The client MUST process that field as described in 269 Section 3.2. 270 The client never sets this field as it never sends the full 271 option but includes the option code in the Parameter Request 272 List as described in Section 3.2. 274 3.2. DHCPv4 Client Behavior 276 A DHCPv4 client SHOULD allow a device administrator to configure 277 IPv6-only preferred mode either for a specific interface (to indicate 278 that the device is IPv6-only capable if connected to a NAT64 network 279 via that interface) or for all interfaces. If only a specific 280 interface is configured as IPv6-only capable the DHCPv4 client MUST 281 NOT consider the host to be an IPv6-only capable for the purpose of 282 sending/receiving DHCPv4 packets over any other interfaces. 284 The DHCPv4 client on an IPv4-requiring host MUST NOT include the 285 IPv6-only Preferred option in the Parameter Request List of any 286 DHCPv4 packets and MUST ignore that option in packets received from 287 DHCPv4 servers. 289 DHCPv4 clients running on IPv6-only capable hosts SHOULD include the 290 IPv6-only Preferred option code in the Parameter Request List in 291 DHCPDISCOVER and DHCPREQUEST messages for interfaces so enabled and 292 follow the processing as described below on a per enabled interface 293 basis. 295 If the client did not include the IPv6-only Preferred option code in 296 the Parameter Request List option in the DHCPDISCOVER or DHCPREQUEST 297 message it MUST ignore the IPv6-only Preferred option in any messages 298 received from the server. 300 If the client includes the IPv6-only Preferred option in the 301 Parameter Request List and the DHCPOFFER message from the server 302 contains a valid IPv6-only Preferred option, the client SHOULD NOT 303 request the IPv4 address provided in the DHCPOFFER. If the IPv6-only 304 Preferred option returned by the server contains a value greater or 305 equal to MIN_V6ONLY_WAIT, the client SHOULD set the V6ONLY_WAIT timer 306 to that value. Otherwise, the client SHOULD set the V6ONLY_WAIT 307 timer to MIN_V6ONLY_WAIT. The client SHOULD stop the DHCPv4 308 configuration process for V6ONLY_WAIT seconds or until a network 309 attachment event, whichever happens first. The host MAY disable the 310 IPv4 stack completely on the affected interface for V6ONLY_WAIT 311 seconds or until the network attachment event, whichever happens 312 first. 314 The IPv6-only Preferred option SHOULD be included in the Parameter 315 Request List option in DHCPREQUEST messages (after receiving a 316 DHCPOFFER without this option, for a INIT-REBOOT, or when renewing or 317 rebinding a leased address). If the DHCPv4 server responds with a 318 DHCPACK that includes the IPv6-only Preferred option, the client 319 behaviour depends on the client's state. If the client is in the 320 INIT-REBOOT state it SHOULD stop the DHCPv4 configuration process or 321 disable IPv4 stack completely for V6ONLY_WAIT seconds or until the 322 network event, whichever happens first. It also MAY send a 323 DHCPRELEASE message. If the client is in any other state it SHOULD 324 continue to use the assigned IPv4 address until further DHCPv4 325 reconfiguration events. 327 If the client includes the IPv6-only Preferred option in the 328 Parameter Request List and the server responds with DHCPOFFER message 329 without a valid IPv6-only Preferred option, the client MUST proceed 330 as normal with a DHCPREQUEST. 332 If the client waits for multiple DHCPOFFER responses and selects one 333 of them, it MUST follow the processing for the IPv6-only Preferred 334 option based on the selected response. A client MAY use the presence 335 of the IPv6-only Preferred option as a selection criteria. 337 When an IPv6-only capable client receives the IPv6-Only Preferred 338 option from the server, the client MAY configurean IPv4 link-local 339 address [RFC3927]. In that case IPv6-only capable devices might 340 still be able to communicate over IPv4 to other devices on the link. 341 The Auto-Configure Option [RFC2563] can be used to control IPv4 link- 342 local addresses autoconfiguration. Section 3.3.1 discusses the 343 interaction between the IPv6-only Preferred and the Auto-Configure 344 options. 346 3.3. DHCPv4 Server Behavior 348 The DHCPv4 server SHOULD be able to configure all or individual pools 349 to include the IPv6-only preferred option in DHCPv4 responses if the 350 client included the option code in the Parameter Request List option. 351 The DHCPv4 server MAY have a configuration option to specify the 352 V6ONLY_WAIT timer for all or individual IPv6-mostly pools. 354 The server MUST NOT include the IPv6-only Preferred option in the 355 DHCPOFFER or DHCPACK message if the YIADDR field in the message does 356 not belong to a pool configured as IPv6-mostly. The server MUST NOT 357 include the IPv6-only Preferred option in the DHCPOFFER or DHCPACK 358 message if the option was not present in the Parameter Request List 359 sent by the client. 361 If the IPv6-only Preferred option is present in the Parameter Request 362 List received from the client and the corresponding DHCPv4 pool is 363 explicitly configured as belonging to an IPv6-mostly network segment, 364 the server MUST include the IPv6-only Preferred option when 365 responding with the DHCPOFFER or DHCPACK message. If the server 366 responds with the IPv6-only Preferred option and the V6ONLY_WAIT 367 timer is configured for the pool, the server MUST copy the configured 368 value to the IPv6-only Preferred option value field. Otherwise it 369 MUST set the field to zero. The server SHOULD NOT assign an address 370 from the pool. Instead it SHOULD return 0.0.0.0 as the offered 371 address. Alternatively, if offering 0.0.0.0 is not feasible, for 372 example due to some limitations of the server or the network 373 infrastructure, the server MAY include an available IPv4 address from 374 the pool into the DHCPOFFER as per recommendations in [RFC2131]. In 375 this case, the offered address MUST be a valid address that is not 376 committed to any other client. Because the client is not expected 377 ever to request this address, the server SHOULD NOT reserve the 378 address and SHOULD NOT verify its uniqueness. If the client then 379 issues a DHCPREQUEST for the address, the server MUST process it per 381 [RFC2131], including replying with a DHCPACK for the address if in 382 the meantime it has not been committed to another client. 384 If a client includes both a Rapid-Commit option [RFC4039] and 385 IPv6-Only Preferred option in the DHCPDISCOVER message the server 386 SHOULD NOT honor the Rapid-Commit option if the response would 387 contain the IPv6-only Preferred option to the client. It SHOULD 388 instead respond with a DHCPOFFER as indicated above. 390 If the server receives a DHCPREQUEST containing the IPv6-only 391 Preferred option for the address from a pool configured as 392 IPv6-mostly, the server MUST process it per [RFC2131]. 394 3.3.1. Interaction with RFC2563 396 [RFC2563] defines an Auto-Configure DHCPv4 option to disable IPv4 397 link-local address configuration for IPv4 clients. Clients can 398 support both, neither or just one of IPv6-Only Preferred and Auto- 399 Configure options. If a client sends both IPv6-Only Preferred and 400 Auto-Configure options the network administrator can prevent the host 401 from configuring an IPv4 link-local address on an IPv6-mostly 402 network. To achieve this the server needs to send DHCPOFFER which 403 contains a 'yiaddr' of 0x00000000, and the Auto-Configure flag saying 404 "DoNotAutoConfigure". 406 However special care should be taken in a situation when a server 407 supports both options and receives just IPv6-Only Preferred option 408 from a client. Section 2.3 of [RFC2563] states that if no address is 409 chosen for the host (which would be the case for IPv6-only capable 410 clients on IPv6-mostly network) then: "If the DHCPDISCOVER does not 411 contain the Auto-Configure option, it is not answered." Such 412 behavior would be undesirable for clients supporting the IPv6-Only 413 Preferred option without supporting the Auto-Configure option as they 414 would not receive any response from the server and would keep asking, 415 instead of disabling DHCPv4 for V6ONLY_WAIT seconds. Therefore the 416 following update is made to Section 2.3 of [RFC2563]" 418 OLD TEXT: 420 --- 422 However, if no address is chosen for the host, a few additional steps 423 MUST be taken. 425 If the DHCPDISCOVER does not contain the Auto-Configure option, it is 426 not answered. 428 --- 429 NEW TEXT: 431 --- 433 However, if no address is chosen for the host, a few additional steps 434 MUST be taken. 436 If the DHCPDISCOVER does not contain the Auto-Configure option and 437 the IPv6-Only Preferred option is not present, it is not answered. 438 If the DHCPDISCOVER does not contain the Auto-Configure option but 439 contains the IPv6-Only Preferred option, the processing rules for the 440 IPv6-Only Preferred option apply. 442 --- 444 3.4. Constants and Configuration Variables 446 V6ONLY_WAIT The time for which the client SHOULD stop the DHCPv4 447 configuration process. The value MUST NOT be less 448 than MIN_V6ONLY_WAIT seconds. Default: 1800 seconds 449 MIN_V6ONLY_WAIT The lower boundary for V6ONLY_WAIT. Value: 300 450 seconds 452 4. IPv6-Only Transition Technologies Considerations 454 Until IPv6 adoption in the Internet reaches 100%, communication 455 between an IPv6-only host and IPv4-only destination requires some 456 form of transition mechanism deployed in the network. At the time of 457 writing, the only such mechanism that is widely supported by end 458 hosts is NAT64 [RFC6146] (either with or without 464XLAT). Therefore 459 the IPv6-only Preferred option is only sent by hosts capable of 460 operating on NAT64 networks. In a typical deployment scenario, a 461 network administrator would not configure the DHCPv4 server to return 462 the IPv6-only Preferred option unless the network provides NAT64 463 service. 465 Hypothetically, it is possible for multiple transition technologies 466 to coexist. In such scenario some form of negotiation would be 467 required between a client and a server to ensure that the transition 468 technology supported by the client is the one the network provides. 469 However it seems unlikely that any new transition technology would 470 arise and be widely adopted in any foreseeable future. Therefore 471 adding support for non-existing technologies seems to be suboptimal 472 and the proposed mechanism implies that NAT64 is used to facilitate 473 connectivity between IPv6 and IPv4. In the unlikely event that a new 474 transition mechanism becomes widely deployed, the applicability of 475 the IPv6-Only-Preferred option to that mechanism will depend on the 476 nature of the new mechanism. If the new mechanism is designed in 477 such a way that it's fully transparent for hosts that support NAT64 478 and the IPv6-Only-Preferred option, then the option can continue to 479 be used with the new mechanism. If the new mechanism is not 480 compatible with NAT64, and implementation on the host side is 481 required to support it, then a new DHCPv4 option needs to be defined. 483 It should be also noted that declaring a host (technically, a host 484 interface) IPv6-only capable is a policy decision. For example, 486 o An operating system vendor may make such decision and configure 487 their DHCPv4 clients to send the IPv6-Only Preferred option by 488 default if the OS has 464XLAT CLAT [RFC6877] enabled. 490 o An enterprise network administrator may provision the corporate 491 hosts as IPv6-only capable if all applications users are supposed 492 to run have been tested in an IPv6-only environment (or if 464XLAT 493 CLAT is enabled on the devices). 495 o IoT devices may be shipped in IPv6-only capable mode if they are 496 designed to connect to IPv6-enabled cloud destination only. 498 5. IANA Considerations 500 The IANA is requested to assign a new DHCPv4 Option code for the 501 IPv6-Only Preferred option from the BOOTP Vendor Extensions and 502 DHCPv4 Options registry, located at https://www.iana.org/assignments/ 503 bootp-dhcp-parameters/bootp-dhcp-parameters.xhtml#options . If 504 possible, please assign option code 108. 506 +-----------+-----------+--------+----------+-----------------------+ 507 | Tag | Name | Data | Meaning | Reference | 508 | | | Length | | | 509 +-----------+-----------+--------+----------+-----------------------+ 510 | TBD | IPv6-only | 4 | Number | draft-ietf-dhc-v6only | 511 | (proposed | Preferred | | of | | 512 | value: | option | | seconds | | 513 | 108) | | | to | | 514 | | | | disable | | 515 | | | | DHCPv4 | | 516 | | | | for | | 517 +-----------+-----------+--------+----------+-----------------------+ 519 Table 1 521 6. Security Considerations 523 An attacker might send a spoofed DHCPOFFER containing IPv6-only 524 Preferred option with the value field set to a large number, such as 525 0xffffffff, effectively disabling DHCPv4 on clients supporting the 526 option. If the network is IPv4-only such clients would lose 527 connectivity, while on a dual-stack network without NAT64 service 528 only connectivity to IPv4-only destinations would be affected. The 529 recovery would require triggering a network attachment event. 530 However it should be noted that if the network does not provide 531 protection from a rogue DHCPv4 server the similar attack vector can 532 be executed by offering an invalid address and setting the Lease Time 533 option value field to 0xffffffff. The latter attack would affect all 534 hosts, not just hosts that support the IPv6-only Preferred option. 535 Therefore the security measures against rogue DHCPv4 servers would be 536 sufficient to prevent the attacks specific to IPv6-only Preferred 537 option. Additionally such attacks can only be executed if the victim 538 prefers the rogue DHCPOFFER over the legitimate ones. Therefore for 539 the attack to be successful the attacker needs to know the selection 540 criteria used by the client and to be able to make its rogue offer 541 more preferable. 543 It should be noted that disabling IPv4 on a host upon receiving the 544 IPv6-only Preferred option from the DHCPv4 server protects the host 545 from IPv4-related attacks and therefore could be considered a 546 security feature as it reduces the attack surface. 548 7. Acknowledgements 550 Thanks to the following people (in alphabetical order) for their 551 review and feedback: Mohamed Boucadair, Martin Duke, Russ Housley, 552 Sheng Jiang, Benjamin Kaduk, Murray Kucherawy, Ted Lemon, Roy 553 Marples, Bjorn Mork, Alvaro Retana, Peng Shuping, Pascal Thubert, 554 Bernie Volz, Eric Vyncke, Robert Wilton. Authors would like to thank 555 Bob Hinden and Brian Carpenter for the initial idea of signaling 556 IPv6-only capability to hosts. Special thanks to Erik Kline, Mark 557 Townsley and Maciej Zenczykowski for the discussion which led to the 558 idea of signalling IPv6-only capability over DHCPv4. 560 8. References 562 8.1. Normative References 564 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 565 Requirement Levels", BCP 14, RFC 2119, 566 DOI 10.17487/RFC2119, March 1997, 567 . 569 [RFC2131] Droms, R., "Dynamic Host Configuration Protocol", 570 RFC 2131, DOI 10.17487/RFC2131, March 1997, 571 . 573 [RFC2563] Troll, R., "DHCP Option to Disable Stateless Auto- 574 Configuration in IPv4 Clients", RFC 2563, 575 DOI 10.17487/RFC2563, May 1999, 576 . 578 [RFC3927] Cheshire, S., Aboba, B., and E. Guttman, "Dynamic 579 Configuration of IPv4 Link-Local Addresses", RFC 3927, 580 DOI 10.17487/RFC3927, May 2005, 581 . 583 [RFC4039] Park, S., Kim, P., and B. Volz, "Rapid Commit Option for 584 the Dynamic Host Configuration Protocol version 4 585 (DHCPv4)", RFC 4039, DOI 10.17487/RFC4039, March 2005, 586 . 588 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 589 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 590 May 2017, . 592 8.2. Informative References 594 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 595 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 596 DOI 10.17487/RFC4861, September 2007, 597 . 599 [RFC4957] Krishnan, S., Ed., Montavont, N., Njedjou, E., Veerepalli, 600 S., and A. Yegin, Ed., "Link-Layer Event Notifications for 601 Detecting Network Attachments", RFC 4957, 602 DOI 10.17487/RFC4957, August 2007, 603 . 605 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 606 NAT64: Network Address and Protocol Translation from IPv6 607 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 608 April 2011, . 610 [RFC6147] Bagnulo, M., Sullivan, A., Matthews, P., and I. van 611 Beijnum, "DNS64: DNS Extensions for Network Address 612 Translation from IPv6 Clients to IPv4 Servers", RFC 6147, 613 DOI 10.17487/RFC6147, April 2011, 614 . 616 [RFC6877] Mawatari, M., Kawashima, M., and C. Byrne, "464XLAT: 617 Combination of Stateful and Stateless Translation", 618 RFC 6877, DOI 10.17487/RFC6877, April 2013, 619 . 621 Authors' Addresses 623 Lorenzo Colitti 624 Google 625 Shibuya 3-21-3 626 Shibuya, Tokyo 150-0002 627 JP 629 Email: lorenzo@google.com 631 Jen Linkova 632 Google 633 1 Darling Island Rd 634 Pyrmont, NSW 2009 635 AU 637 Email: furry@google.com 639 Michael C. Richardson 640 Sandelman Software Works 642 Email: mcr+ietf@sandelman.ca 643 URI: http://www.sandelman.ca/ 645 Tomek Mrugalski 646 Internet Systems Consortium, Inc. 647 950 Charter Street 648 Redwood City, CA 94063 649 USA 651 Email: tomasz.mrugalski@gmail.com