idnits 2.17.1 draft-templin-autoconf-dhcp-14.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1 on line 16. -- Found old boilerplate from RFC 3978, Section 5.5, updated by RFC 4748 on line 946. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 957. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 964. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 970. 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 Copyright Line does not match the current year -- 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 (April 4, 2008) is 5859 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Informational ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 2460 (Obsoleted by RFC 8200) ** Obsolete normative reference: RFC 3315 (Obsoleted by RFC 8415) ** Obsolete normative reference: RFC 3633 (Obsoleted by RFC 8415) == Outdated reference: A later version (-15) exists of draft-cheshire-dnsext-multicastdns-06 == Outdated reference: A later version (-13) exists of draft-ietf-dhc-subnet-alloc-06 == Outdated reference: A later version (-14) exists of draft-ietf-manet-smf-07 == Outdated reference: A later version (-23) exists of draft-templin-seal-03 -- Obsolete informational reference (is this intentional?): RFC 4941 (Obsoleted by RFC 8981) Summary: 4 errors (**), 0 flaws (~~), 5 warnings (==), 9 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group F. Templin, Ed. 3 Internet-Draft S. Russert 4 Intended status: Informational S. Yi 5 Expires: October 6, 2008 Boeing Phantom Works 6 April 4, 2008 8 The MANET Virtual Ethernet (VET) Abstraction 9 draft-templin-autoconf-dhcp-14.txt 11 Status of this Memo 13 By submitting this Internet-Draft, each author represents that any 14 applicable patent or other IPR claims of which he or she is aware 15 have been or will be disclosed, and any of which he or she becomes 16 aware will be disclosed, in accordance with Section 6 of BCP 79. 18 Internet-Drafts are working documents of the Internet Engineering 19 Task Force (IETF), its areas, and its working groups. Note that 20 other groups may also distribute working documents as Internet- 21 Drafts. 23 Internet-Drafts are draft documents valid for a maximum of six months 24 and may be updated, replaced, or obsoleted by other documents at any 25 time. It is inappropriate to use Internet-Drafts as reference 26 material or to cite them other than as "work in progress." 28 The list of current Internet-Drafts can be accessed at 29 http://www.ietf.org/ietf/1id-abstracts.txt. 31 The list of Internet-Draft Shadow Directories can be accessed at 32 http://www.ietf.org/shadow.html. 34 This Internet-Draft will expire on October 6, 2008. 36 Abstract 38 Mobile Ad-hoc Networks (MANETs) connect routers on links with 39 asymmetric reachability characteristics, and may also connect to 40 other networks including the Internet. Routers in MANETs must have a 41 way to automatically provision IP addresses/prefixes and other 42 information. This document specifies a Virtual Ethernet (VET) 43 abstraction for autoconfiguration and operation of routers in MANETs. 45 Table of Contents 47 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 48 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4 49 3. MANET Router Autoconfiguration . . . . . . . . . . . . . . . . 6 50 3.1. MANET Interface Autoconfiguration . . . . . . . . . . . . 6 51 3.2. MNBR/MNGW List Discovery . . . . . . . . . . . . . . . . . 7 52 3.3. VET Interface Autoconfiguration . . . . . . . . . . . . . 8 53 3.4. Ingress Interface Autoconfiguration . . . . . . . . . . . 8 54 3.4.1. Autoconfiguration of IPv4 Prefixes . . . . . . . . . . 9 55 3.4.2. Autoconfiguration of IPv6 Addresses/Prefixes . . . . . 9 56 3.4.3. Prefix and Route Maintenance . . . . . . . . . . . . . 11 57 3.5. Portable and Self-Configured IP Prefixes . . . . . . . . . 11 58 3.6. Separation of IP Addressing Domains . . . . . . . . . . . 12 59 4. Post-Autoconfiguration Operation . . . . . . . . . . . . . . . 12 60 4.1. Forwarding Packets to Off-MANET Destinations . . . . . . . 12 61 4.2. MANET-Local Communications . . . . . . . . . . . . . . . . 13 62 4.3. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 13 63 4.4. Service Discovery . . . . . . . . . . . . . . . . . . . . 13 64 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 13 65 6. Security Considerations . . . . . . . . . . . . . . . . . . . 14 66 7. Related Work . . . . . . . . . . . . . . . . . . . . . . . . . 14 67 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 14 68 9. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 14 69 10. References . . . . . . . . . . . . . . . . . . . . . . . . . . 15 70 10.1. Normative References . . . . . . . . . . . . . . . . . . . 15 71 10.2. Informative References . . . . . . . . . . . . . . . . . . 15 72 Appendix A. Duplicate Address Detection (DAD) Considerations . . 17 73 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 18 74 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 20 75 Intellectual Property and Copyright Statements . . . . . . . . . . 22 77 1. Introduction 79 Mobile Ad-hoc Networks (MANETs) connect MANET Routers (MNRs) on links 80 with asymmetric reachability characteristics (see: [RFC4861], Section 81 2.2). MNRs may participate in a routing protocol over MANET 82 interfaces to discover routes across the MANET using multiple Layer-2 83 or Layer-3 forwarding hops if necessary. MANETs may also connect to 84 other networks via MANET Border Routers (MNBRs) and connect to the 85 Internet via MANET Gateways (MNGWs). A MANET may be as simple as a 86 small collection of MNRs (and their attached networks); a MANET may 87 also contain other MANETs and/or be a subnetwork of a larger MANET. 89 MANETs that comprise homogeneous link types within a single IP subnet 90 can configure the routing protocol to operate as a sub-IP layer 91 mechanism such that IP sees the MANET as an ordinary shared link the 92 same as for a (bridged) campus LAN. In that case, a single IP hop is 93 sufficient to traverse the MANET without IP layer encapsulation. 95 MANETs that comprise heterogeneous link types and/or multiple IP 96 subnets must also provide a routing service that operates as an IP 97 layer mechanism, e.g., to accommodate media types with dissimilar 98 Layer-2 address formats and maximum transmission units (MTUs). In 99 that case, multiple IP hops may be necessary to traverse the MANET 100 such that specific autoconfiguration procedures are necessary to 101 avoid multilink subnet issues [RFC4903]. In particular, we describe 102 herein the use of a virtualized link that spans the MANET, to avoid 103 the multilink subnet issues that arise when MANET interfaces are used 104 directly by applications. 106 Conceptually, a MNR embodies a router entity that connects its 107 attached networks to MANETs and/or connects the MANET to other 108 networks including the Internet (see: Figure 1). The router entity 109 may also connect to an imaginary "Virtual Ethernet" that views other 110 routers in the MANET as single-hop neighbors. For each distinct 111 MANET to which they connect, MNRs discover a list of MNBRs that can 112 be used for forwarding packets to off-MANET destinations. An MNR 113 (and its attached networks) is a "site" unto itself, therefore a 114 MANET is a "site-of-sites". 116 This document specifies a Virtual EThernet (VET) abstraction using 117 IP-in-IP encapsulation for MANET autoconfiguration and operation with 118 multilink subnet avoidance; both IPv4 [RFC0791] and IPv6 [RFC2460] 119 are discussed within this context. The use of standard DHCP 120 [RFC2131][RFC3315] and neighbor discovery [RFC0826][RFC4861] 121 mechanisms is assumed unless otherwise specified. 123 2. Terminology 125 The terms "inner" and "outer" are used throughout this document to 126 respectively refer to the innermost IP {address, protocol, header, 127 packet, etc.} *before* encapsulation, and the outermost IP {address, 128 protocol, header, packet, etc.} *after* encapsulation. (There may 129 also be "mid-layer" encapsulations between the inner and outer 130 layers, including IPSec [RFC4301], the Subnetwork Encapsulation and 131 Adaptation Layer (SEAL) [I-D.templin-seal], etc.) 133 The terminology in [I-D.ietf-autoconf-manetarch] and the normative 134 references apply. The following terms are defined within the scope 135 of this document: 137 subnetwork 138 the same as defined in [RFC3819]. 140 Mobile Ad-hoc Network (MANET) 141 a connected network region of MANET routers that maintain a 142 routing structure among themselves over asymmetric reachability 143 links (see: [RFC4861], Section 2.2). A MANET may be as simple as 144 a small collection of routers (and their attached networks); a 145 MANET may also contain other MANETs and/or be a subnetwork of a 146 larger MANET. A MANET router (and its attached networks) is a 147 site unto itself, and a MANET is therefore a site-of-sites. 149 Further information on the characteristics of MANETs can be found 150 in [RFC2501]. 152 MANET Router (MNR) 153 a mobile router that forwards packets over MANET interfaces. For 154 the purpose of this specification, an MNR comprises a router 155 entity, one or more host entities, one or more MANET interfaces 156 and zero or more ingress/egress/VET interfaces (see: Figure 1). 158 MANET Border Router (MNBR) 159 an MNR that connects other networks to the MANET (via ingress 160 interfaces) and/or connects the MANET to other networks, including 161 the Internet (via egress interfaces). MNBRs also configure a VET 162 interface for automatic tunneling across the MANET. All MNBRs are 163 also MNRs. 165 MANET Gateway (MNGW) 166 a MNBR that connects the MANET to the Internet via egress 167 interfaces and can delegate addresses/prefixes to other MNBRs. 168 All MNGWs are also MNBRs. 170 egress/ingress interface 171 the same as defined in ([RFC3753], Section 3). Note that in some 172 MANET scenarios, an interface may dynamically switch from being an 173 ingress interface to being an egress interface, and vice-versa. 174 Addresses assigned to egress/ingress interfaces are used as 175 *inner* IP addresses during encapsulation. 177 MANET Interface 178 a MNR's attachment to a link in a MANET. A MANET interface is 179 "neutral" in its orientation, i.e., it is inherently neither 180 egress nor ingress. In particular, a packet may need to traverse 181 several MANET interfaces before it is forwarded via either an 182 egress or ingress interface. 184 MANET Local Address (MLA) 185 a MNR's IP address that is assigned to a MANET interface and 186 unique within the MANET. MLAs are used as identifiers for 187 operating the routing protocol and/or locators for packet 188 forwarding within the scope of the MANET; MLAs are also used as 189 *outer* IP addresses during encapsulation. 191 Virtual Ethernet (VET) 192 an imaginary shared link that connects all MNBRs in a MANET. 194 VET interface 195 a MNBR's attachment to a VET. The MNBR configures a VET interface 196 over a set of underlying MANET interface(s) belonging to the same 197 MANET. The VET interface encapsulates each inner IP packet in any 198 mid-layer headers plus an outer IP header then forwards it on an 199 underlying MANET interface such that the TTL/Hop Limit in the 200 inner header is not decremented as the packet traverses the MANET. 201 The VET interface presents an automatic tunneling abstraction that 202 represents the MANET as a unified shared link. 204 The following additional abbreviations are used throughout the 205 document: 207 CGA - Cryptographically Generated Address 208 DHCP[v4,v6] - the Dynamic Host Configuration Protocol 209 IP[v4,v6] - the Internet Protocol 210 ISATAP - Intra-Site Automatic Tunnel Addressing Protocol 211 ND - Neighbor Discovery 212 PIO - Prefix Information Option 213 RIO - Route Information Option 214 RS/RA - IPv6 Neighbor Discovery Router Solicitation/Advertisement 215 SEAL - Subnetwork Encapsulation and Adaptation Layer 216 SLAAC - IPv6 StateLess Address AutoConfiguation 218 3. MANET Router Autoconfiguration 220 Autoconfiguration entails the configuration of addresses/prefixes and 221 other information for routers in MANETs. Figure 1 below depicts the 222 conceptual model for a MNR: 223 Egress Interfaces (to Internet) 224 x x x 225 | | | 226 +------------------------+---+--------+----------+ 227 | Internal hosts | | | | M 228 | and routers | | .... | | A 229 | ,-. | +---+---+--------+---+ | N 230 | (H1 )---+ | /| | E 231 | | `-' | | I /*+------+--< T 232 | . | +---+ | | n|**| | 233 | . +--|R1 |---+-----+ t|**| | I 234 | . | +---+ | | Router V e|**+------+--< n 235 | | ,-. | | E r|**| . | t 236 | (H2 )---+ | Entity T f|**| . | e 237 | `-' | . | a|**| . | r 238 | . | c|**| . | f 239 | ,-. . | e \*+------+--< a 240 | (Hn )---------+ \| | c 241 | `-' +---+---+--------+---+ | e 242 | Ingress Interfaces | | .... | | s 243 | (to internal networks) | | | | 244 +------------------------+---+--------+----------+ 245 | | | 246 x x x 247 Ingress Interfaces (to attached networks) 249 Figure 1: MANET Router 251 MNRs configure one or more MANET interfaces and engage in any MANET 252 routing protocols over those interfaces. They also configure zero or 253 more egress interfaces that connect the MANET to the Internet, and 254 zero or more ingress interfaces (including internal virtual 255 interfaces such as a loopback interface) that attach other networks 256 to the MANET. MNRs that configure ingress/egress interfaces also act 257 as MNBRs, and configure a VET interface over a set of underlying 258 MANET interfaces belonging to the same MANET. MNRs obtain addresses/ 259 prefixes and other autoconfiguration information using the mechanisms 260 specified in the following sections: 262 3.1. MANET Interface Autoconfiguration 264 When a MNR joins a MANET, it first configures a unique IPv6 link- 265 local address on each MANET interface that requires an IPv6 link- 266 local capability and an IPv4 link-local address on each MANET 267 interface that requires an IPv4 link-local capability. IPv6 link- 268 local address generation mechanisms that provide sufficient 269 uniqueness include Cryptographically Generated Addresses (CGAs) 270 [RFC3972], StateLess Address AutoConfiguration (SLAAC) using EUI-64 271 interface identifiers [RFC4862], etc. The mechanisms specified in 272 [RFC3927] provide an IPv4 link-local address generation capability, 273 but with limited uniqueness properties. 275 Next, the MNR configures a MANET Local Address (MLA) of the outer IP 276 protocol version on each of its MANET interfaces and engages in any 277 MANET routing protocols on those interfaces. The MNR can configure 278 an MLA via explicit management, DHCP autoconfiguration, pseudo-random 279 self-generation from a suitably large address pool, or through an 280 alternate autoconfiguration mechanism. 282 DHCP configuration of MLAs may require support from relays within the 283 MANET that have already autoconfigured an MLA. For DHCPv6, relays 284 that do not already know the MLA of a server relay requests to the 285 'All_DHCP_Servers' site-scoped IPv6 multicast group. For DHCPv4, 286 relays that do not already know the MLA of a server relay requests to 287 the IPv4 multicast group address TBD (see: Section 5). DHCPv4 288 servers that delegate MLAs join the TBD multicast group and service 289 any DHCPv4 messages received for that group. 291 Self-generation of MLAs for IPv6 can be from a large IPv6 local-use 292 address range, e.g., IPv6 Unique Local Addresses [RFC4193]. Self- 293 generation of MLAs for IPv4 can be from a large IPv4 private address 294 range, e.g., 240/4 [I-D.fuller-240space]. When self-generation is 295 used alone, the MNR must continuously monitor the self-generated MLAs 296 through an in-service duplicate address detection mechanism, e.g., by 297 monitoring the routing protocol. 299 A combined approach using both DHCP and self-generation is also 300 possible. In this combined approach, the MNR first self-generates a 301 temporary MLA which it will use only for the purpose of procuring an 302 actual MLA from a DHCP server. Acting as a combined client/relay, 303 the MNR then uses the temporary MLA to engage in the routing protocol 304 and performs a relay-server exchange using the temporary MLA as an 305 address for the relay. When the DHCP server delegates an actual MLA, 306 the MNR abandons the temporary MLA, assigns the actual MLA to the 307 MANET interface and re-engages in the routing protocol. 309 3.2. MNBR/MNGW List Discovery 311 After the MNR configures its MANET interfaces, it next discovers the 312 list of MNBRs/MNGWs on the MANET. The list can be discovered through 313 information conveyed in the routing protocol or through the discovery 314 mechanisms outlined in [RFC5214], Section 8.3.2. 316 Identifying names/values for MNBRs/MNGWs (and/or the prefixes that 317 they aggregate) serve as an identifier for the MANET. 319 3.3. VET Interface Autoconfiguration 321 MNBRs configure a VET interface over a set of underlying MANET 322 interfaces belonging to the same MANET, where the VET interface 323 represents an attachment to a "virtual ethernet" that connects all 324 MNBRs in the MANET. Inner IP packets forwarded over the VET 325 interface are encapsulated in any mid-layer headers (e.g., IPsec, the 326 SEAL header, etc.) followed by an outer IP header, then submitted to 327 the outer IP forwarding engine for transmission on an underlying 328 MANET interface. 330 When IPv6 and IPv4 are used as the inner/outer protocols 331 (respectively), the MNBR autoconfigures an ISATAP link-local address 332 ([RFC5214], Section 6.2) on the VET interface to support packet 333 forwarding and operation of the IPv6 neighbor discovery protocol. 334 The ISATAP address embeds an IPv4 MLA assigned to an underlying MANET 335 interface, and need not be checked for uniqueness since the IPv4 MLA 336 itself was already determined to be unique. 338 After the MNBR configures a VET interface, it can communicate with 339 other MNBRs as on-link neighbors on the VET, i.e., it can confirm 340 reachability of other MNBRs through Neighbor Discovery (ND) and/or 341 DHCP exchanges over the VET interface. (The MNBR can also confirm 342 reachability through information conveyed in the MANET routing 343 protocol or through some other means associated with the specific 344 MANET subnetwork technology.) 346 The MNBR must be able to detect and recover from the loss of 347 neighbors on the VET due to e.g., MANET partitions, node failures, 348 etc. Mechanisms such as monitoring the routing protocol, ND 349 beaconing/polling, DHCP renewals/leasequeries, upper layer protocol 350 hints of forward progress, bidirectional forward detection, detection 351 of network attachment, etc. can be used according to the particular 352 deployment scenario. 354 3.4. Ingress Interface Autoconfiguration 356 MNBRs can acquire addresses and/or prefix delegations for assignment 357 on ingress interfaces through autoconfiguration exchanges with MNGWs 358 over the VET interface. These prefixes may be: 360 o global-scope and provider aggregated 362 o global-scope and provider-independent 364 o global-scope and 6to4 [RFC3056] 366 o unique-local scope and centrally administrated 368 o unique-local scope and locally assigned 370 o other non-link-local scope 372 Ingress interface autoconfiguration considerations are discussed in 373 the following sections: 375 3.4.1. Autoconfiguration of IPv4 Prefixes 377 When IPv4 is used as the inner protocol, the MNBR discovers the 378 addresses of one or more MNGWs that delegate IPv4 prefixes then 379 performs a DHCPv4 prefix delegation exchange 380 [I-D.ietf-dhc-subnet-alloc] over the VET interface to obtain IPv4 381 prefixes for assignment and/or sub-delegation on its ingress 382 interfaces. 384 To perform the DHCPv4 prefix delegation exchange, a DHCPv4 client 385 function associated with the MNBR's host entity forwards a 386 DHCPDISCOVER message with a Subnet Allocation option to a relay 387 function associated with its router entity, i.e., the MNBR acts as 388 both client and relay. The relay function then forwards the message 389 over the VET interface to the DHCPv4 server on a MNGW. The forwarded 390 DHCPDISCOVER will elicit a DHCPOFFER from the server containing IPv4 391 prefix delegations, and the MNBR completes the delegation through a 392 DHCPREQUEST/DHCPACK exchange (again using the combined client/relay 393 approach). 395 When the MNBR receives IPv4 prefix delegations, it assigns the 396 prefixes on ingress interfaces; it does not assign them on the VET 397 interface or on MANET interfaces. The MNBR can also obtain /32 398 prefixes using DHCPv4 prefix delegation the same as for any IPv4 399 prefix, and can assign them as IPv4 addresses with /32 netmasks on 400 ingress interfaces (including loopback interfaces). 402 3.4.2. Autoconfiguration of IPv6 Addresses/Prefixes 404 When IPv6 is used as the inner protocol, the MNBR sends unicast IPv6 405 Router Solicitation (RS) messages to MNGWs over the VET interface to 406 receive Router Advertisements (RAs) with Prefix Information Options 407 (PIOs) and/or with the 'M' flag set to signify whether DHCPv6 408 autoconfiguration is available. If the MNBR receives an RA 409 containing PIOs with the 'A' and 'L' bits set to 1, it autoconfigures 410 IPv6 addresses from the prefixes using SLAAC and assigns them to the 411 VET interface. (When IPv4 is used as the outer IP protocol, the 412 addresses are autoconfigured and assigned as ISATAP addresses the 413 same as specified in [RFC5214].) 415 When the MNBR receives an RA with the 'M' flag set to 1, the MNGW 416 that sent the RA also hosts a DHCPv6 server capable of delegating 417 IPv6 prefixes. If the RA also contains PIOs with the 'L' bit set to 418 0, the MNBR can use them as hints of prefixes the server is willing 419 to delegate. For example, a MNGW can include a PIO with a prefix 420 such as 2001::DB8::/64 as a hint of an aggregated prefix from which 421 an MNBR can delegate a /128. Whether or not such hints are 422 available, the MNBR can use DHCPv6 prefix delegation [RFC3633] to 423 obtain IPv6 prefixes from the MNGW for assignment and/or sub- 424 delegation on its ingress interfaces. 426 The MNBR can obtain prefixes using either a 2-message or 4-message 427 DHCPv6 exchange [RFC3315]. To perform the 2-message exchange, a 428 DHCPv6 client function associated with the MNBR's host entity 429 forwards a Solicit message with an IA_PD option to a relay function 430 associated with its router entity, i.e., the MNBR acts as both client 431 and relay. The relay function then forwards the message over the VET 432 interface to the DHCPv6 server. The forwarded Solicit message will 433 elicit a Reply from the server containing IPv6 prefix delegations. 434 When the MNBR receives IPv6 prefix delegations, it assigns the 435 prefixes on ingress interfaces; it does not assign them on the VET 436 interface or on MANET interfaces. 438 The MNBR can also obtain /128 prefixes using DHCPv6 prefix delegation 439 the same as for any IPv6 prefix. When the MNBR receives a prefix 440 delegation hint (see above) it can self-generate an address from the 441 prefix using mechanisms such as CGAs [RFC3972], IPv6 privacy address 442 [RFC4941], etc. without assigning the address to an interface. The 443 MNBR can then perform a DHCPv6 prefix delegation exchange to propose 444 the address as a /128 prefix to the DHCPv6 server per Section 7 of 445 [RFC3633]. The server will check the proposed prefix for consistency 446 and uniqueness, then return it in its reply to the MNBR if it was 447 able to delegate the prefix. 449 When no prefix delegation hints are available, the MNBR can self- 450 generate an address using "prefix gleaning" from a /128 prefix 451 generated by the DHCPv6 server. The MNBR first performs an ordinary 452 DHCPv6 prefix delegation exchange with the server to obtain a server- 453 generated /128 prefix delegation, then interprets the leading 64 bits 454 of the /128 prefix as a /64 prefix delegation hint. As described in 455 the previous paragraph, the MNBR then self-generates an address from 456 the /64 and proposes the resulting /128 prefix to the server, which 457 will in turn delegate the prefix to the MNBR. (The MNBR can instead 458 attempt "prefix substitution" in a 4-message DHCPv6 exchange by 459 requesting its self-generated /128 prefix instead of the one 460 advertised by the server, but some servers may find this confusing.) 462 When the MNBR receives /128 prefix delegations, it can assign them as 463 IPv6 addresses with /128 prefix lengths on ingress interfaces 464 (including loopback interfaces). 466 3.4.3. Prefix and Route Maintenance 468 When DHCP prefix delegation is used, the MNGW's DHCP server ensures 469 that the delegations are unique within the MANET and that its routing 470 function will forward IP packets over the VET interface to the MNBR 471 to which the prefix was delegated. The prefix delegation remains 472 active as long as the MNBR continues to issue renewals over the VET 473 interface before the lease lifetime expires. The lease lifetime also 474 keeps the delegation state active even if the link between the MNBR 475 and MNGW is disrupted for a period of time (e.g., due to a MANET 476 partition) before being reestablished (e.g., due to a MANET merge). 478 Since the DHCP client and relay are co-resident on the same MNBR, no 479 special coordination is necessary for the MNGW to maintain routing 480 information. The MNGW simply retains forwarding information base 481 entries that identify the MNBR as the next-hop toward the prefix via 482 the VET interface, and issues redirects over the VET interface the 483 same as for any link. 485 3.5. Portable and Self-Configured IP Prefixes 487 Independent of any MNGW-aggregated addresses/prefixes (see: 488 Section 3.4), a MNBR can retain portable IP prefixes (e.g., prefixes 489 taken from a home network, IPv6 Unique Local Addresses (ULAs) 490 [RFC4193][I-D.ietf-ipv6-ula-central], etc.) as it travels between 491 visited networks as long it coordinates in some fashion, e.g., with a 492 mapping agent, prefix aggregation authority, etc. MNBRs can sub- 493 delegate portable (and other self-configured) prefixes on networks 494 connected on their ingress interfaces. 496 Portable prefixes are not aggregated, redistributed or advertised by 497 MNGWs and can therefore travel with the MNBR as it moves to new 498 visited networks and/or configures peering arrangements with other 499 nodes. Generation and coordination of portable (and other self- 500 configured) prefixes can therefore occur independently of any other 501 autoconfiguration considerations. 503 3.6. Separation of IP Addressing Domains 505 When the inner and outer IP protocols are different (i.e., IPv6-in- 506 IPv4 or IPv4-in-IPv6), the MNR's dual-stack orientation provides a 507 natural separation between the inner and outer IP addressing domains, 508 and separate default routes can be configured for each domain. 510 When the inner and outer IP protocols are the same (i.e., IPv4-in- 511 IPv4 or IPv6-in-IPv6) separation between inner and outer IP 512 addressing domains can only be determined through the examination of 513 IP prefixes (else, the inner and outer IP addressing domains would 514 overlap). In that case, special configurations/mechanisms may be 515 necessary to support unambiguous determination of when to encapsulate 516 using the VET interface vs when to forward using a MANET interface. 517 In certain use cases, comingling the inner and outer addressing 518 domains directly over MANET interfaces and without invoking VET 519 encapsulation may be acceptable. 521 4. Post-Autoconfiguration Operation 523 After a MNR has been autoconfigured, it participates in any MANET 524 routing protocols and forwards packets over its MANET interfaces and 525 other attached interfaces. It also provides normal routing services 526 during post-autoconfiguration operation as specified in the following 527 sections: 529 4.1. Forwarding Packets to Off-MANET Destinations 531 MNBRs forward IP packets to off-MANET destinations via the VET 532 interface using another MNBR's address as the inner next-hop address. 533 For IPv6-in-IPv4 encapsulation using an ISATAP next-hop address, 534 determination of the outer destination address is through static 535 extraction of the embedded IPv4 address. For other IP-in-IP 536 encapsulations, determination of the outer destination address may 537 require additional supporting mechanisms. 539 MNBRs that use IPv6 as the inner protocol can discover default router 540 preferences and more-specific routes [RFC4191] by sending an RS over 541 the VET interface to elicit an RA from another MNBR. After default 542 and/or more-specific routes are discovered, the MNBR can forward IP 543 packets via a specific MNBR as the next-hop router on the VET 544 interface. When multiple default routers are available on the VET 545 interface, the MNBR can use default router preferences, traffic 546 engineering configurations, etc. to select the best exit router. 548 4.2. MANET-Local Communications 550 When permitted by policy, pairs of MNRs that configure the endpoints 551 of a communications session can avoid VET encapsulation by directly 552 invoking the outer IP protocol using MLAs assigned to their MANET 553 interfaces. For example, when the outer protocol is IPv4 a pair of 554 communicating MNRs can use IPv4 MLAs for direct communications over 555 their MANET interfaces without using the VET interface. 557 4.3. Multicast 559 In multicast-capable deployments, MNRs provide a MANET-wide 560 multicasting service such as Simplified Multicast Forwarding (SMF) 561 [I-D.ietf-manet-smf] over their MANET interfaces so that outer IP 562 multicast messages of site- or greater scope will be propagated 563 across the MANET. For these deployments, MNBRs can also provide an 564 inner IP multicast capability over their VET interfaces through 565 mapping of the inner IP multicast address space to the outer IP 566 multicast address space. 568 MNBRs encapsulate inner IP multicast messages sent over the VET 569 interface in any mid-layer headers (e.g., IPsec, SEAL, etc.) plus an 570 outer IP header with a site-scoped outer IP multicast address as the 571 destination. For the case of IPv6 and IPv4 as the inner/outer 572 protocols (respectively), [RFC2529] provides mappings from the IPv6 573 multicast address space to the IPv4 multicast address space. 575 For multicast-capable MANETs, use of the inner IP multicast service 576 for operating the ND protocol over the VET is available but should be 577 used sparingly to minimize MANET-wide flooding. Therefore, MNBRs 578 should use unicast ND services over the VET interface instead of 579 multicast whenever possible. 581 4.4. Service Discovery 583 MANET-wide service discovery can be accommodated by a suitable name- 584 to-address resolution service. Examples of flooding-based services 585 include the use of LLMNR [RFC4759] over the VET interface or mDNS 586 [I-D.cheshire-dnsext-multicastdns] over an underlying MANET 587 interface. More scalable and efficient service discovery mechanisms 588 for MANETs are for further study. 590 5. IANA Considerations 592 A site-scoped IPv4 multicast group (TBD) for DHCPv4 server discovery 593 is requested. The group should be allocated from the IPv4 multicast 594 site-local scope address block the same as for '239.255.2.2' (i.e., 595 the IPv4 multicast group allocated for the 'rasadv' protocol 596 [RASADV]). 598 6. Security Considerations 600 Security considerations for MANETs are found in 601 [RFC2501][I-D.ietf-autoconf-manetarch] and apply also to the 602 mechanisms and procedures specified in this document. 604 Security considerations for MANET routing protocols that may be used 605 within this context are found in their respective specifications. 607 7. Related Work 609 The authors acknowledge the work done by Brian Carpenter and Cyndi 610 Jung in [RFC2529] that introduced the concept of intra-site automatic 611 tunneling. This concept was later called: "Virtual Ethernet" and 612 investigated by Quang Nguyen under the guidance of Dr. Lixia Zhang. 614 Telcordia has proposed DHCP-related solutions for the CECOM MOSAIC 615 program. The Naval Research Lab (NRL) Information Technology 616 Division uses DHCP in their MANET research testbeds. Various IETF 617 AUTOCONF working group proposals have suggested similar mechanisms. 619 8. Acknowledgements 621 The following individuals gave direct and/or indirect input that was 622 essential to the work: Jari Arkko, Teco Boot, Emmanuel Bacelli, James 623 Bound, Thomas Clausen, Eric Fleischman, Bob Hinden, Joe Macker, 624 Thomas Narten, Alexandru Petrescu, Jinmei Tatuya, Dave Thaler, 625 Michaela Vanderveen and others in the IETF AUTOCONF and MANET working 626 groups. Many others have provided guidance over the course of many 627 years. 629 9. Contributors 631 Thomas Henderson (thomas.r.henderson@boeing.com) contributed to this 632 document. Ian Chakeres (ian.chakeres@gmail.com) contributed to 633 earlier versions of the document. 635 10. References 636 10.1. Normative References 638 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 639 September 1981. 641 [RFC0826] Plummer, D., "Ethernet Address Resolution Protocol: Or 642 converting network protocol addresses to 48.bit Ethernet 643 address for transmission on Ethernet hardware", STD 37, 644 RFC 826, November 1982. 646 [RFC2131] Droms, R., "Dynamic Host Configuration Protocol", 647 RFC 2131, March 1997. 649 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 650 (IPv6) Specification", RFC 2460, December 1998. 652 [RFC3315] Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C., 653 and M. Carney, "Dynamic Host Configuration Protocol for 654 IPv6 (DHCPv6)", RFC 3315, July 2003. 656 [RFC3633] Troan, O. and R. Droms, "IPv6 Prefix Options for Dynamic 657 Host Configuration Protocol (DHCP) version 6", RFC 3633, 658 December 2003. 660 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 661 More-Specific Routes", RFC 4191, November 2005. 663 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 664 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 665 September 2007. 667 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 668 Address Autoconfiguration", RFC 4862, September 2007. 670 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 671 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 672 March 2008. 674 10.2. Informative References 676 [I-D.cheshire-dnsext-multicastdns] 677 Cheshire, S. and M. Krochmal, "Multicast DNS", 678 draft-cheshire-dnsext-multicastdns-06 (work in progress), 679 August 2006. 681 [I-D.fuller-240space] 682 Fuller, V., "Reclassifying 240/4 as usable unicast address 683 space", draft-fuller-240space-02 (work in progress), 684 March 2008. 686 [I-D.ietf-autoconf-manetarch] 687 Chakeres, I., Macker, J., and T. Clausen, "Mobile Ad hoc 688 Network Architecture", draft-ietf-autoconf-manetarch-07 689 (work in progress), November 2007. 691 [I-D.ietf-dhc-subnet-alloc] 692 Johnson, R., "Subnet Allocation Option", 693 draft-ietf-dhc-subnet-alloc-06 (work in progress), 694 November 2007. 696 [I-D.ietf-ipv6-ula-central] 697 Hinden, R., "Centrally Assigned Unique Local IPv6 Unicast 698 Addresses", draft-ietf-ipv6-ula-central-02 (work in 699 progress), June 2007. 701 [I-D.ietf-manet-smf] 702 Macker, J. and S. Team, "Simplified Multicast Forwarding 703 for MANET", draft-ietf-manet-smf-07 (work in progress), 704 February 2008. 706 [I-D.templin-seal] 707 Templin, F., "Subnetwork Encapsulation and Adaptation 708 Layer", draft-templin-seal-03 (work in progress), 709 February 2008. 711 [RASADV] MSDN, "Remote Access Server Advertisement (RASADV) 712 Protocol Specification, 713 http://msdn2.microsoft.com/en-us/library/cc240334.aspx", 714 March 2008. 716 [RFC2501] Corson, M. and J. Macker, "Mobile Ad hoc Networking 717 (MANET): Routing Protocol Performance Issues and 718 Evaluation Considerations", RFC 2501, January 1999. 720 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 721 Domains without Explicit Tunnels", RFC 2529, March 1999. 723 [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains 724 via IPv4 Clouds", RFC 3056, February 2001. 726 [RFC3753] Manner, J. and M. Kojo, "Mobility Related Terminology", 727 RFC 3753, June 2004. 729 [RFC3819] Karn, P., Bormann, C., Fairhurst, G., Grossman, D., 730 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 731 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 732 RFC 3819, July 2004. 734 [RFC3927] Cheshire, S., Aboba, B., and E. Guttman, "Dynamic 735 Configuration of IPv4 Link-Local Addresses", RFC 3927, 736 May 2005. 738 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 739 RFC 3972, March 2005. 741 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 742 Addresses", RFC 4193, October 2005. 744 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 745 Internet Protocol", RFC 4301, December 2005. 747 [RFC4759] Stastny, R., Shockey, R., and L. Conroy, "The ENUM Dip 748 Indicator Parameter for the "tel" URI", RFC 4759, 749 December 2006. 751 [RFC4903] Thaler, D., "Multi-Link Subnet Issues", RFC 4903, 752 June 2007. 754 [RFC4941] Narten, T., Draves, R., and S. Krishnan, "Privacy 755 Extensions for Stateless Address Autoconfiguration in 756 IPv6", RFC 4941, September 2007. 758 Appendix A. Duplicate Address Detection (DAD) Considerations 760 Pre-service DAD for an MLA assigned on a MANET interface (such as 761 specified in [RFC4862]) would require either flooding the entire 762 MANET or somehow discovering a link in the MANET on which a node that 763 configures a duplicate address is attached and performing a localized 764 DAD exchange on that link. But, the control message overhead for 765 such a MANET-wide DAD would be substantial and prone to false- 766 negatives due to packet loss and intermittent connectivity. An 767 alternative to pre-service DAD is to autoconfigure pseudo-random MLAs 768 on MANET interfaces and employ a passive in-service DAD (e.g., one 769 that monitors routing protocol messages for duplicate assignments). 771 Pseudo-random IPv6 MLAs can be generated with mechanisms such as 772 CGAs, IPv6 privacy addresses, etc. with very small probability of 773 collision. Pseudo-random IPv4 MLAs can be generated through random 774 assignment from a suitably large IPv4 prefix space, e.g., the soon- 775 to-be-reclassified 240/4 space [I-D.fuller-240space]. 777 Consistent operational practices can assure uniqueness for MNGW- 778 aggregated addresses/prefixes, while statistical properties for 779 pseudo-random address self-generation can assure uniqueness for the 780 MLAs assigned on a MNR's MANET interfaces. Still, an MLA delegation 781 authority should be used when available, while a passive in-service 782 DAD mechanism should be used to detect MLA duplications when their is 783 no MLA delegation authority. 785 Appendix B. Change Log 787 (Note to RFC editor - this section to be removed before publication 788 as an RFC.) 790 Changes from -12 to 14: 792 o title change to "The MANET Virtual Ethernet Abstraction". 794 o Minor section rearrangement. 796 o Clartifications on portable and self-configured prefixes. 798 o Clarifications on DHCPv6 prefix delegation procedures. 800 Changes from -11 to 12: 802 o title change to "MANET Autoconfiguration using Virtual Ethernet". 804 o DHCP prefix delegation for both IPv4 and IPv6 as primary address 805 delegation mechanism. 807 o IPv6 SLAAC for address autoconfiguration on the VET interface. 809 o fixed editorials based on comments received. 811 Changes from -10 to 11: 813 o removed the transparent/opaque VET portal abstractions. 815 o removed routing header as an option for MANET exit router 816 selection. 818 o included IPv6 SLAAC as an endorsed address configuration mechanism 819 for the VET interface. 821 Changes from -08 to -09: 823 o Introduced the term "VET". 825 o Changed address delegation language to speak of "MNBR-aggregated" 826 instead of global/local. 828 o Updated figures 1-3. 830 o Explained why a MANET interface is "neutral". 832 o Removed DHCPv4 "MLA Address option". Now, MNBRs can only be 833 DHCPv4 servers; not relays. 835 Changes from -07 to -08: 837 o changed terms "unenhanced" and "enhanced" to "transparent" and 838 "opaque". 840 o revised MANET Router diagram. 842 o introduced RFC3753 terminology for Mobile Router; ingress/egress 843 interface. 845 o changed abbreviations to "MNR" and "MNBR". 847 o added text on ULAs and ULA-Cs to "Self-Generated Addresses". 849 o rearranged Section 3.1. 851 o various minor text cleanups 853 Changes from -06 to -07: 855 o added MANET Router diagram. 857 o added new references 859 o various minor text cleanups 861 Changed from -05 to -06: 863 o Changed terms "raw" and "cooked" to "unenhanced" and "enhanced". 865 o minor changes to preserve generality 867 Changed from -04 to -05: 869 o introduced conceptual "virtual ethernet" model. 871 o support "raw" and "cooked" modes as equivalent access methods on 872 the virutal ethernet. 874 Changed from -03 to -04: 876 o introduced conceptual "imaginary shared link" as a representation 877 for a MANET. 879 o discussion of autonomous system and site abstractions for MANETs 881 o discussion of autoconfiguration of CGAs 883 o new appendix on IPv6 StateLess Address AutoConfiguration 885 Changes from -02 to -03: 887 o updated terminology based on RFC2461 "asymmetric reachability" 888 link type; IETF67 MANET Autoconf wg discussions. 890 o added new appendix on IPv6 Neighbor Discovery and Duplicate 891 Address Detection 893 o relaxed DHCP server deployment considerations allow DHCP servers 894 within the MANET itself 896 Changes from -01 to -02: 898 o minor updates for consistency with recent developments 900 Changes from -00 to -01: 902 o new text on DHCPv6 prefix delegation and multilink subnet 903 considerations. 905 o various editorial changes 907 Authors' Addresses 909 Fred L. Templin (editor) 910 Boeing Phantom Works 911 P.O. Box 3707 MC 7L-49 912 Seattle, WA 98124 913 USA 915 Email: fltemplin@acm.org 916 Steven W. Russert 917 Boeing Phantom Works 918 P.O. Box 3707 MC 7L-49 919 Seattle, WA 98124 920 USA 922 Email: steven.w.russert@boeing.com 924 Seung Yi 925 Boeing Phantom Works 926 P.O. Box 3707 MC 7L-49 927 Seattle, WA 98124 928 USA 930 Email: seung.yi@boeing.com 932 Full Copyright Statement 934 Copyright (C) The IETF Trust (2008). 936 This document is subject to the rights, licenses and restrictions 937 contained in BCP 78, and except as set forth therein, the authors 938 retain all their rights. 940 This document and the information contained herein are provided on an 941 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 942 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND 943 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS 944 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF 945 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 946 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 948 Intellectual Property 950 The IETF takes no position regarding the validity or scope of any 951 Intellectual Property Rights or other rights that might be claimed to 952 pertain to the implementation or use of the technology described in 953 this document or the extent to which any license under such rights 954 might or might not be available; nor does it represent that it has 955 made any independent effort to identify any such rights. Information 956 on the procedures with respect to rights in RFC documents can be 957 found in BCP 78 and BCP 79. 959 Copies of IPR disclosures made to the IETF Secretariat and any 960 assurances of licenses to be made available, or the result of an 961 attempt made to obtain a general license or permission for the use of 962 such proprietary rights by implementers or users of this 963 specification can be obtained from the IETF on-line IPR repository at 964 http://www.ietf.org/ipr. 966 The IETF invites any interested party to bring to its attention any 967 copyrights, patents or patent applications, or other proprietary 968 rights that may cover technology that may be required to implement 969 this standard. Please address the information to the IETF at 970 ietf-ipr@ietf.org.