idnits 2.17.1 draft-templin-aerolink-38.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 : ---------------------------------------------------------------------------- == The 'Obsoletes: ' line in the draft header should list only the _numbers_ of the RFCs which will be obsoleted by this document (if approved); it should not include the word 'RFC' in the list. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (September 19, 2014) is 3506 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) == Unused Reference: 'RFC0768' is defined on line 2379, but no explicit reference was found in the text == Unused Reference: 'RFC4862' is defined on line 2418, but no explicit reference was found in the text == Unused Reference: 'RFC0879' is defined on line 2431, but no explicit reference was found in the text == Unused Reference: 'RFC4994' is defined on line 2477, but no explicit reference was found in the text == Unused Reference: 'RFC6146' is defined on line 2510, but no explicit reference was found in the text == Unused Reference: 'RFC6204' is defined on line 2514, but no explicit reference was found in the text == Unused Reference: 'RFC6355' is defined on line 2521, but no explicit reference was found in the text == Unused Reference: 'RFC6691' is defined on line 2529, but no explicit reference was found in the text == Unused Reference: 'RFC6706' is defined on line 2532, but no explicit reference was found in the text == Unused Reference: 'RFC6980' is defined on line 2548, but no explicit reference was found in the text == Unused Reference: 'RFC7078' is defined on line 2551, but no explicit reference was found in the text ** 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) ** Obsolete normative reference: RFC 6434 (Obsoleted by RFC 8504) == Outdated reference: A later version (-21) exists of draft-ietf-dhc-sedhcpv6-03 -- Obsolete informational reference (is this intentional?): RFC 879 (Obsoleted by RFC 7805, RFC 9293) -- Obsolete informational reference (is this intentional?): RFC 5246 (Obsoleted by RFC 8446) -- Obsolete informational reference (is this intentional?): RFC 5996 (Obsoleted by RFC 7296) -- Obsolete informational reference (is this intentional?): RFC 6204 (Obsoleted by RFC 7084) -- Obsolete informational reference (is this intentional?): RFC 6691 (Obsoleted by RFC 9293) Summary: 4 errors (**), 0 flaws (~~), 14 warnings (==), 6 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 Boeing Research & Technology 4 Obsoletes: rfc5320, rfc5558, rfc5720, September 19, 2014 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: March 23, 2015 10 Transmission of IP Packets over AERO Links 11 draft-templin-aerolink-38.txt 13 Abstract 15 This document specifies the operation of IP over tunnel virtual links 16 using Asymmetric Extended Route Optimization (AERO). Nodes attached 17 to AERO links can exchange packets via trusted intermediate routers 18 that provide forwarding services to reach off-link destinations and 19 redirection services for route optimization. AERO provides an IPv6 20 link-local address format known as the AERO address that supports 21 operation of the IPv6 Neighbor Discovery (ND) protocol and links IPv6 22 ND to IP forwarding. Admission control and provisioning are 23 supported by the Dynamic Host Configuration Protocol for IPv6 24 (DHCPv6), and node mobility is naturally supported through dynamic 25 neighbor cache updates. Although DHCPv6 and IPv6 ND messaging is 26 used in the control plane, both IPv4 and IPv6 are supported in the 27 data plane. 29 Status of This Memo 31 This Internet-Draft is submitted in full conformance with the 32 provisions of BCP 78 and BCP 79. 34 Internet-Drafts are working documents of the Internet Engineering 35 Task Force (IETF). Note that other groups may also distribute 36 working documents as Internet-Drafts. The list of current Internet- 37 Drafts is at http://datatracker.ietf.org/drafts/current/. 39 Internet-Drafts are draft documents valid for a maximum of six months 40 and may be updated, replaced, or obsoleted by other documents at any 41 time. It is inappropriate to use Internet-Drafts as reference 42 material or to cite them other than as "work in progress." 44 This Internet-Draft will expire on March 23, 2015. 46 Copyright Notice 48 Copyright (c) 2014 IETF Trust and the persons identified as the 49 document authors. All rights reserved. 51 This document is subject to BCP 78 and the IETF Trust's Legal 52 Provisions Relating to IETF Documents 53 (http://trustee.ietf.org/license-info) in effect on the date of 54 publication of this document. Please review these documents 55 carefully, as they describe your rights and restrictions with respect 56 to this document. Code Components extracted from this document must 57 include Simplified BSD License text as described in Section 4.e of 58 the Trust Legal Provisions and are provided without warranty as 59 described in the Simplified BSD License. 61 Table of Contents 63 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 64 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4 65 3. Asymmetric Extended Route Optimization (AERO) . . . . . . . . 6 66 3.1. AERO Link Reference Model . . . . . . . . . . . . . . . . 6 67 3.2. AERO Node Types . . . . . . . . . . . . . . . . . . . . . 7 68 3.3. AERO Addresses . . . . . . . . . . . . . . . . . . . . . 8 69 3.4. AERO Interface Characteristics . . . . . . . . . . . . . 9 70 3.4.1. Coordination of Multiple Underlying Interfaces . . . 11 71 3.5. AERO Interface Neighbor Cache Maintenace . . . . . . . . 12 72 3.6. AERO Interface Sending Algorithm . . . . . . . . . . . . 14 73 3.7. AERO Interface Encapsulation, Re-encapsulation and 74 Decapsulation . . . . . . . . . . . . . . . . . . . . . . 15 75 3.8. AERO Interface Data Origin Authentication . . . . . . . . 17 76 3.9. AERO Interface MTU Considerations . . . . . . . . . . . . 17 77 3.10. AERO Interface Error Handling . . . . . . . . . . . . . . 21 78 3.11. AERO Router Discovery, Prefix Delegation and Address 79 Configuration . . . . . . . . . . . . . . . . . . . . . . 25 80 3.11.1. AERO DHCPv6 Service Model . . . . . . . . . . . . . 25 81 3.11.2. AERO Client Behavior . . . . . . . . . . . . . . . . 25 82 3.11.3. AERO Server Behavior . . . . . . . . . . . . . . . . 28 83 3.12. AERO Relay/Server Routing System . . . . . . . . . . . . 30 84 3.13. AERO Redirection . . . . . . . . . . . . . . . . . . . . 30 85 3.13.1. Reference Operational Scenario . . . . . . . . . . . 31 86 3.13.2. Concept of Operations . . . . . . . . . . . . . . . 32 87 3.13.3. Message Format . . . . . . . . . . . . . . . . . . . 32 88 3.13.4. Sending Predirects . . . . . . . . . . . . . . . . . 33 89 3.13.5. Re-encapsulating and Relaying Predirects . . . . . . 35 90 3.13.6. Processing Predirects and Sending Redirects . . . . 35 91 3.13.7. Re-encapsulating and Relaying Redirects . . . . . . 37 92 3.13.8. Processing Redirects . . . . . . . . . . . . . . . . 38 93 3.13.9. Server-Oriented Redirection . . . . . . . . . . . . 38 95 3.14. Neighbor Unreachability Detection (NUD) . . . . . . . . . 39 96 3.15. Mobility Management . . . . . . . . . . . . . . . . . . . 40 97 3.15.1. Announcing Link-Layer Address Changes . . . . . . . 40 98 3.15.2. Bringing New Links Into Service . . . . . . . . . . 41 99 3.15.3. Removing Existing Links from Service . . . . . . . . 42 100 3.15.4. Moving to a New Server . . . . . . . . . . . . . . . 42 101 3.16. Encapsulation Protocol Version Considerations . . . . . . 43 102 3.17. Multicast Considerations . . . . . . . . . . . . . . . . 43 103 3.18. Operation on AERO Links Without DHCPv6 Services . . . . . 43 104 3.19. Operation on Server-less AERO Links . . . . . . . . . . . 43 105 3.20. Proxy AERO . . . . . . . . . . . . . . . . . . . . . . . 44 106 3.21. Extending AERO Links Through Security Gateways . . . . . 45 107 3.22. Extending IPv6 AERO Links to the Internet . . . . . . . . 47 108 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 50 109 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 50 110 6. Security Considerations . . . . . . . . . . . . . . . . . . . 50 111 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 51 112 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 51 113 8.1. Normative References . . . . . . . . . . . . . . . . . . 51 114 8.2. Informative References . . . . . . . . . . . . . . . . . 52 115 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 55 117 1. Introduction 119 This document specifies the operation of IP over tunnel virtual links 120 using Asymmetric Extended Route Optimization (AERO). The AERO link 121 can be used for tunneling to neighboring nodes over either IPv6 or 122 IPv4 networks, i.e., AERO views the IPv6 and IPv4 networks as 123 equivalent links for tunneling. Nodes attached to AERO links can 124 exchange packets via trusted intermediate routers that provide 125 forwarding services to reach off-link destinations and redirection 126 services for route optimization that addresses the requirements 127 outlined in [RFC5522]. 129 AERO provides an IPv6 link-local address format known as the AERO 130 address that supports operation of the IPv6 Neighbor Discovery (ND) 131 [RFC4861] protocol and links IPv6 ND to IP forwarding. Admission 132 control and provisioning are supported by the Dynamic Host 133 Configuration Protocol for IPv6 (DHCPv6) [RFC3315], and node mobility 134 is naturally supported through dynamic neighbor cache updates. 135 Although DHCPv6 and IPv6 ND message signalling is used in the control 136 plane, both IPv4 and IPv6 can be used in the data plane. The 137 remainder of this document presents the AERO specification. 139 2. Terminology 141 The terminology in the normative references applies; the following 142 terms are defined within the scope of this document: 144 AERO link 145 a Non-Broadcast, Multiple Access (NBMA) tunnel virtual overlay 146 configured over a node's attached IPv6 and/or IPv4 networks. All 147 nodes on the AERO link appear as single-hop neighbors from the 148 perspective of the virtual overlay. 150 AERO interface 151 a node's attachment to an AERO link. 153 AERO address 154 an IPv6 link-local address constructed as specified in Section 3.2 155 and applied to a Client's AERO interface. 157 AERO node 158 a node that is connected to an AERO link and that participates in 159 IPv6 ND and DHCPv6 messaging over the link. 161 AERO Client ("Client") 162 a node that applies an AERO address to an AERO interface and 163 receives an IP prefix via a DHCPv6 Prefix Delegation (PD) exchange 164 with one or more AERO Servers. 166 AERO Server ("Server") 167 a node that configures an AERO interface to provide default 168 forwarding and DHCPv6 services for AERO Clients. The Server 169 applies the IPv6 link-local subnet router anycast address (fe80::) 170 to the AERO interface and also applies an administratively 171 assigned IPv6 link-local unicast address used for operation of 172 DHCPv6 and the IPv6 ND protocol. 174 AERO Relay ("Relay") 175 a node that configures an AERO interface to relay IP packets 176 between nodes on the same AERO link and/or forward IP packets 177 between the AERO link and the native Internetwork. The Relay 178 applies an administratively assigned IPv6 link-local unicast 179 address to the AERO interface the same as for a Server. 181 ingress tunnel endpoint (ITE) 182 an AERO interface endpoint that injects tunneled packets into an 183 AERO link. 185 egress tunnel endpoint (ETE) 186 an AERO interface endpoint that receives tunneled packets from an 187 AERO link. 189 underlying network 190 a connected IPv6 or IPv4 network routing region over which the 191 tunnel virtual overlay is configured. A typical example is an 192 enterprise network. 194 underlying interface 195 an AERO node's interface point of attachment to an underlying 196 network. 198 link-layer address 199 an IP address assigned to an AERO node's underlying interface. 200 When UDP encapsulation is used, the UDP port number is also 201 considered as part of the link-layer address. Link-layer 202 addresses are used as the encapsulation header source and 203 destination addresses. 205 network layer address 206 the source or destination address of the encapsulated IP packet. 208 end user network (EUN) 209 an internal virtual or external edge IP network that an AERO 210 Client connects to the rest of the network via the AERO interface. 212 AERO Service Prefix (ASP) 213 an IP prefix associated with the AERO link and from which AERO 214 Client Prefixes (ACPs) are derived (for example, the IPv6 ACP 215 2001:db8:1:2::/64 is derived from the IPv6 ASP 2001:db8::/32). 217 AERO Client Prefix (ACP) 218 a more-specific IP prefix taken from an ASP and delegated to a 219 Client. 221 Throughout the document, the simple terms "Client", "Server" and 222 "Relay" refer to "AERO Client", "AERO Server" and "AERO Relay", 223 respectively. Capitalization is used to distinguish these terms from 224 DHCPv6 client/server/relay. 226 Throughout the document, it is said that an address is "applied" to 227 an AERO interface since the address is not "assigned" to the 228 interface from the perspective of the IP layer. However, the address 229 must at least be bound to the interface in some fashion to support 230 the operation of DHCPv6 and the IPv6 ND protocol. 232 The terminology of [RFC4861] (including the names of node variables 233 and protocol constants) applies to this document. Also throughout 234 the document, the term "IP" is used to generically refer to either 235 Internet Protocol version (i.e., IPv4 or IPv6). 237 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 238 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 239 document are to be interpreted as described in [RFC2119]. 241 3. Asymmetric Extended Route Optimization (AERO) 243 The following sections specify the operation of IP over Asymmetric 244 Extended Route Optimization (AERO) links: 246 3.1. AERO Link Reference Model 248 .-(::::::::) 249 .-(:::: IP ::::)-. 250 (:: Internetwork ::) 251 `-(::::::::::::)-' 252 `-(::::::)-' 253 | 254 +--------------+ +--------+-------+ +--------------+ 255 |AERO Server S1| | AERO Relay R1 | |AERO Server S2| 256 | Nbr: C1; R1 | | Nbr: S1; S2 | | Nbr: C2; R1 | 257 | default->R1 | |(H1->S1; H2->S2)| | default->R1 | 258 | H1->C1 | +--------+-------+ | H2->C2 | 259 +-------+------+ | +------+-------+ 260 | | | 261 X---+---+-------------------+------------------+---+---X 262 | AERO Link | 263 +-----+--------+ +--------+-----+ 264 |AERO Client C1| |AERO Client C2| 265 | Nbr: S1 | | Nbr: S2 | 266 | default->S1 | | default->S2 | 267 +--------------+ +--------------+ 268 .-. .-. 269 ,-( _)-. ,-( _)-. 270 .-(_ IP )-. .-(_ IP )-. 271 (__ EUN ) (__ EUN ) 272 `-(______)-' `-(______)-' 273 | | 274 +--------+ +--------+ 275 | Host H1| | Host H2| 276 +--------+ +--------+ 278 Figure 1: AERO Link Reference Model 280 Figure 1 above presents the AERO link reference model. In this 281 model: 283 o Relay R1 acts as a default router for its associated Servers S1 284 and S2, and connects the AERO link to the rest of the IP 285 Internetwork 287 o Servers S1 and S2 associate with Relay R1 and also act as default 288 routers for their associated Clients C1 and C2. 290 o Clients C1 and C2 associate with Servers S1 and S2, respectively 291 and also act as default routers for their associated EUNs 293 o Hosts H1 and H2 attach to the EUNs served by Clients C1 and C2, 294 respectively 296 In common operational practice, there may be many additional Relays, 297 Servers and Clients. 299 3.2. AERO Node Types 301 AERO Relays provide default forwarding services to AERO Servers. 302 Relays forward packets between Servers connected to the same AERO 303 link and also forward packets between the AERO link and the native 304 Internetwork. Relays present the AERO link to the native 305 Internetwork as a set of one or more AERO Service Prefixes (ASPs). 306 Each Relay advertises the ASPs for the AERO link into the native IP 307 Internetwork and serves as a gateway between the AERO link and the 308 Internetwork. AERO Relays maintain an AERO interface neighbor cache 309 entry for each AERO Server, and maintain an IP forwarding table entry 310 for each AERO Client Prefix (ACP). 312 AERO Servers provide default forwarding services to AERO Clients. 313 Each Server also peers with each Relay in a dynamic routing protocol 314 instance to advertise its list of associated ACPs. Servers configure 315 a DHCPv6 server function to facilitate Prefix Delegation (PD) 316 exchanges with Clients. Each delegated prefix becomes an ACP taken 317 from an ASP. Servers forward packets between Clients and Relays, as 318 well as between Clients and other Clients associated with the same 319 Server. AERO Servers maintain an AERO interface neighbor cache entry 320 for each AERO Relay. They also maintain both a neighbor cache entry 321 and an IP forwarding table entry for each of their associated 322 Clients. 324 AERO Clients act as requesting routers to receive ACPs through DHCPv6 325 PD exchanges with AERO Servers over the AERO link and sub-delegate 326 portions of their ACPs to EUN interfaces. (Each Client MAY associate 327 with a single Server or with multiple Servers, e.g., for fault 328 tolerance and/or load balancing.) Each IPv6 Client receives at least 329 a /64 IPv6 ACP, and may receive even shorter prefixes. Similarly, 330 each IPv4 Client receives at least a /32 IPv4 ACP (i.e., a singleton 331 IPv4 address), and may receive even shorter prefixes. AERO Clients 332 maintain an AERO interface neighbor cache entry for each of their 333 associated Servers as well as for each of their correspondent 334 Clients. 336 AERO Clients that act as hosts typically configure a TUN/TAP 337 interface as a point-to-point linkage between the IP layer and the 338 AERO interface. The IP layer therefore sees only the TUN/TAP 339 interface, while the AERO interface provides an intermediate conduit 340 between the TUN/TAP interface and the underlying interfaces. AERO 341 Clients that act as hosts assign one or more IP addresses from their 342 ACPs to the TUN/TAP interface. 344 3.3. AERO Addresses 346 An AERO address is an IPv6 link-local address with an embedded ACP 347 and applied to a Client's AERO interface. The AERO address is formed 348 as follows: 350 fe80::[ACP] 352 For IPv6, the AERO address begins with the prefix fe80::/64 and 353 includes in its interface identifier the base prefix taken from the 354 Client's IPv6 ACP. The base prefix is determined by masking the ACP 355 with the prefix length. For example, if the AERO Client receives the 356 IPv6 ACP: 358 2001:db8:1000:2000::/56 360 it constructs its AERO address as: 362 fe80::2001:db8:1000:2000 364 For IPv4, the AERO address is formed from the lower 64 bits of an 365 IPv4-mapped IPv6 address [RFC4291] that includes the base prefix 366 taken from the Client's IPv4 ACP. For example, if the AERO Client 367 receives the IPv4 ACP: 369 192.0.2.32/28 371 it constructs its AERO address as: 373 fe80::FFFF:192.0.2.32 375 The AERO address remains stable as the Client moves between 376 topological locations, i.e., even if its link-layer addresses change. 378 NOTE: In some cases, prospective neighbors may not have advanced 379 knowledge of the Client's ACP length and may therefore send initial 380 IPv6 ND messages with an AERO destination address that matches the 381 ACP but does not correspond to the base prefix. In that case, the 382 Client MUST accept the address as equivalent to the base address, but 383 then use the base address as the source address of any IPv6 ND 384 message replies. For example, if the Client receives the IPv6 ACP 385 2001:db8:1000:2000::/56 then subsequently receives an IPv6 ND message 386 with destination address fe80::2001:db8:1000:2001, it accepts the 387 message but uses fe80::2001:db8:1000:2000 as the source address of 388 any IPv6 ND replies. 390 3.4. AERO Interface Characteristics 392 AERO interfaces use IP-in-IPv6 encapsulation [RFC2473] to exchange 393 tunneled packets with AERO neighbors attached to an underlying IPv6 394 network, and use IP-in-IPv4 encapsulation [RFC2003][RFC4213] to 395 exchange tunneled packets with AERO neighbors attached to an 396 underlying IPv4 network. AERO interfaces can also coordinate secured 397 tunnel types such as IPsec [RFC4301] or TLS [RFC5246]. When Network 398 Address Translator (NAT) traversal and/or filtering middlebox 399 traversal may be necessary, a UDP header is further inserted 400 immediately above the IP encapsulation header. 402 AERO interfaces maintain a neighbor cache, and AERO Clients and 403 Servers use an adaptation of standard unicast IPv6 ND messaging. 404 AERO interfaces use unicast Neighbor Solicitation (NS), Neighbor 405 Advertisement (NA), Router Solicitation (RS) and Router Advertisement 406 (RA) messages the same as for any IPv6 link. AERO interfaces use two 407 redirection message types -- the first known as a Predirect message 408 and the second being the standard Redirect message (see Section 3.9). 409 AERO links further use link-local-only addressing; hence, AERO nodes 410 ignore any Prefix Information Options (PIOs) they may receive in RA 411 messages over an AERO interface. 413 AERO interface ND messages include one or more Target Link-Layer 414 Address Options (TLLAOs) formatted as shown in Figure 2: 416 0 1 2 3 417 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 1 418 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 419 | Type = 2 | Length = 3 | Reserved | 420 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 421 | Link ID | Preference | UDP Port Number (or 0) | 422 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 423 | | 424 +-- --+ 425 | | 426 +-- IP Address --+ 427 | | 428 +-- --+ 429 | | 430 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 432 Figure 2: AERO Target Link-Layer Address Option (TLLAO) Format 434 In this format, Link ID is an integer value between 0 and 255 435 corresponding to an underlying interface of the target node, and 436 Preference is an integer value between 0 and 255 indicating the 437 node's preference for this underlying interface (with 255 being the 438 highest preference, 1 being the lowest, and 0 meaning "link 439 disabled"). UDP Port Number and IP Address are set to the addresses 440 used by the target node when it sends encapsulated packets over the 441 underlying interface. When no UDP encapsulation is used, UDP Port 442 Number is set to 0. When the encapsulation IP address family is 443 IPv4, IP Address is formed as an IPv4-mapped IPv6 address [RFC4291]. 445 When a Relay enables an AERO interface, it applies an 446 administratively assigned link-local address fe80::ID to the 447 interface. Each fe80::ID address MUST be unique among all Relays and 448 Servers on the link, and MUST NOT collide with any potential AERO 449 addresses. The addresses are typically taken from the range 450 fe80::/96, e.g., as fe80::1, fe80::2, fe80::3, etc. The Relay also 451 maintains an IP forwarding table entry for each Client-Server 452 association and maintains a neighbor cache entry for each Server on 453 the link. Relays do not require the use of IPv6 ND messaging for 454 reachability determination since Relays and Servers engage in a 455 dynamic routing protocol over the AERO interface. At a minimum, 456 however, Relays respond to NS messages by returning an NA. 458 When a Server enables an AERO interface, it applies the address 459 fe80:: to the interface as a link-local Subnet Router Anycast 460 address, and also applies an administratively assigned link-local 461 address fe80::ID the same as for Relays. (The Server then accepts 462 DHCPv6 and IPv6 ND solicitation messages destined to either the 463 fe80:: or fe80::ID addresses, but always uses fe80::ID as the source 464 address in the replies it generates.) The Server further configures 465 a DHCPv6 server function to facilitate DHCPv6 PD exchanges with AERO 466 Clients. The Server maintains a neighbor cache entry for each Relay 467 on the link, and manages per-Client neighbor cache entries and IP 468 forwarding table entries based on DHCPv6 exchanges. When the Server 469 receives an NS/RS message on the AERO interface it returns an NA/RA 470 message but does not update the neighbor cache. Each Server also 471 engages in a dynamic routing protocol with all Relays on the link. 472 Finally, the Server provides a simple conduit between Clients and 473 Relays, or between Clients and other Clients. Therefore, packets 474 enter the Server's AERO interface from the link layer and are 475 forwarded back out the link layer without ever leaving the AERO 476 interface and therefore without ever disturbing the network layer. 478 When a Client enables an AERO interface, it invokes DHCPv6 PD to 479 receive an ACP from an AERO Server. Next, it applies the 480 corresponding AERO address to the AERO interface and creates a 481 neighbor cache entry for the Server, i.e., the PD exchange bootstraps 482 the provisioning of a unique link-local address. The Client 483 maintains a neighbor cache entry for each of its Servers and each of 484 its active correspondent Clients. When the Client receives Redirect/ 485 Predirect messages on the AERO interface it updates or creates 486 neighbor cache entries, including link-layer address information. 487 Unsolicited NA messages update the cached link-layer addresses for 488 correspondent Clients (e.g., following a link-layer address change 489 due to node mobility) but do not create new neighbor cache entries. 490 NS/NA messages used for Neighbor Unreachability Detection (NUD) 491 update timers in existing neighbor cache entires but do not update 492 link-layer addresses nor create new neighbor cache entries. Finally, 493 the Client need not maintain any IP forwarding table entries for its 494 Servers or correspondent Clients. Instead, it can set a single 495 "route-to-interface" default route in the IP forwarding table 496 pointing to the AERO interface, and all forwarding decisions can be 497 made within the AERO interface based on neighbor cache entries. (On 498 systems in which adding a default route would violate security 499 policy, the default route could instead be installed via a 500 "synthesized RA", e.g., as discussed in Section 3.11.2.) 502 3.4.1. Coordination of Multiple Underlying Interfaces 504 AERO interfaces may be configured over multiple underlying 505 interfaces. For example, common mobile handheld devices have both 506 wireless local area network ("WLAN") and cellular wireless links. 507 These links are typically used "one at a time" with low-cost WLAN 508 preferred and highly-available cellular wireless as a standby. In a 509 more complex example, aircraft frequently have many wireless data 510 link types (e.g. satellite-based, terrestrial, air-to-air 511 directional, etc.) with diverse performance and cost properties. 513 If a Client's multiple underlying interfaces are used "one at a time" 514 (i.e., all other interfaces are in standby mode while one interface 515 is active), then Redirect, Predirect and unsolicited NA messages 516 include only a single TLLAO with Link ID set to a constant value. 518 If the Client has multiple active underlying interfaces, then from 519 the perspective of IPv6 ND it would appear to have a single link- 520 local address with multiple link-layer addresses. In that case, 521 Redirect, Predirect and unsolicited NA messages MAY include multiple 522 TLLAOs -- each with a different Link ID that corresponds to a 523 specific underlying interface of the Client. 525 3.5. AERO Interface Neighbor Cache Maintenace 527 Each AERO interface maintains a conceptual neighbor cache that 528 includes an entry for each neighbor it communicates with on the AERO 529 link, the same as for any IPv6 interface [RFC4861]. AERO interface 530 neighbor cache entires are said to be one of "permanent", "static" or 531 "dynamic". 533 Permanent neighbor cache entries are created through explicit 534 administrative action; they have no timeout values and remain in 535 place until explicitly deleted. AERO Relays maintain a permanent 536 neighbor cache entry for each Server on the link, and AERO Servers 537 maintain a permanent neighbor cache entry for each Relay on the link. 539 Static neighbor cache entries are created though DHCPv6 PD exchanges 540 and remain in place for durations bounded by prefix lifetimes. AERO 541 Servers maintain a static neighbor cache entry for each of their 542 associated Clients, and AERO Clients maintain a static neighbor cache 543 for each of their associated Servers. When an AERO Server sends a 544 DHCPv6 Reply message response to a Client's DHCPv6 Solicit/Request or 545 Renew message, it creates or updates a static neighbor cache entry 546 based on the Client's AERO address as the network-layer address, the 547 prefix lifetime as the neighbor cache entry lifetime, the Client's 548 encapsulation IP address and UDP port number as the link-layer 549 address and the prefix length as the length to apply to the AERO 550 address. When an AERO Client receives a DHCPv6 Reply message from a 551 Server, it creates or updates a static neighbor cache entry based on 552 the Reply message link-local source address as the network-layer 553 address, the prefix lifetime as the neighbor cache entry lifetime, 554 and the encapsulation IP source address and UDP source port number as 555 the link-layer address. 557 Dynamic neighbor cache entries are created based on receipt of an 558 IPv6 ND message, and are garbage-collected if not used within a short 559 timescale. AERO Clients maintain dynamic neighbor cache entries for 560 each of their active correspondent Clients with lifetimes based on 561 IPv6 ND messaging constants. When an AERO Client receives a valid 562 Predirect message it creates or updates a dynamic neighbor cache 563 entry for the Predirect target network-layer and link-layer addresses 564 plus prefix length. The node then sets an "AcceptTime" variable in 565 the neighbor cache entry and uses this value to determine whether 566 packets received from the correspondent can be accepted. When an 567 AERO Client receives a valid Redirect message it creates or updates a 568 dynamic neighbor cache entry for the Redirect target network-layer 569 and link-layer addresses plus prefix length. The Client then sets a 570 "ForwardTime" variable in the neighbor cache entry and uses this 571 value to determine whether packets can be sent directly to the 572 correspondent. The Client also maintains a "MaxRetry" variable to 573 limit the number of keepalives sent when a correspondent may have 574 gone unreachable. 576 For dynamic neighbor cache entries, when an AERO Client receives a 577 valid NS message it (re)sets AcceptTime for the neighbor to 578 ACCEPT_TIME. When an AERO Client receives a valid solicited NA 579 message, it (re)sets ForwardTime for the neighbor to FORWARD_TIME and 580 sets MaxRetry to MAX_RETRY. When an AERO Client receives a valid 581 unsolicited NA message, it updates the correspondent's link-layer 582 addresses but DOES NOT reset AcceptTime, ForwardTime or MaxRetry. 584 It is RECOMMENDED that FORWARD_TIME be set to the default constant 585 value 30 seconds to match the default REACHABLE_TIME value specified 586 for IPv6 ND [RFC4861]. 588 It is RECOMMENDED that ACCEPT_TIME be set to the default constant 589 value 40 seconds to allow a 10 second window so that the AERO 590 redirection procedure can converge before AcceptTime decrements below 591 FORWARD_TIME. 593 It is RECOMMENDED that MAX_RETRY be set to 3 the same as described 594 for IPv6 ND address resolution in Section 7.3.3 of [RFC4861]. 596 Different values for FORWARD_TIME, ACCEPT_TIME, and MAX_RETRY MAY be 597 administratively set, if necessary, to better match the AERO link's 598 performance characteristics; however, if different values are chosen, 599 all nodes on the link MUST consistently configure the same values. 600 Most importantly, ACCEPT_TIME SHOULD be set to a value that is 601 sufficiently longer than FORWARD_TIME to allow the AERO redirection 602 procedure to converge. 604 3.6. AERO Interface Sending Algorithm 606 IP packets enter a node's AERO interface either from the network 607 layer (i.e., from a local application or the IP forwarding system), 608 or from the link layer (i.e., from the AERO tunnel virtual link). 609 Packets that enter the AERO interface from the network layer are 610 encapsulated and admitted into the AERO link (i.e., they are 611 tunnelled to an AERO interface neighbor). Packets that enter the 612 AERO interface from the link layer are either re-admitted into the 613 AERO link or delivered to the network layer where they are subject to 614 either local delivery or IP forwarding. Since each AERO node has 615 only partial information about neighbors on the link, AERO interfaces 616 may forward packets with link-local destination addresses at a layer 617 below the network layer. This means that AERO nodes act as both IP 618 routers and sub-IP layer forwarding agents. AERO interface sending 619 considerations for Clients, Servers and Relays are given below. 621 When an IP packet enters a Client's AERO interface from the network 622 layer, if the destination is covered by an ASP the Client searches 623 for a dynamic neighbor cache entry with a non-zero ForwardTime and an 624 AERO address that matches the packet's destination address. (The 625 destination address may be either an address covered by the 626 neighbor's ACP or the (link-local) AERO address itself.) If there is 627 a match, the Client uses a link-layer address in the entry as the 628 link-layer address for encapsulation then admits the packet into the 629 AERO link. If there is no match, the Client instead uses the link- 630 layer address of a neighboring Server as the link-layer address for 631 encapsulation. 633 When an IP packet enters a Server's AERO interface from the link 634 layer, if the destination is covered by an ASP the Server searches 635 for a static neighbor cache entry with an AERO address that matches 636 the packet's destination address. (The destination address may be 637 either an address covered by the neighbor's ACP or the AERO address 638 itself.) If there is a match, the Server uses a link-layer address 639 in the entry as the link-layer address for encapsulation and re- 640 admits the packet into the AERO link. If there is no match, the 641 Server instead uses the link-layer address in any permanent neighbor 642 cache entry as the link-layer address for encapsulation. 644 When an IP packet enters a Relay's AERO interface from the network 645 layer, the Relay searches its IP forwarding table for an entry that 646 is covered by an ASP and also matches the destination. If there is a 647 match, the Relay uses the link-layer address in the neighbor cache 648 entry for the next-hop Server as the link-layer address for 649 encapsulation and admits the packet into the AERO link. When an IP 650 packet enters a Relay's AERO interface from the link-layer, if the 651 destination is not a link-local address and is not covered by an ASP 652 the Relay removes the packet from the AERO interface and uses IP 653 forwarding to forward the packet to the Internetwork. If the 654 destination address is covered by an ASP, and there is a more- 655 specific IP forwarding table entry that matches the destination, the 656 Relay uses the link-layer address in the neighbor cache entry for the 657 next-hop Server as the link-layer address for encapsulation and re- 658 admits the packet into the AERO link. When an IP packet enters a 659 Relay's AERO interface from either the network layer or link-layer, 660 and the packet's destination address matches an ASP but there is no 661 more-specific ACP entry, the Relay drops the packet and returns an 662 ICMP Destination Unreachable message (see: Section 3.10). 664 When an AERO Server receives a packet from a Relay via the AERO 665 interface, the Server MUST NOT forward the packet back to the same or 666 a different Relay. 668 When an AERO Relay receives a packet from a Server via the AERO 669 interface, the Relay MUST NOT forward the packet back to the same 670 Server. 672 When an AERO node re-admits a packet into the AERO link without 673 involving the network layer, the node MUST NOT decrement the network 674 layer TTL/Hop-count. 676 Note that in the above that the link-layer address for encapsulation 677 may be determined through consulting either the neighbor cache or the 678 IP forwarding table. IP forwarding is therefore linked to IPv6 ND 679 via the AERO address. 681 3.7. AERO Interface Encapsulation, Re-encapsulation and Decapsulation 683 AERO interfaces encapsulate IP packets according to whether they are 684 entering the AERO interface from the network layer or if they are 685 being re-admitted into the same AERO link they arrived on. This 686 latter form of encapsulation is known as "re-encapsulation". 688 AERO interfaces encapsulate packets per the base tunneling 689 specifications (e.g., [RFC2003][RFC2473][RFC4213][RFC4301][RFC5246], 690 etc.) except that the interface copies the "TTL/Hop Limit", "Type of 691 Service/Traffic Class" and "Congestion Experienced" values in the 692 packet's IP header into the corresponding fields in the encapsulation 693 header. For packets undergoing re-encapsulation, the AERO interface 694 instead copies the "TTL/Hop Limit", "Type of Service/Traffic Class" 695 and "Congestion Experienced" values in the original encapsulation 696 header into the corresponding fields in the new encapsulation header 697 (i.e., the values are transferred between encapsulation headers and 698 *not* copied from the encapsulated packet's network-layer header). 700 When AERO UDP encapsulation is used, the AERO interface encapsulates 701 the packet per the base tunneling specification except that it 702 inserts a UDP header between the encapsulation header and the 703 packet's IP header. The AERO interface sets the UDP source port to a 704 constant value that it will use in each successive packet it sends 705 and sets the UDP length field to the length of the IP packet plus 8 706 bytes for the UDP header itself. For packets sent via a Server, the 707 AERO interface sets the UDP destination port to 8060 (i.e., the IANA- 708 registered port number for AERO) when AERO-only encapsulation is 709 used. For packets sent to a correspondent Client, the AERO interface 710 sets the UDP destination port to the port value stored in the 711 neighbor cache entry for this correspondent. 713 The AERO interface also sets the UDP checksum field to zero (see: 714 [RFC6935][RFC6936]) for packets that do not require assurance against 715 reassembly errors. For packets that require reassembly checks (see 716 Section 3.9), the AERO interface instead (re)calculates the UDP 717 checksum and writes the resulting value in the UDP checksum field. 719 The AERO interface next sets the IP protocol number in the 720 encapsulation header to the appropriate value for the first protocol 721 layer within the encapsulation (e.g., IPv4, IPv6, UDP, IPsec, etc.). 722 When IPv6 is used as the encapsulation protocol, the interface then 723 sets the flow label value in the encapsulation header the same as 724 described in [RFC6438]. When IPv4 is used as the encapsulation 725 protocol, the AERO interface sets the DF bit as discussed in 726 Section 3.9. 728 AERO interfaces decapsulate packets destined either to the node 729 itself or to a destination reached via an interface other than the 730 AERO interface the packet was received on. When AERO UDP 731 encapsulation is used (i.e., when a UDP header with destination port 732 8060 is present) the interface first verifies the UDP checksum if the 733 UDP checksum was non-zero, then examines the first octet of the 734 encapsulated packet. The packet is accepted if the most significant 735 four bits of the first octet encode the value '0110' (i.e., the 736 version number value for IPv6) or the value '0100' (i.e., the version 737 number value for IPv4). Otherwise, the packet is accepted if the 738 first octet encodes a valid IP protocol number per the IANA 739 "protocol-numbers" registry that matches a supported encapsulation 740 type. Otherwise, the packet is discarded. 742 Further decapsulation then proceeds according to the appropriate base 743 tunneling specification. 745 3.8. AERO Interface Data Origin Authentication 747 AERO nodes employ simple data origin authentication procedures for 748 encapsulated packets they receive from other nodes on the AERO link. 749 In particular: 751 o AERO Relays and Servers accept encapsulated packets with a link- 752 layer source address that matches a permanent neighbor cache 753 entry. 755 o AERO Servers accept authentic encapsulated DHCPv6 messages, and 756 create or update a static neighbor cache entry for the source 757 based on the specific message type. 759 o AERO Servers accept encapsulated packets if there is a static 760 neighbor cache entry with an AERO address that matches the 761 packet's network-layer source address and with a link-layer 762 address that matches the packet's link-layer source address. 764 o AERO Clients accept encapsulated packets if there is a static 765 neighbor cache entry with a link-layer source address that matches 766 the packet's link-layer source address. 768 o AERO Clients and Servers accept encapsulated packets if there is a 769 dynamic neighbor cache entry with an AERO address that matches the 770 packet's network-layer source address, with a link-layer address 771 that matches the packet's link-layer source address, and with a 772 non-zero AcceptTime. 774 Note that this simple data origin authentication only applies to 775 environments in which link-layer addresses cannot be spoofed. 776 Additional security mitigations may be necessary in other 777 environments. 779 3.9. AERO Interface MTU Considerations 781 The AERO interface is the node's point of attachment to the AERO 782 link. AERO links over IP networks have a maximum link MTU of 64KB 783 minus the encapsulation overhead (i.e., "64KB-ENCAPS"), since the 784 maximum packet size in the base IP specifications is 64KB 785 [RFC0791][RFC2460]. AERO links over IPv6 networks have a theoretical 786 maximum link MTU of 4GB-ENCAPS [RFC2675], however IPv6 Jumbograms are 787 considered optional for IPv6 nodes [RFC6434] and therefore out of 788 scope for this document. 790 The IP layer sees the AERO interface as an ordinary interface that 791 configures an MTU that is no larger than the link MTU, i.e., the same 792 as for any interface. Routers MAY set an AERO interface MTU up to 793 the maximum link MTU. Hosts SHOULD set a more conservative MTU so 794 that upper layer protocols will see an appropriate maximum packet 795 size, for example when setting an initial TCP Maximum Segment Size 796 (MSS). In all cases, routers and hosts MUST set an MTU of at least 797 1500 bytes. 799 IPv6 specifies a minimum link MTU of 1280 bytes [RFC2460]. This is 800 the minimum packet size an AERO interface MUST be capable of 801 forwarding without returning an ICMP Packet Too Big (PTB) message. 802 Although IPv4 specifies a smaller minimum link MTU of 68 bytes 803 [RFC0791], AERO interfaces also observe a 1280 byte minimum for IPv4. 804 Additionally, the vast majority of links in the Internet configure an 805 MTU of at least 1500 bytes. Hosts have therefore become conditioned 806 to expect that IP packets up to 1500 bytes in length will either be 807 delivered to the final destination or a suitable ICMP Packet Too Big 808 (PTB) message returned, however such PTB messages are often lost 809 [RFC2923]. Therefore, AERO interfaces MUST pass IP packets of at 810 least 1500 bytes even if the encapsulated packet must be fragmented. 812 PTB messages may be generated by the IP layer of the AERO node if the 813 packet is too large to enter the AERO interface, from within the AERO 814 interface itself if the packet is larger than 1500 bytes and also 815 larger than the MTU of the underlying interface to be used for 816 tunneling minus ENCAPS, or from a router within the AERO link (i.e., 817 the "tunnel") after the encapsulated packet has been admitted into 818 the tunnel. The latter condition would result in a link-layer (L2) 819 PTB message delivered to the AERO interface, while the former two 820 conditions would result in a network-layer (L3) PTB message delivered 821 to the original source. 823 For AERO links over IPv4, the IP ID field is only 16 bits in length, 824 meaning that fragmentation at high data rates could result in 825 dangerous reassembly misassociations [RFC6864][RFC4963]. For this 826 reason, AERO interfaces that send fragmented IPv4-encapsulated 827 packets MUST either institute rate limiting to ensure that the IP ID 828 field will not wrap before all earlier fragments have been processed, 829 or include an integrity check to detect reassembly errors. 831 The AERO interface therefore admits encapsulated packets into the 832 tunnel (using fragmentation as necessary) as follows: 834 o For IP packets that are no larger than (1280-ENCAPS) bytes, the 835 AERO interface admits the packet into the tunnel without 836 fragmentation. For IPv4 AERO links, the AERO interface sets the 837 Don't Fragment (DF) bit to 0 so that these packets will be 838 deterministically delivered even if there is a restricting link in 839 the path. The AERO interface need not perform rate limiting or 840 include integrity checks for these packets, since any IPv4 links 841 in the path that configure an MTU smaller than 1280 bytes are very 842 likely to be slow links [RFC3819]. 844 o For IP packets that are larger than (1280-ENCAPS) bytes but no 845 larger than 1500 bytes, the AERO interface encapsulates the 846 packet. (For IPv4 AERO links, the AERO interface then sets the DF 847 bit to 0 and calculates the UDP checksum for the encapsulated 848 packet as an integrity check to account for the potential for 849 reassembly misassociations. If the encapsulation does not include 850 a UDP header or other integrity check, the AERO interface instead 851 MUST institute rate limiting.) Next, the AERO interface uses IP 852 fragmentation to fragment the encapsulated packet into two 853 fragments where the first fragment is no larger than 1024 bytes 854 and the other fragment is no larger than the first fragment. The 855 AERO interface then admits both fragments into the tunnel. 857 o For IPv4 packets that are larger than 1500 bytes and with the DF 858 bit set to 0, the AERO interface fragments the unencapsulated 859 packet into a minimum number of fragments where the first fragment 860 is no larger than 1024 bytes and all other fragments are no larger 861 than the first fragment. The AERO interface then encapsulates 862 each fragment (and for IPv4 sets the DF bit to 0) and admits the 863 fragments into the tunnel. These encapsulated fragments will be 864 deterministically delivered to the final destination. (The AERO 865 interface need not perform rate limiting or include integrity 866 checks for these packets since it is not the original source of 867 the unencapsulated packet.) 869 o For all other IP packets, if the packet is larger than the AERO 870 interface MTU the AERO node drops the packet and returns an L3 PTB 871 message with MTU set to the AERO interface MTU; otherwise, the 872 node admits the packet into the AERO interface. Next, if the 873 packet length is larger than the MTU of the underlying interface 874 to be used for tunneling minus ENCAPS, the AERO interface drops 875 the packet and returns an L3 PTB message with MTU set to the 876 larger of 1500 or the underlying interface MTU minus ENCAPS. 877 Otherwise, the AERO interface encapsulates the packet and admits 878 it into the tunnel without fragmentation (and for IPv4 sets the DF 879 bit to 1) and translates any L2 PTB messages it may receive from 880 the network into corresponding L3 PTB messages to send to the 881 original source as specified in Section 3.10. Since both L2 and 882 L3 PTB messages may be either lost or contain insufficient 883 information, however, it is RECOMMENDED that sources that send 884 unfragmentable IP packets larger than 1500 bytes use Packetization 885 Layer Path MTU Discovery (PLPMTUD) [RFC4821]. 887 While sending packets according to the above specifications, the AERO 888 interface (i.e., the tunnel ingress) MAY also send 1500 byte probe 889 packets to the tunnel egress to determine whether the probes can 890 traverse the tunnel without fragmentation. If the probes succeed, 891 the tunnel ingress can begin sending packets that are larger than 892 1280-ENCAPS bytes but no larger than 1500 bytes without fragmentation 893 (and for IPv4 with DF set to 1). Since the path MTU within the 894 tunnel may fluctuate due to routing changes, the tunnel ingress 895 SHOULD continually send additional probes subject to rate limiting in 896 case L2 PTB messages are lost. If the path MTU within the tunnel 897 later becomes insufficient, the tunnel ingress must resume 898 fragmentation. 900 To construct a probe, the tunnel ingress prepares an NS message with 901 a Nonce option plus trailing padding octets added to a length of 1500 902 bytes without including the length of the padding in the IPv6 Payload 903 Length field. The tunnel ingress then encapsulates the padded NS 904 message in the encapsulation headers (and for IPv4 sets DF to 1) then 905 sends the message to the tunnel egress. If the tunnel egress returns 906 a solicited NA message with a matching Nonce option, the tunnel 907 ingress deems the probe successful.. Note that the tunnel ingress 908 SHOULD NOT include the trailing padding within the Nonce option 909 itself but rather as padding beyond the last option in the NS 910 message; otherwise, the (large) Nonce option would be echoed back in 911 the solicited NA message and may be lost at a link with a small MTU 912 along the reverse path. 914 In light of the above fragmentation and reassembly recommendations, 915 the tunnel egress MUST be capable of reassembling encapsulated 916 packets up to 1500+ENCAPS bytes in length. It is therefore 917 RECOMMENDED that the tunnel egress be capable of reassembling at 918 least 2KB. Also, in some environments there may be operational 919 assurance that all links within the routing region spanned by the 920 tunnel configure sufficiently large MTUs so that fragmentation and 921 reassembly can be avoided. In those cases, specific tunnel 922 specifications must explain the circumstances under which the above 923 fragmentation and reassembly recommendations need not be applied. 925 Of possible concern is that some network middleboxes hold the 926 fragments of a fragmented UDP packet until all fragments have arrived 927 before forwarding the fragments to the final destination. This means 928 that the network middlebox must also be able to accommodate 929 fragmented UDP packets up to 1500+ENCAPS bytes in length, i.e., and 930 not just the IP protocol minimum reassembly size. However, network 931 middleboxes are already capable of passing fragmented UDP datagrams 932 up to the maximum fragmented IP packet size as evidenced through 933 actual operational experience (see the thread "PMTUD issue 934 discussion" in the IETF v6ops archive dated September 10, 2014). 935 Hence, there is no need for AERO to stipulate a minimum reassembly 936 size for such devices. 938 3.10. AERO Interface Error Handling 940 When an AERO node admits encapsulated packets into the AERO 941 interface, it may receive link-layer (L2) or network-layer (L3) error 942 indications. 944 An L2 error indication is an ICMP error message generated by a router 945 on the path to the neighbor or by the neighbor itself. The message 946 includes an IP header with the address of the node that generated the 947 error as the source address and with the link-layer address of the 948 AERO node as the destination address. 950 The IP header is followed by an ICMP header that includes an error 951 Type, Code and Checksum. For ICMPv6 [RFC4443], the error Types 952 include "Destination Unreachable", "Packet Too Big (PTB)", "Time 953 Exceeded" and "Parameter Problem". For ICMPv4 [RFC0792], the error 954 Types include "Destination Unreachable", "Fragmentation Needed" (a 955 Destination Unreachable Code that is analogous to the ICMPv6 PTB), 956 "Time Exceeded" and "Parameter Problem". 958 The ICMP header is followed by the leading portion of the packet that 959 generated the error, also known as the "packet-in-error". For 960 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 961 much of invoking packet as possible without the ICMPv6 packet 962 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 963 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 964 "Internet Header + 64 bits of Original Data Datagram", however 965 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 966 ICMP datagram SHOULD contain as much of the original datagram as 967 possible without the length of the ICMP datagram exceeding 576 968 bytes". 970 The L2 error message format is shown in Figure 3: 972 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 973 ~ ~ 974 | L2 IP Header of | 975 | error message | 976 ~ ~ 977 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 978 | L2 ICMP Header | 979 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 980 ~ ~ P 981 | IP and other encapsulation | a 982 | headers of original L3 packet | c 983 ~ ~ k 984 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 985 ~ ~ t 986 | IP header of | 987 | original L3 packet | i 988 ~ ~ n 989 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 990 ~ ~ e 991 | Upper layer headers and | r 992 | leading portion of body | r 993 | of the original L3 packet | o 994 ~ ~ r 995 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 997 Figure 3: AERO Interface L2 Error Message Format 999 The AERO node rules for processing these L2 error messages is as 1000 follows: 1002 o When an AERO node receives an L2 "Parameter Problem", it processes 1003 the message the same as described as for ordinary ICMP errors in 1004 the normative references [RFC0792][RFC4443]. 1006 o When an AERO node receives persistent L2 Time Exceeded messages, 1007 it SHOULD reduce its current rate of admitting fragmented 1008 encapsulated packets into the tunnel to ensure that the IP ID 1009 field will not wrap before all earlier fragments have been 1010 processed. If the AERO node includes an integrity check vector, 1011 however, it MAY ignore the messages and continue sending 1012 fragmented encapsulated packets without rate limiting. 1014 o When an AERO Client receives persistent L2 Destination Unreachable 1015 messages in response to tunneled packets that it sends to one of 1016 its dynamic neighbor correspondents, the Client SHOULD test the 1017 path to the correspondent using Neighbor Unreachability Detection 1018 (NUD) (see Section 3.14). If NUD fails, the Client SHOULD set 1019 ForwardTime for the corresponding dynamic neighbor cache entry to 1020 0 and allow future packets destined to the correspondent to flow 1021 through a Server. 1023 o When an AERO Client receives persistent L2 Destination Unreachable 1024 messages in response to tunneled packets that it sends to one of 1025 its static neighbor Servers, the Client SHOULD test the path to 1026 the Server using NUD. If NUD fails, the Client SHOULD delete the 1027 neighbor cache entry and attempt to associate with a new Server. 1029 o When an AERO Server receives persistent L2 Destination Unreachable 1030 messages in response to tunneled packets that it sends to one of 1031 its static neighbor Clients, the Server SHOULD test the path to 1032 the Client using NUD. If NUD fails, the Server SHOULD cancel the 1033 DHCPv6 PD lease for the Client's ACP, withdraw its route for the 1034 ACP from the AERO routing system and delete the neighbor cache 1035 entry (see Sections 3.11 and 3.12). 1037 o When an AERO Relay or Server receives an L2 Destination 1038 Unreachable message in response to a tunneled packet that it sends 1039 to one of its permanent neighbors, it discards the message since 1040 the routing system is likely in a temporary transitional state 1041 that will soon re-converge. 1043 o When an AERO node receives an L2 PTB message, it translates the 1044 message into an L3 PTB message if possible (*) and forwards the 1045 message toward the original source as described below. 1047 To translate an L2 PTB message to an L3 PTB message, the AERO node 1048 first caches the MTU field value of the L2 ICMP header. The node 1049 next discards the L2 IP and ICMP headers, and also discards the 1050 encapsulation headers of the original L3 packet. Next the node 1051 encapsulates the included segment of the original L3 packet in an L3 1052 IP and ICMP header, and sets the ICMP header Type and Code values to 1053 appropriate values for the L3 IP protocol. In the process, the node 1054 writes the maximum of 1500 bytes and (L2 MTU - ENCAPS) into the MTU 1055 field of the L3 ICMP header. 1057 The node next writes the IP source address of the original L3 packet 1058 as the destination address of the L3 PTB message and determines the 1059 next hop to the destination. If the next hop is reached via the AERO 1060 interface, the node uses the IPv6 address "::" or the IPv4 address 1061 "0.0.0.0" as the IP source address of the L3 PTB message. Otherwise, 1062 the node uses one of its non link-local addresses as the source 1063 address of the L3 PTB message. The node finally calculates the ICMP 1064 checksum over the L3 PTB message and writes the Checksum in the 1065 corresponding field of the L3 ICMP header. The L3 PTB message 1066 therefore is formatted as follows: 1068 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1069 ~ ~ 1070 | L3 IP Header of | 1071 | error message | 1072 ~ ~ 1073 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1074 | L3 ICMP Header | 1075 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1076 ~ ~ p 1077 | IP header of | k 1078 | original L3 packet | t 1079 ~ ~ 1080 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ i 1081 ~ ~ n 1082 | Upper layer headers and | 1083 | leading portion of body | e 1084 | of the original L3 packet | r 1085 ~ ~ r 1086 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1088 Figure 4: AERO Interface L3 Error Message Format 1090 After the node has prepared the L3 PTB message, it either forwards 1091 the message via a link outside of the AERO interface without 1092 encapsulation, or encapsulates and forwards the message to the next 1093 hop via the AERO interface. 1095 When an AERO Relay receives an L3 packet for which the destination 1096 address is covered by an ASP, if there is no more-specific routing 1097 information for the destination the Relay drops the packet and 1098 returns an L3 Destination Unreachable message. The Relay first 1099 writes the IP source address of the original L3 packet as the 1100 destination address of the L3 Destination Unreachable message and 1101 determines the next hop to the destination. If the next hop is 1102 reached via the AERO interface, the Relay uses the IPv6 address "::" 1103 or the IPv4 address "0.0.0.0" as the IP source address of the L3 1104 Destination Unreachable message and forwards the message to the next 1105 hop within the AERO interface. Otherwise, the Relay uses one of its 1106 non link-local addresses as the source address of the L3 Destination 1107 Unreachable message and forwards the message via a link outside the 1108 AERO interface. 1110 When an AERO node receives any L3 error message via the AERO 1111 interface, it examines the destination address in the L3 IP header of 1112 the message. If the next hop toward the destination address of the 1113 error message is via the AERO interface, the node re-encapsulates and 1114 forwards the message to the next hop within the AERO interface. 1115 Otherwise, if the source address in the L3 IP header of the message 1116 is the IPv6 address "::" or the IPv4 address "0.0.0.0", the node 1117 writes one of its non link-local addresses as the source address of 1118 the L3 message and recalculates the IP and/or ICMP checksums. The 1119 node finally forwards the message via a link outside of the AERO 1120 interface. 1122 (*) Note that in some instances the packet-in-error field of an L2 1123 PTB message may not include enough information for translation to an 1124 L3 PTB message. In that case, the AERO interface simply discards the 1125 L2 PTB message. It can therefore be said that translation of L2 PTB 1126 messages to L3 PTB messages can provide a useful optimization when 1127 possible, but is not critical for sources that correctly use PLPMTUD. 1129 3.11. AERO Router Discovery, Prefix Delegation and Address 1130 Configuration 1132 3.11.1. AERO DHCPv6 Service Model 1134 Each AERO Server configures a DHCPv6 server function to facilitate PD 1135 requests from Clients. Each Server is pre-configured with an 1136 identical list of ACP-to-Client ID mappings for all Clients enrolled 1137 in the AERO system, as well as any information necessary to 1138 authenticate Clients. The configuration information is maintained by 1139 a central administrative authority for the AERO link and securely 1140 propagated to all Servers whenever a new Client is enrolled or an 1141 existing Client is withdrawn. 1143 With these identical configurations, each Server can function 1144 independently of all other Servers, including the maintenance of 1145 active leases. Therefore, no Server-to-Server DHCPv6 state 1146 synchronization is necessary, and Clients can optionally hold 1147 separate leases for the same ACP from multiple Servers. 1149 In this way, Clients can easily associate with multiple Servers, and 1150 can receive new leases from new Servers before deprecating leases 1151 held through old Servers. This enables a graceful "make-before- 1152 break" capability. 1154 3.11.2. AERO Client Behavior 1156 AERO Clients discover the link-layer addresses of AERO Servers via 1157 static configuration, or through an automated means such as DNS name 1158 resolution. In the absence of other information, the Client resolves 1159 the Fully-Qualified Domain Name (FQDN) "linkupnetworks.[domainname]" 1160 where "linkupnetworks" is a constant text string and "[domainname]" 1161 is the connection-specific DNS suffix for the Client's underlying 1162 network connection (e.g., "example.com"). After discovering the 1163 link-layer addresses, the Client associates with one or more of the 1164 corresponding Servers. 1166 To associate with a Server, the Client acts as a requesting router to 1167 request an ACP through a DHCPv6 PD exchange[RFC3315][RFC3633] in 1168 which the Client's Solicit/Request messages use the IPv6 1169 "unspecified" address (i.e., "::") as the IPv6 source address, 1170 'All_DHCP_Relay_Agents_and_Servers' as the IPv6 destination address 1171 and the link-layer address of the Server as the link-layer 1172 destination address. The Client also includes a Client Identifier 1173 option with a DHCP Unique Identifier (DUID) plus any necessary 1174 authentication options to identify itself to the DHCPv6 server, and 1175 includes a Client Link Layer Address Option (CLLAO) [RFC6939] with 1176 the format shown in Figure 5: 1178 0 1 2 3 1179 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 1180 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1181 | OPTION_CLIENT_LINKLAYER_ADDR | option-length | 1182 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1183 | link-layer type (16 bits) | Link ID | Preference | 1184 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1186 Figure 5: AERO Client Link-Layer Address Option (CLLAO) Format 1188 The Client sets the CLLAO 'option-length' field to 4 and sets the 1189 'link-layer type' field to TBD1 (see: IANA Considerations), then 1190 includes appropriate Link ID and Preference values for the underlying 1191 interface over which the Solicit/Request will be issued (note that 1192 these are the same values that would be included in a TLLAO as shown 1193 in Figure 2). If the Client is pre-provisioned with an ACP 1194 associated with the AERO service, it MAY also include the ACP in the 1195 Solicit/Request message Identity Association (IA) option to indicate 1196 its preferred ACP to the DHCPv6 server. The Client then sends the 1197 encapsulated DHCPv6 request via the underlying interface. 1199 When the Client receives its ACP and the set of ASPs via a DHCPv6 1200 Reply from the AERO Server, it creates a static neighbor cache entry 1201 with the Server's link-local address as the network-layer address and 1202 the Server's encapsulation address as the link-layer address. The 1203 Client then records the lifetime for the ACP in the neighbor cache 1204 entry and marks the neighbor cache entry as "default", i.e., the 1205 Client considers the Server as a default router. If the Reply 1206 message contains a Vendor-Specific Information Option (see: 1207 Section 3.10.3) the Client also caches each ASP in the option. 1209 The Client then applies the AERO address to the AERO interface and 1210 sub-delegates the ACP to nodes and links within its attached EUNs 1211 (the AERO address thereafter remains stable as the Client moves). 1212 The Client also assigns a default IP route to the AERO interface as a 1213 route-to-interface, i.e., with no explicit next-hop. The next hop 1214 will then be determined after a packet has been submitted to the AERO 1215 interface by inspecting the neighbor cache (see above). 1217 On some platforms (e.g., popular cell phone operating systems), the 1218 act of assigning a default IPv6 route to the AERO interface may not 1219 be permitted from a user application due to security policy. 1220 Typically, those platforms include a TUN/TAP interface that acts as a 1221 point-to-point conduit between user applications and the AERO 1222 interface. In that case, the Client can instead generate a 1223 "synthesized RA" message. The message conforms to [RFC4861] and is 1224 prepared as follows: 1226 o the IPv6 source address is fe80:: 1228 o the IPv6 destination address is all-nodes multicast 1230 o the Router Lifetime is set to a time that is no longer than the 1231 ACP DHCPv6 lifetime 1233 o the message does not include a Source Link Layer Address Option 1234 (SLLAO) 1236 o the message includes a Prefix Information Option (PIO) with a /64 1237 prefix taken from the ACP as the prefix for autoconfiguration 1239 The Client then sends the synthesized RA message via the TUN/TAP 1240 interface, where the operating system kernel will interpret it as 1241 though it were generated by an actual router. The operating system 1242 will then install a default route and use StateLess Address 1243 AutoConfiguration (SLAAC) to configure an IPv6 address on the TUN/TAP 1244 interface. Methods for similarly installing an IPv4 default route 1245 and IPv4 address on the TUN/TAP interface are based on synthesized 1246 DHCPv4 messages [RFC2131]. Note that in this method, the Client 1247 appears as a mobility proxy for applications that bind to the (point- 1248 to-point) TUN/TAP interface. The arrangement can be likened to a 1249 Proxy AERO scenario in which the mobile node and Client are located 1250 within the same physical platform (see Section 3.20 for further 1251 details on Proxy AERO). 1253 The Client subsequently renews its ACP delegation through each of its 1254 Servers by performing DHCPv6 Renew/Reply exchanges with its AERO 1255 address as the IPv6 source address, 1256 'All_DHCP_Relay_Agents_and_Servers' as the IPv6 destination address, 1257 the link-layer address of a Server as the link-layer destination 1258 address and the same Client identifier, authentication options and 1259 CLLAO option as was used in the initial PD request. Note that if the 1260 Client does not issue a DHCPv6 Renew before the Server has terminated 1261 the lease (e.g., if the Client has been out of touch with the Server 1262 for a considerable amount of time), the Server's Reply will report 1263 NoBinding and the Client must re-initiate the DHCPv6 PD procedure. 1264 If the Client sends synthesized RA and/or DHCPv4 messages (see 1265 above), it also sends a new synthesized message when issuing a DHCPv6 1266 Renew or when re-initiating the DHCPv6 PD procedure. 1268 Since the Client's AERO address is configured from the unique ACP 1269 delegation it receives, there is no need for Duplicate Address 1270 Detection (DAD) on AERO links. Other nodes maliciously attempting to 1271 hijack an authorized Client's AERO address will be denied access to 1272 the network by the DHCPv6 server due to an unacceptable link-layer 1273 address and/or security parameters (see: Security Considerations). 1275 AERO Clients ignore the IP address and UDP port number in any S/TLLAO 1276 options in ND messages they receive directly from another AERO 1277 Client, but examine the Link ID and Preference values to match the 1278 message with the correct link-layer address information. 1280 When a source Client forwards a packet to a prospective destination 1281 Client (i.e., one for which the packet's destination address is 1282 covered by an ASP), the source Client initiates an AERO route 1283 optimization procedure as specified in Section 3.13. 1285 3.11.3. AERO Server Behavior 1287 AERO Servers configure a DHCPv6 server function on their AERO links. 1288 AERO Servers arrange to add their encapsulation layer IP addresses 1289 (i.e., their link-layer addresses) to the DNS resource records for 1290 the FQDN "linkupnetworks.[domainname]" before entering service. 1292 When an AERO Server receives a prospective Client's DHCPv6 PD 1293 Solicit/Request message, it first authenticates the message. If 1294 authentication succeeds, the Server determines the correct ACP to 1295 delegate to the Client by matching the Client's DUID within an online 1296 directory service (e.g., LDAP). The Server then delegates the ACP 1297 and creates a static neighbor cache entry for the Client's AERO 1298 address with lifetime set to no more than the lease lifetime and the 1299 Client's link-layer address as the link-layer address for the Link ID 1300 specified in the CLLAO option. The Server then creates an IP 1301 forwarding table entry so that the AERO routing system will propagate 1302 the ACP to all Relays (see: Section 3.12). Finally, the Server sends 1303 a DHCPv6 Reply message to the Client while using fe80::ID as the IPv6 1304 source address, the Client's AERO address as the IPv6 destination 1305 address, and the Client's link-layer address as the destination link- 1306 layer address. The Server also includes a Server Unicast option with 1307 server-address set to fe80::ID so that all future Client/Server 1308 transactions will be link-local-only unicast over the AERO link. 1310 When the Server sends the DHCPv6 Reply message, it also includes a 1311 DHCPv6 Vendor-Specific Information Option with 'enterprise-number' 1312 set to "TBD2" (see: IANA Considerations). The option is formatted as 1313 shown in[RFC3315] and with the AERO enterprise-specific format shown 1314 in Figure 6: 1316 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 1317 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1318 | OPTION_VENDOR_OPTS | option-len | 1319 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1320 | enterprise-number ("TBD2") | 1321 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1322 | Reserved | Prefix Length | 1323 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1324 | | 1325 + ASP (1) + 1326 | | 1327 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1328 | Reserved | Prefix Length | 1329 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1330 | | 1331 + ASP (2) + 1332 | | 1333 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1334 | Reserved | Prefix Length | 1335 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1336 | | 1337 + ASP (3) + 1338 | | 1339 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1340 . (etc.) . 1341 . . 1342 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1344 Figure 6: AERO Vendor-Specific Information Option 1346 Per Figure 6, the option includes one or more ASP. The ASP field 1347 contains the IP prefix as it would appear in the interface identifier 1348 portion of the corresponding AERO address (see: Section 3.3). For 1349 IPv6, valid values for the Prefix Length field are 0 through 64; for 1350 IPv4, valid values are 0 through 32. 1352 After the initial DHCPv6 PD exchange, the AERO Server maintains the 1353 neighbor cache entry for the Client as long as the lease lifetime 1354 remains current. If the Client issues a Renew/Reply exchange, the 1355 Server extends the lifetime. If the Client issues a Release/Reply 1356 exchange, or if the Client does not issue a Renew/Reply within the 1357 lease lifetime, the Server deletes the neighbor cache entry for the 1358 Client and withdraws the IP route from the AERO routing system. 1360 3.12. AERO Relay/Server Routing System 1362 Relays require full topology information of all Client/Server 1363 associations, while individual Servers only require partial topology 1364 information, i.e., they only need to know the ACPs associated with 1365 their current set of associated Clients. This is accomplished 1366 through the use of an internal instance of the Border Gateway 1367 Protocol (BGP) [RFC4271] coordinated between Servers and Relays. 1368 This internal BGP instance does not interact with the public Internet 1369 BGP instance; therefore, the AERO link is presented to the IP 1370 Internetwork as a small set of ASPs as opposed to the full set of 1371 individual ACPs. 1373 In a reference BGP arrangement, each AERO Server is configured as an 1374 Autonomous System Border Router (ASBR) for a stub Autonomous System 1375 (AS) (possibly using a private AS Number (ASN) [RFC1930]), and each 1376 Server further peers with each Relay but does not peer with other 1377 Servers. Similarly, Relays need not peer with each other, since they 1378 will receive all updates from all Servers and will therefore have a 1379 consistent view of the AERO link ACP delegations. 1381 Each Server maintains a working set of associated Clients, and 1382 dynamically announces new ACPs and withdraws departed ACPs in its BGP 1383 updates to Relays (this is typically accomplished via a "redistribute 1384 static" routing directive). Relays do not send BGP updates to 1385 Servers, however, such that the BGP route reporting is unidirectional 1386 from the Servers to the Relays. 1388 The Relays therefore discover the full topology of the AERO link in 1389 terms of the working set of ACPs associated with each Server, while 1390 the Servers only discover the ACPs of their associated Clients. 1391 Since Clients are expected to remain associated with their current 1392 set of Servers for extended timeframes, the amount of BGP control 1393 messaging between Servers and Relays should be minimal. However, BGP 1394 peers SHOULD dampen any route oscillations caused by impatient 1395 Clients that repeatedly associate and disassociate with Servers. 1397 3.13. AERO Redirection 1398 3.13.1. Reference Operational Scenario 1400 Figure 7 depicts the AERO redirection reference operational scenario, 1401 using IPv6 addressing as the example (while not shown, a 1402 corresponding example for IPv4 addressing can be easily constructed). 1403 The figure shows an AERO Relay ('R1'), two AERO Servers ('S1', 'S2'), 1404 two AERO Clients ('C1', 'C2') and two ordinary IPv6 hosts ('H1', 1405 'H2'): 1407 +--------------+ +--------------+ +--------------+ 1408 | Server S1 | | Relay R1 | | Server S2 | 1409 +--------------+ +--------------+ +--------------+ 1410 fe80::2 fe80::1 fe80::3 1411 L2(S1) L2(R1) L2(S2) 1412 | | | 1413 X-----+-----+------------------+-----------------+----+----X 1414 | AERO Link | 1415 L2(A) L2(B) 1416 fe80::2001:db8:0:0 fe80::2001:db8:1:0 1417 +--------------+ +--------------+ 1418 |AERO Client C1| |AERO Client C2| 1419 +--------------+ +--------------+ 1420 2001:DB8:0::/48 2001:DB8:1::/48 1421 | | 1422 .-. .-. 1423 ,-( _)-. 2001:db8:0::1 2001:db8:1::1 ,-( _)-. 1424 .-(_ IP )-. +---------+ +---------+ .-(_ IP )-. 1425 (__ EUN )--| Host H1 | | Host H2 |--(__ EUN ) 1426 `-(______)-' +---------+ +---------+ `-(______)-' 1428 Figure 7: AERO Reference Operational Scenario 1430 In Figure 7, Relay ('R1') applies the address fe80::1 to its AERO 1431 interface with link-layer address L2(R1), Server ('S1') applies the 1432 address fe80::2 with link-layer address L2(S1),and Server ('S2') 1433 applies the address fe80::3 with link-layer address L2(S2). Servers 1434 ('S1') and ('S2') next arrange to add their link-layer addresses to a 1435 published list of valid Servers for the AERO link. 1437 AERO Client ('C1') receives the ACP 2001:db8:0::/48 in a DHCPv6 PD 1438 exchange via AERO Server ('S1') then applies the address 1439 fe80::2001:db8:0:0 to its AERO interface with link-layer address 1440 L2(C1). Client ('C1') configures a default route and neighbor cache 1441 entry via the AERO interface with next-hop address fe80::2 and link- 1442 layer address L2(S1), then sub-delegates the ACP to its attached 1443 EUNs. IPv6 host ('H1') connects to the EUN, and configures the 1444 address 2001:db8:0::1. 1446 AERO Client ('C2') receives the ACP 2001:db8:1::/48 in a DHCPv6 PD 1447 exchange via AERO Server ('S2') then applies the address 1448 fe80::2001:db8:1:0 to its AERO interface with link-layer address 1449 L2(C2). Client ('C2') configures a default route and neighbor cache 1450 entry via the AERO interface with next-hop address fe80::3 and link- 1451 layer address L2(S2), then sub-delegates the ACP to its attached 1452 EUNs. IPv6 host ('H1') connects to the EUN, and configures the 1453 address 2001:db8:1::1. 1455 3.13.2. Concept of Operations 1457 Again, with reference to Figure 7, when source host ('H1') sends a 1458 packet to destination host ('H2'), the packet is first forwarded over 1459 the source host's attached EUN to Client ('C1'). Client ('C1') then 1460 forwards the packet via its AERO interface to Server ('S1') and also 1461 sends a Predirect message toward Client ('C2') via Server ('S1'). 1462 Server ('S1') then re-encapsulates and forwards both the packet and 1463 the Predirect message out the same AERO interface toward Client 1464 ('C2') via Relay ('R1'). 1466 When Relay ('R1') receives the packet and Predirect message, it 1467 consults its forwarding table to discover Server ('S2') as the next 1468 hop toward Client ('C2'). Relay ('R1') then forwards both the packet 1469 and the Predirect message to Server ('S2'), which then forwards them 1470 to Client ('C2'). 1472 After Client ('C2') receives the Predirect message, it process the 1473 message and returns a Redirect message toward Client ('C1') via 1474 Server ('S2'). During the process, Client ('C2') also creates or 1475 updates a dynamic neighbor cache entry for Client ('C1'). 1477 When Server ('S2') receives the Redirect message, it re-encapsulates 1478 the message and forwards it on to Relay ('R1'), which forwards the 1479 message on to Server ('S1') which forwards the message on to Client 1480 ('C1'). After Client ('C1') receives the Redirect message, it 1481 processes the message and creates or updates a dynamic neighbor cache 1482 entry for Client ('C2'). 1484 Following the above Predirect/Redirect message exchange, forwarding 1485 of packets from Client ('C1') to Client ('C2') without involving any 1486 intermediate nodes is enabled. The mechanisms that support this 1487 exchange are specified in the following sections. 1489 3.13.3. Message Format 1491 AERO Redirect/Predirect messages use the same format as for ICMPv6 1492 Redirect messages depicted in Section 4.5 of [RFC4861], but also 1493 include a new "Prefix Length" field taken from the low-order 8 bits 1494 of the Redirect message Reserved field. For IPv6, valid values for 1495 the Prefix Length field are 0 through 64; for IPv4, valid values are 1496 0 through 32. The Redirect/Predirect messages are formatted as shown 1497 in Figure 8: 1499 0 1 2 3 1500 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 1501 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1502 | Type (=137) | Code (=0/1) | Checksum | 1503 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1504 | Reserved | Prefix Length | 1505 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1506 | | 1507 + + 1508 | | 1509 + Target Address + 1510 | | 1511 + + 1512 | | 1513 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1514 | | 1515 + + 1516 | | 1517 + Destination Address + 1518 | | 1519 + + 1520 | | 1521 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1522 | Options ... 1523 +-+-+-+-+-+-+-+-+-+-+-+- 1525 Figure 8: AERO Redirect/Predirect Message Format 1527 3.13.4. Sending Predirects 1529 When a Client forwards a packet with a source address from one of its 1530 ACPs toward a destination address covered by an ASP (i.e., toward 1531 another AERO Client connected to the same AERO link), the source 1532 Client MAY send a Predirect message forward toward the destination 1533 Client via the Server. 1535 In the reference operational scenario, when Client ('C1') forwards a 1536 packet toward Client ('C2'), it MAY also send a Predirect message 1537 forward toward Client ('C2'), subject to rate limiting (see 1538 Section 8.2 of [RFC4861]). Client ('C1') prepares the Predirect 1539 message as follows: 1541 o the link-layer source address is set to 'L2(C1)' (i.e., the link- 1542 layer address of Client ('C1')). 1544 o the link-layer destination address is set to 'L2(S1)' (i.e., the 1545 link-layer address of Server ('S1')). 1547 o the network-layer source address is set to fe80::2001:db8:0:0 1548 (i.e., the AERO address of Client ('C1')). 1550 o the network-layer destination address is set to fe80::2001:db8:1:0 1551 (i.e., the AERO address of Client ('C2')). 1553 o the Type is set to 137. 1555 o the Code is set to 1 to indicate "Predirect". 1557 o the Prefix Length is set to the length of the prefix to be applied 1558 to the Target Address. 1560 o the Target Address is set to fe80::2001:db8:0:0 (i.e., the AERO 1561 address of Client ('C1')). 1563 o the Destination Address is set to the source address of the 1564 originating packet that triggered the Predirection event. (If the 1565 originating packet is an IPv4 packet, the address is constructed 1566 in IPv4-compatible IPv6 address format). 1568 o the message includes one or more TLLAOs with Link ID and 1569 Preference set to appropriate values for Client ('C1')'s 1570 underlying interfaces, and with UDP Port Number and IP Address set 1571 to 0'. 1573 o the message SHOULD include a Timestamp option and a Nonce option. 1575 o the message includes a Redirected Header Option (RHO) that 1576 contains the originating packet truncated if necessary to ensure 1577 that at least the network-layer header is included but the size of 1578 the message does not exceed 1280 bytes. 1580 Note that the act of sending Predirect messages is cited as "MAY", 1581 since Client ('C1') may have advanced knowledge that the direct path 1582 to Client ('C2') would be unusable or otherwise undesirable. If the 1583 direct path later becomes unusable after the initial route 1584 optimization, Client ('C1') simply allows packets to again flow 1585 through Server ('S1'). 1587 3.13.5. Re-encapsulating and Relaying Predirects 1589 When Server ('S1') receives a Predirect message from Client ('C1'), 1590 it first verifies that the TLLAOs in the Predirect are a proper 1591 subset of the Link IDs in Client ('C1')'s neighbor cache entry. If 1592 the Client's TLLAOs are not acceptable, Server ('S1') discards the 1593 message. Otherwise, Server ('S1') validates the message according to 1594 the ICMPv6 Redirect message validation rules in Section 8.1 of 1595 [RFC4861], except that the Predirect has Code=1. Server ('S1') also 1596 verifies that Client ('C1') is authorized to use the Prefix Length in 1597 the Predirect when applied to the AERO address in the network-layer 1598 source address by searching for the AERO address in the neighbor 1599 cache. If validation fails, Server ('S1') discards the Predirect; 1600 otherwise, it copies the correct UDP Port numbers and IP Addresses 1601 for Client ('C1')'s links into the (previously empty) TLLAOs. 1603 Server ('S1') then examines the network-layer destination address of 1604 the Predirect to determine the next hop toward Client ('C2') by 1605 searching for the AERO address in the neighbor cache. Since Client 1606 ('C2') is not one of its neighbors, Server ('S1') re-encapsulates the 1607 Predirect and relays it via Relay ('R1') by changing the link-layer 1608 source address of the message to 'L2(S1)' and changing the link-layer 1609 destination address to 'L2(R1)'. Server ('S1') finally forwards the 1610 re-encapsulated message to Relay ('R1') without decrementing the 1611 network-layer TTL/Hop Limit field. 1613 When Relay ('R1') receives the Predirect message from Server ('S1') 1614 it determines that Server ('S2') is the next hop toward Client ('C2') 1615 by consulting its forwarding table. Relay ('R1') then re- 1616 encapsulates the Predirect while changing the link-layer source 1617 address to 'L2(R1)' and changing the link-layer destination address 1618 to 'L2(S2)'. Relay ('R1') then relays the Predirect via Server 1619 ('S2'). 1621 When Server ('S2') receives the Predirect message from Relay ('R1') 1622 it determines that Client ('C2') is a neighbor by consulting its 1623 neighbor cache. Server ('S2') then re-encapsulates the Predirect 1624 while changing the link-layer source address to 'L2(S2)' and changing 1625 the link-layer destination address to 'L2(C2)'. Server ('S2') then 1626 forwards the message to Client ('C2'). 1628 3.13.6. Processing Predirects and Sending Redirects 1630 When Client ('C2') receives the Predirect message, it accepts the 1631 Predirect only if the message has a link-layer source address of one 1632 of its Servers (e.g., L2(S2)). Client ('C2') further accepts the 1633 message only if it is willing to serve as a redirection target. 1634 Next, Client ('C2') validates the message according to the ICMPv6 1635 Redirect message validation rules in Section 8.1 of [RFC4861], except 1636 that it accepts the message even though Code=1 and even though the 1637 network-layer source address is not that of it's current first-hop 1638 router. 1640 In the reference operational scenario, when Client ('C2') receives a 1641 valid Predirect message, it either creates or updates a dynamic 1642 neighbor cache entry that stores the Target Address of the message as 1643 the network-layer address of Client ('C1') , stores the link-layer 1644 addresses found in the TLLAOs as the link-layer addresses of Client 1645 ('C1') and stores the Prefix Length as the length to be applied to 1646 the network-layer address for forwarding purposes. Client ('C2') 1647 then sets AcceptTime for the neighbor cache entry to ACCEPT_TIME. 1649 After processing the message, Client ('C2') prepares a Redirect 1650 message response as follows: 1652 o the link-layer source address is set to 'L2(C2)' (i.e., the link- 1653 layer address of Client ('C2')). 1655 o the link-layer destination address is set to 'L2(S2)' (i.e., the 1656 link-layer address of Server ('S2')). 1658 o the network-layer source address is set to fe80::2001:db8:1:0 1659 (i.e., the AERO address of Client ('C2')). 1661 o the network-layer destination address is set to fe80::2001:db8:0:0 1662 (i.e., the AERO address of Client ('C1')). 1664 o the Type is set to 137. 1666 o the Code is set to 0 to indicate "Redirect". 1668 o the Prefix Length is set to the length of the prefix to be applied 1669 to the Target Address. 1671 o the Target Address is set to fe80::2001:db8:1:0 (i.e., the AERO 1672 address of Client ('C2')). 1674 o the Destination Address is set to the destination address of the 1675 originating packet that triggered the Redirection event. (If the 1676 originating packet is an IPv4 packet, the address is constructed 1677 in IPv4-compatible IPv6 address format). 1679 o the message includes one or more TLLAOs with Link ID and 1680 Preference set to appropriate values for Client ('C2')'s 1681 underlying interfaces, and with UDP Port Number and IP Address set 1682 to '0'. 1684 o the message SHOULD include a Timestamp option and MUST echo the 1685 Nonce option received in the Predirect (i.e., if a Nonce option is 1686 included). 1688 o the message includes as much of the RHO copied from the 1689 corresponding AERO Predirect message as possible such that at 1690 least the network-layer header is included but the size of the 1691 message does not exceed 1280 bytes. 1693 After Client ('C2') prepares the Redirect message, it sends the 1694 message to Server ('S2'). 1696 3.13.7. Re-encapsulating and Relaying Redirects 1698 When Server ('S2') receives a Redirect message from Client ('C2'), it 1699 first verifies that the TLLAOs in the Redirect are a proper subset of 1700 the Link IDs in Client ('C2')'s neighbor cache entry. If the 1701 Client's TLLAOs are not acceptable, Server ('S2') discards the 1702 message. Otherwise, Server ('S2') validates the message according to 1703 the ICMPv6 Redirect message validation rules in Section 8.1 of 1704 [RFC4861]. Server ('S2') also verifies that Client ('C2') is 1705 authorized to use the Prefix Length in the Redirect when applied to 1706 the AERO address in the network-layer source address by searching for 1707 the AERO address in the neighbor cache. If validation fails, Server 1708 ('S2') discards the Predirect; otherwise, it copies the correct UDP 1709 Port numbers and IP Addresses for Client ('C2')'s links into the 1710 (previously empty) TLLAOs. 1712 Server ('S2') then examines the network-layer destination address of 1713 the Predirect to determine the next hop toward Client ('C2') by 1714 searching for the AERO address in the neighbor cache. Since Client 1715 ('C2') is not a neighbor, Server ('S2') re-encapsulates the Predirect 1716 and relays it via Relay ('R1') by changing the link-layer source 1717 address of the message to 'L2(S2)' and changing the link-layer 1718 destination address to 'L2(R1)'. Server ('S2') finally forwards the 1719 re-encapsulated message to Relay ('R1') without decrementing the 1720 network-layer TTL/Hop Limit field. 1722 When Relay ('R1') receives the Predirect message from Server ('S2') 1723 it determines that Server ('S1') is the next hop toward Client ('C1') 1724 by consulting its forwarding table. Relay ('R1') then re- 1725 encapsulates the Predirect while changing the link-layer source 1726 address to 'L2(R1)' and changing the link-layer destination address 1727 to 'L2(S1)'. Relay ('R1') then relays the Predirect via Server 1728 ('S1'). 1730 When Server ('S1') receives the Predirect message from Relay ('R1') 1731 it determines that Client ('C1') is a neighbor by consulting its 1732 neighbor cache. Server ('S1') then re-encapsulates the Predirect 1733 while changing the link-layer source address to 'L2(S1)' and changing 1734 the link-layer destination address to 'L2(C1)'. Server ('S1') then 1735 forwards the message to Client ('C1'). 1737 3.13.8. Processing Redirects 1739 When Client ('C1') receives the Redirect message, it accepts the 1740 message only if it has a link-layer source address of one of its 1741 Servers (e.g., ''L2(S1)'). Next, Client ('C1') validates the message 1742 according to the ICMPv6 Redirect message validation rules in 1743 Section 8.1 of [RFC4861], except that it accepts the message even 1744 though the network-layer source address is not that of it's current 1745 first-hop router. Following validation, Client ('C1') then processes 1746 the message as follows. 1748 In the reference operational scenario, when Client ('C1') receives 1749 the Redirect message, it either creates or updates a dynamic neighbor 1750 cache entry that stores the Target Address of the message as the 1751 network-layer address of Client ('C2'), stores the link-layer 1752 addresses found in the TLLAOs as the link-layer addresses of Client 1753 ('C2') and stores the Prefix Length as the length to be applied to 1754 the network-layer address for forwarding purposes. Client ('C1') 1755 then sets ForwardTime for the neighbor cache entry to FORWARD_TIME. 1757 Now, Client ('C1') has a neighbor cache entry with a valid 1758 ForwardTime value, while Client ('C2') has a neighbor cache entry 1759 with a valid AcceptTime value. Thereafter, Client ('C1') may forward 1760 ordinary network-layer data packets directly to Client ('C2') without 1761 involving any intermediate nodes, and Client ('C2') can verify that 1762 the packets came from an acceptable source. (In order for Client 1763 ('C2') to forward packets to Client ('C1'), a corresponding 1764 Predirect/Redirect message exchange is required in the reverse 1765 direction; hence, the mechanism is asymmetric.) 1767 3.13.9. Server-Oriented Redirection 1769 In some environments, the Server nearest the target Client may need 1770 to serve as the redirection target, e.g., if direct Client-to-Client 1771 communications are not possible. In that case, the Server prepares 1772 the Redirect message the same as if it were the destination Client 1773 (see: Section 3.9.6), except that it writes its own link-layer 1774 address in the TLLAO option. The Server must then maintain a 1775 neighbor cache entry for the redirected source Client. 1777 3.14. Neighbor Unreachability Detection (NUD) 1779 AERO nodes perform Neighbor Unreachability Detection (NUD) by sending 1780 unicast NS messages to elicit solicited NA messages from neighbors 1781 the same as described in [RFC4861]. NUD is performed either 1782 reactively in response to persistent L2 errors (see Section 3.10) or 1783 proactively to refresh existing neighbor cache entries. 1785 When an AERO node sends an NS/NA message, it MUST use its link-local 1786 address as the IPv6 source address and the link-local address of the 1787 neighbor as the IPv6 destination address. When an AERO node receives 1788 an NS message or a solicited NA message, it accepts the message if it 1789 has a neighbor cache entry for the neighbor; otherwise, it ignores 1790 the message. 1792 When a source Client is redirected to a target Client it SHOULD 1793 proactively test the direct path by sending an initial NS message to 1794 elicit a solicited NA response. While testing the path, the source 1795 Client can optionally continue sending packets via the Server, 1796 maintain a small queue of packets until target reachability is 1797 confirmed, or (optimistically) allow packets to flow directly to the 1798 target. The source Client SHOULD thereafter continue to proactively 1799 test the direct path to the target Client (see Section 7.3 of 1800 [RFC4861]) periodically in order to keep dynamic neighbor cache 1801 entries alive. 1803 In particular, while the source Client is actively sending packets to 1804 the target Client it SHOULD also send NS messages separated by 1805 RETRANS_TIMER milliseconds in order to receive solicited NA messages. 1806 If the source Client is unable to elicit a solicited NA response from 1807 the target Client after MAX_RETRY attempts, it SHOULD set ForwardTime 1808 to 0 and resume sending packets via one of its Servers. Otherwise, 1809 the source Client considers the path usable and SHOULD thereafter 1810 process any link-layer errors as a hint that the direct path to the 1811 target Client has either failed or has become intermittent. 1813 When a target Client receives an NS message from a source Client, it 1814 resets AcceptTime to ACCEPT_TIME if a neighbor cache entry exists; 1815 otherwise, it discards the NS message. If ForwardTime is non-zero, 1816 the target Client then sends a solicited NA message to the link-layer 1817 address of the source Client; otherwise, it sends the solicited NA 1818 message to the link-layer address of one of its Servers. 1820 When a source Client receives a solicited NA message from a target 1821 Client, it resets ForwardTime to FORWARD_TIME if a neighbor cache 1822 entry exists; otherwise, it discards the NA message. 1824 When ForwardTime for a dynamic neighbor cache entry expires, the 1825 source Client resumes sending any subsequent packets via a Server and 1826 may (eventually) attempt to re-initiate the AERO redirection process. 1827 When AcceptTime for a dynamic neighbor cache entry expires, the 1828 target Client discards any subsequent packets received directly from 1829 the source Client. When both ForwardTime and AcceptTime for a 1830 dynamic neighbor cache entry expire, the Client deletes the neighbor 1831 cache entry. 1833 3.15. Mobility Management 1835 3.15.1. Announcing Link-Layer Address Changes 1837 When a Client needs to change its link-layer address, e.g., due to a 1838 mobility event, it performs an immediate DHCPv6 Rebind/Reply exchange 1839 via each of its Servers using the new link-layer address as the 1840 source and with a CLLAO that includes the correct Link ID and 1841 Preference values. If authentication succeeds, the Server then 1842 update its neighbor cache and sends a DHCPv6 Reply. Note that if the 1843 Client does not issue a DHCPv6 Rebind before the Server has 1844 terminated the lease (e.g., if the Client has been out of touch with 1845 the Server for a considerable amount of time), the Server's Reply 1846 will report NoBinding and the Client must re-initiate the DHCPv6 PD 1847 procedure. 1849 Next, the Client sends unsolicited NA messages to each of its 1850 correspondent Client neighbors using the same procedures as specified 1851 in Section 7.2.6 of [RFC4861], except that it sends the messages as 1852 unicast to each neighbor via a Server instead of multicast. In this 1853 process, the Client should send no more than 1854 MAX_NEIGHBOR_ADVERTISEMENT messages separated by no less than 1855 RETRANS_TIMER seconds to each neighbor. 1857 With reference to Figure 7, when Client ('C2') needs to change its 1858 link-layer address it sends unicast unsolicited NA messages to Client 1859 ('C1') via Server ('S2') as follows: 1861 o the link-layer source address is set to 'L2(C2)' (i.e., the link- 1862 layer address of Client ('C2')). 1864 o the link-layer destination address is set to 'L2(S2)' (i.e., the 1865 link-layer address of Server ('S2')). 1867 o the network-layer source address is set to fe80::2001:db8:1:0 1868 (i.e., the AERO address of Client ('C2')). 1870 o the network-layer destination address is set to fe80::2001:db8:0:0 1871 (i.e., the AERO address of Client ('C1')). 1873 o the Type is set to 136. 1875 o the Code is set to 0. 1877 o the Solicited flag is set to 0. 1879 o the Override flag is set to 1. 1881 o the Target Address is set to fe80::2001:db8:1:0 (i.e., the AERO 1882 address of Client ('C2')). 1884 o the message includes one or more TLLAOs with Link ID and 1885 Preference set to appropriate values for Client ('C2')'s 1886 underlying interfaces, and with UDP Port Number and IP Address set 1887 to '0'. 1889 o the message SHOULD include a Timestamp option. 1891 When Server ('S1') receives the NA message, it relays the message in 1892 the same way as described for relaying Redirect messages in 1893 Section 3.12.7. In particular, Server ('S1') copies the correct UDP 1894 port numbers and IP addresses into the TLLAOs, changes the link-layer 1895 source address to its own address, changes the link-layer destination 1896 address to the address of Relay ('R1'), then forwards the NA message 1897 via the relaying chain the same as for a Redirect. 1899 When Client ('C1') receives the NA message, it accepts the message 1900 only if it already has a neighbor cache entry for Client ('C2') then 1901 updates the link-layer addresses for Client ('C2') based on the 1902 addresses in the TLLAOs. However, Client ('C1') MUST NOT update 1903 ForwardTime since Client ('C2') will not have updated AcceptTime. 1905 Note that these unsolicited NA messages are unacknowledged; hence, 1906 Client ('C2') has no way of knowing whether Client ('C1') has 1907 received them. If the messages are somehow lost, however, Client 1908 ('C1') will soon learn of the mobility event via the NUD procedures 1909 specified in Section 3.14. 1911 3.15.2. Bringing New Links Into Service 1913 When a Client needs to bring a new underlying interface into service 1914 (e.g., when it activates a new data link), it performs an immediate 1915 Rebind/Reply exchange via each of its Servers using the new link- 1916 layer address as the source address and with a CLLAO that includes 1917 the new Link ID and Preference values. If authentication succeeds, 1918 the Server then updates its neighbor cache and sends a DHCPv6 Reply. 1919 The Client MAY then send unsolicited NA messages to each of its 1920 correspondent Clients to inform them of the new link-layer address as 1921 described in Section 3.15.1. 1923 3.15.3. Removing Existing Links from Service 1925 When a Client needs to remove an existing underlying interface from 1926 service (e.g., when it de-activates an existing data link), it 1927 performs an immediate Rebind/Reply exchange via each of its Servers 1928 over any available link with a CLLAO that includes the deprecated 1929 Link ID and a Preference value of 0. If authentication succeeds, the 1930 Server then updates its neighbor cache and sends a DHCPv6 Reply. The 1931 Client SHOULD then send unsolicited NA messages to each of its 1932 correspondent Clients to inform them of the deprecated link-layer 1933 address as described in Section 3.15.1. 1935 3.15.4. Moving to a New Server 1937 When a Client associates with a new Server, it performs the Client 1938 procedures specified in Section 3.10. 1940 When a Client disassociates with an existing Server, it sends a 1941 DHCPv6 Release message to the unicast link-local network layer 1942 address of the old Server. The Client SHOULD send the message via a 1943 new Server (i.e., by setting the link-layer destination address to 1944 the address of the new Server) in case the old Server is unreachable 1945 at the link layer, e.g., if the old Server is in a different network 1946 partition. The new Server will forward the message to a Relay, which 1947 will in turn forward the message to the old Server. 1949 When the old Server receives the DHCPv6 Release, it first 1950 authenticates the message. If authentication succeeds, the old 1951 Server withdraws the IP route from the AERO routing system and 1952 deletes the neighbor cache entry for the Client. (The old Server MAY 1953 impose a small delay before deleting the neighbor cache entry so that 1954 any packets already in the system can still be delivered to the 1955 Client.) The old Server then returns a DHCPv6 Reply message via a 1956 Relay. The Client can then use the Reply message to verify that the 1957 termination signal has been processed, and can delete both the 1958 default route and the neighbor cache entry for the old Server. (Note 1959 that the Server's Reply to the Client's Release message may be lost, 1960 e.g., if the AERO routing system has not yet converged. Since the 1961 Client is responsible for reliability, however, it will retry until 1962 it gets an indication that the Release was successful.) 1964 Clients SHOULD NOT move rapidly between Servers in order to avoid 1965 causing excessive oscillations in the AERO routing system. Such 1966 oscillations could result in intermittent reachability for the Client 1967 itself, while causing little harm to the network due to routing 1968 protocol dampening. Examples of when a Client might wish to change 1969 to a different Server include a Server that has gone unreachable, 1970 topological movements of significant distance, etc. 1972 3.16. Encapsulation Protocol Version Considerations 1974 A source Client may connect only to an IPvX underlying network, while 1975 the target Client connects only to an IPvY underlying network. In 1976 that case, the target and source Clients have no means for reaching 1977 each other directly (since they connect to underlying networks of 1978 different IP protocol versions) and so must ignore any redirection 1979 messages and continue to send packets via the Server. 1981 3.17. Multicast Considerations 1983 When the underlying network does not support multicast, AERO nodes 1984 map IPv6 link-scoped multicast addresses (including 1985 'All_DHCP_Relay_Agents_and_Servers') to the link-layer address of a 1986 Server. 1988 When the underlying network supports multicast, AERO nodes use the 1989 multicast address mapping specification found in [RFC2529] for IPv4 1990 underlying networks and use a direct multicast mapping for IPv6 1991 underlying networks. (In the latter case, "direct multicast mapping" 1992 means that if the IPv6 multicast destination address of the 1993 encapsulated packet is "M", then the IPv6 multicast destination 1994 address of the encapsulating header is also "M".) 1996 3.18. Operation on AERO Links Without DHCPv6 Services 1998 When Servers on the AERO link do not provide DHCPv6 services, 1999 operation can still be accommodated through administrative 2000 configuration of ACPs on AERO Clients. In that case, administrative 2001 configurations of AERO interface neighbor cache entries on both the 2002 Server and Client are also necessary. However, this may interfere 2003 with the ability for Clients to dynamically change to new Servers, 2004 and can expose the AERO link to misconfigurations unless the 2005 administrative configurations are carefully coordinated. 2007 3.19. Operation on Server-less AERO Links 2009 In some AERO link scenarios, there may be no Servers on the link and/ 2010 or no need for Clients to use a Server as an intermediary trust 2011 anchor. In that case, each Client acts as a Server unto itself to 2012 establish neighbor cache entries by performing direct Client-to- 2013 Client IPv6 ND message exchanges, and some other form of trust basis 2014 must be applied so that each Client can verify that the prospective 2015 neighbor is authorized to use its claimed ACP. 2017 When there is no Server on the link, Clients must arrange to receive 2018 ACPs and publish them via a secure alternate prefix delegation 2019 authority through some means outside the scope of this document. 2021 3.20. Proxy AERO 2023 Proxy Mobile IPv6 (PMIPv6) [RFC5213][RFC5844] presents a localized 2024 mobility management scheme for use within an access network domain. 2025 It is typically used in cellular wireless service provider networks, 2026 and allows mobile nodes to receive and retain a stable IP address 2027 without needing to implement any special mobility protocols. In the 2028 PMIPv6 architecture, access network devices known as Mobility Access 2029 Gateways (MAGs) provide mobile nodes with an access link abstraction 2030 and receive prefixes for the mobile nodes from a Local Mobility 2031 Anchor (LMA). 2033 The AERO Client (acting as a MAG) can similarly provide proxy 2034 services for mobile nodes that do not participate in AERO messaging. 2035 The proxy Client presents an access link abstraction to mobile nodes, 2036 and performs DHCPv6 PD exchanges over the AERO interface with an AERO 2037 Server (acting as an LMA) to receive a prefix for address 2038 provisioning of the mobile node. 2040 When a mobile node comes onto an access link presented by a proxy 2041 Client, the Client authenticates the node and obtains a unique 2042 identifier that it can use as the DUID in its DHCPv6 PD messages to 2043 the Server. When the Server delegates a prefix, the Client creates a 2044 new AERO address for the mobile node and assigns the delegated prefix 2045 to the mobile node's access link. The Client then generates address 2046 autoconfiguration messages (e.g., IPv6 RA, DHCPv6, DHCPv4, etc.) over 2047 the access link and configures itself as a default router for the 2048 mobile node. Since the Client may serve many such mobile nodes 2049 simultaneously, it may configure multiple AERO addresses, i.e., one 2050 for each mobile node. 2052 When two mobile nodes are associated with the same proxy Client, the 2053 Client can forward traffic between the mobiles without involving the 2054 Server since it configures the AERO addresses of each mobile and 2055 therefore also has the necessary routing information. When two 2056 mobiles are associated with different Clients, the first mobile 2057 node's Client can initiate standard AERO route optimization using the 2058 mobile's AERO address as the source for route optimization messaging. 2059 This may result in a route optimization where the first mobile node's 2060 Client discovers a direct path to the second mobile node's Client. 2062 When a mobile node moves to a new proxy Client, the old proxy Client 2063 issues a DHCPv6 Release message and sends unsolicited NA messages to 2064 any of the mobile node's correspondents the same as specified for 2065 announcing link-layer address changes in Section 3.15.1. However, 2066 since the old Client has no way of knowing where the mobile has moved 2067 to, it sets the Code field in the NA message to 1. When the 2068 correspondent receives such an NA message, it deletes the neighbor 2069 cache entry for the departed mobile node and again allows packets to 2070 flow through its Server. 2072 In addition to the use of DHCPv6 PD signaling, the AERO approach 2073 differs from PMIPv6 in its use of the NBMA virtual link model instead 2074 of point-to-point tunnels. This provides a more agile interface for 2075 Client-to-Server coordinations, and also facilitates simple route 2076 optimization. The AERO routing system is also arranged in such a 2077 fashion that Clients get the same service from any Server they happen 2078 to associate with. This provides a natural fault tolerance and load 2079 balancing capability such as desired for distributed mobility 2080 management. All other considerations are the same as specified in 2081 [RFC5213][RFC5844]. 2083 3.21. Extending AERO Links Through Security Gateways 2085 When an enterprise mobile device moves from a campus LAN connection 2086 to a public Internet link, it must re-enter the enterprise via a 2087 security gateway that has both an physical interface connection to 2088 the Internet and a physical interface connection to the enterprise 2089 internetwork. This most often entails the establishment of a Virtual 2090 Private Network (VPN) link over the public Internet from the mobile 2091 device to the security gateway. During this process, the mobile 2092 device supplies the security gateway with its public Internet address 2093 as the link-layer address for the VPN. The mobile device then acts 2094 as an AERO Client to negotiate with the security gateway to obtain 2095 its ACP. 2097 In order to satisfy this need, the security gateway also operates as 2098 an AERO Server with support for AERO Client proxying. In particular, 2099 when a mobile device (i.e., the Client) connects via the security 2100 gateway (i.e., the Server), the Server provides the Client with an 2101 ACP in a DHCPv6 PD exchange the same as if it were attached to an 2102 enterprise campus access link. The Server then replaces the Client's 2103 link-layer source address with the Server's enterprise-facing link- 2104 layer address in all AERO messages the Client sends toward neighbors 2105 on the AERO link. The AERO messages are then delivered to other 2106 devices on the AERO link as if they were originated by the security 2107 gateway instead of by the AERO Client. In the reverse direction, the 2108 AERO messages sourced by devices within the enterprise network can be 2109 forwarded to the security gateway, which then replaces the link-layer 2110 destination address with the Client's link-layer address and replaces 2111 the link-layer source address with its own (Internet-facing) link- 2112 layer address. 2114 After receiving the ACP, the Client can send IP packets that use an 2115 address taken from the ACP as the network layer source address, the 2116 Client's link-layer address as the link-layer source address, and the 2117 Server's Internet-facing link-layer address as the link-layer 2118 destination address. The Server will then rewrite the link-layer 2119 source address with the Server's own enterprise-facing link-layer 2120 address and rewrite the link-layer destination address with the 2121 target AERO node's link-layer address, and the packets will enter the 2122 enterprise network as though they were sourced from a device located 2123 within the enterprise. In the reverse direction, when a packet 2124 sourced by a node within the enterprise network uses a destination 2125 address from the Client's ACP, the packet will be delivered to the 2126 security gateway which then rewrites the link-layer destination 2127 address to the Client's link-layer address and rewrites the link- 2128 layer source address to the Server's Internet-facing link-layer 2129 address. The Server then delivers the packet across the VPN to the 2130 AERO Client. In this way, the AERO virtual link is essentially 2131 extended *through* the security gateway to the point at which the VPN 2132 link and AERO link are effectively grafted together by the link-layer 2133 address rewriting performed by the security gateway. All AERO 2134 messaging services (including route optimization and mobility 2135 signaling) are therefore extended to the Client. 2137 In order to support this virtual link grafting, the security gateway 2138 (acting as an AERO Server) must keep static neighbor cache entries 2139 for all of its associated Clients located on the public Internet. 2140 The neighbor cache entry is keyed by the AERO Client's AERO address 2141 the same as if the Client were located within the enterprise 2142 internetwork. The neighbor cache is then managed in all ways as 2143 though the Client were an ordinary AERO Client. This includes the 2144 AERO IPv6 ND messaging signaling for Route Optimization and Neighbor 2145 Unreachability Detection. 2147 Note that the main difference between a security gateway acting as an 2148 AERO Server and an enterprise-internal AERO Server is that the 2149 security gateway has at least one enterprise-internal physical 2150 interface and at least one public Internet physical interface. 2151 Conversely, the enterprise-internal AERO Server has only enterprise- 2152 internal physical interfaces. For this reason security gateway 2153 proxying is needed to ensure that the public Internet link-layer 2154 addressing space is kept separate from the enterprise-internal link- 2155 layer addressing space. This is afforded through a natural extension 2156 of the security association caching already performed for each VPN 2157 client by the security gateway. 2159 3.22. Extending IPv6 AERO Links to the Internet 2161 When an IPv6 host ('H1') with address 2001:db8:1::1 serviced by AERO 2162 Client ('C1') in the public IPv6 Internet sends packets to a 2163 correspondent IPv6 host ('H2') with address 2001:db8:2::1 that is not 2164 covered by one of its own ASPs, the packets will be forwarded through 2165 ('C1')s Server and then through a Relay before finally being 2166 forwarded to the public IPv6 Internet. This could lead to sub- 2167 optimal performance when the correspondent could instead be reached 2168 via a more direct route over the IPv6 Internet, i.e., without 2169 involving ('C1')s home network. 2171 Consider the case when Client ('C1') has the ACP 2001:db8:1::/64 and 2172 has an underlying IPv6 Internet address of 2001:db8:1000::1, while 2173 Client ('C2') (unbeknownst to ('C1')) has the ACP 2001:db8:2::/64 and 2174 has an underlying IPv6 Internet address of 2001:db8:2000::1. A first 2175 point for ('C1') to determine is whether ('H2') is serviced by a 2176 Client ('C2') associated with a different home network. To do this, 2177 Client ('C1') creates a specially-crafted encapsulated AERO Predirect 2178 message that will be routed through ('C1')s home network to the 2179 public Internet, where it will be routed through potential Client 2180 ('C2')s home network and then finally to ('C2') itself. Client 2181 ('C1') prepares the message as follows: 2183 o The encapsulating IPv6 header source address is set to 2184 2001:db8:1:: (i.e., the IPv6 subnet router anycast address for 2185 ('C1')s ACP) 2187 o The encapsulating IPv6 header destination address is set to 2188 2001:db8:2:: (i.e., the presumed IPv6 subnet router anycast 2189 address for ('C2')s ACP) 2191 o The encapsulating IPv6 header is followed by a UDP header with 2192 source and destination port set to 8060 2194 o The encapsulated IPv6 header source address is set to 2195 fe80::2001:db8:1:0 (i.e., the AERO address for ('C1')) 2197 o The encapsulated IPv6 header destination address is set to 2198 fe80::2001:db8:2:0 (i.e., the presumed AERO address for ('C2')) 2200 o The encapsulated AERO Predirect message includes all of the 2201 securing information that would occur in a Mobile IPv6 (MIPv6) 2202 [RFC6275] "Home Test Init" message (format TBD) 2204 Client ('C1') then further encapsulates the message in the 2205 encapsulating headers necessary to convey the packet to its AERO 2206 Server (e.g., through IPsec encapsulation) so that the message now 2207 appears "double-encapsulated". ('C1') then sends the message to its 2208 Server, which further re-encapsulates and forwards it to a Relay, 2209 which in turn removes the outermost encapsulation layer and forwards 2210 the (now single-encapsulated) message to the Internet. 2212 At the same time, ('C1') creates a second encapsulated AERO Predirect 2213 message that will be routed through the Internet to potential Client 2214 ('C2')s home network and then finally to ('C2') itself. Client 2215 ('C2') prepares the message as follows: 2217 o The encapsulating IPv6 header source address is set to 2218 2001:db8:1000:1 (i.e., the Internet IPv6 address of ('C1')) 2220 o The encapsulating IPv6 header destination address is set to 2221 2001:db8:2:: (i.e., the presumed IPv6 subnet router anycast 2222 address for ('H2')) 2224 o The encapsulating IPv6 header is followed by a UDP header with 2225 source and destination port set to 8060 2227 o The encapsulated IPv6 header source address is set to 2228 fe80::2001:db8:1:0 (i.e., the AERO address for ('C1')) 2230 o The encapsulated IPv6 header destination address is set to 2231 fe80::2001:db8:2:0 (i.e., the presumed AERO address for ('C2')) 2233 o The encapsulated AERO Predirect message includes all of the 2234 securing information that would occur in a MIPv6 "Care-of Test 2235 Init" message (format TBD) 2237 If ('C2') is indeed an AERO Client, it will receive both Predirect 2238 messages through its home network. ('C2') then returns a 2239 corresponding Redirect for each of the Predirect messages with the 2240 source and destination addresses in the inner and outer headers 2241 reversed. The first message includes all of the securing information 2242 that would occur in a MIPv6 "Home Test" message, while the second 2243 message includes all of the securing information that would occur in 2244 a MIPv6 "Care-of Test" message (format TBD). 2246 When ('C1') receives the Redirect messages that serve the purposes of 2247 the Home/Care-of tests, it performs the necessary security procedures 2248 per the MIPv6 specification. It then prepares an encapsulated NS 2249 message that includes the same source and destination addresses as 2250 for the "Home Test Init" Predirect message, and includes all of the 2251 securing information that would occur in a MIPv6 "Binding Update" 2252 message (format TBD) and sends the message to ('C2'). 2254 When ('C2') receives the NS message, if the securing information is 2255 correct it creates or updates a neighbor cache entry for ('C1') with 2256 fe80::2001:db8:1:0 as the network-layer address, 2001:db8:1000::1 as 2257 the link-layer address and with AcceptTime set to ACCEPT_TIME. 2258 ('C2') then sends an encapsulated NA message back to ('C1') that 2259 includes the same source and destination addresses as for the "Home 2260 Test" Redirect message, and includes all of the securing information 2261 that would occur in a MIPv6 "Binding Acknowledgement" message (format 2262 TBD) and sends the message to ('C1'). 2264 When ('C1') receives the NA message, it creates or updates a neighbor 2265 cache entry for ('C2') with fe80::2001:db8:2:0 as the network-layer 2266 address and 2001:db8:2:: as the link-layer address and with 2267 ForwardTime set to FORWARD_TIME, thus completing the route 2268 optimization in the forward direction. 2270 ('C1') subsequently forwards encapsulated messages with outer source 2271 address 2001:db8:1000::1, with outer destination address 2272 2001:db8:2::, with inner source address taken from the 2001:db8:1::, 2273 and with inner destination address taken from 2001:db8:2:: due to the 2274 fact that it has a securely-established neighbor cache entry with 2275 non-zero ForwardTime. ('C2') subsequently accepts any such 2276 encapsulated messages due to the fact that it has a securely- 2277 established neighbor cache entry with non-zero AcceptTime.. 2279 In order to keep neighbor cache entries alive, ('C1') periodically 2280 sends additional NS messages to ('C2') and receives any NA responses. 2281 If ('C1') moves to a different point of attachment after the initial 2282 route optimization, it sends a new secured NS message to ('C2') as 2283 above to update ('C2')s neighbor cache. 2285 If ('C2') has packets to send to ('C1'), it performs a corresponding 2286 route optimization in the opposite direction following the same 2287 procedures described above. In the process, the already-established 2288 unidirectional neighbor cache entries within ('C1') and ('C2') are 2289 updated to include the now-bidirectional information. In particular, 2290 the AcceptTime and ForwardTime variables for both neighbor cache 2291 entries are updated to non-zero values, and the link-layer address 2292 for ('C1')s neighbor cache entry for ('C2') is reset to 2293 2001:db8:2000::1. 2295 In order to support this process, Clients MUST process any UDP 2296 packets with destination port 8060 that have a subnet router anycast 2297 address that corresponds to any of the /64 prefixes covered by the 2298 Client's ACP. 2300 4. Implementation Status 2302 An application-layer implementation is in progress. 2304 5. IANA Considerations 2306 IANA is instructed to assign a new 2-octet Hardware Type number 2307 "TBD1" for AERO in the "arp-parameters" registry per Section 2 of 2308 [RFC5494]. The number is assigned from the 2-octet Unassigned range 2309 with Hardware Type "AERO" and with this document as the reference. 2311 IANA is instructed to assign a 4-octet Enterprise Number "TBD2" for 2312 AERO in the "enterprise-numbers" registry per [RFC3315]. 2314 6. Security Considerations 2316 AERO link security considerations are the same as for standard IPv6 2317 Neighbor Discovery [RFC4861] except that AERO improves on some 2318 aspects. In particular, AERO uses a trust basis between Clients and 2319 Servers, where the Clients only engage in the AERO mechanism when it 2320 is facilitated by a trust anchor. Unless there is some other means 2321 of authenticating the Client's identity (e.g., link-layer security), 2322 AERO nodes SHOULD also use DHCPv6 securing services (e.g., DHCPv6 2323 authentication, Secure DHCPv6 [I-D.ietf-dhc-sedhcpv6], etc.) for 2324 Client authentication and network admission control. 2326 AERO Redirect, Predirect and unsolicited NA messages SHOULD include a 2327 Timestamp option (see Section 5.3 of [RFC3971]) that other AERO nodes 2328 can use to verify the message time of origin. AERO Predirect, NS and 2329 RS messages SHOULD include a Nonce option (see Section 5.3 of 2330 [RFC3971]) that recipients echo back in corresponding responses. 2332 AERO links must be protected against link-layer address spoofing 2333 attacks in which an attacker on the link pretends to be a trusted 2334 neighbor. Links that provide link-layer securing mechanisms (e.g., 2335 IEEE 802.1X WLANs) and links that provide physical security (e.g., 2336 enterprise network wired LANs) provide a first line of defense that 2337 is often sufficient. In other instances, additional securing 2338 mechanisms such as Secure Neighbor Discovery (SeND) [RFC3971], IPsec 2339 [RFC4301] or TLS [RFC5246] may be necessary. 2341 AERO Clients MUST ensure that their connectivity is not used by 2342 unauthorized nodes on their EUNs to gain access to a protected 2343 network, i.e., AERO Clients that act as routers MUST NOT provide 2344 routing services for unauthorized nodes. (This concern is no 2345 different than for ordinary hosts that receive an IP address 2346 delegation but then "share" the address with unauthorized nodes via a 2347 NAT function.) 2348 On some AERO links, establishment and maintenance of a direct path 2349 between neighbors requires secured coordination such as through the 2350 Internet Key Exchange (IKEv2) protocol [RFC5996] to establish a 2351 security association. 2353 7. Acknowledgements 2355 Discussions both on IETF lists and in private exchanges helped shape 2356 some of the concepts in this work. Individuals who contributed 2357 insights include Mikael Abrahamsson, Mark Andrews, Fred Baker, 2358 Stewart Bryant, Brian Carpenter, Wojciech Dec, Ralph Droms, Sri 2359 Gundavelli, Brian Haberman, Joel Halpern, Sascha Hlusiak, Lee Howard, 2360 Andre Kostur, Ted Lemon, Joe Touch and Bernie Volz. Members of the 2361 IESG also provided valuable input during their review process that 2362 greatly improved the document. Special thanks go to Stewart Bryant, 2363 Joel Halpern and Brian Haberman for their shepherding guidance. 2365 This work has further been encouraged and supported by Boeing 2366 colleagues including Keith Bartley, Dave Bernhardt, Cam Brodie, 2367 Balaguruna Chidambaram, Claudiu Danilov, Wen Fang, Anthony Gregory, 2368 Jeff Holland, Ed King, Gen MacLean, Kent Shuey, Brian Skeen, Mike 2369 Slane, Julie Wulff, Yueli Yang, and other members of the BR&T and BIT 2370 mobile networking teams. 2372 Earlier works on NBMA tunneling approaches are found in 2373 [RFC2529][RFC5214][RFC5569]. 2375 8. References 2377 8.1. Normative References 2379 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 2380 August 1980. 2382 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, September 2383 1981. 2385 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 2386 RFC 792, September 1981. 2388 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 2389 October 1996. 2391 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2392 Requirement Levels", BCP 14, RFC 2119, March 1997. 2394 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 2395 (IPv6) Specification", RFC 2460, December 1998. 2397 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 2398 IPv6 Specification", RFC 2473, December 1998. 2400 [RFC3315] Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C., 2401 and M. Carney, "Dynamic Host Configuration Protocol for 2402 IPv6 (DHCPv6)", RFC 3315, July 2003. 2404 [RFC3633] Troan, O. and R. Droms, "IPv6 Prefix Options for Dynamic 2405 Host Configuration Protocol (DHCP) version 6", RFC 3633, 2406 December 2003. 2408 [RFC3971] Arkko, J., Kempf, J., Zill, B., and P. Nikander, "SEcure 2409 Neighbor Discovery (SEND)", RFC 3971, March 2005. 2411 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 2412 for IPv6 Hosts and Routers", RFC 4213, October 2005. 2414 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 2415 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 2416 September 2007. 2418 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 2419 Address Autoconfiguration", RFC 4862, September 2007. 2421 [RFC6434] Jankiewicz, E., Loughney, J., and T. Narten, "IPv6 Node 2422 Requirements", RFC 6434, December 2011. 2424 8.2. Informative References 2426 [I-D.ietf-dhc-sedhcpv6] 2427 Jiang, S., Shen, S., Zhang, D., and T. Jinmei, "Secure 2428 DHCPv6 with Public Key", draft-ietf-dhc-sedhcpv6-03 (work 2429 in progress), June 2014. 2431 [RFC0879] Postel, J., "TCP maximum segment size and related topics", 2432 RFC 879, November 1983. 2434 [RFC1812] Baker, F., "Requirements for IP Version 4 Routers", RFC 2435 1812, June 1995. 2437 [RFC1930] Hawkinson, J. and T. Bates, "Guidelines for creation, 2438 selection, and registration of an Autonomous System (AS)", 2439 BCP 6, RFC 1930, March 1996. 2441 [RFC2131] Droms, R., "Dynamic Host Configuration Protocol", RFC 2442 2131, March 1997. 2444 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 2445 Domains without Explicit Tunnels", RFC 2529, March 1999. 2447 [RFC2675] Borman, D., Deering, S., and R. Hinden, "IPv6 Jumbograms", 2448 RFC 2675, August 1999. 2450 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", RFC 2451 2923, September 2000. 2453 [RFC3819] Karn, P., Bormann, C., Fairhurst, G., Grossman, D., 2454 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 2455 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 2456 RFC 3819, July 2004. 2458 [RFC4271] Rekhter, Y., Li, T., and S. Hares, "A Border Gateway 2459 Protocol 4 (BGP-4)", RFC 4271, January 2006. 2461 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 2462 Architecture", RFC 4291, February 2006. 2464 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 2465 Internet Protocol", RFC 4301, December 2005. 2467 [RFC4443] Conta, A., Deering, S., and M. Gupta, "Internet Control 2468 Message Protocol (ICMPv6) for the Internet Protocol 2469 Version 6 (IPv6) Specification", RFC 4443, March 2006. 2471 [RFC4821] Mathis, M. and J. Heffner, "Packetization Layer Path MTU 2472 Discovery", RFC 4821, March 2007. 2474 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 2475 Errors at High Data Rates", RFC 4963, July 2007. 2477 [RFC4994] Zeng, S., Volz, B., Kinnear, K., and J. Brzozowski, 2478 "DHCPv6 Relay Agent Echo Request Option", RFC 4994, 2479 September 2007. 2481 [RFC5213] Gundavelli, S., Leung, K., Devarapalli, V., Chowdhury, K., 2482 and B. Patil, "Proxy Mobile IPv6", RFC 5213, August 2008. 2484 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 2485 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 2486 March 2008. 2488 [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security 2489 (TLS) Protocol Version 1.2", RFC 5246, August 2008. 2491 [RFC5494] Arkko, J. and C. Pignataro, "IANA Allocation Guidelines 2492 for the Address Resolution Protocol (ARP)", RFC 5494, 2493 April 2009. 2495 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 2496 Route Optimization Requirements for Operational Use in 2497 Aeronautics and Space Exploration Mobile Networks", RFC 2498 5522, October 2009. 2500 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 2501 Infrastructures (6rd)", RFC 5569, January 2010. 2503 [RFC5844] Wakikawa, R. and S. Gundavelli, "IPv4 Support for Proxy 2504 Mobile IPv6", RFC 5844, May 2010. 2506 [RFC5996] Kaufman, C., Hoffman, P., Nir, Y., and P. Eronen, 2507 "Internet Key Exchange Protocol Version 2 (IKEv2)", RFC 2508 5996, September 2010. 2510 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 2511 NAT64: Network Address and Protocol Translation from IPv6 2512 Clients to IPv4 Servers", RFC 6146, April 2011. 2514 [RFC6204] Singh, H., Beebee, W., Donley, C., Stark, B., and O. 2515 Troan, "Basic Requirements for IPv6 Customer Edge 2516 Routers", RFC 6204, April 2011. 2518 [RFC6275] Perkins, C., Johnson, D., and J. Arkko, "Mobility Support 2519 in IPv6", RFC 6275, July 2011. 2521 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 2522 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, August 2523 2011. 2525 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 2526 for Equal Cost Multipath Routing and Link Aggregation in 2527 Tunnels", RFC 6438, November 2011. 2529 [RFC6691] Borman, D., "TCP Options and Maximum Segment Size (MSS)", 2530 RFC 6691, July 2012. 2532 [RFC6706] Templin, F., "Asymmetric Extended Route Optimization 2533 (AERO)", RFC 6706, August 2012. 2535 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID Field", 2536 RFC 6864, February 2013. 2538 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 2539 UDP Checksums for Tunneled Packets", RFC 6935, April 2013. 2541 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 2542 for the Use of IPv6 UDP Datagrams with Zero Checksums", 2543 RFC 6936, April 2013. 2545 [RFC6939] Halwasia, G., Bhandari, S., and W. Dec, "Client Link-Layer 2546 Address Option in DHCPv6", RFC 6939, May 2013. 2548 [RFC6980] Gont, F., "Security Implications of IPv6 Fragmentation 2549 with IPv6 Neighbor Discovery", RFC 6980, August 2013. 2551 [RFC7078] Matsumoto, A., Fujisaki, T., and T. Chown, "Distributing 2552 Address Selection Policy Using DHCPv6", RFC 7078, January 2553 2014. 2555 Author's Address 2557 Fred L. Templin (editor) 2558 Boeing Research & Technology 2559 P.O. Box 3707 2560 Seattle, WA 98124 2561 USA 2563 Email: fltemplin@acm.org