idnits 2.17.1 draft-templin-aerolink-49.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 == Line 810 has weird spacing: '...Version a 4-b...' -- The document date (December 18, 2014) is 3416 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) == Missing Reference: 'I-D.taylor-v6ops-fragdrop' is mentioned on line 952, but not defined == Unused Reference: 'RFC0768' is defined on line 2710, but no explicit reference was found in the text == Unused Reference: 'RFC4862' is defined on line 2749, but no explicit reference was found in the text == Unused Reference: 'RFC0879' is defined on line 2781, but no explicit reference was found in the text == Unused Reference: 'RFC4994' is defined on line 2857, but no explicit reference was found in the text == Unused Reference: 'RFC5720' is defined on line 2889, but no explicit reference was found in the text == Unused Reference: 'RFC6146' is defined on line 2904, but no explicit reference was found in the text == Unused Reference: 'RFC6204' is defined on line 2911, but no explicit reference was found in the text == Unused Reference: 'RFC6355' is defined on line 2922, but no explicit reference was found in the text == Unused Reference: 'RFC6691' is defined on line 2930, but no explicit reference was found in the text == Unused Reference: 'RFC6980' is defined on line 2949, but no explicit reference was found in the text == Unused Reference: 'RFC7078' is defined on line 2952, 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-sedhcpv6-05 == Outdated reference: A later version (-09) exists of draft-vandevelde-idr-remote-next-hop-08 -- Obsolete informational reference (is this intentional?): RFC 879 (Obsoleted by RFC 7805, RFC 9293) -- Obsolete informational reference (is this intentional?): RFC 1981 (Obsoleted by RFC 8201) -- 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 (~~), 17 warnings (==), 7 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, December 18, 2014 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: June 21, 2015 10 Transmission of IP Packets over AERO Links 11 draft-templin-aerolink-49.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 June 21, 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 Link Node Types . . . . . . . . . . . . . . . . . . 7 68 3.3. AERO Addresses . . . . . . . . . . . . . . . . . . . . . 8 69 3.4. AERO Interface Characteristics . . . . . . . . . . . . . 9 70 3.5. AERO Link Registration . . . . . . . . . . . . . . . . . 11 71 3.6. AERO Interface Initialization . . . . . . . . . . . . . . 11 72 3.6.1. AERO Relay Behavior . . . . . . . . . . . . . . . . . 11 73 3.6.2. AERO Server Behavior . . . . . . . . . . . . . . . . 12 74 3.6.3. AERO Client Behavior . . . . . . . . . . . . . . . . 12 75 3.6.4. AERO Forwarding Agent Behavior . . . . . . . . . . . 13 76 3.7. AERO Link Routing System . . . . . . . . . . . . . . . . 13 77 3.8. AERO Interface Neighbor Cache Maintenace . . . . . . . . 14 78 3.9. AERO Interface Sending Algorithm . . . . . . . . . . . . 15 79 3.10. AERO Interface Encapsulation and Re-encapsulation . . . . 17 80 3.11. AERO Interface Decapsulation . . . . . . . . . . . . . . 19 81 3.12. AERO Interface Data Origin Authentication . . . . . . . . 19 82 3.13. AERO Interface MTU and Fragmentation . . . . . . . . . . 20 83 3.13.1. Accommodating Large Control Messages . . . . . . . . 23 84 3.13.2. Integrity . . . . . . . . . . . . . . . . . . . . . 23 85 3.14. AERO Interface Error Handling . . . . . . . . . . . . . . 25 86 3.15. AERO Router Discovery, Prefix Delegation and Address 87 Configuration . . . . . . . . . . . . . . . . . . . . . . 29 88 3.15.1. AERO DHCPv6 Service Model . . . . . . . . . . . . . 29 89 3.15.2. AERO Client Behavior . . . . . . . . . . . . . . . . 29 90 3.15.3. AERO Server Behavior . . . . . . . . . . . . . . . . 32 91 3.16. AERO Forwarding Agent Discovery . . . . . . . . . . . . . 34 92 3.17. AERO Intradomain Route Optimization . . . . . . . . . . . 34 93 3.17.1. Reference Operational Scenario . . . . . . . . . . . 35 94 3.17.2. Concept of Operations . . . . . . . . . . . . . . . 36 95 3.17.3. Message Format . . . . . . . . . . . . . . . . . . . 36 96 3.17.4. Sending Predirects . . . . . . . . . . . . . . . . . 37 97 3.17.5. Re-encapsulating and Relaying Predirects . . . . . . 39 98 3.17.6. Processing Predirects and Sending Redirects . . . . 39 99 3.17.7. Re-encapsulating and Relaying Redirects . . . . . . 41 100 3.17.8. Processing Redirects . . . . . . . . . . . . . . . . 42 101 3.17.9. Server-Oriented Redirection . . . . . . . . . . . . 42 102 3.18. Neighbor Unreachability Detection (NUD) . . . . . . . . . 43 103 3.19. Mobility Management . . . . . . . . . . . . . . . . . . . 44 104 3.19.1. Announcing Link-Layer Address Changes . . . . . . . 44 105 3.19.2. Bringing New Links Into Service . . . . . . . . . . 45 106 3.19.3. Removing Existing Links from Service . . . . . . . . 46 107 3.19.4. Moving to a New Server . . . . . . . . . . . . . . . 46 108 3.20. Proxy AERO . . . . . . . . . . . . . . . . . . . . . . . 47 109 3.21. Extending AERO Links Through Security Gateways . . . . . 49 110 3.22. Extending IPv6 AERO Links to the Internet . . . . . . . . 51 111 3.23. Encapsulation Protocol Version Considerations . . . . . . 54 112 3.24. Multicast Considerations . . . . . . . . . . . . . . . . 55 113 3.25. Operation on AERO Links Without DHCPv6 Services . . . . . 55 114 3.26. Operation on Server-less AERO Links . . . . . . . . . . . 55 115 3.27. Manually-Configured AERO Tunnels . . . . . . . . . . . . 55 116 3.28. Intradomain Routing . . . . . . . . . . . . . . . . . . . 56 117 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 56 118 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 56 119 6. Security Considerations . . . . . . . . . . . . . . . . . . . 56 120 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 57 121 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 58 122 8.1. Normative References . . . . . . . . . . . . . . . . . . 58 123 8.2. Informative References . . . . . . . . . . . . . . . . . 59 124 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 63 126 1. Introduction 128 This document specifies the operation of IP over tunnel virtual links 129 using Asymmetric Extended Route Optimization (AERO). The AERO link 130 can be used for tunneling to neighboring nodes over either IPv6 or 131 IPv4 networks, i.e., AERO views the IPv6 and IPv4 networks as 132 equivalent links for tunneling. Nodes attached to AERO links can 133 exchange packets via trusted intermediate routers that provide 134 forwarding services to reach off-link destinations and redirection 135 services for route optimization that addresses the requirements 136 outlined in [RFC5522]. 138 AERO provides an IPv6 link-local address format known as the AERO 139 address that supports operation of the IPv6 Neighbor Discovery (ND) 140 [RFC4861] protocol and links IPv6 ND to IP forwarding. Admission 141 control and provisioning are supported by the Dynamic Host 142 Configuration Protocol for IPv6 (DHCPv6) [RFC3315], and node mobility 143 is naturally supported through dynamic neighbor cache updates. 144 Although DHCPv6 and IPv6 ND message signalling is used in the control 145 plane, both IPv4 and IPv6 can be used in the data plane. The 146 remainder of this document presents the AERO specification. 148 2. Terminology 150 The terminology in the normative references applies; the following 151 terms are defined within the scope of this document: 153 AERO link 154 a Non-Broadcast, Multiple Access (NBMA) tunnel virtual overlay 155 configured over a node's attached IPv6 and/or IPv4 networks. All 156 nodes on the AERO link appear as single-hop neighbors from the 157 perspective of the virtual overlay. 159 AERO interface 160 a node's attachment to an AERO link. 162 AERO address 163 an IPv6 link-local address constructed as specified in Section 3.3 164 and assigned to a Client's AERO interface. 166 AERO node 167 a node that is connected to an AERO link and that participates in 168 IPv6 ND and DHCPv6 messaging over the link. 170 AERO Client ("Client") 171 a node that assigns AERO addresses to an AERO interface and 172 receives IP prefixes via a DHCPv6 Prefix Delegation (PD) exchanges 173 with one or more AERO Servers. 175 AERO Server ("Server") 176 a node that configures an AERO interface to provide default 177 forwarding and DHCPv6 services for AERO Clients. The Server 178 assigns the IPv6 link-local subnet router anycast address (fe80::) 179 to the AERO interface and also assigns an administratively 180 assigned IPv6 link-local unicast address used for operation of 181 DHCPv6 and the IPv6 ND protocol. 183 AERO Relay ("Relay") 184 a node that configures an AERO interface to relay IP packets 185 between nodes on the same AERO link and/or forward IP packets 186 between the AERO link and the native Internetwork. The Relay 187 assigns an administratively assigned IPv6 link-local unicast 188 address to the AERO interface the same as for a Server. 190 AERO Forwarding Agent ("Forwarding Agent") 191 a node that performs data plane forwarding services as a companion 192 to other AERO nodes. 194 ingress tunnel endpoint (ITE) 195 an AERO interface endpoint that injects tunneled packets into an 196 AERO link. 198 egress tunnel endpoint (ETE) 199 an AERO interface endpoint that receives tunneled packets from an 200 AERO link. 202 underlying network 203 a connected IPv6 or IPv4 network routing region over which the 204 tunnel virtual overlay is configured. A typical example is an 205 enterprise network. 207 underlying interface 208 an AERO node's interface point of attachment to an underlying 209 network. 211 link-layer address 212 an IP address assigned to an AERO node's underlying interface. 213 When UDP encapsulation is used, the UDP port number is also 214 considered as part of the link-layer address. Link-layer 215 addresses are used as the encapsulation header source and 216 destination addresses. 218 network layer address 219 the source or destination address of the encapsulated IP packet. 221 end user network (EUN) 222 an internal virtual or external edge IP network that an AERO 223 Client connects to the rest of the network via the AERO interface. 225 AERO Service Prefix (ASP) 226 an IP prefix associated with the AERO link and from which AERO 227 Client Prefixes (ACPs) are derived (for example, the IPv6 ACP 228 2001:db8:1:2::/64 is derived from the IPv6 ASP 2001:db8::/32). 230 AERO Client Prefix (ACP) 231 a more-specific IP prefix taken from an ASP and delegated to a 232 Client. 234 Throughout the document, the simple terms "Client", "Server" and 235 "Relay" refer to "AERO Client", "AERO Server" and "AERO Relay", 236 respectively. Capitalization is used to distinguish these terms from 237 DHCPv6 client/server/relay [RFC3315]. 239 The terminology of [RFC4861] (including the names of node variables 240 and protocol constants) applies to this document. Also throughout 241 the document, the term "IP" is used to generically refer to either 242 Internet Protocol version (i.e., IPv4 or IPv6). 244 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 245 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 246 document are to be interpreted as described in [RFC2119]. 248 3. Asymmetric Extended Route Optimization (AERO) 250 The following sections specify the operation of IP over Asymmetric 251 Extended Route Optimization (AERO) links: 253 3.1. AERO Link Reference Model 255 .-(::::::::) 256 .-(:::: IP ::::)-. 257 (:: Internetwork ::) 258 `-(::::::::::::)-' 259 `-(::::::)-' 260 | 261 +--------------+ +--------+-------+ +--------------+ 262 |AERO Server S1| | AERO Relay R1 | |AERO Server S2| 263 | Nbr: C1; R1 | | Nbr: S1; S2 | | Nbr: C2; R1 | 264 | default->R1 | |(H1->S1; H2->S2)| | default->R1 | 265 | H1->C1 | +--------+-------+ | H2->C2 | 266 +-------+------+ | +------+-------+ 267 | | | 268 X---+---+-------------------+------------------+---+---X 269 | AERO Link | 270 +-----+--------+ +--------+-----+ 271 |AERO Client C1| |AERO Client C2| 272 | Nbr: S1 | | Nbr: S2 | 273 | default->S1 | | default->S2 | 274 +--------------+ +--------------+ 275 .-. .-. 276 ,-( _)-. ,-( _)-. 277 .-(_ IP )-. .-(_ IP )-. 278 (__ EUN ) (__ EUN ) 279 `-(______)-' `-(______)-' 280 | | 281 +--------+ +--------+ 282 | Host H1| | Host H2| 283 +--------+ +--------+ 285 Figure 1: AERO Link Reference Model 287 Figure 1 above presents the AERO link reference model. In this 288 model: 290 o Relay R1 acts as a default router for its associated Servers S1 291 and S2, and connects the AERO link to the rest of the IP 292 Internetwork 294 o Servers S1 and S2 associate with Relay R1 and also act as default 295 routers for their associated Clients C1 and C2. 297 o Clients C1 and C2 associate with Servers S1 and S2, respectively 298 and also act as default routers for their associated EUNs 300 o Hosts H1 and H2 attach to the EUNs served by Clients C1 and C2, 301 respectively 303 In common operational practice, there may be many additional Relays, 304 Servers and Clients. (Although not shown in the figure, AERO 305 Forwarding Agents may also be provided for data plane forwarding 306 offload services.) 308 3.2. AERO Link Node Types 310 AERO Relays provide default forwarding services to AERO Servers. 311 Relays forward packets between Servers connected to the same AERO 312 link and also forward packets between the AERO link and the native IP 313 Internetwork. Relays present the AERO link to the native 314 Internetwork as a set of one or more AERO Service Prefixes (ASPs) and 315 serve as a gateway between the AERO link and the Internetwork. AERO 316 Relays maintain an AERO interface neighbor cache entry for each AERO 317 Server, and maintain an IP forwarding table entry for each AERO 318 Client Prefix (ACP). 320 AERO Servers provide default forwarding services to AERO Clients. 321 Each Server also peers with each Relay in a dynamic routing protocol 322 instance to advertise its list of associated ACPs. Servers configure 323 a DHCPv6 server function to facilitate Prefix Delegation (PD) 324 exchanges with Clients. Each delegated prefix becomes an ACP taken 325 from an ASP. Servers forward packets between AERO interface 326 neighbors only, i.e., and not between the AERO link and the native IP 327 Internetwork unless they are also configured as a Relay. AERO 328 Servers maintain an AERO interface neighbor cache entry for each AERO 329 Relay. They also maintain both a neighbor cache entry and an IP 330 forwarding table entry for each of their associated Clients. 332 AERO Clients act as requesting routers to receive ACPs through DHCPv6 333 PD exchanges with AERO Servers over the AERO link and sub-delegate 334 portions of their ACPs to EUN interfaces. (Each Client MAY associate 335 with a single Server or with multiple Servers, e.g., for fault 336 tolerance, load balancing, etc.) Each IPv6 Client receives at least 337 a /64 IPv6 ACP, and may receive even shorter prefixes. Similarly, 338 each IPv4 Client receives at least a /32 IPv4 ACP (i.e., a singleton 339 IPv4 address), and may receive even shorter prefixes. AERO Clients 340 maintain an AERO interface neighbor cache entry for each of their 341 associated Servers as well as for each of their correspondent 342 Clients. 344 AERO Clients that act as hosts typically configure a TUN/TAP 345 interface [TUNTAP] as a point-to-point linkage between the IP layer 346 and the AERO interface. The IP layer therefore sees only the TUN/TAP 347 interface, while the AERO interface provides an intermediate conduit 348 between the TUN/TAP interface and the underlying interfaces. AERO 349 Clients that act as hosts assign one or more IP addresses from their 350 ACPs to the TUN/TAP interface, i.e., and not to the AERO interface. 352 AERO Forwarding Agents provide data plane forwarding services as 353 companions to other AERO nodes. Note that while all Relays, Servers 354 and Clients are required to perform both control and data plane 355 operations on their own behalf, they may optionally enlist the 356 services of special-purpose Forwarding Agents to offload performance- 357 intensive traffic. 359 3.3. AERO Addresses 361 An AERO address is an IPv6 link-local address with an embedded ACP 362 and assigned to a Client's AERO interface. The AERO address is 363 formed as follows: 365 fe80::[ACP] 367 For IPv6, the AERO address begins with the prefix fe80::/64 and 368 includes in its interface identifier the base prefix taken from the 369 Client's IPv6 ACP. The base prefix is determined by masking the ACP 370 with the prefix length. For example, if the AERO Client receives the 371 IPv6 ACP: 373 2001:db8:1000:2000::/56 375 it constructs its AERO address as: 377 fe80::2001:db8:1000:2000 379 For IPv4, the AERO address is formed from the lower 64 bits of an 380 IPv4-mapped IPv6 address [RFC4291] that includes the base prefix 381 taken from the Client's IPv4 ACP. For example, if the AERO Client 382 receives the IPv4 ACP: 384 192.0.2.32/28 386 it constructs its AERO address as: 388 fe80::FFFF:192.0.2.32 390 The AERO address remains stable as the Client moves between 391 topological locations, i.e., even if its link-layer addresses change. 393 NOTE: In some cases, prospective neighbors may not have advanced 394 knowledge of the Client's ACP length and may therefore send initial 395 IPv6 ND messages with an AERO destination address that matches the 396 ACP but does not correspond to the base prefix. In that case, the 397 Client MUST accept the address as equivalent to the base address, but 398 then use the base address as the source address of any IPv6 ND 399 message replies. For example, if the Client receives the IPv6 ACP 400 2001:db8:1000:2000::/56 then subsequently receives an IPv6 ND message 401 with destination address fe80::2001:db8:1000:2001, it accepts the 402 message but uses fe80::2001:db8:1000:2000 as the source address of 403 any IPv6 ND replies. 405 3.4. AERO Interface Characteristics 407 AERO interfaces use encapsulation (see Section 3.10) to exchange 408 packets with neighbors attached to the AERO link. AERO interfaces 409 maintain a neighbor cache, and AERO Clients and Servers use an 410 adaptation of standard unicast IPv6 ND messaging. AERO interfaces 411 use unicast Neighbor Solicitation (NS), Neighbor Advertisement (NA), 412 Router Solicitation (RS) and Router Advertisement (RA) messages the 413 same as for any IPv6 link. AERO interfaces use two redirection 414 message types -- the first known as a Predirect message and the 415 second being the standard Redirect message (see Section 3.17). AERO 416 links further use link-local-only addressing; hence, AERO nodes 417 ignore any Prefix Information Options (PIOs) they may receive in RA 418 messages over an AERO interface. 420 AERO interface ND messages include one or more Source/Target Link- 421 Layer Address Options (S/TLLAOs) formatted as shown in Figure 2: 423 0 1 2 3 424 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 1 425 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 426 | Type = 2 | Length = 3 | Reserved | 427 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 428 | Link ID | Preference | UDP Port Number | 429 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 430 | | 431 +-- --+ 432 | | 433 +-- IP Address --+ 434 | | 435 +-- --+ 436 | | 437 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 439 Figure 2: AERO Source/Target Link-Layer Address Option (S/TLLAO) 440 Format 442 In this format, Link ID is an integer value between 0 and 255 443 corresponding to an underlying interface of the target node, and 444 Preference is an integer value between 0 and 255 indicating the 445 node's preference for this underlying interface (with 255 being the 446 highest preference, 1 being the lowest, and 0 meaning "link 447 disabled"). UDP Port Number and IP Address are set to the addresses 448 used by the target node when it sends encapsulated packets over the 449 underlying interface. When the encapsulation IP address family is 450 IPv4, IP Address is formed as an IPv4-mapped IPv6 address [RFC4291]. 452 AERO interfaces may be configured over multiple underlying 453 interfaces. For example, common mobile handheld devices have both 454 wireless local area network ("WLAN") and cellular wireless links. 455 These links are typically used "one at a time" with low-cost WLAN 456 preferred and highly-available cellular wireless as a standby. In a 457 more complex example, aircraft frequently have many wireless data 458 link types (e.g. satellite-based, terrestrial, air-to-air 459 directional, etc.) with diverse performance and cost properties. 461 If a Client's multiple underlying interfaces are used "one at a time" 462 (i.e., all other interfaces are in standby mode while one interface 463 is active), then Redirect, Predirect and unsolicited NA messages 464 include only a single TLLAO with Link ID set to a constant value. 466 If the Client has multiple active underlying interfaces, then from 467 the perspective of IPv6 ND it would appear to have a single link- 468 local address with multiple link-layer addresses. In that case, 469 Redirect, Predirect and unsolicited NA messages MAY include multiple 470 TLLAOs -- each with a different Link ID that corresponds to a 471 specific underlying interface of the Client. 473 3.5. AERO Link Registration 475 When an administrative authority first deploys a set of AERO Relays 476 and Servers that comprise an AERO link, they also assign a unique 477 domain name for the link, e.g., "example.com". Next, if 478 administrative policy permits Clients within the domain to serve as 479 correspondent nodes for Internet mobile nodes, the administrative 480 authority adds a Fully Qualified Domain Name (FQDN) for each of the 481 AERO link's ASPs to the Domain Name System (DNS) [RFC1035]. The FQDN 482 is based on the suffix "aero.linkupnetworks.net" with a wildcard- 483 terminated reverse mapping of the ASP [RFC3596][RFC4592], and 484 resolves to a DNS PTR resource record. For example, for the ASP 485 '2001:db8:1::/48' within the domain name "example.com", the DNS 486 database contains: 488 '*.1.0.0.0.8.b.d.0.1.0.0.2.aero.linkupnetworks.net. PTR example.com' 490 This DNS registration advertises the AERO link's ASPs to prospective 491 mobile nodes. 493 3.6. AERO Interface Initialization 495 3.6.1. AERO Relay Behavior 497 When a Relay enables an AERO interface, it first assigns an 498 administratively-assigned link-local address fe80::ID to the 499 interface. Each fe80::ID address MUST be unique among all Relays and 500 Servers on the link, and MUST NOT collide with any potential AERO 501 addresses. The addresses are typically taken from the range 502 fe80::/96, e.g., as fe80::1, fe80::2, fe80::3, etc. The Relay then 503 engages in a dynamic routing protocol session with all Servers on the 504 link (see: Section 3.7), and advertises the set of ASPs into the 505 native IP Internetwork. 507 Each Relay subsequently maintains an IP forwarding table entry for 508 each Client-Server association, and maintains a neighbor cache entry 509 for each Server on the link. Relays do not require the use of 510 Neighbor Unreachability Detection (NUD) messaging (see: Section 3.18) 511 since the dynamic routing protocol already provides Relay-Server 512 reachability information. At a minimum, however, Relays perform NS/ 513 NA, e.g., to support AERO interface MTU determination (see: 514 Section 3.13). 516 3.6.2. AERO Server Behavior 518 When a Server enables an AERO interface, it assigns the address 519 fe80:: to the interface as a link-local Subnet Router Anycast 520 address, and also assigns an administratively assigned link-local 521 address fe80::ID the same as for Relays. (The Server then accepts 522 DHCPv6 and IPv6 ND control messages destined to either the fe80:: or 523 fe80::ID addresses, but always uses fe80::ID as the source address in 524 the control messages it generates.) The Server further configures a 525 DHCPv6 server function to facilitate DHCPv6 PD exchanges with AERO 526 Clients. The Server maintains a neighbor cache entry for each Relay 527 on the link, and manages per-Client neighbor cache entries and IP 528 forwarding table entries based on control message exchanges. Each 529 Server also engages in a dynamic routing protocol with each Relay on 530 the link (see: Section 3.7). 532 When the Server receives an NS/RS message on the AERO interface it 533 returns an NA/RA message but does not update the neighbor cache. The 534 Server further provides a simple conduit between AERO interface 535 neighbors. Therefore, packets enter the Server's AERO interface from 536 the link layer and are forwarded back out the link layer without ever 537 leaving the AERO interface and therefore without ever disturbing the 538 network layer. 540 3.6.3. AERO Client Behavior 542 When a Client enables an AERO interface, it invokes DHCPv6 PD to 543 receive an ACP from an AERO Server. Next, it assigns the 544 corresponding AERO address to the AERO interface and creates a 545 neighbor cache entry for the Server, i.e., the PD exchange bootstraps 546 the provisioning of a unique link-local address. The Client 547 maintains a neighbor cache entry for each of its Servers and each of 548 its active correspondent Clients. When the Client receives Redirect/ 549 Predirect messages on the AERO interface it updates or creates 550 neighbor cache entries, including link-layer address information. 551 Unsolicited NA messages update the cached link-layer addresses for 552 correspondent Clients (e.g., following a link-layer address change 553 due to node mobility) but do not create new neighbor cache entries. 554 NS/NA messages used for NUD update timers in existing neighbor cache 555 entires but do not update link-layer addresses nor create new 556 neighbor cache entries. 558 Finally, the Client need not maintain any IP forwarding table entries 559 for its Servers or correspondent Clients. Instead, it can set a 560 single "route-to-interface" default route in the IP forwarding table, 561 and all forwarding decisions can be made within the AERO interface 562 based on neighbor cache entries. (On systems in which adding a 563 default route would violate security policy, the default route could 564 instead be installed via a "synthesized RA", e.g., as discussed in 565 Section 3.15.2.) 567 3.6.4. AERO Forwarding Agent Behavior 569 When a Forwarding Agent enables an AERO interface, it assigns the 570 same link-local address(es) as the companion AERO node that manages 571 AERO control messaging services. The Forwarding Agent thereafter 572 provides data plane forwarding services based solely on the 573 forwarding information assigned to it by the companion AERO node. 574 AERO Forwarding Agents perform NS/NA messaging, i.e., the same as for 575 AERO Relays. 577 3.7. AERO Link Routing System 579 Relays require full topology knowledge of all ACP/Server 580 associations, while individual Servers at a minimum only need to know 581 the ACPs for their current set of associated Clients. This is 582 accomplished through the use of an internal instance of the Border 583 Gateway Protocol (BGP) [RFC4271] coordinated between Servers and 584 Relays. This internal BGP instance does not interact with the public 585 Internet BGP instance; therefore, the AERO link is presented to the 586 IP Internetwork as a small set of ASPs as opposed to the full set of 587 individual ACPs. 589 In a reference BGP arrangement, each AERO Server is configured as an 590 Autonomous System Border Router (ASBR) for a stub Autonomous System 591 (AS) (possibly using a private AS Number (ASN) [RFC1930]), and each 592 Server further peers with each Relay but does not peer with other 593 Servers. Similarly, Relays do not peer with each other, since they 594 will reliably receive all updates from all Servers and will therefore 595 have a consistent view of the AERO link ACP delegations. 597 Each Server maintains a working set of associated ACPs, and 598 dynamically announces new ACPs and withdraws departed ACPs in its BGP 599 updates to Relays. Clients are expected to remain associated with 600 their current Servers for extended timeframes, however Servers SHOULD 601 selectively suppress BGP updates for impatient Clients that 602 repeatedly associate and disassociate with them in order to reduce 603 routing churn. 605 In some environments, Relays need not send BGP updates to Servers 606 since Servers can always use Relays as default routers, however this 607 presents a data/control plane performance tradeoff. In environments 608 where sustained packet forwarding over Relays is undesirable, Relays 609 can instead report ACPs to Servers while including a BGP Remote-Next- 610 Hop [I-D.vandevelde-idr-remote-next-hop]. The Server then creates a 611 neighbor cache entry for each ACP with the Remote-Next-Hop as the 612 link-layer address to enable Server-to-Server route optimization. 614 3.8. AERO Interface Neighbor Cache Maintenace 616 Each AERO interface maintains a conceptual neighbor cache that 617 includes an entry for each neighbor it communicates with on the AERO 618 link, the same as for any IPv6 interface [RFC4861]. AERO interface 619 neighbor cache entires are said to be one of "permanent", "static" or 620 "dynamic". 622 Permanent neighbor cache entries are created through explicit 623 administrative action; they have no timeout values and remain in 624 place until explicitly deleted. AERO Relays maintain a permanent 625 neighbor cache entry for each Server on the link, and AERO Servers 626 maintain a permanent neighbor cache entry for each Relay. Each entry 627 maintains the mapping between the neighbor's fe80::ID network-layer 628 address and corresponding link-layer address. 630 Static neighbor cache entries are created though DHCPv6 PD exchanges 631 and remain in place for durations bounded by prefix lifetimes. AERO 632 Servers maintain a static neighbor cache entry for each ACP of each 633 of their associated Clients, and AERO Clients maintain a static 634 neighbor cache for each of their associated Servers. When an AERO 635 Server sends a DHCPv6 Reply message response to a Client's DHCPv6 636 Solicit/Request or Renew message, it creates or updates a static 637 neighbor cache entry based on the AERO address corresponding to the 638 Client's ACP as the network-layer address, the prefix lifetime as the 639 neighbor cache entry lifetime, the Client's encapsulation IP address 640 and UDP port number as the link-layer address and the prefix length 641 as the length to apply to the AERO address. When an AERO Client 642 receives a DHCPv6 Reply message from a Server, it creates or updates 643 a static neighbor cache entry based on the Reply message link-local 644 source address as the network-layer address, the prefix lifetime as 645 the neighbor cache entry lifetime, and the encapsulation IP source 646 address and UDP source port number as the link-layer address. 648 Dynamic neighbor cache entries are created or updated based on 649 receipt of an IPv6 ND message, and are garbage-collected if not used 650 within a short timescale. AERO Clients maintain dynamic neighbor 651 cache entries for each of their active correspondent Client ACPs with 652 lifetimes based on IPv6 ND messaging constants. When an AERO Client 653 receives a valid Predirect message it creates or updates a dynamic 654 neighbor cache entry for the Predirect target network-layer and link- 655 layer addresses plus prefix length. The node then sets an 656 "AcceptTime" variable in the neighbor cache entry to ACCEPT_TIME 657 seconds and uses this value to determine whether packets received 658 from the correspondent can be accepted. When an AERO Client receives 659 a valid Redirect message it creates or updates a dynamic neighbor 660 cache entry for the Redirect target network-layer and link-layer 661 addresses plus prefix length. The Client then sets a "ForwardTime" 662 variable in the neighbor cache entry to FORWARD_TIME seconds and uses 663 this value to determine whether packets can be sent directly to the 664 correspondent. The Client also sets a "MaxRetry" variable to 665 MAX_RETRY to limit the number of keepalives sent when a correspondent 666 may have gone unreachable. 668 For dynamic neighbor cache entries, when an AERO Client receives a 669 valid NS message it (re)sets AcceptTime for the neighbor to 670 ACCEPT_TIME. When an AERO Client receives a valid solicited NA 671 message, it (re)sets ForwardTime for the neighbor to FORWARD_TIME and 672 sets MaxRetry to MAX_RETRY. When an AERO Client receives a valid 673 unsolicited NA message, it updates the correspondent's link-layer 674 addresses but DOES NOT reset AcceptTime, ForwardTime or MaxRetry. 676 It is RECOMMENDED that FORWARD_TIME be set to the default constant 677 value 30 seconds to match the default REACHABLE_TIME value specified 678 for IPv6 ND [RFC4861]. 680 It is RECOMMENDED that ACCEPT_TIME be set to the default constant 681 value 40 seconds to allow a 10 second window so that the AERO 682 redirection procedure can converge before AcceptTime decrements below 683 FORWARD_TIME. 685 It is RECOMMENDED that MAX_RETRY be set to 3 the same as described 686 for IPv6 ND address resolution in Section 7.3.3 of [RFC4861]. 688 Different values for FORWARD_TIME, ACCEPT_TIME, and MAX_RETRY MAY be 689 administratively set, if necessary, to better match the AERO link's 690 performance characteristics; however, if different values are chosen, 691 all nodes on the link MUST consistently configure the same values. 692 Most importantly, ACCEPT_TIME SHOULD be set to a value that is 693 sufficiently longer than FORWARD_TIME to allow the AERO redirection 694 procedure to converge. 696 3.9. AERO Interface Sending Algorithm 698 IP packets enter a node's AERO interface either from the network 699 layer (i.e., from a local application or the IP forwarding system), 700 or from the link layer (i.e., from the AERO tunnel virtual link). 701 Packets that enter the AERO interface from the network layer are 702 encapsulated and admitted into the AERO link, i.e., they are 703 tunnelled to an AERO interface neighbor. Packets that enter the AERO 704 interface from the link layer are either re-admitted into the AERO 705 link or delivered to the network layer where they are subject to 706 either local delivery or IP forwarding. Since each AERO node may 707 have only partial information about neighbors on the link, AERO 708 interfaces may forward packets with link-local destination addresses 709 at a layer below the network layer. This means that AERO nodes act 710 as both IP routers and sub-IP layer forwarding agents. AERO 711 interface sending considerations for Clients, Servers and Relays are 712 given below. 714 When an IP packet enters a Client's AERO interface from the network 715 layer, if the destination is covered by an ASP the Client searches 716 for a dynamic neighbor cache entry with a non-zero ForwardTime and an 717 AERO address that matches the packet's destination address. (The 718 destination address may be either an address covered by the 719 neighbor's ACP or the (link-local) AERO address itself.) If there is 720 a match, the Client uses a link-layer address in the entry as the 721 link-layer address for encapsulation then admits the packet into the 722 AERO link. If there is no match, the Client instead uses the link- 723 layer address of a neighboring Server as the link-layer address for 724 encapsulation. 726 When an IP packet enters a Server's AERO interface from the link 727 layer, if the destination is covered by an ASP the Server searches 728 for a neighbor cache entry with an AERO address that matches the 729 packet's destination address. (The destination address may be either 730 an address covered by the neighbor's ACP or the AERO address itself.) 731 If there is a match, the Server uses a link-layer address in the 732 entry as the link-layer address for encapsulation and re-admits the 733 packet into the AERO link. If there is no match, the Server instead 734 uses the link-layer address in a permanent neighbor cache entry for a 735 Relay as the link-layer address for encapsulation. 737 When an IP packet enters a Relay's AERO interface from the network 738 layer, the Relay searches its IP forwarding table for an entry that 739 is covered by an ASP and also matches the destination. If there is a 740 match, the Relay uses the link-layer address in a permanent neighbor 741 cache entry for a Server as the link-layer address for encapsulation 742 and admits the packet into the AERO link. When an IP packet enters a 743 Relay's AERO interface from the link-layer, if the destination is not 744 a link-local address and does not match an ASP the Relay removes the 745 packet from the AERO interface and uses IP forwarding to forward the 746 packet to the Internetwork. If the destination address is a link- 747 local address or a non-link-local address that matches an ASP, and 748 there is a more-specific ACP entry in the IP forwarding table, the 749 Relay uses the link-layer address in the corresponding neighbor cache 750 entry as the link-layer address for encapsulation and re-admits the 751 packet into the AERO link. When an IP packet enters a Relay's AERO 752 interface from either the network layer or link-layer, and the 753 packet's destination address matches an ASP but there is no more- 754 specific ACP entry, the Relay drops the packet and returns an ICMP 755 Destination Unreachable message (see: Section 3.14). 757 When an AERO Server receives a packet from a Relay via the AERO 758 interface, the Server MUST NOT forward the packet back to the same or 759 a different Relay. 761 When an AERO Relay receives a packet from a Server via the AERO 762 interface, the Relay MUST NOT forward the packet back to the same 763 Server. 765 When an AERO node re-admits a packet into the AERO link without 766 involving the network layer, the node MUST NOT decrement the network 767 layer TTL/Hop-count. 769 When an AERO node forwards a data packet to the primary link-layer 770 address of a neighbor, it may receive RA messages with one or more 771 SLLAOs that include the link-layer addresses of AERO Forwarding 772 Agents. The AERO node SHOULD record the link-layer addresses in the 773 neighbor cache entry for the neighbor and send subsequent data 774 packets via one of these addresses instead of the neighbor's primary 775 address (see: Section 3.16). 777 3.10. AERO Interface Encapsulation and Re-encapsulation 779 AERO interfaces encapsulate IP packets according to whether they are 780 entering the AERO interface from the network layer or if they are 781 being re-admitted into the same AERO link they arrived on. This 782 latter form of encapsulation is known as "re-encapsulation". 784 The AERO interface encapsulates packets per the base tunneling 785 specifications (e.g., [RFC2003], [RFC2473], [RFC2784], [RFC4213], 786 [RFC4301], [RFC5246], etc.) except that it inserts a UDP header 787 immediately following the IP encapsulation header. If there are no 788 additional encapsulation headers (and no fragmentation, 789 identification, checksum or signature is needed), the AERO interface 790 next encapsulates the IPv4 or IPv6 packet immediately following the 791 UDP header. In that case, the most significant four bits of the 792 encapsulated packet encode the value '4' for IPv4 or '6' for IPv6. 794 For all other encapsulations, the AERO interface MUST insert an AERO 795 Header between the UDP header and the next encapsulation header as 796 shown in Figure 3: 798 0 1 2 3 799 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 1 800 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 801 |Version|N|F|C|S| Next Header |Fragment Offset (13 bits)|Res|M| 802 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 803 | Identification (32 bits) | 804 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 805 | Checksum (16 bits) | Signature (variable length) : 806 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 808 Figure 3: AERO Header 810 Version a 4-bit "Version" field. MUST be 0. 812 N a 1-bit "Next Header" flag. Set to '1' if the "Next Header" field 813 is present; otherwise, set to '0' for "Experimental" per 814 [RFC6706]. "Next Header" encodes the IP protocol number 815 corresponding to the next header in the encapsulation immediately 816 following the AERO header. For example, "Next Header" encodes the 817 value '4' for IPv4, '17' for UDP, '41' for IPv6, '47' for GRE, 818 '50' for ESP, '51' for AH, etc. 820 F a 1-bit "Fragment Header" flag. Set to '1' if the "Fragment 821 Offset", "Res", "M", and "Identification" fields are present and 822 collectively referred to as the "AERO Fragment Header"; otherwise, 823 set to '0'. 825 C a 1-bit "Checksum" flag. Set to '1' if the "Checksum" field is 826 present; otherwise, set to '0'. The Checksum field itself then 827 contains a checksum of the IP/UDP/AERO encapsulation headers prior 828 to the Checksum field. 830 S a 1-bit "Signature" flag. Set to '1' if the "Signature" field is 831 present; otherwise, set to '0'. The Signature field itself then 832 contains a cryptographic signature of the encapsulated packet 833 following the Signature field. The signature is applied prior to 834 any fragmentation; hence' the Signature field only appears in the 835 first fragment of a fragmented packet. 837 During encapsulation, the AERO interface copies the "TTL/Hop Limit", 838 Type of Service/Traffic Class" [RFC2983] and "Congestion Experienced" 839 [RFC3168] values in the packet's IP header into the corresponding 840 fields in the encapsulation IP header. (When IPv6 is used as the 841 encapsulation protocol, the interface also sets the Flow Label value 842 in the encapsulation header per [RFC6438].) For packets undergoing 843 re-encapsulation, the AERO interface instead copies the "TTL/Hop 844 Limit", "Type of Service/Traffic Class", "Flow Label" and "Congestion 845 Experienced" values in the original encapsulation IP header into the 846 corresponding fields in the new encapsulation IP header, i.e., the 847 values are transferred between encapsulation headers and *not* copied 848 from the encapsulated packet's network-layer header. 850 The AERO interface next sets the UDP source port to a constant value 851 that it will use in each successive packet it sends, and sets the UDP 852 length field to the length of the encapsulated packet plus 8 bytes 853 for the UDP header itself, plus the length of the AERO header. For 854 packets sent via a Server, the AERO interface sets the UDP 855 destination port to 8060, i.e., the IANA-registered port number for 856 AERO. For packets sent to a correspondent Client, the AERO interface 857 sets the UDP destination port to the port value stored in the 858 neighbor cache entry for this correspondent. The AERO interface also 859 sets the UDP checksum field to zero (see: [RFC6935][RFC6936]) unless 860 an integrity check is required (see: Section 3.13.2). 862 The AERO interface next sets the IP protocol number in the 863 encapsulation header to 17 (i.e., the IP protocol number for UDP). 864 When IPv4 is used as the encapsulation protocol, the AERO interface 865 sets the DF bit as discussed in Section 3.13. The AERO interface 866 finally sets the AERO header fields as described in Figure 3. 868 3.11. AERO Interface Decapsulation 870 AERO interfaces decapsulate packets destined either to the node 871 itself or to a destination reached via an interface other than the 872 AERO interface the packet was received on. When the AERO interface 873 receives a UDP packet, it examines the first octet of the 874 encapsulated packet. 876 If the most significant four bits of the first octet encode the value 877 '4' (i.e., the IP version number value for IPv4) or the value '6' 878 (i.e., the IP version number value for IPv6), the AERO interface 879 discards the encapsulation headers and accepts the encapsulated 880 packet as an ordinary IPv6 or IPv4 data packet, respectively. If the 881 most significant four bits encode the value '0', however, the AERO 882 interface processes the packet according to the appropriate AERO 883 Header fields as specified in Figure 3. 885 3.12. AERO Interface Data Origin Authentication 887 AERO nodes employ simple data origin authentication procedures for 888 encapsulated packets they receive from other nodes on the AERO link. 889 In particular: 891 o AERO Relays and Servers accept encapsulated packets with a link- 892 layer source address that matches a permanent neighbor cache 893 entry. 895 o AERO Servers accept authentic encapsulated DHCPv6 messages from 896 Clients, and create or update a static neighbor cache entry for 897 the source based on the specific message type. 899 o AERO Servers accept encapsulated packets if there is a neighbor 900 cache entry with an AERO address that matches the packet's 901 network-layer source address and with a link-layer address that 902 matches the packet's link-layer source address. 904 o AERO Clients accept encapsulated packets if there is a static 905 neighbor cache entry with a link-layer source address that matches 906 the packet's link-layer source address. 908 o AERO Clients and Servers accept encapsulated packets if there is a 909 dynamic neighbor cache entry with an AERO address that matches the 910 packet's network-layer source address, with a link-layer address 911 that matches the packet's link-layer source address, and with a 912 non-zero AcceptTime. 914 Note that this simple data origin authentication is effective in 915 environments in which link-layer addresses cannot be spoofed. In 916 other environments, each AERO message must include a signature that 917 the recipient can use to authenticate the message origin. 919 3.13. AERO Interface MTU and Fragmentation 921 The AERO interface is the node's point of attachment to the AERO 922 link. AERO links over IP networks have a maximum link MTU of 64KB 923 minus the encapsulation overhead (termed here "ENCAPS"), since the 924 maximum packet size in the base IP specifications is 64KB 925 [RFC0791][RFC2460] (while IPv6 jumbograms can be up to 4GB, they are 926 considered optional for IPv6 nodes [RFC2675][RFC6434]). 928 IPv6 specifies a minimum link MTU of 1280 bytes [RFC2460]. This is 929 the minimum packet size the AERO interface MUST admit without 930 returning an ICMP Packet Too Big (PTB) message. Although IPv4 931 specifies a smaller minimum link MTU of 68 bytes [RFC0791], AERO 932 interfaces also observe a 1280 byte minimum for IPv4. Additionally, 933 the vast majority of links in the Internet configure an MTU of at 934 least 1500 bytes. Original source hosts have therefore become 935 conditioned to expect that IP packets up to 1500 bytes in length will 936 either be delivered to the final destination or a suitable PTB 937 message returned. However, PTB messages may be lost in the network 939 [RFC2923] resulting in failure of the IP MTU discovery mechanisms 940 [RFC1191][RFC1981]. 942 For these reasons, AERO interfaces MUST admit packets up to 1500 943 bytes in length even if some fragmentation is necessary. AERO 944 interfaces MAY admit even larger packets as long as they can be 945 accommodated without fragmentation. 947 For AERO links over IPv4, the IP ID field is only 16 bits in length, 948 meaning that fragmentation at high data rates could result in data 949 corruption due to reassembly misassociations [RFC6864][RFC4963] (see: 950 Section 3.13.2). For AERO links over both IPv4 and IPv6, studies 951 have also shown that IP fragments are dropped unconditionally over 952 some network paths [I-D.taylor-v6ops-fragdrop]. For these reasons, 953 when fragmentation is needed it is performed through the insertion of 954 an AERO fragmentation shim header (see: Section 3.10) using the 955 fragmentation procedures specified for IPv6 fragmentation in 956 [RFC2460]. Since the fragmentation shim header reduces the room 957 available for packet data, but the original source has no way to 958 control its insertion, the shim header length MUST be included in the 959 ENCAPS length even for packets in which the header does not appear. 961 The source AERO interface (i.e., the tunnel ingress) therefore sends 962 encapsulated packets to the destination AERO interface (i.e., the 963 tunnel egress) according to the following algorithm: 965 o For IP packets that are no larger than (1280-ENCAPS) bytes, the 966 tunnel ingress encapsulates the packet and admits it into the 967 tunnel without fragmentation. For IPv4 AERO links, the tunnel 968 ingress sets the Don't Fragment (DF) bit to 0 so that these 969 packets will be delivered to the tunnel egress even if there is a 970 restricting link in the path, i.e., unless lost due to congestion 971 or routing errors. 973 o For IP packets that are larger than (1280-ENCAPS) bytes but no 974 larger than 1500 bytes, the tunnel ingress encapsulates the packet 975 and inserts a fragmentation shim header above the UDP/IP 976 encapsulation headers. Next, the tunnel ingress uses the 977 fragmentation algorithm in [RFC2460] to break the packet into two 978 non-overlapping fragments where the first fragment (including 979 ENCAPS) is no larger than 1024 bytes and the second is no larger 980 than the first. Each fragment consists of identical UDP/IP 981 encapsulation headers, followed by the fragmentation shim header 982 followed by the fragment of the encapsulated packet itself. The 983 tunnel ingress then admits both fragments into the tunnel, and for 984 IPv4 sets the DF bit to 0 in the IP encapsulation header. These 985 fragmented encapsulated packets will be delivered to the tunnel 986 egress. 988 o For IPv4 packets that are larger than 1500 bytes and with the DF 989 bit set to 0, the tunnel ingress uses ordinary IPv4 fragmentation 990 to break the unencapsulated packet into a minimum number of non- 991 overlapping fragments where the first fragment is no larger than 992 1024-ENCAPS and all other fragments are no larger than the first 993 fragment. The tunnel ingress then encapsulates each fragment (and 994 for IPv4 sets the DF bit to 0) then admits them into the tunnel. 995 These encapsulated fragments will be delivered to the final 996 destination via the tunnel egress. 998 o For all other IP packets, if the packet is too large to enter the 999 underlying interface following encapsulation, the tunnel ingress 1000 drops the packet and returns a network-layer (L3) PTB message to 1001 the original source with MTU set to the larger of 1500 bytes or 1002 the underlying interface MTU minus ENCAPS. Otherwise, the tunnel 1003 ingress encapsulates the packet and admits it into the tunnel 1004 without fragmentation (and for IPv4 sets the DF bit to 1) and 1005 translates any link-layer (L2) PTB messages it may receive from 1006 the network into corresponding L3 PTB messages to send to the 1007 original source as specified in Section 3.14. Since both L2 and 1008 L3 PTB messages may be either lost or contain insufficient 1009 information, however, it is RECOMMENDED that original sources that 1010 send unfragmentable IP packets larger than 1500 bytes use 1011 Packetization Layer Path MTU Discovery (PLPMTUD) [RFC4821]. 1013 While sending packets according to the above algorithm, the tunnel 1014 ingress MAY also send 1500 byte or larger probe packets to determine 1015 whether they can reach the tunnel egress without fragmentation. If 1016 the probes succeed, the tunnel ingress can discontinue fragmentation 1017 and (for IPv4) set DF to 1. Since the path MTU within the tunnel may 1018 fluctuate due to routing changes, the tunnel ingress SHOULD continue 1019 to send additional probes subject to rate limiting and SHOULD process 1020 any L2 PTB messages as an indication that the path MTU may have 1021 decreased. If the path MTU within the tunnel becomes insufficient, 1022 the source MUST resume fragmentation. 1024 To construct a probe, the tunnel ingress prepares an NS message with 1025 a Nonce option plus trailing NULL padding octets added to the probe 1026 length without including the length of the padding in the IPv6 1027 Payload Length field, but with the length included in the 1028 encapsulating IP header. The tunnel ingress then encapsulates the 1029 padded NS message in the encapsulation headers (and for IPv4 sets DF 1030 to 1) then sends the message to the tunnel egress. If the tunnel 1031 egress returns a solicited NA message with a matching Nonce option, 1032 the tunnel ingress deems the probe successful. Note that in this 1033 process it is essential that probes follow equivalent paths to those 1034 used to convey actual data packets. This means that Equal Cost 1035 MultiPath (ECMP) and Link Aggregation Gateway (LAG) equipment is 1036 assumed to support identical MTUs along all paths. 1038 When the tunnel egress receives the fragments of a fragmented packet, 1039 it reassembles them into a whole packet per the reassembly algorithm 1040 in [RFC2460] then discards the fragmentation shim header. The tunnel 1041 egress therefore MUST be capable of reassembling packets up to 1042 1500+ENCAPS bytes in length; hence, it is RECOMMENDED that the tunnel 1043 egress be capable of reassembling at least 2KB. 1045 3.13.1. Accommodating Large Control Messages 1047 Control messages (i.e., IPv6 ND, DHCPv6, etc.) MUST be accommodated 1048 even if some fragmentation is necessary. These packets are therefore 1049 accommodated through a modification of the second rule in the above 1050 algorithm as follows: 1052 o For control messages that are larger than (1280-ENCAPS) bytes, the 1053 tunnel ingress encapsulates the packet and inserts a fragmentation 1054 shim header above the UDP/IP encapsulation headers. Next, the 1055 tunnel ingress uses the fragmentation algorithm in [RFC2460] to 1056 break the packet into a minimum number of non-overlapping 1057 fragments where the first fragment (including ENCAPS) is no larger 1058 than 1024 bytes and the remaining fragments are no larger than the 1059 first. The tunnel ingress then encapsulates each fragment (and 1060 for IPv4 sets the DF bit to 0) then admits them into the tunnel. 1062 Control messages that exceed the minimum reassembly size listed above 1063 rarely occur in the modern era, however the tunnel egress SHOULD be 1064 able to reassemble them if they do. This means that the tunnel 1065 egress SHOULD include a configuration knob allowing the operator to 1066 set a larger reassembly buffer size if large control messages become 1067 more common in the future. 1069 The tunnel ingress can send large control messages without 1070 fragmentation if there is assurance that large packets can traverse 1071 the tunnel without fragmentation. The tunnel ingress MAY send 1500 1072 byte or larger probe packets as specified above to determine a size 1073 for which fragmentation can be avoided. 1075 3.13.2. Integrity 1077 When fragmentation is needed, there must be assurance that reassembly 1078 can be safely conducted without incurring data corruption. Sources 1079 of corruption can include implementation errors, memory errors and 1080 misassociation of fragments from a first datagram with fragments of 1081 another datagram. The first two conditions (implementation and 1082 memory errors) are mitigated by modern systems and implementations 1083 that have demonstrated integrity through decades of operational 1084 practice. The third condition (reassembly misassociations) must be 1085 accounted for by AERO. 1087 The AERO fragmentation procedure described in the above algorithms 1088 reuses standard IPv6 fragmentation and reassembly code. Since the 1089 fragmentation shim header includes a 32-bit ID field, there would 1090 need to be 2^32 packets alive in the network before a second packet 1091 with a duplicate ID enters the system with the (remote) possibility 1092 for a reassembly misassociation. For 1280 byte packets, and for a 1093 maximum network lifetime value of 60 seconds[RFC2460], this means 1094 that the tunnel ingress would need to produce ~(7 *10^12) bits/sec in 1095 order for a duplication event to be possible. This exceeds the 1096 bandwidth of data link technologies of the modern era, but not 1097 necessarily so going forward into the future. Although wireless data 1098 links commonly used by AERO Clients support vastly lower data rates, 1099 the aggregate data rates between AERO Servers and Relays may be 1100 substantial. However, high speed data links in the network core are 1101 expected to configure larger MTUs, e.g., 4KB, 8KB or even larger such 1102 that unfragmented packets can be used. Hence, no integrity check is 1103 included to cover the AERO fragmentation and reassembly procedures. 1105 When the tunnel ingress sends an IPv4-encapsulated packet with the DF 1106 bit set to 0 in the above algorithms, there is a chance that the 1107 packet may be fragmented by an IPv4 router somewhere within the 1108 tunnel. Since the largest such packet is only 1280 bytes, however, 1109 it is very likely that the packet will traverse the tunnel without 1110 incurring a restricting link. Even when a link within the tunnel 1111 configures an MTU smaller than 1280 bytes, it is very likely that it 1112 does so due to limited performance characteristics [RFC3819]. This 1113 means that the tunnel would not be able to convey fragmented 1114 IPv4-encapsulated packets fast enough to produce reassembly 1115 misassociations, as discussed above. However, AERO must also account 1116 for the possibility of tunnel paths that include "poorly managed" 1117 IPv4 link MTUs since configurable MTUs can often be misconfigured. 1119 Since the IPv4 header includes only a 16-bit ID field, there would 1120 only need to be 2^16 packets alive in the network before a second 1121 packet with a duplicate ID enters the system. For 1280 byte packets, 1122 and for a maximum network lifetime value of 120 seconds[RFC0791], 1123 this means that the tunnel ingress would only need to produce ~(5 1124 *10^6) bits/sec in order for a duplication event to be possible - a 1125 value that is well within range for many modern wired and wireless 1126 data link technologies. 1128 Therefore, if there is strong operational assurance that no IPv4 1129 links capable of supporting data rates of 5Mbps or more configure an 1130 MTU smaller than 1280 the tunnel ingress MAY omit an integrity check 1131 for the IPv4 fragmentation and reassembly procedures; otherwise, the 1132 tunnel ingress SHOULD include an integrity check. When an upper 1133 layer encapsulation (e.g., IPsec) already includes an integrity 1134 check, the tunnel ingress need not include an additional check. 1135 Otherwise, the tunnel ingress calculates the UDP checksum over the 1136 encapsulated packet and writes the value into the UDP encapsulation 1137 header, i.e., instead of writing the value 0. The tunnel egress will 1138 then verify the UDP checksum and discard the packet if the checksum 1139 is incorrect. 1141 3.14. AERO Interface Error Handling 1143 When an AERO node admits encapsulated packets into the AERO 1144 interface, it may receive link-layer (L2) or network-layer (L3) error 1145 indications. 1147 An L2 error indication is an ICMP error message generated by a router 1148 on the path to the neighbor or by the neighbor itself. The message 1149 includes an IP header with the address of the node that generated the 1150 error as the source address and with the link-layer address of the 1151 AERO node as the destination address. 1153 The IP header is followed by an ICMP header that includes an error 1154 Type, Code and Checksum. For ICMPv6 [RFC4443], the error Types 1155 include "Destination Unreachable", "Packet Too Big (PTB)", "Time 1156 Exceeded" and "Parameter Problem". For ICMPv4 [RFC0792], the error 1157 Types include "Destination Unreachable", "Fragmentation Needed" (a 1158 Destination Unreachable Code that is analogous to the ICMPv6 PTB), 1159 "Time Exceeded" and "Parameter Problem". 1161 The ICMP header is followed by the leading portion of the packet that 1162 generated the error, also known as the "packet-in-error". For 1163 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1164 much of invoking packet as possible without the ICMPv6 packet 1165 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1166 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1167 "Internet Header + 64 bits of Original Data Datagram", however 1168 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1169 ICMP datagram SHOULD contain as much of the original datagram as 1170 possible without the length of the ICMP datagram exceeding 576 1171 bytes". 1173 The L2 error message format is shown in Figure 4: 1175 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1176 ~ ~ 1177 | L2 IP Header of | 1178 | error message | 1179 ~ ~ 1180 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1181 | L2 ICMP Header | 1182 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1183 ~ ~ P 1184 | IP and other encapsulation | a 1185 | headers of original L3 packet | c 1186 ~ ~ k 1187 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1188 ~ ~ t 1189 | IP header of | 1190 | original L3 packet | i 1191 ~ ~ n 1192 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1193 ~ ~ e 1194 | Upper layer headers and | r 1195 | leading portion of body | r 1196 | of the original L3 packet | o 1197 ~ ~ r 1198 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1200 Figure 4: AERO Interface L2 Error Message Format 1202 The AERO node rules for processing these L2 error messages is as 1203 follows: 1205 o When an AERO node receives an L2 Parameter Problem message, it 1206 processes the message the same as described as for ordinary ICMP 1207 errors in the normative references [RFC0792][RFC4443]. 1209 o When an AERO node receives persistent L2 IPv4 Time Exceeded 1210 messages, the IP ID field may be wrapping before earlier fragments 1211 have been processed. In that case, the node SHOULD begin 1212 including IPv4 integrity checks (see: Section 3.13.2). 1214 o When an AERO Client receives persistent L2 Destination Unreachable 1215 messages in response to tunneled packets that it sends to one of 1216 its dynamic neighbor correspondents, the Client SHOULD test the 1217 path to the correspondent using Neighbor Unreachability Detection 1218 (NUD) (see Section 3.18). If NUD fails, the Client SHOULD set 1219 ForwardTime for the corresponding dynamic neighbor cache entry to 1220 0 and allow future packets destined to the correspondent to flow 1221 through a Server. 1223 o When an AERO Client receives persistent L2 Destination Unreachable 1224 messages in response to tunneled packets that it sends to one of 1225 its static neighbor Servers, the Client SHOULD test the path to 1226 the Server using NUD. If NUD fails, the Client SHOULD delete the 1227 neighbor cache entry and attempt to associate with a new Server. 1229 o When an AERO Server receives persistent L2 Destination Unreachable 1230 messages in response to tunneled packets that it sends to one of 1231 its static neighbor Clients, the Server SHOULD test the path to 1232 the Client using NUD. If NUD fails, the Server SHOULD cancel the 1233 DHCPv6 PD lease for the Client's ACP, withdraw its route for the 1234 ACP from the AERO routing system and delete the neighbor cache 1235 entry (see Section 3.18 and Section 3.19). 1237 o When an AERO Relay or Server receives an L2 Destination 1238 Unreachable message in response to a tunneled packet that it sends 1239 to one of its permanent neighbors, it discards the message since 1240 the routing system is likely in a temporary transitional state 1241 that will soon re-converge. 1243 o When an AERO node receives an L2 PTB message, it translates the 1244 message into an L3 PTB message if possible (*) and forwards the 1245 message toward the original source as described below. 1247 To translate an L2 PTB message to an L3 PTB message, the AERO node 1248 first caches the MTU field value of the L2 ICMP header. The node 1249 next discards the L2 IP and ICMP headers, and also discards the 1250 encapsulation headers of the original L3 packet. Next the node 1251 encapsulates the included segment of the original L3 packet in an L3 1252 IP and ICMP header, and sets the ICMP header Type and Code values to 1253 appropriate values for the L3 IP protocol. In the process, the node 1254 writes the maximum of 1500 bytes and (L2 MTU - ENCAPS) into the MTU 1255 field of the L3 ICMP header. 1257 The node next writes the IP source address of the original L3 packet 1258 as the destination address of the L3 PTB message and determines the 1259 next hop to the destination. If the next hop is reached via the AERO 1260 interface, the node uses the IPv6 address "::" or the IPv4 address 1261 "0.0.0.0" as the IP source address of the L3 PTB message. Otherwise, 1262 the node uses one of its non link-local addresses as the source 1263 address of the L3 PTB message. The node finally calculates the ICMP 1264 checksum over the L3 PTB message and writes the Checksum in the 1265 corresponding field of the L3 ICMP header. The L3 PTB message 1266 therefore is formatted as follows: 1268 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1269 ~ ~ 1270 | L3 IP Header of | 1271 | error message | 1272 ~ ~ 1273 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1274 | L3 ICMP Header | 1275 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1276 ~ ~ p 1277 | IP header of | k 1278 | original L3 packet | t 1279 ~ ~ 1280 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ i 1281 ~ ~ n 1282 | Upper layer headers and | 1283 | leading portion of body | e 1284 | of the original L3 packet | r 1285 ~ ~ r 1286 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1288 Figure 5: AERO Interface L3 Error Message Format 1290 After the node has prepared the L3 PTB message, it either forwards 1291 the message via a link outside of the AERO interface without 1292 encapsulation, or encapsulates and forwards the message to the next 1293 hop via the AERO interface. 1295 When an AERO Relay receives an L3 packet for which the destination 1296 address is covered by an ASP, if there is no more-specific routing 1297 information for the destination the Relay drops the packet and 1298 returns an L3 Destination Unreachable message. The Relay first 1299 writes the IP source address of the original L3 packet as the 1300 destination address of the L3 Destination Unreachable message and 1301 determines the next hop to the destination. If the next hop is 1302 reached via the AERO interface, the Relay uses the IPv6 address "::" 1303 or the IPv4 address "0.0.0.0" as the IP source address of the L3 1304 Destination Unreachable message and forwards the message to the next 1305 hop within the AERO interface. Otherwise, the Relay uses one of its 1306 non link-local addresses as the source address of the L3 Destination 1307 Unreachable message and forwards the message via a link outside the 1308 AERO interface. 1310 When an AERO node receives any L3 error message via the AERO 1311 interface, it examines the destination address in the L3 IP header of 1312 the message. If the next hop toward the destination address of the 1313 error message is via the AERO interface, the node re-encapsulates and 1314 forwards the message to the next hop within the AERO interface. 1315 Otherwise, if the source address in the L3 IP header of the message 1316 is the IPv6 address "::" or the IPv4 address "0.0.0.0", the node 1317 writes one of its non link-local addresses as the source address of 1318 the L3 message and recalculates the IP and/or ICMP checksums. The 1319 node finally forwards the message via a link outside of the AERO 1320 interface. 1322 (*) Note that in some instances the packet-in-error field of an L2 1323 PTB message may not include enough information for translation to an 1324 L3 PTB message. In that case, the AERO interface simply discards the 1325 L2 PTB message. It can therefore be said that translation of L2 PTB 1326 messages to L3 PTB messages can provide a useful optimization when 1327 possible, but is not critical for sources that correctly use PLPMTUD. 1329 3.15. AERO Router Discovery, Prefix Delegation and Address 1330 Configuration 1332 3.15.1. AERO DHCPv6 Service Model 1334 Each AERO Server configures a DHCPv6 server function to facilitate PD 1335 requests from Clients. Each Server is pre-configured with an 1336 identical list of ACP-to-Client ID mappings for all Clients enrolled 1337 in the AERO system, as well as any information necessary to 1338 authenticate Clients. The configuration information is maintained by 1339 a central administrative authority for the AERO link and securely 1340 propagated to all Servers whenever a new Client is enrolled or an 1341 existing Client is withdrawn. 1343 With these identical configurations, each Server can function 1344 independently of all other Servers, including the maintenance of 1345 active leases. Therefore, no Server-to-Server DHCPv6 state 1346 synchronization is necessary, and Clients can optionally hold 1347 separate leases for the same ACP from multiple Servers. 1349 In this way, Clients can easily associate with multiple Servers, and 1350 can receive new leases from new Servers before deprecating leases 1351 held through old Servers. This enables a graceful "make-before- 1352 break" capability. 1354 3.15.2. AERO Client Behavior 1356 AERO Clients discover the link-layer addresses of AERO Servers via 1357 static configuration, or through an automated means such as DNS name 1358 resolution. In the absence of other information, the Client resolves 1359 the FQDN "linkupnetworks.[domainname]" where "linkupnetworks" is a 1360 constant text string and "[domainname]" is the connection-specific 1361 DNS suffix for the Client's underlying network connection (e.g., 1362 "example.com"). After discovering the link-layer addresses, the 1363 Client associates with one or more of the corresponding Servers. 1365 To associate with a Server, the Client acts as a requesting router to 1366 request an ACP through a DHCPv6 PD exchange[RFC3315][RFC3633] in 1367 which the Client's Solicit/Request messages use the IPv6 1368 "unspecified" address (i.e., "::") as the IPv6 source address, 1369 'All_DHCP_Relay_Agents_and_Servers' as the IPv6 destination address 1370 and the link-layer address of the Server as the link-layer 1371 destination address. The Client also includes a Client Identifier 1372 option with a DHCP Unique Identifier (DUID) plus any necessary 1373 authentication options to identify itself to the DHCPv6 server, and 1374 includes a Client Link Layer Address Option (CLLAO) [RFC6939] with 1375 the format shown in Figure 6: 1377 0 1 2 3 1378 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 1379 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1380 | OPTION_CLIENT_LINKLAYER_ADDR | option-length | 1381 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1382 | link-layer type (16 bits) | Link ID | Preference | 1383 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1385 Figure 6: AERO Client Link-Layer Address Option (CLLAO) Format 1387 The Client sets the CLLAO 'option-length' field to 4 and sets the 1388 'link-layer type' field to TBD1 (see: IANA Considerations), then 1389 includes appropriate Link ID and Preference values for the underlying 1390 interface over which the Solicit/Request will be issued (note that 1391 these are the same values that would be included in an S/TLLAO as 1392 shown in Figure 2). If the Client is pre-provisioned with an ACP 1393 associated with the AERO service, it MAY also include the ACP in a 1394 Solicit/Request message Identity Association for Prefix Delegation 1395 (IA_PD) option to indicate its preferred ACP to the DHCPv6 server. 1396 The Client then sends the encapsulated DHCPv6 request via the 1397 underlying interface. 1399 When the Client receives its ACP via a DHCPv6 Reply from the AERO 1400 Server, it creates a static neighbor cache entry with the Server's 1401 link-local address as the network-layer address and the Server's 1402 encapsulation address as the link-layer address. If the Reply 1403 message contains a Vendor-Specific Information Option (see: 1404 Section 3.15.3) the Client also caches each ASP in the option. The 1405 Client then considers the link-layer address of the Server as the 1406 primary default encapsulation address for forwarding packets for 1407 which there is no more-specific forwarding information. 1409 Next, the Client applies the AERO address to the AERO interface and 1410 sub-delegates the ACP to nodes and links within its attached EUNs 1411 (the AERO address thereafter remains stable as the Client moves). 1412 The Client also assigns a default IP route to the AERO interface as a 1413 route-to-interface, i.e., with no explicit next-hop. The next hop 1414 will then be determined after a packet has been submitted to the AERO 1415 interface by inspecting the neighbor cache (see above). 1417 On some platforms (e.g., popular cell phone operating systems), the 1418 act of assigning a default IPv6 route may not be permitted from a 1419 user application due to security policy. Typically, those platforms 1420 include a TUN/TAP interface that acts as a point-to-point conduit 1421 between user applications and the AERO interface. In that case, the 1422 Client can instead generate a "synthesized RA" message. The message 1423 conforms to [RFC4861] and is prepared as follows: 1425 o the IPv6 source address is fe80:: 1427 o the IPv6 destination address is all-nodes multicast 1429 o the Router Lifetime is set to a time that is no longer than the 1430 ACP DHCPv6 lifetime 1432 o the message does not include a Source Link Layer Address Option 1433 (SLLAO) 1435 o the message includes a Prefix Information Option (PIO) with a /64 1436 prefix taken from the ACP as the prefix for autoconfiguration 1438 The Client then sends the synthesized RA message via the TUN/TAP 1439 interface, where the operating system kernel will interpret it as 1440 though it were generated by an actual router. The operating system 1441 will then install a default route and use StateLess Address 1442 AutoConfiguration (SLAAC) to configure an IPv6 address on the TUN/TAP 1443 interface. Methods for similarly installing an IPv4 default route 1444 and IPv4 address on the TUN/TAP interface are based on synthesized 1445 DHCPv4 messages [RFC2131]. Note that in this method, the Client 1446 appears as a mobility proxy for applications that bind to the (point- 1447 to-point) TUN/TAP interface. The arrangement can be likened to a 1448 Proxy AERO scenario in which the mobile node and Client are located 1449 within the same physical platform (see Section 3.20 for further 1450 details on Proxy AERO). 1452 The Client subsequently renews its ACP delegation through each of its 1453 Servers by performing DHCPv6 Renew/Reply exchanges with its AERO 1454 address as the IPv6 source address, 1455 'All_DHCP_Relay_Agents_and_Servers' as the IPv6 destination address, 1456 the link-layer address of a Server as the link-layer destination 1457 address and the same Client identifier, authentication options and 1458 CLLAO option as was used in the initial PD request. Note that if the 1459 Client does not issue a DHCPv6 Renew before the Server has terminated 1460 the lease (e.g., if the Client has been out of touch with the Server 1461 for a considerable amount of time), the Server's Reply will report 1462 NoBinding and the Client must re-initiate the DHCPv6 PD procedure. 1463 If the Client sends synthesized RA and/or DHCPv4 messages (see 1464 above), it also sends a new synthesized message when issuing a DHCPv6 1465 Renew or when re-initiating the DHCPv6 PD procedure. 1467 Since the Client's AERO address is configured from the unique ACP 1468 delegation it receives, there is no need for Duplicate Address 1469 Detection (DAD) on AERO links. Other nodes maliciously attempting to 1470 hijack an authorized Client's AERO address will be denied access to 1471 the network by the DHCPv6 server due to an unacceptable link-layer 1472 address and/or security parameters (see: Security Considerations). 1474 3.15.3. AERO Server Behavior 1476 AERO Servers configure a DHCPv6 server function on their AERO links. 1477 AERO Servers arrange to add their encapsulation layer IP addresses 1478 (i.e., their link-layer addresses) to the DNS resource records for 1479 the FQDN "linkupnetworks.[domainname]" before entering service. 1481 When an AERO Server receives a prospective Client's DHCPv6 PD 1482 Solicit/Request message, it first authenticates the message. If 1483 authentication succeeds, the Server determines the correct ACP to 1484 delegate to the Client by searching an online directory service 1485 (e.g., LDAP) based on a Client identification value. In environments 1486 where spoofing is not considered a threat, the Server MAY use the 1487 Client's DUID as the identification value. Otherwise, the Server 1488 SHOULD use a signed certificate provided by the Client. 1490 The Server then delegates the ACP and creates a static neighbor cache 1491 entry for the Client's AERO address with lifetime set to no more than 1492 the lease lifetime and the Client's link-layer address as the link- 1493 layer address for the Link ID specified in the CLLAO option. The 1494 Server then creates an IP forwarding table entry so that the AERO 1495 routing system will propagate the ACP to all Relays (see: 1496 Section 3.7). Finally, the Server sends a DHCPv6 Reply message to 1497 the Client while using fe80::ID as the IPv6 source address, the 1498 Client's AERO address as the IPv6 destination address, and the 1499 Client's link-layer address as the destination link-layer address. 1500 The Server also includes a Server Unicast option with server-address 1501 set to fe80::ID so that all future Client/Server transactions will be 1502 link-local-only unicast over the AERO link. 1504 When the Server sends the DHCPv6 Reply message, it also includes a 1505 DHCPv6 Vendor-Specific Information Option with 'enterprise-number' 1506 set to "TBD2" (see: IANA Considerations). The option is formatted as 1507 shown in[RFC3315] and with the AERO enterprise-specific format shown 1508 in Figure 7: 1510 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 1511 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1512 | OPTION_VENDOR_OPTS | option-len | 1513 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1514 | enterprise-number ("TBD2") | 1515 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1516 | Reserved | Prefix Length | 1517 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1518 | | 1519 + ASP (1) + 1520 | | 1521 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1522 | Reserved | Prefix Length | 1523 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1524 | | 1525 + ASP (2) + 1526 | | 1527 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1528 | Reserved | Prefix Length | 1529 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1530 | | 1531 + ASP (3) + 1532 | | 1533 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1534 . (etc.) . 1535 . . 1536 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1538 Figure 7: AERO Vendor-Specific Information Option 1540 Per Figure 7, the option includes one or more ASP. The ASP field 1541 contains the IP prefix as it would appear in the interface identifier 1542 portion of the corresponding AERO address (see: Section 3.3). For 1543 IPv6, valid values for the Prefix Length field are 0 through 64; for 1544 IPv4, valid values are 0 through 32. 1546 After the initial DHCPv6 PD exchange, the AERO Server maintains the 1547 neighbor cache entry for the Client until the lease lifetime expires. 1548 If the Client issues a Renew/Reply exchange, the Server extends the 1549 lifetime. If the Client issues a Release/Reply, or if the Client 1550 does not issue a Renew/Reply before the lifetime expires, the Server 1551 deletes the neighbor cache entry for the Client and withdraws the IP 1552 route from the AERO routing system. 1554 3.16. AERO Forwarding Agent Discovery 1556 AERO Relays, Servers and Clients MAY associate with one or more 1557 companion AERO Forwarding Agents as platforms for offloading high- 1558 speed data plane traffic. AERO nodes distribute forwarding 1559 information to Forwarding Agents via an out-of-band messaging service 1560 (e.g., NETCONF [RFC6241], etc.). 1562 When an AERO node receives a data packet on an AERO interface with a 1563 network layer destination address for which it has distributed 1564 forwarding information to one or more Forwarding Agents, the node 1565 returns an RA message to the source neighbor (subject to rate 1566 limiting) then forwards the data packet as usual. The RA message 1567 includes one or more SLLAOs with the link-layer addresses of 1568 candidate Forwarding Engines. 1570 If the forwarding information pertains only to a specific ACP, the 1571 AERO node sets the network-layer source address of the RA to the AERO 1572 address corresponding to the ACP, and sets the default router 1573 lifetime to 0. If the forwarding information pertains to all 1574 addresses, the AERO node instead sets the network-layer source 1575 address of the RA to its own link-local address and sets the default 1576 router lifetime to a non-zero value. 1578 When the source neighbor receives the RA message, it SHOULD record 1579 the link-layer addresses in the SLLAOs as the encapsulation addresses 1580 to use for sending subsequent data packets with addresses that match 1581 the information in the RA. However, the source MUST continue to use 1582 the primary link-layer address of the AERO node as the encapsulation 1583 address for sending control messages. 1585 3.17. AERO Intradomain Route Optimization 1587 When a source Client forwards packets to a prospective correspondent 1588 Client within the same AERO link domain (i.e., one for which the 1589 packet's destination address is covered by an ASP), the source Client 1590 initiates an intra-domain AERO route optimization procedure. The 1591 procedure is based on an exchange of IPv6 ND messages using a chain 1592 of AERO Servers and Relays as a trust basis. This procedure is in 1593 contrast to the Return Routability procedure required for route 1594 optimization to a correspondent Client located in the Internet as 1595 described in Section 3.22. The following sections specify the AERO 1596 intradomain route optimization procedure. 1598 3.17.1. Reference Operational Scenario 1600 Figure 8 depicts the AERO intradomain route optimization reference 1601 operational scenario, using IPv6 addressing as the example (while not 1602 shown, a corresponding example for IPv4 addressing can be easily 1603 constructed). The figure shows an AERO Relay ('R1'), two AERO 1604 Servers ('S1', 'S2'), two AERO Clients ('C1', 'C2') and two ordinary 1605 IPv6 hosts ('H1', 'H2'): 1607 +--------------+ +--------------+ +--------------+ 1608 | Server S1 | | Relay R1 | | Server S2 | 1609 +--------------+ +--------------+ +--------------+ 1610 fe80::2 fe80::1 fe80::3 1611 L2(S1) L2(R1) L2(S2) 1612 | | | 1613 X-----+-----+------------------+-----------------+----+----X 1614 | AERO Link | 1615 L2(A) L2(B) 1616 fe80::2001:db8:0:0 fe80::2001:db8:1:0 1617 +--------------+ +--------------+ 1618 |AERO Client C1| |AERO Client C2| 1619 +--------------+ +--------------+ 1620 2001:DB8:0::/48 2001:DB8:1::/48 1621 | | 1622 .-. .-. 1623 ,-( _)-. 2001:db8:0::1 2001:db8:1::1 ,-( _)-. 1624 .-(_ IP )-. +---------+ +---------+ .-(_ IP )-. 1625 (__ EUN )--| Host H1 | | Host H2 |--(__ EUN ) 1626 `-(______)-' +---------+ +---------+ `-(______)-' 1628 Figure 8: AERO Reference Operational Scenario 1630 In Figure 8, Relay ('R1') assigns the address fe80::1 to its AERO 1631 interface with link-layer address L2(R1), Server ('S1') assigns the 1632 address fe80::2 with link-layer address L2(S1),and Server ('S2') 1633 assigns the address fe80::3 with link-layer address L2(S2). Servers 1634 ('S1') and ('S2') next arrange to add their link-layer addresses to a 1635 published list of valid Servers for the AERO link. 1637 AERO Client ('C1') receives the ACP 2001:db8:0::/48 in a DHCPv6 PD 1638 exchange via AERO Server ('S1') then assigns the address 1639 fe80::2001:db8:0:0 to its AERO interface with link-layer address 1640 L2(C1). Client ('C1') configures a default route and neighbor cache 1641 entry via the AERO interface with next-hop address fe80::2 and link- 1642 layer address L2(S1), then sub-delegates the ACP to its attached 1643 EUNs. IPv6 host ('H1') connects to the EUN, and configures the 1644 address 2001:db8:0::1. 1646 AERO Client ('C2') receives the ACP 2001:db8:1::/48 in a DHCPv6 PD 1647 exchange via AERO Server ('S2') then assigns the address 1648 fe80::2001:db8:1:0 to its AERO interface with link-layer address 1649 L2(C2). Client ('C2') configures a default route and neighbor cache 1650 entry via the AERO interface with next-hop address fe80::3 and link- 1651 layer address L2(S2), then sub-delegates the ACP to its attached 1652 EUNs. IPv6 host ('H1') connects to the EUN, and configures the 1653 address 2001:db8:1::1. 1655 3.17.2. Concept of Operations 1657 Again, with reference to Figure 8, when source host ('H1') sends a 1658 packet to destination host ('H2'), the packet is first forwarded over 1659 the source host's attached EUN to Client ('C1'). Client ('C1') then 1660 forwards the packet via its AERO interface to Server ('S1') and also 1661 sends a Predirect message toward Client ('C2') via Server ('S1'). 1662 Server ('S1') then re-encapsulates and forwards both the packet and 1663 the Predirect message out the same AERO interface toward Client 1664 ('C2') via Relay ('R1'). 1666 When Relay ('R1') receives the packet and Predirect message, it 1667 consults its forwarding table to discover Server ('S2') as the next 1668 hop toward Client ('C2'). Relay ('R1') then forwards both the packet 1669 and the Predirect message to Server ('S2'), which then forwards them 1670 to Client ('C2'). 1672 After Client ('C2') receives the Predirect message, it process the 1673 message and returns a Redirect message toward Client ('C1') via 1674 Server ('S2'). During the process, Client ('C2') also creates or 1675 updates a dynamic neighbor cache entry for Client ('C1'). 1677 When Server ('S2') receives the Redirect message, it re-encapsulates 1678 the message and forwards it on to Relay ('R1'), which forwards the 1679 message on to Server ('S1') which forwards the message on to Client 1680 ('C1'). After Client ('C1') receives the Redirect message, it 1681 processes the message and creates or updates a dynamic neighbor cache 1682 entry for Client ('C2'). 1684 Following the above Predirect/Redirect message exchange, forwarding 1685 of packets from Client ('C1') to Client ('C2') without involving any 1686 intermediate nodes is enabled. The mechanisms that support this 1687 exchange are specified in the following sections. 1689 3.17.3. Message Format 1691 AERO Redirect/Predirect messages use the same format as for ICMPv6 1692 Redirect messages depicted in Section 4.5 of [RFC4861], but also 1693 include a new "Prefix Length" field taken from the low-order 8 bits 1694 of the Redirect message Reserved field. For IPv6, valid values for 1695 the Prefix Length field are 0 through 64; for IPv4, valid values are 1696 0 through 32. The Redirect/Predirect messages are formatted as shown 1697 in Figure 9: 1699 0 1 2 3 1700 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 1701 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1702 | Type (=137) | Code (=0/1) | Checksum | 1703 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1704 | Reserved | Prefix Length | 1705 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1706 | | 1707 + + 1708 | | 1709 + Target Address + 1710 | | 1711 + + 1712 | | 1713 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1714 | | 1715 + + 1716 | | 1717 + Destination Address + 1718 | | 1719 + + 1720 | | 1721 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1722 | Options ... 1723 +-+-+-+-+-+-+-+-+-+-+-+- 1725 Figure 9: AERO Redirect/Predirect Message Format 1727 3.17.4. Sending Predirects 1729 When a Client forwards a packet with a source address from one of its 1730 ACPs toward a destination address covered by an ASP (i.e., toward 1731 another AERO Client connected to the same AERO link), the source 1732 Client MAY send a Predirect message forward toward the destination 1733 Client via the Server. 1735 In the reference operational scenario, when Client ('C1') forwards a 1736 packet toward Client ('C2'), it MAY also send a Predirect message 1737 forward toward Client ('C2'), subject to rate limiting (see 1738 Section 8.2 of [RFC4861]). Client ('C1') prepares the Predirect 1739 message as follows: 1741 o the link-layer source address is set to 'L2(C1)' (i.e., the link- 1742 layer address of Client ('C1')). 1744 o the link-layer destination address is set to 'L2(S1)' (i.e., the 1745 link-layer address of Server ('S1')). 1747 o the network-layer source address is set to fe80::2001:db8:0:0 1748 (i.e., the AERO address of Client ('C1')). 1750 o the network-layer destination address is set to fe80::2001:db8:1:0 1751 (i.e., the AERO address of Client ('C2')). 1753 o the Type is set to 137. 1755 o the Code is set to 1 to indicate "Predirect". 1757 o the Prefix Length is set to the length of the prefix to be 1758 assigned to the Target Address. 1760 o the Target Address is set to fe80::2001:db8:0:0 (i.e., the AERO 1761 address of Client ('C1')). 1763 o the Destination Address is set to the source address of the 1764 originating packet that triggered the Predirection event. (If the 1765 originating packet is an IPv4 packet, the address is constructed 1766 in IPv4-compatible IPv6 address format). 1768 o the message includes one or more TLLAOs with Link ID and 1769 Preference set to appropriate values for Client ('C1')'s 1770 underlying interfaces, and with UDP Port Number and IP Address set 1771 to 0'. 1773 o the message SHOULD include a Timestamp option and a Nonce option. 1775 o the message includes a Redirected Header Option (RHO) that 1776 contains the originating packet truncated if necessary to ensure 1777 that at least the network-layer header is included but the size of 1778 the message does not exceed 1280 bytes. 1780 Note that the act of sending Predirect messages is cited as "MAY", 1781 since Client ('C1') may have advanced knowledge that the direct path 1782 to Client ('C2') would be unusable or otherwise undesirable. If the 1783 direct path later becomes unusable after the initial route 1784 optimization, Client ('C1') simply allows packets to again flow 1785 through Server ('S1'). 1787 3.17.5. Re-encapsulating and Relaying Predirects 1789 When Server ('S1') receives a Predirect message from Client ('C1'), 1790 it first verifies that the TLLAOs in the Predirect are a proper 1791 subset of the Link IDs in Client ('C1')'s neighbor cache entry. If 1792 the Client's TLLAOs are not acceptable, Server ('S1') discards the 1793 message. Otherwise, Server ('S1') validates the message according to 1794 the ICMPv6 Redirect message validation rules in Section 8.1 of 1795 [RFC4861], except that the Predirect has Code=1. Server ('S1') also 1796 verifies that Client ('C1') is authorized to use the Prefix Length in 1797 the Predirect when applied to the AERO address in the network-layer 1798 source address by searching for the AERO address in the neighbor 1799 cache. If validation fails, Server ('S1') discards the Predirect; 1800 otherwise, it copies the correct UDP Port numbers and IP Addresses 1801 for Client ('C1')'s links into the (previously empty) TLLAOs. 1803 Server ('S1') then examines the network-layer destination address of 1804 the Predirect to determine the next hop toward Client ('C2') by 1805 searching for the AERO address in the neighbor cache. Since Client 1806 ('C2') is not one of its neighbors, Server ('S1') re-encapsulates the 1807 Predirect and relays it via Relay ('R1') by changing the link-layer 1808 source address of the message to 'L2(S1)' and changing the link-layer 1809 destination address to 'L2(R1)'. Server ('S1') finally forwards the 1810 re-encapsulated message to Relay ('R1') without decrementing the 1811 network-layer TTL/Hop Limit field. 1813 When Relay ('R1') receives the Predirect message from Server ('S1') 1814 it determines that Server ('S2') is the next hop toward Client ('C2') 1815 by consulting its forwarding table. Relay ('R1') then re- 1816 encapsulates the Predirect while changing the link-layer source 1817 address to 'L2(R1)' and changing the link-layer destination address 1818 to 'L2(S2)'. Relay ('R1') then relays the Predirect via Server 1819 ('S2'). 1821 When Server ('S2') receives the Predirect message from Relay ('R1') 1822 it determines that Client ('C2') is a neighbor by consulting its 1823 neighbor cache. Server ('S2') then re-encapsulates the Predirect 1824 while changing the link-layer source address to 'L2(S2)' and changing 1825 the link-layer destination address to 'L2(C2)'. Server ('S2') then 1826 forwards the message to Client ('C2'). 1828 3.17.6. Processing Predirects and Sending Redirects 1830 When Client ('C2') receives the Predirect message, it accepts the 1831 Predirect only if the message has a link-layer source address of one 1832 of its Servers (e.g., L2(S2)). Client ('C2') further accepts the 1833 message only if it is willing to serve as a redirection target. 1834 Next, Client ('C2') validates the message according to the ICMPv6 1835 Redirect message validation rules in Section 8.1 of [RFC4861], except 1836 that it accepts the message even though Code=1 and even though the 1837 network-layer source address is not that of it's current first-hop 1838 router. 1840 In the reference operational scenario, when Client ('C2') receives a 1841 valid Predirect message, it either creates or updates a dynamic 1842 neighbor cache entry that stores the Target Address of the message as 1843 the network-layer address of Client ('C1') , stores the link-layer 1844 addresses found in the TLLAOs as the link-layer addresses of Client 1845 ('C1') and stores the Prefix Length as the length to be applied to 1846 the network-layer address for forwarding purposes. Client ('C2') 1847 then sets AcceptTime for the neighbor cache entry to ACCEPT_TIME. 1849 After processing the message, Client ('C2') prepares a Redirect 1850 message response as follows: 1852 o the link-layer source address is set to 'L2(C2)' (i.e., the link- 1853 layer address of Client ('C2')). 1855 o the link-layer destination address is set to 'L2(S2)' (i.e., the 1856 link-layer address of Server ('S2')). 1858 o the network-layer source address is set to fe80::2001:db8:1:0 1859 (i.e., the AERO address of Client ('C2')). 1861 o the network-layer destination address is set to fe80::2001:db8:0:0 1862 (i.e., the AERO address of Client ('C1')). 1864 o the Type is set to 137. 1866 o the Code is set to 0 to indicate "Redirect". 1868 o the Prefix Length is set to the length of the prefix to be applied 1869 to the Target Address. 1871 o the Target Address is set to fe80::2001:db8:1:0 (i.e., the AERO 1872 address of Client ('C2')). 1874 o the Destination Address is set to the destination address of the 1875 originating packet that triggered the Redirection event. (If the 1876 originating packet is an IPv4 packet, the address is constructed 1877 in IPv4-compatible IPv6 address format). 1879 o the message includes one or more TLLAOs with Link ID and 1880 Preference set to appropriate values for Client ('C2')'s 1881 underlying interfaces, and with UDP Port Number and IP Address set 1882 to '0'. 1884 o the message SHOULD include a Timestamp option and MUST echo the 1885 Nonce option received in the Predirect (i.e., if a Nonce option is 1886 included). 1888 o the message includes as much of the RHO copied from the 1889 corresponding AERO Predirect message as possible such that at 1890 least the network-layer header is included but the size of the 1891 message does not exceed 1280 bytes. 1893 After Client ('C2') prepares the Redirect message, it sends the 1894 message to Server ('S2'). 1896 3.17.7. Re-encapsulating and Relaying Redirects 1898 When Server ('S2') receives a Redirect message from Client ('C2'), it 1899 first verifies that the TLLAOs in the Redirect are a proper subset of 1900 the Link IDs in Client ('C2')'s neighbor cache entry. If the 1901 Client's TLLAOs are not acceptable, Server ('S2') discards the 1902 message. Otherwise, Server ('S2') validates the message according to 1903 the ICMPv6 Redirect message validation rules in Section 8.1 of 1904 [RFC4861]. Server ('S2') also verifies that Client ('C2') is 1905 authorized to use the Prefix Length in the Redirect when applied to 1906 the AERO address in the network-layer source address by searching for 1907 the AERO address in the neighbor cache. If validation fails, Server 1908 ('S2') discards the Predirect; otherwise, it copies the correct UDP 1909 Port numbers and IP Addresses for Client ('C2')'s links into the 1910 (previously empty) TLLAOs. 1912 Server ('S2') then examines the network-layer destination address of 1913 the Predirect to determine the next hop toward Client ('C2') by 1914 searching for the AERO address in the neighbor cache. Since Client 1915 ('C2') is not a neighbor, Server ('S2') re-encapsulates the Predirect 1916 and relays it via Relay ('R1') by changing the link-layer source 1917 address of the message to 'L2(S2)' and changing the link-layer 1918 destination address to 'L2(R1)'. Server ('S2') finally forwards the 1919 re-encapsulated message to Relay ('R1') without decrementing the 1920 network-layer TTL/Hop Limit field. 1922 When Relay ('R1') receives the Predirect message from Server ('S2') 1923 it determines that Server ('S1') is the next hop toward Client ('C1') 1924 by consulting its forwarding table. Relay ('R1') then re- 1925 encapsulates the Predirect while changing the link-layer source 1926 address to 'L2(R1)' and changing the link-layer destination address 1927 to 'L2(S1)'. Relay ('R1') then relays the Predirect via Server 1928 ('S1'). 1930 When Server ('S1') receives the Predirect message from Relay ('R1') 1931 it determines that Client ('C1') is a neighbor by consulting its 1932 neighbor cache. Server ('S1') then re-encapsulates the Predirect 1933 while changing the link-layer source address to 'L2(S1)' and changing 1934 the link-layer destination address to 'L2(C1)'. Server ('S1') then 1935 forwards the message to Client ('C1'). 1937 3.17.8. Processing Redirects 1939 When Client ('C1') receives the Redirect message, it accepts the 1940 message only if it has a link-layer source address of one of its 1941 Servers (e.g., ''L2(S1)'). Next, Client ('C1') validates the message 1942 according to the ICMPv6 Redirect message validation rules in 1943 Section 8.1 of [RFC4861], except that it accepts the message even 1944 though the network-layer source address is not that of it's current 1945 first-hop router. Following validation, Client ('C1') then processes 1946 the message as follows. 1948 In the reference operational scenario, when Client ('C1') receives 1949 the Redirect message, it either creates or updates a dynamic neighbor 1950 cache entry that stores the Target Address of the message as the 1951 network-layer address of Client ('C2'), stores the link-layer 1952 addresses found in the TLLAOs as the link-layer addresses of Client 1953 ('C2') and stores the Prefix Length as the length to be applied to 1954 the network-layer address for forwarding purposes. Client ('C1') 1955 then sets ForwardTime for the neighbor cache entry to FORWARD_TIME. 1957 Now, Client ('C1') has a neighbor cache entry with a valid 1958 ForwardTime value, while Client ('C2') has a neighbor cache entry 1959 with a valid AcceptTime value. Thereafter, Client ('C1') may forward 1960 ordinary network-layer data packets directly to Client ('C2') without 1961 involving any intermediate nodes, and Client ('C2') can verify that 1962 the packets came from an acceptable source. (In order for Client 1963 ('C2') to forward packets to Client ('C1'), a corresponding 1964 Predirect/Redirect message exchange is required in the reverse 1965 direction; hence, the mechanism is asymmetric.) 1967 3.17.9. Server-Oriented Redirection 1969 In some environments, the Server nearest the target Client may need 1970 to serve as the redirection target, e.g., if direct Client-to-Client 1971 communications are not possible. In that case, the Server prepares 1972 the Redirect message the same as if it were the destination Client 1973 (see: Section 3.17.6), except that it writes its own link-layer 1974 address in the TLLAO option. The Server must then maintain a dynamic 1975 neighbor cache entry for the redirected source Client. 1977 3.18. Neighbor Unreachability Detection (NUD) 1979 AERO nodes perform Neighbor Unreachability Detection (NUD) by sending 1980 unicast NS messages to elicit solicited NA messages from neighbors 1981 the same as described in [RFC4861]. NUD is performed either 1982 reactively in response to persistent L2 errors (see Section 3.14) or 1983 proactively to refresh existing neighbor cache entries. 1985 When an AERO node sends an NS/NA message, it MUST use its link-local 1986 address as the IPv6 source address and the link-local address of the 1987 neighbor as the IPv6 destination address. When an AERO node receives 1988 an NS message or a solicited NA message, it accepts the message if it 1989 has a neighbor cache entry for the neighbor; otherwise, it ignores 1990 the message. 1992 When a source Client is redirected to a target Client it SHOULD 1993 proactively test the direct path by sending an initial NS message to 1994 elicit a solicited NA response. While testing the path, the source 1995 Client can optionally continue sending packets via the Server, 1996 maintain a small queue of packets until target reachability is 1997 confirmed, or (optimistically) allow packets to flow directly to the 1998 target. The source Client SHOULD thereafter continue to proactively 1999 test the direct path to the target Client (see Section 7.3 of 2000 [RFC4861]) periodically in order to keep dynamic neighbor cache 2001 entries alive. 2003 In particular, while the source Client is actively sending packets to 2004 the target Client it SHOULD also send NS messages separated by 2005 RETRANS_TIMER milliseconds in order to receive solicited NA messages. 2006 If the source Client is unable to elicit a solicited NA response from 2007 the target Client after MAX_RETRY attempts, it SHOULD set ForwardTime 2008 to 0 and resume sending packets via one of its Servers. Otherwise, 2009 the source Client considers the path usable and SHOULD thereafter 2010 process any link-layer errors as a hint that the direct path to the 2011 target Client has either failed or has become intermittent. 2013 When a target Client receives an NS message from a source Client, it 2014 resets AcceptTime to ACCEPT_TIME if a neighbor cache entry exists; 2015 otherwise, it discards the NS message. If ForwardTime is non-zero, 2016 the target Client then sends a solicited NA message to the link-layer 2017 address of the source Client; otherwise, it sends the solicited NA 2018 message to the link-layer address of one of its Servers. 2020 When a source Client receives a solicited NA message from a target 2021 Client, it resets ForwardTime to FORWARD_TIME if a neighbor cache 2022 entry exists; otherwise, it discards the NA message. 2024 When ForwardTime for a dynamic neighbor cache entry expires, the 2025 source Client resumes sending any subsequent packets via a Server and 2026 may (eventually) attempt to re-initiate the AERO redirection process. 2027 When AcceptTime for a dynamic neighbor cache entry expires, the 2028 target Client discards any subsequent packets received directly from 2029 the source Client. When both ForwardTime and AcceptTime for a 2030 dynamic neighbor cache entry expire, the Client deletes the neighbor 2031 cache entry. 2033 3.19. Mobility Management 2035 3.19.1. Announcing Link-Layer Address Changes 2037 When a Client needs to change its link-layer address, e.g., due to a 2038 mobility event, it performs an immediate DHCPv6 Rebind/Reply exchange 2039 via each of its Servers using the new link-layer address as the 2040 source and with a CLLAO that includes the correct Link ID and 2041 Preference values. If authentication succeeds, the Server then 2042 update its neighbor cache and sends a DHCPv6 Reply. Note that if the 2043 Client does not issue a DHCPv6 Rebind before the lease lifetime 2044 expires (e.g., if the Client has been out of touch with the Server 2045 for a considerable amount of time), the Server's Reply will report 2046 NoBinding and the Client must re-initiate the DHCPv6 PD procedure. 2048 Next, the Client sends unsolicited NA messages to each of its 2049 correspondent Client neighbors using the same procedures as specified 2050 in Section 7.2.6 of [RFC4861], except that it sends the messages as 2051 unicast to each neighbor via a Server instead of multicast. In this 2052 process, the Client should send no more than 2053 MAX_NEIGHBOR_ADVERTISEMENT messages separated by no less than 2054 RETRANS_TIMER seconds to each neighbor. 2056 With reference to Figure 8, when Client ('C2') needs to change its 2057 link-layer address it sends unicast unsolicited NA messages to Client 2058 ('C1') via Server ('S2') as follows: 2060 o the link-layer source address is set to 'L2(C2)' (i.e., the link- 2061 layer address of Client ('C2')). 2063 o the link-layer destination address is set to 'L2(S2)' (i.e., the 2064 link-layer address of Server ('S2')). 2066 o the network-layer source address is set to fe80::2001:db8:1:0 2067 (i.e., the AERO address of Client ('C2')). 2069 o the network-layer destination address is set to fe80::2001:db8:0:0 2070 (i.e., the AERO address of Client ('C1')). 2072 o the Type is set to 136. 2074 o the Code is set to 0. 2076 o the Solicited flag is set to 0. 2078 o the Override flag is set to 1. 2080 o the Target Address is set to fe80::2001:db8:1:0 (i.e., the AERO 2081 address of Client ('C2')). 2083 o the message includes one or more TLLAOs with Link ID and 2084 Preference set to appropriate values for Client ('C2')'s 2085 underlying interfaces, and with UDP Port Number and IP Address set 2086 to '0'. 2088 o the message SHOULD include a Timestamp option. 2090 When Server ('S1') receives the NA message, it relays the message in 2091 the same way as described for relaying Redirect messages in 2092 Section 3.17.7. In particular, Server ('S1') copies the correct UDP 2093 port numbers and IP addresses into the TLLAOs, changes the link-layer 2094 source address to its own address, changes the link-layer destination 2095 address to the address of Relay ('R1'), then forwards the NA message 2096 via the relaying chain the same as for a Redirect. 2098 When Client ('C1') receives the NA message, it accepts the message 2099 only if it already has a neighbor cache entry for Client ('C2') then 2100 updates the link-layer addresses for Client ('C2') based on the 2101 addresses in the TLLAOs. Next, Client ('C1') SHOULD initiate the NUD 2102 procedures specified in Section 3.18 to provide Client ('C2') with an 2103 indication that the link-layer source address has been updated, and 2104 to refresh ('C2')'s AcceptTime and ('C1')'s ForwardTime timers. 2106 If Client ('C2') receives an NS message from Client ('C1') indicating 2107 that an unsolicited NA has updated its neighbor cache, Client ('C2') 2108 need not send additional unsolicited NAs. If Client ('C2')'s 2109 unsolicited NA messages are somehow lost, however, Client ('C1') will 2110 soon learn of the mobility event via NUD. 2112 3.19.2. Bringing New Links Into Service 2114 When a Client needs to bring a new underlying interface into service 2115 (e.g., when it activates a new data link), it performs an immediate 2116 Rebind/Reply exchange via each of its Servers using the new link- 2117 layer address as the source address and with a CLLAO that includes 2118 the new Link ID and Preference values. If authentication succeeds, 2119 the Server then updates its neighbor cache and sends a DHCPv6 Reply. 2121 The Client MAY then send unsolicited NA messages to each of its 2122 correspondent Clients to inform them of the new link-layer address as 2123 described in Section 3.19.1. 2125 3.19.3. Removing Existing Links from Service 2127 When a Client needs to remove an existing underlying interface from 2128 service (e.g., when it de-activates an existing data link), it 2129 performs an immediate Rebind/Reply exchange via each of its Servers 2130 over any available link with a CLLAO that includes the deprecated 2131 Link ID and a Preference value of 0. If authentication succeeds, the 2132 Server then updates its neighbor cache and sends a DHCPv6 Reply. The 2133 Client SHOULD then send unsolicited NA messages to each of its 2134 correspondent Clients to inform them of the deprecated link-layer 2135 address as described in Section 3.19.1. 2137 3.19.4. Moving to a New Server 2139 When a Client associates with a new Server, it performs the Client 2140 procedures specified in Section 3.15.2. 2142 When a Client disassociates with an existing Server, it sends a 2143 DHCPv6 Release message via a new Server to the unicast link-local 2144 network layer address of the old Server. The new Server then writes 2145 its own link-layer address in the DHCPv6 release message IP source 2146 address and forwards the message to the old Server. 2148 When the old Server receives the DHCPv6 Release, it first 2149 authenticates the message. The Server then resets the Client's 2150 neighbor cache entry lifetime to 5 seconds, rewrites the link-layer 2151 address in the neighbor cache entry to the address of the new Server, 2152 then returns a DHCPv6 Reply message to the Client via the old Server. 2153 When the lifetime expires, the old Server withdraws the IP route from 2154 the AERO routing system and deletes the neighbor cache entry for the 2155 Client. The Client can then use the Reply message to verify that the 2156 termination signal has been processed, and can delete both the 2157 default route and the neighbor cache entry for the old Server. (Note 2158 that since Release/Reply messages may be lost in the network the 2159 Client MUST retry until it gets Reply indicating that the Release was 2160 successful.) 2162 Clients SHOULD NOT move rapidly between Servers in order to avoid 2163 causing excessive oscillations in the AERO routing system. Such 2164 oscillations could result in intermittent reachability for the Client 2165 itself, while causing little harm to the network. Examples of when a 2166 Client might wish to change to a different Server include a Server 2167 that has gone unreachable, topological movements of significant 2168 distance, etc. 2170 3.20. Proxy AERO 2172 Proxy Mobile IPv6 (PMIPv6) [RFC5213][RFC5844][RFC5949] presents a 2173 localized mobility management scheme for use within an access network 2174 domain. It is typically used in WiFi and cellular wireless access 2175 networks, and allows Mobile Nodes (MNs) to receive and retain an IP 2176 address that remains stable within the access network domain without 2177 needing to implement any special mobility protocols. In the PMIPv6 2178 architecture, access network devices known as Mobility Access 2179 Gateways (MAGs) provide MNs with an access link abstraction and 2180 receive prefixes for the MNs from a Local Mobility Anchor (LMA). 2182 In a proxy AERO domain, a proxy AERO Client (acting as a MAG) can 2183 similarly provide proxy services for MNs that do not participate in 2184 AERO messaging. The proxy Client presents an access link abstraction 2185 to MNs, and performs DHCPv6 PD exchanges over the AERO interface with 2186 an AERO Server (acting as an LMA) to receive ACPs for address 2187 provisioning of new MNs that come onto an access link. This scheme 2188 assumes that proxy Clients act as fixed (non-mobile) infrastructure 2189 elements under the same administrative trust basis as for Relays and 2190 Servers. 2192 When an MN comes onto an access link within a proxy AERO domain for 2193 the first time, the proxy Client authenticates the MN and obtains a 2194 unique identifier that it can use as a DHCPv6 DUID then issues a 2195 DHCPv6 PD Request to its Server. When the Server delegates an ACP, 2196 the proxy Client creates an AERO address for the MN and assigns the 2197 ACP to the MN's access link. The proxy Client then configures itself 2198 as a default router for the MN and provides address autoconfiguration 2199 services (e.g., SLAAC, DHCPv6, DHCPv4, etc.) for provisioning MN 2200 addresses from the ACP over the access link. Since the proxy Client 2201 may serve many such MNs simultaneously, it may receive multiple ACP 2202 prefix delegations and configure multiple AERO addresses, i.e., one 2203 for each MN. 2205 When two MNs are associated with the same proxy Client, the Client 2206 can forward traffic between the MNs without involving a Server since 2207 it configures the AERO addresses of both MNs and therefore also has 2208 the necessary routing information. When two MNs are associated with 2209 different proxy Clients, the source MN's Client can initiate standard 2210 AERO route optimization to discover a direct path to the target MN's 2211 Client through the exchange of Predirect/Redirect messages. 2213 When an MN in a proxy AERO domain leaves an access link provided by 2214 an old proxy Client, the MN issues an access link-specific "leave" 2215 message that informs the old Client of the link-layer address of a 2216 new Client on the planned new access link. This is known as a 2217 "predictive handover". When an MN comes onto an access link provided 2218 by a new proxy Client, the MN issues an access link-specific "join" 2219 message that informs the new Client of the link-layer address of the 2220 old Client on the actual old access link. This is known as a 2221 "reactive handover". 2223 Upon receiving a predictive handover indication, the old proxy Client 2224 sends a DHCPv6 PD Request message directly to the new Client and 2225 queues any arriving data packets addressed to the departed MN. The 2226 Request message includes the MN's ID as the DUID, the ACP in an IA_PD 2227 option, the AERO address derived from the MN's ACP as the network- 2228 layer source address, 'All_DHCP_Relay_Agents_and_Servers' as the 2229 network-layer destination address, the old Client's address as the 2230 link-layer source address and the new Client's address as the link- 2231 layer destination address. When the new Client receives the Request 2232 message, it changes the link-layer source address to its own address, 2233 changes the link-layer destination address to the address of its 2234 Server, and forwards the message to the Server. At the same time, 2235 the new Client creates access link state for the ACP in anticipation 2236 of the MN's arrival (while queuing any data packets until the MN 2237 arrives), creates a neighbor cache entry for the old Client with 2238 AcceptTime set to ACCEPT_TIME, then sends a Redirect message back to 2239 the old Client. When the old Client receives the Redirect message, 2240 it creates a neighbor cache entry for new Client with ForwardTime set 2241 to FORWARD_TIME, then forwards any queued data packets to the new 2242 Client. At the same time, the old Client sends a DHCPv6 PD Release 2243 message to its Server. Finally, the old Client sends unsolicited NA 2244 messages to any of the ACP's correspondents with a TLLAO containing 2245 the link-layer address of the new Client. This follows the procedure 2246 specified in Section 3.19.1, except that it is the old Client and not 2247 the Server that supplies the link-layer address. 2249 Upon receiving a reactive handover indication, the new proxy Client 2250 creates access link state for the MN's ACP, sends a DHCPv6 PD Request 2251 message to its Server, and sends a DHCPv6 PD Release message directly 2252 to the old Client. The Release message includes the MN's ID as the 2253 DUID, the ACP in an IA_PD option, the AERO address derived from the 2254 MN's ACP as the network-layer source address, 2255 'All_DHCP_Relay_Agents_and_Servers' as the network-layer destination 2256 address, the new Client's address as the link-layer source address 2257 and the old Client's address as the link-layer destination address. 2258 When the old Client receives the Release message, it changes the 2259 link-layer source address to its own address, changes the link-layer 2260 destination address to the address of its Server, and forwards the 2261 message to the Server. At the same time, the old Client sends a 2262 Predirect message back to the new Client and queues any arriving data 2263 packets addressed to the departed MN. When the new Client receives 2264 the Predirect, it creates a neighbor cache entry for the old Client 2265 with AcceptTime set to ACCEPT_TIME, then sends a Redirect message 2266 back to the old Client. When the old Client receives the Redirect 2267 message, it creates a neighbor cache entry for the new Client with 2268 ForwardTime set to FORWARD_TIME, then forwards any queued data 2269 packets to the new Client. Finally, the old Client sends unsolicited 2270 NA messages to correspondents the same as for the predictive case. 2272 When a Server processes a DHCPv6 Request message, it creates a 2273 neighbor cache entry for this ACP if none currently exists. If a 2274 neighbor cache entry already exists, however, the Server changes the 2275 link-layer address to the address of the new proxy Client (this 2276 satisfies the case of both the old Client and new Client using the 2277 same Server). 2279 When a Server processes a DHCPv6 Release message, it resets the 2280 neighbor cache entry lifetime for this ACP to 5 seconds if the cached 2281 link-layer address matches the old proxy Client's address. 2282 Otherwise, the Server ignores the Release message (this satisfies the 2283 case of both the old Client and new Client using the same Server). 2285 When a correspondent Client receives an unsolicited NA message, it 2286 changes the link-layer address for the ACP's neighbor cache entry to 2287 the address of the new proxy Client. The correspondent Client then 2288 issues a Predirect/Redirect exchange to establish a new neighbor 2289 cache entry in the new Client. 2291 From an architectural perspective, in addition to the use of DHCPv6 2292 PD and IPv6 ND signaling the AERO approach differs from PMIPv6 in its 2293 use of the NBMA virtual link model instead of point-to-point tunnels. 2294 This provides a more agile interface for Client/Server and Client/ 2295 Client coordinations, and also facilitates simple route optimization. 2296 The AERO routing system is also arranged in such a fashion that 2297 Clients get the same service from any Server they happen to associate 2298 with. This provides a natural fault tolerance and load balancing 2299 capability such as desired for distributed mobility management. 2301 3.21. Extending AERO Links Through Security Gateways 2303 When an enterprise mobile device moves from a campus LAN connection 2304 to a public Internet link, it must re-enter the enterprise via a 2305 security gateway that has both a physical interface connection to the 2306 Internet and a physical interface connection to the enterprise 2307 internetwork. This most often entails the establishment of a Virtual 2308 Private Network (VPN) link over the public Internet from the mobile 2309 device to the security gateway. During this process, the mobile 2310 device supplies the security gateway with its public Internet address 2311 as the link-layer address for the VPN. The mobile device then acts 2312 as an AERO Client to negotiate with the security gateway to obtain 2313 its ACP. 2315 In order to satisfy this need, the security gateway also operates as 2316 an AERO Server with support for AERO Client proxying. In particular, 2317 when a mobile device (i.e., the Client) connects via the security 2318 gateway (i.e., the Server), the Server provides the Client with an 2319 ACP in a DHCPv6 PD exchange the same as if it were attached to an 2320 enterprise campus access link. The Server then replaces the Client's 2321 link-layer source address with the Server's enterprise-facing link- 2322 layer address in all AERO messages the Client sends toward neighbors 2323 on the AERO link. The AERO messages are then delivered to other 2324 devices on the AERO link as if they were originated by the security 2325 gateway instead of by the AERO Client. In the reverse direction, the 2326 AERO messages sourced by devices within the enterprise network can be 2327 forwarded to the security gateway, which then replaces the link-layer 2328 destination address with the Client's link-layer address and replaces 2329 the link-layer source address with its own (Internet-facing) link- 2330 layer address. 2332 After receiving the ACP, the Client can send IP packets that use an 2333 address taken from the ACP as the network layer source address, the 2334 Client's link-layer address as the link-layer source address, and the 2335 Server's Internet-facing link-layer address as the link-layer 2336 destination address. The Server will then rewrite the link-layer 2337 source address with the Server's own enterprise-facing link-layer 2338 address and rewrite the link-layer destination address with the 2339 target AERO node's link-layer address, and the packets will enter the 2340 enterprise network as though they were sourced from a device located 2341 within the enterprise. In the reverse direction, when a packet 2342 sourced by a node within the enterprise network uses a destination 2343 address from the Client's ACP, the packet will be delivered to the 2344 security gateway which then rewrites the link-layer destination 2345 address to the Client's link-layer address and rewrites the link- 2346 layer source address to the Server's Internet-facing link-layer 2347 address. The Server then delivers the packet across the VPN to the 2348 AERO Client. In this way, the AERO virtual link is essentially 2349 extended *through* the security gateway to the point at which the VPN 2350 link and AERO link are effectively grafted together by the link-layer 2351 address rewriting performed by the security gateway. All AERO 2352 messaging services (including route optimization and mobility 2353 signaling) are therefore extended to the Client. 2355 In order to support this virtual link grafting, the security gateway 2356 (acting as an AERO Server) must keep static neighbor cache entries 2357 for all of its associated Clients located on the public Internet. 2358 The neighbor cache entry is keyed by the AERO Client's AERO address 2359 the same as if the Client were located within the enterprise 2360 internetwork. The neighbor cache is then managed in all ways as 2361 though the Client were an ordinary AERO Client. This includes the 2362 AERO IPv6 ND messaging signaling for Route Optimization and Neighbor 2363 Unreachability Detection. 2365 Note that the main difference between a security gateway acting as an 2366 AERO Server and an enterprise-internal AERO Server is that the 2367 security gateway has at least one enterprise-internal physical 2368 interface and at least one public Internet physical interface. 2369 Conversely, the enterprise-internal AERO Server has only enterprise- 2370 internal physical interfaces. For this reason security gateway 2371 proxying is needed to ensure that the public Internet link-layer 2372 addressing space is kept separate from the enterprise-internal link- 2373 layer addressing space. This is afforded through a natural extension 2374 of the security association caching already performed for each VPN 2375 client by the security gateway. 2377 3.22. Extending IPv6 AERO Links to the Internet 2379 When an IPv6 host ('H1') with an address from an ACP owned by AERO 2380 Client ('C1') sends packets to a correspondent IPv6 host ('H2'), the 2381 packets eventually arrive at the IPv6 router that owns ('H2')s 2382 prefix. This IPv6 router may or may not be an AERO Client ('C2') 2383 either within the same home network as ('C1') or in a different home 2384 network. 2386 If Client ('C1') is currently located outside the boundaries of its 2387 home network, it will connect back into the home network via a 2388 security gateway acting as an AERO Server. The packets sent by 2389 ('H1') via ('C1') will then be forwarded through the security gateway 2390 then through the home network and finally to ('C2') where they will 2391 be delivered to ('H2'). This could lead to sub-optimal performance 2392 when ('C2') could instead be reached via a more direct route without 2393 involving the security gateway. 2395 Consider the case when host ('H1') has the IPv6 address 2396 2001:db8:1::1, and Client ('C1') has the ACP 2001:db8:1::/64 with 2397 underlying IPv6 Internet address of 2001:db8:1000::1. Also, host 2398 ('H2') has the IPv6 address 2001:db8:2::1, and Client ('C2') has the 2399 ACP 2001:db8:2::/64 with underlying IPv6 address of 2001:db8:2000::1. 2400 Client ('C1') can determine whether 'C2' is indeed also an AERO 2401 Client willing to serve as a route optimization correspondent by 2402 resolving the AAAA records for the DNS FQDN that matches ('H2')s 2403 prefix, i.e.: 2405 '0.0.0.0.2.0.0.0.8.b.d.0.1.0.0.2.aero.linkupnetworks.net' 2407 If ('C2') is indeed a candidate correspondent, the FQDN lookup will 2408 return a PTR resource record that contains the domain name for the 2409 AERO link that manages ('C2')s ASP. Client ('C1') can then attempt 2410 route optimization using an approach similar to the Return 2411 Routability procedure specified for Mobile IPv6 (MIPv6) [RFC6275]. 2412 In order to support this process, both Clients MUST intercept and 2413 decapsulate packets that have a subnet router anycast address 2414 corresponding to any of the /64 prefixes covered by their respective 2415 ACPs. 2417 To initiate the process, Client ('C1') creates a specially-crafted 2418 encapsulated AERO Predirect message that will be routed through its 2419 home network then through ('C2')s home network and finally to ('C2') 2420 itself. Client ('C1') prepares the initial message in the exchange 2421 as follows: 2423 o The encapsulating IPv6 header source address is set to 2424 2001:db8:1:: (i.e., the IPv6 subnet router anycast address for 2425 ('C1')s ACP) 2427 o The encapsulating IPv6 header destination address is set to 2428 2001:db8:2:: (i.e., the IPv6 subnet router anycast address for 2429 ('C2')s ACP) 2431 o The encapsulating IPv6 header is followed by a UDP header with 2432 source and destination port set to 8060 2434 o The encapsulated IPv6 header source address is set to 2435 fe80::2001:db8:1:0 (i.e., the AERO address for ('C1')) 2437 o The encapsulated IPv6 header destination address is set to 2438 fe80::2001:db8:2:0 (i.e., the AERO address for ('C2')) 2440 o The encapsulated AERO Predirect message includes all of the 2441 securing information that would occur in a MIPv6 "Home Test Init" 2442 message (format TBD) 2444 Client ('C1') then further encapsulates the message in the 2445 encapsulating headers necessary to convey the packet to the security 2446 gateway (e.g., through IPsec encapsulation) so that the message now 2447 appears "double-encapsulated". ('C1') then sends the message to the 2448 security gateway, which re-encapsulates and forwards it over the home 2449 network from where it will eventually reach ('C2'). 2451 At the same time, ('C1') creates and sends a second encapsulated AERO 2452 Predirect message that will be routed through the IPv6 Internet 2453 without involving the security gateway. Client ('C1') prepares the 2454 message as follows: 2456 o The encapsulating IPv6 header source address is set to 2457 2001:db8:1000:1 (i.e., the Internet IPv6 address of ('C1')) 2459 o The encapsulating IPv6 header destination address is set to 2460 2001:db8:2:: (i.e., the IPv6 subnet router anycast address for 2461 ('C2')s ACP) 2463 o The encapsulating IPv6 header is followed by a UDP header with 2464 source and destination port set to 8060 2466 o The encapsulated IPv6 header source address is set to 2467 fe80::2001:db8:1:0 (i.e., the AERO address for ('C1')) 2469 o The encapsulated IPv6 header destination address is set to 2470 fe80::2001:db8:2:0 (i.e., the AERO address for ('C2')) 2472 o The encapsulated AERO Predirect message includes all of the 2473 securing information that would occur in a MIPv6 "Care-of Test 2474 Init" message (format TBD) 2476 ('C2') will receive both Predirect messages through its home network 2477 then return a corresponding Redirect for each of the Predirect 2478 messages with the source and destination addresses in the inner and 2479 outer headers reversed. The first message includes all of the 2480 securing information that would occur in a MIPv6 "Home Test" message, 2481 while the second message includes all of the securing information 2482 that would occur in a MIPv6 "Care-of Test" message (formats TBD). 2484 When ('C1') receives the Redirect messages, it performs the necessary 2485 security procedures per the MIPv6 specification. It then prepares an 2486 encapsulated NS message that includes the same source and destination 2487 addresses as for the "Care-of Test Init" Predirect message, and 2488 includes all of the securing information that would occur in a MIPv6 2489 "Binding Update" message (format TBD) and sends the message to 2490 ('C2'). 2492 When ('C2') receives the NS message, if the securing information is 2493 correct it creates or updates a neighbor cache entry for ('C1') with 2494 fe80::2001:db8:1:0 as the network-layer address, 2001:db8:1000::1 as 2495 the link-layer address and with AcceptTime set to ACCEPT_TIME. 2496 ('C2') then sends an encapsulated NA message back to ('C1') that 2497 includes the same source and destination addresses as for the "Care- 2498 of Test" Redirect message, and includes all of the securing 2499 information that would occur in a MIPv6 "Binding Acknowledgement" 2500 message (format TBD) and sends the message to ('C1'). 2502 When ('C1') receives the NA message, it creates or updates a neighbor 2503 cache entry for ('C2') with fe80::2001:db8:2:0 as the network-layer 2504 address and 2001:db8:2:: as the link-layer address and with 2505 ForwardTime set to FORWARD_TIME, thus completing the route 2506 optimization in the forward direction. 2508 ('C1') subsequently forwards encapsulated packets with outer source 2509 address 2001:db8:1000::1, with outer destination address 2510 2001:db8:2::, with inner source address taken from the 2001:db8:1::, 2511 and with inner destination address taken from 2001:db8:2:: due to the 2512 fact that it has a securely-established neighbor cache entry with 2513 non-zero ForwardTime. ('C2') subsequently accepts any such 2514 encapsulated packets due to the fact that it has a securely- 2515 established neighbor cache entry with non-zero AcceptTime. 2517 In order to keep neighbor cache entries alive, ('C1') periodically 2518 sends additional NS messages to ('C2') and receives any NA responses. 2519 If ('C1') moves to a different point of attachment after the initial 2520 route optimization, it sends a new secured NS message to ('C2') as 2521 above to update ('C2')s neighbor cache. 2523 If ('C2') has packets to send to ('C1'), it performs a corresponding 2524 route optimization in the opposite direction following the same 2525 procedures described above. In the process, the already-established 2526 unidirectional neighbor cache entries within ('C1') and ('C2') are 2527 updated to include the now-bidirectional information. In particular, 2528 the AcceptTime and ForwardTime variables for both neighbor cache 2529 entries are updated to non-zero values, and the link-layer address 2530 for ('C1')s neighbor cache entry for ('C2') is reset to 2531 2001:db8:2000::1. 2533 Note that two AERO Clients can use full security protocol messaging 2534 instead of Return Routability, e.g., if strong authentication and/or 2535 confidentiality are desired. In that case, security protocol key 2536 exchanges such as specified for MOBIKE [RFC4555] would be used to 2537 establish security associations and neighbor cache entries between 2538 the AERO clients. Thereafter, AERO NS/NA messaging can be used to 2539 maintain neighbor cache entries, test reachability, and to announce 2540 mobility events. If reachability testing fails, e.g., if both 2541 Clients move at roughly the same time, the Clients can tear down the 2542 security association and neighbor cache entries and again allow 2543 packets to flow through their home network. 2545 3.23. Encapsulation Protocol Version Considerations 2547 A source Client may connect only to an IPvX underlying network, while 2548 the target Client connects only to an IPvY underlying network. In 2549 that case, the target and source Clients have no means for reaching 2550 each other directly (since they connect to underlying networks of 2551 different IP protocol versions) and so must ignore any redirection 2552 messages and continue to send packets via the Server. 2554 3.24. Multicast Considerations 2556 When the underlying network does not support multicast, AERO nodes 2557 map IPv6 link-scoped multicast addresses (including 2558 'All_DHCP_Relay_Agents_and_Servers') to the link-layer address of a 2559 Server. 2561 When the underlying network supports multicast, AERO nodes use the 2562 multicast address mapping specification found in [RFC2529] for IPv4 2563 underlying networks and use a direct multicast mapping for IPv6 2564 underlying networks. (In the latter case, "direct multicast mapping" 2565 means that if the IPv6 multicast destination address of the 2566 encapsulated packet is "M", then the IPv6 multicast destination 2567 address of the encapsulating header is also "M".) 2569 3.25. Operation on AERO Links Without DHCPv6 Services 2571 When Servers on the AERO link do not provide DHCPv6 services, 2572 operation can still be accommodated through administrative 2573 configuration of ACPs on AERO Clients. In that case, administrative 2574 configurations of AERO interface neighbor cache entries on both the 2575 Server and Client are also necessary. However, this may interfere 2576 with the ability for Clients to dynamically change to new Servers, 2577 and can expose the AERO link to misconfigurations unless the 2578 administrative configurations are carefully coordinated. 2580 3.26. Operation on Server-less AERO Links 2582 In some AERO link scenarios, there may be no Servers on the link and/ 2583 or no need for Clients to use a Server as an intermediary trust 2584 anchor. In that case, each Client acts as a Server unto itself to 2585 establish neighbor cache entries by performing direct Client-to- 2586 Client IPv6 ND message exchanges, and some other form of trust basis 2587 must be applied so that each Client can verify that the prospective 2588 neighbor is authorized to use its claimed ACP. 2590 When there is no Server on the link, Clients must arrange to receive 2591 ACPs and publish them via a secure alternate prefix delegation 2592 authority through some means outside the scope of this document. 2594 3.27. Manually-Configured AERO Tunnels 2596 In addition to the dynamic neighbor discovery procedures for AERO 2597 link neighbors described above, AERO encapsulation can be applied to 2598 manually-configured tunnels. In that case, the tunnel endpoints use 2599 an administratively-assigned link-local address and exchange NS/NA 2600 messages the same as for dynamically-established tunnels. 2602 3.28. Intradomain Routing 2604 After a tunnel neighbor relationship has been established, neighbors 2605 can use a traditional dynamic routing protocol over the tunnel to 2606 exchange routing information without having to inject the routes into 2607 the AERO routing system. 2609 4. Implementation Status 2611 An application-layer implementation is in progress. 2613 5. IANA Considerations 2615 IANA is instructed to assign a new 2-octet Hardware Type number 2616 "TBD1" for AERO in the "arp-parameters" registry per Section 2 of 2617 [RFC5494]. The number is assigned from the 2-octet Unassigned range 2618 with Hardware Type "AERO" and with this document as the reference. 2620 IANA is instructed to assign a 4-octet Enterprise Number "TBD2" for 2621 AERO in the "enterprise-numbers" registry per [RFC3315]. 2623 6. Security Considerations 2625 AERO link security considerations are the same as for standard IPv6 2626 Neighbor Discovery [RFC4861] except that AERO improves on some 2627 aspects. In particular, AERO uses a trust basis between Clients and 2628 Servers, where the Clients only engage in the AERO mechanism when it 2629 is facilitated by a trust anchor. Unless there is some other means 2630 of authenticating the Client's identity (e.g., link-layer security), 2631 AERO nodes SHOULD also use DHCPv6 securing services (e.g., DHCPv6 2632 authentication, Secure DHCPv6 [I-D.ietf-dhc-sedhcpv6], etc.) for 2633 Client authentication and network admission control. 2635 AERO Redirect, Predirect and unsolicited NA messages SHOULD include a 2636 Timestamp option (see Section 5.3 of [RFC3971]) that other AERO nodes 2637 can use to verify the message time of origin. AERO Predirect, NS and 2638 RS messages SHOULD include a Nonce option (see Section 5.3 of 2639 [RFC3971]) that recipients echo back in corresponding responses. 2641 AERO links must be protected against link-layer address spoofing 2642 attacks in which an attacker on the link pretends to be a trusted 2643 neighbor. Links that provide link-layer securing mechanisms (e.g., 2644 IEEE 802.1X WLANs) and links that provide physical security (e.g., 2645 enterprise network wired LANs) provide a first line of defense that 2646 is often sufficient. In other instances, additional securing 2647 mechanisms such as Secure Neighbor Discovery (SeND) [RFC3971], IPsec 2648 [RFC4301] or TLS [RFC5246] may be necessary. 2650 AERO Clients MUST ensure that their connectivity is not used by 2651 unauthorized nodes on their EUNs to gain access to a protected 2652 network, i.e., AERO Clients that act as routers MUST NOT provide 2653 routing services for unauthorized nodes. (This concern is no 2654 different than for ordinary hosts that receive an IP address 2655 delegation but then "share" the address with unauthorized nodes via a 2656 NAT function.) 2658 On some AERO links, establishment and maintenance of a direct path 2659 between neighbors requires secured coordination such as through the 2660 Internet Key Exchange (IKEv2) protocol [RFC5996] to establish a 2661 security association. 2663 7. Acknowledgements 2665 Discussions both on IETF lists and in private exchanges helped shape 2666 some of the concepts in this work. Individuals who contributed 2667 insights include Mikael Abrahamsson, Mark Andrews, Fred Baker, 2668 Stewart Bryant, Brian Carpenter, Wojciech Dec, Ralph Droms, Adrian 2669 Farrel, Sri Gundavelli, Brian Haberman, Joel Halpern, Tom Herbert, 2670 Sascha Hlusiak, Lee Howard, Andre Kostur, Ted Lemon, Satoru 2671 Matsushima, Behcet Saikaya, Joe Touch, Bernie Volz, Ryuji Wakikawa 2672 and Lloyd Wood. Members of the IESG also provided valuable input 2673 during their review process that greatly improved the document. 2674 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 2675 for their shepherding guidance. 2677 This work has further been encouraged and supported by Boeing 2678 colleagues including Keith Bartley, Dave Bernhardt, Cam Brodie, 2679 Balaguruna Chidambaram, Bruce Cornish, Claudiu Danilov, Wen Fang, 2680 Anthony Gregory, Jeff Holland, Ed King, Gen MacLean, Rob Muszkiewicz, 2681 Sean O'Sullivan, Kent Shuey, Brian Skeen, Mike Slane, Brendan 2682 Williams, Julie Wulff, Yueli Yang, and other members of the BR&T and 2683 BIT mobile networking teams. 2685 Earlier works on NBMA tunneling approaches are found in 2686 [RFC2529][RFC5214][RFC5569]. 2688 Many of the constructs presented in this second edition of AERO are 2689 based on the author's earlier works, including: 2691 o The Internet Routing Overlay Network (IRON) 2692 [RFC6179][I-D.templin-ironbis] 2694 o Virtual Enterprise Traversal (VET) 2695 [RFC5558][I-D.templin-intarea-vet] 2697 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 2698 [RFC5320][I-D.templin-intarea-seal] 2700 o AERO, First Edition [RFC6706] 2702 Note that these works cite numerous earlier efforts that are not also 2703 cited here due to space limitations. The authors of those earlier 2704 works are acknowledged for their insights. 2706 8. References 2708 8.1. Normative References 2710 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 2711 August 1980. 2713 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, September 2714 1981. 2716 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 2717 RFC 792, September 1981. 2719 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 2720 October 1996. 2722 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2723 Requirement Levels", BCP 14, RFC 2119, March 1997. 2725 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 2726 (IPv6) Specification", RFC 2460, December 1998. 2728 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 2729 IPv6 Specification", RFC 2473, December 1998. 2731 [RFC3315] Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C., 2732 and M. Carney, "Dynamic Host Configuration Protocol for 2733 IPv6 (DHCPv6)", RFC 3315, July 2003. 2735 [RFC3633] Troan, O. and R. Droms, "IPv6 Prefix Options for Dynamic 2736 Host Configuration Protocol (DHCP) version 6", RFC 3633, 2737 December 2003. 2739 [RFC3971] Arkko, J., Kempf, J., Zill, B., and P. Nikander, "SEcure 2740 Neighbor Discovery (SEND)", RFC 3971, March 2005. 2742 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 2743 for IPv6 Hosts and Routers", RFC 4213, October 2005. 2745 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 2746 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 2747 September 2007. 2749 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 2750 Address Autoconfiguration", RFC 4862, September 2007. 2752 [RFC6434] Jankiewicz, E., Loughney, J., and T. Narten, "IPv6 Node 2753 Requirements", RFC 6434, December 2011. 2755 8.2. Informative References 2757 [I-D.ietf-dhc-sedhcpv6] 2758 Jiang, S., Shen, S., Zhang, D., and T. Jinmei, "Secure 2759 DHCPv6", draft-ietf-dhc-sedhcpv6-05 (work in progress), 2760 December 2014. 2762 [I-D.templin-intarea-seal] 2763 Templin, F., "The Subnetwork Encapsulation and Adaptation 2764 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 2765 progress), January 2014. 2767 [I-D.templin-intarea-vet] 2768 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 2769 templin-intarea-vet-40 (work in progress), May 2013. 2771 [I-D.templin-ironbis] 2772 Templin, F., "The Interior Routing Overlay Network 2773 (IRON)", draft-templin-ironbis-16 (work in progress), 2774 March 2014. 2776 [I-D.vandevelde-idr-remote-next-hop] 2777 Velde, G., Patel, K., Rao, D., Raszuk, R., and R. Bush, 2778 "BGP Remote-Next-Hop", draft-vandevelde-idr-remote-next- 2779 hop-08 (work in progress), October 2014. 2781 [RFC0879] Postel, J., "TCP maximum segment size and related topics", 2782 RFC 879, November 1983. 2784 [RFC1035] Mockapetris, P., "Domain names - implementation and 2785 specification", STD 13, RFC 1035, November 1987. 2787 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 2788 November 1990. 2790 [RFC1812] Baker, F., "Requirements for IP Version 4 Routers", RFC 2791 1812, June 1995. 2793 [RFC1930] Hawkinson, J. and T. Bates, "Guidelines for creation, 2794 selection, and registration of an Autonomous System (AS)", 2795 BCP 6, RFC 1930, March 1996. 2797 [RFC1981] McCann, J., Deering, S., and J. Mogul, "Path MTU Discovery 2798 for IP version 6", RFC 1981, August 1996. 2800 [RFC2131] Droms, R., "Dynamic Host Configuration Protocol", RFC 2801 2131, March 1997. 2803 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 2804 Domains without Explicit Tunnels", RFC 2529, March 1999. 2806 [RFC2675] Borman, D., Deering, S., and R. Hinden, "IPv6 Jumbograms", 2807 RFC 2675, August 1999. 2809 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 2810 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 2811 March 2000. 2813 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", RFC 2814 2923, September 2000. 2816 [RFC2983] Black, D., "Differentiated Services and Tunnels", RFC 2817 2983, October 2000. 2819 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 2820 of Explicit Congestion Notification (ECN) to IP", RFC 2821 3168, September 2001. 2823 [RFC3596] Thomson, S., Huitema, C., Ksinant, V., and M. Souissi, 2824 "DNS Extensions to Support IP Version 6", RFC 3596, 2825 October 2003. 2827 [RFC3819] Karn, P., Bormann, C., Fairhurst, G., Grossman, D., 2828 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 2829 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 2830 RFC 3819, July 2004. 2832 [RFC4271] Rekhter, Y., Li, T., and S. Hares, "A Border Gateway 2833 Protocol 4 (BGP-4)", RFC 4271, January 2006. 2835 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 2836 Architecture", RFC 4291, February 2006. 2838 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 2839 Internet Protocol", RFC 4301, December 2005. 2841 [RFC4443] Conta, A., Deering, S., and M. Gupta, "Internet Control 2842 Message Protocol (ICMPv6) for the Internet Protocol 2843 Version 6 (IPv6) Specification", RFC 4443, March 2006. 2845 [RFC4555] Eronen, P., "IKEv2 Mobility and Multihoming Protocol 2846 (MOBIKE)", RFC 4555, June 2006. 2848 [RFC4592] Lewis, E., "The Role of Wildcards in the Domain Name 2849 System", RFC 4592, July 2006. 2851 [RFC4821] Mathis, M. and J. Heffner, "Packetization Layer Path MTU 2852 Discovery", RFC 4821, March 2007. 2854 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 2855 Errors at High Data Rates", RFC 4963, July 2007. 2857 [RFC4994] Zeng, S., Volz, B., Kinnear, K., and J. Brzozowski, 2858 "DHCPv6 Relay Agent Echo Request Option", RFC 4994, 2859 September 2007. 2861 [RFC5213] Gundavelli, S., Leung, K., Devarapalli, V., Chowdhury, K., 2862 and B. Patil, "Proxy Mobile IPv6", RFC 5213, August 2008. 2864 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 2865 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 2866 March 2008. 2868 [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security 2869 (TLS) Protocol Version 1.2", RFC 5246, August 2008. 2871 [RFC5320] Templin, F., "The Subnetwork Encapsulation and Adaptation 2872 Layer (SEAL)", RFC 5320, February 2010. 2874 [RFC5494] Arkko, J. and C. Pignataro, "IANA Allocation Guidelines 2875 for the Address Resolution Protocol (ARP)", RFC 5494, 2876 April 2009. 2878 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 2879 Route Optimization Requirements for Operational Use in 2880 Aeronautics and Space Exploration Mobile Networks", RFC 2881 5522, October 2009. 2883 [RFC5558] Templin, F., "Virtual Enterprise Traversal (VET)", RFC 2884 5558, February 2010. 2886 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 2887 Infrastructures (6rd)", RFC 5569, January 2010. 2889 [RFC5720] Templin, F., "Routing and Addressing in Networks with 2890 Global Enterprise Recursion (RANGER)", RFC 5720, February 2891 2010. 2893 [RFC5844] Wakikawa, R. and S. Gundavelli, "IPv4 Support for Proxy 2894 Mobile IPv6", RFC 5844, May 2010. 2896 [RFC5949] Yokota, H., Chowdhury, K., Koodli, R., Patil, B., and F. 2897 Xia, "Fast Handovers for Proxy Mobile IPv6", RFC 5949, 2898 September 2010. 2900 [RFC5996] Kaufman, C., Hoffman, P., Nir, Y., and P. Eronen, 2901 "Internet Key Exchange Protocol Version 2 (IKEv2)", RFC 2902 5996, September 2010. 2904 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 2905 NAT64: Network Address and Protocol Translation from IPv6 2906 Clients to IPv4 Servers", RFC 6146, April 2011. 2908 [RFC6179] Templin, F., "The Internet Routing Overlay Network 2909 (IRON)", RFC 6179, March 2011. 2911 [RFC6204] Singh, H., Beebee, W., Donley, C., Stark, B., and O. 2912 Troan, "Basic Requirements for IPv6 Customer Edge 2913 Routers", RFC 6204, April 2011. 2915 [RFC6241] Enns, R., Bjorklund, M., Schoenwaelder, J., and A. 2916 Bierman, "Network Configuration Protocol (NETCONF)", RFC 2917 6241, June 2011. 2919 [RFC6275] Perkins, C., Johnson, D., and J. Arkko, "Mobility Support 2920 in IPv6", RFC 6275, July 2011. 2922 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 2923 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, August 2924 2011. 2926 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 2927 for Equal Cost Multipath Routing and Link Aggregation in 2928 Tunnels", RFC 6438, November 2011. 2930 [RFC6691] Borman, D., "TCP Options and Maximum Segment Size (MSS)", 2931 RFC 6691, July 2012. 2933 [RFC6706] Templin, F., "Asymmetric Extended Route Optimization 2934 (AERO)", RFC 6706, August 2012. 2936 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID Field", 2937 RFC 6864, February 2013. 2939 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 2940 UDP Checksums for Tunneled Packets", RFC 6935, April 2013. 2942 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 2943 for the Use of IPv6 UDP Datagrams with Zero Checksums", 2944 RFC 6936, April 2013. 2946 [RFC6939] Halwasia, G., Bhandari, S., and W. Dec, "Client Link-Layer 2947 Address Option in DHCPv6", RFC 6939, May 2013. 2949 [RFC6980] Gont, F., "Security Implications of IPv6 Fragmentation 2950 with IPv6 Neighbor Discovery", RFC 6980, August 2013. 2952 [RFC7078] Matsumoto, A., Fujisaki, T., and T. Chown, "Distributing 2953 Address Selection Policy Using DHCPv6", RFC 7078, January 2954 2014. 2956 [TUNTAP] Wikipedia, W., "http://en.wikipedia.org/wiki/TUN/TAP", 2957 October 2014. 2959 Author's Address 2961 Fred L. Templin (editor) 2962 Boeing Research & Technology 2963 P.O. Box 3707 2964 Seattle, WA 98124 2965 USA 2967 Email: fltemplin@acm.org