idnits 2.17.1 draft-templin-aerolink-46.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- == The 'Obsoletes: ' line in the draft header should list only the _numbers_ of the RFCs which will be obsoleted by this document (if approved); it should not include the word 'RFC' in the list. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (October 24, 2014) is 3471 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 931, but not defined == Unused Reference: 'RFC0768' is defined on line 2666, but no explicit reference was found in the text == Unused Reference: 'RFC4862' is defined on line 2705, but no explicit reference was found in the text == Unused Reference: 'RFC0879' is defined on line 2737, but no explicit reference was found in the text == Unused Reference: 'RFC4994' is defined on line 2806, but no explicit reference was found in the text == Unused Reference: 'RFC5720' is defined on line 2838, but no explicit reference was found in the text == Unused Reference: 'RFC6146' is defined on line 2853, but no explicit reference was found in the text == Unused Reference: 'RFC6204' is defined on line 2860, but no explicit reference was found in the text == Unused Reference: 'RFC6355' is defined on line 2871, but no explicit reference was found in the text == Unused Reference: 'RFC6691' is defined on line 2879, but no explicit reference was found in the text == Unused Reference: 'RFC6980' is defined on line 2898, but no explicit reference was found in the text == Unused Reference: 'RFC7078' is defined on line 2901, 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-sedhcpv6-04 == 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 (~~), 16 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, October 24, 2014 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: April 27, 2015 10 Transmission of IP Packets over AERO Links 11 draft-templin-aerolink-46.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 April 27, 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, Re-encapsulation and 80 Decapsulation . . . . . . . . . . . . . . . . . . . . . . 17 81 3.11. AERO Interface Data Origin Authentication . . . . . . . . 19 82 3.12. AERO Interface MTU and Fragmentation . . . . . . . . . . 20 83 3.12.1. Accommodating Large Control Messages . . . . . . . . 22 84 3.12.2. Integrity . . . . . . . . . . . . . . . . . . . . . 23 85 3.13. AERO Interface Error Handling . . . . . . . . . . . . . . 24 86 3.14. AERO Router Discovery, Prefix Delegation and Address 87 Configuration . . . . . . . . . . . . . . . . . . . . . . 29 88 3.14.1. AERO DHCPv6 Service Model . . . . . . . . . . . . . 29 89 3.14.2. AERO Client Behavior . . . . . . . . . . . . . . . . 29 90 3.14.3. AERO Server Behavior . . . . . . . . . . . . . . . . 32 91 3.15. AERO Forwarding Agent Discovery . . . . . . . . . . . . . 34 92 3.16. AERO Intradomain Route Optimization . . . . . . . . . . . 34 93 3.16.1. Reference Operational Scenario . . . . . . . . . . . 35 94 3.16.2. Concept of Operations . . . . . . . . . . . . . . . 36 95 3.16.3. Message Format . . . . . . . . . . . . . . . . . . . 36 96 3.16.4. Sending Predirects . . . . . . . . . . . . . . . . . 37 97 3.16.5. Re-encapsulating and Relaying Predirects . . . . . . 39 98 3.16.6. Processing Predirects and Sending Redirects . . . . 39 99 3.16.7. Re-encapsulating and Relaying Redirects . . . . . . 41 100 3.16.8. Processing Redirects . . . . . . . . . . . . . . . . 42 101 3.16.9. Server-Oriented Redirection . . . . . . . . . . . . 42 102 3.17. Neighbor Unreachability Detection (NUD) . . . . . . . . . 43 103 3.18. Mobility Management . . . . . . . . . . . . . . . . . . . 44 104 3.18.1. Announcing Link-Layer Address Changes . . . . . . . 44 105 3.18.2. Bringing New Links Into Service . . . . . . . . . . 45 106 3.18.3. Removing Existing Links from Service . . . . . . . . 46 107 3.18.4. Moving to a New Server . . . . . . . . . . . . . . . 46 108 3.19. Encapsulation Protocol Version Considerations . . . . . . 47 109 3.20. Multicast Considerations . . . . . . . . . . . . . . . . 47 110 3.21. Operation on AERO Links Without DHCPv6 Services . . . . . 47 111 3.22. Operation on Server-less AERO Links . . . . . . . . . . . 47 112 3.23. Proxy AERO . . . . . . . . . . . . . . . . . . . . . . . 48 113 3.24. Extending AERO Links Through Security Gateways . . . . . 50 114 3.25. Extending IPv6 AERO Links to the Internet . . . . . . . . 52 115 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 55 116 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 55 117 6. Security Considerations . . . . . . . . . . . . . . . . . . . 56 118 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 56 119 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 57 120 8.1. Normative References . . . . . . . . . . . . . . . . . . 57 121 8.2. Informative References . . . . . . . . . . . . . . . . . 58 122 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 62 124 1. Introduction 126 This document specifies the operation of IP over tunnel virtual links 127 using Asymmetric Extended Route Optimization (AERO). The AERO link 128 can be used for tunneling to neighboring nodes over either IPv6 or 129 IPv4 networks, i.e., AERO views the IPv6 and IPv4 networks as 130 equivalent links for tunneling. Nodes attached to AERO links can 131 exchange packets via trusted intermediate routers that provide 132 forwarding services to reach off-link destinations and redirection 133 services for route optimization that addresses the requirements 134 outlined in [RFC5522]. 136 AERO provides an IPv6 link-local address format known as the AERO 137 address that supports operation of the IPv6 Neighbor Discovery (ND) 138 [RFC4861] protocol and links IPv6 ND to IP forwarding. Admission 139 control and provisioning are supported by the Dynamic Host 140 Configuration Protocol for IPv6 (DHCPv6) [RFC3315], and node mobility 141 is naturally supported through dynamic neighbor cache updates. 143 Although DHCPv6 and IPv6 ND message signalling is used in the control 144 plane, both IPv4 and IPv6 can be used in the data plane. The 145 remainder of this document presents the AERO specification. 147 2. Terminology 149 The terminology in the normative references applies; the following 150 terms are defined within the scope of this document: 152 AERO link 153 a Non-Broadcast, Multiple Access (NBMA) tunnel virtual overlay 154 configured over a node's attached IPv6 and/or IPv4 networks. All 155 nodes on the AERO link appear as single-hop neighbors from the 156 perspective of the virtual overlay. 158 AERO interface 159 a node's attachment to an AERO link. 161 AERO address 162 an IPv6 link-local address constructed as specified in Section 3.3 163 and assigned to a Client's AERO interface. 165 AERO node 166 a node that is connected to an AERO link and that participates in 167 IPv6 ND and DHCPv6 messaging over the link. 169 AERO Client ("Client") 170 a node that assigns AERO addresses to an AERO interface and 171 receives IP prefixes via a DHCPv6 Prefix Delegation (PD) exchanges 172 with one or more AERO Servers. 174 AERO Server ("Server") 175 a node that configures an AERO interface to provide default 176 forwarding and DHCPv6 services for AERO Clients. The Server 177 assigns the IPv6 link-local subnet router anycast address (fe80::) 178 to the AERO interface and also assigns an administratively 179 assigned IPv6 link-local unicast address used for operation of 180 DHCPv6 and the IPv6 ND protocol. 182 AERO Relay ("Relay") 183 a node that configures an AERO interface to relay IP packets 184 between nodes on the same AERO link and/or forward IP packets 185 between the AERO link and the native Internetwork. The Relay 186 assigns an administratively assigned IPv6 link-local unicast 187 address to the AERO interface the same as for a Server. 189 AERO Forwarding Agent ("Forwarding Agent") 190 a node that performs data plane forwarding services as a companion 191 to other AERO nodes. 193 ingress tunnel endpoint (ITE) 194 an AERO interface endpoint that injects tunneled packets into an 195 AERO link. 197 egress tunnel endpoint (ETE) 198 an AERO interface endpoint that receives tunneled packets from an 199 AERO link. 201 underlying network 202 a connected IPv6 or IPv4 network routing region over which the 203 tunnel virtual overlay is configured. A typical example is an 204 enterprise network. 206 underlying interface 207 an AERO node's interface point of attachment to an underlying 208 network. 210 link-layer address 211 an IP address assigned to an AERO node's underlying interface. 212 When UDP encapsulation is used, the UDP port number is also 213 considered as part of the link-layer address. Link-layer 214 addresses are used as the encapsulation header source and 215 destination addresses. 217 network layer address 218 the source or destination address of the encapsulated IP packet. 220 end user network (EUN) 221 an internal virtual or external edge IP network that an AERO 222 Client connects to the rest of the network via the AERO interface. 224 AERO Service Prefix (ASP) 225 an IP prefix associated with the AERO link and from which AERO 226 Client Prefixes (ACPs) are derived (for example, the IPv6 ACP 227 2001:db8:1:2::/64 is derived from the IPv6 ASP 2001:db8::/32). 229 AERO Client Prefix (ACP) 230 a more-specific IP prefix taken from an ASP and delegated to a 231 Client. 233 Throughout the document, the simple terms "Client", "Server" and 234 "Relay" refer to "AERO Client", "AERO Server" and "AERO Relay", 235 respectively. Capitalization is used to distinguish these terms from 236 DHCPv6 client/server/relay [RFC3315]. 238 The terminology of [RFC4861] (including the names of node variables 239 and protocol constants) applies to this document. Also throughout 240 the document, the term "IP" is used to generically refer to either 241 Internet Protocol version (i.e., IPv4 or IPv6). 243 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 244 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 245 document are to be interpreted as described in [RFC2119]. 247 3. Asymmetric Extended Route Optimization (AERO) 249 The following sections specify the operation of IP over Asymmetric 250 Extended Route Optimization (AERO) links: 252 3.1. AERO Link Reference Model 254 .-(::::::::) 255 .-(:::: IP ::::)-. 256 (:: Internetwork ::) 257 `-(::::::::::::)-' 258 `-(::::::)-' 259 | 260 +--------------+ +--------+-------+ +--------------+ 261 |AERO Server S1| | AERO Relay R1 | |AERO Server S2| 262 | Nbr: C1; R1 | | Nbr: S1; S2 | | Nbr: C2; R1 | 263 | default->R1 | |(H1->S1; H2->S2)| | default->R1 | 264 | H1->C1 | +--------+-------+ | H2->C2 | 265 +-------+------+ | +------+-------+ 266 | | | 267 X---+---+-------------------+------------------+---+---X 268 | AERO Link | 269 +-----+--------+ +--------+-----+ 270 |AERO Client C1| |AERO Client C2| 271 | Nbr: S1 | | Nbr: S2 | 272 | default->S1 | | default->S2 | 273 +--------------+ +--------------+ 274 .-. .-. 275 ,-( _)-. ,-( _)-. 276 .-(_ IP )-. .-(_ IP )-. 277 (__ EUN ) (__ EUN ) 278 `-(______)-' `-(______)-' 279 | | 280 +--------+ +--------+ 281 | Host H1| | Host H2| 282 +--------+ +--------+ 284 Figure 1: AERO Link Reference Model 286 Figure 1 above presents the AERO link reference model. In this 287 model: 289 o Relay R1 acts as a default router for its associated Servers S1 290 and S2, and connects the AERO link to the rest of the IP 291 Internetwork 293 o Servers S1 and S2 associate with Relay R1 and also act as default 294 routers for their associated Clients C1 and C2. 296 o Clients C1 and C2 associate with Servers S1 and S2, respectively 297 and also act as default routers for their associated EUNs 299 o Hosts H1 and H2 attach to the EUNs served by Clients C1 and C2, 300 respectively 302 In common operational practice, there may be many additional Relays, 303 Servers and Clients. (Although not shown in the figure, AERO 304 Forwarding Agents may also be provided for data plane forwarding 305 offload services.) 307 3.2. AERO Link Node Types 309 AERO Relays provide default forwarding services to AERO Servers. 310 Relays forward packets between Servers connected to the same AERO 311 link and also forward packets between the AERO link and the native IP 312 Internetwork. Relays present the AERO link to the native 313 Internetwork as a set of one or more AERO Service Prefixes (ASPs) and 314 serve as a gateway between the AERO link and the Internetwork. AERO 315 Relays maintain an AERO interface neighbor cache entry for each AERO 316 Server, and maintain an IP forwarding table entry for each AERO 317 Client Prefix (ACP). 319 AERO Servers provide default forwarding services to AERO Clients. 320 Each Server also peers with each Relay in a dynamic routing protocol 321 instance to advertise its list of associated ACPs. Servers configure 322 a DHCPv6 server function to facilitate Prefix Delegation (PD) 323 exchanges with Clients. Each delegated prefix becomes an ACP taken 324 from an ASP. Servers forward packets between AERO interface 325 neighbors only, i.e., and not between the AERO link and the native IP 326 Internetwork unless they are also configured as a Relay. AERO 327 Servers maintain an AERO interface neighbor cache entry for each AERO 328 Relay. They also maintain both a neighbor cache entry and an IP 329 forwarding table entry for each of their associated Clients. 331 AERO Clients act as requesting routers to receive ACPs through DHCPv6 332 PD exchanges with AERO Servers over the AERO link and sub-delegate 333 portions of their ACPs to EUN interfaces. (Each Client MAY associate 334 with a single Server or with multiple Servers, e.g., for fault 335 tolerance, load balancing, etc.) Each IPv6 Client receives at least 336 a /64 IPv6 ACP, and may receive even shorter prefixes. Similarly, 337 each IPv4 Client receives at least a /32 IPv4 ACP (i.e., a singleton 338 IPv4 address), and may receive even shorter prefixes. AERO Clients 339 maintain an AERO interface neighbor cache entry for each of their 340 associated Servers as well as for each of their correspondent 341 Clients. 343 AERO Clients that act as hosts typically configure a TUN/TAP 344 interface [TUNTAP] as a point-to-point linkage between the IP layer 345 and the AERO interface. The IP layer therefore sees only the TUN/TAP 346 interface, while the AERO interface provides an intermediate conduit 347 between the TUN/TAP interface and the underlying interfaces. AERO 348 Clients that act as hosts assign one or more IP addresses from their 349 ACPs to the TUN/TAP interface, i.e., and not to the AERO interface. 351 AERO Forwarding Agents provide data plane forwarding services as 352 companions to other AERO nodes. Note that while all Relays, Servers 353 and Clients are required to perform both control and data plane 354 operations on their own behalf, they may optionally enlist the 355 services of special-purpose Forwarding Agents to offload performance- 356 intensive traffic. 358 3.3. AERO Addresses 360 An AERO address is an IPv6 link-local address with an embedded ACP 361 and assigned to a Client's AERO interface. The AERO address is 362 formed as follows: 364 fe80::[ACP] 366 For IPv6, the AERO address begins with the prefix fe80::/64 and 367 includes in its interface identifier the base prefix taken from the 368 Client's IPv6 ACP. The base prefix is determined by masking the ACP 369 with the prefix length. For example, if the AERO Client receives the 370 IPv6 ACP: 372 2001:db8:1000:2000::/56 374 it constructs its AERO address as: 376 fe80::2001:db8:1000:2000 378 For IPv4, the AERO address is formed from the lower 64 bits of an 379 IPv4-mapped IPv6 address [RFC4291] that includes the base prefix 380 taken from the Client's IPv4 ACP. For example, if the AERO Client 381 receives the IPv4 ACP: 383 192.0.2.32/28 385 it constructs its AERO address as: 387 fe80::FFFF:192.0.2.32 389 The AERO address remains stable as the Client moves between 390 topological locations, i.e., even if its link-layer addresses change. 392 NOTE: In some cases, prospective neighbors may not have advanced 393 knowledge of the Client's ACP length and may therefore send initial 394 IPv6 ND messages with an AERO destination address that matches the 395 ACP but does not correspond to the base prefix. In that case, the 396 Client MUST accept the address as equivalent to the base address, but 397 then use the base address as the source address of any IPv6 ND 398 message replies. For example, if the Client receives the IPv6 ACP 399 2001:db8:1000:2000::/56 then subsequently receives an IPv6 ND message 400 with destination address fe80::2001:db8:1000:2001, it accepts the 401 message but uses fe80::2001:db8:1000:2000 as the source address of 402 any IPv6 ND replies. 404 3.4. AERO Interface Characteristics 406 AERO interfaces use encapsulation (see Section 3.10) to exchange 407 packets with neighbors attached to the AERO link. AERO interfaces 408 maintain a neighbor cache, and AERO Clients and Servers use an 409 adaptation of standard unicast IPv6 ND messaging. AERO interfaces 410 use unicast Neighbor Solicitation (NS), Neighbor Advertisement (NA), 411 Router Solicitation (RS) and Router Advertisement (RA) messages the 412 same as for any IPv6 link. AERO interfaces use two redirection 413 message types -- the first known as a Predirect message and the 414 second being the standard Redirect message (see Section 3.16). AERO 415 links further use link-local-only addressing; hence, AERO nodes 416 ignore any Prefix Information Options (PIOs) they may receive in RA 417 messages over an AERO interface. 419 AERO interface ND messages include one or more Source/Target Link- 420 Layer Address Options (S/TLLAOs) formatted as shown in Figure 2: 422 0 1 2 3 423 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 1 424 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 425 | Type = 2 | Length = 3 | Reserved | 426 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 427 | Link ID | Preference | UDP Port Number | 428 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 429 | | 430 +-- --+ 431 | | 432 +-- IP Address --+ 433 | | 434 +-- --+ 435 | | 436 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 438 Figure 2: AERO Source/Target Link-Layer Address Option (S/TLLAO) 439 Format 441 In this format, Link ID is an integer value between 0 and 255 442 corresponding to an underlying interface of the target node, and 443 Preference is an integer value between 0 and 255 indicating the 444 node's preference for this underlying interface (with 255 being the 445 highest preference, 1 being the lowest, and 0 meaning "link 446 disabled"). UDP Port Number and IP Address are set to the addresses 447 used by the target node when it sends encapsulated packets over the 448 underlying interface. When the encapsulation IP address family is 449 IPv4, IP Address is formed as an IPv4-mapped IPv6 address [RFC4291]. 451 AERO interfaces may be configured over multiple underlying 452 interfaces. For example, common mobile handheld devices have both 453 wireless local area network ("WLAN") and cellular wireless links. 454 These links are typically used "one at a time" with low-cost WLAN 455 preferred and highly-available cellular wireless as a standby. In a 456 more complex example, aircraft frequently have many wireless data 457 link types (e.g. satellite-based, terrestrial, air-to-air 458 directional, etc.) with diverse performance and cost properties. 460 If a Client's multiple underlying interfaces are used "one at a time" 461 (i.e., all other interfaces are in standby mode while one interface 462 is active), then Redirect, Predirect and unsolicited NA messages 463 include only a single TLLAO with Link ID set to a constant value. 465 If the Client has multiple active underlying interfaces, then from 466 the perspective of IPv6 ND it would appear to have a single link- 467 local address with multiple link-layer addresses. In that case, 468 Redirect, Predirect and unsolicited NA messages MAY include multiple 469 TLLAOs -- each with a different Link ID that corresponds to a 470 specific underlying interface of the Client. 472 3.5. AERO Link Registration 474 When an administrative authority first deploys a set of AERO Relays 475 and Servers that comprise an AERO link, they also assign a unique 476 domain name for the link, e.g., "example.com". Next, if 477 administrative policy permits Clients within the domain to serve as 478 correspondent nodes for Internet mobile nodes, the administrative 479 authority adds a Fully Qualified Domain Name (FQDN) for each of the 480 AERO link's ASPs to the Domain Name System (DNS) [RFC1035]. The FQDN 481 is based on the suffix "aero.linkupnetworks.net" with a wildcard- 482 terminated reverse mapping of the ASP [RFC3596][RFC4592], and 483 resolves to a DNS PTR resource record. For example, for the ASP 484 '2001:db8:1::/48' within the domain name "example.com", the DNS 485 database contains: 487 '*.1.0.0.0.8.b.d.0.1.0.0.2.aero.linkupnetworks.net. PTR example.com' 489 This DNS registration advertises the AERO link's ASPs to prospective 490 mobile nodes. 492 3.6. AERO Interface Initialization 494 3.6.1. AERO Relay Behavior 496 When a Relay enables an AERO interface, it first assigns an 497 administratively-assigned link-local address fe80::ID to the 498 interface. Each fe80::ID address MUST be unique among all Relays and 499 Servers on the link, and MUST NOT collide with any potential AERO 500 addresses. The addresses are typically taken from the range 501 fe80::/96, e.g., as fe80::1, fe80::2, fe80::3, etc. The Relay then 502 engages in a dynamic routing protocol session with all Servers on the 503 link (see: Section 3.7), and advertises the set of ASPs into the 504 native IP Internetwork. 506 Each Relay subsequently maintains an IP forwarding table entry for 507 each Client-Server association, and maintains a neighbor cache entry 508 for each Server on the link. Relays do not require the use of 509 Neighbor Unreachability Detection (NUD) messaging (see: Section 3.17) 510 since the dynamic routing protocol already provides Relay-Server 511 reachability information. At a minimum, however, Relays perform NS/ 512 NA, e.g., to support AERO interface MTU determination (see: 513 Section 3.12). 515 3.6.2. AERO Server Behavior 517 When a Server enables an AERO interface, it assigns the address 518 fe80:: to the interface as a link-local Subnet Router Anycast 519 address, and also assigns an administratively assigned link-local 520 address fe80::ID the same as for Relays. (The Server then accepts 521 DHCPv6 and IPv6 ND control messages destined to either the fe80:: or 522 fe80::ID addresses, but always uses fe80::ID as the source address in 523 the control messages it generates.) The Server further configures a 524 DHCPv6 server function to facilitate DHCPv6 PD exchanges with AERO 525 Clients. The Server maintains a neighbor cache entry for each Relay 526 on the link, and manages per-Client neighbor cache entries and IP 527 forwarding table entries based on control message exchanges. Each 528 Server also engages in a dynamic routing protocol with each Relay on 529 the link (see: Section 3.7). 531 When the Server receives an NS/RS message on the AERO interface it 532 returns an NA/RA message but does not update the neighbor cache. The 533 Server further provides a simple conduit between AERO interface 534 neighbors. Therefore, packets enter the Server's AERO interface from 535 the link layer and are forwarded back out the link layer without ever 536 leaving the AERO interface and therefore without ever disturbing the 537 network layer. 539 3.6.3. AERO Client Behavior 541 When a Client enables an AERO interface, it invokes DHCPv6 PD to 542 receive an ACP from an AERO Server. Next, it assigns the 543 corresponding AERO address to the AERO interface and creates a 544 neighbor cache entry for the Server, i.e., the PD exchange bootstraps 545 the provisioning of a unique link-local address. The Client 546 maintains a neighbor cache entry for each of its Servers and each of 547 its active correspondent Clients. When the Client receives Redirect/ 548 Predirect messages on the AERO interface it updates or creates 549 neighbor cache entries, including link-layer address information. 550 Unsolicited NA messages update the cached link-layer addresses for 551 correspondent Clients (e.g., following a link-layer address change 552 due to node mobility) but do not create new neighbor cache entries. 553 NS/NA messages used for NUD update timers in existing neighbor cache 554 entires but do not update link-layer addresses nor create new 555 neighbor cache entries. 557 Finally, the Client need not maintain any IP forwarding table entries 558 for its Servers or correspondent Clients. Instead, it can set a 559 single "route-to-interface" default route in the IP forwarding table, 560 and all forwarding decisions can be made within the AERO interface 561 based on neighbor cache entries. (On systems in which adding a 562 default route would violate security policy, the default route could 563 instead be installed via a "synthesized RA", e.g., as discussed in 564 Section 3.14.2.) 566 3.6.4. AERO Forwarding Agent Behavior 568 When a Forwarding Agent enables an AERO interface, it assigns the 569 same link-local address(es) as the companion AERO node that manages 570 AERO control messaging services. The Forwarding Agent thereafter 571 provides data plane forwarding services based solely on the 572 forwarding information assigned to it by the companion AERO node. 573 AERO Forwarding Agents perform NS/NA messaging, i.e., the same as for 574 AERO Relays. 576 3.7. AERO Link Routing System 578 Relays require full topology knowledge of all ACP/Server 579 associations, while individual Servers at a minimum only need to know 580 the ACPs for their current set of associated Clients. This is 581 accomplished through the use of an internal instance of the Border 582 Gateway Protocol (BGP) [RFC4271] coordinated between Servers and 583 Relays. This internal BGP instance does not interact with the public 584 Internet BGP instance; therefore, the AERO link is presented to the 585 IP Internetwork as a small set of ASPs as opposed to the full set of 586 individual ACPs. 588 In a reference BGP arrangement, each AERO Server is configured as an 589 Autonomous System Border Router (ASBR) for a stub Autonomous System 590 (AS) (possibly using a private AS Number (ASN) [RFC1930]), and each 591 Server further peers with each Relay but does not peer with other 592 Servers. Similarly, Relays do not peer with each other, since they 593 will reliably receive all updates from all Servers and will therefore 594 have a consistent view of the AERO link ACP delegations. 596 Each Server maintains a working set of associated ACPs, and 597 dynamically announces new ACPs and withdraws departed ACPs in its BGP 598 updates to Relays. Clients are expected to remain associated with 599 their current Servers for extended timeframes, however Servers SHOULD 600 selectively suppress BGP updates for impatient Clients that 601 repeatedly associate and disassociate with them in order to reduce 602 routing churn. 604 In some environments, Relays need not send BGP updates to Servers 605 since Servers can always use Relays as default routers, however this 606 presents a data/control plane performance tradeoff. In environments 607 where sustained packet forwarding over Relays is undesirable, Relays 608 can instead report ACPs to Servers while including a BGP Remote-Next- 609 Hop [I-D.vandevelde-idr-remote-next-hop]. The Server then creates a 610 neighbor cache entry for each ACP with the Remote-Next-Hop as the 611 link-layer address to enable Server-to-Server route optimization. 613 3.8. AERO Interface Neighbor Cache Maintenace 615 Each AERO interface maintains a conceptual neighbor cache that 616 includes an entry for each neighbor it communicates with on the AERO 617 link, the same as for any IPv6 interface [RFC4861]. AERO interface 618 neighbor cache entires are said to be one of "permanent", "static" or 619 "dynamic". 621 Permanent neighbor cache entries are created through explicit 622 administrative action; they have no timeout values and remain in 623 place until explicitly deleted. AERO Relays maintain a permanent 624 neighbor cache entry for each Server on the link, and AERO Servers 625 maintain a permanent neighbor cache entry for each Relay. Each entry 626 maintains the mapping between the neighbor's fe80::ID network-layer 627 address and corresponding link-layer address. 629 Static neighbor cache entries are created though DHCPv6 PD exchanges 630 and remain in place for durations bounded by prefix lifetimes. AERO 631 Servers maintain a static neighbor cache entry for each ACP of each 632 of their associated Clients, and AERO Clients maintain a static 633 neighbor cache for each of their associated Servers. When an AERO 634 Server sends a DHCPv6 Reply message response to a Client's DHCPv6 635 Solicit/Request or Renew message, it creates or updates a static 636 neighbor cache entry based on the AERO address corresponding to the 637 Client's ACP as the network-layer address, the prefix lifetime as the 638 neighbor cache entry lifetime, the Client's encapsulation IP address 639 and UDP port number as the link-layer address and the prefix length 640 as the length to apply to the AERO address. When an AERO Client 641 receives a DHCPv6 Reply message from a Server, it creates or updates 642 a static neighbor cache entry based on the Reply message link-local 643 source address as the network-layer address, the prefix lifetime as 644 the neighbor cache entry lifetime, and the encapsulation IP source 645 address and UDP source port number as the link-layer address. 647 Dynamic neighbor cache entries are created or updated based on 648 receipt of an IPv6 ND message, and are garbage-collected if not used 649 within a short timescale. AERO Clients maintain dynamic neighbor 650 cache entries for each of their active correspondent Client ACPs with 651 lifetimes based on IPv6 ND messaging constants. When an AERO Client 652 receives a valid Predirect message it creates or updates a dynamic 653 neighbor cache entry for the Predirect target network-layer and link- 654 layer addresses plus prefix length. The node then sets an 655 "AcceptTime" variable in the neighbor cache entry to ACCEPT_TIME 656 seconds and uses this value to determine whether packets received 657 from the correspondent can be accepted. When an AERO Client receives 658 a valid Redirect message it creates or updates a dynamic neighbor 659 cache entry for the Redirect target network-layer and link-layer 660 addresses plus prefix length. The Client then sets a "ForwardTime" 661 variable in the neighbor cache entry to FORWARD_TIME seconds and uses 662 this value to determine whether packets can be sent directly to the 663 correspondent. The Client also sets a "MaxRetry" variable to 664 MAX_RETRY to limit the number of keepalives sent when a correspondent 665 may have gone unreachable. 667 For dynamic neighbor cache entries, when an AERO Client receives a 668 valid NS message it (re)sets AcceptTime for the neighbor to 669 ACCEPT_TIME. When an AERO Client receives a valid solicited NA 670 message, it (re)sets ForwardTime for the neighbor to FORWARD_TIME and 671 sets MaxRetry to MAX_RETRY. When an AERO Client receives a valid 672 unsolicited NA message, it updates the correspondent's link-layer 673 addresses but DOES NOT reset AcceptTime, ForwardTime or MaxRetry. 675 It is RECOMMENDED that FORWARD_TIME be set to the default constant 676 value 30 seconds to match the default REACHABLE_TIME value specified 677 for IPv6 ND [RFC4861]. 679 It is RECOMMENDED that ACCEPT_TIME be set to the default constant 680 value 40 seconds to allow a 10 second window so that the AERO 681 redirection procedure can converge before AcceptTime decrements below 682 FORWARD_TIME. 684 It is RECOMMENDED that MAX_RETRY be set to 3 the same as described 685 for IPv6 ND address resolution in Section 7.3.3 of [RFC4861]. 687 Different values for FORWARD_TIME, ACCEPT_TIME, and MAX_RETRY MAY be 688 administratively set, if necessary, to better match the AERO link's 689 performance characteristics; however, if different values are chosen, 690 all nodes on the link MUST consistently configure the same values. 691 Most importantly, ACCEPT_TIME SHOULD be set to a value that is 692 sufficiently longer than FORWARD_TIME to allow the AERO redirection 693 procedure to converge. 695 3.9. AERO Interface Sending Algorithm 697 IP packets enter a node's AERO interface either from the network 698 layer (i.e., from a local application or the IP forwarding system), 699 or from the link layer (i.e., from the AERO tunnel virtual link). 700 Packets that enter the AERO interface from the network layer are 701 encapsulated and admitted into the AERO link, i.e., they are 702 tunnelled to an AERO interface neighbor. Packets that enter the AERO 703 interface from the link layer are either re-admitted into the AERO 704 link or delivered to the network layer where they are subject to 705 either local delivery or IP forwarding. Since each AERO node may 706 have only partial information about neighbors on the link, AERO 707 interfaces may forward packets with link-local destination addresses 708 at a layer below the network layer. This means that AERO nodes act 709 as both IP routers and sub-IP layer forwarding agents. AERO 710 interface sending considerations for Clients, Servers and Relays are 711 given below. 713 When an IP packet enters a Client's AERO interface from the network 714 layer, if the destination is covered by an ASP the Client searches 715 for a dynamic neighbor cache entry with a non-zero ForwardTime and an 716 AERO address that matches the packet's destination address. (The 717 destination address may be either an address covered by the 718 neighbor's ACP or the (link-local) AERO address itself.) If there is 719 a match, the Client uses a link-layer address in the entry as the 720 link-layer address for encapsulation then admits the packet into the 721 AERO link. If there is no match, the Client instead uses the link- 722 layer address of a neighboring Server as the link-layer address for 723 encapsulation. 725 When an IP packet enters a Server's AERO interface from the link 726 layer, if the destination is covered by an ASP the Server searches 727 for a neighbor cache entry with an AERO address that matches the 728 packet's destination address. (The destination address may be either 729 an address covered by the neighbor's ACP or the AERO address itself.) 730 If there is a match, the Server uses a link-layer address in the 731 entry as the link-layer address for encapsulation and re-admits the 732 packet into the AERO link. If there is no match, the Server instead 733 uses the link-layer address in a permanent neighbor cache entry for a 734 Relay as the link-layer address for encapsulation. 736 When an IP packet enters a Relay's AERO interface from the network 737 layer, the Relay searches its IP forwarding table for an entry that 738 is covered by an ASP and also matches the destination. If there is a 739 match, the Relay uses the link-layer address in a permanent neighbor 740 cache entry for a Server as the link-layer address for encapsulation 741 and admits the packet into the AERO link. When an IP packet enters a 742 Relay's AERO interface from the link-layer, if the destination is not 743 a link-local address and does not match an ASP the Relay removes the 744 packet from the AERO interface and uses IP forwarding to forward the 745 packet to the Internetwork. If the destination address is a link- 746 local address or a non-link-local address that matches an ASP, and 747 there is a more-specific ACP entry in the IP forwarding table, the 748 Relay uses the link-layer address in the corresponding neighbor cache 749 entry as the link-layer address for encapsulation and re-admits the 750 packet into the AERO link. When an IP packet enters a Relay's AERO 751 interface from either the network layer or link-layer, and the 752 packet's destination address matches an ASP but there is no more- 753 specific ACP entry, the Relay drops the packet and returns an ICMP 754 Destination Unreachable message (see: Section 3.13). 756 When an AERO Server receives a packet from a Relay via the AERO 757 interface, the Server MUST NOT forward the packet back to the same or 758 a different Relay. 760 When an AERO Relay receives a packet from a Server via the AERO 761 interface, the Relay MUST NOT forward the packet back to the same 762 Server. 764 When an AERO node re-admits a packet into the AERO link without 765 involving the network layer, the node MUST NOT decrement the network 766 layer TTL/Hop-count. 768 When an AERO node forwards a data packet to the primary link-layer 769 address of a neighbor, it may receive RA messages with one or more 770 SLLAOs that include the link-layer addresses of AERO Forwarding 771 Agents. The AERO node SHOULD record the link-layer addresses in the 772 neighbor cache entry for the neighbor and send subsequent data 773 packets via one of these addresses instead of the neighbor's primary 774 address (see: Section 3.15). 776 3.10. AERO Interface Encapsulation, Re-encapsulation and Decapsulation 778 AERO interfaces encapsulate IP packets according to whether they are 779 entering the AERO interface from the network layer or if they are 780 being re-admitted into the same AERO link they arrived on. This 781 latter form of encapsulation is known as "re-encapsulation". 783 The AERO interface encapsulates packets per the base tunneling 784 specifications (e.g., 785 [RFC2003][RFC2473][RFC2784][RFC4213][RFC4301][RFC5246], etc.) except 786 that it inserts a UDP header immediately following the IP 787 encapsulation header and immediately before the next header. 789 If the next header is an IPv4 or IPv6 header and the packet is an 790 unfragmented ordinary data packet, no other encapsulations are 791 necessary. For all others (including IPv6 ND and DHCPv6 messages), 792 the AERO interface MUST insert an AERO shim header immediately 793 following the UDP header formatted as shown in Figure 3: 795 0 1 796 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 797 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 798 | Vers1 | Vers2 | Next Header | 799 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 801 Figure 3: AERO Shim Header 803 In the AERO shim header, "Vers1" encodes the value '0', "Vers2" 804 encodes the value '1', and "Next Header" encodes the IP protocol 805 number corresponding to the next header in the encapsulation. For 806 example, "Next Header" encodes the value '4' for an IPv4 header, '17' 807 for a UDP header, '41' for an IPv6 header, '44' for the IPv6 Fragment 808 Header, '47' for GRE, '50' for ESP, '51' for AH, etc. (other Next 809 Header values are found in the IANA "protocol numbers" registry). 811 During encapsulation, the AERO interface copies the "TTL/Hop Limit", 812 "Type of Service/Traffic Class" and "Congestion Experienced" values 813 in the packet's IP header into the corresponding fields in the 814 encapsulation IP header. For packets undergoing re-encapsulation, 815 the AERO interface instead copies the "TTL/Hop Limit", "Type of 816 Service/Traffic Class" and "Congestion Experienced" values in the 817 original encapsulation IP header into the corresponding fields in the 818 new encapsulation IP header, i.e., the values are transferred between 819 encapsulation headers and *not* copied from the encapsulated packet's 820 network-layer header. (Note that these instructions may deviate 821 slightly from those found in the base tunneling specifications.) 823 The AERO interface next sets the UDP source port to a constant value 824 that it will use in each successive packet it sends, and sets the UDP 825 length field to the length of the encapsulated packet plus 8 bytes 826 for the UDP header itself (or plus 10 bytes if an AERO shim header is 827 also included). For packets sent via a Server, the AERO interface 828 sets the UDP destination port to 8060, i.e., the IANA-registered port 829 number for AERO. For packets sent to a correspondent Client, the 830 AERO interface sets the UDP destination port to the port value stored 831 in the neighbor cache entry for this correspondent. The AERO 832 interface also sets the UDP checksum field to zero (see: 833 [RFC6935][RFC6936]) unless an integrity check is required (see: 834 Section 3.12.2). 836 The AERO interface next sets the IP protocol number in the 837 encapsulation header to 17 (i.e., the IP protocol number for UDP). 838 When IPv6 is used as the encapsulation protocol, the interface then 839 sets the flow label value in the encapsulation header the same as 840 described in [RFC6438]. When IPv4 is used as the encapsulation 841 protocol, the AERO interface sets the DF bit as discussed in 842 Section 3.12. 844 AERO interfaces decapsulate packets destined either to the node 845 itself or to a destination reached via an interface other than the 846 AERO interface the packet was received on. When the AERO interface 847 receives a UDP packet, it examines the first octet of the 848 encapsulated packet. If the most significant four bits of the first 849 octet encode the value '6' (i.e., the IP version number value for 850 IPv6) or the value '4' (i.e., the IP version number value for IPv4), 851 the AERO interface discards the encapsulation headers and accepts the 852 encapsulated packet as an ordinary IPv6 or IPv4 data packet, 853 respectively. 855 If the most significant four bits encode the value '0' and the next 856 four bits encode the value '1', however, the AERO interface processes 857 the next octet as a "Next Header" field, i.e., the interface treats 858 the first two octets of the encapsulated packet as an AERO shim 859 header as shown in Figure 3 (note that the "Vers2" value is set to 1 860 to distinguish AERO encapsulations from the experimental message 861 formats specified in [RFC6706]). Processing then proceeds according 862 to the appropriate base tunneling specification and/or control 863 message type. 865 3.11. AERO Interface Data Origin Authentication 867 AERO nodes employ simple data origin authentication procedures for 868 encapsulated packets they receive from other nodes on the AERO link. 869 In particular: 871 o AERO Relays and Servers accept encapsulated packets with a link- 872 layer source address that matches a permanent neighbor cache 873 entry. 875 o AERO Servers accept authentic encapsulated DHCPv6 messages from 876 Clients, and create or update a static neighbor cache entry for 877 the source based on the specific message type. 879 o AERO Servers accept encapsulated packets if there is a neighbor 880 cache entry with an AERO address that matches the packet's 881 network-layer source address and with a link-layer address that 882 matches the packet's link-layer source address. 884 o AERO Clients accept encapsulated packets if there is a static 885 neighbor cache entry with a link-layer source address that matches 886 the packet's link-layer source address. 888 o AERO Clients and Servers accept encapsulated packets if there is a 889 dynamic neighbor cache entry with an AERO address that matches the 890 packet's network-layer source address, with a link-layer address 891 that matches the packet's link-layer source address, and with a 892 non-zero AcceptTime. 894 Note that this simple data origin authentication is effective in 895 environments in which link-layer addresses cannot be spoofed. 896 Additional security mitigations may be necessary in other 897 environments. 899 3.12. AERO Interface MTU and Fragmentation 901 The AERO interface is the node's point of attachment to the AERO 902 link. AERO links over IP networks have a maximum link MTU of 64KB 903 minus the encapsulation overhead (termed here "ENCAPS"), since the 904 maximum packet size in the base IP specifications is 64KB 905 [RFC0791][RFC2460] (while IPv6 jumbograms can be up to 4GB, they are 906 considered optional for IPv6 nodes [RFC2675][RFC6434]). 908 IPv6 specifies a minimum link MTU of 1280 bytes [RFC2460]. This is 909 the minimum packet size the AERO interface MUST admit without 910 returning an ICMP Packet Too Big (PTB) message. Although IPv4 911 specifies a smaller minimum link MTU of 68 bytes [RFC0791], AERO 912 interfaces also observe a 1280 byte minimum for IPv4. Additionally, 913 the vast majority of links in the Internet configure an MTU of at 914 least 1500 bytes. Original source hosts have therefore become 915 conditioned to expect that IP packets up to 1500 bytes in length will 916 either be delivered to the final destination or a suitable PTB 917 message returned. However, PTB messages may be lost in the network 918 [RFC2923] resulting in failure of the IP MTU discovery mechanisms 919 [RFC1191][RFC1981]. 921 For these reasons, AERO interfaces MUST admit packets up to 1500 922 bytes in length even if some fragmentation is necessary. AERO 923 interfaces MAY admit even larger packets as long as they can be 924 accommodated without fragmentation. 926 For AERO links over IPv4, the IP ID field is only 16 bits in length, 927 meaning that fragmentation at high data rates could result in data 928 corruption due to reassembly misassociations [RFC6864][RFC4963] (see: 929 Section 3.12.2). For AERO links over both IPv4 and IPv6, studies 930 have also shown that IP fragments are dropped unconditionally over 931 some network paths [I-D.taylor-v6ops-fragdrop]. For these reasons, 932 when fragmentation is needed it is performed within the AERO 933 interface (i.e., instead of at the encapsulating IP layer) through 934 the insertion of an IPv6 Fragment Header [RFC2460]. Since the 935 Fragment Header reduces the room available for packet data, but the 936 original source has no way to control its insertion, the Fragment 937 Header length plus the length of the AERO shim header (see: 939 Section 3.10) MUST be included in the ENCAPS length even for packets 940 in which the headers do not appear. 942 The source AERO interface (i.e., the tunnel ingress) therefore sends 943 encapsulated packets to the destination AERO interface (i.e., the 944 tunnel egress) according to the following algorithm: 946 o For IP packets that are no larger than (1280-ENCAPS) bytes, the 947 tunnel ingress encapsulates the packet and admits it into the 948 tunnel without fragmentation. For IPv4 AERO links, the tunnel 949 ingress sets the Don't Fragment (DF) bit to 0 so that these 950 packets will be delivered to the tunnel egress even if there is a 951 restricting link in the path, i.e., unless lost due to congestion 952 or routing errors. 954 o For IP packets that are larger than (1280-ENCAPS) bytes but no 955 larger than 1500 bytes, the tunnel ingress encapsulates the packet 956 and inserts a Fragment Header and AERO shim header above the UDP/ 957 IP encapsulation headers. Next, the tunnel ingress uses the 958 fragmentation algorithm in [RFC2460] to break the packet into two 959 non-overlapping fragments where the first fragment (including 960 ENCAPS) is no larger than 1024 bytes and the second is no larger 961 than the first. Each fragment consists of identical AERO/UDP/IP 962 encapsulation headers, followed by the Fragment Header followed by 963 the fragment of the encapsulated packet itself. The tunnel 964 ingress then admits both fragments into the tunnel, and for IPv4 965 sets the DF bit to 0 in the IP encapsulation header. These 966 fragmented encapsulated packets will be delivered to the tunnel 967 egress. 969 o For IPv4 packets that are larger than 1500 bytes and with the DF 970 bit set to 0, the tunnel ingress uses ordinary IP fragmentation to 971 break the unencapsulated packet into a minimum number of non- 972 overlapping fragments where the first fragment is no larger than 973 1024-ENCAPS and all other fragments are no larger than the first 974 fragment. The tunnel ingress then encapsulates each fragment (and 975 for IPv4 sets the DF bit to 0) then admits them into the tunnel. 976 These encapsulated fragments will be delivered to the final 977 destination via the tunnel egress. 979 o For all other IP packets, if the packet is too large to enter the 980 underlying interface following encapsulation, the tunnel ingress 981 drops the packet and returns a network-layer (L3) PTB message to 982 the original source with MTU set to the larger of 1500 bytes or 983 the underlying interface MTU minus ENCAPS. Otherwise, the tunnel 984 ingress encapsulates the packet and admits it into the tunnel 985 without fragmentation (and for IPv4 sets the DF bit to 1) and 986 translates any link-layer (L2) PTB messages it may receive from 987 the network into corresponding L3 PTB messages to send to the 988 original source as specified in Section 3.13. Since both L2 and 989 L3 PTB messages may be either lost or contain insufficient 990 information, however, it is RECOMMENDED that original sources that 991 send unfragmentable IP packets larger than 1500 bytes use 992 Packetization Layer Path MTU Discovery (PLPMTUD) [RFC4821]. 994 While sending packets according to the above algorithm, the tunnel 995 ingress MAY also send 1500 byte probe packets to determine whether 996 they can reach the tunnel egress without fragmentation. If the 997 probes succeed, the tunnel ingress can begin sending packets that are 998 larger than (1280-ENCAPS) bytes but no larger than 1500 bytes without 999 fragmentation (and for IPv4 with DF set to 1). Since the path MTU 1000 within the tunnel may fluctuate due to routing changes, the tunnel 1001 ingress SHOULD continue to send additional probes subject to rate 1002 limiting and SHOULD process any L2 PTB messages as an indication that 1003 the path MTU may have decreased. If the path MTU within the tunnel 1004 becomes insufficient, the source MUST resume fragmentation. 1006 To construct a probe, the tunnel ingress prepares an NS message with 1007 a Nonce option plus trailing NULL padding octets added to a length of 1008 1500 bytes without including the length of the padding in the IPv6 1009 Payload Length field, but with the length included in the 1010 encapsulating IP header. The tunnel ingress then encapsulates the 1011 padded NS message in the encapsulation headers (and for IPv4 sets DF 1012 to 1) then sends the message to the tunnel egress. If the tunnel 1013 egress returns a solicited NA message with a matching Nonce option, 1014 the tunnel ingress deems the probe successful. 1016 When the tunnel egress receives the fragments of a fragmented packet, 1017 it reassembles them into a whole packet per the reassembly algorithm 1018 in [RFC2460] then discards the Fragment Header. The tunnel egress 1019 therefore MUST be capable of reassembling packets up to 1500+ENCAPS 1020 bytes in length; hence, it is RECOMMENDED that the tunnel egress be 1021 capable of reassembling at least 2KB. 1023 3.12.1. Accommodating Large Control Messages 1025 Control messages (i.e., IPv6 ND, DHCPv6, etc.) MUST be accommodated 1026 even if some fragmentation is necessary. These packets are therefore 1027 accommodated through a modification of the second rule in the above 1028 algorithm as follows: 1030 o For control messages that are larger than (1280-ENCAPS) bytes, the 1031 tunnel ingress encapsulates the packet and inserts a Fragment 1032 Header and AERO shim header above the UDP/IP encapsulation 1033 headers. Next, the tunnel ingress uses the fragmentation 1034 algorithm in [RFC2460] to break the packet into a minimum number 1035 of non-overlapping fragments where the first fragment (including 1036 ENCAPS) is no larger than 1024 bytes and the remaining fragments 1037 are no larger than the first. The tunnel ingress then 1038 encapsulates each fragment (and for IPv4 sets the DF bit to 0) 1039 then admits them into the tunnel. 1041 Control messages that exceed the minimum reassembly size listed above 1042 rarely occur in the modern era, however the tunnel egress SHOULD be 1043 able to reassemble them if they do. This means that the tunnel 1044 egress SHOULD include a configuration knob allowing the operator to 1045 set a larger reassembly buffer size if large control messages become 1046 more common in the future. 1048 The tunnel ingress can send large control messages without 1049 fragmentation if there is assurance that large packets can traverse 1050 the tunnel without fragmentation. The tunnel ingress MAY send probe 1051 packets of 1500 bytes or larger as specified above to determine a 1052 size for which fragmentation can be avoided. 1054 3.12.2. Integrity 1056 When fragmentation is needed, there must be assurance that reassembly 1057 can be safely conducted without incurring data corruption. Sources 1058 of corruption can include implementation errors, memory errors and 1059 misassociation of fragments from a first datagram with fragments of 1060 another datagram. The first two conditions (implementation and 1061 memory errors) are mitigated by modern systems and implementations 1062 that have demonstrated integrity through decades of operational 1063 practice. The third condition (reassembly misassociations) must be 1064 accounted for by AERO. 1066 The AERO fragmentation procedure described in the above algorithms 1067 uses the IPv6 Fragment Header and reuses standard IPv6 fragmentation 1068 and reassembly code. Since the Fragment Header includes a 32-bit ID 1069 field, there would need to be 2^32 packets alive in the network 1070 before a second packet with a duplicate ID enters the system with the 1071 (remote) possibility for a reassembly misassociation. For 1280 byte 1072 packets, and for a maximum network lifetime value of 60 1073 seconds[RFC2460], this means that the tunnel ingress would need to 1074 produce ~(7 *10^12) bits/sec in order for a duplication event to be 1075 possible. This exceeds the bandwidth of data link technologies of 1076 the modern era, but not necessarily so going forward into the future. 1077 Although wireless data links commonly used by AERO Clients support 1078 vastly lower data rates, the aggregate data rates between AERO 1079 Servers and Relays may be substantial. However, high speed data 1080 links in the network core are expected to configure larger MTUs, 1081 e.g., 4KB, 8KB or even larger such that unfragmented packets can be 1082 used. Hence, no integrity check is included to cover the AERO 1083 fragmentation and reassembly procedures. 1085 When the tunnel ingress sends an IPv4-encapsulated packet with the DF 1086 bit set to 0 in the above algorithms, there is a chance that the 1087 packet may be fragmented by an IPv4 router somewhere within the 1088 tunnel. Since the largest such packet is only 1280 bytes, however, 1089 it is very likely that the packet will traverse the tunnel without 1090 incurring a restricting link. Even when a link within the tunnel 1091 configures an MTU smaller than 1280 bytes, it is very likely that it 1092 does so due to limited performance characteristics [RFC3819]. This 1093 means that the tunnel would not be able to convey fragmented 1094 IPv4-encapsulated packets fast enough to produce reassembly 1095 misassociations, as discussed above. However, AERO must also account 1096 for the possibility of tunnel paths that include "poorly managed" 1097 IPv4 link MTUs. 1099 Since the IPv4 header includes only a 16-bit ID field, there would 1100 only need to be 2^16 packets alive in the network before a second 1101 packet with a duplicate ID enters the system. For 1280 byte packets, 1102 and for a maximum network lifetime value of 120 seconds[RFC0791], 1103 this means that the tunnel ingress would only need to produce ~(5 1104 *10^6) bits/sec in order for a duplication event to be possible - a 1105 value that is well within range for many modern wired and wireless 1106 data link technologies. 1108 Therefore, if there is strong operational assurance that no IPv4 1109 links capable of supporting data rates of 5Mbps or more configure an 1110 MTU smaller than 1280 the tunnel ingress MAY omit an integrity check 1111 for the IPv4 fragmentation and reassembly procedures; otherwise, the 1112 tunnel ingress SHOULD include an integrity check. When an upper 1113 layer encapsulation (e.g., IPsec) already includes an integrity 1114 check, the tunnel ingress need not include an additional check. 1115 Otherwise, the tunnel ingress calculates the UDP checksum over the 1116 encapsulated packet and writes the value into the UDP encapsulation 1117 header, i.e., instead of writing the value 0. The tunnel egress will 1118 then verify the UDP checksum and discard the packet if the checksum 1119 is incorrect. 1121 3.13. AERO Interface Error Handling 1123 When an AERO node admits encapsulated packets into the AERO 1124 interface, it may receive link-layer (L2) or network-layer (L3) error 1125 indications. 1127 An L2 error indication is an ICMP error message generated by a router 1128 on the path to the neighbor or by the neighbor itself. The message 1129 includes an IP header with the address of the node that generated the 1130 error as the source address and with the link-layer address of the 1131 AERO node as the destination address. 1133 The IP header is followed by an ICMP header that includes an error 1134 Type, Code and Checksum. For ICMPv6 [RFC4443], the error Types 1135 include "Destination Unreachable", "Packet Too Big (PTB)", "Time 1136 Exceeded" and "Parameter Problem". For ICMPv4 [RFC0792], the error 1137 Types include "Destination Unreachable", "Fragmentation Needed" (a 1138 Destination Unreachable Code that is analogous to the ICMPv6 PTB), 1139 "Time Exceeded" and "Parameter Problem". 1141 The ICMP header is followed by the leading portion of the packet that 1142 generated the error, also known as the "packet-in-error". For 1143 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1144 much of invoking packet as possible without the ICMPv6 packet 1145 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1146 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1147 "Internet Header + 64 bits of Original Data Datagram", however 1148 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1149 ICMP datagram SHOULD contain as much of the original datagram as 1150 possible without the length of the ICMP datagram exceeding 576 1151 bytes". 1153 The L2 error message format is shown in Figure 4: 1155 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1156 ~ ~ 1157 | L2 IP Header of | 1158 | error message | 1159 ~ ~ 1160 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1161 | L2 ICMP Header | 1162 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1163 ~ ~ P 1164 | IP and other encapsulation | a 1165 | headers of original L3 packet | c 1166 ~ ~ k 1167 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1168 ~ ~ t 1169 | IP header of | 1170 | original L3 packet | i 1171 ~ ~ n 1172 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1173 ~ ~ e 1174 | Upper layer headers and | r 1175 | leading portion of body | r 1176 | of the original L3 packet | o 1177 ~ ~ r 1178 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1180 Figure 4: AERO Interface L2 Error Message Format 1182 The AERO node rules for processing these L2 error messages is as 1183 follows: 1185 o When an AERO node receives an L2 Parameter Problem message, it 1186 processes the message the same as described as for ordinary ICMP 1187 errors in the normative references [RFC0792][RFC4443]. 1189 o When an AERO node receives persistent L2 IPv4 Time Exceeded 1190 messages, the IP ID field may be wrapping before earlier fragments 1191 have been processed. In that case, the node SHOULD begin 1192 including IPv4 integrity checks (see: Section 3.12.2). 1194 o When an AERO Client receives persistent L2 Destination Unreachable 1195 messages in response to tunneled packets that it sends to one of 1196 its dynamic neighbor correspondents, the Client SHOULD test the 1197 path to the correspondent using Neighbor Unreachability Detection 1198 (NUD) (see Section 3.17). If NUD fails, the Client SHOULD set 1199 ForwardTime for the corresponding dynamic neighbor cache entry to 1200 0 and allow future packets destined to the correspondent to flow 1201 through a Server. 1203 o When an AERO Client receives persistent L2 Destination Unreachable 1204 messages in response to tunneled packets that it sends to one of 1205 its static neighbor Servers, the Client SHOULD test the path to 1206 the Server using NUD. If NUD fails, the Client SHOULD delete the 1207 neighbor cache entry and attempt to associate with a new Server. 1209 o When an AERO Server receives persistent L2 Destination Unreachable 1210 messages in response to tunneled packets that it sends to one of 1211 its static neighbor Clients, the Server SHOULD test the path to 1212 the Client using NUD. If NUD fails, the Server SHOULD cancel the 1213 DHCPv6 PD lease for the Client's ACP, withdraw its route for the 1214 ACP from the AERO routing system and delete the neighbor cache 1215 entry (see Section 3.17 and Section 3.18). 1217 o When an AERO Relay or Server receives an L2 Destination 1218 Unreachable message in response to a tunneled packet that it sends 1219 to one of its permanent neighbors, it discards the message since 1220 the routing system is likely in a temporary transitional state 1221 that will soon re-converge. 1223 o When an AERO node receives an L2 PTB message, it translates the 1224 message into an L3 PTB message if possible (*) and forwards the 1225 message toward the original source as described below. 1227 To translate an L2 PTB message to an L3 PTB message, the AERO node 1228 first caches the MTU field value of the L2 ICMP header. The node 1229 next discards the L2 IP and ICMP headers, and also discards the 1230 encapsulation headers of the original L3 packet. Next the node 1231 encapsulates the included segment of the original L3 packet in an L3 1232 IP and ICMP header, and sets the ICMP header Type and Code values to 1233 appropriate values for the L3 IP protocol. In the process, the node 1234 writes the maximum of 1500 bytes and (L2 MTU - ENCAPS) into the MTU 1235 field of the L3 ICMP header. 1237 The node next writes the IP source address of the original L3 packet 1238 as the destination address of the L3 PTB message and determines the 1239 next hop to the destination. If the next hop is reached via the AERO 1240 interface, the node uses the IPv6 address "::" or the IPv4 address 1241 "0.0.0.0" as the IP source address of the L3 PTB message. Otherwise, 1242 the node uses one of its non link-local addresses as the source 1243 address of the L3 PTB message. The node finally calculates the ICMP 1244 checksum over the L3 PTB message and writes the Checksum in the 1245 corresponding field of the L3 ICMP header. The L3 PTB message 1246 therefore is formatted as follows: 1248 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1249 ~ ~ 1250 | L3 IP Header of | 1251 | error message | 1252 ~ ~ 1253 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1254 | L3 ICMP Header | 1255 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1256 ~ ~ p 1257 | IP header of | k 1258 | original L3 packet | t 1259 ~ ~ 1260 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ i 1261 ~ ~ n 1262 | Upper layer headers and | 1263 | leading portion of body | e 1264 | of the original L3 packet | r 1265 ~ ~ r 1266 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1268 Figure 5: AERO Interface L3 Error Message Format 1270 After the node has prepared the L3 PTB message, it either forwards 1271 the message via a link outside of the AERO interface without 1272 encapsulation, or encapsulates and forwards the message to the next 1273 hop via the AERO interface. 1275 When an AERO Relay receives an L3 packet for which the destination 1276 address is covered by an ASP, if there is no more-specific routing 1277 information for the destination the Relay drops the packet and 1278 returns an L3 Destination Unreachable message. The Relay first 1279 writes the IP source address of the original L3 packet as the 1280 destination address of the L3 Destination Unreachable message and 1281 determines the next hop to the destination. If the next hop is 1282 reached via the AERO interface, the Relay uses the IPv6 address "::" 1283 or the IPv4 address "0.0.0.0" as the IP source address of the L3 1284 Destination Unreachable message and forwards the message to the next 1285 hop within the AERO interface. Otherwise, the Relay uses one of its 1286 non link-local addresses as the source address of the L3 Destination 1287 Unreachable message and forwards the message via a link outside the 1288 AERO interface. 1290 When an AERO node receives any L3 error message via the AERO 1291 interface, it examines the destination address in the L3 IP header of 1292 the message. If the next hop toward the destination address of the 1293 error message is via the AERO interface, the node re-encapsulates and 1294 forwards the message to the next hop within the AERO interface. 1295 Otherwise, if the source address in the L3 IP header of the message 1296 is the IPv6 address "::" or the IPv4 address "0.0.0.0", the node 1297 writes one of its non link-local addresses as the source address of 1298 the L3 message and recalculates the IP and/or ICMP checksums. The 1299 node finally forwards the message via a link outside of the AERO 1300 interface. 1302 (*) Note that in some instances the packet-in-error field of an L2 1303 PTB message may not include enough information for translation to an 1304 L3 PTB message. In that case, the AERO interface simply discards the 1305 L2 PTB message. It can therefore be said that translation of L2 PTB 1306 messages to L3 PTB messages can provide a useful optimization when 1307 possible, but is not critical for sources that correctly use PLPMTUD. 1309 3.14. AERO Router Discovery, Prefix Delegation and Address 1310 Configuration 1312 3.14.1. AERO DHCPv6 Service Model 1314 Each AERO Server configures a DHCPv6 server function to facilitate PD 1315 requests from Clients. Each Server is pre-configured with an 1316 identical list of ACP-to-Client ID mappings for all Clients enrolled 1317 in the AERO system, as well as any information necessary to 1318 authenticate Clients. The configuration information is maintained by 1319 a central administrative authority for the AERO link and securely 1320 propagated to all Servers whenever a new Client is enrolled or an 1321 existing Client is withdrawn. 1323 With these identical configurations, each Server can function 1324 independently of all other Servers, including the maintenance of 1325 active leases. Therefore, no Server-to-Server DHCPv6 state 1326 synchronization is necessary, and Clients can optionally hold 1327 separate leases for the same ACP from multiple Servers. 1329 In this way, Clients can easily associate with multiple Servers, and 1330 can receive new leases from new Servers before deprecating leases 1331 held through old Servers. This enables a graceful "make-before- 1332 break" capability. 1334 3.14.2. AERO Client Behavior 1336 AERO Clients discover the link-layer addresses of AERO Servers via 1337 static configuration, or through an automated means such as DNS name 1338 resolution. In the absence of other information, the Client resolves 1339 the FQDN "linkupnetworks.[domainname]" where "linkupnetworks" is a 1340 constant text string and "[domainname]" is the connection-specific 1341 DNS suffix for the Client's underlying network connection (e.g., 1342 "example.com"). After discovering the link-layer addresses, the 1343 Client associates with one or more of the corresponding Servers. 1345 To associate with a Server, the Client acts as a requesting router to 1346 request an ACP through a DHCPv6 PD exchange[RFC3315][RFC3633] in 1347 which the Client's Solicit/Request messages use the IPv6 1348 "unspecified" address (i.e., "::") as the IPv6 source address, 1349 'All_DHCP_Relay_Agents_and_Servers' as the IPv6 destination address 1350 and the link-layer address of the Server as the link-layer 1351 destination address. The Client also includes a Client Identifier 1352 option with a DHCP Unique Identifier (DUID) plus any necessary 1353 authentication options to identify itself to the DHCPv6 server, and 1354 includes a Client Link Layer Address Option (CLLAO) [RFC6939] with 1355 the format shown in Figure 6: 1357 0 1 2 3 1358 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 1359 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1360 | OPTION_CLIENT_LINKLAYER_ADDR | option-length | 1361 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1362 | link-layer type (16 bits) | Link ID | Preference | 1363 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1365 Figure 6: AERO Client Link-Layer Address Option (CLLAO) Format 1367 The Client sets the CLLAO 'option-length' field to 4 and sets the 1368 'link-layer type' field to TBD1 (see: IANA Considerations), then 1369 includes appropriate Link ID and Preference values for the underlying 1370 interface over which the Solicit/Request will be issued (note that 1371 these are the same values that would be included in an S/TLLAO as 1372 shown in Figure 2). If the Client is pre-provisioned with an ACP 1373 associated with the AERO service, it MAY also include the ACP in a 1374 Solicit/Request message Identity Association for Prefix Delegation 1375 (IA_PD) option to indicate its preferred ACP to the DHCPv6 server. 1376 The Client then sends the encapsulated DHCPv6 request via the 1377 underlying interface. 1379 When the Client receives its ACP via a DHCPv6 Reply from the AERO 1380 Server, it creates a static neighbor cache entry with the Server's 1381 link-local address as the network-layer address and the Server's 1382 encapsulation address as the link-layer address. If the Reply 1383 message contains a Vendor-Specific Information Option (see: 1384 Section 3.14.3) the Client also caches each ASP in the option. The 1385 Client then considers the link-layer address of the Server as the 1386 primary default encapsulation address for forwarding packets for 1387 which there is no more-specific forwarding information. 1389 Next, the Client applies the AERO address to the AERO interface and 1390 sub-delegates the ACP to nodes and links within its attached EUNs 1391 (the AERO address thereafter remains stable as the Client moves). 1392 The Client also assigns a default IP route to the AERO interface as a 1393 route-to-interface, i.e., with no explicit next-hop. The next hop 1394 will then be determined after a packet has been submitted to the AERO 1395 interface by inspecting the neighbor cache (see above). 1397 On some platforms (e.g., popular cell phone operating systems), the 1398 act of assigning a default IPv6 route may not be permitted from a 1399 user application due to security policy. Typically, those platforms 1400 include a TUN/TAP interface that acts as a point-to-point conduit 1401 between user applications and the AERO interface. In that case, the 1402 Client can instead generate a "synthesized RA" message. The message 1403 conforms to [RFC4861] and is prepared as follows: 1405 o the IPv6 source address is fe80:: 1407 o the IPv6 destination address is all-nodes multicast 1409 o the Router Lifetime is set to a time that is no longer than the 1410 ACP DHCPv6 lifetime 1412 o the message does not include a Source Link Layer Address Option 1413 (SLLAO) 1415 o the message includes a Prefix Information Option (PIO) with a /64 1416 prefix taken from the ACP as the prefix for autoconfiguration 1418 The Client then sends the synthesized RA message via the TUN/TAP 1419 interface, where the operating system kernel will interpret it as 1420 though it were generated by an actual router. The operating system 1421 will then install a default route and use StateLess Address 1422 AutoConfiguration (SLAAC) to configure an IPv6 address on the TUN/TAP 1423 interface. Methods for similarly installing an IPv4 default route 1424 and IPv4 address on the TUN/TAP interface are based on synthesized 1425 DHCPv4 messages [RFC2131]. Note that in this method, the Client 1426 appears as a mobility proxy for applications that bind to the (point- 1427 to-point) TUN/TAP interface. The arrangement can be likened to a 1428 Proxy AERO scenario in which the mobile node and Client are located 1429 within the same physical platform (see Section 3.23 for further 1430 details on Proxy AERO). 1432 The Client subsequently renews its ACP delegation through each of its 1433 Servers by performing DHCPv6 Renew/Reply exchanges with its AERO 1434 address as the IPv6 source address, 1435 'All_DHCP_Relay_Agents_and_Servers' as the IPv6 destination address, 1436 the link-layer address of a Server as the link-layer destination 1437 address and the same Client identifier, authentication options and 1438 CLLAO option as was used in the initial PD request. Note that if the 1439 Client does not issue a DHCPv6 Renew before the Server has terminated 1440 the lease (e.g., if the Client has been out of touch with the Server 1441 for a considerable amount of time), the Server's Reply will report 1442 NoBinding and the Client must re-initiate the DHCPv6 PD procedure. 1443 If the Client sends synthesized RA and/or DHCPv4 messages (see 1444 above), it also sends a new synthesized message when issuing a DHCPv6 1445 Renew or when re-initiating the DHCPv6 PD procedure. 1447 Since the Client's AERO address is configured from the unique ACP 1448 delegation it receives, there is no need for Duplicate Address 1449 Detection (DAD) on AERO links. Other nodes maliciously attempting to 1450 hijack an authorized Client's AERO address will be denied access to 1451 the network by the DHCPv6 server due to an unacceptable link-layer 1452 address and/or security parameters (see: Security Considerations). 1454 3.14.3. AERO Server Behavior 1456 AERO Servers configure a DHCPv6 server function on their AERO links. 1457 AERO Servers arrange to add their encapsulation layer IP addresses 1458 (i.e., their link-layer addresses) to the DNS resource records for 1459 the FQDN "linkupnetworks.[domainname]" before entering service. 1461 When an AERO Server receives a prospective Client's DHCPv6 PD 1462 Solicit/Request message, it first authenticates the message. If 1463 authentication succeeds, the Server determines the correct ACP to 1464 delegate to the Client by matching the Client's DUID within an online 1465 directory service (e.g., LDAP). The Server then delegates the ACP 1466 and creates a static neighbor cache entry for the Client's AERO 1467 address with lifetime set to no more than the lease lifetime and the 1468 Client's link-layer address as the link-layer address for the Link ID 1469 specified in the CLLAO option. The Server then creates an IP 1470 forwarding table entry so that the AERO routing system will propagate 1471 the ACP to all Relays (see: Section 3.7). Finally, the Server sends 1472 a DHCPv6 Reply message to the Client while using fe80::ID as the IPv6 1473 source address, the Client's AERO address as the IPv6 destination 1474 address, and the Client's link-layer address as the destination link- 1475 layer address. The Server also includes a Server Unicast option with 1476 server-address set to fe80::ID so that all future Client/Server 1477 transactions will be link-local-only unicast over the AERO link. 1479 When the Server sends the DHCPv6 Reply message, it also includes a 1480 DHCPv6 Vendor-Specific Information Option with 'enterprise-number' 1481 set to "TBD2" (see: IANA Considerations). The option is formatted as 1482 shown in[RFC3315] and with the AERO enterprise-specific format shown 1483 in Figure 7: 1485 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 1486 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1487 | OPTION_VENDOR_OPTS | option-len | 1488 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1489 | enterprise-number ("TBD2") | 1490 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1491 | Reserved | Prefix Length | 1492 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1493 | | 1494 + ASP (1) + 1495 | | 1496 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1497 | Reserved | Prefix Length | 1498 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1499 | | 1500 + ASP (2) + 1501 | | 1502 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1503 | Reserved | Prefix Length | 1504 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1505 | | 1506 + ASP (3) + 1507 | | 1508 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1509 . (etc.) . 1510 . . 1511 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1513 Figure 7: AERO Vendor-Specific Information Option 1515 Per Figure 7, the option includes one or more ASP. The ASP field 1516 contains the IP prefix as it would appear in the interface identifier 1517 portion of the corresponding AERO address (see: Section 3.3). For 1518 IPv6, valid values for the Prefix Length field are 0 through 64; for 1519 IPv4, valid values are 0 through 32. 1521 After the initial DHCPv6 PD exchange, the AERO Server maintains the 1522 neighbor cache entry for the Client until the lease lifetime expires. 1523 If the Client issues a Renew/Reply exchange, the Server extends the 1524 lifetime. If the Client issues a Release/Reply, or if the Client 1525 does not issue a Renew/Reply before the lifetime expires, the Server 1526 deletes the neighbor cache entry for the Client and withdraws the IP 1527 route from the AERO routing system. 1529 3.15. AERO Forwarding Agent Discovery 1531 AERO Relays, Servers and Clients MAY associate with one or more 1532 companion AERO Forwarding Agents as platforms for offloading high- 1533 speed data plane traffic. AERO nodes distribute forwarding 1534 information to Forwarding Agents via an out-of-band messaging service 1535 (e.g., NETCONF [RFC6241], etc.). 1537 When an AERO node receives a data packet on an AERO interface with a 1538 network layer destination address for which it has distributed 1539 forwarding information to one or more Forwarding Agents, the node 1540 returns an RA message to the source neighbor (subject to rate 1541 limiting) then forwards the data packet as usual. The RA message 1542 includes one or more SLLAOs with the link-layer addresses of 1543 candidate Forwarding Engines. 1545 If the forwarding information pertains only to a specific ACP, the 1546 AERO node sets the network-layer source address of the RA to the AERO 1547 address corresponding to the ACP, and sets the default router 1548 lifetime to 0. If the forwarding information pertains to all 1549 addresses, the AERO node instead sets the network-layer source 1550 address of the RA to its own link-local address and sets the default 1551 router lifetime to a non-zero value. 1553 When the source neighbor receives the RA message, it SHOULD record 1554 the link-layer addresses in the SLLAOs as the encapsulation addresses 1555 to use for sending subsequent data packets with addresses that match 1556 the information in the RA. However, the source MUST continue to use 1557 the primary link-layer address of the AERO node as the encapsulation 1558 address for sending control messages. 1560 3.16. AERO Intradomain Route Optimization 1562 When a source Client forwards packets to a prospective correspondent 1563 Client within the same AERO link domain (i.e., one for which the 1564 packet's destination address is covered by an ASP), the source Client 1565 initiates an intra-domain AERO route optimization procedure. The 1566 procedure is based on an exchange of IPv6 ND messages using a chain 1567 of AERO Servers and Relays as a trust basis. This procedure is in 1568 contrast to the Return Routability procedure required for route 1569 optimization to a correspondent Client located in the Internet as 1570 described in Section 3.25. The following sections specify the AERO 1571 intradomain route optimization procedure. 1573 3.16.1. Reference Operational Scenario 1575 Figure 8 depicts the AERO intradomain route optimization reference 1576 operational scenario, using IPv6 addressing as the example (while not 1577 shown, a corresponding example for IPv4 addressing can be easily 1578 constructed). The figure shows an AERO Relay ('R1'), two AERO 1579 Servers ('S1', 'S2'), two AERO Clients ('C1', 'C2') and two ordinary 1580 IPv6 hosts ('H1', 'H2'): 1582 +--------------+ +--------------+ +--------------+ 1583 | Server S1 | | Relay R1 | | Server S2 | 1584 +--------------+ +--------------+ +--------------+ 1585 fe80::2 fe80::1 fe80::3 1586 L2(S1) L2(R1) L2(S2) 1587 | | | 1588 X-----+-----+------------------+-----------------+----+----X 1589 | AERO Link | 1590 L2(A) L2(B) 1591 fe80::2001:db8:0:0 fe80::2001:db8:1:0 1592 +--------------+ +--------------+ 1593 |AERO Client C1| |AERO Client C2| 1594 +--------------+ +--------------+ 1595 2001:DB8:0::/48 2001:DB8:1::/48 1596 | | 1597 .-. .-. 1598 ,-( _)-. 2001:db8:0::1 2001:db8:1::1 ,-( _)-. 1599 .-(_ IP )-. +---------+ +---------+ .-(_ IP )-. 1600 (__ EUN )--| Host H1 | | Host H2 |--(__ EUN ) 1601 `-(______)-' +---------+ +---------+ `-(______)-' 1603 Figure 8: AERO Reference Operational Scenario 1605 In Figure 8, Relay ('R1') assigns the address fe80::1 to its AERO 1606 interface with link-layer address L2(R1), Server ('S1') assigns the 1607 address fe80::2 with link-layer address L2(S1),and Server ('S2') 1608 assigns the address fe80::3 with link-layer address L2(S2). Servers 1609 ('S1') and ('S2') next arrange to add their link-layer addresses to a 1610 published list of valid Servers for the AERO link. 1612 AERO Client ('C1') receives the ACP 2001:db8:0::/48 in a DHCPv6 PD 1613 exchange via AERO Server ('S1') then assigns the address 1614 fe80::2001:db8:0:0 to its AERO interface with link-layer address 1615 L2(C1). Client ('C1') configures a default route and neighbor cache 1616 entry via the AERO interface with next-hop address fe80::2 and link- 1617 layer address L2(S1), then sub-delegates the ACP to its attached 1618 EUNs. IPv6 host ('H1') connects to the EUN, and configures the 1619 address 2001:db8:0::1. 1621 AERO Client ('C2') receives the ACP 2001:db8:1::/48 in a DHCPv6 PD 1622 exchange via AERO Server ('S2') then assigns the address 1623 fe80::2001:db8:1:0 to its AERO interface with link-layer address 1624 L2(C2). Client ('C2') configures a default route and neighbor cache 1625 entry via the AERO interface with next-hop address fe80::3 and link- 1626 layer address L2(S2), then sub-delegates the ACP to its attached 1627 EUNs. IPv6 host ('H1') connects to the EUN, and configures the 1628 address 2001:db8:1::1. 1630 3.16.2. Concept of Operations 1632 Again, with reference to Figure 8, when source host ('H1') sends a 1633 packet to destination host ('H2'), the packet is first forwarded over 1634 the source host's attached EUN to Client ('C1'). Client ('C1') then 1635 forwards the packet via its AERO interface to Server ('S1') and also 1636 sends a Predirect message toward Client ('C2') via Server ('S1'). 1637 Server ('S1') then re-encapsulates and forwards both the packet and 1638 the Predirect message out the same AERO interface toward Client 1639 ('C2') via Relay ('R1'). 1641 When Relay ('R1') receives the packet and Predirect message, it 1642 consults its forwarding table to discover Server ('S2') as the next 1643 hop toward Client ('C2'). Relay ('R1') then forwards both the packet 1644 and the Predirect message to Server ('S2'), which then forwards them 1645 to Client ('C2'). 1647 After Client ('C2') receives the Predirect message, it process the 1648 message and returns a Redirect message toward Client ('C1') via 1649 Server ('S2'). During the process, Client ('C2') also creates or 1650 updates a dynamic neighbor cache entry for Client ('C1'). 1652 When Server ('S2') receives the Redirect message, it re-encapsulates 1653 the message and forwards it on to Relay ('R1'), which forwards the 1654 message on to Server ('S1') which forwards the message on to Client 1655 ('C1'). After Client ('C1') receives the Redirect message, it 1656 processes the message and creates or updates a dynamic neighbor cache 1657 entry for Client ('C2'). 1659 Following the above Predirect/Redirect message exchange, forwarding 1660 of packets from Client ('C1') to Client ('C2') without involving any 1661 intermediate nodes is enabled. The mechanisms that support this 1662 exchange are specified in the following sections. 1664 3.16.3. Message Format 1666 AERO Redirect/Predirect messages use the same format as for ICMPv6 1667 Redirect messages depicted in Section 4.5 of [RFC4861], but also 1668 include a new "Prefix Length" field taken from the low-order 8 bits 1669 of the Redirect message Reserved field. For IPv6, valid values for 1670 the Prefix Length field are 0 through 64; for IPv4, valid values are 1671 0 through 32. The Redirect/Predirect messages are formatted as shown 1672 in Figure 9: 1674 0 1 2 3 1675 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 1676 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1677 | Type (=137) | Code (=0/1) | Checksum | 1678 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1679 | Reserved | Prefix Length | 1680 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1681 | | 1682 + + 1683 | | 1684 + Target Address + 1685 | | 1686 + + 1687 | | 1688 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1689 | | 1690 + + 1691 | | 1692 + Destination Address + 1693 | | 1694 + + 1695 | | 1696 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1697 | Options ... 1698 +-+-+-+-+-+-+-+-+-+-+-+- 1700 Figure 9: AERO Redirect/Predirect Message Format 1702 3.16.4. Sending Predirects 1704 When a Client forwards a packet with a source address from one of its 1705 ACPs toward a destination address covered by an ASP (i.e., toward 1706 another AERO Client connected to the same AERO link), the source 1707 Client MAY send a Predirect message forward toward the destination 1708 Client via the Server. 1710 In the reference operational scenario, when Client ('C1') forwards a 1711 packet toward Client ('C2'), it MAY also send a Predirect message 1712 forward toward Client ('C2'), subject to rate limiting (see 1713 Section 8.2 of [RFC4861]). Client ('C1') prepares the Predirect 1714 message as follows: 1716 o the link-layer source address is set to 'L2(C1)' (i.e., the link- 1717 layer address of Client ('C1')). 1719 o the link-layer destination address is set to 'L2(S1)' (i.e., the 1720 link-layer address of Server ('S1')). 1722 o the network-layer source address is set to fe80::2001:db8:0:0 1723 (i.e., the AERO address of Client ('C1')). 1725 o the network-layer destination address is set to fe80::2001:db8:1:0 1726 (i.e., the AERO address of Client ('C2')). 1728 o the Type is set to 137. 1730 o the Code is set to 1 to indicate "Predirect". 1732 o the Prefix Length is set to the length of the prefix to be 1733 assigned to the Target Address. 1735 o the Target Address is set to fe80::2001:db8:0:0 (i.e., the AERO 1736 address of Client ('C1')). 1738 o the Destination Address is set to the source address of the 1739 originating packet that triggered the Predirection event. (If the 1740 originating packet is an IPv4 packet, the address is constructed 1741 in IPv4-compatible IPv6 address format). 1743 o the message includes one or more TLLAOs with Link ID and 1744 Preference set to appropriate values for Client ('C1')'s 1745 underlying interfaces, and with UDP Port Number and IP Address set 1746 to 0'. 1748 o the message SHOULD include a Timestamp option and a Nonce option. 1750 o the message includes a Redirected Header Option (RHO) that 1751 contains the originating packet truncated if necessary to ensure 1752 that at least the network-layer header is included but the size of 1753 the message does not exceed 1280 bytes. 1755 Note that the act of sending Predirect messages is cited as "MAY", 1756 since Client ('C1') may have advanced knowledge that the direct path 1757 to Client ('C2') would be unusable or otherwise undesirable. If the 1758 direct path later becomes unusable after the initial route 1759 optimization, Client ('C1') simply allows packets to again flow 1760 through Server ('S1'). 1762 3.16.5. Re-encapsulating and Relaying Predirects 1764 When Server ('S1') receives a Predirect message from Client ('C1'), 1765 it first verifies that the TLLAOs in the Predirect are a proper 1766 subset of the Link IDs in Client ('C1')'s neighbor cache entry. If 1767 the Client's TLLAOs are not acceptable, Server ('S1') discards the 1768 message. Otherwise, Server ('S1') validates the message according to 1769 the ICMPv6 Redirect message validation rules in Section 8.1 of 1770 [RFC4861], except that the Predirect has Code=1. Server ('S1') also 1771 verifies that Client ('C1') is authorized to use the Prefix Length in 1772 the Predirect when applied to the AERO address in the network-layer 1773 source address by searching for the AERO address in the neighbor 1774 cache. If validation fails, Server ('S1') discards the Predirect; 1775 otherwise, it copies the correct UDP Port numbers and IP Addresses 1776 for Client ('C1')'s links into the (previously empty) TLLAOs. 1778 Server ('S1') then examines the network-layer destination address of 1779 the Predirect to determine the next hop toward Client ('C2') by 1780 searching for the AERO address in the neighbor cache. Since Client 1781 ('C2') is not one of its neighbors, Server ('S1') re-encapsulates the 1782 Predirect and relays it via Relay ('R1') by changing the link-layer 1783 source address of the message to 'L2(S1)' and changing the link-layer 1784 destination address to 'L2(R1)'. Server ('S1') finally forwards the 1785 re-encapsulated message to Relay ('R1') without decrementing the 1786 network-layer TTL/Hop Limit field. 1788 When Relay ('R1') receives the Predirect message from Server ('S1') 1789 it determines that Server ('S2') is the next hop toward Client ('C2') 1790 by consulting its forwarding table. Relay ('R1') then re- 1791 encapsulates the Predirect while changing the link-layer source 1792 address to 'L2(R1)' and changing the link-layer destination address 1793 to 'L2(S2)'. Relay ('R1') then relays the Predirect via Server 1794 ('S2'). 1796 When Server ('S2') receives the Predirect message from Relay ('R1') 1797 it determines that Client ('C2') is a neighbor by consulting its 1798 neighbor cache. Server ('S2') then re-encapsulates the Predirect 1799 while changing the link-layer source address to 'L2(S2)' and changing 1800 the link-layer destination address to 'L2(C2)'. Server ('S2') then 1801 forwards the message to Client ('C2'). 1803 3.16.6. Processing Predirects and Sending Redirects 1805 When Client ('C2') receives the Predirect message, it accepts the 1806 Predirect only if the message has a link-layer source address of one 1807 of its Servers (e.g., L2(S2)). Client ('C2') further accepts the 1808 message only if it is willing to serve as a redirection target. 1809 Next, Client ('C2') validates the message according to the ICMPv6 1810 Redirect message validation rules in Section 8.1 of [RFC4861], except 1811 that it accepts the message even though Code=1 and even though the 1812 network-layer source address is not that of it's current first-hop 1813 router. 1815 In the reference operational scenario, when Client ('C2') receives a 1816 valid Predirect message, it either creates or updates a dynamic 1817 neighbor cache entry that stores the Target Address of the message as 1818 the network-layer address of Client ('C1') , stores the link-layer 1819 addresses found in the TLLAOs as the link-layer addresses of Client 1820 ('C1') and stores the Prefix Length as the length to be applied to 1821 the network-layer address for forwarding purposes. Client ('C2') 1822 then sets AcceptTime for the neighbor cache entry to ACCEPT_TIME. 1824 After processing the message, Client ('C2') prepares a Redirect 1825 message response as follows: 1827 o the link-layer source address is set to 'L2(C2)' (i.e., the link- 1828 layer address of Client ('C2')). 1830 o the link-layer destination address is set to 'L2(S2)' (i.e., the 1831 link-layer address of Server ('S2')). 1833 o the network-layer source address is set to fe80::2001:db8:1:0 1834 (i.e., the AERO address of Client ('C2')). 1836 o the network-layer destination address is set to fe80::2001:db8:0:0 1837 (i.e., the AERO address of Client ('C1')). 1839 o the Type is set to 137. 1841 o the Code is set to 0 to indicate "Redirect". 1843 o the Prefix Length is set to the length of the prefix to be applied 1844 to the Target Address. 1846 o the Target Address is set to fe80::2001:db8:1:0 (i.e., the AERO 1847 address of Client ('C2')). 1849 o the Destination Address is set to the destination address of the 1850 originating packet that triggered the Redirection event. (If the 1851 originating packet is an IPv4 packet, the address is constructed 1852 in IPv4-compatible IPv6 address format). 1854 o the message includes one or more TLLAOs with Link ID and 1855 Preference set to appropriate values for Client ('C2')'s 1856 underlying interfaces, and with UDP Port Number and IP Address set 1857 to '0'. 1859 o the message SHOULD include a Timestamp option and MUST echo the 1860 Nonce option received in the Predirect (i.e., if a Nonce option is 1861 included). 1863 o the message includes as much of the RHO copied from the 1864 corresponding AERO Predirect message as possible such that at 1865 least the network-layer header is included but the size of the 1866 message does not exceed 1280 bytes. 1868 After Client ('C2') prepares the Redirect message, it sends the 1869 message to Server ('S2'). 1871 3.16.7. Re-encapsulating and Relaying Redirects 1873 When Server ('S2') receives a Redirect message from Client ('C2'), it 1874 first verifies that the TLLAOs in the Redirect are a proper subset of 1875 the Link IDs in Client ('C2')'s neighbor cache entry. If the 1876 Client's TLLAOs are not acceptable, Server ('S2') discards the 1877 message. Otherwise, Server ('S2') validates the message according to 1878 the ICMPv6 Redirect message validation rules in Section 8.1 of 1879 [RFC4861]. Server ('S2') also verifies that Client ('C2') is 1880 authorized to use the Prefix Length in the Redirect when applied to 1881 the AERO address in the network-layer source address by searching for 1882 the AERO address in the neighbor cache. If validation fails, Server 1883 ('S2') discards the Predirect; otherwise, it copies the correct UDP 1884 Port numbers and IP Addresses for Client ('C2')'s links into the 1885 (previously empty) TLLAOs. 1887 Server ('S2') then examines the network-layer destination address of 1888 the Predirect to determine the next hop toward Client ('C2') by 1889 searching for the AERO address in the neighbor cache. Since Client 1890 ('C2') is not a neighbor, Server ('S2') re-encapsulates the Predirect 1891 and relays it via Relay ('R1') by changing the link-layer source 1892 address of the message to 'L2(S2)' and changing the link-layer 1893 destination address to 'L2(R1)'. Server ('S2') finally forwards the 1894 re-encapsulated message to Relay ('R1') without decrementing the 1895 network-layer TTL/Hop Limit field. 1897 When Relay ('R1') receives the Predirect message from Server ('S2') 1898 it determines that Server ('S1') is the next hop toward Client ('C1') 1899 by consulting its forwarding table. Relay ('R1') then re- 1900 encapsulates the Predirect while changing the link-layer source 1901 address to 'L2(R1)' and changing the link-layer destination address 1902 to 'L2(S1)'. Relay ('R1') then relays the Predirect via Server 1903 ('S1'). 1905 When Server ('S1') receives the Predirect message from Relay ('R1') 1906 it determines that Client ('C1') is a neighbor by consulting its 1907 neighbor cache. Server ('S1') then re-encapsulates the Predirect 1908 while changing the link-layer source address to 'L2(S1)' and changing 1909 the link-layer destination address to 'L2(C1)'. Server ('S1') then 1910 forwards the message to Client ('C1'). 1912 3.16.8. Processing Redirects 1914 When Client ('C1') receives the Redirect message, it accepts the 1915 message only if it has a link-layer source address of one of its 1916 Servers (e.g., ''L2(S1)'). Next, Client ('C1') validates the message 1917 according to the ICMPv6 Redirect message validation rules in 1918 Section 8.1 of [RFC4861], except that it accepts the message even 1919 though the network-layer source address is not that of it's current 1920 first-hop router. Following validation, Client ('C1') then processes 1921 the message as follows. 1923 In the reference operational scenario, when Client ('C1') receives 1924 the Redirect message, it either creates or updates a dynamic neighbor 1925 cache entry that stores the Target Address of the message as the 1926 network-layer address of Client ('C2'), stores the link-layer 1927 addresses found in the TLLAOs as the link-layer addresses of Client 1928 ('C2') and stores the Prefix Length as the length to be applied to 1929 the network-layer address for forwarding purposes. Client ('C1') 1930 then sets ForwardTime for the neighbor cache entry to FORWARD_TIME. 1932 Now, Client ('C1') has a neighbor cache entry with a valid 1933 ForwardTime value, while Client ('C2') has a neighbor cache entry 1934 with a valid AcceptTime value. Thereafter, Client ('C1') may forward 1935 ordinary network-layer data packets directly to Client ('C2') without 1936 involving any intermediate nodes, and Client ('C2') can verify that 1937 the packets came from an acceptable source. (In order for Client 1938 ('C2') to forward packets to Client ('C1'), a corresponding 1939 Predirect/Redirect message exchange is required in the reverse 1940 direction; hence, the mechanism is asymmetric.) 1942 3.16.9. Server-Oriented Redirection 1944 In some environments, the Server nearest the target Client may need 1945 to serve as the redirection target, e.g., if direct Client-to-Client 1946 communications are not possible. In that case, the Server prepares 1947 the Redirect message the same as if it were the destination Client 1948 (see: Section 3.16.6), except that it writes its own link-layer 1949 address in the TLLAO option. The Server must then maintain a dynamic 1950 neighbor cache entry for the redirected source Client. 1952 3.17. Neighbor Unreachability Detection (NUD) 1954 AERO nodes perform Neighbor Unreachability Detection (NUD) by sending 1955 unicast NS messages to elicit solicited NA messages from neighbors 1956 the same as described in [RFC4861]. NUD is performed either 1957 reactively in response to persistent L2 errors (see Section 3.13) or 1958 proactively to refresh existing neighbor cache entries. 1960 When an AERO node sends an NS/NA message, it MUST use its link-local 1961 address as the IPv6 source address and the link-local address of the 1962 neighbor as the IPv6 destination address. When an AERO node receives 1963 an NS message or a solicited NA message, it accepts the message if it 1964 has a neighbor cache entry for the neighbor; otherwise, it ignores 1965 the message. 1967 When a source Client is redirected to a target Client it SHOULD 1968 proactively test the direct path by sending an initial NS message to 1969 elicit a solicited NA response. While testing the path, the source 1970 Client can optionally continue sending packets via the Server, 1971 maintain a small queue of packets until target reachability is 1972 confirmed, or (optimistically) allow packets to flow directly to the 1973 target. The source Client SHOULD thereafter continue to proactively 1974 test the direct path to the target Client (see Section 7.3 of 1975 [RFC4861]) periodically in order to keep dynamic neighbor cache 1976 entries alive. 1978 In particular, while the source Client is actively sending packets to 1979 the target Client it SHOULD also send NS messages separated by 1980 RETRANS_TIMER milliseconds in order to receive solicited NA messages. 1981 If the source Client is unable to elicit a solicited NA response from 1982 the target Client after MAX_RETRY attempts, it SHOULD set ForwardTime 1983 to 0 and resume sending packets via one of its Servers. Otherwise, 1984 the source Client considers the path usable and SHOULD thereafter 1985 process any link-layer errors as a hint that the direct path to the 1986 target Client has either failed or has become intermittent. 1988 When a target Client receives an NS message from a source Client, it 1989 resets AcceptTime to ACCEPT_TIME if a neighbor cache entry exists; 1990 otherwise, it discards the NS message. If ForwardTime is non-zero, 1991 the target Client then sends a solicited NA message to the link-layer 1992 address of the source Client; otherwise, it sends the solicited NA 1993 message to the link-layer address of one of its Servers. 1995 When a source Client receives a solicited NA message from a target 1996 Client, it resets ForwardTime to FORWARD_TIME if a neighbor cache 1997 entry exists; otherwise, it discards the NA message. 1999 When ForwardTime for a dynamic neighbor cache entry expires, the 2000 source Client resumes sending any subsequent packets via a Server and 2001 may (eventually) attempt to re-initiate the AERO redirection process. 2002 When AcceptTime for a dynamic neighbor cache entry expires, the 2003 target Client discards any subsequent packets received directly from 2004 the source Client. When both ForwardTime and AcceptTime for a 2005 dynamic neighbor cache entry expire, the Client deletes the neighbor 2006 cache entry. 2008 3.18. Mobility Management 2010 3.18.1. Announcing Link-Layer Address Changes 2012 When a Client needs to change its link-layer address, e.g., due to a 2013 mobility event, it performs an immediate DHCPv6 Rebind/Reply exchange 2014 via each of its Servers using the new link-layer address as the 2015 source and with a CLLAO that includes the correct Link ID and 2016 Preference values. If authentication succeeds, the Server then 2017 update its neighbor cache and sends a DHCPv6 Reply. Note that if the 2018 Client does not issue a DHCPv6 Rebind before the lease lifetime 2019 expires (e.g., if the Client has been out of touch with the Server 2020 for a considerable amount of time), the Server's Reply will report 2021 NoBinding and the Client must re-initiate the DHCPv6 PD procedure. 2023 Next, the Client sends unsolicited NA messages to each of its 2024 correspondent Client neighbors using the same procedures as specified 2025 in Section 7.2.6 of [RFC4861], except that it sends the messages as 2026 unicast to each neighbor via a Server instead of multicast. In this 2027 process, the Client should send no more than 2028 MAX_NEIGHBOR_ADVERTISEMENT messages separated by no less than 2029 RETRANS_TIMER seconds to each neighbor. 2031 With reference to Figure 8, when Client ('C2') needs to change its 2032 link-layer address it sends unicast unsolicited NA messages to Client 2033 ('C1') via Server ('S2') as follows: 2035 o the link-layer source address is set to 'L2(C2)' (i.e., the link- 2036 layer address of Client ('C2')). 2038 o the link-layer destination address is set to 'L2(S2)' (i.e., the 2039 link-layer address of Server ('S2')). 2041 o the network-layer source address is set to fe80::2001:db8:1:0 2042 (i.e., the AERO address of Client ('C2')). 2044 o the network-layer destination address is set to fe80::2001:db8:0:0 2045 (i.e., the AERO address of Client ('C1')). 2047 o the Type is set to 136. 2049 o the Code is set to 0. 2051 o the Solicited flag is set to 0. 2053 o the Override flag is set to 1. 2055 o the Target Address is set to fe80::2001:db8:1:0 (i.e., the AERO 2056 address of Client ('C2')). 2058 o the message includes one or more TLLAOs with Link ID and 2059 Preference set to appropriate values for Client ('C2')'s 2060 underlying interfaces, and with UDP Port Number and IP Address set 2061 to '0'. 2063 o the message SHOULD include a Timestamp option. 2065 When Server ('S1') receives the NA message, it relays the message in 2066 the same way as described for relaying Redirect messages in 2067 Section 3.16.7. In particular, Server ('S1') copies the correct UDP 2068 port numbers and IP addresses into the TLLAOs, changes the link-layer 2069 source address to its own address, changes the link-layer destination 2070 address to the address of Relay ('R1'), then forwards the NA message 2071 via the relaying chain the same as for a Redirect. 2073 When Client ('C1') receives the NA message, it accepts the message 2074 only if it already has a neighbor cache entry for Client ('C2') then 2075 updates the link-layer addresses for Client ('C2') based on the 2076 addresses in the TLLAOs. Next, Client ('C1') SHOULD initiate the NUD 2077 procedures specified in Section 3.17 to provide Client ('C2') with an 2078 indication that the link-layer source address has been updated, and 2079 to refresh ('C2')'s AcceptTime and ('C1')'s ForwardTime timers. 2081 If Client ('C2') receives an NS message from Client ('C1') indicating 2082 that an unsolicited NA has updated its neighbor cache, Client ('C2') 2083 need not send additional unsolicited NAs. If Client ('C2')'s 2084 unsolicited NA messages are somehow lost, however, Client ('C1') will 2085 soon learn of the mobility event via NUD. 2087 3.18.2. Bringing New Links Into Service 2089 When a Client needs to bring a new underlying interface into service 2090 (e.g., when it activates a new data link), it performs an immediate 2091 Rebind/Reply exchange via each of its Servers using the new link- 2092 layer address as the source address and with a CLLAO that includes 2093 the new Link ID and Preference values. If authentication succeeds, 2094 the Server then updates its neighbor cache and sends a DHCPv6 Reply. 2096 The Client MAY then send unsolicited NA messages to each of its 2097 correspondent Clients to inform them of the new link-layer address as 2098 described in Section 3.18.1. 2100 3.18.3. Removing Existing Links from Service 2102 When a Client needs to remove an existing underlying interface from 2103 service (e.g., when it de-activates an existing data link), it 2104 performs an immediate Rebind/Reply exchange via each of its Servers 2105 over any available link with a CLLAO that includes the deprecated 2106 Link ID and a Preference value of 0. If authentication succeeds, the 2107 Server then updates its neighbor cache and sends a DHCPv6 Reply. The 2108 Client SHOULD then send unsolicited NA messages to each of its 2109 correspondent Clients to inform them of the deprecated link-layer 2110 address as described in Section 3.18.1. 2112 3.18.4. Moving to a New Server 2114 When a Client associates with a new Server, it performs the Client 2115 procedures specified in Section 3.14.2. 2117 When a Client disassociates with an existing Server, it sends a 2118 DHCPv6 Release message via a new Server to the unicast link-local 2119 network layer address of the old Server. The new Server then writes 2120 its own link-layer address in the DHCPv6 release message IP source 2121 address and forwards the message to the old Server. 2123 When the old Server receives the DHCPv6 Release, it first 2124 authenticates the message. The Server then resets the Client's 2125 neighbor cache entry lifetime to 5 seconds, rewrites the link-layer 2126 address in the neighbor cache entry to the address of the new Server, 2127 then returns a DHCPv6 Reply message to the Client via the old Server. 2128 When the lifetime expires, the old Server withdraws the IP route from 2129 the AERO routing system and deletes the neighbor cache entry for the 2130 Client. The Client can then use the Reply message to verify that the 2131 termination signal has been processed, and can delete both the 2132 default route and the neighbor cache entry for the old Server. (Note 2133 that since Release/Reply messages may be lost in the network the 2134 Client MUST retry until it gets Reply indicating that the Release was 2135 successful.) 2137 Clients SHOULD NOT move rapidly between Servers in order to avoid 2138 causing excessive oscillations in the AERO routing system. Such 2139 oscillations could result in intermittent reachability for the Client 2140 itself, while causing little harm to the network. Examples of when a 2141 Client might wish to change to a different Server include a Server 2142 that has gone unreachable, topological movements of significant 2143 distance, etc. 2145 3.19. Encapsulation Protocol Version Considerations 2147 A source Client may connect only to an IPvX underlying network, while 2148 the target Client connects only to an IPvY underlying network. In 2149 that case, the target and source Clients have no means for reaching 2150 each other directly (since they connect to underlying networks of 2151 different IP protocol versions) and so must ignore any redirection 2152 messages and continue to send packets via the Server. 2154 3.20. Multicast Considerations 2156 When the underlying network does not support multicast, AERO nodes 2157 map IPv6 link-scoped multicast addresses (including 2158 'All_DHCP_Relay_Agents_and_Servers') to the link-layer address of a 2159 Server. 2161 When the underlying network supports multicast, AERO nodes use the 2162 multicast address mapping specification found in [RFC2529] for IPv4 2163 underlying networks and use a direct multicast mapping for IPv6 2164 underlying networks. (In the latter case, "direct multicast mapping" 2165 means that if the IPv6 multicast destination address of the 2166 encapsulated packet is "M", then the IPv6 multicast destination 2167 address of the encapsulating header is also "M".) 2169 3.21. Operation on AERO Links Without DHCPv6 Services 2171 When Servers on the AERO link do not provide DHCPv6 services, 2172 operation can still be accommodated through administrative 2173 configuration of ACPs on AERO Clients. In that case, administrative 2174 configurations of AERO interface neighbor cache entries on both the 2175 Server and Client are also necessary. However, this may interfere 2176 with the ability for Clients to dynamically change to new Servers, 2177 and can expose the AERO link to misconfigurations unless the 2178 administrative configurations are carefully coordinated. 2180 3.22. Operation on Server-less AERO Links 2182 In some AERO link scenarios, there may be no Servers on the link and/ 2183 or no need for Clients to use a Server as an intermediary trust 2184 anchor. In that case, each Client acts as a Server unto itself to 2185 establish neighbor cache entries by performing direct Client-to- 2186 Client IPv6 ND message exchanges, and some other form of trust basis 2187 must be applied so that each Client can verify that the prospective 2188 neighbor is authorized to use its claimed ACP. 2190 When there is no Server on the link, Clients must arrange to receive 2191 ACPs and publish them via a secure alternate prefix delegation 2192 authority through some means outside the scope of this document. 2194 3.23. Proxy AERO 2196 Proxy Mobile IPv6 (PMIPv6) [RFC5213][RFC5844][RFC5949] presents a 2197 localized mobility management scheme for use within an access network 2198 domain. It is typically used in WiFi and cellular wireless access 2199 networks, and allows Mobile Nodes (MNs) to receive and retain an IP 2200 address that remains stable within the access network domain without 2201 needing to implement any special mobility protocols. In the PMIPv6 2202 architecture, access network devices known as Mobility Access 2203 Gateways (MAGs) provide MNs with an access link abstraction and 2204 receive prefixes for the MNs from a Local Mobility Anchor (LMA). 2206 In a proxy AERO domain, a proxy AERO Client (acting as a MAG) can 2207 similarly provide proxy services for MNs that do not participate in 2208 AERO messaging. The proxy Client presents an access link abstraction 2209 to MNs, and performs DHCPv6 PD exchanges over the AERO interface with 2210 an AERO Server (acting as an LMA) to receive ACPs for address 2211 provisioning of new MNs that come onto an access link. This scheme 2212 assumes that proxy Clients act as fixed (non-mobile) infrastructure 2213 elements under the same administrative trust basis as for Relays and 2214 Servers. 2216 When an MN comes onto an access link within a proxy AERO domain for 2217 the first time, the proxy Client authenticates the MN and obtains a 2218 unique identifier that it can use as a DHCPv6 DUID then issues a 2219 DHCPv6 PD Request to its Server. When the Server delegates an ACP, 2220 the proxy Client creates an AERO address for the MN and assigns the 2221 ACP to the MN's access link. The proxy Client then configures itself 2222 as a default router for the MN and provides address autoconfiguration 2223 services (e.g., SLAAC, DHCPv6, DHCPv4, etc.) for provisioning MN 2224 addresses from the ACP over the access link. Since the proxy Client 2225 may serve many such MNs simultaneously, it may receive multiple ACP 2226 prefix delegations and configure multiple AERO addresses, i.e., one 2227 for each MN. 2229 When two MNs are associated with the same proxy Client, the Client 2230 can forward traffic between the MNs without involving a Server since 2231 it configures the AERO addresses of both MNs and therefore also has 2232 the necessary routing information. When two MNs are associated with 2233 different proxy Clients, the source MN's Client can initiate standard 2234 AERO route optimization to discover a direct path to the target MN's 2235 Client through the exchange of Predirect/Redirect messages. 2237 When an MN in a proxy AERO domain leaves an access link provided by 2238 an old proxy Client, the MN issues an access link-specific "leave" 2239 message that informs the old Client of the link-layer address of a 2240 new Client on the planned new access link. This is known as a 2241 "predictive handover". When an MN comes onto an access link provided 2242 by a new proxy Client, the MN issues an access link-specific "join" 2243 message that informs the new Client of the link-layer address of the 2244 old Client on the actual old access link. This is known as a 2245 "reactive handover". 2247 Upon receiving a predictive handover indication, the old proxy Client 2248 sends a DHCPv6 PD Request message directly to the new Client and 2249 queues any arriving data packets addressed to the departed MN. The 2250 Request message includes the MN's ID as the DUID, the ACP in an IA_PD 2251 option, the AERO address derived from the MN's ACP as the network- 2252 layer source address, 'All_DHCP_Relay_Agents_and_Servers' as the 2253 network-layer destination address, the old Client's address as the 2254 link-layer source address and the new Client's address as the link- 2255 layer destination address. When the new Client receives the Request 2256 message, it changes the link-layer source address to its own address, 2257 changes the link-layer destination address to the address of its 2258 Server, and forwards the message to the Server. At the same time, 2259 the new Client creates access link state for the ACP in anticipation 2260 of the MN's arrival (while queuing any data packets until the MN 2261 arrives), creates a neighbor cache entry for the old Client with 2262 AcceptTime set to ACCEPT_TIME, then sends a Redirect message back to 2263 the old Client. When the old Client receives the Redirect message, 2264 it creates a neighbor cache entry for new Client with ForwardTime set 2265 to FORWARD_TIME, then forwards any queued data packets to the new 2266 Client. At the same time, the old Client sends a DHCPv6 PD Release 2267 message to its Server. Finally, the old Client sends unsolicited NA 2268 messages to any of the ACP's correspondents with a TLLAO containing 2269 the link-layer address of the new Client. This follows the procedure 2270 specified in Section 3.18.1, except that it is the old Client and not 2271 the Server that supplies the link-layer address. 2273 Upon receiving a reactive handover indication, the new proxy Client 2274 creates access link state for the MN's ACP, sends a DHCPv6 PD Request 2275 message to its Server, and sends a DHCPv6 PD Release message directly 2276 to the old Client. The Release message includes the MN's ID as the 2277 DUID, the ACP in an IA_PD option, the AERO address derived from the 2278 MN's ACP as the network-layer source address, 2279 'All_DHCP_Relay_Agents_and_Servers' as the network-layer destination 2280 address, the new Client's address as the link-layer source address 2281 and the old Client's address as the link-layer destination address. 2282 When the old Client receives the Release message, it changes the 2283 link-layer source address to its own address, changes the link-layer 2284 destination address to the address of its Server, and forwards the 2285 message to the Server. At the same time, the old Client sends a 2286 Predirect message back to the new Client and queues any arriving data 2287 packets addressed to the departed MN. When the new Client receives 2288 the Predirect, it creates a neighbor cache entry for the old Client 2289 with AcceptTime set to ACCEPT_TIME, then sends a Redirect message 2290 back to the old Client. When the old Client receives the Redirect 2291 message, it creates a neighbor cache entry for the new Client with 2292 ForwardTime set to FORWARD_TIME, then forwards any queued data 2293 packets to the new Client. Finally, the old Client sends unsolicited 2294 NA messages to correspondents the same as for the predictive case. 2296 When a Server processes a DHCPv6 Request message, it creates a 2297 neighbor cache entry for this ACP if none currently exists. If a 2298 neighbor cache entry already exists, however, the Server changes the 2299 link-layer address to the address of the new proxy Client (this 2300 satisfies the case of both the old Client and new Client using the 2301 same Server). 2303 When a Server processes a DHCPv6 Release message, it resets the 2304 neighbor cache entry lifetime for this ACP to 5 seconds if the cached 2305 link-layer address matches the old proxy Client's address. 2306 Otherwise, the Server ignores the Release message (this satisfies the 2307 case of both the old Client and new Client using the same Server). 2309 When a correspondent Client receives an unsolicited NA message, it 2310 changes the link-layer address for the ACP's neighbor cache entry to 2311 the address of the new proxy Client. The correspondent Client then 2312 issues a Predirect/Redirect exchange to establish a new neighbor 2313 cache entry in the new Client. 2315 From an architectural perspective, in addition to the use of DHCPv6 2316 PD and IPv6 ND signaling the AERO approach differs from PMIPv6 in its 2317 use of the NBMA virtual link model instead of point-to-point tunnels. 2318 This provides a more agile interface for Client/Server and Client/ 2319 Client coordinations, and also facilitates simple route optimization. 2320 The AERO routing system is also arranged in such a fashion that 2321 Clients get the same service from any Server they happen to associate 2322 with. This provides a natural fault tolerance and load balancing 2323 capability such as desired for distributed mobility management. 2325 3.24. Extending AERO Links Through Security Gateways 2327 When an enterprise mobile device moves from a campus LAN connection 2328 to a public Internet link, it must re-enter the enterprise via a 2329 security gateway that has both a physical interface connection to the 2330 Internet and a physical interface connection to the enterprise 2331 internetwork. This most often entails the establishment of a Virtual 2332 Private Network (VPN) link over the public Internet from the mobile 2333 device to the security gateway. During this process, the mobile 2334 device supplies the security gateway with its public Internet address 2335 as the link-layer address for the VPN. The mobile device then acts 2336 as an AERO Client to negotiate with the security gateway to obtain 2337 its ACP. 2339 In order to satisfy this need, the security gateway also operates as 2340 an AERO Server with support for AERO Client proxying. In particular, 2341 when a mobile device (i.e., the Client) connects via the security 2342 gateway (i.e., the Server), the Server provides the Client with an 2343 ACP in a DHCPv6 PD exchange the same as if it were attached to an 2344 enterprise campus access link. The Server then replaces the Client's 2345 link-layer source address with the Server's enterprise-facing link- 2346 layer address in all AERO messages the Client sends toward neighbors 2347 on the AERO link. The AERO messages are then delivered to other 2348 devices on the AERO link as if they were originated by the security 2349 gateway instead of by the AERO Client. In the reverse direction, the 2350 AERO messages sourced by devices within the enterprise network can be 2351 forwarded to the security gateway, which then replaces the link-layer 2352 destination address with the Client's link-layer address and replaces 2353 the link-layer source address with its own (Internet-facing) link- 2354 layer address. 2356 After receiving the ACP, the Client can send IP packets that use an 2357 address taken from the ACP as the network layer source address, the 2358 Client's link-layer address as the link-layer source address, and the 2359 Server's Internet-facing link-layer address as the link-layer 2360 destination address. The Server will then rewrite the link-layer 2361 source address with the Server's own enterprise-facing link-layer 2362 address and rewrite the link-layer destination address with the 2363 target AERO node's link-layer address, and the packets will enter the 2364 enterprise network as though they were sourced from a device located 2365 within the enterprise. In the reverse direction, when a packet 2366 sourced by a node within the enterprise network uses a destination 2367 address from the Client's ACP, the packet will be delivered to the 2368 security gateway which then rewrites the link-layer destination 2369 address to the Client's link-layer address and rewrites the link- 2370 layer source address to the Server's Internet-facing link-layer 2371 address. The Server then delivers the packet across the VPN to the 2372 AERO Client. In this way, the AERO virtual link is essentially 2373 extended *through* the security gateway to the point at which the VPN 2374 link and AERO link are effectively grafted together by the link-layer 2375 address rewriting performed by the security gateway. All AERO 2376 messaging services (including route optimization and mobility 2377 signaling) are therefore extended to the Client. 2379 In order to support this virtual link grafting, the security gateway 2380 (acting as an AERO Server) must keep static neighbor cache entries 2381 for all of its associated Clients located on the public Internet. 2382 The neighbor cache entry is keyed by the AERO Client's AERO address 2383 the same as if the Client were located within the enterprise 2384 internetwork. The neighbor cache is then managed in all ways as 2385 though the Client were an ordinary AERO Client. This includes the 2386 AERO IPv6 ND messaging signaling for Route Optimization and Neighbor 2387 Unreachability Detection. 2389 Note that the main difference between a security gateway acting as an 2390 AERO Server and an enterprise-internal AERO Server is that the 2391 security gateway has at least one enterprise-internal physical 2392 interface and at least one public Internet physical interface. 2393 Conversely, the enterprise-internal AERO Server has only enterprise- 2394 internal physical interfaces. For this reason security gateway 2395 proxying is needed to ensure that the public Internet link-layer 2396 addressing space is kept separate from the enterprise-internal link- 2397 layer addressing space. This is afforded through a natural extension 2398 of the security association caching already performed for each VPN 2399 client by the security gateway. 2401 3.25. Extending IPv6 AERO Links to the Internet 2403 When an IPv6 host ('H1') with an address from an ACP owned by AERO 2404 Client ('C1') sends packets to a correspondent IPv6 host ('H2'), the 2405 packets eventually arrive at the IPv6 router that owns ('H2')s 2406 prefix. This IPv6 router may or may not be an AERO Client ('C2') 2407 either within the same home network as ('C1') or in a different home 2408 network. 2410 If Client ('C1') is currently located outside the boundaries of its 2411 home network, it will connect back into the home network via a 2412 security gateway acting as an AERO Server. The packets sent by 2413 ('H1') via ('C1') will then be forwarded through the security gateway 2414 then through the home network and finally to ('C2') where they will 2415 be delivered to ('H2'). This could lead to sub-optimal performance 2416 when ('C2') could instead be reached via a more direct route without 2417 involving the security gateway. 2419 Consider the case when host ('H1') has the IPv6 address 2420 2001:db8:1::1, and Client ('C1') has the ACP 2001:db8:1::/64 with 2421 underlying IPv6 Internet address of 2001:db8:1000::1. Also, host 2422 ('H2') has the IPv6 address 2001:db8:2::1, and Client ('C2') has the 2423 ACP 2001:db8:2::/64 with underlying IPv6 address of 2001:db8:2000::1. 2424 Client ('C1') can determine whether 'C2' is indeed also an AERO 2425 Client willing to serve as a route optimization correspondent by 2426 resolving the AAAA records for the DNS FQDN that matches ('H2')s 2427 prefix, i.e.: 2429 '0.0.0.0.2.0.0.0.8.b.d.0.1.0.0.2.aero.linkupnetworks.net' 2431 If ('C2') is indeed a candidate correspondent, the FQDN lookup will 2432 return a PTR resource record that contains the domain name for the 2433 AERO link that manages ('C2')s ASP. Client ('C1') can then attempt 2434 route optimization using an approach similar to the Return 2435 Routability procedure specified for Mobile IPv6 (MIPv6) [RFC6275]. 2436 In order to support this process, both Clients MUST intercept and 2437 decapsulate packets that have a subnet router anycast address 2438 corresponding to any of the /64 prefixes covered by their respective 2439 ACPs. 2441 To initiate the process, Client ('C1') creates a specially-crafted 2442 encapsulated AERO Predirect message that will be routed through its 2443 home network then through ('C2')s home network and finally to ('C2') 2444 itself. Client ('C1') prepares the initial message in the exchange 2445 as follows: 2447 o The encapsulating IPv6 header source address is set to 2448 2001:db8:1:: (i.e., the IPv6 subnet router anycast address for 2449 ('C1')s ACP) 2451 o The encapsulating IPv6 header destination address is set to 2452 2001:db8:2:: (i.e., the IPv6 subnet router anycast address for 2453 ('C2')s ACP) 2455 o The encapsulating IPv6 header is followed by a UDP header with 2456 source and destination port set to 8060 2458 o The encapsulated IPv6 header source address is set to 2459 fe80::2001:db8:1:0 (i.e., the AERO address for ('C1')) 2461 o The encapsulated IPv6 header destination address is set to 2462 fe80::2001:db8:2:0 (i.e., the AERO address for ('C2')) 2464 o The encapsulated AERO Predirect message includes all of the 2465 securing information that would occur in a MIPv6 "Home Test Init" 2466 message (format TBD) 2468 Client ('C1') then further encapsulates the message in the 2469 encapsulating headers necessary to convey the packet to the security 2470 gateway (e.g., through IPsec encapsulation) so that the message now 2471 appears "double-encapsulated". ('C1') then sends the message to the 2472 security gateway, which re-encapsulates and forwards it over the home 2473 network from where it will eventually reach ('C2'). 2475 At the same time, ('C1') creates and sends a second encapsulated AERO 2476 Predirect message that will be routed through the IPv6 Internet 2477 without involving the security gateway. Client ('C1') prepares the 2478 message as follows: 2480 o The encapsulating IPv6 header source address is set to 2481 2001:db8:1000:1 (i.e., the Internet IPv6 address of ('C1')) 2483 o The encapsulating IPv6 header destination address is set to 2484 2001:db8:2:: (i.e., the IPv6 subnet router anycast address for 2485 ('C2')s ACP) 2487 o The encapsulating IPv6 header is followed by a UDP header with 2488 source and destination port set to 8060 2490 o The encapsulated IPv6 header source address is set to 2491 fe80::2001:db8:1:0 (i.e., the AERO address for ('C1')) 2493 o The encapsulated IPv6 header destination address is set to 2494 fe80::2001:db8:2:0 (i.e., the AERO address for ('C2')) 2496 o The encapsulated AERO Predirect message includes all of the 2497 securing information that would occur in a MIPv6 "Care-of Test 2498 Init" message (format TBD) 2500 ('C2') will receive both Predirect messages through its home network 2501 then return a corresponding Redirect for each of the Predirect 2502 messages with the source and destination addresses in the inner and 2503 outer headers reversed. The first message includes all of the 2504 securing information that would occur in a MIPv6 "Home Test" message, 2505 while the second message includes all of the securing information 2506 that would occur in a MIPv6 "Care-of Test" message (formats TBD). 2508 When ('C1') receives the Redirect messages, it performs the necessary 2509 security procedures per the MIPv6 specification. It then prepares an 2510 encapsulated NS message that includes the same source and destination 2511 addresses as for the "Care-of Test Init" Predirect message, and 2512 includes all of the securing information that would occur in a MIPv6 2513 "Binding Update" message (format TBD) and sends the message to 2514 ('C2'). 2516 When ('C2') receives the NS message, if the securing information is 2517 correct it creates or updates a neighbor cache entry for ('C1') with 2518 fe80::2001:db8:1:0 as the network-layer address, 2001:db8:1000::1 as 2519 the link-layer address and with AcceptTime set to ACCEPT_TIME. 2520 ('C2') then sends an encapsulated NA message back to ('C1') that 2521 includes the same source and destination addresses as for the "Care- 2522 of Test" Redirect message, and includes all of the securing 2523 information that would occur in a MIPv6 "Binding Acknowledgement" 2524 message (format TBD) and sends the message to ('C1'). 2526 When ('C1') receives the NA message, it creates or updates a neighbor 2527 cache entry for ('C2') with fe80::2001:db8:2:0 as the network-layer 2528 address and 2001:db8:2:: as the link-layer address and with 2529 ForwardTime set to FORWARD_TIME, thus completing the route 2530 optimization in the forward direction. 2532 ('C1') subsequently forwards encapsulated packets with outer source 2533 address 2001:db8:1000::1, with outer destination address 2534 2001:db8:2::, with inner source address taken from the 2001:db8:1::, 2535 and with inner destination address taken from 2001:db8:2:: due to the 2536 fact that it has a securely-established neighbor cache entry with 2537 non-zero ForwardTime. ('C2') subsequently accepts any such 2538 encapsulated packets due to the fact that it has a securely- 2539 established neighbor cache entry with non-zero AcceptTime. 2541 In order to keep neighbor cache entries alive, ('C1') periodically 2542 sends additional NS messages to ('C2') and receives any NA responses. 2543 If ('C1') moves to a different point of attachment after the initial 2544 route optimization, it sends a new secured NS message to ('C2') as 2545 above to update ('C2')s neighbor cache. 2547 If ('C2') has packets to send to ('C1'), it performs a corresponding 2548 route optimization in the opposite direction following the same 2549 procedures described above. In the process, the already-established 2550 unidirectional neighbor cache entries within ('C1') and ('C2') are 2551 updated to include the now-bidirectional information. In particular, 2552 the AcceptTime and ForwardTime variables for both neighbor cache 2553 entries are updated to non-zero values, and the link-layer address 2554 for ('C1')s neighbor cache entry for ('C2') is reset to 2555 2001:db8:2000::1. 2557 Note that two AERO Clients can use full security protocol messaging 2558 instead of Return Routability, e.g., if strong authentication and/or 2559 confidentiality are desired. In that case, security protocol key 2560 exchanges such as specified for MOBIKE [RFC4555] would be used to 2561 establish security associations and neighbor cache entries between 2562 the AERO clients. Thereafter, AERO NS/NA messaging can be used to 2563 maintain neighbor cache entries, test reachability, and to announce 2564 mobility events. If reachability testing fails, e.g., if both 2565 Clients move at roughly the same time, the Clients can tear down the 2566 security association and neighbor cache entries and again allow 2567 packets to flow through their home network. 2569 4. Implementation Status 2571 An application-layer implementation is in progress. 2573 5. IANA Considerations 2575 IANA is instructed to assign a new 2-octet Hardware Type number 2576 "TBD1" for AERO in the "arp-parameters" registry per Section 2 of 2577 [RFC5494]. The number is assigned from the 2-octet Unassigned range 2578 with Hardware Type "AERO" and with this document as the reference. 2580 IANA is instructed to assign a 4-octet Enterprise Number "TBD2" for 2581 AERO in the "enterprise-numbers" registry per [RFC3315]. 2583 6. Security Considerations 2585 AERO link security considerations are the same as for standard IPv6 2586 Neighbor Discovery [RFC4861] except that AERO improves on some 2587 aspects. In particular, AERO uses a trust basis between Clients and 2588 Servers, where the Clients only engage in the AERO mechanism when it 2589 is facilitated by a trust anchor. Unless there is some other means 2590 of authenticating the Client's identity (e.g., link-layer security), 2591 AERO nodes SHOULD also use DHCPv6 securing services (e.g., DHCPv6 2592 authentication, Secure DHCPv6 [I-D.ietf-dhc-sedhcpv6], etc.) for 2593 Client authentication and network admission control. 2595 AERO Redirect, Predirect and unsolicited NA messages SHOULD include a 2596 Timestamp option (see Section 5.3 of [RFC3971]) that other AERO nodes 2597 can use to verify the message time of origin. AERO Predirect, NS and 2598 RS messages SHOULD include a Nonce option (see Section 5.3 of 2599 [RFC3971]) that recipients echo back in corresponding responses. 2601 AERO links must be protected against link-layer address spoofing 2602 attacks in which an attacker on the link pretends to be a trusted 2603 neighbor. Links that provide link-layer securing mechanisms (e.g., 2604 IEEE 802.1X WLANs) and links that provide physical security (e.g., 2605 enterprise network wired LANs) provide a first line of defense that 2606 is often sufficient. In other instances, additional securing 2607 mechanisms such as Secure Neighbor Discovery (SeND) [RFC3971], IPsec 2608 [RFC4301] or TLS [RFC5246] may be necessary. 2610 AERO Clients MUST ensure that their connectivity is not used by 2611 unauthorized nodes on their EUNs to gain access to a protected 2612 network, i.e., AERO Clients that act as routers MUST NOT provide 2613 routing services for unauthorized nodes. (This concern is no 2614 different than for ordinary hosts that receive an IP address 2615 delegation but then "share" the address with unauthorized nodes via a 2616 NAT function.) 2618 On some AERO links, establishment and maintenance of a direct path 2619 between neighbors requires secured coordination such as through the 2620 Internet Key Exchange (IKEv2) protocol [RFC5996] to establish a 2621 security association. 2623 7. Acknowledgements 2625 Discussions both on IETF lists and in private exchanges helped shape 2626 some of the concepts in this work. Individuals who contributed 2627 insights include Mikael Abrahamsson, Mark Andrews, Fred Baker, 2628 Stewart Bryant, Brian Carpenter, Wojciech Dec, Ralph Droms, Sri 2629 Gundavelli, Brian Haberman, Joel Halpern, Sascha Hlusiak, Lee Howard, 2630 Andre Kostur, Ted Lemon, Joe Touch and Bernie Volz. Members of the 2631 IESG also provided valuable input during their review process that 2632 greatly improved the document. Special thanks go to Stewart Bryant, 2633 Joel Halpern and Brian Haberman for their shepherding guidance. 2635 This work has further been encouraged and supported by Boeing 2636 colleagues including Keith Bartley, Dave Bernhardt, Cam Brodie, 2637 Balaguruna Chidambaram, Claudiu Danilov, Wen Fang, Anthony Gregory, 2638 Jeff Holland, Ed King, Gen MacLean, Kent Shuey, Brian Skeen, Mike 2639 Slane, Julie Wulff, Yueli Yang, and other members of the BR&T and BIT 2640 mobile networking teams. 2642 Earlier works on NBMA tunneling approaches are found in 2643 [RFC2529][RFC5214][RFC5569]. 2645 Many of the constructs presented in this second edition of AERO are 2646 based on the author's earlier works, including: 2648 o The Internet Routing Overlay Network (IRON) 2649 [RFC6179][I-D.templin-ironbis] 2651 o Virtual Enterprise Traversal (VET) 2652 [RFC5558][I-D.templin-intarea-vet] 2654 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 2655 [RFC5320][I-D.templin-intarea-seal] 2657 o AERO, First Edition [RFC6706] 2659 Finally, Satoru Matsushima and Ryuji Wakikawa also discuss BGP in 2660 their work. Behcet Sarikaya made comments on the dmm list. 2662 8. References 2664 8.1. Normative References 2666 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 2667 August 1980. 2669 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, September 2670 1981. 2672 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 2673 RFC 792, September 1981. 2675 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 2676 October 1996. 2678 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2679 Requirement Levels", BCP 14, RFC 2119, March 1997. 2681 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 2682 (IPv6) Specification", RFC 2460, December 1998. 2684 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 2685 IPv6 Specification", RFC 2473, December 1998. 2687 [RFC3315] Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C., 2688 and M. Carney, "Dynamic Host Configuration Protocol for 2689 IPv6 (DHCPv6)", RFC 3315, July 2003. 2691 [RFC3633] Troan, O. and R. Droms, "IPv6 Prefix Options for Dynamic 2692 Host Configuration Protocol (DHCP) version 6", RFC 3633, 2693 December 2003. 2695 [RFC3971] Arkko, J., Kempf, J., Zill, B., and P. Nikander, "SEcure 2696 Neighbor Discovery (SEND)", RFC 3971, March 2005. 2698 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 2699 for IPv6 Hosts and Routers", RFC 4213, October 2005. 2701 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 2702 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 2703 September 2007. 2705 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 2706 Address Autoconfiguration", RFC 4862, September 2007. 2708 [RFC6434] Jankiewicz, E., Loughney, J., and T. Narten, "IPv6 Node 2709 Requirements", RFC 6434, December 2011. 2711 8.2. Informative References 2713 [I-D.ietf-dhc-sedhcpv6] 2714 Jiang, S., Shen, S., Zhang, D., and T. Jinmei, "Secure 2715 DHCPv6", draft-ietf-dhc-sedhcpv6-04 (work in progress), 2716 September 2014. 2718 [I-D.templin-intarea-seal] 2719 Templin, F., "The Subnetwork Encapsulation and Adaptation 2720 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 2721 progress), January 2014. 2723 [I-D.templin-intarea-vet] 2724 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 2725 templin-intarea-vet-40 (work in progress), May 2013. 2727 [I-D.templin-ironbis] 2728 Templin, F., "The Interior Routing Overlay Network 2729 (IRON)", draft-templin-ironbis-16 (work in progress), 2730 March 2014. 2732 [I-D.vandevelde-idr-remote-next-hop] 2733 Velde, G., Patel, K., Rao, D., Raszuk, R., and R. Bush, 2734 "BGP Remote-Next-Hop", draft-vandevelde-idr-remote-next- 2735 hop-08 (work in progress), October 2014. 2737 [RFC0879] Postel, J., "TCP maximum segment size and related topics", 2738 RFC 879, November 1983. 2740 [RFC1035] Mockapetris, P., "Domain names - implementation and 2741 specification", STD 13, RFC 1035, November 1987. 2743 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 2744 November 1990. 2746 [RFC1812] Baker, F., "Requirements for IP Version 4 Routers", RFC 2747 1812, June 1995. 2749 [RFC1930] Hawkinson, J. and T. Bates, "Guidelines for creation, 2750 selection, and registration of an Autonomous System (AS)", 2751 BCP 6, RFC 1930, March 1996. 2753 [RFC1981] McCann, J., Deering, S., and J. Mogul, "Path MTU Discovery 2754 for IP version 6", RFC 1981, August 1996. 2756 [RFC2131] Droms, R., "Dynamic Host Configuration Protocol", RFC 2757 2131, March 1997. 2759 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 2760 Domains without Explicit Tunnels", RFC 2529, March 1999. 2762 [RFC2675] Borman, D., Deering, S., and R. Hinden, "IPv6 Jumbograms", 2763 RFC 2675, August 1999. 2765 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 2766 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 2767 March 2000. 2769 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", RFC 2770 2923, September 2000. 2772 [RFC3596] Thomson, S., Huitema, C., Ksinant, V., and M. Souissi, 2773 "DNS Extensions to Support IP Version 6", RFC 3596, 2774 October 2003. 2776 [RFC3819] Karn, P., Bormann, C., Fairhurst, G., Grossman, D., 2777 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 2778 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 2779 RFC 3819, July 2004. 2781 [RFC4271] Rekhter, Y., Li, T., and S. Hares, "A Border Gateway 2782 Protocol 4 (BGP-4)", RFC 4271, January 2006. 2784 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 2785 Architecture", RFC 4291, February 2006. 2787 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 2788 Internet Protocol", RFC 4301, December 2005. 2790 [RFC4443] Conta, A., Deering, S., and M. Gupta, "Internet Control 2791 Message Protocol (ICMPv6) for the Internet Protocol 2792 Version 6 (IPv6) Specification", RFC 4443, March 2006. 2794 [RFC4555] Eronen, P., "IKEv2 Mobility and Multihoming Protocol 2795 (MOBIKE)", RFC 4555, June 2006. 2797 [RFC4592] Lewis, E., "The Role of Wildcards in the Domain Name 2798 System", RFC 4592, July 2006. 2800 [RFC4821] Mathis, M. and J. Heffner, "Packetization Layer Path MTU 2801 Discovery", RFC 4821, March 2007. 2803 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 2804 Errors at High Data Rates", RFC 4963, July 2007. 2806 [RFC4994] Zeng, S., Volz, B., Kinnear, K., and J. Brzozowski, 2807 "DHCPv6 Relay Agent Echo Request Option", RFC 4994, 2808 September 2007. 2810 [RFC5213] Gundavelli, S., Leung, K., Devarapalli, V., Chowdhury, K., 2811 and B. Patil, "Proxy Mobile IPv6", RFC 5213, August 2008. 2813 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 2814 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 2815 March 2008. 2817 [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security 2818 (TLS) Protocol Version 1.2", RFC 5246, August 2008. 2820 [RFC5320] Templin, F., "The Subnetwork Encapsulation and Adaptation 2821 Layer (SEAL)", RFC 5320, February 2010. 2823 [RFC5494] Arkko, J. and C. Pignataro, "IANA Allocation Guidelines 2824 for the Address Resolution Protocol (ARP)", RFC 5494, 2825 April 2009. 2827 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 2828 Route Optimization Requirements for Operational Use in 2829 Aeronautics and Space Exploration Mobile Networks", RFC 2830 5522, October 2009. 2832 [RFC5558] Templin, F., "Virtual Enterprise Traversal (VET)", RFC 2833 5558, February 2010. 2835 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 2836 Infrastructures (6rd)", RFC 5569, January 2010. 2838 [RFC5720] Templin, F., "Routing and Addressing in Networks with 2839 Global Enterprise Recursion (RANGER)", RFC 5720, February 2840 2010. 2842 [RFC5844] Wakikawa, R. and S. Gundavelli, "IPv4 Support for Proxy 2843 Mobile IPv6", RFC 5844, May 2010. 2845 [RFC5949] Yokota, H., Chowdhury, K., Koodli, R., Patil, B., and F. 2846 Xia, "Fast Handovers for Proxy Mobile IPv6", RFC 5949, 2847 September 2010. 2849 [RFC5996] Kaufman, C., Hoffman, P., Nir, Y., and P. Eronen, 2850 "Internet Key Exchange Protocol Version 2 (IKEv2)", RFC 2851 5996, September 2010. 2853 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 2854 NAT64: Network Address and Protocol Translation from IPv6 2855 Clients to IPv4 Servers", RFC 6146, April 2011. 2857 [RFC6179] Templin, F., "The Internet Routing Overlay Network 2858 (IRON)", RFC 6179, March 2011. 2860 [RFC6204] Singh, H., Beebee, W., Donley, C., Stark, B., and O. 2861 Troan, "Basic Requirements for IPv6 Customer Edge 2862 Routers", RFC 6204, April 2011. 2864 [RFC6241] Enns, R., Bjorklund, M., Schoenwaelder, J., and A. 2865 Bierman, "Network Configuration Protocol (NETCONF)", RFC 2866 6241, June 2011. 2868 [RFC6275] Perkins, C., Johnson, D., and J. Arkko, "Mobility Support 2869 in IPv6", RFC 6275, July 2011. 2871 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 2872 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, August 2873 2011. 2875 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 2876 for Equal Cost Multipath Routing and Link Aggregation in 2877 Tunnels", RFC 6438, November 2011. 2879 [RFC6691] Borman, D., "TCP Options and Maximum Segment Size (MSS)", 2880 RFC 6691, July 2012. 2882 [RFC6706] Templin, F., "Asymmetric Extended Route Optimization 2883 (AERO)", RFC 6706, August 2012. 2885 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID Field", 2886 RFC 6864, February 2013. 2888 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 2889 UDP Checksums for Tunneled Packets", RFC 6935, April 2013. 2891 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 2892 for the Use of IPv6 UDP Datagrams with Zero Checksums", 2893 RFC 6936, April 2013. 2895 [RFC6939] Halwasia, G., Bhandari, S., and W. Dec, "Client Link-Layer 2896 Address Option in DHCPv6", RFC 6939, May 2013. 2898 [RFC6980] Gont, F., "Security Implications of IPv6 Fragmentation 2899 with IPv6 Neighbor Discovery", RFC 6980, August 2013. 2901 [RFC7078] Matsumoto, A., Fujisaki, T., and T. Chown, "Distributing 2902 Address Selection Policy Using DHCPv6", RFC 7078, January 2903 2014. 2905 [TUNTAP] Wikipedia, W., "http://en.wikipedia.org/wiki/TUN/TAP", 2906 October 2014. 2908 Author's Address 2909 Fred L. Templin (editor) 2910 Boeing Research & Technology 2911 P.O. Box 3707 2912 Seattle, WA 98124 2913 USA 2915 Email: fltemplin@acm.org