idnits 2.17.1 draft-templin-aerolink-56.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- == The 'Obsoletes: ' line in the draft header should list only the _numbers_ of the RFCs which will be obsoleted by this document (if approved); it should not include the word 'RFC' in the list. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 844 has weird spacing: '...Version a 4-b...' -- The document date (June 9, 2015) is 3216 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 1007, but not defined == Unused Reference: 'RFC0768' is defined on line 2889, but no explicit reference was found in the text == Unused Reference: 'RFC4862' is defined on line 2933, but no explicit reference was found in the text == Unused Reference: 'RFC0879' is defined on line 2975, but no explicit reference was found in the text == Unused Reference: 'RFC1930' is defined on line 2987, but no explicit reference was found in the text == Unused Reference: 'RFC4994' is defined on line 3058, but no explicit reference was found in the text == Unused Reference: 'RFC5494' is defined on line 3075, but no explicit reference was found in the text == Unused Reference: 'RFC5720' is defined on line 3090, but no explicit reference was found in the text == Unused Reference: 'RFC6146' is defined on line 3105, but no explicit reference was found in the text == Unused Reference: 'RFC6204' is defined on line 3112, but no explicit reference was found in the text == Unused Reference: 'RFC6241' is defined on line 3120, but no explicit reference was found in the text == Unused Reference: 'RFC6355' is defined on line 3127, but no explicit reference was found in the text == Unused Reference: 'RFC6691' is defined on line 3138, but no explicit reference was found in the text == Unused Reference: 'RFC6980' is defined on line 3157, but no explicit reference was found in the text == Unused Reference: 'RFC7078' is defined on line 3160, 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 (-02) exists of draft-herbert-gue-fragmentation-00 == Outdated reference: A later version (-21) exists of draft-ietf-dhc-sedhcpv6-07 == Outdated reference: A later version (-05) exists of draft-ietf-nvo3-gue-00 -- 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 (~~), 21 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, June 9, 2015 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: December 11, 2015 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-aerolink-56.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 are 26 used in the control plane, both IPv4 and IPv6 are supported in the 27 data plane. AERO is a widely-applicable tunneling solution using 28 standard control messaging exchanges as described in this document. 30 Status of This Memo 32 This Internet-Draft is submitted in full conformance with the 33 provisions of BCP 78 and BCP 79. 35 Internet-Drafts are working documents of the Internet Engineering 36 Task Force (IETF). Note that other groups may also distribute 37 working documents as Internet-Drafts. The list of current Internet- 38 Drafts is at http://datatracker.ietf.org/drafts/current/. 40 Internet-Drafts are draft documents valid for a maximum of six months 41 and may be updated, replaced, or obsoleted by other documents at any 42 time. It is inappropriate to use Internet-Drafts as reference 43 material or to cite them other than as "work in progress." 45 This Internet-Draft will expire on December 11, 2015. 47 Copyright Notice 49 Copyright (c) 2015 IETF Trust and the persons identified as the 50 document authors. All rights reserved. 52 This document is subject to BCP 78 and the IETF Trust's Legal 53 Provisions Relating to IETF Documents 54 (http://trustee.ietf.org/license-info) in effect on the date of 55 publication of this document. Please review these documents 56 carefully, as they describe your rights and restrictions with respect 57 to this document. Code Components extracted from this document must 58 include Simplified BSD License text as described in Section 4.e of 59 the Trust Legal Provisions and are provided without warranty as 60 described in the Simplified BSD License. 62 Table of Contents 64 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 65 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4 66 3. Asymmetric Extended Route Optimization (AERO) . . . . . . . . 6 67 3.1. AERO Link Reference Model . . . . . . . . . . . . . . . . 6 68 3.2. AERO Link Node Types . . . . . . . . . . . . . . . . . . 8 69 3.3. AERO Addresses . . . . . . . . . . . . . . . . . . . . . 9 70 3.4. AERO Interface Characteristics . . . . . . . . . . . . . 10 71 3.5. AERO Link Registration . . . . . . . . . . . . . . . . . 11 72 3.6. AERO Interface Initialization . . . . . . . . . . . . . . 12 73 3.6.1. AERO Relay Behavior . . . . . . . . . . . . . . . . . 12 74 3.6.2. AERO Server Behavior . . . . . . . . . . . . . . . . 12 75 3.6.3. AERO Client Behavior . . . . . . . . . . . . . . . . 13 76 3.6.4. AERO Forwarding Agent Behavior . . . . . . . . . . . 13 77 3.7. AERO Link Routing System . . . . . . . . . . . . . . . . 13 78 3.8. AERO Interface Neighbor Cache Maintenace . . . . . . . . 15 79 3.9. AERO Interface Sending Algorithm . . . . . . . . . . . . 16 80 3.10. AERO Interface Encapsulation and Re-encapsulation . . . . 18 81 3.11. AERO Interface Decapsulation . . . . . . . . . . . . . . 20 82 3.12. AERO Interface Data Origin Authentication . . . . . . . . 21 83 3.13. AERO Interface MTU and Fragmentation . . . . . . . . . . 21 84 3.13.1. Accommodating Large Control Messages . . . . . . . . 24 85 3.13.2. Integrity . . . . . . . . . . . . . . . . . . . . . 25 86 3.14. AERO Interface Error Handling . . . . . . . . . . . . . . 26 87 3.15. AERO Router Discovery, Prefix Delegation and Address 88 Configuration . . . . . . . . . . . . . . . . . . . . . . 30 89 3.15.1. AERO DHCPv6 Service Model . . . . . . . . . . . . . 30 90 3.15.2. AERO Client Behavior . . . . . . . . . . . . . . . . 31 91 3.15.3. AERO Server Behavior . . . . . . . . . . . . . . . . 34 92 3.15.4. Deleting Link Registrations . . . . . . . . . . . . 37 93 3.16. AERO Forwarding Agent Behavior . . . . . . . . . . . . . 37 94 3.17. AERO Intradomain Route Optimization . . . . . . . . . . . 38 95 3.17.1. Reference Operational Scenario . . . . . . . . . . . 38 96 3.17.2. Concept of Operations . . . . . . . . . . . . . . . 40 97 3.17.3. Message Format . . . . . . . . . . . . . . . . . . . 40 98 3.17.4. Sending Predirects . . . . . . . . . . . . . . . . . 41 99 3.17.5. Re-encapsulating and Relaying Predirects . . . . . . 42 100 3.17.6. Processing Predirects and Sending Redirects . . . . 43 101 3.17.7. Re-encapsulating and Relaying Redirects . . . . . . 45 102 3.17.8. Processing Redirects . . . . . . . . . . . . . . . . 46 103 3.17.9. Server-Oriented Redirection . . . . . . . . . . . . 46 104 3.18. Neighbor Unreachability Detection (NUD) . . . . . . . . . 46 105 3.19. Mobility Management . . . . . . . . . . . . . . . . . . . 48 106 3.19.1. Announcing Link-Layer Address Changes . . . . . . . 48 107 3.19.2. Bringing New Links Into Service . . . . . . . . . . 49 108 3.19.3. Removing Existing Links from Service . . . . . . . . 49 109 3.19.4. Moving to a New Server . . . . . . . . . . . . . . . 50 110 3.20. Proxy AERO . . . . . . . . . . . . . . . . . . . . . . . 50 111 3.21. Extending AERO Links Through Security Gateways . . . . . 53 112 3.22. Extending IPv6 AERO Links to the Internet . . . . . . . . 55 113 3.23. Encapsulation Protocol Version Considerations . . . . . . 58 114 3.24. Multicast Considerations . . . . . . . . . . . . . . . . 58 115 3.25. Operation on AERO Links Without DHCPv6 Services . . . . . 59 116 3.26. Operation on Server-less AERO Links . . . . . . . . . . . 59 117 3.27. Manually-Configured AERO Tunnels . . . . . . . . . . . . 59 118 3.28. Intradomain Routing . . . . . . . . . . . . . . . . . . . 59 119 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 59 120 5. Next Steps . . . . . . . . . . . . . . . . . . . . . . . . . 60 121 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 60 122 7. Security Considerations . . . . . . . . . . . . . . . . . . . 60 123 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 61 124 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 62 125 9.1. Normative References . . . . . . . . . . . . . . . . . . 62 126 9.2. Informative References . . . . . . . . . . . . . . . . . 63 127 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 68 129 1. Introduction 131 This document specifies the operation of IP over tunnel virtual links 132 using Asymmetric Extended Route Optimization (AERO). The AERO link 133 can be used for tunneling to neighboring nodes over either IPv6 or 134 IPv4 networks, i.e., AERO views the IPv6 and IPv4 networks as 135 equivalent links for tunneling. Nodes attached to AERO links can 136 exchange packets via trusted intermediate routers that provide 137 forwarding services to reach off-link destinations and redirection 138 services for route optimization that addresses the requirements 139 outlined in [RFC5522]. 141 AERO provides an IPv6 link-local address format known as the AERO 142 address that supports operation of the IPv6 Neighbor Discovery (ND) 144 [RFC4861] protocol and links IPv6 ND to IP forwarding. Admission 145 control and provisioning are supported by the Dynamic Host 146 Configuration Protocol for IPv6 (DHCPv6) [RFC3315], and node mobility 147 is naturally supported through dynamic neighbor cache updates. 148 Although DHCPv6 and IPv6 ND messaging are used in the control plane, 149 both IPv4 and IPv6 can be used in the data plane. AERO is a widely- 150 applicable tunneling solution using standard control messaging 151 exchanges as described in this document. The remainder of this 152 document presents the AERO specification. 154 2. Terminology 156 The terminology in the normative references applies; the following 157 terms are defined within the scope of this document: 159 AERO link 160 a Non-Broadcast, Multiple Access (NBMA) tunnel virtual overlay 161 configured over a node's attached IPv6 and/or IPv4 networks. All 162 nodes on the AERO link appear as single-hop neighbors from the 163 perspective of the virtual overlay. 165 AERO interface 166 a node's attachment to an AERO link. Nodes typically have a 167 single AERO interface; support for multiple AERO interfaces is 168 also possible but out of scope for this document. 170 AERO address 171 an IPv6 link-local address constructed as specified in Section 3.3 172 and assigned to a Client's AERO interface. 174 AERO node 175 a node that is connected to an AERO link and that participates in 176 IPv6 ND and DHCPv6 messaging over the link. 178 AERO Client ("Client") 179 a node that issues DHCPv6 messages using the special IPv6 link- 180 local address 'fe80::ffff:ffff:ffff:ffff' to receive IP Prefix 181 Delegations (PD) from one or more AERO Servers. Following PD, the 182 Client assigns an AERO address to the AERO interface which it uses 183 in IPv6 ND messaging to coordinate with other AERO nodes. 185 AERO Server ("Server") 186 a node that configures an AERO interface to provide default 187 forwarding and DHCPv6 services for AERO Clients. The Server 188 assigns an administratively provisioned IPv6 link-local unicast 189 address to support the operation of DHCPv6 and the IPv6 ND 190 protocol. An AERO Server can also act as an AERO Relay. 192 AERO Relay ("Relay") 193 a node that configures an AERO interface to relay IP packets 194 between nodes on the same AERO link and/or forward IP packets 195 between the AERO link and the native Internetwork. The Relay 196 assigns an administratively provisioned IPv6 link-local unicast 197 address to the AERO interface the same as for a Server. An AERO 198 Relay can also act as an AERO Server. 200 AERO Forwarding Agent ("Forwarding Agent") 201 a node that performs data plane forwarding services as a companion 202 to an AERO Server. 204 ingress tunnel endpoint (ITE) 205 an AERO interface endpoint that injects tunneled packets into an 206 AERO link. 208 egress tunnel endpoint (ETE) 209 an AERO interface endpoint that receives tunneled packets from an 210 AERO link. 212 underlying network 213 a connected IPv6 or IPv4 network routing region over which the 214 tunnel virtual overlay is configured. A typical example is an 215 enterprise network. 217 underlying interface 218 an AERO node's interface point of attachment to an underlying 219 network. 221 link-layer address 222 an IP address assigned to an AERO node's underlying interface. 223 When UDP encapsulation is used, the UDP port number is also 224 considered as part of the link-layer address. Link-layer 225 addresses are used as the encapsulation header source and 226 destination addresses. 228 network layer address 229 the source or destination address of the encapsulated IP packet. 231 end user network (EUN) 232 an internal virtual or external edge IP network that an AERO 233 Client connects to the rest of the network via the AERO interface. 235 AERO Service Prefix (ASP) 236 an IP prefix associated with the AERO link and from which AERO 237 Client Prefixes (ACPs) are derived (for example, the IPv6 ACP 238 2001:db8:1:2::/64 is derived from the IPv6 ASP 2001:db8::/32). 240 AERO Client Prefix (ACP) 241 a more-specific IP prefix taken from an ASP and delegated to a 242 Client. 244 Throughout the document, the simple terms "Client", "Server" and 245 "Relay" refer to "AERO Client", "AERO Server" and "AERO Relay", 246 respectively. Capitalization is used to distinguish these terms from 247 DHCPv6 client/server/relay [RFC3315]. 249 The terminology of [RFC4861] (including the names of node variables 250 and protocol constants) applies to this document. Also throughout 251 the document, the term "IP" is used to generically refer to either 252 Internet Protocol version (i.e., IPv4 or IPv6). 254 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 255 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 256 document are to be interpreted as described in [RFC2119]. Lower case 257 uses of these words are not to be interpreted as carrying RFC2119 258 significance. 260 3. Asymmetric Extended Route Optimization (AERO) 262 The following sections specify the operation of IP over Asymmetric 263 Extended Route Optimization (AERO) links: 265 3.1. AERO Link Reference Model 266 .-(::::::::) 267 .-(:::: IP ::::)-. 268 (:: Internetwork ::) 269 `-(::::::::::::)-' 270 `-(::::::)-' 271 | 272 +--------------+ +--------+-------+ +--------------+ 273 |AERO Server S1| | AERO Relay R1 | |AERO Server S2| 274 | Nbr: C1; R1 | | Nbr: S1; S2 | | Nbr: C2; R1 | 275 | default->R1 | |(H1->S1; H2->S2)| | default->R1 | 276 | H1->C1 | +--------+-------+ | H2->C2 | 277 +-------+------+ | +------+-------+ 278 | | | 279 X---+---+-------------------+------------------+---+---X 280 | AERO Link | 281 +-----+--------+ +--------+-----+ 282 |AERO Client C1| |AERO Client C2| 283 | Nbr: S1 | | Nbr: S2 | 284 | default->S1 | | default->S2 | 285 +--------------+ +--------------+ 286 .-. .-. 287 ,-( _)-. ,-( _)-. 288 .-(_ IP )-. .-(_ IP )-. 289 (__ EUN ) (__ EUN ) 290 `-(______)-' `-(______)-' 291 | | 292 +--------+ +--------+ 293 | Host H1| | Host H2| 294 +--------+ +--------+ 296 Figure 1: AERO Link Reference Model 298 Figure 1 presents the AERO link reference model. In this model: 300 o Relay R1 acts as a default router for its associated Servers S1 301 and S2, and connects the AERO link to the rest of the IP 302 Internetwork 304 o Servers S1 and S2 associate with Relay R1 and also act as default 305 routers for their associated Clients C1 and C2. 307 o Clients C1 and C2 associate with Servers S1 and S2, respectively 308 and also act as default routers for their associated EUNs 310 o Hosts H1 and H2 attach to the EUNs served by Clients C1 and C2, 311 respectively 313 Each node maintains a neighbor cache and IP forwarding table. (For 314 example, AERO Relay R1 in the diagram has neighbor cache entries for 315 Servers S1 and S2 and IP forwarding table entries for ACPs H1 and 316 H2.) In common operational practice, there may be many additional 317 Relays, Servers and Clients. (Although not shown in the figure, AERO 318 Forwarding Agents may also be provided for data plane forwarding 319 offload services.) 321 3.2. AERO Link Node Types 323 AERO Relays provide default forwarding services to AERO Servers. 324 Relays forward packets between Servers connected to the same AERO 325 link and also forward packets between the AERO link and the native IP 326 Internetwork. Relays present the AERO link to the native 327 Internetwork as a set of one or more AERO Service Prefixes (ASPs) and 328 serve as a gateway between the AERO link and the Internetwork. AERO 329 Relays maintain an AERO interface neighbor cache entry for each AERO 330 Server, and maintain an IP forwarding table entry for each AERO 331 Client Prefix (ACP). AERO Relays can also be configured to act as 332 AERO Servers. 334 AERO Servers provide default forwarding services to AERO Clients. 335 Each Server also peers with each Relay in a dynamic routing protocol 336 instance to advertise its list of associated ACPs. Servers configure 337 a DHCPv6 server function to facilitate Prefix Delegation (PD) 338 exchanges with Clients. Each delegated prefix becomes an ACP taken 339 from an ASP. Servers forward packets between AERO interface 340 neighbors only, i.e., and not between the AERO link and the native IP 341 Internetwork. AERO Servers maintain an AERO interface neighbor cache 342 entry for each AERO Relay. They also maintain both a neighbor cache 343 entry and an IP forwarding table entry for each of their associated 344 Clients. AERO Servers can also be configured to act as AERO Relays. 346 AERO Clients act as requesting routers to receive ACPs through DHCPv6 347 PD exchanges with AERO Servers over the AERO link and sub-delegate 348 portions of their ACPs to EUN interfaces. (Each Client MAY associate 349 with a single Server or with multiple Servers, e.g., for fault 350 tolerance, load balancing, etc.) Each IPv6 Client receives at least 351 a /64 IPv6 ACP, and may receive even shorter prefixes. Similarly, 352 each IPv4 Client receives at least a /32 IPv4 ACP (i.e., a singleton 353 IPv4 address), and may receive even shorter prefixes. AERO Clients 354 maintain an AERO interface neighbor cache entry for each of their 355 associated Servers as well as for each of their correspondent 356 Clients. 358 AERO Clients typically configure a TUN/TAP interface [TUNTAP] as a 359 point-to-point linkage between the IP layer and the AERO interface. 360 The IP layer therefore sees only the TUN/TAP interface, while the 361 AERO interface provides an intermediate conduit between the TUN/TAP 362 interface and the underlying interfaces. AERO Clients that act as 363 hosts assign one or more IP addresses from their ACPs to the TUN/TAP 364 interface, i.e., and not to the AERO interface. 366 AERO Forwarding Agents provide data plane forwarding services as 367 companions to AERO Servers. Note that while Servers are required to 368 perform both control and data plane operations on their own behalf, 369 they may optionally enlist the services of special-purpose Forwarding 370 Agents to offload data plane traffic. 372 3.3. AERO Addresses 374 An AERO address is an IPv6 link-local address with an embedded ACP 375 and assigned to a Client's AERO interface. The AERO address is 376 formed as follows: 378 fe80::[ACP] 380 For IPv6, the AERO address begins with the prefix fe80::/64 and 381 includes in its interface identifier the base prefix taken from the 382 Client's IPv6 ACP. The base prefix is determined by masking the ACP 383 with the prefix length. For example, if the AERO Client receives the 384 IPv6 ACP: 386 2001:db8:1000:2000::/56 388 it constructs its AERO address as: 390 fe80::2001:db8:1000:2000 392 For IPv4, the AERO address is formed from the lower 64 bits of an 393 IPv4-mapped IPv6 address [RFC4291] that includes the base prefix 394 taken from the Client's IPv4 ACP. For example, if the AERO Client 395 receives the IPv4 ACP: 397 192.0.2.32/28 399 it constructs its AERO address as: 401 fe80::FFFF:192.0.2.32 403 The AERO address remains stable as the Client moves between 404 topological locations, i.e., even if its link-layer addresses change. 406 NOTE: In some cases, prospective neighbors may not have advanced 407 knowledge of the Client's ACP length and may therefore send initial 408 IPv6 ND messages with an AERO destination address that matches the 409 ACP but does not correspond to the base prefix. In that case, the 410 Client MUST accept the address as equivalent to the base address, but 411 then use the base address as the source address of any IPv6 ND 412 message replies. For example, if the Client receives the IPv6 ACP 413 2001:db8:1000:2000::/56 then subsequently receives an IPv6 ND message 414 with destination address fe80::2001:db8:1000:2001, it accepts the 415 message but uses fe80::2001:db8:1000:2000 as the source address of 416 any IPv6 ND replies. 418 3.4. AERO Interface Characteristics 420 AERO interfaces use encapsulation (see Section 3.10) to exchange 421 packets with neighbors attached to the AERO link. AERO interfaces 422 maintain a neighbor cache, and AERO Clients and Servers use unicast 423 IPv6 ND messaging. AERO interfaces use unicast Neighbor Solicitation 424 (NS), Neighbor Advertisement (NA), Router Solicitation (RS) and 425 Router Advertisement (RA) messages the same as for any IPv6 link. 426 AERO interfaces use two redirection message types -- the first known 427 as a Predirect message and the second being the standard Redirect 428 message (see Section 3.17). AERO links further use link-local-only 429 addressing; hence, AERO nodes ignore any Prefix Information Options 430 (PIOs) they may receive in RA messages over an AERO interface. 432 AERO interface ND messages include one or more Source/Target Link- 433 Layer Address Options (S/TLLAOs) formatted as shown in Figure 2: 435 0 1 2 3 436 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 1 437 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 438 | Type = 2 | Length = 3 | Reserved | 439 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 440 | Link ID | NDSCPs | DSCP #1 |Prf| DSCP #2 |Prf| 441 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 442 | DSCP #3 |Prf| DSCP #4 |Prf| .... 443 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 444 | UDP Port Number | | 445 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + 446 | | 447 + + 448 | IP Address | 449 + + 450 | | 451 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 452 | | 453 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 455 Figure 2: AERO Source/Target Link-Layer Address Option (S/TLLAO) 456 Format 458 In this format, Link ID is an integer value between 0 and 255 459 corresponding to an underlying interface of the target node, NDSCPs 460 encodes an integer value between 1 and 64 indicating the number of 461 Differentiated Services Code Point (DSCP) octets that follow. Each 462 DSCP octet is a 6-bit integer DSCP value followed by a 2-bit 463 Preference ("Prf") value. Each DSCP value encodes an integer between 464 0 and 63 associated with this Link ID, where the value 0 means 465 "default" and other values are interpreted as specified in [RFC2474]. 466 The 'Prf' qualifier for each DSCP value is set to the value 0 467 ("deprecated'), 1 ("low"), 2 ("medium"), or 3 ("high") to indicate a 468 preference level for packet forwarding purposes. UDP Port Number and 469 IP Address are set to the addresses used by the target node when it 470 sends encapsulated packets over the underlying interface. When the 471 encapsulation IP address family is IPv4, IP Address is formed as an 472 IPv4-mapped IPv6 address [RFC4291]. 474 AERO interfaces may be configured over multiple underlying 475 interfaces. For example, common mobile handheld devices have both 476 wireless local area network ("WLAN") and cellular wireless links. 477 These links are typically used "one at a time" with low-cost WLAN 478 preferred and highly-available cellular wireless as a standby. In a 479 more complex example, aircraft frequently have many wireless data 480 link types (e.g. satellite-based, terrestrial, air-to-air 481 directional, etc.) with diverse performance and cost properties. 483 If a Client's multiple underlying interfaces are used "one at a time" 484 (i.e., all other interfaces are in standby mode while one interface 485 is active), then Redirect, Predirect and unsolicited NA messages 486 include only a single TLLAO with Link ID set to a constant value. 488 If the Client has multiple active underlying interfaces, then from 489 the perspective of IPv6 ND it would appear to have a single link- 490 local address with multiple link-layer addresses. In that case, 491 Redirect, Predirect and unsolicited NA messages MAY include multiple 492 TLLAOs -- each with a different Link ID that corresponds to a 493 specific underlying interface of the Client. 495 3.5. AERO Link Registration 497 When an administrative authority first deploys a set of AERO Relays 498 and Servers that comprise an AERO link, they also assign a unique 499 domain name for the link, e.g., "linkupnetworks.example.com". Next, 500 if administrative policy permits Clients within the domain to serve 501 as correspondent nodes for Internet mobile nodes, the administrative 502 authority adds a Fully Qualified Domain Name (FQDN) for each of the 503 AERO link's ASPs to the Domain Name System (DNS) [RFC1035]. The FQDN 504 is based on the suffix "aero.linkupnetworks.net" with a prefix formed 505 from the wildcard-terminated reverse mapping of the ASP 507 [RFC3596][RFC4592], and resolves to a DNS PTR resource record. For 508 example, for the ASP '2001:db8:1::/48' within the domain name 509 "linkupnetworks.example.com", the DNS database contains: 511 '*.1.0.0.0.8.b.d.0.1.0.0.2.aero.linkupnetworks.net. PTR 512 linkupnetworks.example.com' 514 This DNS registration advertises the AERO link's ASPs to prospective 515 correspondent nodes. 517 3.6. AERO Interface Initialization 519 3.6.1. AERO Relay Behavior 521 When a Relay enables an AERO interface, it first assigns an 522 administratively provisioned link-local address fe80::ID to the 523 interface. Each fe80::ID address MUST be unique among all AERO nodes 524 on the link, and MUST NOT collide with any potential AERO addresses 525 nor the special addresses fe80:: and fe80::ffff:ffff:ffff:ffff. (The 526 fe80::ID addresses are typically taken from the available range 527 fe80::/96, e.g., as fe80::1, fe80::2, fe80::3, etc.) The Relay then 528 engages in a dynamic routing protocol session with all Servers on the 529 link (see: Section 3.7), and advertises its assigned ASP prefixes 530 into the native IP Internetwork. 532 Each Relay subsequently maintains an IP forwarding table entry for 533 each Client-Server association, and maintains a neighbor cache entry 534 for each Server on the link. Relays exchange NS/NA messages with 535 AERO link neighbors the same as for any AERO node, however they 536 typically do not perform explicit Neighbor Unreachability Detection 537 (NUD) (see: Section 3.18) since the dynamic routing protocol already 538 provides reachability confirmation. 540 3.6.2. AERO Server Behavior 542 When a Server enables an AERO interface, it assigns an 543 administratively provisioned link-local address fe80::ID the same as 544 for Relays. The Server further configures a DHCPv6 server function 545 to facilitate DHCPv6 PD exchanges with AERO Clients. The Server 546 maintains a neighbor cache entry for each Relay on the link, and 547 manages per-Client neighbor cache entries and IP forwarding table 548 entries based on control message exchanges. Each Server also engages 549 in a dynamic routing protocol with each Relay on the link (see: 550 Section 3.7). 552 When the Server receives an NS/RS message on the AERO interface it 553 returns an NA/RA message but does not update the neighbor cache. The 554 Server further provides a simple conduit between AERO interface 555 neighbors. Therefore, packets enter the Server's AERO interface from 556 the link layer and are forwarded back out the link layer without ever 557 leaving the AERO interface and therefore without ever disturbing the 558 network layer. 560 3.6.3. AERO Client Behavior 562 When a Client enables an AERO interface, it uses the special address 563 fe80::ffff:ffff:ffff:ffff to obtain an ACP from an AERO Server via 564 DHCPv6 PD. Next, it assigns the corresponding AERO address to the 565 AERO interface and creates a neighbor cache entry for the Server, 566 i.e., the PD exchange bootstraps autoconfiguration of a unique link- 567 local address. The Client maintains a neighbor cache entry for each 568 of its Servers and each of its active correspondent Clients. When 569 the Client receives Redirect/Predirect messages on the AERO interface 570 it updates or creates neighbor cache entries, including link-layer 571 address information. Unsolicited NA messages update the cached link- 572 layer addresses for correspondent Clients (e.g., following a link- 573 layer address change due to node mobility) but do not create new 574 neighbor cache entries. NS/NA messages used for NUD update timers in 575 existing neighbor cache entires but do not update link-layer 576 addresses nor create new neighbor cache entries. 578 Finally, the Client need not maintain any IP forwarding table entries 579 for its Servers or correspondent Clients. Instead, it can set a 580 single "route-to-interface" default route in the IP forwarding table, 581 and all forwarding decisions can be made within the AERO interface 582 based on neighbor cache entries. (On systems in which adding a 583 default route would violate security policy, the default route could 584 instead be installed via a "synthesized RA", e.g., as discussed in 585 Section 3.15.2.) 587 3.6.4. AERO Forwarding Agent Behavior 589 When a Forwarding Agent enables an AERO interface, it assigns the 590 same link-local address(es) as the companion AERO Server. The 591 Forwarding Agent thereafter provides data plane forwarding services 592 based solely on the forwarding information assigned to it by the 593 companion AERO Server. 595 3.7. AERO Link Routing System 597 Relays require full topology knowledge of all ACP/Server 598 associations, while individual Servers at a minimum only need to know 599 the ACPs for their current set of associated Clients. This is 600 accomplished through the use of an internal instance of the Border 601 Gateway Protocol (BGP) [RFC4271] coordinated between Servers and 602 Relays. This internal BGP instance does not interact with the public 603 Internet BGP instance; therefore, the AERO link is presented to the 604 IP Internetwork as a small set of ASPs as opposed to the full set of 605 individual ACPs. 607 In a reference BGP arrangement, each AERO Server is configured as an 608 Autonomous System Border Router (ASBR) for a stub Autonomous System 609 (AS) using an AS Number (ASN) that is unique within the BGP instance, 610 and each Server further peers with each Relay but does not peer with 611 other Servers. Similarly, Relays do not peer with each other, since 612 they will reliably receive all updates from all Servers and will 613 therefore have a consistent view of the AERO link ACP delegations. 615 Each Server maintains a working set of associated ACPs, and 616 dynamically announces new ACPs and withdraws departed ACPs in its BGP 617 updates to Relays. Clients are expected to remain associated with 618 their current Servers for extended timeframes, however Servers SHOULD 619 selectively suppress BGP updates for impatient Clients that 620 repeatedly associate and disassociate with them in order to dampen 621 routing churn. 623 In some environments, Relays need not send BGP updates to Servers 624 since Servers can always use Relays as default routers, however this 625 presents a data/control plane performance tradeoff. In environments 626 where sustained packet forwarding over Relays is undesirable, Relays 627 can instead report ACPs to Servers while including a BGP Remote-Next- 628 Hop [I-D.vandevelde-idr-remote-next-hop]. The Server then creates a 629 neighbor cache entry for each ACP with the Remote-Next-Hop as the 630 link-layer address to enable Server-to-Server route optimization. 632 Scaling properties of the AERO routing system are therefore limited 633 by the number of BGP routes that can be carried by Relays. Assuming 634 O(10^6) as a reasonable maximum number of BGP routes, this means that 635 O(10^6) Clients can be serviced by a single Relay. A means of 636 increasing scaling would be to assign a different set of Relays for 637 each set of ASPs. In that case, each Server still peers with each 638 Relay, but the Server institutes route filters so that each set of 639 Relays only receives BGP updates for the ASPs they aggregate. 641 Assuming up to O(10^3) sets of Relays, the system can then 642 accommodate O(10^9) Clients with no additional overhead for Servers 643 and Relays. In this way, each set of Relays services a specific set 644 of ASPs that they advertise to the native routing system outside of 645 the AERO link, and each Server configures ASP-specific routes that 646 list the correct set of Relays as next hops. 648 3.8. AERO Interface Neighbor Cache Maintenace 650 Each AERO interface maintains a conceptual neighbor cache that 651 includes an entry for each neighbor it communicates with on the AERO 652 link, the same as for any IPv6 interface [RFC4861]. AERO interface 653 neighbor cache entires are said to be one of "permanent", "static" or 654 "dynamic". 656 Permanent neighbor cache entries are created through explicit 657 administrative action; they have no timeout values and remain in 658 place until explicitly deleted. AERO Relays maintain a permanent 659 neighbor cache entry for each Server on the link, and AERO Servers 660 maintain a permanent neighbor cache entry for each Relay. Each entry 661 maintains the mapping between the neighbor's fe80::ID network-layer 662 address and corresponding link-layer address. 664 Static neighbor cache entries are created though DHCPv6 PD exchanges 665 and remain in place for durations bounded by prefix lifetimes. AERO 666 Servers maintain static neighbor cache entries for the ACPs of each 667 of their associated Clients, and AERO Clients maintain a static 668 neighbor cache entry for each of their associated Servers. When an 669 AERO Server sends a DHCPv6 Reply message response to a Client's 670 DHCPv6 Request, Rebind or Renew message, it creates or updates a 671 static neighbor cache entry based on the AERO address corresponding 672 to the Client's ACP as the network-layer address, the prefix lifetime 673 as the neighbor cache entry lifetime, the Client's encapsulation IP 674 address and UDP port number as the link-layer address and the prefix 675 length as the length to apply to the AERO address. When an AERO 676 Client receives a DHCPv6 Reply message from a Server, it creates or 677 updates a static neighbor cache entry based on the Reply message 678 link-local source address as the network-layer address, the prefix 679 lifetime as the neighbor cache entry lifetime, and the encapsulation 680 IP source address and UDP source port number as the link-layer 681 address. 683 Dynamic neighbor cache entries are created or updated based on 684 receipt of an IPv6 ND message, and are garbage-collected if not used 685 within a bounded timescale. AERO Clients maintain dynamic neighbor 686 cache entries for each of their active correspondent Client ACPs with 687 lifetimes based on IPv6 ND messaging constants. When an AERO Client 688 receives a valid Predirect message it creates or updates a dynamic 689 neighbor cache entry for the Predirect target network-layer and link- 690 layer addresses plus prefix length. The node then sets an 691 "AcceptTime" variable in the neighbor cache entry to ACCEPT_TIME 692 seconds and uses this value to determine whether packets received 693 from the correspondent can be accepted. When an AERO Client receives 694 a valid Redirect message it creates or updates a dynamic neighbor 695 cache entry for the Redirect target network-layer and link-layer 696 addresses plus prefix length. The Client then sets a "ForwardTime" 697 variable in the neighbor cache entry to FORWARD_TIME seconds and uses 698 this value to determine whether packets can be sent directly to the 699 correspondent. The Client also sets a "MaxRetry" variable to 700 MAX_RETRY to limit the number of keepalives sent when a correspondent 701 may have gone unreachable. 703 For dynamic neighbor cache entries, when an AERO Client receives a 704 valid NS message it (re)sets AcceptTime for the neighbor to 705 ACCEPT_TIME. When an AERO Client receives a valid solicited NA 706 message, it (re)sets ForwardTime for the neighbor to FORWARD_TIME and 707 sets MaxRetry to MAX_RETRY. When an AERO Client receives a valid 708 unsolicited NA message, it updates the correspondent's link-layer 709 addresses but DOES NOT reset AcceptTime, ForwardTime or MaxRetry. 711 It is RECOMMENDED that FORWARD_TIME be set to the default constant 712 value 30 seconds to match the default REACHABLE_TIME value specified 713 for IPv6 ND [RFC4861]. 715 It is RECOMMENDED that ACCEPT_TIME be set to the default constant 716 value 40 seconds to allow a 10 second window so that the AERO 717 redirection procedure can converge before AcceptTime decrements below 718 FORWARD_TIME. 720 It is RECOMMENDED that MAX_RETRY be set to 3 the same as described 721 for IPv6 ND address resolution in Section 7.3.3 of [RFC4861]. 723 Different values for FORWARD_TIME, ACCEPT_TIME, and MAX_RETRY MAY be 724 administratively set, if necessary, to better match the AERO link's 725 performance characteristics; however, if different values are chosen, 726 all nodes on the link MUST consistently configure the same values. 727 Most importantly, ACCEPT_TIME SHOULD be set to a value that is 728 sufficiently longer than FORWARD_TIME to allow the AERO redirection 729 procedure to converge. 731 3.9. AERO Interface Sending Algorithm 733 IP packets enter a node's AERO interface either from the network 734 layer (i.e., from a local application or the IP forwarding system), 735 or from the link layer (i.e., from the AERO tunnel virtual link). 736 Packets that enter the AERO interface from the network layer are 737 encapsulated and admitted into the AERO link, i.e., they are 738 tunnelled to an AERO interface neighbor. Packets that enter the AERO 739 interface from the link layer are either re-admitted into the AERO 740 link or delivered to the network layer where they are subject to 741 either local delivery or IP forwarding. Since each AERO node may 742 have only partial information about neighbors on the link, AERO 743 interfaces may forward packets with link-local destination addresses 744 at a layer below the network layer. This means that AERO nodes act 745 as both IP routers and sub-IP layer forwarding agents. AERO 746 interface sending considerations for Clients, Servers and Relays are 747 given below. 749 When an IP packet enters a Client's AERO interface from the network 750 layer, if the destination is covered by an ASP the Client searches 751 for a dynamic neighbor cache entry with a non-zero ForwardTime and an 752 AERO address that matches the packet's destination address. (The 753 destination address may be either an address covered by the 754 neighbor's ACP or the (link-local) AERO address itself.) If there is 755 a match, the Client uses a link-layer address in the entry as the 756 link-layer address for encapsulation then admits the packet into the 757 AERO link. If there is no match, the Client instead uses the link- 758 layer address of a neighboring Server as the link-layer address for 759 encapsulation. 761 When an IP packet enters a Server's AERO interface from the link 762 layer, if the destination is covered by an ASP the Server searches 763 for a neighbor cache entry with an AERO address that matches the 764 packet's destination address. (The destination address may be either 765 an address covered by the neighbor's ACP or the AERO address itself.) 766 If there is a match, the Server uses a link-layer address in the 767 entry as the link-layer address for encapsulation and re-admits the 768 packet into the AERO link. If there is no match, the Server instead 769 uses the link-layer address in a permanent neighbor cache entry for a 770 Relay as the link-layer address for encapsulation. 772 When an IP packet enters a Relay's AERO interface from the network 773 layer, the Relay searches its IP forwarding table for an entry that 774 is covered by an ASP and also matches the destination. If there is a 775 match, the Relay uses the link-layer address in a permanent neighbor 776 cache entry for a Server as the link-layer address for encapsulation 777 and admits the packet into the AERO link. When an IP packet enters a 778 Relay's AERO interface from the link-layer, if the destination is not 779 a link-local address and does not match an ASP the Relay removes the 780 packet from the AERO interface and uses IP forwarding to forward the 781 packet to the Internetwork. If the destination address is a link- 782 local address or a non-link-local address that matches an ASP, and 783 there is a more-specific ACP entry in the IP forwarding table, the 784 Relay uses the link-layer address in the corresponding neighbor cache 785 entry as the link-layer address for encapsulation and re-admits the 786 packet into the AERO link. When an IP packet enters a Relay's AERO 787 interface from either the network layer or link-layer, and the 788 packet's destination address matches an ASP but there is no more- 789 specific ACP entry, the Relay drops the packet and returns an ICMP 790 Destination Unreachable message (see: Section 3.14). 792 When an AERO Server receives a packet from a Relay via the AERO 793 interface, the Server MUST NOT forward the packet back to the same or 794 a different Relay. 796 When an AERO Relay receives a packet from a Server via the AERO 797 interface, the Relay MUST NOT forward the packet back to the same 798 Server. 800 When an AERO node re-admits a packet into the AERO link without 801 involving the network layer, the node MUST NOT decrement the network 802 layer TTL/Hop-count. 804 When an AERO node forwards a data packet to the primary link-layer 805 address of a Server, it may receive Redirect messages with an SLLAO 806 that include the link-layer address of an AERO Forwarding Agent. The 807 AERO node SHOULD record the link-layer address in the neighbor cache 808 entry for the neighbor and send subsequent data packets via this 809 address instead of the Server's primary address (see: Section 3.16). 811 3.10. AERO Interface Encapsulation and Re-encapsulation 813 AERO interfaces encapsulate IP packets according to whether they are 814 entering the AERO interface from the network layer or if they are 815 being re-admitted into the same AERO link they arrived on. This 816 latter form of encapsulation is known as "re-encapsulation". 818 The AERO interface encapsulates packets per the base tunneling 819 specifications (e.g., [RFC2003], [RFC2473], [RFC2784], [RFC4213], 820 [RFC4301], [RFC5246], etc.) except that it inserts a UDP header 821 immediately following the IP encapsulation header. If there are no 822 additional encapsulation headers (and no fragmentation, 823 identification, checksum or signature is needed), the AERO interface 824 next encapsulates the IPv4 or IPv6 packet immediately following the 825 UDP header. In that case, the most significant four bits of the 826 encapsulated packet encode the value '4' for IPv4 or '6' for IPv6. 828 For all other encapsulations, the AERO interface MUST insert an AERO 829 Header between the UDP header and the next encapsulation header as 830 shown in Figure 3: 832 0 1 2 3 833 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 1 834 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 835 |Version|N|F|C|S| Next Header |Fragment Offset (13 bits)|Res|M| 836 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 837 | Identification (32 bits) | 838 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 839 | Checksum (16 bits) | Signature (variable length) : 840 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 842 Figure 3: AERO Header 844 Version a 4-bit "Version" field. MUST be 0 for the purpose of this 845 specification. 847 N a 1-bit "Next Header" flag. MUST be 1 for the purpose of this 848 specification to indicate that "Next Header" field is present. 849 "Next Header" encodes the IP protocol number corresponding to the 850 next header in the encapsulation immediately following the AERO 851 header. For example, "Next Header" encodes the value '4' for 852 IPv4, '17' for UDP, '41' for IPv6, '47' for GRE, '50' for ESP, 853 '51' for AH, etc. 855 F a 1-bit "Fragment Header" flag. Set to '1' if the "Fragment 856 Offset", "Res", "M", and "Identification" fields are present and 857 collectively referred to as the "AERO Fragment Header"; otherwise, 858 set to '0'. 860 C a 1-bit "Checksum" flag. Set to '1' if the "Checksum" field is 861 present; otherwise, set to '0'. When present, the Checksum field 862 contains a checksum of the IP/UDP/AERO encapsulation headers prior 863 to the Checksum field. 865 S a 1-bit "Signature" flag. Set to '1' if the "Signature" field is 866 present; otherwise, set to '0'. When present, the Signature field 867 contains a cryptographic signature of the encapsulated packet 868 following the Signature field. The signature is applied prior to 869 any fragmentation; hence' the Signature field only appears in the 870 first fragment of a fragmented packet. 872 (Note: [RFC6706] defines an experimental use in which the bits 873 corresponding to (Version, N, F, C, S) are all zero, which can be 874 unambiguously distinguished from the values permitted by this 875 specification.) 877 During encapsulation, the AERO interface copies the "TTL/Hop Limit", 878 "Type of Service/Traffic Class" [RFC2983] and "Congestion 879 Experienced" [RFC3168] values in the packet's IP header into the 880 corresponding fields in the encapsulation IP header. (When IPv6 is 881 used as the encapsulation protocol, the interface also sets the Flow 882 Label value in the encapsulation header per [RFC6438].) For packets 883 undergoing re-encapsulation, the AERO interface instead copies the 884 "TTL/Hop Limit", "Type of Service/Traffic Class", "Flow Label" and 885 "Congestion Experienced" values in the original encapsulation IP 886 header into the corresponding fields in the new encapsulation IP 887 header, i.e., the values are transferred between encapsulation 888 headers and *not* copied from the encapsulated packet's network-layer 889 header. 891 The AERO interface next sets the UDP source port to a constant value 892 that it will use in each successive packet it sends, and sets the UDP 893 length field to the length of the encapsulated packet plus 8 bytes 894 for the UDP header itself, plus the length of the AERO header. For 895 packets sent via a Server, the AERO interface sets the UDP 896 destination port to 8060, i.e., the IANA-registered port number for 897 AERO. For packets sent to a correspondent Client, the AERO interface 898 sets the UDP destination port to the port value stored in the 899 neighbor cache entry for this correspondent. The AERO interface also 900 sets the UDP checksum field to zero (see: [RFC6935][RFC6936]) unless 901 an integrity check is required (see: Section 3.13.2). 903 The AERO interface next sets the IP protocol number in the 904 encapsulation header to 17 (i.e., the IP protocol number for UDP). 905 When IPv4 is used as the encapsulation protocol, the AERO interface 906 sets the DF bit as discussed in Section 3.13. The AERO interface 907 finally sets the AERO header fields as described in Figure 3. 909 3.11. AERO Interface Decapsulation 911 AERO interfaces decapsulate packets destined either to the node 912 itself or to a destination reached via an interface other than the 913 AERO interface the packet was received on. When the AERO interface 914 receives a UDP packet, it examines the first octet of the 915 encapsulated packet. 917 If the most significant four bits of the first octet encode the value 918 '4' (i.e., the IP version number value for IPv4) or the value '6' 919 (i.e., the IP version number value for IPv6), the AERO interface 920 discards the encapsulation headers and accepts the encapsulated 921 packet as an ordinary IPv6 or IPv4 data packet, respectively. If the 922 most significant four bits encode the value '0', however, the AERO 923 interface processes the packet according to the appropriate AERO 924 Header fields as specified in Figure 3. 926 3.12. AERO Interface Data Origin Authentication 928 AERO nodes employ simple data origin authentication procedures for 929 encapsulated packets they receive from other nodes on the AERO link. 930 In particular: 932 o AERO Relays and Servers accept encapsulated packets with a link- 933 layer source address that matches a permanent neighbor cache 934 entry. 936 o AERO Servers accept authentic encapsulated DHCPv6 messages from 937 Clients, and create or update a static neighbor cache entry for 938 the source based on the specific message type. 940 o AERO Servers accept encapsulated packets if there is a neighbor 941 cache entry with an AERO address that matches the packet's 942 network-layer source address and with a link-layer address that 943 matches the packet's link-layer source address. 945 o AERO Clients accept encapsulated packets if there is a static 946 neighbor cache entry with a link-layer source address that matches 947 the packet's link-layer source address. 949 o AERO Clients and Servers accept encapsulated packets if there is a 950 dynamic neighbor cache entry with an AERO address that matches the 951 packet's network-layer source address, with a link-layer address 952 that matches the packet's link-layer source address, and with a 953 non-zero AcceptTime. 955 Note that this simple data origin authentication is effective in 956 environments in which link-layer addresses cannot be spoofed. In 957 other environments, each AERO message must include a signature that 958 the recipient can use to authenticate the message origin. 960 3.13. AERO Interface MTU and Fragmentation 962 The AERO interface is the node's point of attachment to the AERO 963 link. AERO links over IP networks have a maximum link MTU of 64KB 964 minus the encapsulation overhead (termed here "ENCAPS"), since the 965 maximum packet size in the base IP specifications is 64KB 966 [RFC0791][RFC2460] (while IPv6 jumbograms can be up to 4GB, they are 967 considered optional for IPv6 nodes [RFC2675][RFC6434]). 969 IPv6 specifies a minimum link MTU of 1280 bytes [RFC2460]. This is 970 the minimum packet size the AERO interface MUST admit without 971 returning an ICMP Packet Too Big (PTB) message. Although IPv4 972 specifies a smaller minimum link MTU of 68 bytes [RFC0791], AERO 973 interfaces also observe a 1280 byte minimum for IPv4. Additionally, 974 the vast majority of links in the Internet configure an MTU of at 975 least 1500 bytes. Original source hosts have therefore become 976 conditioned to expect that IP packets up to 1500 bytes in length will 977 either be delivered to the final destination or a suitable PTB 978 message returned. However, PTB messages may be lost in the network 979 [RFC2923] resulting in failure of the IP Path MTU Discovery (PMTUD) 980 mechanisms [RFC1191][RFC1981]. 982 For these reasons, the source AERO interface (i.e., the tunnel 983 ingress)admit packets into the tunnel subject to their reasonable 984 expectation that PMTUD will convey the correct information to the 985 original source in the event that the packet is too large. In 986 particular, if the original source is within the same well-managed 987 administrative domain as the tunnel ingress, the ingress drops the 988 packet and sends a PTB message back to the original source if the 989 packet is too large to traverse the tunnel in one piece. Similarly, 990 if the tunnel ingress is within the same well-managed administrative 991 domain as the to the destination AERO interface (i.e., the tunnel 992 egress), the ingress can cache MTU values reported in PTB messages 993 received from a router on the path to the egress. 995 In all other cases, AERO interfaces admit all packets up to 1500 996 bytes in length even if some fragmentation is necessary, and admit 997 larger packets without fragmentation in case they are able to 998 traverse the tunnel in one piece. AERO interfaces are therefore 999 considered to have an indefinite MTU, i.e., instead of clamping the 1000 MTU to a finite size. 1002 For AERO links over IPv4, the IP ID field is only 16 bits in length, 1003 meaning that fragmentation at high data rates could result in data 1004 corruption due to reassembly misassociations [RFC6864][RFC4963] (see: 1005 Section 3.13.2). For AERO links over both IPv4 and IPv6, studies 1006 have also shown that IP fragments are dropped unconditionally over 1007 some network paths [I-D.taylor-v6ops-fragdrop]. For these reasons, 1008 when fragmentation is needed it is performed through insertion of an 1009 AERO fragment header (see: Section 3.10) and application of tunnel 1010 fragmentation as described in Section 3.1.7 of [RFC2764]. Since the 1011 AERO fragment header reduces the room available for packet data, but 1012 the original source has no way to control its insertion, the header 1013 length MUST be included in the ENCAPS length even for packets in 1014 which the header does not appear. 1016 The tunnel ingress therefore sends encapsulated packets to the tunnel 1017 egress according to the following algorithm: 1019 o For IP packets that are no larger than (1280-ENCAPS) bytes, the 1020 tunnel ingress encapsulates the packet and admits it into the 1021 tunnel without fragmentation. For IPv4 AERO links, the tunnel 1022 ingress sets the Don't Fragment (DF) bit to 0 so that these 1023 packets will be delivered to the tunnel egress even if there is a 1024 restricting link in the path, i.e., unless lost due to congestion 1025 or routing errors. 1027 o For IP packets that are larger than (1280-ENCAPS) bytes but no 1028 larger than 1500 bytes, the tunnel ingress encapsulates the packet 1029 and inserts an AERO fragment header. Next, the tunnel ingress 1030 uses the fragmentation algorithm in [RFC2460] to break the packet 1031 into two non-overlapping fragments where the first fragment 1032 (including ENCAPS) is no larger than 1024 bytes and the second is 1033 no larger than the first. Each fragment consists of identical 1034 UDP/IP encapsulation headers, followed by the AERO header followed 1035 by the fragment of the encapsulated packet itself. The tunnel 1036 ingress then admits both fragments into the tunnel, and for IPv4 1037 sets the DF bit to 0 in the IP encapsulation header. These 1038 fragmented encapsulated packets will be delivered to the tunnel 1039 egress. When the tunnel egress receives the fragments, it 1040 reassembles them into a whole packet per the reassembly algorithm 1041 in [RFC2460]. The tunnel egress therefore MUST be capable of 1042 reassembling packets up to 1500+ENCAPS bytes in length; hence, it 1043 is RECOMMENDED that the tunnel egress be capable of reassembling 1044 at least 2KB. 1046 o For IPv4 packets that are larger than 1500 bytes and with the DF 1047 bit set to 0, the tunnel ingress uses ordinary IPv4 fragmentation 1048 to break the unencapsulated packet into a minimum number of non- 1049 overlapping fragments where the first fragment is no larger than 1050 1024-ENCAPS and all other fragments are no larger than the first 1051 fragment. The tunnel ingress then encapsulates each fragment (and 1052 for IPv4 sets the DF bit to 0) then admits them into the tunnel. 1053 These fragments will be delivered to the final destination via the 1054 tunnel egress. 1056 o For all other IP packets, if the packet is too large to enter the 1057 underlying interface following encapsulation, the tunnel ingress 1058 drops the packet and returns a network-layer (L3) PTB message to 1059 the original source with MTU set to the larger of 1500 bytes or 1060 the underlying interface MTU minus ENCAPS. Otherwise, the tunnel 1061 ingress encapsulates the packet and admits it into the tunnel 1062 without fragmentation (and for IPv4 sets the DF bit to 1) and 1063 translates any link-layer (L2) PTB messages it may receive from 1064 the network into corresponding L3 PTB messages to send to the 1065 original source as specified in Section 3.14. Since both L2 and 1066 L3 PTB messages may be either lost or contain insufficient 1067 information, however, it is RECOMMENDED that original sources that 1068 send unfragmentable IP packets larger than 1500 bytes use 1069 Packetization Layer Path MTU Discovery (PLPMTUD) [RFC4821]. 1071 While sending packets according to the above algorithm, the tunnel 1072 ingress MAY also send 1500 byte or larger probe packets to determine 1073 whether they can reach the tunnel egress without fragmentation. If 1074 the probes succeed, the tunnel ingress can discontinue fragmentation 1075 and (for IPv4) set DF to 1. Since the path MTU within the tunnel may 1076 fluctuate due to routing changes, the tunnel ingress SHOULD continue 1077 to send additional probes subject to rate limiting and SHOULD process 1078 any L2 PTB messages as an indication that the path MTU may have 1079 decreased. If the path MTU within the tunnel becomes insufficient, 1080 the source MUST resume fragmentation. 1082 To construct a probe, the tunnel ingress prepares an NS message with 1083 a Nonce option plus trailing NULL padding octets added to the probe 1084 length without including the length of the padding in the IPv6 1085 Payload Length field, but with the length included in the 1086 encapsulating IP header. The tunnel ingress then encapsulates the 1087 padded NS message in the encapsulation headers (and for IPv4 sets DF 1088 to 1) then sends the message to the tunnel egress. If the tunnel 1089 egress returns a solicited NA message with a matching Nonce option, 1090 the tunnel ingress deems the probe successful. Note that in this 1091 process it is essential that probes follow equivalent paths to those 1092 used to convey actual data packets. This means that Equal Cost 1093 MultiPath (ECMP) and Link Aggregation Gateway (LAG) equipment in the 1094 path would need to ensure that probes and data packets follow the 1095 same path, which is outside the scope of this specification. 1097 3.13.1. Accommodating Large Control Messages 1099 Control messages (i.e., IPv6 ND, DHCPv6, etc.) MUST be accommodated 1100 even if some fragmentation is necessary. These packets are therefore 1101 accommodated through a modification of the second rule in the above 1102 algorithm as follows: 1104 o For control messages that are larger than (1280-ENCAPS) bytes, the 1105 tunnel ingress encapsulates the packet and inserts an AERO 1106 fragment header. Next, the tunnel ingress uses the fragmentation 1107 algorithm in [RFC2460] to break the packet into a minimum number 1108 of non-overlapping fragments where the first fragment (including 1109 ENCAPS) is no larger than 1024 bytes and the remaining fragments 1110 are no larger than the first. The tunnel ingress then 1111 encapsulates each fragment (and for IPv4 sets the DF bit to 0) 1112 then admits them into the tunnel. 1114 Control messages that exceed the 2KB minimum reassembly size rarely 1115 occur in the modern era, however the tunnel egress SHOULD be able to 1116 reassemble them if they do. This means that the tunnel egress SHOULD 1117 include a configuration knob allowing the operator to set a larger 1118 reassembly buffer size if large control messages become more common 1119 in the future. 1121 The tunnel ingress can send large control messages without 1122 fragmentation if there is assurance that large packets can traverse 1123 the tunnel without fragmentation. The tunnel ingress MAY send 1500 1124 byte or larger probe packets as specified above to determine a size 1125 for which fragmentation can be avoided. 1127 3.13.2. Integrity 1129 When fragmentation is needed, there must be assurance that reassembly 1130 can be safely conducted without incurring data corruption. Sources 1131 of corruption can include implementation errors, memory errors and 1132 misassociation of fragments from a first datagram with fragments of 1133 another datagram. The first two conditions (implementation and 1134 memory errors) are mitigated by modern systems and implementations 1135 that have demonstrated integrity through decades of operational 1136 practice. The third condition (reassembly misassociations) must be 1137 accounted for by AERO. 1139 The AERO fragmentation procedure described in the above algorithms 1140 reuses standard IPv6 fragmentation and reassembly code. Since the 1141 AERO fragment header includes a 32-bit ID field, there would need to 1142 be 2^32 packets alive in the network before a second packet with a 1143 duplicate ID enters the system with the (remote) possibility for a 1144 reassembly misassociation. For 1280 byte packets, and for a maximum 1145 network lifetime value of 60 seconds[RFC2460], this means that the 1146 tunnel ingress would need to produce ~(7 *10^12) bits/sec in order 1147 for a duplication event to be possible. This exceeds the bandwidth 1148 of data link technologies of the modern era, but not necessarily so 1149 going forward into the future. Although wireless data links commonly 1150 used by AERO Clients support vastly lower data rates, the aggregate 1151 data rates between AERO Servers and Relays may be substantial. 1152 However, high speed data links in the network core are expected to 1153 configure larger MTUs, e.g., 4KB, 8KB or even larger such that 1154 unfragmented packets can be used. Hence, no integrity check is 1155 included to cover the AERO fragmentation and reassembly procedures. 1157 When the tunnel ingress sends an IPv4-encapsulated packet with the DF 1158 bit set to 0 in the above algorithms, there is a chance that the 1159 packet may be fragmented by an IPv4 router somewhere within the 1160 tunnel. Since the largest such packet is only 1280 bytes, however, 1161 it is very likely that the packet will traverse the tunnel without 1162 incurring a restricting link. Even when a link within the tunnel 1163 configures an MTU smaller than 1280 bytes, it is very likely that it 1164 does so due to limited performance characteristics [RFC3819]. This 1165 means that the tunnel would not be able to convey fragmented 1166 IPv4-encapsulated packets fast enough to produce reassembly 1167 misassociations, as discussed above. However, AERO must also account 1168 for the possibility of tunnel paths that include "poorly managed" 1169 IPv4 link MTUs due to misconfigurations. 1171 Since the IPv4 header includes only a 16-bit ID field, there would 1172 only need to be 2^16 packets alive in the network before a second 1173 packet with a duplicate ID enters the system. For 1280 byte packets, 1174 and for a maximum network lifetime value of 120 seconds[RFC0791], 1175 this means that the tunnel ingress would only need to produce ~(5 1176 *10^6) bits/sec in order for a duplication event to be possible - a 1177 value that is well within range for many modern wired and wireless 1178 data link technologies. 1180 Therefore, if there is strong operational assurance that no IPv4 1181 links capable of supporting data rates of 5Mbps or more configure an 1182 MTU smaller than 1280 the tunnel ingress MAY omit an integrity check 1183 for the IPv4 fragmentation and reassembly procedures; otherwise, the 1184 tunnel ingress SHOULD include an integrity check. When an upper 1185 layer encapsulation (e.g., IPsec) already includes an integrity 1186 check, the tunnel ingress need not include an additional check. 1187 Otherwise, the tunnel ingress calculates the UDP checksum over the 1188 encapsulated packet and writes the value into the UDP encapsulation 1189 header, i.e., instead of writing the value 0. The tunnel egress will 1190 then verify the UDP checksum and discard the packet if the checksum 1191 is incorrect. 1193 3.14. AERO Interface Error Handling 1195 When an AERO node admits encapsulated packets into the AERO 1196 interface, it may receive link-layer (L2) or network-layer (L3) error 1197 indications. 1199 An L2 error indication is an ICMP error message generated by a router 1200 on the path to the neighbor or by the neighbor itself. The message 1201 includes an IP header with the address of the node that generated the 1202 error as the source address and with the link-layer address of the 1203 AERO node as the destination address. 1205 The IP header is followed by an ICMP header that includes an error 1206 Type, Code and Checksum. For ICMPv6 [RFC4443], the error Types 1207 include "Destination Unreachable", "Packet Too Big (PTB)", "Time 1208 Exceeded" and "Parameter Problem". For ICMPv4 [RFC0792], the error 1209 Types include "Destination Unreachable", "Fragmentation Needed" (a 1210 Destination Unreachable Code that is analogous to the ICMPv6 PTB), 1211 "Time Exceeded" and "Parameter Problem". 1213 The ICMP header is followed by the leading portion of the packet that 1214 generated the error, also known as the "packet-in-error". For 1215 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1216 much of invoking packet as possible without the ICMPv6 packet 1217 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1218 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1219 "Internet Header + 64 bits of Original Data Datagram", however 1220 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1221 ICMP datagram SHOULD contain as much of the original datagram as 1222 possible without the length of the ICMP datagram exceeding 576 1223 bytes". 1225 The L2 error message format is shown in Figure 4: 1227 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1228 ~ ~ 1229 | L2 IP Header of | 1230 | error message | 1231 ~ ~ 1232 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1233 | L2 ICMP Header | 1234 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1235 ~ ~ P 1236 | IP and other encapsulation | a 1237 | headers of original L3 packet | c 1238 ~ ~ k 1239 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1240 ~ ~ t 1241 | IP header of | 1242 | original L3 packet | i 1243 ~ ~ n 1244 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1245 ~ ~ e 1246 | Upper layer headers and | r 1247 | leading portion of body | r 1248 | of the original L3 packet | o 1249 ~ ~ r 1250 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1252 Figure 4: AERO Interface L2 Error Message Format 1254 The AERO node rules for processing these L2 error messages is as 1255 follows: 1257 o When an AERO node receives an L2 Parameter Problem message, it 1258 processes the message the same as described as for ordinary ICMP 1259 errors in the normative references [RFC0792][RFC4443]. 1261 o When an AERO node receives persistent L2 IPv4 Time Exceeded 1262 messages, the IP ID field may be wrapping before earlier fragments 1263 have been processed. In that case, the node SHOULD begin 1264 including IPv4 integrity checks (see: Section 3.13.2). 1266 o When an AERO Client receives persistent L2 Destination Unreachable 1267 messages in response to tunneled packets that it sends to one of 1268 its dynamic neighbor correspondents, the Client SHOULD test the 1269 path to the correspondent using Neighbor Unreachability Detection 1270 (NUD) (see Section 3.18). If NUD fails, the Client SHOULD set 1271 ForwardTime for the corresponding dynamic neighbor cache entry to 1272 0 and allow future packets destined to the correspondent to flow 1273 through a Server. 1275 o When an AERO Client receives persistent L2 Destination Unreachable 1276 messages in response to tunneled packets that it sends to one of 1277 its static neighbor Servers, the Client SHOULD test the path to 1278 the Server using NUD. If NUD fails, the Client SHOULD delete the 1279 neighbor cache entry and attempt to associate with a new Server. 1281 o When an AERO Server receives persistent L2 Destination Unreachable 1282 messages in response to tunneled packets that it sends to one of 1283 its static neighbor Clients, the Server SHOULD test the path to 1284 the Client using NUD. If NUD fails, the Server SHOULD cancel the 1285 DHCPv6 PD for the Client's ACP, withdraw its route for the ACP 1286 from the AERO routing system and delete the neighbor cache entry 1287 (see Section 3.18 and Section 3.19). 1289 o When an AERO Relay or Server receives an L2 Destination 1290 Unreachable message in response to a tunneled packet that it sends 1291 to one of its permanent neighbors, it discards the message since 1292 the routing system is likely in a temporary transitional state 1293 that will soon re-converge. 1295 o When an AERO node receives an L2 PTB message, it translates the 1296 message into an L3 PTB message if possible (*) and forwards the 1297 message toward the original source as described below. 1299 To translate an L2 PTB message to an L3 PTB message, the AERO node 1300 first caches the MTU field value of the L2 ICMP header. The node 1301 next discards the L2 IP and ICMP headers, and also discards the 1302 encapsulation headers of the original L3 packet. Next the node 1303 encapsulates the included segment of the original L3 packet in an L3 1304 IP and ICMP header, and sets the ICMP header Type and Code values to 1305 appropriate values for the L3 IP protocol. In the process, the node 1306 writes the maximum of 1500 bytes and (L2 MTU - ENCAPS) into the MTU 1307 field of the L3 ICMP header. 1309 The node next writes the IP source address of the original L3 packet 1310 as the destination address of the L3 PTB message and determines the 1311 next hop to the destination. If the next hop is reached via the AERO 1312 interface, the node uses the IPv6 address "::" or the IPv4 address 1313 "0.0.0.0" as the IP source address of the L3 PTB message. Otherwise, 1314 the node uses one of its non link-local addresses as the source 1315 address of the L3 PTB message. The node finally calculates the ICMP 1316 checksum over the L3 PTB message and writes the Checksum in the 1317 corresponding field of the L3 ICMP header. The L3 PTB message 1318 therefore is formatted as follows: 1320 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1321 ~ ~ 1322 | L3 IP Header of | 1323 | error message | 1324 ~ ~ 1325 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1326 | L3 ICMP Header | 1327 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1328 ~ ~ p 1329 | IP header of | k 1330 | original L3 packet | t 1331 ~ ~ 1332 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ i 1333 ~ ~ n 1334 | Upper layer headers and | 1335 | leading portion of body | e 1336 | of the original L3 packet | r 1337 ~ ~ r 1338 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1340 Figure 5: AERO Interface L3 Error Message Format 1342 After the node has prepared the L3 PTB message, it either forwards 1343 the message via a link outside of the AERO interface without 1344 encapsulation, or encapsulates and forwards the message to the next 1345 hop via the AERO interface. 1347 When an AERO Relay receives an L3 packet for which the destination 1348 address is covered by an ASP, if there is no more-specific routing 1349 information for the destination the Relay drops the packet and 1350 returns an L3 Destination Unreachable message. The Relay first 1351 writes the IP source address of the original L3 packet as the 1352 destination address of the L3 Destination Unreachable message and 1353 determines the next hop to the destination. If the next hop is 1354 reached via the AERO interface, the Relay uses the IPv6 address "::" 1355 or the IPv4 address "0.0.0.0" as the IP source address of the L3 1356 Destination Unreachable message and forwards the message to the next 1357 hop within the AERO interface. Otherwise, the Relay uses one of its 1358 non link-local addresses as the source address of the L3 Destination 1359 Unreachable message and forwards the message via a link outside the 1360 AERO interface. 1362 When an AERO node receives any L3 error message via the AERO 1363 interface, it examines the destination address in the L3 IP header of 1364 the message. If the next hop toward the destination address of the 1365 error message is via the AERO interface, the node re-encapsulates and 1366 forwards the message to the next hop within the AERO interface. 1367 Otherwise, if the source address in the L3 IP header of the message 1368 is the IPv6 address "::" or the IPv4 address "0.0.0.0", the node 1369 writes one of its non link-local addresses as the source address of 1370 the L3 message and recalculates the IP and/or ICMP checksums. The 1371 node finally forwards the message via a link outside of the AERO 1372 interface. 1374 (*) Note that in some instances the packet-in-error field of an L2 1375 PTB message may not include enough information for translation to an 1376 L3 PTB message. In that case, the AERO interface simply discards the 1377 L2 PTB message. It can therefore be said that translation of L2 PTB 1378 messages to L3 PTB messages can provide a useful optimization when 1379 possible, but is not critical for sources that correctly use PLPMTUD. 1381 3.15. AERO Router Discovery, Prefix Delegation and Address 1382 Configuration 1384 3.15.1. AERO DHCPv6 Service Model 1386 Each AERO Server configures a DHCPv6 server function to facilitate PD 1387 requests from Clients. Each Server is provisioned with a database of 1388 ACP-to-Client ID mappings for all Clients enrolled in the AERO 1389 system, as well as any information necessary to authenticate each 1390 Client. The Client database is maintained by a central 1391 administrative authority for the AERO link and securely distributed 1392 to all Servers, e.g., via the Lightweight Directory Access Protocol 1393 (LDAP) [RFC4511] or a similar distributed database service. 1395 Therefore, no Server-to-Server DHCPv6 PD delegation state 1396 synchronization is necessary, and Clients can optionally hold 1397 separate delegations for the same ACP from multiple Servers. In this 1398 way, Clients can associate with multiple Servers, and can receive new 1399 delegations from new Servers before deprecating delegations received 1400 from existing Servers. 1402 AERO Clients and Servers exchange Client link-layer address 1403 information using an option format similar to the Client Link Layer 1404 Address Option (CLLAO) defined in [RFC6939]. Due to practical 1405 limitations of CLLAO, however, AERO interfaces instead use Vendor- 1406 Specific Information Options as described in the following sections. 1408 3.15.2. AERO Client Behavior 1410 AERO Clients discover the link-layer addresses of AERO Servers via 1411 static configuration, or through an automated means such as DNS name 1412 resolution. In the absence of other information, the Client resolves 1413 the FQDN "linkupnetworks.[domainname]" where "linkupnetworks" is a 1414 constant text string and "[domainname]" is the connection-specific 1415 DNS suffix for the Client's underlying network connection (e.g., 1416 "example.com"). After discovering the link-layer addresses, the 1417 Client associates with one or more of the corresponding Servers. 1419 To associate with a Server, the Client acts as a requesting router to 1420 request an ACP through a two-message (i.e., Request/Reply) DHCPv6 PD 1421 exchange [RFC3315][RFC3633]. The Client's Request message includes 1422 fe80::ffff:ffff:ffff:ffff as the IPv6 source address, 1423 'All_DHCP_Relay_Agents_and_Servers' as the IPv6 destination address 1424 and the link-layer address of the Server as the link-layer 1425 destination address. The Request message also includes a Client 1426 Identifier option with a DHCP Unique Identifier (DUID) and an 1427 Identity Association for Prefix Delegation (IA_PD) option. If the 1428 Client is pre-provisioned with an ACP associated with the AERO 1429 service, it MAY also include the ACP in the IA_PD to indicate its 1430 preference to the DHCPv6 server. 1432 The Client also SHOULD include an AERO Link-registration Request 1433 (ALREQ) option to register one or more links with the Server. The 1434 Server will include an AERO Link-registration Reply (ALREP) option in 1435 the corresponding DHCPv6 Reply message as specified in 1436 Section 3.15.3. (The Client MAY omit the ALREQ option, in which case 1437 the Server will still include an ALREP option in its Reply with "Link 1438 ID" set to 0, "DSCP" set to 0, and "Prf" set to 3.) 1440 The format for the ALREQ option is shown in Figure 6: 1442 0 1 2 3 1443 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 1444 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1445 | OPTION_VENDOR_OPTS | option-len (1) | 1446 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1447 | enterprise-number = 45282 | 1448 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1449 | opt-code = OPTION_ALREQ (0) | option-len (2) | 1450 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1451 | Link ID | DSCP #1 |Prf| DSCP #2 |Prf| ... 1452 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1454 Figure 6: AERO Link-registration Request (ALREQ) Option 1456 In the above format, the Client sets 'option-code' to 1457 OPTION_VENDOR_OPTS, sets 'option-len (1)' to the length of the option 1458 following this field, sets 'enterprise-number' to 45282 (see: "IANA 1459 Considerations"), sets opt-code to the value 0 ("OPTION_ALREQ") and 1460 sets 'option-len (2)' to the length of the remainder of the option. 1461 The Client includes appropriate 'Link ID, 'DSCP' and 'Prf' values for 1462 the underlying interface over which the DHCPv6 PD Request will be 1463 issued the same as specified for an S/TLLAO Section 3.4. The Client 1464 MAY include multiple (DSCP, Prf) values with this Link ID, with the 1465 number of values indicated by option-len (2). The Server will 1466 register each value with the Link ID in the Client's neighbor cache 1467 entry. The Client finally includes any necessary authentication 1468 options to identify itself to the DHCPv6 server, and sends the 1469 encapsulated DHCPv6 PD Request via the underlying interface 1470 corresponding to Link ID. (Note that this implies that the Client 1471 must perform additional Renew/Reply DHCPv6 exchanges with the server 1472 following the initial Request/Reply using different underlying 1473 interfaces and their corresponding Link IDs if it wishes to register 1474 additional link-layer addresses and their associated DSCPs.) 1476 When the Client receives its ACP via a DHCPv6 Reply from the AERO 1477 Server, it creates a static neighbor cache entry with the Server's 1478 link-local address as the network-layer address and the Server's 1479 encapsulation address as the link-layer address. The Client then 1480 considers the link-layer address of the Server as the primary default 1481 encapsulation address for forwarding packets for which no more- 1482 specific forwarding information is available. The Client further 1483 caches any ASPs included in the ALREP option as ASPs to apply to the 1484 AERO link. 1486 Next, the Client autoconfigures an AERO address from the delegated 1487 ACP, assigns the AERO address to the AERO interface and sub-delegates 1488 the ACP to its attached EUNs and/or the Client's own internal virtual 1489 interfaces. The Client also assigns a default IP route to the AERO 1490 interface as a route-to-interface, i.e., with no explicit next-hop. 1491 The Client can then determine the correct next hops for packets 1492 submitted to the AERO interface by inspecting the neighbor cache. 1494 The Client subsequently renews its ACP delegation through each of its 1495 Servers by performing DHCPv6 Renew/Reply exchanges with the link- 1496 layer address of a Server as the link-layer destination address and 1497 the same options that were used in the initial PD request. Note that 1498 if the Client does not issue a DHCPv6 Renew before the delegation 1499 expires (e.g., if the Client has been out of touch with the Server 1500 for a considerable amount of time) it must re-initiate the DHCPv6 PD 1501 procedure. 1503 Since the Client's AERO address is obtained from the unique ACP 1504 delegation it receives, there is no need for Duplicate Address 1505 Detection (DAD) on AERO links. Other nodes maliciously attempting to 1506 hijack an authorized Client's AERO address will be denied access to 1507 the network by the DHCPv6 server due to an unacceptable link-layer 1508 address and/or security parameters (see: Security Considerations). 1510 3.15.2.1. Autoconfiguration for Constrained Platforms 1512 On some platforms (e.g., popular cell phone operating systems), the 1513 act of assigning a default IPv6 route and/or assigning an address to 1514 an interface may not be permitted from a user application due to 1515 security policy. Typically, those platforms include a TUN/TAP 1516 interface that acts as a point-to-point conduit between user 1517 applications and the AERO interface. In that case, the Client can 1518 instead generate a "synthesized RA" message. The message conforms to 1519 [RFC4861] and is prepared as follows: 1521 o the IPv6 source address is the Client's AERO address 1523 o the IPv6 destination address is all-nodes multicast 1525 o the Router Lifetime is set to a time that is no longer than the 1526 ACP DHCPv6 lifetime 1528 o the message does not include a Source Link Layer Address Option 1529 (SLLAO) 1531 o the message includes a Prefix Information Option (PIO) with a /64 1532 prefix taken from the ACP as the prefix for autoconfiguration 1534 The Client then sends the synthesized RA message via the TUN/TAP 1535 interface, where the operating system kernel will interpret it as 1536 though it were generated by an actual router. The operating system 1537 will then install a default route and use StateLess Address 1538 AutoConfiguration (SLAAC) to configure an IPv6 address on the TUN/TAP 1539 interface. Methods for similarly installing an IPv4 default route 1540 and IPv4 address on the TUN/TAP interface are based on synthesized 1541 DHCPv4 messages [RFC2131]. 1543 3.15.2.2. Client DHCPv6 Message Source Address 1545 In the initial DHCPv6 PD message exchanges, AERO Clients use the 1546 special IPv6 source address 'fe80::ffff:ffff:ffff:ffff' since their 1547 AERO addresses are not yet configured. After AERO address 1548 autoconfiguration, however, AERO Clients can either continue to use 1549 'fe80::ffff:ffff:ffff:ffff' as the source address for further DHCPv6 1550 messaging or begin using their AERO address as the source address. 1552 3.15.3. AERO Server Behavior 1554 AERO Servers configure a DHCPv6 server function on their AERO links. 1555 AERO Servers arrange to add their encapsulation layer IP addresses 1556 (i.e., their link-layer addresses) to the DNS resource records for 1557 the FQDN "linkupnetworks.[domainname]" before entering service. 1559 When an AERO Server receives a prospective Client's DHCPv6 PD Request 1560 on its AERO interface, it first authenticates the message. If 1561 authentication succeeds, the Server determines the correct ACP to 1562 delegate to the Client by searching the Client database. In 1563 environments where spoofing is not considered a threat, the Server 1564 MAY use the Client's DUID as the identification value. Otherwise, 1565 the Server SHOULD use a signed certificate provided by the Client. 1567 When the Server delegates the ACP, it also creates an IP forwarding 1568 table entry so that the AERO routing system will propagate the ACP to 1569 all Relays that aggregate the corresponding ASP (see: Section 3.7). 1570 Next, the Server prepares a DHCPv6 Reply message to send to the 1571 Client while using fe80::ID as the IPv6 source address, the link- 1572 local address taken from the Client's Request as the IPv6 destination 1573 address, the Server's link-layer address as the source link-layer 1574 address, and the Client's link-layer address as the destination link- 1575 layer address. The server also includes an IA_PD option with the 1576 delegated ACP. 1578 The Server also includes an ALREP option that includes the UDP Port 1579 Number and IP Address values it observed when it received the ALREQ 1580 in the Client's original DHCPv6 message (if present) followed by the 1581 ASP(s) for the AERO link. The ALREP option is formatted as shown in 1582 Figure 7: 1584 0 1 2 3 1585 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 1586 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1587 | OPTION_VENDOR_OPTS | option-len (1) | 1588 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1589 | enterprise-number = 45282 | 1590 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1591 | opt-code = OPTION_ALREP (1) | option-len (2) | 1592 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1593 | Link ID | Reserved | UDP Port Number | 1594 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1595 | | 1596 + + 1597 | | 1598 + IP Address + 1599 | | 1600 + + 1601 | | 1602 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1603 | | 1604 + AERO Service Prefix (ASP) #1 +-+-+-+-+-+-+-+-+ 1605 | | Prefix Len | 1606 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1607 | | 1608 + AERO Service Prefix (ASP) #2 +-+-+-+-+-+-+-+-+ 1609 | | Prefix Len | 1610 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1611 ~ ~ 1612 ~ ~ 1614 Figure 7: AERO Link-registration Reply (ALREP) Option 1616 In the ALREP, the Server sets 'option-code' to OPTION_VENDOR_OPTS, 1617 sets 'option-length (1)' to the length of the option, sets 1618 'enterprise-number' to 45282 (see: "IANA Considerations"), sets opt- 1619 code to OPTION_ALREP (1), and sets 'option-len (2)' to the length of 1620 the remainder of the option. Next, the Server sets 'Link ID' to the 1621 same value that appeared in the ALREQ, sets Reserved to 0 and sets 1622 'UDP Port Number' and 'IP address' to the Client's link-layer 1623 address. The Server next includes one or more ASP with the IP prefix 1624 as it would appear in the interface identifier portion of the 1625 corresponding AERO address (see: Section 3.3), except that the low- 1626 order 8 bits of the ASP field encode the prefix length instead of the 1627 low-order 8 bits of the prefix. The longest prefix that can 1628 therefore appear as an ASP is /56 for IPv6 or /24 for IPv4. (Note 1629 that if the Client did not include an ALREQ option in its DHCPv6 1630 message, the Server MUST still include an ALREP option in the 1631 corresponding reply with 'Link ID' set to 0.) 1633 When the Server admits the DHCPv6 Reply message into the AERO 1634 interface, it creates a static neighbor cache entry for the Client's 1635 AERO address with lifetime set to no more than the delegation 1636 lifetime and the Client's link-layer address as the link-layer 1637 address for the Link ID specified in the ALREQ. The Server then uses 1638 the Client link-layer address information in the ALREQ option as the 1639 link-layer address for encapsulation based on the (DSCP, Prf) 1640 information. 1642 After the initial DHCPv6 PD exchange, the AERO Server maintains the 1643 neighbor cache entry for the Client until the delegation lifetime 1644 expires. If the Client issues a Renew/Reply exchange, the Server 1645 extends the lifetime. If the Client issues a Release/Reply, or if 1646 the Client does not issue a Renew/Reply before the lifetime expires, 1647 the Server deletes the neighbor cache entry for the Client and 1648 withdraws the IP route from the AERO routing system. 1650 3.15.3.1. Lightweight DHCPv6 Relay Agent (LDRA) 1652 AERO Clients and Servers are always on the same link (i.e., the AERO 1653 link) from the perspective of DHCPv6. However, in some 1654 implementations the DHCPv6 server and AERO interface driver may be 1655 located in separate modules. In that case, the Server's AERO 1656 interface driver module acts as a Lightweight DHCPv6 Relay Agent 1657 (LDRA)[RFC6221] to relay DHCPv6 messages to and from the DHCPv6 1658 server module. 1660 When the LDRA receives a DHCPv6 message from a client, it prepares an 1661 ALREP option the same as described above then wraps the option in a 1662 Relay-Supplied DHCP Option option (RSOO) [RFC6422]. The LDRA then 1663 incorporates the option into the Relay-Forward message and forwards 1664 the message to the DHCPv6 server. 1666 When the DHCPv6 server receives the Relay-Forward message, it caches 1667 the ALREP option and authenticates the encapsulated DHCPv6 message. 1668 The DHCPv6 server subsequently ignores the ALREQ option itself, since 1669 the relay has already included the ALREP option. 1671 When the DHCPv6 server prepares a Reply message, it then includes the 1672 ALREP option in the body of the message along with any other options, 1673 then wraps the message in a Relay-Reply message. The DHCPv6 server 1674 then delivers the Relay-Reply message to the LDRA, which discards the 1675 Relay-Reply wrapper and delivers the DHCPv6 message to the Client. 1677 3.15.4. Deleting Link Registrations 1679 After an AERO Client registers its Link IDs and their associated 1680 (DSCP,Prf) values with the AERO Server, the Client may wish to delete 1681 one or more Link registrations, e.g., if an underlying link becomes 1682 unavailable. To do so, the Client prepares a DHCPv6 Renew message 1683 that includes an AERO Link-registration Delete (ALDEL) option and 1684 sends the Renew message to the Server over any available underlying 1685 link. The ALDEL option is formatted as shown in Figure 8: 1687 0 1 2 3 1688 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 1689 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1690 | OPTION_VENDOR_OPTS | option-len (1) | 1691 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1692 | enterprise-number = 45282 | 1693 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1694 | opt-code = OPTION_ALDEL (2) | option-len (2) | 1695 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1696 | Link ID #1 | Link ID #2 | Link ID #3 | ... 1697 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1699 Figure 8: AERO Link-registration Delete (ALDEL) Option 1701 In the ALDEL, the Client sets 'option-code' to OPTION_VENDOR_OPTS, 1702 sets 'option-length (1)' to the length of the option, sets 1703 'enterprise-number' to 45282 (see: "IANA Considerations"), sets 1704 optcode to OPTION_ALDEL (2), and sets 'option-len (2)' to the length 1705 of the remainder of the option. Next, the Server includes each 'Link 1706 ID' value that it wishes to delete. 1708 If the Client wishes to discontinue use of a Server and thereby 1709 delete all of its Link ID associations, it must use a DHCPv6 Release/ 1710 Reply exchange to delete the entire neighbor cache entry, i.e., 1711 instead of using a DHCPv6 Renew/Reply exchange with one or more ALDEL 1712 options. 1714 3.16. AERO Forwarding Agent Behavior 1716 AERO Servers MAY associate with one or more companion AERO Forwarding 1717 Agents as platforms for offloading high-speed data plane traffic. 1718 When an AERO Server receives a Client's DHCPv6 Request/Renew/Rebind/ 1719 Release message, it services the message then forwards the 1720 corresponding Reply message to the Forwarding Agent. When the 1721 Forwarding Agent receives the Reply message, it creates, updates or 1722 deletes a neighbor cache entry with the Client's AERO address and 1723 link-layer information included in the Reply message. The Forwarding 1724 Agent then forwards the Reply message back to the AERO Server, which 1725 forwards the message to the Client. In this way, Forwarding Agent 1726 state is managed in conjunction with Server state, with the Client 1727 responsible for reliability. If the Client subsequently disappears 1728 without issuing a Release, the Server is responsible for purging 1729 stale state by sending synthesized Reply messages to the Forwarding 1730 Agent. 1732 When an AERO Server receives a data packet on an AERO interface with 1733 a network layer destination address for which it has distributed 1734 forwarding information to a Forwarding Agent, the Server returns a 1735 Redirect message to the source neighbor (subject to rate limiting) 1736 then forwards the data packet as usual. The Redirect message 1737 includes a TLLAO with the link-layer address of the Forwarding 1738 Engine. 1740 When the source neighbor receives the Redirect message, it SHOULD 1741 record the link-layer address in the TLLAO as the encapsulation 1742 addresses to use for sending subsequent data packets. However, the 1743 source MUST continue to use the primary link-layer address of the 1744 Server as the encapsulation address for sending control messages. 1746 3.17. AERO Intradomain Route Optimization 1748 When a source Client forwards packets to a prospective correspondent 1749 Client within the same AERO link domain (i.e., one for which the 1750 packet's destination address is covered by an ASP), the source Client 1751 initiates an intra-domain AERO route optimization procedure. It is 1752 important to note that this procedure is initiated by the Client; if 1753 the procedure were initiated by the Server, the Server would have no 1754 way of knowing whether the Client was actually able to contact the 1755 correspondent over the route-optimized path. 1757 The procedure is based on an exchange of IPv6 ND messages using a 1758 chain of AERO Servers and Relays as a trust basis. This procedure is 1759 in contrast to the Return Routability procedure required for route 1760 optimization to a correspondent Client located in the Internet as 1761 described in Section 3.22. The following sections specify the AERO 1762 intradomain route optimization procedure. 1764 3.17.1. Reference Operational Scenario 1766 Figure 9 depicts the AERO intradomain route optimization reference 1767 operational scenario, using IPv6 addressing as the example (while not 1768 shown, a corresponding example for IPv4 addressing can be easily 1769 constructed). The figure shows an AERO Relay ('R1'), two AERO 1770 Servers ('S1', 'S2'), two AERO Clients ('C1', 'C2') and two ordinary 1771 IPv6 hosts ('H1', 'H2'): 1773 +--------------+ +--------------+ +--------------+ 1774 | Server S1 | | Relay R1 | | Server S2 | 1775 +--------------+ +--------------+ +--------------+ 1776 fe80::2 fe80::1 fe80::3 1777 L2(S1) L2(R1) L2(S2) 1778 | | | 1779 X-----+-----+------------------+-----------------+----+----X 1780 | AERO Link | 1781 L2(A) L2(B) 1782 fe80::2001:db8:0:0 fe80::2001:db8:1:0 1783 +--------------+ +--------------+ 1784 |AERO Client C1| |AERO Client C2| 1785 +--------------+ +--------------+ 1786 2001:DB8:0::/48 2001:DB8:1::/48 1787 | | 1788 .-. .-. 1789 ,-( _)-. 2001:db8:0::1 2001:db8:1::1 ,-( _)-. 1790 .-(_ IP )-. +---------+ +---------+ .-(_ IP )-. 1791 (__ EUN )--| Host H1 | | Host H2 |--(__ EUN ) 1792 `-(______)-' +---------+ +---------+ `-(______)-' 1794 Figure 9: AERO Reference Operational Scenario 1796 In Figure 9, Relay ('R1') assigns the address fe80::1 to its AERO 1797 interface with link-layer address L2(R1), Server ('S1') assigns the 1798 address fe80::2 with link-layer address L2(S1),and Server ('S2') 1799 assigns the address fe80::3 with link-layer address L2(S2). Servers 1800 ('S1') and ('S2') next arrange to add their link-layer addresses to a 1801 published list of valid Servers for the AERO link. 1803 AERO Client ('C1') receives the ACP 2001:db8:0::/48 in a DHCPv6 PD 1804 exchange via AERO Server ('S1') then assigns the address 1805 fe80::2001:db8:0:0 to its AERO interface with link-layer address 1806 L2(C1). Client ('C1') configures a default route and neighbor cache 1807 entry via the AERO interface with next-hop address fe80::2 and link- 1808 layer address L2(S1), then sub-delegates the ACP to its attached 1809 EUNs. IPv6 host ('H1') connects to the EUN, and configures the 1810 address 2001:db8:0::1. 1812 AERO Client ('C2') receives the ACP 2001:db8:1::/48 in a DHCPv6 PD 1813 exchange via AERO Server ('S2') then assigns the address 1814 fe80::2001:db8:1:0 to its AERO interface with link-layer address 1815 L2(C2). Client ('C2') configures a default route and neighbor cache 1816 entry via the AERO interface with next-hop address fe80::3 and link- 1817 layer address L2(S2), then sub-delegates the ACP to its attached 1818 EUNs. IPv6 host ('H1') connects to the EUN, and configures the 1819 address 2001:db8:1::1. 1821 3.17.2. Concept of Operations 1823 Again, with reference to Figure 9, when source host ('H1') sends a 1824 packet to destination host ('H2'), the packet is first forwarded over 1825 the source host's attached EUN to Client ('C1'). Client ('C1') then 1826 forwards the packet via its AERO interface to Server ('S1') and also 1827 sends a Predirect message toward Client ('C2') via Server ('S1'). 1828 Server ('S1') then re-encapsulates and forwards both the packet and 1829 the Predirect message out the same AERO interface toward Client 1830 ('C2') via Relay ('R1'). 1832 When Relay ('R1') receives the packet and Predirect message, it 1833 consults its forwarding table to discover Server ('S2') as the next 1834 hop toward Client ('C2'). Relay ('R1') then forwards both the packet 1835 and the Predirect message to Server ('S2'), which then forwards them 1836 to Client ('C2'). 1838 After Client ('C2') receives the Predirect message, it process the 1839 message and returns a Redirect message toward Client ('C1') via 1840 Server ('S2'). During the process, Client ('C2') also creates or 1841 updates a dynamic neighbor cache entry for Client ('C1'). 1843 When Server ('S2') receives the Redirect message, it re-encapsulates 1844 the message and forwards it on to Relay ('R1'), which forwards the 1845 message on to Server ('S1') which forwards the message on to Client 1846 ('C1'). After Client ('C1') receives the Redirect message, it 1847 processes the message and creates or updates a dynamic neighbor cache 1848 entry for Client ('C2'). 1850 Following the above Predirect/Redirect message exchange, forwarding 1851 of packets from Client ('C1') to Client ('C2') without involving any 1852 intermediate nodes is enabled. The mechanisms that support this 1853 exchange are specified in the following sections. 1855 3.17.3. Message Format 1857 AERO Redirect/Predirect messages use the same format as for ICMPv6 1858 Redirect messages depicted in Section 4.5 of [RFC4861], but also 1859 include a new "Prefix Length" field taken from the low-order 8 bits 1860 of the Redirect message Reserved field. For IPv6, valid values for 1861 the Prefix Length field are 0 through 64; for IPv4, valid values are 1862 0 through 32. The Redirect/Predirect messages are formatted as shown 1863 in Figure 10: 1865 0 1 2 3 1866 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 1867 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1868 | Type (=137) | Code (=0/1) | Checksum | 1869 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1870 | Reserved | Prefix Length | 1871 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1872 | | 1873 + + 1874 | | 1875 + Target Address + 1876 | | 1877 + + 1878 | | 1879 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1880 | | 1881 + + 1882 | | 1883 + Destination Address + 1884 | | 1885 + + 1886 | | 1887 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1888 | Options ... 1889 +-+-+-+-+-+-+-+-+-+-+-+- 1891 Figure 10: AERO Redirect/Predirect Message Format 1893 3.17.4. Sending Predirects 1895 When a Client forwards a packet with a source address from one of its 1896 ACPs toward a destination address covered by an ASP (i.e., toward 1897 another AERO Client connected to the same AERO link), the source 1898 Client MAY send a Predirect message forward toward the destination 1899 Client via the Server. 1901 In the reference operational scenario, when Client ('C1') forwards a 1902 packet toward Client ('C2'), it MAY also send a Predirect message 1903 forward toward Client ('C2'), subject to rate limiting (see 1904 Section 8.2 of [RFC4861]). Client ('C1') prepares the Predirect 1905 message as follows: 1907 o the link-layer source address is set to 'L2(C1)' (i.e., the link- 1908 layer address of Client ('C1')). 1910 o the link-layer destination address is set to 'L2(S1)' (i.e., the 1911 link-layer address of Server ('S1')). 1913 o the network-layer source address is set to fe80::2001:db8:0:0 1914 (i.e., the AERO address of Client ('C1')). 1916 o the network-layer destination address is set to fe80::2001:db8:1:0 1917 (i.e., the AERO address of Client ('C2')). 1919 o the Type is set to 137. 1921 o the Code is set to 1 to indicate "Predirect". 1923 o the Prefix Length is set to the length of the prefix to be 1924 assigned to the Target Address. 1926 o the Target Address is set to fe80::2001:db8:0:0 (i.e., the AERO 1927 address of Client ('C1')). 1929 o the Destination Address is set to the source address of the 1930 originating packet that triggered the Predirection event. (If the 1931 originating packet is an IPv4 packet, the address is constructed 1932 in IPv4-compatible IPv6 address format). 1934 o the message includes one or more TLLAOs with Link ID and DSCPs set 1935 to appropriate values for Client ('C1')'s underlying interfaces, 1936 and with UDP Port Number and IP Address set to 0'. 1938 o the message SHOULD include a Timestamp option and a Nonce option. 1940 o the message includes a Redirected Header Option (RHO) that 1941 contains the originating packet truncated if necessary to ensure 1942 that at least the network-layer header is included but the size of 1943 the message does not exceed 1280 bytes. 1945 Note that the act of sending Predirect messages is cited as "MAY", 1946 since Client ('C1') may have advanced knowledge that the direct path 1947 to Client ('C2') would be unusable or otherwise undesirable. If the 1948 direct path later becomes unusable after the initial route 1949 optimization, Client ('C1') simply allows packets to again flow 1950 through Server ('S1'). 1952 3.17.5. Re-encapsulating and Relaying Predirects 1954 When Server ('S1') receives a Predirect message from Client ('C1'), 1955 it first verifies that the TLLAOs in the Predirect are a proper 1956 subset of the Link IDs in Client ('C1')'s neighbor cache entry. If 1957 the Client's TLLAOs are not acceptable, Server ('S1') discards the 1958 message. Otherwise, Server ('S1') validates the message according to 1959 the ICMPv6 Redirect message validation rules in Section 8.1 of 1960 [RFC4861], except that the Predirect has Code=1. Server ('S1') also 1961 verifies that Client ('C1') is authorized to use the Prefix Length in 1962 the Predirect when applied to the AERO address in the network-layer 1963 source address by searching for the AERO address in the neighbor 1964 cache. If validation fails, Server ('S1') discards the Predirect; 1965 otherwise, it copies the correct UDP Port numbers and IP Addresses 1966 for Client ('C1')'s links into the (previously empty) TLLAOs. 1968 Server ('S1') then examines the network-layer destination address of 1969 the Predirect to determine the next hop toward Client ('C2') by 1970 searching for the AERO address in the neighbor cache. Since Client 1971 ('C2') is not one of its neighbors, Server ('S1') re-encapsulates the 1972 Predirect and relays it via Relay ('R1') by changing the link-layer 1973 source address of the message to 'L2(S1)' and changing the link-layer 1974 destination address to 'L2(R1)'. Server ('S1') finally forwards the 1975 re-encapsulated message to Relay ('R1') without decrementing the 1976 network-layer TTL/Hop Limit field. 1978 When Relay ('R1') receives the Predirect message from Server ('S1') 1979 it determines that Server ('S2') is the next hop toward Client ('C2') 1980 by consulting its forwarding table. Relay ('R1') then re- 1981 encapsulates the Predirect while changing the link-layer source 1982 address to 'L2(R1)' and changing the link-layer destination address 1983 to 'L2(S2)'. Relay ('R1') then relays the Predirect via Server 1984 ('S2'). 1986 When Server ('S2') receives the Predirect message from Relay ('R1') 1987 it determines that Client ('C2') is a neighbor by consulting its 1988 neighbor cache. Server ('S2') then re-encapsulates the Predirect 1989 while changing the link-layer source address to 'L2(S2)' and changing 1990 the link-layer destination address to 'L2(C2)'. Server ('S2') then 1991 forwards the message to Client ('C2'). 1993 3.17.6. Processing Predirects and Sending Redirects 1995 When Client ('C2') receives the Predirect message, it accepts the 1996 Predirect only if the message has a link-layer source address of one 1997 of its Servers (e.g., L2(S2)). Client ('C2') further accepts the 1998 message only if it is willing to serve as a redirection target. 1999 Next, Client ('C2') validates the message according to the ICMPv6 2000 Redirect message validation rules in Section 8.1 of [RFC4861], except 2001 that it accepts the message even though Code=1 and even though the 2002 network-layer source address is not that of it's current first-hop 2003 router. 2005 In the reference operational scenario, when Client ('C2') receives a 2006 valid Predirect message, it either creates or updates a dynamic 2007 neighbor cache entry that stores the Target Address of the message as 2008 the network-layer address of Client ('C1') , stores the link-layer 2009 addresses found in the TLLAOs as the link-layer addresses of Client 2010 ('C1') and stores the Prefix Length as the length to be applied to 2011 the network-layer address for forwarding purposes. Client ('C2') 2012 then sets AcceptTime for the neighbor cache entry to ACCEPT_TIME. 2014 After processing the message, Client ('C2') prepares a Redirect 2015 message response as follows: 2017 o the link-layer source address is set to 'L2(C2)' (i.e., the link- 2018 layer address of Client ('C2')). 2020 o the link-layer destination address is set to 'L2(S2)' (i.e., the 2021 link-layer address of Server ('S2')). 2023 o the network-layer source address is set to fe80::2001:db8:1:0 2024 (i.e., the AERO address of Client ('C2')). 2026 o the network-layer destination address is set to fe80::2001:db8:0:0 2027 (i.e., the AERO address of Client ('C1')). 2029 o the Type is set to 137. 2031 o the Code is set to 0 to indicate "Redirect". 2033 o the Prefix Length is set to the length of the prefix to be applied 2034 to the Target Address. 2036 o the Target Address is set to fe80::2001:db8:1:0 (i.e., the AERO 2037 address of Client ('C2')). 2039 o the Destination Address is set to the destination address of the 2040 originating packet that triggered the Redirection event. (If the 2041 originating packet is an IPv4 packet, the address is constructed 2042 in IPv4-compatible IPv6 address format). 2044 o the message includes one or more TLLAOs with Link ID and DSCPs set 2045 to appropriate values for Client ('C2')'s underlying interfaces, 2046 and with UDP Port Number and IP Address set to '0'. 2048 o the message SHOULD include a Timestamp option and MUST echo the 2049 Nonce option received in the Predirect (i.e., if a Nonce option is 2050 included). 2052 o the message includes as much of the RHO copied from the 2053 corresponding AERO Predirect message as possible such that at 2054 least the network-layer header is included but the size of the 2055 message does not exceed 1280 bytes. 2057 After Client ('C2') prepares the Redirect message, it sends the 2058 message to Server ('S2'). 2060 3.17.7. Re-encapsulating and Relaying Redirects 2062 When Server ('S2') receives a Redirect message from Client ('C2'), it 2063 first verifies that the TLLAOs in the Redirect are a proper subset of 2064 the Link IDs in Client ('C2')'s neighbor cache entry. If the 2065 Client's TLLAOs are not acceptable, Server ('S2') discards the 2066 message. Otherwise, Server ('S2') validates the message according to 2067 the ICMPv6 Redirect message validation rules in Section 8.1 of 2068 [RFC4861]. Server ('S2') also verifies that Client ('C2') is 2069 authorized to use the Prefix Length in the Redirect when applied to 2070 the AERO address in the network-layer source address by searching for 2071 the AERO address in the neighbor cache. If validation fails, Server 2072 ('S2') discards the Predirect; otherwise, it copies the correct UDP 2073 Port numbers and IP Addresses for Client ('C2')'s links into the 2074 (previously empty) TLLAOs. 2076 Server ('S2') then examines the network-layer destination address of 2077 the Predirect to determine the next hop toward Client ('C2') by 2078 searching for the AERO address in the neighbor cache. Since Client 2079 ('C2') is not a neighbor, Server ('S2') re-encapsulates the Predirect 2080 and relays it via Relay ('R1') by changing the link-layer source 2081 address of the message to 'L2(S2)' and changing the link-layer 2082 destination address to 'L2(R1)'. Server ('S2') finally forwards the 2083 re-encapsulated message to Relay ('R1') without decrementing the 2084 network-layer TTL/Hop Limit field. 2086 When Relay ('R1') receives the Predirect message from Server ('S2') 2087 it determines that Server ('S1') is the next hop toward Client ('C1') 2088 by consulting its forwarding table. Relay ('R1') then re- 2089 encapsulates the Predirect while changing the link-layer source 2090 address to 'L2(R1)' and changing the link-layer destination address 2091 to 'L2(S1)'. Relay ('R1') then relays the Predirect via Server 2092 ('S1'). 2094 When Server ('S1') receives the Predirect message from Relay ('R1') 2095 it determines that Client ('C1') is a neighbor by consulting its 2096 neighbor cache. Server ('S1') then re-encapsulates the Predirect 2097 while changing the link-layer source address to 'L2(S1)' and changing 2098 the link-layer destination address to 'L2(C1)'. Server ('S1') then 2099 forwards the message to Client ('C1'). 2101 3.17.8. Processing Redirects 2103 When Client ('C1') receives the Redirect message, it accepts the 2104 message only if it has a link-layer source address of one of its 2105 Servers (e.g., ''L2(S1)'). Next, Client ('C1') validates the message 2106 according to the ICMPv6 Redirect message validation rules in 2107 Section 8.1 of [RFC4861], except that it accepts the message even 2108 though the network-layer source address is not that of it's current 2109 first-hop router. Following validation, Client ('C1') then processes 2110 the message as follows. 2112 In the reference operational scenario, when Client ('C1') receives 2113 the Redirect message, it either creates or updates a dynamic neighbor 2114 cache entry that stores the Target Address of the message as the 2115 network-layer address of Client ('C2'), stores the link-layer 2116 addresses found in the TLLAOs as the link-layer addresses of Client 2117 ('C2') and stores the Prefix Length as the length to be applied to 2118 the network-layer address for forwarding purposes. Client ('C1') 2119 then sets ForwardTime for the neighbor cache entry to FORWARD_TIME. 2121 Now, Client ('C1') has a neighbor cache entry with a valid 2122 ForwardTime value, while Client ('C2') has a neighbor cache entry 2123 with a valid AcceptTime value. Thereafter, Client ('C1') may forward 2124 ordinary network-layer data packets directly to Client ('C2') without 2125 involving any intermediate nodes, and Client ('C2') can verify that 2126 the packets came from an acceptable source. (In order for Client 2127 ('C2') to forward packets to Client ('C1'), a corresponding 2128 Predirect/Redirect message exchange is required in the reverse 2129 direction; hence, the mechanism is asymmetric.) 2131 3.17.9. Server-Oriented Redirection 2133 In some environments, the Server nearest the target Client may need 2134 to serve as the redirection target, e.g., if direct Client-to-Client 2135 communications are not possible. In that case, the Server prepares 2136 the Redirect message the same as if it were the destination Client 2137 (see: Section 3.17.6), except that it writes its own link-layer 2138 address in the TLLAO option. The Server must then maintain a dynamic 2139 neighbor cache entry for the redirected source Client. 2141 3.18. Neighbor Unreachability Detection (NUD) 2143 AERO nodes perform Neighbor Unreachability Detection (NUD) by sending 2144 unicast NS messages to elicit solicited NA messages from neighbors 2145 the same as described in [RFC4861]. NUD is performed either 2146 reactively in response to persistent L2 errors (see Section 3.14) or 2147 proactively to refresh existing neighbor cache entries. 2149 When an AERO node sends an NS/NA message, it MUST use its link-local 2150 address as the IPv6 source address and the link-local address of the 2151 neighbor as the IPv6 destination address. When an AERO node receives 2152 an NS message or a solicited NA message, it accepts the message if it 2153 has a neighbor cache entry for the neighbor; otherwise, it ignores 2154 the message. 2156 When a source Client is redirected to a target Client it SHOULD 2157 proactively test the direct path by sending an initial NS message to 2158 elicit a solicited NA response. While testing the path, the source 2159 Client can optionally continue sending packets via the Server, 2160 maintain a small queue of packets until target reachability is 2161 confirmed, or (optimistically) allow packets to flow directly to the 2162 target. The source Client SHOULD thereafter continue to proactively 2163 test the direct path to the target Client (see Section 7.3 of 2164 [RFC4861]) periodically in order to keep dynamic neighbor cache 2165 entries alive. 2167 In particular, while the source Client is actively sending packets to 2168 the target Client it SHOULD also send NS messages separated by 2169 RETRANS_TIMER milliseconds in order to receive solicited NA messages. 2170 If the source Client is unable to elicit a solicited NA response from 2171 the target Client after MAX_RETRY attempts, it SHOULD set ForwardTime 2172 to 0 and resume sending packets via one of its Servers. Otherwise, 2173 the source Client considers the path usable and SHOULD thereafter 2174 process any link-layer errors as a hint that the direct path to the 2175 target Client has either failed or has become intermittent. 2177 When a target Client receives an NS message from a source Client, it 2178 resets AcceptTime to ACCEPT_TIME if a neighbor cache entry exists; 2179 otherwise, it discards the NS message. If ForwardTime is non-zero, 2180 the target Client then sends a solicited NA message to the link-layer 2181 address of the source Client; otherwise, it sends the solicited NA 2182 message to the link-layer address of one of its Servers. 2184 When a source Client receives a solicited NA message from a target 2185 Client, it resets ForwardTime to FORWARD_TIME if a neighbor cache 2186 entry exists; otherwise, it discards the NA message. 2188 When ForwardTime for a dynamic neighbor cache entry expires, the 2189 source Client resumes sending any subsequent packets via a Server and 2190 may (eventually) attempt to re-initiate the AERO redirection process. 2191 When AcceptTime for a dynamic neighbor cache entry expires, the 2192 target Client discards any subsequent packets received directly from 2193 the source Client. When both ForwardTime and AcceptTime for a 2194 dynamic neighbor cache entry expire, the Client deletes the neighbor 2195 cache entry. 2197 3.19. Mobility Management 2199 3.19.1. Announcing Link-Layer Address Changes 2201 When a Client needs to change its link-layer address, e.g., due to a 2202 mobility event, it performs an immediate DHCPv6 Rebind/Reply exchange 2203 via each of its Servers using the new link-layer address as the 2204 source address and with an ALREQ that includes the correct Link ID 2205 and DSCP values. If authentication succeeds, the Server then update 2206 its neighbor cache and sends a DHCPv6 Reply. Note that if the Client 2207 does not issue a DHCPv6 Rebind before the prefix delegation lifetime 2208 expires (e.g., if the Client has been out of touch with the Server 2209 for a considerable amount of time), the Server's Reply will report 2210 NoBinding and the Client must re-initiate the DHCPv6 PD procedure. 2212 Next, the Client sends unsolicited NA messages to each of its 2213 correspondent Client neighbors using the same procedures as specified 2214 in Section 7.2.6 of [RFC4861], except that it sends the messages as 2215 unicast to each neighbor via a Server instead of multicast. In this 2216 process, the Client should send no more than 2217 MAX_NEIGHBOR_ADVERTISEMENT messages separated by no less than 2218 RETRANS_TIMER seconds to each neighbor. 2220 With reference to Figure 9, when Client ('C2') needs to change its 2221 link-layer address it sends unicast unsolicited NA messages to Client 2222 ('C1') via Server ('S2') as follows: 2224 o the link-layer source address is set to 'L2(C2)' (i.e., the link- 2225 layer address of Client ('C2')). 2227 o the link-layer destination address is set to 'L2(S2)' (i.e., the 2228 link-layer address of Server ('S2')). 2230 o the network-layer source address is set to fe80::2001:db8:1:0 2231 (i.e., the AERO address of Client ('C2')). 2233 o the network-layer destination address is set to fe80::2001:db8:0:0 2234 (i.e., the AERO address of Client ('C1')). 2236 o the Type is set to 136. 2238 o the Code is set to 0. 2240 o the Solicited flag is set to 0. 2242 o the Override flag is set to 1. 2244 o the Target Address is set to fe80::2001:db8:1:0 (i.e., the AERO 2245 address of Client ('C2')). 2247 o the message includes one or more TLLAOs with Link ID and DSCPs set 2248 to appropriate values for Client ('C2')'s underlying interfaces, 2249 and with UDP Port Number and IP Address set to '0'. 2251 o the message SHOULD include a Timestamp option. 2253 When Server ('S1') receives the NA message, it relays the message in 2254 the same way as described for relaying Redirect messages in 2255 Section 3.17.7. In particular, Server ('S1') copies the correct UDP 2256 port numbers and IP addresses into the TLLAOs, changes the link-layer 2257 source address to its own address, changes the link-layer destination 2258 address to the address of Relay ('R1'), then forwards the NA message 2259 via the relaying chain the same as for a Redirect. 2261 When Client ('C1') receives the NA message, it accepts the message 2262 only if it already has a neighbor cache entry for Client ('C2') then 2263 updates the link-layer addresses for Client ('C2') based on the 2264 addresses in the TLLAOs. Next, Client ('C1') SHOULD initiate the NUD 2265 procedures specified in Section 3.18 to provide Client ('C2') with an 2266 indication that the link-layer source address has been updated, and 2267 to refresh ('C2')'s AcceptTime and ('C1')'s ForwardTime timers. 2269 If Client ('C2') receives an NS message from Client ('C1') indicating 2270 that an unsolicited NA has updated its neighbor cache, Client ('C2') 2271 need not send additional unsolicited NAs. If Client ('C2')'s 2272 unsolicited NA messages are somehow lost, however, Client ('C1') will 2273 soon learn of the mobility event via NUD. 2275 3.19.2. Bringing New Links Into Service 2277 When a Client needs to bring a new underlying interface into service 2278 (e.g., when it activates a new data link), it performs an immediate 2279 Renew/Reply exchange via each of its Servers using the new link-layer 2280 address as the source address and with an ALREQ that includes the new 2281 Link ID and DSCP values. If authentication succeeds, the Server then 2282 updates its neighbor cache and sends a DHCPv6 Reply. The Client MAY 2283 then send unsolicited NA messages to each of its correspondent 2284 Clients to inform them of the new link-layer address as described in 2285 Section 3.19.1. 2287 3.19.3. Removing Existing Links from Service 2289 When a Client needs to remove an existing underlying interface from 2290 service (e.g., when it de-activates an existing data link), it 2291 performs an immediate Renew/Reply exchange via each of its Servers 2292 over any available link with an ALDEL that includes the deprecated 2293 Link ID. If authentication succeeds, the Server then updates its 2294 neighbor cache and sends a DHCPv6 Reply. The Client SHOULD then send 2295 unsolicited NA messages to each of its correspondent Clients to 2296 inform them of the deprecated link-layer address as described in 2297 Section 3.19.1. 2299 3.19.4. Moving to a New Server 2301 When a Client associates with a new Server, it performs the Client 2302 procedures specified in Section 3.15.2. 2304 When a Client disassociates with an existing Server, it sends a 2305 DHCPv6 Release message via a new Server to the unicast link-local 2306 network layer address of the old Server. The new Server then writes 2307 its own link-layer address in the DHCPv6 Release message IP source 2308 address and forwards the message to the old Server. 2310 When the old Server receives the DHCPv6 Release, it first 2311 authenticates the message. The Server then resets the Client's 2312 neighbor cache entry lifetime to 5 seconds, rewrites the link-layer 2313 address in the neighbor cache entry to the address of the new Server, 2314 then returns a DHCPv6 Reply message to the Client via the old Server. 2315 When the lifetime expires, the old Server withdraws the IP route from 2316 the AERO routing system and deletes the neighbor cache entry for the 2317 Client. The Client can then use the Reply message to verify that the 2318 termination signal has been processed, and can delete both the 2319 default route and the neighbor cache entry for the old Server. (Note 2320 that since Release/Reply messages may be lost in the network the 2321 Client MUST retry until it gets Reply indicating that the Release was 2322 successful.) 2324 Clients SHOULD NOT move rapidly between Servers in order to avoid 2325 causing excessive oscillations in the AERO routing system. Such 2326 oscillations could result in intermittent reachability for the Client 2327 itself, while causing little harm to the network. Examples of when a 2328 Client might wish to change to a different Server include a Server 2329 that has gone unreachable, topological movements of significant 2330 distance, etc. 2332 3.20. Proxy AERO 2334 Proxy Mobile IPv6 (PMIPv6) [RFC5213][RFC5844][RFC5949] presents a 2335 localized mobility management scheme for use within an access network 2336 domain. It is typically used in WiFi and cellular wireless access 2337 networks, and allows Mobile Nodes (MNs) to receive and retain an IP 2338 address that remains stable within the access network domain without 2339 needing to implement any special mobility protocols. In the PMIPv6 2340 architecture, access network devices known as Mobility Access 2341 Gateways (MAGs) provide MNs with an access link abstraction and 2342 receive prefixes for the MNs from a Local Mobility Anchor (LMA). 2344 In a proxy AERO domain, a proxy AERO Client (acting as a MAG) can 2345 similarly provide proxy services for MNs that do not participate in 2346 AERO messaging. The proxy Client presents an access link abstraction 2347 to MNs, and performs DHCPv6 PD exchanges over the AERO interface with 2348 an AERO Server (acting as an LMA) to receive ACPs for address 2349 provisioning of new MNs that come onto an access link. This scheme 2350 assumes that proxy Clients act as fixed (non-mobile) infrastructure 2351 elements under the same administrative trust basis as for Relays and 2352 Servers. 2354 When an MN comes onto an access link within a proxy AERO domain for 2355 the first time, the proxy Client authenticates the MN and obtains a 2356 unique identifier that it can use as a DHCPv6 DUID then issues a 2357 DHCPv6 PD Request to its Server. When the Server delegates an ACP, 2358 the proxy Client creates an AERO address for the MN and assigns the 2359 ACP to the MN's access link. The proxy Client then configures itself 2360 as a default router for the MN and provides address autoconfiguration 2361 services (e.g., SLAAC, DHCPv6, DHCPv4, etc.) for provisioning MN 2362 addresses from the ACP over the access link. Since the proxy Client 2363 may serve many such MNs simultaneously, it may receive multiple ACP 2364 prefix delegations and configure multiple AERO addresses, i.e., one 2365 for each MN. 2367 When two MNs are associated with the same proxy Client, the Client 2368 can forward traffic between the MNs without involving a Server since 2369 it configures the AERO addresses of both MNs and therefore also has 2370 the necessary routing information. When two MNs are associated with 2371 different proxy Clients, the source MN's Client can initiate standard 2372 AERO route optimization to discover a direct path to the target MN's 2373 Client through the exchange of Predirect/Redirect messages. 2375 When an MN in a proxy AERO domain leaves an access link provided by 2376 an old proxy Client, the MN issues an access link-specific "leave" 2377 message that informs the old Client of the link-layer address of a 2378 new Client on the planned new access link. This is known as a 2379 "predictive handover". When an MN comes onto an access link provided 2380 by a new proxy Client, the MN issues an access link-specific "join" 2381 message that informs the new Client of the link-layer address of the 2382 old Client on the actual old access link. This is known as a 2383 "reactive handover". 2385 Upon receiving a predictive handover indication, the old proxy Client 2386 sends a DHCPv6 PD Request message directly to the new Client and 2387 queues any arriving data packets addressed to the departed MN. The 2388 Request message includes the MN's ID as the DUID, the ACP in an IA_PD 2389 option, the old Client's address as the link-layer source address and 2390 the new Client's address as the link-layer destination address. When 2391 the new Client receives the Request message, it changes the link- 2392 layer source address to its own address, changes the link-layer 2393 destination address to the address of its Server, and forwards the 2394 message to the Server. At the same time, the new Client creates 2395 access link state for the ACP in anticipation of the MN's arrival 2396 (while queuing any data packets until the MN arrives), creates a 2397 neighbor cache entry for the old Client with AcceptTime set to 2398 ACCEPT_TIME, then sends a Redirect message back to the old Client. 2399 When the old Client receives the Redirect message, it creates a 2400 neighbor cache entry for the new Client with ForwardTime set to 2401 FORWARD_TIME, then forwards any queued data packets to the new 2402 Client. At the same time, the old Client sends a DHCPv6 PD Release 2403 message to its Server. Finally, the old Client sends unsolicited NA 2404 messages to any of the ACP's correspondents with a TLLAO containing 2405 the link-layer address of the new Client. This follows the procedure 2406 specified in Section 3.19.1, except that it is the old Client and not 2407 the Server that supplies the link-layer address. 2409 Upon receiving a reactive handover indication, the new proxy Client 2410 creates access link state for the MN's ACP, sends a DHCPv6 PD Request 2411 message to its Server, and sends a DHCPv6 PD Release message directly 2412 to the old Client. The Release message includes the MN's ID as the 2413 DUID, the ACP in an IA_PD option, the new Client's address as the 2414 link-layer source address and the old Client's address as the link- 2415 layer destination address. When the old Client receives the Release 2416 message, it changes the link-layer source address to its own address, 2417 changes the link-layer destination address to the address of its 2418 Server, and forwards the message to the Server. At the same time, 2419 the old Client sends a Predirect message back to the new Client and 2420 queues any arriving data packets addressed to the departed MN. When 2421 the new Client receives the Predirect, it creates a neighbor cache 2422 entry for the old Client with AcceptTime set to ACCEPT_TIME, then 2423 sends a Redirect message back to the old Client. When the old Client 2424 receives the Redirect message, it creates a neighbor cache entry for 2425 the new Client with ForwardTime set to FORWARD_TIME, then forwards 2426 any queued data packets to the new Client. Finally, the old Client 2427 sends unsolicited NA messages to correspondents the same as for the 2428 predictive case. 2430 When a Server processes a DHCPv6 Request message, it creates a 2431 neighbor cache entry for this ACP if none currently exists. If a 2432 neighbor cache entry already exists, however, the Server changes the 2433 link-layer address to the address of the new proxy Client (this 2434 satisfies the case of both the old Client and new Client using the 2435 same Server). 2437 When a Server processes a DHCPv6 Release message, it resets the 2438 neighbor cache entry lifetime for this ACP to 5 seconds if the cached 2439 link-layer address matches the old proxy Client's address. 2440 Otherwise, the Server ignores the Release message (this satisfies the 2441 case of both the old Client and new Client using the same Server). 2443 When a correspondent Client receives an unsolicited NA message, it 2444 changes the link-layer address for the ACP's neighbor cache entry to 2445 the address of the new proxy Client. The correspondent Client then 2446 issues a Predirect/Redirect exchange to establish a new neighbor 2447 cache entry in the new Client. 2449 From an architectural perspective, in addition to the use of DHCPv6 2450 PD and IPv6 ND signaling the AERO approach differs from PMIPv6 in its 2451 use of the NBMA virtual link model instead of point-to-point tunnels. 2452 This provides a more agile interface for Client/Server and Client/ 2453 Client coordinations, and also facilitates simple route optimization. 2454 The AERO routing system is also arranged in such a fashion that 2455 Clients get the same service from any Server they happen to associate 2456 with. This provides a natural fault tolerance and load balancing 2457 capability such as desired for distributed mobility management. 2459 3.21. Extending AERO Links Through Security Gateways 2461 When an enterprise mobile device moves from a campus LAN connection 2462 to a public Internet link, it must re-enter the enterprise via a 2463 security gateway that has both a physical interface connection to the 2464 Internet and a physical interface connection to the enterprise 2465 internetwork. This most often entails the establishment of a Virtual 2466 Private Network (VPN) link over the public Internet from the mobile 2467 device to the security gateway. During this process, the mobile 2468 device supplies the security gateway with its public Internet address 2469 as the link-layer address for the VPN. The mobile device then acts 2470 as an AERO Client to negotiate with the security gateway to obtain 2471 its ACP. 2473 In order to satisfy this need, the security gateway also operates as 2474 an AERO Server with support for AERO Client proxying. In particular, 2475 when a mobile device (i.e., the Client) connects via the security 2476 gateway (i.e., the Server), the Server provides the Client with an 2477 ACP in a DHCPv6 PD exchange the same as if it were attached to an 2478 enterprise campus access link. The Server then replaces the Client's 2479 link-layer source address with the Server's enterprise-facing link- 2480 layer address in all AERO messages the Client sends toward neighbors 2481 on the AERO link. The AERO messages are then delivered to other 2482 devices on the AERO link as if they were originated by the security 2483 gateway instead of by the AERO Client. In the reverse direction, the 2484 AERO messages sourced by devices within the enterprise network can be 2485 forwarded to the security gateway, which then replaces the link-layer 2486 destination address with the Client's link-layer address and replaces 2487 the link-layer source address with its own (Internet-facing) link- 2488 layer address. 2490 After receiving the ACP, the Client can send IP packets that use an 2491 address taken from the ACP as the network layer source address, the 2492 Client's link-layer address as the link-layer source address, and the 2493 Server's Internet-facing link-layer address as the link-layer 2494 destination address. The Server will then rewrite the link-layer 2495 source address with the Server's own enterprise-facing link-layer 2496 address and rewrite the link-layer destination address with the 2497 target AERO node's link-layer address, and the packets will enter the 2498 enterprise network as though they were sourced from a device located 2499 within the enterprise. In the reverse direction, when a packet 2500 sourced by a node within the enterprise network uses a destination 2501 address from the Client's ACP, the packet will be delivered to the 2502 security gateway which then rewrites the link-layer destination 2503 address to the Client's link-layer address and rewrites the link- 2504 layer source address to the Server's Internet-facing link-layer 2505 address. The Server then delivers the packet across the VPN to the 2506 AERO Client. In this way, the AERO virtual link is essentially 2507 extended *through* the security gateway to the point at which the VPN 2508 link and AERO link are effectively grafted together by the link-layer 2509 address rewriting performed by the security gateway. All AERO 2510 messaging services (including route optimization and mobility 2511 signaling) are therefore extended to the Client. 2513 In order to support this virtual link grafting, the security gateway 2514 (acting as an AERO Server) must keep static neighbor cache entries 2515 for all of its associated Clients located on the public Internet. 2516 The neighbor cache entry is keyed by the AERO Client's AERO address 2517 the same as if the Client were located within the enterprise 2518 internetwork. The neighbor cache is then managed in all ways as 2519 though the Client were an ordinary AERO Client. This includes the 2520 AERO IPv6 ND messaging signaling for Route Optimization and Neighbor 2521 Unreachability Detection. 2523 Note that the main difference between a security gateway acting as an 2524 AERO Server and an enterprise-internal AERO Server is that the 2525 security gateway has at least one enterprise-internal physical 2526 interface and at least one public Internet physical interface. 2527 Conversely, the enterprise-internal AERO Server has only enterprise- 2528 internal physical interfaces. For this reason security gateway 2529 proxying is needed to ensure that the public Internet link-layer 2530 addressing space is kept separate from the enterprise-internal link- 2531 layer addressing space. This is afforded through a natural extension 2532 of the security association caching already performed for each VPN 2533 client by the security gateway. 2535 3.22. Extending IPv6 AERO Links to the Internet 2537 When an IPv6 host ('H1') with an address from an ACP owned by AERO 2538 Client ('C1') sends packets to a correspondent IPv6 host ('H2'), the 2539 packets eventually arrive at the IPv6 router that owns ('H2')s 2540 prefix. This IPv6 router may or may not be an AERO Client ('C2') 2541 either within the same home network as ('C1') or in a different home 2542 network. 2544 If Client ('C1') is currently located outside the boundaries of its 2545 home network, it will connect back into the home network via a 2546 security gateway acting as an AERO Server. The packets sent by 2547 ('H1') via ('C1') will then be forwarded through the security gateway 2548 then through the home network and finally to ('C2') where they will 2549 be delivered to ('H2'). This could lead to sub-optimal performance 2550 when ('C2') could instead be reached via a more direct route without 2551 involving the security gateway. 2553 Consider the case when host ('H1') has the IPv6 address 2554 2001:db8:1::1, and Client ('C1') has the ACP 2001:db8:1::/64 with 2555 underlying IPv6 Internet address of 2001:db8:1000::1. Also, host 2556 ('H2') has the IPv6 address 2001:db8:2::1, and Client ('C2') has the 2557 ACP 2001:db8:2::/64 with underlying IPv6 address of 2001:db8:2000::1. 2558 Client ('C1') can determine whether 'C2' is indeed also an AERO 2559 Client willing to serve as a route optimization correspondent by 2560 resolving the AAAA records for the DNS FQDN that matches ('H2')s 2561 prefix, i.e.: 2563 '0.0.0.0.2.0.0.0.8.b.d.0.1.0.0.2.aero.linkupnetworks.net' 2565 If ('C2') is indeed a candidate correspondent, the FQDN lookup will 2566 return a PTR resource record that contains the domain name for the 2567 AERO link that manages ('C2')s ASP. Client ('C1') can then attempt 2568 route optimization using an approach similar to the Return 2569 Routability procedure specified for Mobile IPv6 (MIPv6) [RFC6275]. 2570 In order to support this process, both Clients MUST intercept and 2571 decapsulate packets that have a subnet router anycast address 2572 corresponding to any of the /64 prefixes covered by their respective 2573 ACPs. 2575 To initiate the process, Client ('C1') creates a specially-crafted 2576 encapsulated AERO Predirect message that will be routed through its 2577 home network then through ('C2')s home network and finally to ('C2') 2578 itself. Client ('C1') prepares the initial message in the exchange 2579 as follows: 2581 o The encapsulating IPv6 header source address is set to 2582 2001:db8:1:: (i.e., the IPv6 subnet router anycast address for 2583 ('C1')s ACP) 2585 o The encapsulating IPv6 header destination address is set to 2586 2001:db8:2:: (i.e., the IPv6 subnet router anycast address for 2587 ('C2')s ACP) 2589 o The encapsulating IPv6 header is followed by a UDP header with 2590 source and destination port set to 8060 2592 o The encapsulated IPv6 header source address is set to 2593 fe80::2001:db8:1:0 (i.e., the AERO address for ('C1')) 2595 o The encapsulated IPv6 header destination address is set to 2596 fe80::2001:db8:2:0 (i.e., the AERO address for ('C2')) 2598 o The encapsulated AERO Predirect message includes all of the 2599 securing information that would occur in a MIPv6 "Home Test Init" 2600 message (format TBD) 2602 Client ('C1') then further encapsulates the message in the 2603 encapsulating headers necessary to convey the packet to the security 2604 gateway (e.g., through IPsec encapsulation) so that the message now 2605 appears "double-encapsulated". ('C1') then sends the message to the 2606 security gateway, which re-encapsulates and forwards it over the home 2607 network from where it will eventually reach ('C2'). 2609 At the same time, ('C1') creates and sends a second encapsulated AERO 2610 Predirect message that will be routed through the IPv6 Internet 2611 without involving the security gateway. Client ('C1') prepares the 2612 message as follows: 2614 o The encapsulating IPv6 header source address is set to 2615 2001:db8:1000:1 (i.e., the Internet IPv6 address of ('C1')) 2617 o The encapsulating IPv6 header destination address is set to 2618 2001:db8:2:: (i.e., the IPv6 subnet router anycast address for 2619 ('C2')s ACP) 2621 o The encapsulating IPv6 header is followed by a UDP header with 2622 source and destination port set to 8060 2624 o The encapsulated IPv6 header source address is set to 2625 fe80::2001:db8:1:0 (i.e., the AERO address for ('C1')) 2627 o The encapsulated IPv6 header destination address is set to 2628 fe80::2001:db8:2:0 (i.e., the AERO address for ('C2')) 2630 o The encapsulated AERO Predirect message includes all of the 2631 securing information that would occur in a MIPv6 "Care-of Test 2632 Init" message (format TBD) 2634 ('C2') will receive both Predirect messages through its home network 2635 then return a corresponding Redirect for each of the Predirect 2636 messages with the source and destination addresses in the inner and 2637 outer headers reversed. The first message includes all of the 2638 securing information that would occur in a MIPv6 "Home Test" message, 2639 while the second message includes all of the securing information 2640 that would occur in a MIPv6 "Care-of Test" message (formats TBD). 2642 When ('C1') receives the Redirect messages, it performs the necessary 2643 security procedures per the MIPv6 specification. It then prepares an 2644 encapsulated NS message that includes the same source and destination 2645 addresses as for the "Care-of Test Init" Predirect message, and 2646 includes all of the securing information that would occur in a MIPv6 2647 "Binding Update" message (format TBD) and sends the message to 2648 ('C2'). 2650 When ('C2') receives the NS message, if the securing information is 2651 correct it creates or updates a neighbor cache entry for ('C1') with 2652 fe80::2001:db8:1:0 as the network-layer address, 2001:db8:1000::1 as 2653 the link-layer address and with AcceptTime set to ACCEPT_TIME. 2654 ('C2') then sends an encapsulated NA message back to ('C1') that 2655 includes the same source and destination addresses as for the "Care- 2656 of Test" Redirect message, and includes all of the securing 2657 information that would occur in a MIPv6 "Binding Acknowledgement" 2658 message (format TBD) and sends the message to ('C1'). 2660 When ('C1') receives the NA message, it creates or updates a neighbor 2661 cache entry for ('C2') with fe80::2001:db8:2:0 as the network-layer 2662 address and 2001:db8:2:: as the link-layer address and with 2663 ForwardTime set to FORWARD_TIME, thus completing the route 2664 optimization in the forward direction. 2666 ('C1') subsequently forwards encapsulated packets with outer source 2667 address 2001:db8:1000::1, with outer destination address 2668 2001:db8:2::, with inner source address taken from the 2001:db8:1::, 2669 and with inner destination address taken from 2001:db8:2:: due to the 2670 fact that it has a securely-established neighbor cache entry with 2671 non-zero ForwardTime. ('C2') subsequently accepts any such 2672 encapsulated packets due to the fact that it has a securely- 2673 established neighbor cache entry with non-zero AcceptTime. 2675 In order to keep neighbor cache entries alive, ('C1') periodically 2676 sends additional NS messages to ('C2') and receives any NA responses. 2677 If ('C1') moves to a different point of attachment after the initial 2678 route optimization, it sends a new secured NS message to ('C2') as 2679 above to update ('C2')s neighbor cache. 2681 If ('C2') has packets to send to ('C1'), it performs a corresponding 2682 route optimization in the opposite direction following the same 2683 procedures described above. In the process, the already-established 2684 unidirectional neighbor cache entries within ('C1') and ('C2') are 2685 updated to include the now-bidirectional information. In particular, 2686 the AcceptTime and ForwardTime variables for both neighbor cache 2687 entries are updated to non-zero values, and the link-layer address 2688 for ('C1')s neighbor cache entry for ('C2') is reset to 2689 2001:db8:2000::1. 2691 Note that two AERO Clients can use full security protocol messaging 2692 instead of Return Routability, e.g., if strong authentication and/or 2693 confidentiality are desired. In that case, security protocol key 2694 exchanges such as specified for MOBIKE [RFC4555] would be used to 2695 establish security associations and neighbor cache entries between 2696 the AERO clients. Thereafter, AERO NS/NA messaging can be used to 2697 maintain neighbor cache entries, test reachability, and to announce 2698 mobility events. If reachability testing fails, e.g., if both 2699 Clients move at roughly the same time, the Clients can tear down the 2700 security association and neighbor cache entries and again allow 2701 packets to flow through their home network. 2703 3.23. Encapsulation Protocol Version Considerations 2705 A source Client may connect only to an IPvX underlying network, while 2706 the target Client connects only to an IPvY underlying network. In 2707 that case, the target and source Clients have no means for reaching 2708 each other directly (since they connect to underlying networks of 2709 different IP protocol versions) and so must ignore any redirection 2710 messages and continue to send packets via the Server. 2712 3.24. Multicast Considerations 2714 When the underlying network does not support multicast, AERO nodes 2715 map IPv6 link-scoped multicast addresses (including 2716 'All_DHCP_Relay_Agents_and_Servers') to the link-layer address of a 2717 Server. 2719 When the underlying network supports multicast, AERO nodes use the 2720 multicast address mapping specification found in [RFC2529] for IPv4 2721 underlying networks and use a direct multicast mapping for IPv6 2722 underlying networks. (In the latter case, "direct multicast mapping" 2723 means that if the IPv6 multicast destination address of the 2724 encapsulated packet is "M", then the IPv6 multicast destination 2725 address of the encapsulating header is also "M".) 2727 3.25. Operation on AERO Links Without DHCPv6 Services 2729 When Servers on the AERO link do not provide DHCPv6 services, 2730 operation can still be accommodated through administrative 2731 configuration of ACPs on AERO Clients. In that case, administrative 2732 configurations of AERO interface neighbor cache entries on both the 2733 Server and Client are also necessary. However, this may interfere 2734 with the ability for Clients to dynamically change to new Servers, 2735 and can expose the AERO link to misconfigurations unless the 2736 administrative configurations are carefully coordinated. 2738 3.26. Operation on Server-less AERO Links 2740 In some AERO link scenarios, there may be no Servers on the link and/ 2741 or no need for Clients to use a Server as an intermediary trust 2742 anchor. In that case, each Client acts as a Server unto itself to 2743 establish neighbor cache entries by performing direct Client-to- 2744 Client IPv6 ND message exchanges, and some other form of trust basis 2745 must be applied so that each Client can verify that the prospective 2746 neighbor is authorized to use its claimed ACP. 2748 When there is no Server on the link, Clients must arrange to receive 2749 ACPs and publish them via a secure alternate prefix delegation 2750 authority through some means outside the scope of this document. 2752 3.27. Manually-Configured AERO Tunnels 2754 In addition to the dynamic neighbor discovery procedures for AERO 2755 link neighbors described above, AERO encapsulation can be applied to 2756 manually-configured tunnels. In that case, the tunnel endpoints use 2757 an administratively-assigned link-local address and exchange NS/NA 2758 messages the same as for dynamically-established tunnels. 2760 3.28. Intradomain Routing 2762 After a tunnel neighbor relationship has been established, neighbors 2763 can use a traditional dynamic routing protocol over the tunnel to 2764 exchange routing information without having to inject the routes into 2765 the AERO routing system. 2767 4. Implementation Status 2769 User-level and kernel-level AERO implementations have been developed 2770 and are undergoing internal testing within Boeing. 2772 5. Next Steps 2774 A new Generic UDP Encapsulation (GUE) format has been specified in 2775 [I-D.herbert-gue-fragmentation] [I-D.ietf-nvo3-gue]. The GUE 2776 encapsulation format will eventually supplant the native AERO UDP 2777 encapsulation format. 2779 Future versions of the spec will explore the subject of DSCP marking 2780 in more detail. 2782 6. IANA Considerations 2784 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 2785 AERO in the "enterprise-numbers" registry. 2787 The IANA has assigned the UDP port number "8060" for an earlier 2788 experimental version of AERO [RFC6706]. This document obsoletes 2789 [RFC6706] and claims the UDP port number "8060" for all future use. 2791 No further IANA actions are required. 2793 7. Security Considerations 2795 AERO link security considerations are the same as for standard IPv6 2796 Neighbor Discovery [RFC4861] except that AERO improves on some 2797 aspects. In particular, AERO uses a trust basis between Clients and 2798 Servers, where the Clients only engage in the AERO mechanism when it 2799 is facilitated by a trust anchor. Unless there is some other means 2800 of authenticating the Client's identity (e.g., link-layer security), 2801 AERO nodes SHOULD also use DHCPv6 securing services (e.g., DHCPv6 2802 authentication, Secure DHCPv6 [I-D.ietf-dhc-sedhcpv6], etc.) for 2803 Client authentication and network admission control. 2805 AERO Redirect, Predirect and unsolicited NA messages SHOULD include a 2806 Timestamp option (see Section 5.3 of [RFC3971]) that other AERO nodes 2807 can use to verify the message time of origin. AERO Predirect, NS and 2808 RS messages SHOULD include a Nonce option (see Section 5.3 of 2809 [RFC3971]) that recipients echo back in corresponding responses. 2811 AERO links must be protected against link-layer address spoofing 2812 attacks in which an attacker on the link pretends to be a trusted 2813 neighbor. Links that provide link-layer securing mechanisms (e.g., 2814 IEEE 802.1X WLANs) and links that provide physical security (e.g., 2815 enterprise network wired LANs) provide a first line of defense that 2816 is often sufficient. In other instances, additional securing 2817 mechanisms such as Secure Neighbor Discovery (SeND) [RFC3971], IPsec 2818 [RFC4301] or TLS [RFC5246] may be necessary. 2820 AERO Clients MUST ensure that their connectivity is not used by 2821 unauthorized nodes on their EUNs to gain access to a protected 2822 network, i.e., AERO Clients that act as routers MUST NOT provide 2823 routing services for unauthorized nodes. (This concern is no 2824 different than for ordinary hosts that receive an IP address 2825 delegation but then "share" the address with unauthorized nodes via a 2826 NAT function.) 2828 On some AERO links, establishment and maintenance of a direct path 2829 between neighbors requires secured coordination such as through the 2830 Internet Key Exchange (IKEv2) protocol [RFC5996] to establish a 2831 security association. 2833 An AERO Client's link-layer address could be rewritten by a link- 2834 layer switching element on the path from the Client to the Server and 2835 not detected by the DHCPv6 security mechanism. However, such a 2836 condition would only be a matter of concern on unmanaged/unsecured 2837 links where the link-layer switching elements themselves present a 2838 man-in-the-middle attack threat. For this reason, IP security MUST 2839 be used when AERO is employed over unmanaged/unsecured links. 2841 8. Acknowledgements 2843 Discussions both on IETF lists and in private exchanges helped shape 2844 some of the concepts in this work. Individuals who contributed 2845 insights include Mikael Abrahamsson, Mark Andrews, Fred Baker, 2846 Stewart Bryant, Brian Carpenter, Wojciech Dec, Ralph Droms, Adrian 2847 Farrel, Sri Gundavelli, Brian Haberman, Joel Halpern, Tom Herbert, 2848 Sascha Hlusiak, Lee Howard, Andre Kostur, Ted Lemon, Andy Malis, 2849 Satoru Matsushima, Tomek Mrugalski, Alexandru Petrescu, Behcet 2850 Saikaya, Joe Touch, Bernie Volz, Ryuji Wakikawa and Lloyd Wood. 2851 Members of the IESG also provided valuable input during their review 2852 process that greatly improved the document. Special thanks go to 2853 Stewart Bryant, Joel Halpern and Brian Haberman for their shepherding 2854 guidance. 2856 This work has further been encouraged and supported by Boeing 2857 colleagues including Dave Bernhardt, Cam Brodie, Balaguruna 2858 Chidambaram, Bruce Cornish, Claudiu Danilov, Wen Fang, Anthony 2859 Gregory, Jeff Holland, Ed King, Gen MacLean, Rob Muszkiewicz, Sean 2860 O'Sullivan, Kent Shuey, Brian Skeen, Mike Slane, Brendan Williams, 2861 Julie Wulff, Yueli Yang, and other members of the BR&T and BIT mobile 2862 networking teams. 2864 Earlier works on NBMA tunneling approaches are found in 2865 [RFC2529][RFC5214][RFC5569]. 2867 Many of the constructs presented in this second edition of AERO are 2868 based on the author's earlier works, including: 2870 o The Internet Routing Overlay Network (IRON) 2871 [RFC6179][I-D.templin-ironbis] 2873 o Virtual Enterprise Traversal (VET) 2874 [RFC5558][I-D.templin-intarea-vet] 2876 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 2877 [RFC5320][I-D.templin-intarea-seal] 2879 o AERO, First Edition [RFC6706] 2881 Note that these works cite numerous earlier efforts that are not also 2882 cited here due to space limitations. The authors of those earlier 2883 works are acknowledged for their insights. 2885 9. References 2887 9.1. Normative References 2889 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 2890 August 1980. 2892 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, September 2893 1981. 2895 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 2896 RFC 792, September 1981. 2898 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 2899 October 1996. 2901 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2902 Requirement Levels", BCP 14, RFC 2119, March 1997. 2904 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 2905 (IPv6) Specification", RFC 2460, December 1998. 2907 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 2908 IPv6 Specification", RFC 2473, December 1998. 2910 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 2911 "Definition of the Differentiated Services Field (DS 2912 Field) in the IPv4 and IPv6 Headers", RFC 2474, December 2913 1998. 2915 [RFC3315] Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C., 2916 and M. Carney, "Dynamic Host Configuration Protocol for 2917 IPv6 (DHCPv6)", RFC 3315, July 2003. 2919 [RFC3633] Troan, O. and R. Droms, "IPv6 Prefix Options for Dynamic 2920 Host Configuration Protocol (DHCP) version 6", RFC 3633, 2921 December 2003. 2923 [RFC3971] Arkko, J., Kempf, J., Zill, B., and P. Nikander, "SEcure 2924 Neighbor Discovery (SEND)", RFC 3971, March 2005. 2926 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 2927 for IPv6 Hosts and Routers", RFC 4213, October 2005. 2929 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 2930 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 2931 September 2007. 2933 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 2934 Address Autoconfiguration", RFC 4862, September 2007. 2936 [RFC6434] Jankiewicz, E., Loughney, J., and T. Narten, "IPv6 Node 2937 Requirements", RFC 6434, December 2011. 2939 9.2. Informative References 2941 [I-D.herbert-gue-fragmentation] 2942 Herbert, T. and F. Templin, "Fragmentation option for 2943 Generic UDP Encapsulation", draft-herbert-gue- 2944 fragmentation-00 (work in progress), March 2015. 2946 [I-D.ietf-dhc-sedhcpv6] 2947 Jiang, S., Shen, S., Zhang, D., and T. Jinmei, "Secure 2948 DHCPv6", draft-ietf-dhc-sedhcpv6-07 (work in progress), 2949 March 2015. 2951 [I-D.ietf-nvo3-gue] 2952 Herbert, T., Yong, L., and O. Zia, "Generic UDP 2953 Encapsulation", draft-ietf-nvo3-gue-00 (work in progress), 2954 April 2015. 2956 [I-D.templin-intarea-seal] 2957 Templin, F., "The Subnetwork Encapsulation and Adaptation 2958 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 2959 progress), January 2014. 2961 [I-D.templin-intarea-vet] 2962 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 2963 templin-intarea-vet-40 (work in progress), May 2013. 2965 [I-D.templin-ironbis] 2966 Templin, F., "The Interior Routing Overlay Network 2967 (IRON)", draft-templin-ironbis-16 (work in progress), 2968 March 2014. 2970 [I-D.vandevelde-idr-remote-next-hop] 2971 Velde, G., Patel, K., Rao, D., Raszuk, R., and R. Bush, 2972 "BGP Remote-Next-Hop", draft-vandevelde-idr-remote-next- 2973 hop-09 (work in progress), March 2015. 2975 [RFC0879] Postel, J., "TCP maximum segment size and related topics", 2976 RFC 879, November 1983. 2978 [RFC1035] Mockapetris, P., "Domain names - implementation and 2979 specification", STD 13, RFC 1035, November 1987. 2981 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 2982 November 1990. 2984 [RFC1812] Baker, F., "Requirements for IP Version 4 Routers", RFC 2985 1812, June 1995. 2987 [RFC1930] Hawkinson, J. and T. Bates, "Guidelines for creation, 2988 selection, and registration of an Autonomous System (AS)", 2989 BCP 6, RFC 1930, March 1996. 2991 [RFC1981] McCann, J., Deering, S., and J. Mogul, "Path MTU Discovery 2992 for IP version 6", RFC 1981, August 1996. 2994 [RFC2131] Droms, R., "Dynamic Host Configuration Protocol", RFC 2995 2131, March 1997. 2997 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 2998 Domains without Explicit Tunnels", RFC 2529, March 1999. 3000 [RFC2675] Borman, D., Deering, S., and R. Hinden, "IPv6 Jumbograms", 3001 RFC 2675, August 1999. 3003 [RFC2764] Gleeson, B., Heinanen, J., Lin, A., Armitage, G., and A. 3004 Malis, "A Framework for IP Based Virtual Private 3005 Networks", RFC 2764, February 2000. 3007 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 3008 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 3009 March 2000. 3011 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", RFC 3012 2923, September 2000. 3014 [RFC2983] Black, D., "Differentiated Services and Tunnels", RFC 3015 2983, October 2000. 3017 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3018 of Explicit Congestion Notification (ECN) to IP", RFC 3019 3168, September 2001. 3021 [RFC3596] Thomson, S., Huitema, C., Ksinant, V., and M. Souissi, 3022 "DNS Extensions to Support IP Version 6", RFC 3596, 3023 October 2003. 3025 [RFC3819] Karn, P., Bormann, C., Fairhurst, G., Grossman, D., 3026 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 3027 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 3028 RFC 3819, July 2004. 3030 [RFC4271] Rekhter, Y., Li, T., and S. Hares, "A Border Gateway 3031 Protocol 4 (BGP-4)", RFC 4271, January 2006. 3033 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3034 Architecture", RFC 4291, February 2006. 3036 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3037 Internet Protocol", RFC 4301, December 2005. 3039 [RFC4443] Conta, A., Deering, S., and M. Gupta, "Internet Control 3040 Message Protocol (ICMPv6) for the Internet Protocol 3041 Version 6 (IPv6) Specification", RFC 4443, March 2006. 3043 [RFC4511] Sermersheim, J., "Lightweight Directory Access Protocol 3044 (LDAP): The Protocol", RFC 4511, June 2006. 3046 [RFC4555] Eronen, P., "IKEv2 Mobility and Multihoming Protocol 3047 (MOBIKE)", RFC 4555, June 2006. 3049 [RFC4592] Lewis, E., "The Role of Wildcards in the Domain Name 3050 System", RFC 4592, July 2006. 3052 [RFC4821] Mathis, M. and J. Heffner, "Packetization Layer Path MTU 3053 Discovery", RFC 4821, March 2007. 3055 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 3056 Errors at High Data Rates", RFC 4963, July 2007. 3058 [RFC4994] Zeng, S., Volz, B., Kinnear, K., and J. Brzozowski, 3059 "DHCPv6 Relay Agent Echo Request Option", RFC 4994, 3060 September 2007. 3062 [RFC5213] Gundavelli, S., Leung, K., Devarapalli, V., Chowdhury, K., 3063 and B. Patil, "Proxy Mobile IPv6", RFC 5213, August 2008. 3065 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3066 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3067 March 2008. 3069 [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security 3070 (TLS) Protocol Version 1.2", RFC 5246, August 2008. 3072 [RFC5320] Templin, F., "The Subnetwork Encapsulation and Adaptation 3073 Layer (SEAL)", RFC 5320, February 2010. 3075 [RFC5494] Arkko, J. and C. Pignataro, "IANA Allocation Guidelines 3076 for the Address Resolution Protocol (ARP)", RFC 5494, 3077 April 2009. 3079 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3080 Route Optimization Requirements for Operational Use in 3081 Aeronautics and Space Exploration Mobile Networks", RFC 3082 5522, October 2009. 3084 [RFC5558] Templin, F., "Virtual Enterprise Traversal (VET)", RFC 3085 5558, February 2010. 3087 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3088 Infrastructures (6rd)", RFC 5569, January 2010. 3090 [RFC5720] Templin, F., "Routing and Addressing in Networks with 3091 Global Enterprise Recursion (RANGER)", RFC 5720, February 3092 2010. 3094 [RFC5844] Wakikawa, R. and S. Gundavelli, "IPv4 Support for Proxy 3095 Mobile IPv6", RFC 5844, May 2010. 3097 [RFC5949] Yokota, H., Chowdhury, K., Koodli, R., Patil, B., and F. 3098 Xia, "Fast Handovers for Proxy Mobile IPv6", RFC 5949, 3099 September 2010. 3101 [RFC5996] Kaufman, C., Hoffman, P., Nir, Y., and P. Eronen, 3102 "Internet Key Exchange Protocol Version 2 (IKEv2)", RFC 3103 5996, September 2010. 3105 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3106 NAT64: Network Address and Protocol Translation from IPv6 3107 Clients to IPv4 Servers", RFC 6146, April 2011. 3109 [RFC6179] Templin, F., "The Internet Routing Overlay Network 3110 (IRON)", RFC 6179, March 2011. 3112 [RFC6204] Singh, H., Beebee, W., Donley, C., Stark, B., and O. 3113 Troan, "Basic Requirements for IPv6 Customer Edge 3114 Routers", RFC 6204, April 2011. 3116 [RFC6221] Miles, D., Ooghe, S., Dec, W., Krishnan, S., and A. 3117 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, May 3118 2011. 3120 [RFC6241] Enns, R., Bjorklund, M., Schoenwaelder, J., and A. 3121 Bierman, "Network Configuration Protocol (NETCONF)", RFC 3122 6241, June 2011. 3124 [RFC6275] Perkins, C., Johnson, D., and J. Arkko, "Mobility Support 3125 in IPv6", RFC 6275, July 2011. 3127 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 3128 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, August 3129 2011. 3131 [RFC6422] Lemon, T. and Q. Wu, "Relay-Supplied DHCP Options", RFC 3132 6422, December 2011. 3134 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3135 for Equal Cost Multipath Routing and Link Aggregation in 3136 Tunnels", RFC 6438, November 2011. 3138 [RFC6691] Borman, D., "TCP Options and Maximum Segment Size (MSS)", 3139 RFC 6691, July 2012. 3141 [RFC6706] Templin, F., "Asymmetric Extended Route Optimization 3142 (AERO)", RFC 6706, August 2012. 3144 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID Field", 3145 RFC 6864, February 2013. 3147 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 3148 UDP Checksums for Tunneled Packets", RFC 6935, April 2013. 3150 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 3151 for the Use of IPv6 UDP Datagrams with Zero Checksums", 3152 RFC 6936, April 2013. 3154 [RFC6939] Halwasia, G., Bhandari, S., and W. Dec, "Client Link-Layer 3155 Address Option in DHCPv6", RFC 6939, May 2013. 3157 [RFC6980] Gont, F., "Security Implications of IPv6 Fragmentation 3158 with IPv6 Neighbor Discovery", RFC 6980, August 2013. 3160 [RFC7078] Matsumoto, A., Fujisaki, T., and T. Chown, "Distributing 3161 Address Selection Policy Using DHCPv6", RFC 7078, January 3162 2014. 3164 [TUNTAP] Wikipedia, W., "http://en.wikipedia.org/wiki/TUN/TAP", 3165 October 2014. 3167 Author's Address 3169 Fred L. Templin (editor) 3170 Boeing Research & Technology 3171 P.O. Box 3707 3172 Seattle, WA 98124 3173 USA 3175 Email: fltemplin@acm.org