idnits 2.17.1 draft-templin-intarea-6706bis-14.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- == The 'Obsoletes: ' line in the draft header should list only the _numbers_ of the RFCs which will be obsoleted by this document (if approved); it should not include the word 'RFC' in the list. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (June 3, 2019) is 1788 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'RFC2764' is defined on line 3145, but no explicit reference was found in the text == Unused Reference: 'RFC4607' is defined on line 3232, but no explicit reference was found in the text == Unused Reference: 'RFC7269' is defined on line 3315, but no explicit reference was found in the text == Unused Reference: 'RFC8086' is defined on line 3331, but no explicit reference was found in the text == Outdated reference: A later version (-26) exists of draft-ietf-6man-segment-routing-header-19 == Outdated reference: A later version (-15) exists of draft-ietf-dmm-distributed-mobility-anchoring-13 == Outdated reference: A later version (-09) exists of draft-ietf-intarea-gue-07 == Outdated reference: A later version (-13) exists of draft-ietf-intarea-tunnels-09 == Outdated reference: A later version (-26) exists of draft-ietf-rtgwg-atn-bgp-02 == Outdated reference: A later version (-11) exists of draft-templin-6man-dhcpv6-ndopt-07 == Outdated reference: A later version (-27) exists of draft-templin-v6ops-pdhost-23 -- Obsolete informational reference (is this intentional?): RFC 6106 (Obsoleted by RFC 8106) -- Obsolete informational reference (is this intentional?): RFC 6347 (Obsoleted by RFC 9147) Summary: 0 errors (**), 0 flaws (~~), 13 warnings (==), 3 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 3, 2019 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: December 5, 2019 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-intarea-6706bis-14.txt 13 Abstract 15 This document specifies the operation of IP over tunnel virtual links 16 using Asymmetric Extended Route Optimization (AERO). AERO interfaces 17 use an IPv6 link-local address format that supports operation of the 18 IPv6 Neighbor Discovery (ND) protocol and links ND to IP forwarding. 19 Prefix delegation services are employed for network admission and to 20 manage the routing system. Multilink operation, mobility management, 21 quality of service (QoS) signaling and route optimization are 22 naturally supported through dynamic neighbor cache updates. Standard 23 IP multicasting services are also supported. AERO is a widely- 24 applicable tunneling solution especially well-suited to aviation 25 services, mobile Virtual Private Networks (VPNs) and many other 26 applications. 28 Status of This Memo 30 This Internet-Draft is submitted in full conformance with the 31 provisions of BCP 78 and BCP 79. 33 Internet-Drafts are working documents of the Internet Engineering 34 Task Force (IETF). Note that other groups may also distribute 35 working documents as Internet-Drafts. The list of current Internet- 36 Drafts is at https://datatracker.ietf.org/drafts/current/. 38 Internet-Drafts are draft documents valid for a maximum of six months 39 and may be updated, replaced, or obsoleted by other documents at any 40 time. It is inappropriate to use Internet-Drafts as reference 41 material or to cite them other than as "work in progress." 43 This Internet-Draft will expire on December 5, 2019. 45 Copyright Notice 47 Copyright (c) 2019 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents 52 (https://trustee.ietf.org/license-info) in effect on the date of 53 publication of this document. Please review these documents 54 carefully, as they describe your rights and restrictions with respect 55 to this document. Code Components extracted from this document must 56 include Simplified BSD License text as described in Section 4.e of 57 the Trust Legal Provisions and are provided without warranty as 58 described in the Simplified BSD License. 60 Table of Contents 62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 63 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 64 3. Asymmetric Extended Route Optimization (AERO) . . . . . . . . 10 65 3.1. AERO Link Reference Model . . . . . . . . . . . . . . . . 10 66 3.2. AERO Node Types . . . . . . . . . . . . . . . . . . . . . 12 67 3.3. AERO Routing System . . . . . . . . . . . . . . . . . . . 13 68 3.4. AERO Addresses . . . . . . . . . . . . . . . . . . . . . 15 69 3.5. Spanning Partitioned AERO Networks (SPAN) . . . . . . . . 17 70 3.6. AERO Interface Characteristics . . . . . . . . . . . . . 21 71 3.7. AERO Interface Initialization . . . . . . . . . . . . . . 24 72 3.7.1. AERO Server/Gateway Behavior . . . . . . . . . . . . 25 73 3.7.2. AERO Proxy Behavior . . . . . . . . . . . . . . . . . 25 74 3.7.3. AERO Client Behavior . . . . . . . . . . . . . . . . 25 75 3.7.4. AERO Relay Behavior . . . . . . . . . . . . . . . . . 26 76 3.8. AERO Interface Neighbor Cache Maintenance . . . . . . . . 26 77 3.9. AERO Interface Encapsulation and Re-encapsulation . . . . 28 78 3.10. AERO Interface Decapsulation . . . . . . . . . . . . . . 29 79 3.11. AERO Interface Data Origin Authentication . . . . . . . . 29 80 3.12. AERO Interface Forwarding Algorithm . . . . . . . . . . . 30 81 3.12.1. Client Forwarding Algorithm . . . . . . . . . . . . 31 82 3.12.2. Proxy Forwarding Algorithm . . . . . . . . . . . . . 31 83 3.12.3. Server/Gateway Forwarding Algorithm . . . . . . . . 32 84 3.12.4. Relay Forwarding Algorithm . . . . . . . . . . . . . 33 85 3.13. AERO Interface Packet Size Issues . . . . . . . . . . . . 34 86 3.14. AERO Interface Error Handling . . . . . . . . . . . . . . 36 87 3.15. AERO Router Discovery, Prefix Delegation and 88 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 39 89 3.15.1. AERO ND/PD Service Model . . . . . . . . . . . . . . 39 90 3.15.2. AERO Client Behavior . . . . . . . . . . . . . . . . 39 91 3.15.3. AERO Server Behavior . . . . . . . . . . . . . . . . 42 92 3.16. The AERO Proxy . . . . . . . . . . . . . . . . . . . . . 44 93 3.17. AERO Route Optimization . . . . . . . . . . . . . . . . . 46 94 3.17.1. Route Optimization Initiation . . . . . . . . . . . 46 95 3.17.2. Relaying the NS . . . . . . . . . . . . . . . . . . 47 96 3.17.3. Processing the NS and Sending the NA . . . . . . . . 47 97 3.17.4. Relaying the NA . . . . . . . . . . . . . . . . . . 48 98 3.17.5. Processing the NA . . . . . . . . . . . . . . . . . 48 99 3.17.6. Route Optimization Maintenance . . . . . . . . . . . 48 100 3.18. Neighbor Unreachability Detection (NUD) . . . . . . . . . 49 101 3.19. Mobility Management and Quality of Service (QoS) . . . . 50 102 3.19.1. Mobility Update Messaging . . . . . . . . . . . . . 50 103 3.19.2. Forwarding Packets on Behalf of Departed Clients . . 51 104 3.19.3. Announcing Link-Layer Address and/or QoS Preference 105 Changes . . . . . . . . . . . . . . . . . . . . . . 52 106 3.19.4. Bringing New Links Into Service . . . . . . . . . . 52 107 3.19.5. Removing Existing Links from Service . . . . . . . . 52 108 3.19.6. Moving to a New Server . . . . . . . . . . . . . . . 52 109 3.20. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 53 110 3.20.1. Source-Specific Multicast (SSM) . . . . . . . . . . 53 111 3.20.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 55 112 3.20.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 56 113 3.21. Operation over Multiple AERO Links (VLANs) . . . . . . . 56 114 4. Direct Underlying Interfaces . . . . . . . . . . . . . . . . 57 115 5. AERO Clients on the Open Internetwork . . . . . . . . . . . . 57 116 6. Operation on AERO Links with /64 ASPs . . . . . . . . . . . . 57 117 7. AERO Adaptations for SEcure Neighbor Discovery (SEND) . . . . 58 118 8. AERO Critical Infrastructure Considerations . . . . . . . . . 59 119 9. DNS Considerations . . . . . . . . . . . . . . . . . . . . . 59 120 10. Transition Considerations . . . . . . . . . . . . . . . . . . 60 121 11. Implementation Status . . . . . . . . . . . . . . . . . . . . 60 122 12. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 60 123 13. Security Considerations . . . . . . . . . . . . . . . . . . . 61 124 14. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 62 125 15. References . . . . . . . . . . . . . . . . . . . . . . . . . 64 126 15.1. Normative References . . . . . . . . . . . . . . . . . . 64 127 15.2. Informative References . . . . . . . . . . . . . . . . . 65 128 Appendix A. AERO Alternate Encapsulations . . . . . . . . . . . 71 129 Appendix B. S/TLLAO Extensions for Special-Purpose Links . . . . 73 130 Appendix C. Implicit Mobility Management . . . . . . . . . . . . 75 131 Appendix D. Implementation Strategies for Route Optimization . . 75 132 Appendix E. Change Log . . . . . . . . . . . . . . . . . . . . . 76 133 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 80 135 1. Introduction 137 Asymmetric Extended Route Optimization (AERO) fulfills the 138 requirements of Distributed Mobility Management (DMM) [RFC7333] and 139 route optimization [RFC5522] for aeronautical networking and other 140 network mobility use cases. AERO is based on a Non-Broadcast, 141 Multiple Access (NBMA) virtual link model known as the AERO link. 142 The AERO link is configured over one or more underlying 143 Internetworks, and nodes on the link can exchange IP packets via 144 tunneling. Multilink operation allows for increased reliability, 145 bandwidth optimization and traffic path diversity. 147 The AERO service comprises Clients, Proxys, Servers, and Gateways 148 that are seen as AERO link neighbors. Each node's AERO interface 149 uses an IPv6 link-local address format (known as the AERO address) 150 that supports operation of the IPv6 Neighbor Discovery (ND) protocol 151 [RFC4861] and links ND to IP forwarding. A node's AERO interface can 152 be configured over multiple underlying interfaces, and may therefore 153 may appear as a single interface with multiple link-layer addresses. 154 Each link-layer address is subject to change due to mobility and/or 155 QoS fluctuations, and link-layer address changes are signaled by ND 156 messaging the same as for any IPv6 link. 158 AERO links provide a cloud-based service where mobile nodes may use 159 any Server acting as a Mobility Anchor Point (MAP) and fixed nodes 160 may use any Gateway on the link for efficient communications. Fixed 161 nodes forward packets destined to other AERO nodes to the nearest 162 Gateway, which forwards them through the cloud. A mobile node's 163 initial packets are forwarded through the MAP, while direct routing 164 is supported through asymmetric route optimization while data packets 165 are flowing. Both unicast and multicast communications are 166 supported, and mobile nodes may efficiently move between locations 167 while maintaining continuous communications with correspondents and 168 without changing their IP Address. 170 AERO Relays are interconnected in a secured private BGP overlay 171 routing instance known as the "SPAN". The SPAN provides a hybrid 172 routing/bridging service to join the underlying Internetworks of 173 multiple disjoint administrative domains into a single unified AERO 174 link. Each AERO link instance is characterized by the set of 175 Mobility Service Prefixes (MSPs) common to all mobile nodes. The 176 link should extend to the point where a Gateway/MAP is on the optimal 177 route from any correspondent node on the link, and provides a gateway 178 between the underlying Internetwork and the SPAN. To the underlying 179 Internetwork, the Gateway/MAP is the source of a route to its MSP, 180 and hence uplink traffic to the mobile node is naturally routed to 181 the nearest Gateway/MAP. 183 AERO assumes the use of PIM Sparse Mode in support of multicast 184 communication. In support of Source Specific Multicast (SSM) when a 185 Mobile Node is the source, AERO route optimization ensures that a 186 shortest-path multicast tree is established with provisions for 187 mobility and multilink operation. In all other multicast scenarios 188 there are no AERO dependencies. 190 AERO was designed for aeronautical networking for both manned and 191 unmanned aircraft, where the aircraft is treated as a mobile node 192 that can connect an Internet of Things (IoT). AERO is also 193 applicable to a wide variety of other use cases. For example, it can 194 be used to coordinate the Virtual Private Network (VPN) links of 195 mobile nodes (e.g., cellphones, tablets, laptop computers, etc.) that 196 connect into a home enterprise network via public access networks 197 using services such as OpenVPN [OVPN]. Other applicable use cases 198 are also in scope. 200 The following numbered sections present the AERO specification. The 201 appendices at the end of the document are non-normative. 203 2. Terminology 205 The terminology in the normative references applies; the following 206 terms are defined within the scope of this document: 208 IPv6 Neighbor Discovery (ND) 209 an IPv6 control message service for coordinating neighbor 210 relationships between nodes connected to a common link. AERO 211 interfaces use the ND service specified in [RFC4861]. 213 IPv6 Prefix Delegation (PD) 214 a networking service for delegating IPv6 prefixes to nodes on the 215 link. The nominal PD service is DHCPv6 [RFC8415], however 216 alternate services (e.g., based on ND messaging) are also in scope 217 [I-D.templin-v6ops-pdhost][I-D.templin-6man-dhcpv6-ndopt]. Most 218 notably, a form of PD known as "Prefix Assertion" can be used if 219 the prefix can be represented in the IPv6 source address of an ND 220 message. 222 Access Network (ANET) 223 a node's first-hop data link service network, e.g., a radio access 224 network, cellular service provider network, corporate enterprise 225 network, or the public Internet itself. For secured ANETs, link- 226 layer security services such as IEEE 802.1X and physical-layer 227 security prevent unauthorized access internally while border 228 network-layer security services such as firewalls and proxies 229 prevent unauthorized outside access. 231 ANET interface 232 a node's attachment to a link in an ANET. 234 ANET address 235 an IP address assigned to a node's interface connection to an 236 ANET. 238 Internetwork (INET) 239 a connected IP network topology with a coherent routing and 240 addressing plan and that provides an Internetworking backbone 241 service. INETs also provide an underlay service over which the 242 AERO virtual link is configured. Example INETs include corporate 243 enterprise networks, aviation networks, and the public Internet 244 itself. When there is no administrative boundary between an ANET 245 and the INET, the ANET and INET are one and the same. 247 INET Partition 248 frequently, INETs such as large corporate enterprise networks are 249 sub-divided internally into separate isolated partitions. Each 250 partition is fully connected internally but disconnected form 251 other partitions, and there is no requirement that separate 252 partitions maintain consistent Internet Protocol and/or addressing 253 plans. (An INET partition is the same as a SPAN segment discussed 254 below.) 256 INET interface 257 a node's attachment to a link in an INET. 259 INET address 260 an IP address assigned to a node's interface connection to an 261 INET. 263 AERO link 264 a Non-Broadcast, Multiple Access (NBMA) tunnel virtual overlay 265 configured over one or more underlying INETs. Nodes on the AERO 266 link appear as single-hop neighbors from the perspective of the 267 virtual overlay even though they may be separated by many 268 underlying INET hops. 270 AERO interface 271 a node's attachment to an AERO link. Since the addresses assigned 272 to an AERO interface are managed for uniqueness, AERO interfaces 273 do not require Duplicate Address Detection (DAD) and therefore set 274 the administrative variable 'DupAddrDetectTransmits' to zero 275 [RFC4862]. 277 AERO address 278 an IPv6 link-local address assigned to an AERO interface and 279 constructed as specified in Section 3.4. 281 base AERO address 282 the lowest-numbered AERO address aggregated by the MNP (see 283 Section 3.4). 285 Mobility Service Prefix (MSP) 286 an IP prefix assigned to the AERO link and from which more- 287 specific Mobile Network Prefixes (MNPs) are derived. 289 Mobile Network Prefix (MNP) 290 an IP prefix allocated from an MSP and delegated to an AERO Client 291 or Gateway. 293 AERO node 294 a node that is connected to an AERO link, or that provides 295 services to other nodes on an AERO link. 297 AERO Client ("Client") 298 an AERO node that connects to one or more ANETs and requests MNP 299 PDs from one or more AERO Servers. Following PD, the Client 300 assigns a Client AERO address to the AERO interface for use in ND 301 exchanges with other AERO nodes. A Client can also be deployed on 302 the same platform as a Server, and a node that acts as a Client on 303 one AERO interface can also act as an AERO Server on a different 304 AERO interface. 306 AERO Server ("Server") 307 an INET node that configures an AERO interface to provide default 308 forwarding services and a Mobility Anchor Point (MAP) for AERO 309 Clients. The Server assigns an administratively-provisioned AERO 310 address to the AERO interface to support the operation of the ND/ 311 PD services, and advertises all of its associated MNPs via BGP 312 peerings with Relays. 314 AERO Gateway ("Gateway") 315 an AERO Server that also provides forwarding services between 316 nodes reached via the AERO link and correspondents on other links. 317 AERO Gateways are provisioned with MNPs (i.e., the same as for an 318 AERO Client) and run a dynamic routing protocol to discover any 319 non-MNP IP routes. In both cases, the Gateway advertises the 320 MSP(s) over INET interfaces, and distributes all of its associated 321 MNPs and non-MNP IP routes via BGP peerings with Relays (i.e., the 322 same as for an AERO Server). 324 AERO Relay ("Relay") 325 a node that provides hybrid routing/bridging services (as well as 326 a security trust anchor) for nodes on an AERO link. As a router, 327 the Relay forwards packets using standard IP forwarding. As a 328 bridge, the Relay forwards packets over the AERO link without 329 decrementing the IPv6 Hop Limit. AERO Relays peer with Servers 330 and other Relays to discover the full set of MNPs for the link as 331 well as any non-MNPs that are reachable via Gateways. 333 AERO Proxy ("Proxy") 334 a node that provides proxying services between Clients in an ANET 335 and Servers in external INETs. The AERO Proxy is a conduit 336 between the ANET and external INETs in the same manner as for 337 common web proxies, and behaves in a similar fashion as for ND 338 proxies [RFC4389]. 340 Spanning Partitioned AERO Networks (SPAN) 341 a means for bridging disjoint INET partitions as segments of a 342 unified AERO link the same as for a bridged campus LAN. The SPAN 343 is a mid-layer IPv6 encapsulation service in the AERO routing 344 system that supports a unified AERO link view for all segments. 345 Each segment in the SPAN is a distinct INET partition. 347 SPAN Service Prefix (SSP) 348 a global or unique local /96 IPv6 prefix assigned to the AERO link 349 to support SPAN services. 351 SPAN Partition Prefix (SPP) 352 a sub-prefix of the SPAN Service Prefix uniquely assigned to a 353 single AERO link segment. 355 SPAN Address 356 a global or unique local IPv6 address taken from a SPAN Partition 357 Prefix and constructed as specified in Section 3.5. SPAN 358 addresses are statelessly derived from AERO addresses, and vice- 359 versa. 361 ingress tunnel endpoint (ITE) 362 an AERO interface endpoint that injects encapsulated packets into 363 an AERO link. 365 egress tunnel endpoint (ETE) 366 an AERO interface endpoint that receives encapsulated packets from 367 an AERO link. 369 link-layer address 370 an IP address used as an encapsulation header source or 371 destination address from the perspective of the AERO interface. 372 When UDP encapsulation is used, the UDP port number is also 373 considered as part of the link-layer address. From the 374 perspective of the AERO interface, the link-layer address is 375 either an INET address for intra-segment encapsulation or a SPAN 376 address for inter-segment encapsulation. 378 network layer address 379 the source or destination address of an encapsulated IP packet 380 presented to the AERO interface. 382 end user network (EUN) 383 an internal virtual or external edge IP network that an AERO 384 Client or Gateway connects to the rest of the network via the AERO 385 interface. The Client/Gateway sees each EUN as a "downstream" 386 network, and sees the AERO interface as the point of attachment to 387 the "upstream" network. 389 Mobile Node (MN) 390 an AERO Client and all of its downstream-attached networks that 391 move together as a single unit, i.e., an end system that connects 392 an Internet of Things. 394 Mobile Router (MR) 395 a MN's on-board router that forwards packets between any 396 downstream-attached networks and the AERO link. 398 Mobility Anchor Point (MAP) 399 an AERO Server that is currently tracking and reporting the 400 mobility events of its associated Mobile Node Clients. 402 Route Optimization Source (ROS) 403 the AERO node nearest the source that initiates route 404 optimization. The ROS may be a Server or Proxy acting on behalf 405 of the source Client. 407 Route Optimization responder (ROR) 408 the AERO node nearest the target destination that responds to 409 route optimization requests. The ROR may be a Server acting as a 410 MAP on behalf of a target MNP Client, or a Gateway for a non-MNP 411 destination. 413 MAP List 414 a geographically and/or topologically referenced list of AERO 415 addresses of all MAPs within the same AERO link. There is a 416 single MAP list for the entire AERO link. 418 ROS List 419 a list of AERO-to-INET address mappings of all ROSes within the 420 same AERO link segment. There is a distinct ROS list for each 421 AERO link segment. 423 Distributed Mobility Management (DMM) 424 a BGP-based overlay routing service coordinated by Servers and 425 Relays that tracks all MAP-to-Client associations. 427 Throughout the document, the simple terms "Client", "Server", 428 "Relay", "Proxy" and "Gateway" refer to "AERO Client", "AERO Server", 429 "AERO Relay", "AERO Proxy" and "AERO Gateway", respectively. 431 Capitalization is used to distinguish these terms from other common 432 Internetworking uses in which they appear without capitalization. 434 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 435 the names of node variables, messages and protocol constants) is used 436 throughout this document. Also, the term "IP" is used to generically 437 refer to either Internet Protocol version, i.e., IPv4 [RFC0791] or 438 IPv6 [RFC8200]. 440 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 441 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 442 document are to be interpreted as described in [RFC2119]. Lower case 443 uses of these words are not to be interpreted as carrying RFC2119 444 significance. 446 3. Asymmetric Extended Route Optimization (AERO) 448 The following sections specify the operation of IP over Asymmetric 449 Extended Route Optimization (AERO) links: 451 3.1. AERO Link Reference Model 452 +----------------+ 453 | AERO Relay R1 | 454 | Nbr: S1, S2, P1| 455 |(X1->S1; X2->S2)| 456 | MSP M1 | 457 +-+---------+--+-+ 458 +--------------+ | Secured | | +--------------+ 459 |AERO Server S1| | tunnels | | |AERO Server S2| 460 | Nbr: C1, R1 +-----+ | +-----+ Nbr: C2, R1 | 461 | default->R1 | | | default->R1 | 462 | X1->C1 | | | X2->C2 | 463 +-------+------+ | +------+-------+ 464 | AERO Link | | 465 X---+---+-------------------+--)---------------+---+---X 466 | | | | 467 +-----+--------+ +--------+--+-----+ +--------+-----+ 468 |AERO Client C1| | AERO Proxy P1 | |AERO Client C2| 469 | Nbr: S1 | |(Proxy Nbr Cache)| | Nbr: S2 | 470 | default->S1 | +--------+--------+ | default->S2 | 471 | MNP X1 | | | MNP X2 | 472 +------+-------+ .--------+------. +-----+--------+ 473 | (- Proxyed Clients -) | 474 .-. `---------------' .-. 475 ,-( _)-. ,-( _)-. 476 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 477 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 478 `-(______)-' +-------+ +-------+ `-(______)-' 480 Figure 1: AERO Link Reference Model 482 Figure 1 presents the AERO link reference model. In this model: 484 o the AERO link is an overlay network service configured over one or 485 more underlying INET partitions which may be managed by different 486 administrative authorities and have incompatible protocols and/or 487 addressing plans. 489 o AERO Relay R1 aggregates Mobility Service Prefix (MSP) M1, 490 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 491 via BGP peerings over secured tunnels to Servers (S1, S2). Relays 492 use the SPAN service to bridge disjoint segments of a partitioned 493 AERO link. 495 o AERO Servers S1 and S2 configure secured tunnels with Relay R1 and 496 also act as Mobility Anchor Points (MAPs) and default routers for 497 their associated Clients C1 and C2. 499 o AERO Clients C1 and C2 associate with Servers S1 and S2, 500 respectively. They receive Mobile Network Prefix (MNP) 501 delegations X1 and X2, and also act as default routers for their 502 associated physical or internal virtual EUNs. Simple hosts H1 and 503 H2 attach to the EUNs served by Clients C1 and C2, respectively. 505 o AERO Proxy P1 configures a secured tunnel with Relay R1 and 506 provides proxy services for AERO Clients in secured enclaves that 507 cannot associate directly with other AERO link neighbors. 509 Each node on the AERO link maintains an AERO interface neighbor cache 510 and an IP forwarding table the same as for any link. Although the 511 figure shows a limited deployment, in common operational practice 512 there will normally be many additional Relays, Servers, Clients and 513 Proxys. 515 3.2. AERO Node Types 517 AERO Relays provide hybrid routing/bridging services (as well as a 518 security trust anchor) for nodes on an AERO link. Relays use 519 standard IPv6 routing to forward packets both within the same INET 520 partitions and between disjoint INET partitions based on a mid-layer 521 IPv6 encapsulation known as the SPAN header. The inner IP layer 522 experiences a virtual bridging service since the inner IP TTL/Hop 523 Limit is not decremented during forwarding. Each Relay also peers 524 with Servers and other Relays in a dynamic routing protocol instance 525 to provide a Distributed Mobility Management (DMM) service for the 526 list of active MNPs (see Section 3.3). Relays present the AERO link 527 as a set of one or more Mobility Service Prefixes (MSPs) but as link- 528 layer devices need not connect directly to the AERO link themselves 529 unless an administrative interface is desired. Relays configure 530 secured tunnels with Servers, Proxys and other Relays; they further 531 maintain IP forwarding table entries for each Mobile Network Prefix 532 (MNP) and any other reachable non-MNP prefixes. 534 AERO Servers provide default forwarding services and a Mobility 535 Anchor Point (MAP) for AERO Client Mobile Nodes (MNs). Each Server 536 also peers with Relays in a dynamic routing protocol instance to 537 advertise its list of associated MNPs (see Section 3.3). Servers 538 facilitate PD exchanges with Clients, where each delegated prefix 539 becomes an MNP taken from an MSP. Servers forward packets between 540 AERO interface neighbors and track each Client's mobility profiles. 542 AERO Clients receive MNPs through PD exchanges with AERO Servers over 543 the AERO link, and distribute the MNPs to nodes on EUNs. Each Client 544 can associate with a single Server or with multiple Servers (e.g., 545 for fault tolerance, load balancing, etc). A Client may also be co- 546 resident on the same physical or virtual platform as a Server; in 547 that case, the Client and Server behave as a single functional unit 548 and without the need for any Client/Server control messaging. 550 AERO Proxys provide a conduit for AERO Clients in ANETs to associate 551 with AERO Servers in external INETs. Client and Servers exchange 552 control plane messages via the Proxy, which intercepts them at the 553 ANET/INET boundary. The Proxy forwards data packets to and from 554 Clients according to forwarding information in the neighbor cache. 555 The Proxy function is specified in Section 3.16. 557 AERO Gateways are Servers that provide forwarding services between 558 the AERO interface and INET/EUN interfaces. Gateways are provisioned 559 with MNPs the same as for an AERO Client, and also run a dynamic 560 routing protocol to discover any non-MNP IP routes. The Gateway 561 advertises the MSP(s) to INETs, and distributes all of its associated 562 MNPs and non-MNP IP routes via BGP peerings with Relays. 564 AERO Relays, Servers, Proxys and Gateways are critical infrastructure 565 elements in fixed (i.e., non-mobile) INET deployments and hence have 566 permanent and unchanging INET addresses. AERO Clients are MNs that 567 connect via ANET interfaces, i.e., their ANET addresses may change 568 when the Client moves to a new ANET connection. 570 3.3. AERO Routing System 572 The AERO routing system comprises a private instance of the Border 573 Gateway Protocol (BGP) [RFC4271] that is coordinated between Relays 574 and Servers and does not interact with either the public Internet BGP 575 routing system or any underlying INET routing systems. 577 In a reference deployment, each Server is configured as an Autonomous 578 System Border Router (ASBR) for a stub Autonomous System (AS) using 579 an AS Number (ASN) that is unique within the BGP instance, and each 580 Server further uses eBGP to peer with one or more Relays but does not 581 peer with other Servers. Each INET of a multi-segment AERO link must 582 include one or more Relays, which peer with the Servers and Proxys 583 within that INET. All Relays within the same INET are members of the 584 same hub AS using a common ASN, and use iBGP to maintain a consistent 585 view of all active MNPs currently in service. The Relays of 586 different INETs peer with one another using eBGP. 588 Relays advertise the AERO link's MSPs and any non-MNP routes to each 589 of their Servers. This means that any aggregated non-MNPs (including 590 "default") are advertised to all Servers. Each Relay configures a 591 black-hole route for each of its MSPs. By black-holing the MSPs, the 592 Relay will maintain forwarding table entries only for the MNPs that 593 are currently active, and packets destined to all other MNPs will 594 correctly incur Destination Unreachable messages due to the black- 595 hole route. In this way, Servers have only partial topology 596 knowledge (i.e., they know only about the MNPs of their directly 597 associated Clients) and they forward all other packets to Relays 598 which have full topology knowledge. 600 Servers maintain a working set of associated MNPs, and dynamically 601 announce new MNPs and withdraw departed MNPs in eBGP updates to 602 Relays. Servers that are configured as Gateways also redistribute 603 non-MNP routes learned from non-AERO interfaces via their eBGP Relay 604 peerings. 606 Clients are expected to remain associated with their current Servers 607 for extended timeframes, however Servers SHOULD selectively suppress 608 updates for impatient Clients that repeatedly associate and 609 disassociate with them in order to dampen routing churn. Servers 610 that are configured as Gateways advertise the MSPs via INET/EUN 611 interfaces, and forward packets between INET/EUN interfaces and the 612 AERO interface using standard IP forwarding. 614 For IPv6 MNPs, the AERO routing system includes only IPv6 routes. 615 For IPv4 MNPs, the AERO routing system includes both IPv4 routes and 616 also IPv6 routes based on the IPv4-mapped IPv6 address corresponding 617 to the MNP and with prefix length set to 96 plus the length of the 618 IPv4 prefix. (For example, if the IPv4 MNP is 192.0.2.0/24 then the 619 IPv4-mapped prefix is 0:0:0:0:0:FFFF:192.0.2.0/120.) 621 Scaling properties of the AERO routing system are limited by the 622 number of BGP routes that can be carried by Relays. As of 2015, the 623 global public Internet BGP routing system manages more than 500K 624 routes with linear growth and no signs of router resource exhaustion 625 [BGP]. More recent network emulation studies have also shown that a 626 single Relay can accommodate at least 1M dynamically changing BGP 627 routes even on a lightweight virtual machine, i.e., and without 628 requiring high-end dedicated router hardware. 630 Therefore, assuming each Relay can carry 1M or more routes, this 631 means that at least 1M Clients can be serviced by a single set of 632 Relays. A means of increasing scaling would be to assign a different 633 set of Relays for each set of MSPs. In that case, each Server still 634 peers with one or more Relays, but institutes route filters so that 635 BGP updates are only sent to the specific set of Relays that 636 aggregate the MSP. For example, if the MSP for the AERO link is 637 2001:db8::/32, a first set of Relays could service the MSP segment 638 2001:db8::/40, a second set of Relays could service 639 2001:db8:0100::/40, a third set could service 2001:db8:0200::/40, 640 etc. 642 Assuming up to 1K sets of Relays, the AERO routing system can then 643 accommodate 1B or more MNPs with no additional overhead (for example, 644 it should be possible to service 1B /64 MNPs taken from a /34 MSP and 645 even more for shorter prefixes). In this way, each set of Relays 646 services a specific set of MSPs that they advertise to the native 647 Internetwork routing system, and each Server configures MSP-specific 648 routes that list the correct set of Relays as next hops. This 649 arrangement also allows for natural incremental deployment, and can 650 support small scale initial deployments followed by dynamic 651 deployment of additional Clients, Servers and Relays without 652 disturbing the already-deployed base. 654 A full discussion of the BGP-based routing system used by AERO is 655 found in [I-D.ietf-rtgwg-atn-bgp]. The system provides for 656 Distributed Mobility Management (DMM) per the distributed mobility 657 anchoring architecture [I-D.ietf-dmm-distributed-mobility-anchoring]. 659 3.4. AERO Addresses 661 A Client's AERO address is an IPv6 link-local address with an 662 interface identifier based on the Client's delegated MNP. Relay, 663 Server and Proxy AERO addresses are assigned from the range fe80::/96 664 and include an administratively-provisioned value in the lower 32 665 bits. 667 For IPv6, Client AERO addresses begin with the prefix fe80::/64 and 668 include in the interface identifier (i.e., the lower 64 bits) a 669 64-bit prefix taken from one of the Client's IPv6 MNPs. For example, 670 if the AERO Client receives the IPv6 MNP: 672 2001:db8:1000:2000::/56 674 it constructs its corresponding AERO addresses as: 676 fe80::2001:db8:1000:2000 678 fe80::2001:db8:1000:2001 680 fe80::2001:db8:1000:2002 682 ... etc. ... 684 fe80::2001:db8:1000:20ff 686 For IPv4, Client AERO addresses are based on an IPv4-mapped IPv6 687 address formed from an IPv4 MNP and with a Prefix Length of 96 plus 688 the MNP prefix length. For example, for the IPv4 MNP 192.0.2.32/28 689 the IPv4-mapped IPv6 MNP is: 691 0:0:0:0:0:FFFF:192.0.2.16/124 693 The Client then constructs its AERO addresses with the prefix 694 fe80::/64 and with the lower 64 bits of the IPv4-mapped IPv6 address 695 in the interface identifier as: 697 fe80::FFFF:192.0.2.16 699 fe80::FFFF:192.0.2.17 701 fe80::FFFF:192.0.2.18 703 ... etc. ... 705 fe80:FFFF:192.0.2.31 707 Relay, Server and Proxy AERO addresses are allocated from the range 708 fe80::/96, and MUST be managed for uniqueness. The lower 32 bits of 709 the AERO address includes a unique integer value (e.g., fe80::1, 710 fe80::2, fe80::3, etc.) as assigned by the administrative authority 711 for the link. If the link spans multiple segments (i.e., multiple 712 INETs), the AERO addresses are assigned to each INET in 1x1 713 correspondence with SPAN addresses (see: Section 3.5). The address 714 fe80:: is reserved as the IPv6 link-local Subnet Router Anycast 715 address [RFC4291], and the address fe80::ffff:ffff is reserved as the 716 unspecified AERO address; hence, these values are not available 717 general assignment. 719 The lowest-numbered AERO address from a Client's MNP delegation 720 serves as the "base" AERO address (for example, for the MNP 721 2001:db8:1000:2000::/56 the base AERO address is 722 fe80::2001:db8:1000:2000). The Client then assigns the base AERO 723 address to the AERO interface and uses it for the purpose of 724 maintaining the neighbor cache entry. The Server likewise uses the 725 AERO address as its index into the neighbor cache for this Client. 727 If the Client has multiple AERO addresses (i.e., when there are 728 multiple MNPs and/or MNPs with prefix lengths shorter than /64), the 729 Client originates ND messages using the base AERO address as the 730 source address and accepts and responds to ND messages destined to 731 any of its AERO addresses as equivalent to the base AERO address. In 732 this way, the Client maintains a single neighbor cache entry that may 733 be indexed by multiple AERO addresses. 735 The Client's Subnet Router Anycast address can be statelessly 736 determined from its AERO address by simply transposing the AERO 737 address into the upper N bits of the Anycast address followed by 738 128-N bits of zeros. For example, for the AERO address 739 fe80::2001:db8:1:2 the subnet router anycast address is 740 2001:db8:1:2::/64. 742 AERO addresses for mobile node Clients embed a MNP as discussed 743 above, while AERO addresses for non-MNP destinations are constructed 744 in exactly the same way. A Client AERO address is therefore encodes 745 either an MNP if the prefix is reached via the SPAN or a non-MNP if 746 the prefix is reached via a Gateway. 748 3.5. Spanning Partitioned AERO Networks (SPAN) 750 In the simplest case, an AERO link configured over a single INET 751 appears as a single unified link with a consistent underlying network 752 addressing plan. In that case, all nodes on the link can exchange 753 packets via encapsulation with INET addresses, since the underlying 754 INET is connected. In common practice, however, an AERO link may be 755 partitioned into multiple "segments", where each segment is a 756 distinct INET potentially managed under a different administrative 757 authority (e.g., as for worldwide aviation service providers such as 758 ARINC, SITA, Inmarsat, etc.). Individual INETs may themselves be 759 partitioned internally, in which case each internal partition is seen 760 as a separate segment. 762 The addressing plan of each segment is consistent internally but will 763 often bear no relation to the addressing plans of other segments. 764 Each segment is also likely to be separated from others by network 765 security devices (e.g., firewalls, proxies, packet filtering 766 gateways, etc.), and in many cases disjoint segments may not even 767 have any common physical link connections at all. Therefore, nodes 768 can only be assured of exchanging packets directly with 769 correspondents in the same segment, and not with those in other 770 segments. The only means for joining the segments therefore is 771 through inter-domain peerings between AERO Relays. 773 The same as for traditional campus LANs, multiple AERO link segments 774 can be joined into a single unified link via a virtual bridging 775 service termed the "SPAN". The SPAN performs link-layer packet 776 forwarding between segments (i.e., bridging) without decrementing the 777 network-layer TTL/Hop Limit. The SPAN model is depicted in Figure 2: 779 . . . . . . . . . . . . . . . . . . . . . . . 780 . . 781 . .-(::::::::) . 782 . .-(::::::::::::)-. +-+ . 783 . (:::: Segment A :::)--|R|---+ . 784 . `-(::::::::::::)-' +-+ | . 785 . `-(::::::)-' | . 786 . | . 787 . .-(::::::::) | . 788 . .-(::::::::::::)-. +-+ | . 789 . (:::: Segment B :::)--|R|---+ . 790 . `-(::::::::::::)-' +-+ | . 791 . `-(::::::)-' | . 792 . | . 793 . .-(::::::::) | . 794 . .-(::::::::::::)-. +-+ | . 795 . (:::: Segment C :::)--|R|---+ . 796 . `-(::::::::::::)-' +-+ | . 797 . `-(::::::)-' | . 798 . | . 799 . ..(etc).. x . 800 . . 801 . . 802 . <- AERO Link Bridged by the SPAN -> . 803 . . . . . . . . . . . . . .. . . . . . . . . 805 Figure 2: The SPAN 807 To support the SPAN, AERO links require a reserved /96 IPv6 "SPAN 808 Service Prefix (SSP)". Although any routable IPv6 prefix can be 809 used, a Unique Local Address (ULA) prefix (e.g., fd00::/96) [RFC4389] 810 is recommended since border routers are commonly configured to 811 prevent packets with ULAs from being injected into the AERO link by 812 an external IPv6 node and from leaking out of the AERO link to the 813 outside world. 815 Each segment in the SPAN assigns a unique sub-prefix of the SSP 816 termed a "SPAN Partition Prefix (SPP)". For example, a first segment 817 could assign fd00::1000/116, a second could assign fd00::2000/116, a 818 third could assign fd00::3000/116, etc. The administrative 819 authorities for each segment must therefore coordinate to assure 820 mutually-exclusive SPP assignments, but internal provisioning of the 821 SPP is a local consideration for each administrative authority. 823 A "SPAN address" is an address taken from a SPP and assigned to a 824 Relay, Server or Proxy interface. SPAN addresses are formed by 825 simply replacing the upper portion of an administratively-assigned 826 AERO address with the SPP. For example, if the SPP is 827 fd00::1000/116, the SPAN address formed from the AERO address 828 fe80::1001 is simply fd00::1001. 830 An "INET address" is an address of a node's interface connection to 831 an INET. AERO/SPAN/INET address mappings are maintained as permanent 832 neighbor cache entires as discussed in Section 3.8. 834 AERO Relays serve as bridges to join multiple segments into a unified 835 AERO link over multiple diverse administrative domains. They support 836 the bridging function by first establishing forwarding table entries 837 for their SPPs either via standard BGP routing or static routes. For 838 example, if three Relays ('A', 'B' and 'C') from different segments 839 serviced the SPPs fd00::1000/116, fd00::2000/116 and fd00::3000/116 840 respectively, then the forwarding tables in each Relay are as 841 follows: 843 A: fd00::1000/116->local, fd00::2000/116->B, fd00::3000/116->C 845 B: fd00::1000/116->A, fd00::2000/116->local, fd00::3000/116->C 847 C: fd00::1000/116->A, fd00::2000/116->B, fd00::3000/116->local 849 These forwarding table entries are permanent and never change, since 850 they correspond to fixed infrastructure elements in their respective 851 segments. This provides the basis for a link-layer forwarding 852 service that cannot be disrupted by routing updates due to node 853 mobility. 855 With the SPPs in place in each Relay's forwarding table, control and 856 data packets sent between AERO nodes in different segments can 857 therefore be carried over the SPAN via encapsulation. For example, 858 when a source node in segment A forwards a packet with IPv6 address 859 2001:db8:1:2::1 to a destination node in segment C with IPv6 address 860 2001:db8:1000:2000::1, it first encapsulates the packet in a SPAN 861 header with source SPAN address taken from fd00::1000/116 (e.g., 862 fd00::1001) and destination SPAN address taken from fd00::3000/116 863 (e.g., fd00::3001). Next, it encapsulates the SPAN message in an 864 INET header with source address set to its own INET address (e.g., 865 192.0.2.100) and destination set to the INET address of a Relay 866 (e.g., 192.0.2.1). 868 SPAN encapsulation is based on Generic Packet Tunneling in IPv6 869 [RFC2473]; the encapsulation format in the above example is shown 870 inFigure 3: 872 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 873 | INET Header | 874 | src = 192.0.2.100 | 875 | dst = 192.0.2.1 | 876 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 877 | SPAN Header | 878 | src = fd00::1001 | 879 | dst = fd00::3001 | 880 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 881 | Inner IP Header | 882 | src = 2001:db8:1:2::1 | 883 | dst = 2001:db8:1000:2000::1 | 884 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 885 | | 886 ~ ~ 887 ~ Inner Packet Body ~ 888 ~ ~ 889 | | 890 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 892 Figure 3: SPAN Encapsulation 894 In this format, the inner IP header and packet body are the original 895 IP packet, the SPAN header is an IPv6 header prepared according to 896 [RFC2473], and the INET header is prepared according to Section 3.9. 897 A packet is said to be "forwarded/sent into the SPAN" when it is 898 encapsulated as described above then forwarded via a secured tunnel 899 to a neighboring Relay. 901 This gives rise to a routing system that contains both MNP routes 902 that may change dynamically due to regional node mobility and SPAN 903 routes that never change. The Relays can therefore provide link- 904 layer bridging by sending packets into the SPAN instead of network- 905 layer routing according to MNP routes. As a result, opportunities 906 for packet loss due to node mobility between different segments are 907 mitigated. 909 With reference to Figure 3, for a Client's AERO address the SPAN 910 address is simply set to the Subnet Router Anycast address. For non- 911 link-local addresses, the destination SPAN address may not be known 912 in advance for the first few packets of a flow sent via the SPAN. In 913 that case, the SPAN destination address is set to the original 914 packet's destination, and the SPAN routing system will direct the 915 packet to the correct SPAN egress node. (In the above example, the 916 SPAN destination address is simply 2001:db8:1000:2000::1.) 918 3.6. AERO Interface Characteristics 920 AERO interfaces use encapsulation (see: Section 3.9) to exchange 921 packets with neighbors attached to the AERO link. 923 AERO interfaces maintain a neighbor cache for tracking per-neighbor 924 state the same as for any interface. AERO interfaces use ND messages 925 including Router Solicitation (RS), Router Advertisement (RA), 926 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 927 neighbor cache management. 929 AERO interface ND messages include one or more Source/Target Link- 930 Layer Address Options (S/TLLAOs) formatted as shown in Figure 4: 932 0 1 2 3 933 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 1 934 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 935 | Type | Length = 5 | Prefix Length |S|R|D|X|N|Resvd| 936 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 937 | Interface ID | Port Number | 938 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 939 | | 940 + + 941 | | 942 + Link Layer Address + 943 | | 944 + + 945 | | 946 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 947 |P00|P01|P02|P03|P04|P05|P06|P07|P08|P09|P10|P11|P12|P13|P14|P15| 948 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 949 |P16|P17|P18|P19|P20|P21|P22|P23|P24|P25|P26|P27|P28|P29|P30|P31| 950 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 951 |P32|P33|P34|P35|P36|P37|P38|P39|P40|P41|P42|P43|P44|P45|P46|P47| 952 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 953 |P48|P49|P50|P51|P52|P53|P54|P55|P56|P57|P58|P59|P60|P61|P62|P63| 954 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 956 Figure 4: AERO Source/Target Link-Layer Address Option (S/TLLAO) 957 Format 959 In this format: 961 o Type is set to '1' for SLLAO or '2' for TLLAO. 963 o Length is set to the constant value '5' (i.e., 5 units of 8 964 octets). 966 o Prefix Length is set to the MNP prefix length in an ND message for 967 the Client AERO address found in the source (RS), destination (RA) 968 or target (NA) address; otherwise set to 0. If the message 969 contains multiple SLLAOs, only the Prefix Length value in the 970 SLLAO with S set to 1 is consulted and the values in other SLLAOs 971 are ignored. 973 o S (the 'Source' bit) is set to '1' in the S/TLLAO of an ND message 974 that corresponds to the ANET/INET interface over which the ND 975 message is sent, and set to 0 in all other S/TLLAOs. 977 o R (the "Release" bit) is set to '1' in an S/TLLAO in an RS/NA sent 978 for the purpose of departing from a Server; otherwise, set to '0'. 979 The recipient places the corresponding neighbor cache entry in the 980 DEPARTED state. For RS message, the recipient then releases the 981 corresponding PD and returns an RA with Router Lifetime set to '0' 983 o D (the "Disable" bit) is set to '1' in the S/TLLAOs of an RS/NA 984 message for each Interface ID that is to be disabled in the 985 neighbor cache entry; otherwise, set to '0'. 987 o X (the "proXy" bit) is set to '1' in the SLLAO of an RS/RA message 988 by the Proxy when there is a Proxy in the path; otherwise, set to 989 '0'. If the message contains multiple SLLAOs, only the X value in 990 the first SLLAO is consulted and the values in other SLLAOs are 991 ignored. 993 o N (the "(Network Address) Translator (NAT)" bit) is set to '1' in 994 the SLLAO of an RA message by the Server if there is a translator 995 in the path; otherwise, set to '0'. If the message contains 996 multiple SLLAOs, only the N value in the first SLLAO is consulted 997 and the values in other SLLAOs are ignored. 999 o Resvd is set to the value '0' on transmission and ignored on 1000 receipt. 1002 o Interface ID is set to a 16-bit integer value corresponding to an 1003 AERO node's ANET/INET interface. Once the node has assigned an 1004 Interface ID to an ANET interface, the assignment must remain 1005 unchanged until the node fully detaches from the AERO link. The 1006 value 0xffff is reserved as the Server's INET Interface ID, i.e., 1007 Servers MUST use Interface ID 0xffff, and Clients MUST number 1008 their ANET Interface IDs with values in the range of 0-0xfffe. 1010 o Port Number and Link Layer Address are set to the encapsulation 1011 addresses required to send packets via the target node (or to '0' 1012 when the addresses are left unspecified). When UDP is not used as 1013 part of the encapsulation, Port Number is set to '0'. When the 1014 encapsulation IP address family is IPv4, IP Address is formed as 1015 an IPv4-mapped IPv6 address as specified in Section 3.4. 1017 o P(i) is a set of Preferences that correspond to the 64 1018 Differentiated Service Code Point (DSCP) values [RFC2474]. Each 1019 P(i) is set to the value '0' ("disabled"), '1' ("low"), '2' 1020 ("medium") or '3' ("high") to indicate a QoS preference level for 1021 packet forwarding purposes. 1023 A Client's AERO interface may be configured over multiple ANET 1024 interface connections. For example, common mobile handheld devices 1025 have both wireless local area network ("WLAN") and cellular wireless 1026 links. These links are typically used "one at a time" with low-cost 1027 WLAN preferred and highly-available cellular wireless as a standby. 1028 In a more complex example, aircraft frequently have many wireless 1029 data link types (e.g. satellite-based, cellular, terrestrial, air-to- 1030 air directional, etc.) with diverse performance and cost properties. 1032 A Client's ANET interfaces are classified as follows: 1034 o Native interfaces connect to the open INET, and have a global IP 1035 address that is reachable from any INET correspondent. 1037 o NATed interfaces connect to an ANET behind a Network Address 1038 Translator (NAT). The NAT does not participate in any AERO 1039 control message signaling, but the Server can issue control 1040 messages on behalf of the Client. Clients that are behind a NAT 1041 are required to send periodic keepalive messages to keep NAT state 1042 alive when there are no data packets flowing. If no other 1043 periodic messaging service is available, the Client can send RS 1044 messages to receive RA replies from its Server(s). 1046 o VPNed interfaces use security encapsulation over the ANET to a 1047 Virtual Private Network (VPN) server that also acts as an AERO 1048 Server. As with NATed links, the Server can issue control 1049 messages on behalf of the Client, but the Client need not send 1050 periodic keepalives in addition to those already used to maintain 1051 the VPN connection. 1053 o Proxyed interfaces connect to an ANET that is separated from the 1054 open INET by an AERO Proxy. Unlike NATed and VPNed interfaces, 1055 the Proxy can actively issue control messages on behalf of the 1056 Client. 1058 o Direct interfaces connect the Client directly to a neighbor 1059 without crossing any ANET/INET paths. An example is a line-of- 1060 sight link between a remote pilot and an unmanned aircraft. 1062 If a Client's multiple ANET interfaces are used "one at a time" 1063 (i.e., all other interfaces are in standby mode while one interface 1064 is active), then ND messages include only a single S/TLLAO with 1065 Interface ID set to a constant value. In that case, the Client would 1066 appear to have a single ANET interface but with a dynamically 1067 changing ANET address. 1069 If the Client has multiple active ANET interfaces, then from the 1070 perspective of ND it would appear to have multiple link-layer 1071 addresses. In that case, ND messages MAY include multiple S/TLLAOs 1072 -- each with an Interface ID that corresponds to a specific ANET 1073 interface. S must be set to 1 in the S/TLLAO corresponding to the 1074 AERO node's ANET interface used to transmit the message and set to 0 1075 in all other S/TLLAOs. 1077 When the Client includes an S/TLLAO for an ANET interface for which 1078 it is aware that there is a NAT on the path to the Server, or when a 1079 node includes an S/TLLAO solely for the purpose of announcing new QoS 1080 preferences, the node MAY set both Port Number and Link-Layer Address 1081 to 0 to indicate that the addresses are unspecified at the network 1082 layer and must instead be derived from the link-layer encapsulation 1083 headers. 1085 Relay, Server and Proxy AERO interfaces may be configured over one or 1086 more secured tunnel interfaces. The AERO interface configures both 1087 an AERO address and its corresponding SPAN address, while the 1088 underlying secured tunnel interfaces also configure the same SPAN 1089 address. The AERO interface encapsulates each packet in a SPAN 1090 header if necessary and presents the packet to the underlying secured 1091 tunnel interface. For Relays that do not configure an AERO 1092 interface, the secured tunnel interfaces themselves are exposed to 1093 the IP layer with each interface configuring the same SPAN address. 1094 Routing protocols such as BGP therefore run directly over the secured 1095 tunnel interfaces. For nodes that configure an AERO interface, 1096 routing protocols such as BGP run over the AERO interface but do not 1097 employ SPAN encapsulation. Instead, the AERO interface presents the 1098 routing protocol packets directly to the underlying secured tunnels 1099 without applying encapsulation and while using the SPAN address as 1100 the source address. This distinction must be honored consistently 1101 according to each node's configuration so that the IP forwarding 1102 table will associate discovered IP routes with the correct interface. 1104 3.7. AERO Interface Initialization 1106 AERO Servers, Proxys and Clients configure AERO interfaces as their 1107 point of attachment to the AERO link. AERO nodes assign the MSPs for 1108 the link to their AERO interfaces (i.e., as a "route-to-interface") 1109 to ensure that packets with destination addresses covered by an MNP 1110 not explicitly assigned to a non-AERO interface are directed to the 1111 AERO interface. 1113 AERO interface initialization procedures for Servers, Proxys, Clients 1114 and Relays are discussed in the following sections. 1116 3.7.1. AERO Server/Gateway Behavior 1118 When a Server enables an AERO interface, it assigns AERO/SPAN 1119 addresses and configures permanent neighbor cache entries for 1120 neighbors in the same SPAN segment by consulting the ROS list for the 1121 segment. The Server also configures secured tunnels with one or more 1122 neighboring Relays and engages in a BGP routing protocol session with 1123 each Relay. 1125 The AERO interface provides a single interface abstraction to the IP 1126 layer, but internally comprises multiple secured tunnels as well as 1127 an NBMA nexus for sending encapsulated data packets to AERO interface 1128 neighbors. The Server further configures a service to facilitate ND/ 1129 PD exchanges with AERO Clients and manages per-Client neighbor cache 1130 entries and IP forwarding table entries based on control message 1131 exchanges. 1133 Gateways are simply Servers that run a dynamic routing protocol 1134 between the AERO interface and INET/EUN interfaces (see: 1135 Section 3.3). The Gateway provisions MNPs to networks on the INET/ 1136 EUN interfaces (i.e., the same as a Client would do) and advertises 1137 the MSP(s) for the AERO link over the INET/EUN interfaces. The 1138 Gateway further provides an attachment point of the AERO link to the 1139 non-MNP-based global topology. 1141 3.7.2. AERO Proxy Behavior 1143 When a Proxy enables an AERO interface, it assigns AERO/SPAN 1144 addresses and configures permanent neighbor cache entries the same as 1145 for Servers. The Proxy also maintains per-Client neighbor cache 1146 entries based on control message exchanges. 1148 3.7.3. AERO Client Behavior 1150 When a Client enables an AERO interface, it sends RS messages with 1151 ND/PD parameters over an ANET interface to one or more Servers in the 1152 MAP list, which return RA messages with corresponding PD parameters. 1153 (The RS/RA messages may pass through a Proxy in the case of a 1154 Client's Proxyed interface.) 1156 After the initial ND/PD message exchange, the Client assigns AERO 1157 addresses to the AERO interface based on the delegated prefix(es). 1159 The Client can then register additional ANET interfaces with the 1160 Server by sending an RS message over each ANET interface. 1162 3.7.4. AERO Relay Behavior 1164 AERO Relays need not connect directly to the AERO link, since they 1165 operate as link-layer forwarding devices instead of network layer 1166 routers. Configuration of AERO interfaces on Relays is therefore 1167 OPTIONAL, e.g., if an administrative interface is needed. Relays 1168 configure secured tunnels with Servers, Proxys and other Relays; they 1169 also configure AERO/SPAN addresses and permanent neighbor cache 1170 entries the same as Servers. Relays engage in a BGP routing protocol 1171 session with a subset of the Servers on the local segment, and with 1172 other Relays on the SPAN (see: Section 3.3). 1174 3.8. AERO Interface Neighbor Cache Maintenance 1176 Each AERO interface maintains a conceptual neighbor cache that 1177 includes an entry for each neighbor it communicates with on the AERO 1178 link per [RFC4861]. AERO interface neighbor cache entries are said 1179 to be one of "permanent", "symmetric", "asymmetric" or "proxy". 1181 Permanent neighbor cache entries are created through explicit 1182 administrative action; they have no timeout values and remain in 1183 place until explicitly deleted. AERO Servers and Proxys maintain 1184 permanent neighbor cache entries for all other Servers and Proxys 1185 within the same SPAN segment. Each entry maintains the mapping 1186 between the neighbor's network-layer AERO address and corresponding 1187 INET address. The list of all permanent neighbor cache entries for 1188 the SPAN segment is maintained in the segment's ROS list. 1190 Symmetric neighbor cache entries are created and maintained through 1191 RS/RA exchanges as specified in Section 3.15, and remain in place for 1192 durations bounded by ND/PD lifetimes. AERO Servers maintain 1193 symmetric neighbor cache entries for each of their associated 1194 Clients, and AERO Clients maintain symmetric neighbor cache entries 1195 for each of their associated Servers. The list of all Servers on the 1196 AERO link is maintained in the link's MAP list. 1198 Asymmetric neighbor cache entries are created or updated based on 1199 route optimization messaging as specified in Section 3.17, and are 1200 garbage-collected when keepalive timers expire. AERO route 1201 optimization sources (ROSs) maintain asymmetric neighbor cache 1202 entries for each of their active target Clients with lifetimes based 1203 on ND messaging constants. Asymmetric neighbor cache entries are 1204 unidirectional since only the ROS and not the target (i.e., the 1205 Client's MAP) creates an entry. 1207 Proxy neighbor cache entries are created and maintained by AERO 1208 Proxys when they process Client/Server ND/PD exchanges, and remain in 1209 place for durations bounded by ND/PD lifetimes. AERO Proxys maintain 1210 proxy neighbor cache entries for each of their associated Clients. 1211 Proxy neighbor cache entries track the Client state and the state of 1212 each of the Client's associated Servers. 1214 To the list of neighbor cache entry states in Section 7.3.2 of 1215 [RFC4861], AERO interfaces add an additional state DEPARTED that 1216 applies to symmetric and proxy neighbor cache entries for Clients 1217 that have recently departed. The interface sets a "DepartTime" 1218 variable for the neighbor cache entry to "DEPARTTIME" seconds. 1219 DepartTime is decremented unless a new ND message causes the state to 1220 return to REACHABLE. While a neighbor cache entry is in the DEPARTED 1221 state, packets destined to the target Client are forwarded to the 1222 Client's new location instead of being dropped. When DepartTime 1223 decrements to 0, the neighbor cache entry is deleted. It is 1224 RECOMMENDED that DEPARTTIME be set to the default constant value 40 1225 seconds to allow for packets in flight to be delivered while stale 1226 route optimization state may be present. 1228 When a target Server (acting as a Mobility Anchor Point (MAP)) 1229 receives a valid NS message used for route optimization, it searches 1230 for a symmetric neighbor cache entry for the target Client. The MAP 1231 then returns a solicited NA message without creating a neighbor cache 1232 entry for the ROS, but creates a target Client "Report List" entry 1233 for the ROS and sets a "ReportTime" variable for the entry to 1234 REPORTTIME seconds. The MAP resets ReportTime when it receives a new 1235 authentic NS message, and otherwise decrements ReportTime while no NS 1236 messages have been received. It is RECOMMENDED that REPORTTIME be 1237 set to the default constant value 40 seconds to allow a 10 second 1238 window so that route optimization can converge before ReportTime 1239 decrements below REACHABLETIME. 1241 When the ROS receives a solicited NA message response to its NS 1242 message, it creates or updates an asymmetric neighbor cache entry for 1243 the target network-layer and link-layer addresses. The ROS then 1244 (re)sets ReachableTime for the neighbor cache entry to REACHABLETIME 1245 seconds and uses this value to determine whether packets can be 1246 forwarded directly to the target, i.e., instead of via a default 1247 route. The ROS otherwise decrements ReachableTime while no further 1248 solicited NA messages arrive. It is RECOMMENDED that REACHABLETIME 1249 be set to the default constant value 30 seconds as specified in 1250 [RFC4861]. 1252 The ROS also uses the value MAX_UNICAST_SOLICIT to limit the number 1253 of NS keepalives sent when a correspondent may have gone unreachable, 1254 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1255 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1256 to limit the number of unsolicited NAs that can be sent based on a 1257 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1258 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1259 same as specified in [RFC4861]. 1261 Different values for DEPARTTIME, REPORTTIME, REACHABLETIME, 1262 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1263 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1264 different values are chosen, all nodes on the link MUST consistently 1265 configure the same values. Most importantly, DEPARTTIME and 1266 REPORTTIME SHOULD be set to a value that is sufficiently longer than 1267 REACHABLETIME to avoid packet loss due to stale route optimization 1268 state. 1270 3.9. AERO Interface Encapsulation and Re-encapsulation 1272 AERO interfaces encapsulate packets according to whether they are 1273 entering the AERO interface from the network layer or if they are 1274 being re-admitted into the same AERO link they arrived on. This 1275 latter form of encapsulation is known as "re-encapsulation". Note 1276 that Clients can avoid encapsulation when the first-hop access router 1277 is AERO-aware. 1279 For packets entering the AERO interface from the network layer, the 1280 AERO interface copies the "TTL/Hop Limit", "Type of Service/Traffic 1281 Class" [RFC2983], "Flow Label"[RFC6438] (for IPv6) and "Congestion 1282 Experienced" [RFC3168] values in the packet's IP header into the 1283 corresponding fields in the encapsulation header(s). 1285 For packets undergoing re-encapsulation, the AERO interface instead 1286 copies these values from the original encapsulation header into the 1287 new encapsulation header, i.e., the values are transferred between 1288 encapsulation headers and *not* copied from the encapsulated packet's 1289 network-layer header. (Note especially that by copying the TTL/Hop 1290 Limit between encapsulation headers the value will eventually 1291 decrement to 0 if there is a (temporary) routing loop.) For IPv4 1292 encapsulation/re-encapsulation, the AERO interface sets the DF bit as 1293 discussed in Section 3.13. 1295 The AERO interface encapsulates the packet according to the next hop 1296 determined in the forwarding algorithm in Section 3.12. If the next 1297 hop is reached via a secured tunnel, the AERO interface encapsulates 1298 the packet in a SPAN header and uses an INET encapsulation format 1299 specific to the secured tunnel type (see: Section 13). If the next 1300 hop is reached via an unsecured underlying interface, the AERO 1301 interface instead encapsulates the packet per the Generic UDP 1302 Encapsulation (GUE) procedures in 1304 [I-D.ietf-intarea-gue][I-D.ietf-intarea-gue-extensions], or through 1305 an alternate encapsulation format (see: Appendix A). 1307 When GUE encapsulation is used, the AERO interface next sets the UDP 1308 source port to a constant value that it will use in each successive 1309 packet it sends, and sets the UDP length field to the length of the 1310 encapsulated packet plus 8 bytes for the UDP header itself plus the 1311 length of the GUE header (or 0 if GUE direct IP encapsulation is 1312 used). For packets sent to a Server or Relay, the AERO interface 1313 sets the UDP destination port to 8060, i.e., the IANA-registered port 1314 number for AERO. For packets sent to a Client, the AERO interface 1315 sets the UDP destination port to the port value stored in the 1316 neighbor cache entry for this Client. The AERO interface then either 1317 includes or omits the UDP checksum according to the GUE 1318 specification. 1320 As the final aspect of encapsulation, the AERO interface observes the 1321 packet sizing and fragmentation considerations found in Section 3.13. 1323 3.10. AERO Interface Decapsulation 1325 AERO interfaces decapsulate packets destined either to the AERO node 1326 itself or to a destination reached via an interface other than the 1327 AERO interface the packet was received on. When the encapsulated 1328 packet arrives in multiple fragments, the AERO interface reassembles 1329 as discussed in Section 3.13. Further decapsulation steps are 1330 performed according to the appropriate encapsulation format 1331 specification. 1333 3.11. AERO Interface Data Origin Authentication 1335 AERO nodes employ simple data origin authentication procedures. In 1336 particular: 1338 o AERO Relays, Servers and Proxys accept encapsulated data packets 1339 and control messages received from secured tunnels. 1341 o AERO Servers and Proxys accept encapsulated data packets and NS 1342 messages used for Neighbor Unreachability Detection (NUD) received 1343 from an INET source found in the ROS list. 1345 o AERO Proxys and Clients accept packets that originate from within 1346 the same secured ANET. 1348 o AERO Clients and Gateways accept packets from downstream network 1349 correspondents based on ingress filtering. 1351 AERO nodes silently drop any packets that do not satisfy the above 1352 data origin authentication procedures. Further security 1353 considerations are discussed Section 13. 1355 3.12. AERO Interface Forwarding Algorithm 1357 IP packets enter a node's AERO interface either from the network 1358 layer (i.e., from a local application or the IP forwarding system) or 1359 from the link layer (i.e., from an AERO interface neighbor). All 1360 packets entering a node's AERO interface first undergo data origin 1361 authentication as discussed in Section 3.11. Packets that satisfy 1362 data origin authentication are processed further, while all others 1363 are dropped silently. 1365 Packets that enter the AERO interface from the network layer are 1366 encapsulated and forwarded into the AERO link, i.e., they are 1367 tunneled to an AERO interface neighbor. Packets that enter the AERO 1368 interface from the link layer are either re-admitted into the AERO 1369 link or forwarded to the network layer where they are subject to 1370 either local delivery or IP forwarding. In all cases, the AERO 1371 interface itself MUST NOT decrement the network layer TTL/Hop-count 1372 since its forwarding actions occur below the network layer. 1374 AERO interfaces may have multiple underlying ANET/INET interfaces 1375 and/or neighbor cache entries for neighbors with multiple Interface 1376 ID registrations (see Section 3.6). The AERO interface uses each 1377 packet's DSCP value (and/or port number) to select an outgoing ANET/ 1378 INET interface based on the node's own QoS preferences, and also to 1379 select a destination link-layer address based on the neighbor's ANET/ 1380 INET interface with the highest preference. AERO implementations 1381 SHOULD allow for QoS preference values to be modified at runtime 1382 through network management. 1384 If multiple outgoing interfaces and/or neighbor interfaces have a 1385 preference of "high", the AERO node replicates the packet and sends 1386 one copy via each of the (outgoing / neighbor) interface pairs; 1387 otherwise, the node sends a single copy of the packet via the 1388 interface with the highest preference. AERO nodes keep track of 1389 which ANET/INET interfaces are currently "reachable" or 1390 "unreachable", and only use "reachable" interfaces for forwarding 1391 purposes. 1393 For control messages, the source node encapsulates the message in 1394 SPAN/INET headers and forwards the message into the SPAN. For data 1395 packets and NS NUD messages, if there is an asymmetric neighbor 1396 within the same SPAN segment, the source uses INET encapsulation for 1397 forwarding within the local segment. Otherwise, the source node 1398 forwards the packets/messages into the SPAN. 1400 The following sections discuss the AERO interface forwarding 1401 algorithms for Clients, Proxys, Servers and Relays. In the following 1402 discussion, a packet's destination address is said to "match" if it 1403 is the same as a cached address, or if it is covered by a cached 1404 prefix (which may be encoded in an AERO address). 1406 3.12.1. Client Forwarding Algorithm 1408 When an IP packet enters a Client's AERO interface from the network 1409 layer the Client searches for an asymmetric neighbor cache entry that 1410 matches the destination. If there is a match, the Client uses one or 1411 more "reachable" neighbor interfaces in the entry for packet 1412 forwarding. If there is no asymmetric neighbor cache entry, the 1413 Client instead forwards the packet to a Server (the packet is 1414 intercepted by a Proxy if there is a Proxy on the path). 1416 When an IP packet enters a Client's AERO interface from the link- 1417 layer, if the destination matches one of the Client's MNPs or link- 1418 local addresses the Client decapsulates the packet (if necessary) and 1419 delivers it to the network layer. Otherwise, the Client drops the 1420 packet and MAY return a network-layer ICMP Destination Unreachable 1421 message subject to rate limiting (see: Section 3.14). 1423 3.12.2. Proxy Forwarding Algorithm 1425 For control messages originating from or destined to a Client, the 1426 Proxy intercepts the message and updates its proxy neighbor cache 1427 entry for the Client. The Proxy then forwards a (proxyed) copy of 1428 the control message. (For example, the Proxy forwards a proxied 1429 version of a Client's NS/RS message to the target neighbor, and 1430 forwards a proxied version of the NA/RA reply to the Client.) 1432 When the Proxy receives a data packet from a Client within the ANET, 1433 the Proxy searches for an asymmetric neighbor cache entry that 1434 matches the destination and forwards the packet as follows: 1436 o if the destination matches an asymmetric neighbor cache entry, the 1437 Proxy uses one or more "reachable" neighbor interfaces in the 1438 entry for packet forwarding via encapsulation. If the neighbor 1439 interface is in the same SPAN segment as the Proxy, the Proxy uses 1440 simple INET encapsulation; otherwise the Proxy forwards the packet 1441 into the SPAN. 1443 o else, the Proxy forwards the packet into the SPAN while using the 1444 packet's destination address as the SPAN destination address. (If 1445 the destination is an AERO address, the Proxy instead uses the 1446 corresponding Subnet Router Anycast address for Client AERO 1447 addresses and the SPAN address for administratively-provisioned 1448 AERO addresses.) 1450 When the Proxy receives an encapsulated data packet from an INET 1451 neighbor or from a secured tunnel, it searches for a proxy neighbor 1452 cache entry that matches the destination. If there is a proxy 1453 neighbor cache entry in the REACHABLE state, the Proxy decapsulates 1454 and forwards the packet to the Client; if the neighbor cache entry is 1455 in the DEPARTED state, the Proxy instead forwards the packet to the 1456 Client's Server and returns an unsolicited NA message as discussed in 1457 Section 3.19. If there is no neighbor cache entry, the Proxy 1458 discards the packet. 1460 3.12.3. Server/Gateway Forwarding Algorithm 1462 For control messages destined to a target Client's AERO address that 1463 are received from a secured tunnel, the Server (acting as a MAP) 1464 intercepts the message and sends an appropriate response on behalf of 1465 the Client. (For example, the Server sends an NA/RA message reply 1466 via the SPAN in response to an NS/RS message directed to one of its 1467 associated Clients.) If the Client's neighbor cache entry is in the 1468 DEPARTED state, however, the Server instead forwards the packet to 1469 the Client's new Server as discussed in Section 3.19. 1471 When the Server's AERO interface receives a data packet or a NUD NS 1472 from the link-layer (i.e., from an INET neighbor or from a secured 1473 tunnel), it decapsulates and processes the packet according to the 1474 network-layer destination address as follows: 1476 o if the destination matches a symmetric neighbor cache entry the 1477 Server forwards the packet according to the neighbor cache state 1478 and link-layer address information. If the neighbor cache entry 1479 is in the DEPARTED state, the Server forwards the packet to the 1480 Client's new Server. If the neighbor cache entry is in the 1481 REACHABLE state, the Server instead forwards the packet according 1482 to the cached link-layer information. If the packet is destined 1483 to the same Client from which it arrived, however, the Server 1484 forwards the packet via a different "reachable" neighbor interface 1485 than the one the packet arrived on. If there are no "reachable" 1486 neighbor interfaces, the Server drops the packet. 1488 o else, if the destination matches an asymmetric neighbor cache 1489 entry, the Server uses one or more "reachable" neighbor interfaces 1490 in the entry for packet forwarding using INET encapsulation if the 1491 neighbor is in the same SPAN segment or SPAN encapsulation 1492 otherwise. 1494 o else, if the destination is an administrative AERO address that is 1495 not assigned on the AERO interface the Server forwards the packet 1496 into the SPAN while using the SPAN address corresponding to the 1497 destination as the SPAN destination address. If the packet 1498 arrived from the SPAN, however, the Server instead drops the 1499 packet to avoid looping. 1501 o else, the Server (acting as a Gateway) releases the packet to the 1502 network layer for local delivery or IP forwarding. Based on the 1503 information in the forwarding table, the network layer may return 1504 the packet to the same AERO interface in which case further 1505 processing occurs as below. (Note that this arrangement 1506 accommodates common implementations in which the IP forwarding 1507 table is not accessible from within the AERO interface. If the 1508 AERO interface can directly access the IP forwarding table, the 1509 forwarding table lookup can instead be performed internally from 1510 within the AERO interface itself.) 1512 When the Server's AERO interface receives a data packet from the 1513 network layer, it processes the packet according to the network-layer 1514 destination address as follows: 1516 o if the destination matches a symmetric or asymmetric neighbor 1517 cache entry the Server processes the packet as above. 1519 o else, the Server forwards the packet into the SPAN. For 1520 administratively-assigned AERO address destinations, the Server 1521 uses the SPAN address corresponding to the destination as the SPAN 1522 destination address. For Client AERO address destinations, the 1523 Server uses the Subnet Router Anycast address corresponding to the 1524 destination as the SPAN destination address. For all others, the 1525 Server uses the packet's destination IP address as the SPAN 1526 destination address. 1528 3.12.4. Relay Forwarding Algorithm 1530 Relays forward packets over secured tunnels the same as any IP 1531 router. When the Relay receives an encapsulated packet via a secured 1532 tunnel, it removes the INET header and searches for a forwarding 1533 table entry that matches the destination address in the next header. 1534 The Relay then processes the packet as follows: 1536 o if the destination matches one of the Relay's own addresses, the 1537 Relay submits the packet for local delivery. 1539 o else, if the destination matches a forwarding table entry the 1540 Relay forwards the packet via a secured tunnel to the next hop. 1541 If the destination matches an MSP without matching an MNP, 1542 however, the Relay instead drops the packet and returns an ICMP 1543 Destination Unreachable message subject to rate limiting (see: 1544 Section 3.14). 1546 o else, the Relay drops the packet and returns an ICMP Destination 1547 Unreachable as above. 1549 As for any IP router, the Relay decrements the TTL/Hop Limit when it 1550 forwards the packet. If the packet is encapsulated in a SPAN header, 1551 only the Hop Limit in the SPAN header is decremented, and not the 1552 TTL/Hop Limit in the inner packet header. 1554 3.13. AERO Interface Packet Size Issues 1556 The AERO interface is the node's attachment to the AERO link. The 1557 AERO interface acts as a tunnel ingress when it sends a packet to an 1558 AERO link neighbor and as a tunnel egress when it receives a packet 1559 from an AERO link neighbor. AERO interfaces observe the packet 1560 sizing considerations for tunnels discussed in 1561 [I-D.ietf-intarea-tunnels] and as specified below. 1563 The Internet Protocol expects that IP packets will either be 1564 delivered to the destination or a suitable Packet Too Big (PTB) 1565 message returned to support the process known as IP Path MTU 1566 Discovery (PMTUD) [RFC1191][RFC8201]. However, PTB messages may be 1567 crafted for malicious purposes such as denial of service, or lost in 1568 the network [RFC2923]. This can be especially problematic for 1569 tunnels, where a condition known as a PMTUD "black hole" can result. 1570 For these reasons, AERO interfaces employ operational procedures that 1571 avoid interactions with PMTUD, including the use of fragmentation 1572 when necessary. 1574 AERO interfaces observe two different types of fragmentation. Source 1575 fragmentation occurs when the AERO interface (acting as a tunnel 1576 ingress) fragments the encapsulated packet into multiple fragments 1577 before admitting each fragment into the tunnel. Network 1578 fragmentation occurs when an encapsulated packet admitted into the 1579 tunnel by the ingress is fragmented by an IPv4 router on the path to 1580 the egress. Note that an IPv4 packet that incurs source 1581 fragmentation may also incur network fragmentation. 1583 IPv6 specifies a minimum link Maximum Transmission Unit (MTU) of 1280 1584 bytes [RFC8200]. Although IPv4 specifies a smaller minimum link MTU 1585 of 68 bytes [RFC0791], AERO interfaces also observe the IPv6 minimum 1586 for IPv4 even if encapsulated packets may incur network 1587 fragmentation. 1589 IPv6 specifies a minimum Maximum Reassembly Unit (MRU) of 1500 bytes 1590 [RFC8200], while the minimum MRU for IPv4 is only 576 bytes [RFC1122] 1591 (but, note that many standard IPv6 over IPv4 tunnel types already 1592 assume a larger MRU than the IPv4 minimum). 1594 AERO interfaces therefore configure an MTU that MUST NOT be smaller 1595 than 1280 bytes, MUST NOT be larger than the minimum MRU among all 1596 nodes on the AERO link minus the encapsulation overhead ("ENCAPS"), 1597 and SHOULD NOT be smaller than 1500 bytes. AERO interfaces also 1598 configure a Maximum Segment Unit (MSU) as the maximum-sized 1599 encapsulated packet that the ingress can inject into the tunnel 1600 without source fragmentation. The MSU value MUST NOT be larger than 1601 (MTU+ENCAPS) and MUST NOT be larger than 1280 bytes unless there is 1602 operational assurance that a larger size can traverse the link along 1603 all paths. 1605 All AERO nodes MUST configure the same MTU value for reasons cited in 1606 [RFC3819][RFC4861]; in particular, multicast support requires a 1607 common MTU value among all nodes on the link. All AERO nodes MUST 1608 configure an MRU large enough to reassemble packets up to 1609 (MTU+ENCAPS) bytes in length; nodes that cannot configure a large- 1610 enough MRU MUST NOT enable an AERO interface. For example, for an 1611 MTU of 1500 bytes (or slightly larger) an appropriate MRU might be 1612 2KB. 1614 The network layer proceeds as follows when it presents an IP packet 1615 to the AERO interface. For each IPv4 packet that is larger than the 1616 AERO interface MTU and with the DF bit set to 0, the network layer 1617 uses IPv4 fragmentation to break the packet into a minimum number of 1618 non-overlapping fragments where the first fragment is no larger than 1619 the MTU and the remaining fragments are no larger than the first. 1620 For all other IP packets, if the packet is larger than the AERO 1621 interface MTU, the network layer drops the packet and returns a PTB 1622 message to the original source. Otherwise, the network layer admits 1623 each IP packet or fragment into the AERO interface. 1625 For each IP packet admitted into the AERO interface, the interface 1626 (acting as a tunnel ingress) encapsulates the packet. If the 1627 encapsulated packet is larger than the MSU the ingress source- 1628 fragments the encapsulated packet into a minimum number of non- 1629 overlapping fragments where the first fragment is no larger than the 1630 MSU and the remaining fragments are no larger than the first. The 1631 ingress then admits each encapsulated packet or fragment into the 1632 tunnel, and for IPv4 sets the DF bit to 0 in the IP encapsulation 1633 header in case any network fragmentation is necessary. The 1634 encapsulated packets will be delivered to the egress, which 1635 reassembles them into a whole packet if necessary. 1637 Several factors must be considered when fragmentation is needed. For 1638 AERO links over IPv4, the IP ID field is only 16 bits in length, 1639 meaning that fragmentation at high data rates could result in data 1640 corruption due to reassembly misassociations [RFC6864][RFC4963]. In 1641 environments where IP fragmentation issues could result in 1642 operational problems, the ingress SHOULD employ intermediate-layer 1643 source fragmentation (see: [RFC2473] and 1644 [I-D.ietf-intarea-gue-extensions]) before appending the outer 1645 encapsulation headers to each fragment. Since the encapsulation 1646 fragment header reduces the room available for packet data, but the 1647 original source has no way to control its insertion, the ingress MUST 1648 include the fragment header length in the ENCAPS length even for 1649 packets in which the header is absent. 1651 3.14. AERO Interface Error Handling 1653 When an AERO node admits encapsulated packets into the AERO 1654 interface, it may receive link-layer or network-layer error 1655 indications. 1657 A link-layer error indication is an ICMP error message generated by a 1658 router in the INET on the path to the neighbor or by the neighbor 1659 itself. The message includes an IP header with the address of the 1660 node that generated the error as the source address and with the 1661 link-layer address of the AERO node as the destination address. 1663 The IP header is followed by an ICMP header that includes an error 1664 Type, Code and Checksum. Valid type values include "Destination 1665 Unreachable", "Time Exceeded" and "Parameter Problem" 1666 [RFC0792][RFC4443]. (AERO interfaces ignore all link-layer IPv4 1667 "Fragmentation Needed" and IPv6 "Packet Too Big" messages since they 1668 only emit packets that are guaranteed to be no larger than the IP 1669 minimum link MTU as discussed in Section 3.13.) 1671 The ICMP header is followed by the leading portion of the packet that 1672 generated the error, also known as the "packet-in-error". For 1673 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1674 much of invoking packet as possible without the ICMPv6 packet 1675 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1676 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1677 "Internet Header + 64 bits of Original Data Datagram", however 1678 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1679 ICMP datagram SHOULD contain as much of the original datagram as 1680 possible without the length of the ICMP datagram exceeding 576 1681 bytes". 1683 The link-layer error message format is shown in Figure 5 (where, "L2" 1684 and "L3" refer to link-layer and network-layer, respectively): 1686 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1687 ~ ~ 1688 | L2 IP Header of | 1689 | error message | 1690 ~ ~ 1691 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1692 | L2 ICMP Header | 1693 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1694 ~ ~ P 1695 | IP and other encapsulation | a 1696 | headers of original L3 packet | c 1697 ~ ~ k 1698 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1699 ~ ~ t 1700 | IP header of | 1701 | original L3 packet | i 1702 ~ ~ n 1703 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1704 ~ ~ e 1705 | Upper layer headers and | r 1706 | leading portion of body | r 1707 | of the original L3 packet | o 1708 ~ ~ r 1709 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1711 Figure 5: AERO Interface Link-Layer Error Message Format 1713 The AERO node rules for processing these link-layer error messages 1714 are as follows: 1716 o When an AERO node receives a link-layer Parameter Problem message, 1717 it processes the message the same as described as for ordinary 1718 ICMP errors in the normative references [RFC0792][RFC4443]. 1720 o When an AERO node receives persistent link-layer Time Exceeded 1721 messages, the IP ID field may be wrapping before earlier fragments 1722 awaiting reassembly have been processed. In that case, the node 1723 SHOULD begin including integrity checks and/or institute rate 1724 limits for subsequent packets. 1726 o When an AERO node receives persistent link-layer Destination 1727 Unreachable messages in response to encapsulated packets that it 1728 sends to one of its asymmetric neighbor correspondents, the node 1729 SHOULD process the message as an indication that a path may be 1730 failing, and MAY initiate NUD over that path. If it receives 1731 Destination Unreachable messages on many or all paths, the node 1732 SHOULD set ReachableTime for the corresponding asymmetric neighbor 1733 cache entry to 0 and allow future packets destined to the 1734 correspondent to flow through a default route. 1736 o When an AERO Client receives persistent link-layer Destination 1737 Unreachable messages in response to encapsulated packets that it 1738 sends to one of its symmetric neighbor Servers, the Client SHOULD 1739 mark the path as unusable and use another path. If it receives 1740 Destination Unreachable messages on many or all paths, the Client 1741 SHOULD associate with a new Server and release its association 1742 with the old Server as specified in Section 3.19.6. 1744 o When an AERO Server receives persistent link-layer Destination 1745 Unreachable messages in response to encapsulated packets that it 1746 sends to one of its symmetric neighbor Clients, the Server SHOULD 1747 mark the underlying path as unusable and use another underlying 1748 path. If it receives Destination Unreachable messages on multiple 1749 paths, the Server should take no further actions unless it 1750 receives an explicit ND/PD release message or if the PD lifetime 1751 expires. In that case, the Server MUST release the Client's 1752 delegated MNP, withdraw the MNP from the AERO routing system and 1753 delete the neighbor cache entry. 1755 o When an AERO Server or Proxy receives link-layer Destination 1756 Unreachable messages in response to an encapsulated packet that it 1757 sends to one of its permanent neighbors, it treats the messages as 1758 an indication that the path to the neighbor may be failing. 1759 However, the dynamic routing protocol should soon reconverge and 1760 correct the temporary outage. 1762 When an AERO Relay receives a packet for which the network-layer 1763 destination address is covered by an MSP, if there is no more- 1764 specific routing information for the destination the Relay drops the 1765 packet and returns a network-layer Destination Unreachable message 1766 subject to rate limiting. The Relay writes the network-layer source 1767 address of the original packet as the destination address and uses 1768 one of its non link-local addresses as the source address of the 1769 message. 1771 When an AERO node receives an encapsulated packet for which the 1772 reassembly buffer it too small, it drops the packet and returns a 1773 network-layer Packet Too Big (PTB) message. The node first writes 1774 the MRU value into the PTB message MTU field, writes the network- 1775 layer source address of the original packet as the destination 1776 address and writes one of its non link-local addresses as the source 1777 address. 1779 3.15. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1781 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1782 coordinated as discussed in the following Sections. 1784 3.15.1. AERO ND/PD Service Model 1786 Each AERO Server on the link configures a PD service to facilitate 1787 Client requests. Each Server is provisioned with a database of MNP- 1788 to-Client ID mappings for all Clients enrolled in the AERO service, 1789 as well as any information necessary to authenticate each Client. 1790 The Client database is maintained by a central administrative 1791 authority for the AERO link and securely distributed to all Servers, 1792 e.g., via the Lightweight Directory Access Protocol (LDAP) [RFC4511], 1793 via static configuration, etc. Therefore, no Server-to-Server PD 1794 state synchronization is necessary, and Clients can optionally hold 1795 separate PDs for the same MNPs from multiple Servers. Clients can 1796 receive new PDs from new Servers before releasing PDs received from 1797 existing Servers for service continuity. Clients receive the same 1798 service regardless of the Servers they select, although selecting 1799 Servers that are topologically nearby may provide better routing. 1801 AERO Clients and Servers use ND messages to maintain neighbor cache 1802 entries. AERO Servers configure their AERO interfaces as advertising 1803 interfaces, and therefore send unicast RA messages with configuration 1804 information in response to a Client's RS message. Thereafter, 1805 Clients send additional RS messages to refresh prefix and/or router 1806 lifetimes. 1808 AERO Clients and Servers include PD parameters in RS/RA messages (see 1809 [I-D.templin-6man-dhcpv6-ndopt] for ND/PD alternatives). The unified 1810 ND/PD messages are exchanged between Client and Server according to 1811 the prefix management schedule required by the PD service. If the 1812 Client knows its MNP in advance, it can include its AERO address as 1813 the source address of an RS message and with an SLLAO with a valid 1814 Prefix Length for the MNP. If the Server (and Proxy) accept the 1815 Client's MNP assertion, they inject the prefix into the routing 1816 system and establish the necessary neighbor cache state. 1818 The following sections specify the Client and Server behavior. 1820 3.15.2. AERO Client Behavior 1822 AERO Clients can discover the INET and AERO addresses of Servers in 1823 the MAP list via static configuration (e.g., from a flat-file map of 1824 Server addresses and locations), or through an automated means such 1825 as Domain Name System (DNS) name resolution [RFC1035]. In the 1826 absence of other information, the Client can resolve the DNS Fully- 1827 Qualified Domain Name (FQDN) "linkupnetworks.[domainname]" where 1828 "linkupnetworks" is a constant text string and "[domainname]" is a 1829 DNS suffix for the AERO link (e.g., "example.com"). Alternatively, 1830 the Client can discover the Server's address through a multicast RS 1831 as described below. 1833 To associate with a Server, the Client acts as a requesting router to 1834 request MNPs. The Client prepares an RS message with PD parameters 1835 (e.g., with an SLLAO with non-zero Prefix Length) and SHOULD include 1836 a Nonce and Timestamp option if the Client needs to correlate RA 1837 replies. If the Client already knows the Server's AERO address, it 1838 includes the AERO address as the network-layer destination address; 1839 otherwise, it includes all-routers multicast (ff02::2) as the 1840 network-layer destination address. If the Client already knows its 1841 own AERO address, it uses the AERO address as the network-layer 1842 source address; otherwise, it uses the unspecified AERO address 1843 (fe80::ffff:ffff) as the network-layer source address. 1845 The Client next includes an SLLAO in the RS message formatted as 1846 described in Section 3.6 to register its link-layer information with 1847 the Server. The SLLAO corresponding to the ANET interface over which 1848 the Client will send the RS message MUST set S to 1. The Client MAY 1849 include additional SLLAOs specific to other underlying interfaces, 1850 but if so it MUST set their S, Port Number and Link Layer Address 1851 fields to 0. If the Client is connected to an ANET for which 1852 encapsulation is required, the Client finally encapsulates the RS 1853 message in an ANET header with its own ANET address as the source 1854 address and the INET address of the Server as the destination. 1856 The Client then sends the RS message (either via a VPN for VPNed 1857 interfaces, via a Proxy for proxyed interfaces or via the SPAN for 1858 native interfaces) and waits for an RA message reply (see 1859 Section 3.15.3) while retrying up to MAX_RTR_SOLICITATIONS times 1860 until an RA is received. If the Client receives no RAs, or if it 1861 receives an RA with Router Lifetime set to 0, the Client SHOULD 1862 abandon this Server and try another Server. Otherwise, the Client 1863 processes the PD information found in the RA message. 1865 Next, the Client creates a symmetric neighbor cache entry with the 1866 Server's AERO address as the network-layer address and the address in 1867 the first SLLAO as the Server's INET address. The Client records the 1868 RA Router Lifetime field value in the neighbor cache entry as the 1869 time for which the Server has committed to maintaining the MNP in the 1870 routing system. The Client then autoconfigures AERO addresses for 1871 each of the delegated MNPs and assigns them to the AERO interface. 1872 The Client also caches any MSPs included in Route Information Options 1873 (RIOs) [RFC4191] as MSPs to associate with the AERO link, and assigns 1874 the MTU value in the MTU option to its AERO interface while 1875 configuring an appropriate MRU. 1877 The Client then registers additional ANET interfaces with the Server 1878 by sending RS messages via each additional ANET interface. The RS 1879 messages include the same parameters as for the initial RS/RA 1880 exchange, but with destination address set to the Server's AERO 1881 address and with an SLLAO specific to the ANET interface. (The RS 1882 messages include PD parameters the same as for the initial exchange 1883 so that the additional ANETs can register the PD information.) 1885 The Client examines the X and N bits in the SLLAO with S set to 1 in 1886 each RA message it receives. If X is 1 the Client infers that there 1887 is a Proxy on the path, and if N is 1 the Client infers that there is 1888 a NAT on the path. If N is 1, the Client SHOULD set Port Number and 1889 Link-Layer Address to 0 in the first S/TLLAO of any subsequent ND 1890 messages it sends to the Server over that link. 1892 Following autoconfiguration, the Client sub-delegates the MNPs to its 1893 attached EUNs and/or the Client's own internal virtual interfaces as 1894 described in [I-D.templin-v6ops-pdhost] to support the Client's 1895 downstream attached "Internet of Things (IoT)". The Client 1896 subsequently maintains its MNP delegations through each of its 1897 Servers by sending additional RS messages before Router Lifetime 1898 expires. 1900 After the Client registers its ANET interfaces, it may wish to change 1901 one or more registrations, e.g., if an ANET interface changes address 1902 or becomes unavailable, if QoS preferences change, etc. To do so, 1903 the Client prepares an RS message to send over any available ANET 1904 interface. The RS MUST include an SLLAO with S set to 1 for the 1905 selected ANET interface and MAY include any additional SLLAOs 1906 specific to other ANET interfaces. The Client includes fresh P(i) 1907 values in each SLLAO to update the Server's neighbor cache entry. If 1908 the Client wishes to update only the P(i) values, it sets the Port 1909 Number and Link-Layer Address fields to 0. If the Client wishes to 1910 disable the underlying interface, it sets D to 1. When the Client 1911 receives the Server's RA response, it has assurance that the Server 1912 has been updated with the new information. 1914 If the Client wishes to associate with multiple Servers, it repeats 1915 the same procedures above for each additional Server. If the Client 1916 wishes to discontinue use of a Server it issues an RS message over 1917 any underlying interface with an SLLAO with R set to 1 . When the 1918 Server processes the message, it releases the MNP, sets the symmetric 1919 neighbor cache entry state for the Client to DEPARTED, withdraws the 1920 IP route from the routing system and returns an RA reply with Router 1921 Lifetime set to 0. 1923 3.15.3. AERO Server Behavior 1925 AERO Servers act as IP routers and support a PD service for Clients. 1926 Servers arrange to add their AERO and INET addresses to a static map 1927 of Server addresses for the link and/or the DNS resource records for 1928 the FQDN "linkupnetworks.[domainname]" before entering service. The 1929 list of Server addresses should be geographically and/or 1930 topologically referenced, and forms the MAP list for the AERO link. 1932 When a Server receives a prospective Client's RS message on its AERO 1933 interface, it SHOULD return an immediate RA reply with Router 1934 Lifetime set to 0 if it is currently too busy or otherwise unable to 1935 service the Client. Otherwise, the Server authenticates the RS 1936 message and processes the PD parameters. The Server first determines 1937 the correct MNPs to delegate to the Client by searching the Client 1938 database. When the Server delegates the MNPs, it also creates an IP 1939 forwarding table entry for each MNP so that the MNPs are propagated 1940 into the routing system (see: Section 3.3). For IPv6, the Server 1941 creates a single IPv6 forwarding table entry for each MNP. For IPv4, 1942 the Server creates both an IPv4 forwarding table entry and an IPv6 1943 forwarding table entry with the IPv4-mapped IPv6 address 1944 corresponding to the IPv4 address. 1946 The Server next creates a symmetric neighbor cache entry for the 1947 Client using the base AERO address as the network-layer address and 1948 with lifetime set to no more than the smallest PD lifetime. Next, 1949 the Server updates the neighbor cache entry by recording the 1950 information in each SLLAO in the RS indexed by the Interface ID and 1951 including the Port Number, Link Layer Address and P(i) values. For 1952 the SLLAO with S set to 1, however, the Server records the actual 1953 INET header source addresses instead of those that appear in the 1954 SLLAO in case there was a NAT in the path. The Server also records 1955 the value of the X bit to indicate whether there is a Proxy on the 1956 path. 1958 Next, the Server prepares an RA message using its AERO address as the 1959 network-layer source address and the network-layer source address of 1960 the RS message as the network-layer destination address. The Server 1961 includes the delegated MNPs, any other PD parameters and an SLLAO 1962 with the Link Layer Address set to the Server's SPAN address and with 1963 Interface ID set to 0xffff. The Server then includes one or more 1964 RIOs that encode the MSPs for the AERO link, plus an MTU option for 1965 the link MTU (see Section 3.13). The Server finally encapsulates the 1966 message in a SPAN header with source address set to its own SPAN 1967 address and destination address set to the Client's (or Proxy's) SPAN 1968 address, then forwards the message into the SPAN. 1970 After the initial RS/RA exchange, the Server maintains the symmetric 1971 neighbor cache entry for the Client. If the Client (or Proxy) issues 1972 additional NS/RS messages, the Server resets ReachableTime. If the 1973 Client (or Proxy) issues an RS with PD release parameters (e.g., by 1974 including an SLLAO with R set to 1), or if the Client becomes 1975 unreachable, the Server sets the Client's symmetric neighbor cache 1976 entry to the DEPARTED state and withdraws the IP routes from the AERO 1977 routing system. 1979 The Server processes these and any other Client ND/PD messages, and 1980 returns an NA/RA reply. The Server may also issue unsolicited RA 1981 messages, e.g., with PD reconfigure parameters to cause the Client to 1982 renegotiate its PDs, with Router Lifetime set to 0 if it can no 1983 longer service this Client, etc. Finally, If the symmetric neighbor 1984 cache entry is in the DEPARTED state, the Server deletes the entry 1985 after DepartTime expires. 1987 3.15.3.1. Lightweight DHCPv6 Relay Agent (LDRA) 1989 When DHCPv6 is used as the ND/PD service back end, AERO Clients and 1990 Servers are always on the same link (i.e., the AERO link) from the 1991 perspective of DHCPv6. However, in some implementations the DHCPv6 1992 server and ND function may be located in separate modules. In that 1993 case, the Server's AERO interface module can act as a Lightweight 1994 DHCPv6 Relay Agent (LDRA)[RFC6221] to relay PD messages to and from 1995 the DHCPv6 server module. 1997 When the LDRA receives an authentic RS message, it extracts the PD 1998 message parameters and uses them to construct an IPv6/UDP/DHCPv6 1999 message. It sets the IPv6 source address to the source address of 2000 the RS message, sets the IPv6 destination address to 2001 'All_DHCP_Relay_Agents_and_Servers' and sets the UDP fields to values 2002 that will be understood by the DHCPv6 server. 2004 The LDRA then wraps the message in a DHCPv6 'Relay-Forward' message 2005 header and includes an 'Interface-Id' option that includes enough 2006 information to allow the LDRA to forward the resulting Reply message 2007 back to the Client (e.g., the Client's link-layer addresses, a 2008 security association identifier, etc.). The LDRA also wraps the 2009 information in all of the SLLAOs from the RS message into the 2010 Interface-Id option, then forwards the message to the DHCPv6 server. 2012 When the DHCPv6 server prepares a Reply message, it wraps the message 2013 in a 'Relay-Reply' message and echoes the Interface-Id option. The 2014 DHCPv6 server then delivers the Relay-Reply message to the LDRA, 2015 which discards the Relay-Reply wrapper and IPv6/UDP headers, then 2016 uses the DHCPv6 message to construct an RA response to the Client. 2017 The Server uses the information in the Interface-Id option to prepare 2018 the RA message and to cache the link-layer addresses taken from the 2019 SLLAOs echoed in the Interface-Id option. 2021 3.16. The AERO Proxy 2023 Clients may connect to ANETs that do not permit direct communications 2024 to Servers in outside INETs. In that case, the ANET can employ an 2025 AERO Proxy. The Proxy is located at the ANET/INET border and listens 2026 for RS messages originating from or RA messages destined to ANET 2027 Clients. The Proxy acts on these control messages as follows: 2029 o when the Proxy receives an RS message from a new ANET Client, it 2030 first authenticates the message then examines the network-layer 2031 destination address. If the destination address is a Server's 2032 AERO address, the Proxy proceeds to the next step. Otherwise, if 2033 the destination is all-routers multicast the Proxy selects a 2034 "nearby" Server that is likely to be a good candidate to serve the 2035 Client and replaces the destination address with the Server's AERO 2036 address. Next, the Proxy creates a proxy neighbor cache entry and 2037 caches the Client and Server addresses along with any identifying 2038 information including Transaction IDs, Client Identifiers, Nonce 2039 values, etc. The Proxy then examines the address in the RS 2040 message SLLAO with S set to 1. If the address is different than 2041 the Client's ANET address, the Proxy notes that the Client is 2042 behind a NAT. The Proxy then sets the X to 1 and changes the Link 2043 Layer Address to its own SPAN address. The Proxy finally 2044 encapsulates the RS message in a SPAN header with destination set 2045 to the Server's SPAN address then forwards the message into the 2046 SPAN. 2048 o when the Server receives the RS message, it authenticates the 2049 message then creates or updates a symmetric neighbor cache entry 2050 for the Client with the Proxy's SPAN address as the link-layer 2051 address. The Server then sends an RA message with a single SLLAO 2052 back to the Proxy via the SPAN. 2054 o when the Proxy receives the RA message, it matches the message 2055 with the RS that created the proxy neighbor cache entry. The 2056 Proxy then caches the PD route information as a mapping from the 2057 Client's MNPs to the Client's ANET address, and sets the neighbor 2058 cache entry state to REACHABLE. The Proxy then changes the SLLAO 2059 Link Layer Address to its own ANET address, sets X to 1, sets N to 2060 1 if the Client is behind a NAT, then re-encapsulates the RA 2061 message in an ANET header and forwards it to the Client. 2063 After the initial RS/RA exchange, the Proxy forwards any Client data 2064 packets for which there is no matching asymmetric neighbor cache 2065 entry to a Relay via the SPAN. Finally, the Proxy forwards any 2066 Client data destined to an asymmetric neighbor cache target directly 2067 to the target according to the link-layer information - the process 2068 of establishing asymmetric neighbor cache entries is specified in 2069 Section 3.17. 2071 While the Client is still attached to the ANET, the Proxy 2072 periodically sends NS/RS messages to update each Server's symmetric 2073 neighbor cache entries on behalf of the Client and/or to convey QoS 2074 updates. If the Server ceases to send solicited NA/RA responses, the 2075 Proxy marks the Server as unreachable and sends an unsolicited RA 2076 with Router Lifetime set to zero to inform the Client that this 2077 Server is no longer able to provide Service. If the Client becomes 2078 unreachable, the Proxy sets the neighbor cache entry state to 2079 DEPARTED and sends an RS message to each Server with an SLLAO with D 2080 set to 1 and with Interface ID set to the Client's interface ID so 2081 that the Server will de-register this Interface ID. Although the 2082 Proxy engages in these ND exchanges on behalf of the Client, the 2083 Client can also send ND messages on its own behalf, e.g., if it is in 2084 a better position than the Proxy to convey QoS changes, etc. 2086 In some ANETs that employ a Proxy, the Client's MNP can be injected 2087 into the ANET routing system. In that case, the Client can send data 2088 messages without encapsulation so that the ANET native routing system 2089 transports the unencapsulated packets to the Proxy. This can be very 2090 beneficial, e.g., if the Client connects to the ANET via low-end data 2091 links such as some aviation wireless links. 2093 If the first-hop ANET access router is AERO-aware, the Client can 2094 avoid encapsulation for both its control and data messages. When the 2095 Client connects to the link, it can send an unencapsulated RS message 2096 with source address set to its AERO address and with destination 2097 address set to the AERO address of the Client's selected Server or to 2098 all-routers multicast. The Client includes an SLLAO with Interface 2099 ID, Prefix Length and P(i) information but with Port Number and Link- 2100 Layer Address set to 0. 2102 The Client then sends the unencapsulated RS message, which will be 2103 intercepted by the AERO-Aware access router. The access router then 2104 encapsulates the RS message in an ANET header with its own address as 2105 the source address and the address of a Proxy as the destination 2106 address. The access router further remembers the address of the 2107 Proxy so that it can encapsulate future data packets from the Client 2108 via the same Proxy. If the access router needs to change to a new 2109 Proxy, it simply sends another RS message toward the Server via the 2110 new Proxy on behalf of the Client. 2112 In some cases, the access router and Proxy may be one and the same 2113 node. In that case, the node would be located on the same physical 2114 link as the Client, but its message exchanges with the Server would 2115 need to pass through a security gateway at the ANET/INET border. The 2116 method for deploying access routers and Proxys (i.e. as a single node 2117 or multiple nodes) is an ANET-local administrative consideration. 2119 3.17. AERO Route Optimization 2121 While data packets are flowing between a source and target node, 2122 route optimization SHOULD be used. Route optimization is initiated 2123 by the first eligible Route Optimization Source (ROS) closest to the 2124 source as follows: 2126 o For Clients on VPNed, NATed and Direct interfaces, the Server is 2127 the ROS. 2129 o For Clients on Proxyed interfaces, the Proxy is the ROS. 2131 o For Clients on native interfaces, the Client itself is the ROS. 2133 o For correspondent nodes on INET/EUN interfaces serviced by a 2134 Gateway, the Gateway is the ROS. 2136 The route optimization procedure is conducted between the ROS and the 2137 target Server/Gateway acting as a Route Optimization Responder (ROR) 2138 in the same manner as for IPv6 ND Address Resolution and using the 2139 same NS/NA messaging. The target may either be a MNP Client serviced 2140 by a Server, or a non-MNP correspondent reachable via a Gateway. 2142 The procedures are specified in the following sections. 2144 3.17.1. Route Optimization Initiation 2146 While data packets are flowing from the source node toward a target 2147 node, the ROS performs address resolution by sending an NS message to 2148 receive a solicited NA message from the ROR. 2150 When the ROS sends an NS, it includes the AERO address of the ROS as 2151 the source address (e.g., fe80::1) and the AERO address corresponding 2152 to the data packet's destination address as the destination address 2153 (e.g., if the destination address is 2001:db8:1:2::1 then the 2154 corresponding AERO address is fe80::2001:db8:1:2). The NS message 2155 includes an SLLAO with Link Layer Address set to the SPAN address of 2156 the ROS and with all other fields set to 0. The message SHOULD also 2157 include a Nonce and Timestamp option if the ROS needs to correlate NA 2158 replies. 2160 The ROS then encapsulates the NS message in a SPAN header with source 2161 set to its own SPAN address and destination set to the data packet's 2162 destination address, then sends it into the SPAN without decrementing 2163 the network-layer TTL/Hop Limit field. 2165 3.17.2. Relaying the NS 2167 When the Relay receives the (double-encapsulated) NS message from the 2168 ROS, it discards the INET header and determines that the ROR is the 2169 next hop by consulting its standard IPv6 forwarding table for the 2170 SPAN header destination address. The Relay then forwards the SPAN 2171 message toward the ROR the same as for any IP router. The final-hop 2172 Relay in the SPAN will deliver the message via a secured tunnel to 2173 the ROR. 2175 3.17.3. Processing the NS and Sending the NA 2177 When the ROR receives the (double-encapsulated) NS message, it 2178 examines the AERO destination address to determine whether it has a 2179 neighbor cache entry and/or route that matches the target; if not, it 2180 drops the NS message and returns from processing. Next, if the 2181 target belongs to an MNP Client neighbor in the DEPARTED state the 2182 ROR changes the NS message SPAN destination address to the address of 2183 the Client's new Server, forwards the message into the SPAN and 2184 returns from processing. If the target belongs to an MNP Client 2185 neighbor in the REACHABLE state, the ROR instead adds the AERO source 2186 address to the target Client's Report List with time set to 2187 ReportTime. If the target belongs to a non-MNP route, the ROR 2188 continues processing without adding an entry to the Report List. 2190 The ROR then prepares a solicited NA message to send back to the ROS 2191 but does not create a neighbor cache entry. The ROR sets the NA 2192 source address to the destination AERO address of the NS, and 2193 includes the Nonce value received in the NS plus the current 2194 Timestamp. The ROR next includes a TLLAO with Interface ID set to 2195 0xffff, with S set to 1, with all P(i) values set to "low", and with 2196 Link Layer Address set to the ROR's SPAN address. If the target 2197 belongs to an MNP Client, the ROR sets the Prefix Length to the MNP 2198 prefix length; otherwise, it sets Prefix Length to the maximum of the 2199 non-MNP prefix length and 64. (Note that a /64 limit is imposed to 2200 avoid causing the ROS to set short prefixes (e.g., "default") that 2201 would match destinations for which the routing system includes more- 2202 specific prefixes. Note also that prefix lengths longer than /64 are 2203 out of scope for this specification.) 2205 If the target belongs to an MNP Client, the ROR next includes 2206 additional TLLAOs for all of the target Client's Interface IDs. For 2207 NATed, VPNed and Direct interfaces, the TLLAO Link Layer Addresses 2208 are the SPAN address of the ROR. For Proxyed and native interfaces, 2209 the TLLAO Link Layer Addresses are the SPAN addresses of the Proxys 2210 and the Client's native interfaces. The ROR finally encapsulates the 2211 NA message in a SPAN header with source set to its own SPAN address 2212 and destination set to the source SPAN address of the NS message, 2213 then forwards the message into the SPAN without decrementing the 2214 network-layer TTL/Hop Limit field. 2216 3.17.4. Relaying the NA 2218 When the Relay receives the (double-encapsulated) NA message from the 2219 ROR, it discards the INET header and determines that the ROS is the 2220 next hop by consulting its standard IPv6 forwarding table for the 2221 SPAN header destination address. The Relay then forwards the SPAN- 2222 encapsulated NA message toward the ROS the same as for any IPv6 2223 router. The final-hop Relay in the SPAN will deliver the message via 2224 a secured tunnel to the ROS. 2226 3.17.5. Processing the NA 2228 When the ROS receives the (double-encapsulated) solicited NA message, 2229 it discards the INET and SPAN headers. The ROS next verifies the 2230 Nonce and Timestamp values, then creates an asymmetric neighbor cache 2231 entry for the ROR and caches all information found in the solicited 2232 NA TLLAOs. The ROS finally sets the asymmetric neighbor cache entry 2233 lifetime to ReachableTime seconds. 2235 3.17.6. Route Optimization Maintenance 2237 Following route optimization, the ROS forwards future data packets 2238 destined to the target via the addresses found in the cached link- 2239 layer information. The route optimization is shared by all sources 2240 that send packets to the target via the ROS, i.e., and not just the 2241 source on behalf of which the route optimization was initiated. 2243 While new data packets destined to the target are flowing through the 2244 ROS, it sends additional NS messages to the ROR before ReachableTime 2245 expires to receive a fresh solicited NA message the same as described 2246 in the previous sections. (Route optimization refreshment strategies 2247 are an implementation matter, with a non-normative example given in 2248 Appendix D). 2250 The ROS then updates the asymmetric neighbor cache entry to refresh 2251 ReachableTime, while (for MNP destinations) the ROR adds or updates 2252 the ROS address to the target Client's Report List and with time set 2253 to ReportTime. While no data packets are flowing, the ROS instead 2254 allows ReachableTime for the asymmetric neighbor cache entry to 2255 expire. When ReachableTime expires, the ROS deletes the asymmetric 2256 neighbor cache entry. Future data packets flowing through the ROS 2257 will again trigger a new route optimization exchange while initial 2258 data packets travel over a suboptimal route via Servers and/or 2259 Relays. 2261 The ROS may also receive unsolicited NA messages from the ROR at any 2262 time. If there is an asymmetric neighbor cache entry for the target, 2263 the ROS updates the link-layer information but does not update 2264 ReachableTime since the receipt of an unsolicited NA does not confirm 2265 that the forward path is still working. If there is no asymmetric 2266 neighbor cache entry, the route optimization source simply discards 2267 the unsolicited NA. Cases in which unsolicited NA messages are 2268 generated are specified in Section 3.19. 2270 In this arrangement, the ROS holds an asymmetric neighbor cache entry 2271 for the ROR, but the ROR does not hold an asymmetric neighbor cache 2272 entry for the ROS. The route optimization neighbor relationship is 2273 therefore asymmetric and unidirectional. If the target node also has 2274 packets to send back to the source node, then a separate route 2275 optimization procedure is performed in the reverse direction. But, 2276 there is no requirement that the forward and reverse paths be 2277 symmetric. 2279 3.18. Neighbor Unreachability Detection (NUD) 2281 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2282 [RFC4861]. NUD is performed either reactively in response to 2283 persistent link-layer errors (see Section 3.14) or proactively to 2284 confirm reachability. The NUD algorithm may further be seeded by ND 2285 hints of forward progress, but care must be taken to avoid inferring 2286 reachability based on spoofed information. 2288 When an ROR directs an ROS to a neighbor with one or more target 2289 link-layer addresses, the ROS can proactively test each direct path 2290 by sending an initial NS message to elicit a solicited NA response. 2291 While testing the paths, the ROS can optionally continue sending 2292 packets via the SPAN, maintain a small queue of packets until target 2293 reachability is confirmed, or (optimistically) allow packets to flow 2294 via the direct paths. In any case, the ROS should only consider the 2295 neighbor unreachable if NUD fails over multiple target link-layer 2296 address paths. 2298 When a ROS sends an NS message used for NUD, it uses its AERO 2299 addresses as the IPv6 source address and the AERO address 2300 corresponding to a target link-layer address as the destination. For 2301 each target link-layer address, if the address is not located within 2302 the same AERO link segment the source node encapsulates the NS 2303 message in a SPAN header with its own SPAN address as the source and 2304 the SPAN address of the target as the destination, then forwards the 2305 message into the SPAN. 2307 If the target address is located within the same segment, however, 2308 the source node omits the SPAN header and encapsulates the message in 2309 an INET header with its own INET address as the source and the INET 2310 address of the target as the destination, then sends the message 2311 directly to the target. 2313 Paths that pass NUD tests are marked as "reachable", while those that 2314 do not are marked as "unreachable". These markings inform the AERO 2315 interface forwarding algorithm specified in Section 3.12. 2317 Proxys can perform NUD to verify Server reachability on behalf of 2318 their proxyed Clients so that the Clients need not engage in NUD 2319 messaging themselves. 2321 3.19. Mobility Management and Quality of Service (QoS) 2323 AERO is a Distributed Mobility Management (DMM) service. Each Server 2324 is responsible for only a subset of the Clients on the AERO link, as 2325 opposed to a Centralized Mobility Management (CMM) service where 2326 there is a single network mobility service for all Clients. Clients 2327 coordinate with their associated Servers via RS/RA exchanges to 2328 maintain the DMM profile, and the AERO routing system tracks all 2329 current Client/Server peering relationships. 2331 Servers provide a Mobility Anchor Point (MAP) for their dependent 2332 Clients. Clients are responsible for maintaining neighbor 2333 relationships with their Servers through periodic RS/RA exchanges, 2334 which also serves to confirm neighbor reachability. When a Client's 2335 underlying interface address and/or QoS information changes, the 2336 Client is responsible for updating the Server with this new 2337 information. Note that for Proxyed interfaces, however, the Proxy 2338 can perform the RS/RA exchanges on the Client's behalf. 2340 Mobility management considerations are specified in the following 2341 sections. 2343 3.19.1. Mobility Update Messaging 2345 Servers acting as MAPs accommodate mobility and/or QoS change events 2346 by sending an unsolicited NA message to each ROS in the target 2347 Client's Report List. When a MAP sends an unsolicited NA message, it 2348 sets the IPv6 source address to the Client's AERO address and sets 2349 the IPv6 destination address to all-nodes multicast (ff02::1). The 2350 MAP also includes a TLLAO with Interface ID 0xffff, S set to 1 and 2351 Link Layer address set to the MAP's SPAN address, and includes 2352 additional TLLAOs for all of the target Client's Interface IDs with 2353 Link Layer Addresses set to the corresponding SPAN addresses. The 2354 MAP finally encapsulates the message in a SPAN header with source set 2355 to its own SPAN address and destination set to the SPAN address of 2356 the ROS, then sends the message into the SPAN. 2358 As for the hot-swap of interface cards discussed in Section 7.2.6 of 2359 [RFC4861], the transmission and reception of unsolicited NA messages 2360 is unreliable but provides a useful optimization. In well-connected 2361 Internetworks with robust data links unsolicited NA messages will be 2362 delivered with high probability, but in any case the MAP can 2363 optionally send up to MAX_NEIGHBOR_ADVERTISEMENT unsolicited NAs to 2364 each ROS to increase the likelihood that at least one will be 2365 received. 2367 When an ROS receives an unsolicited NA message, it ignores the 2368 message if there is no existing neighbor cache entry for the Client. 2369 Otherwise, it uses the included TLLAOs to update the Link Layer 2370 Address and QoS information in the neighbor cache entry, but does not 2371 reset ReachableTime since the receipt of an unsolicited NA message 2372 from the target Server does not provide confirmation that any forward 2373 paths to the target Client are working. 2375 If unsolicited NA messages are lost, the ROS may be left with stale 2376 address and/or QoS information for the Client for up to ReachableTime 2377 seconds. During this time, the ROS can continue sending packets to 2378 the target Client according to its current neighbor cache information 2379 but may receive persistent unsolicited NA messages as discussed in 2380 Section 3.19.2. 2382 3.19.2. Forwarding Packets on Behalf of Departed Clients 2384 When a Server acting as a MAP receives packets with destination 2385 addresses that match a symmetric neighbor cache entry in the DEPARTED 2386 state, it forwards the packets to the SPAN address corresponding to 2387 the Client's new MAP. If the ROS is in the Report List, the old MAP 2388 also sends an unsolicited NA message via the SPAN (subject to rate 2389 limiting) with a TLLAO with Interface ID 0xffff and with R set to 1. 2390 When the ROS receives the NA, it SHOULD delete the asymmetric 2391 neighbor cache entry and re-initiate route optimization. 2393 When a Proxy receives packets with destination addresses that match a 2394 proxy neighbor cache entry in the DEPARTED state, it forwards the 2395 packets to one of the target Client's MAPs. If the ROS is not one of 2396 its proxy neighbor Clients, the Proxy also returns an unsolicited NA 2397 message via the SPAN (subject to rate limiting) with a single TLLAO 2398 with the target Client's Interface ID and with D set to 1. The ROS 2399 will then realize that it needs to mark its neighbor cache entry 2400 Interface ID for the Proxy as "unreachable", and SHOULD re-initiate 2401 route optimization while continuing to forward packets according to 2402 the remaining neighbor cache entry state. 2404 3.19.3. Announcing Link-Layer Address and/or QoS Preference Changes 2406 When a Client needs to change its ANET addresses and/or QoS 2407 preferences (e.g., due to a mobility event), either the Client or its 2408 Proxys send RS messages to its Servers via the SPAN with SLLAOs that 2409 include the new Client Port Number, Link Layer Address and P(i) 2410 values. If the RS messages are sent solely for the purpose of 2411 updating QoS preferences, Port Number and Link-Layer Address are set 2412 to 0. 2414 Up to MAX_RTR_SOLICITATION RS messages MAY be sent in parallel with 2415 sending actual data packets in case one or more RAs are lost. If all 2416 RAs are lost, the Client SHOULD re-associate with a new Server. 2418 When the Server receives the Client's changes, it sends unsolicited 2419 NA messages to all nodes in the Report List the same as described in 2420 the previous section. 2422 3.19.4. Bringing New Links Into Service 2424 When a Client needs to bring new ANET interfaces into service (e.g., 2425 when it activates a new data link), it sends RS messages to its 2426 Servers via the ANET interface with SLLAOs that include the new 2427 Client Link Layer Address information. 2429 3.19.5. Removing Existing Links from Service 2431 When a Client needs to remove existing ANET interfaces from service 2432 (e.g., when it de-activates an existing data link), it sends RS 2433 messages to its Servers with SLLAOs with D set to 1. 2435 If the Client needs to send RS messages over an ANET interface other 2436 than the one being removed from service, it MUST include a current 2437 SLLAO with S set to 1 for the sending interface and include 2438 additional SLLAOs with S set to 0 for any ANET interfaces being 2439 removed from service. 2441 3.19.6. Moving to a New Server 2443 When a Client associates with a new Server, it performs the Client 2444 procedures specified in Section 3.15.2. The Client then sends an RS 2445 message over any working ANET interface with destination set to the 2446 old Server's AERO address and with an SLLAO with R set to 1 to fully 2447 release itself from the old Server. The SLLAO also includes the SPAN 2448 address of the new Server in the Link Layer Address. If the Client 2449 does not receive an RA reply after MAX_RTR_SOLICITATIONS attempts 2450 over multiple ANET interfaces, the old Server may have failed and the 2451 Client should discontinue its release attempts. 2453 When the old Server processes the RS, it sends unsolicited NA 2454 messages with a single TLLAO with Interface ID set to 0xffff and with 2455 R and S set to 1 to all ROSs in the Client's Report List. The Server 2456 also changes the symmetric neighbor cache entry state to DEPARTED, 2457 sets the link-layer address of the Client to the address found in the 2458 RS SLLAO (i.e., the SPAN address of the new Server), and sets a timer 2459 to DepartTime seconds. The old Server then returns an RA message to 2460 the Client with Router Lifetime set to 0. After DepartTime seconds 2461 expires, the old Server deletes the symmetric neighbor cache entry. 2463 When the Client receives the RA message with Router Lifetime set to 2464 0, it still must inform each of its remaining Proxys that it has 2465 released the old Server from service. To do so, it sends an RS over 2466 each remaining proxyed ANET interface with destination set to the old 2467 Server's AERO address, with R set to 1 and with no SLLAO. The Proxy 2468 will mark this Server as DEAPARTED and return an immediate RA without 2469 first performing an RS/RA exchange with the old Server. 2471 Clients SHOULD NOT move rapidly between Servers in order to avoid 2472 causing excessive oscillations in the AERO routing system. Examples 2473 of when a Client might wish to change to a different Server include a 2474 Server that has gone unreachable, topological movements of 2475 significant distance, movement to a new geographic region, movement 2476 to a new segment, etc. 2478 3.20. Multicast 2480 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 2481 [RFC3810] proxy service for its EUNs and/or hosted applications 2482 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 2483 ANET interfaces for which group membership is required. The IGMP/MLD 2484 messages may be further forwarded by a first-hop ANET access router 2485 acting as an IGMP/MLD-snooping switch [RFC4541], then ultimately 2486 delivered to an AERO Proxy/Server acting as a Protocol Independent 2487 Multicast - Sparse-Mode (PIM-SM, or simply "PIM") Designated Router 2488 (DR) [RFC7761]. AERO Gateways also act as PIM routers (i.e., the 2489 same as AERO Proxys/Servers) on behalf of nodes on INET/EUN networks. 2490 The behaviors identified in the following sections correspond to 2491 Source-Specific Multicast (SSM) and Any-Source Multicast (ASM) 2492 operational modes. 2494 3.20.1. Source-Specific Multicast (SSM) 2496 When an ROS (i.e., an AERO Proxy/Server/Gateway) "X" acting as PIM 2497 router receives a Join/Prune message from a node on its downstream 2498 interfaces containing one or more ((S)ource, (G)roup) pairs, it 2499 updates its Multicast Routing Information Base (MRIB) accordingly. 2500 For each S belonging to a prefix reachable via X's non-AERO 2501 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 2502 on those interfaces per [RFC7761]. 2504 For each S belonging to a prefix reachable via X's AERO interface, X 2505 originates a separate copy of the Join/Prune for each (S,G) in the 2506 message using its own AERO address as the source address and ALL-PIM- 2507 ROUTERS as the destination address. X then encapsulates each message 2508 in a SPAN header with source address set to the SPAN address of X and 2509 destination address set to S then forwards the message into the SPAN. 2510 The SPAN in turn forwards the message to AERO Server/Gateway "Y" that 2511 services S. At the same time, if the message was a Join, X sends a 2512 route-optimization NS message toward each S the same as discussed in 2513 Section 3.17. The resulting NAs will return the AERO address for the 2514 prefix that matches S as the network-layer source address and TLLAOs 2515 with the SPAN addresses corresponding to any Interface IDs that are 2516 currently servicing S. 2518 When Y processes the Join/Prune message, if S located behind any 2519 Native, Direct, VPNed or NATed interfaces Y acts as a PIM router and 2520 updates its MRIB to list X as the next hop in the reverse path. If S 2521 is located behind any Proxys "Z"*, Y also forwards the message to 2522 each Z* over the SPAN while continuing to use the AERO address of X 2523 as the source address. Each Z* then updates its MRIB accordingly and 2524 maintains the AERO address of X as the next hop in the reverse path. 2525 Since the Relays in the SPAN do not examine network layer control 2526 messages, this means that the (reverse) multicast tree path is simply 2527 from each Z* (and/or Y) to X with no other multicast-aware routers in 2528 the path. If any Z* (and/or Y) is located on the same SPAN segment 2529 as X, the multicast data traffic sent to X can use simple INET 2530 encapsulation and need not go over the SPAN. 2532 Following the initial Join/Prune and NS/NA messaging, X maintains an 2533 asymmetric neighbor cache entry for each S the same as if X was 2534 sending unicast data traffic to S. In particular, X performs 2535 additional NS/NA exchanges to keep the neighbor cache entry alive for 2536 up to t_periodic seconds [RFC7761]. If no new Joins are received 2537 within t_periodic seconds, X allows the neighbor cache entry to 2538 expire. Finally, if X receives any additional Join/Prune messages 2539 for (S,G) it forwards the messages to each Y and Z* in the neighbor 2540 cache entry over the SPAN. 2542 At some later time, Client C that holds an MNP for source S may 2543 depart from a first Proxy Z1 and/or connect via a new Proxy Z2. In 2544 that case, Y sends an unsolicited NA message to X the same as 2545 specified for unicast mobility in Section 3.19. When X receives the 2546 unsolicited NA message, it updates its asymmetric neighbor cache 2547 entry for the AERO address for source S and sends new Join messages 2548 to any new Proxys Z2. There is no requirement to send any Prune 2549 messages to old Proxys Z1 since source S will no longer source any 2550 multicast data traffic via Z1. Instead, the multicast state for 2551 (S,G) in Proxy Z1 will soon time out since no new Joins will arrive. 2553 After some later time, C may move to a new Server Y2 and depart from 2554 old Sever Y1. In that case, Y1 sends Join messages for any of C's 2555 active (S,G) groups to Y2 while including its own AERO address as the 2556 source address. This causes Y2 to include Y1 in the multicast 2557 forwarding tree during the interim time that Y1's symmetric neighbor 2558 cache entry for C is in the DEPARTED state. At the same time, Y1 2559 sends an unsolicited NA message to X with an Interface ID 0xffff and 2560 R set to 1 to cause X to release its asymmetric neighbor cache entry. 2561 X then sends a new Join message to S via the SPAN and re-initiates 2562 route optimization the same as if it were receiving a fresh Join 2563 message from a node on a downstream link. 2565 3.20.2. Any-Source Multicast (ASM) 2567 When an ROS X acting as a PIM router receives a Join/Prune from a 2568 node on its downstream interfaces containing one or more (*,G) pairs, 2569 it updates its Multicast Routing Information Base (MRIB) accordingly. 2570 X then forwards a copy of the message to the Rendezvous Point (RP) R 2571 for each G over the SPAN. X uses its own AERO address as the source 2572 address and ALL-PIM-ROUTERS as the destination address, then 2573 encapsulates each message in a SPAN header with source address set to 2574 the SPAN address of X and destination address set to R, then sends 2575 the message into the SPAN. At the same time, if the message was a 2576 Join X initiates NS/NA route optimization the same as for the SSM 2577 case discussed in Section 3.20.1. 2579 For each source S that sends multicast traffic to group G via R, the 2580 Proxy/Server Z* for the Client that aggregates S encapsulates the 2581 packets in PIM Register messages and forwards them to R via the SPAN. 2582 R may then elect to send a PIM Join to Z* over the SPAN. This will 2583 result in an (S,G) tree rooted at Z* with R as the next hop so that R 2584 will begin to receive two copies of the packet; one native copy from 2585 the (S, G) tree and a second copy from the pre-existing (*, G) tree 2586 that still uses PIM Register encapsulation. R can then issue a PIM 2587 Register-stop message to suppress the Register-encapsulated stream. 2588 At some later time, if C moves to a new Proxy/Server Z*, it resumes 2589 sending packets via PIM Register encapsulation via the new Z*. 2591 At the same time, as multicast listeners discover individual S's for 2592 a given G, they can initiate an (S,G) Join for each S under the same 2593 procedures discussed in Section 3.20.1. Once the (S,G) tree is 2594 established, the listeners can send (S, G) Prune messages to R so 2595 that multicast packets for group G sourced by S will only be 2596 delivered via the (S, G) tree and not from the (*, G) tree rooted at 2597 R. All mobility considerations discussed for SSM apply. 2599 3.20.3. Bi-Directional PIM (BIDIR-PIM) 2601 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 2602 approach to ASM that treats the Rendezvous Point (RP) as a Designated 2603 Forwarder (DF). Further considerations for BIDIR-PIM are out of 2604 scope. 2606 3.21. Operation over Multiple AERO Links (VLANs) 2608 An AERO Client can connect to multiple AERO links the same as for any 2609 data link service. In that case, the Client maintains a distinct 2610 AERO interface for each link, e.g., 'aero0' for the first link, 2611 'aero1' for the second, 'aero2' for the third, etc. Each AERO link 2612 would include its own distinct set of Relays, Servers and Proxys, 2613 thereby providing redundancy in case of failures. 2615 The Relays, Servers and Proxys on each AERO link can assign AERO and 2616 SPAN addresses that use the same or different numberings from those 2617 on other links. Since the links are mutually independent there is no 2618 requirement for avoiding inter-link address duplication, e.g., the 2619 same AERO address such as fe80::1000 could be used to number distinct 2620 nodes that connect to different links. 2622 Each AERO link could utilize the same or different ANET connections. 2623 The links can be distinguished at the link-layer via Virtual Local 2624 Area Network (VLAN) tagging (e.g., IEEE 802.1Q) and/or through 2625 assignment of distinct sets of MSPs on each link. This gives rise to 2626 the opportunity for supporting multiple redundant networked paths, 2627 where each VLAN is distinguished by a different label (e.g., colors 2628 such as Red, Green, Blue, etc.). In particular, the Client can tag 2629 its RS messages with the appropriate label to cause the network to 2630 select the desired VLAN. 2632 Clients that connect to multiple AERO interfaces can select the 2633 outgoing interface appropriate for a given Red/Blue/Green/etc. 2634 traffic profile while (in the reverse direction) correspondent nodes 2635 must have some way of steering their packets destined to a target via 2636 the correct AERO link. This can be accomplished in one of two ways. 2638 In a first alternative, if each AERO link services different MSPs, 2639 then the Client can receive a distinct MNP from each of the links. 2640 IP routing will therefore assure that the correct Red/Green/Blue/etc. 2641 network is used for both outbound and inbound traffic. This can be 2642 accomplished using existing technologies and approaches, and without 2643 requiring any special supporting code in correspondent nodes or 2644 Relays. 2646 In a second alternative, if each AERO link services the same MSP(s) 2647 then each link could assign an "AERO Link Anycast" address that is 2648 configured by all Relays on the link. Correspondent nodes then 2649 include a "type 4" routing header with the Anycast address for the 2650 AERO link as the IPv6 destination and with the address of the target 2651 encoded as the "next segment" in the routing header 2652 [RFC8402][I-D.ietf-6man-segment-routing-header]. Standard IP routing 2653 will then direct the packet to the nearest Relay for the correct AERO 2654 link, which will replace the destination address with the target 2655 address then forward the packet to the target. 2657 4. Direct Underlying Interfaces 2659 When a Client's AERO interface is configured over a Direct interface, 2660 the neighbor at the other end of the Direct link can receive packets 2661 without any encapsulation. In that case, the Client sends packets 2662 over the Direct link according to QoS preferences. If the Direct 2663 interface has the highest QoS preference, then the Client's IP 2664 packets are transmitted directly to the peer without going through an 2665 ANET/INET. If other interfaces have higher QoS preferences, then the 2666 Client's IP packets are transmitted via a different interface, which 2667 may result in the inclusion of Proxys, Servers and Relays in the 2668 communications path. Direct interfaces must be tested periodically 2669 for reachability, e.g., via NUD. 2671 5. AERO Clients on the Open Internetwork 2673 AERO Clients that connect to the open Internetwork via either a 2674 native or NATed interface can establish a VPN to securely connect to 2675 a Server. Alternatively, the Client can exchange ND messages 2676 directly with other AERO nodes on the same segment using INET 2677 encapsulation only and without joining the SPAN. In that case, 2678 however, the Client must apply asymmetric security for ND messages to 2679 ensure routing and neighbor cache integrity (see: Section 13). 2681 6. Operation on AERO Links with /64 ASPs 2683 IPv6 AERO links typically have MSPs that aggregate many candidate 2684 MNPs of length /64 or shorter. However, in some cases it may be 2685 desirable to use AERO over links that have only a /64 MSP. This can 2686 be accommodated by treating all Clients on the AERO link as simple 2687 hosts that receive /128 prefix delegations. 2689 In that case, the Client sends an RS message to the Server the same 2690 as for ordinary AERO links. The Server responds with an RA message 2691 that includes one or more /128 prefixes (i.e., singleton addresses) 2692 that include the /64 MSP prefix along with an interface identifier 2693 portion to be assigned to the Client. The Client and Server then 2694 configure their AERO addresses based on the interface identifier 2695 portions of the /128s (i.e., the lower 64 bits) and not based on the 2696 /64 prefix (i.e., the upper 64 bits). 2698 For example, if the MSP for the host-only IPv6 AERO link is 2699 2001:db8:1000:2000::/64, each Client will receive one or more /128 2700 IPv6 prefix delegations such as 2001:db8:1000:2000::1/128, 2701 2001:db8:1000:2000::2/128, etc. When the Client receives the prefix 2702 delegations, it assigns the AERO addresses fe80::1, fe80::2, etc. to 2703 the AERO interface, and assigns the global IPv6 addresses (i.e., the 2704 /128s) to either the AERO interface or an internal virtual interface 2705 such as a loopback. In this arrangement, the Client conducts route 2706 optimization in the same sense as discussed in Section 3.17. 2708 This specification has applicability for nodes that act as a Client 2709 on an "upstream" AERO link, but also act as a Server on "downstream" 2710 AERO links. More specifically, if the node acts as a Client to 2711 receive a /64 prefix from the upstream AERO link it can then act as a 2712 Server to provision /128s to Clients on downstream AERO links. 2714 7. AERO Adaptations for SEcure Neighbor Discovery (SEND) 2716 SEcure Neighbor Discovery (SEND) [RFC3971] and Cryptographically 2717 Generated Addresses (CGAs) [RFC3972] were designed to secure IPv6 ND 2718 messaging in environments where symmetric network and/or transport- 2719 layer security services are impractical (see: Section 13). AERO 2720 nodes that use SEND/CGA employ the following adaptations. 2722 When a source AERO node prepares a SEND-protected ND message, it uses 2723 a link-local CGA as the IPv6 source address and writes the prefix 2724 embedded in its AERO address (i.e., instead of fe80::/64) in the CGA 2725 parameters Subnet Prefix field. When the neighbor receives the ND 2726 message, it first verifies the message checksum and SEND/CGA 2727 parameters while using the link-local prefix fe80::/64 (i.e., instead 2728 of the value in the Subnet Prefix field) to match against the IPv6 2729 source address of the ND message. 2731 The neighbor then derives the AERO address of the source by using the 2732 value in the Subnet Prefix field as the interface identifier of an 2733 AERO address. For example, if the Subnet Prefix field contains 2734 2001:db8:1:2, the neighbor constructs the AERO address as 2735 fe80::2001:db8:1:2. The neighbor then caches the AERO address in the 2736 neighbor cache entry it creates for the source, and uses the AERO 2737 address as the IPv6 destination address of any ND message replies. 2739 8. AERO Critical Infrastructure Considerations 2741 AERO Relays can be either Commercial off-the Shelf (COTS) standard IP 2742 routers or virtual machines in the cloud. Relays must be 2743 provisioned, supported and managed by the INET administrative 2744 authority, and connected to the Relays of other INETs via inter- 2745 domain peerings. Cost for purchasing, configuring and managing 2746 Relays is nominal even for very large AERO links. 2748 AERO Servers can be standard dedicated server platforms, but most 2749 often will be deployed as virtual machines in the cloud. The only 2750 requirements for Servers are that they can run the AERO user-level 2751 code and have at least one network interface connection to the INET. 2752 As with Relays, Servers must be provisioned, supported and managed by 2753 the INET administrative authority. Cost for purchasing, configuring 2754 and managing Servers is nominal especially for virtual Servers hosted 2755 in the cloud. 2757 AERO Proxys are most often standard dedicated server platforms with 2758 one network interface connected to the ANET and a second interface 2759 connected to an INET. As with Servers, the only requirements are 2760 that they can run the AERO user-level code and have at least one 2761 interface connection to the INET. Proxys must be provisioned, 2762 supported and managed by the ANET administrative authority. Cost for 2763 purchasing, configuring and managing Proxys is nominal, and borne by 2764 the ANET administrative authority. 2766 AERO Gateways can be any dedicated server or COTS router platform 2767 connected to INETs and/or EUNs. The Gateway joins the SPAN and 2768 engages in eBGP peering with one or more Relays as a stub AS. The 2769 Gateway then injects its MNPs and/or non-MNP prefixes into the BGP 2770 routing system, and provisions the prefixes to its downstream- 2771 attached networks. The Gateway can perform ROS and MAP services the 2772 same as for any Server, and can route between the MNP and non-MNP 2773 address spaces. 2775 9. DNS Considerations 2777 AERO Client MNs and INET correspondent nodes consult the Domain Name 2778 System (DNS) the same as for any Internetworking node. When 2779 correspondent nodes and Client MNs use different IP protocol versions 2780 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 2781 A records for IPv4 address mappings to MNs which must then be 2782 populated in Gateway NAT64 mapping caches. In that way, an IPv4 2783 correspondent node can send packets to the IPv4 address mapping of 2784 the target MN, and the Gateway will translate the IPv4 header and 2785 destination address into an IPv6 header and IPv6 destination address 2786 of the MN. 2788 When an AERO Client registers with an AERO Server, the Server returns 2789 the address(es) of DNS servers in RDNSS options [RFC6106]. The DNS 2790 server provides the IP addresses of other MNs and correspondent nodes 2791 in AAAA records for IPv6 or A records for IPv4. 2793 10. Transition Considerations 2795 The SPAN ensures that dissimilar INET partitions can be joined into a 2796 single unified AERO link, even though the partitions themselves may 2797 have differing protocol versions and/or incompatible addressing 2798 plans. However, a commonality can be achieved by incrementally 2799 distributing globally routable (i.e., native) IP prefixes to 2800 eventually reach all nodes (both mobile and fixed) in all segments. 2801 This can be accomplished by incrementally deploying AERO Gateways on 2802 each INET partition, with each Gateway distributing its MNPs and/or 2803 discovering non-MNP prefixes on its INET links. 2805 This gives rise to the opportunity to eventually distribute native IP 2806 addresses to all nodes, and to present a unified AERO link view 2807 (bridged by the SPAN) even if the INET partitions remain in their 2808 current protocol and addressing plans. In that way, the AERO link 2809 can serve the dual purpose of providing a mobility service and a 2810 transition service. Or, if an INET partition is transitioned to a 2811 native IP protocol version and addressing scheme that is compatible 2812 with the AERO link MNP-based addressing scheme, the partition and 2813 AERO link can be joined by Gateways. 2815 Gateways that connect INETs/EUNs with dissimilar IP protocol versions 2816 must employ a network address and protocol translation function such 2817 as NAT64[RFC6146]. 2819 11. Implementation Status 2821 An AERO implementation based on OpenVPN (https://openvpn.net/) was 2822 announced on the v6ops mailing list on January 10, 2018 and an 2823 initial public release of the AERO proof-of-concept source code was 2824 announced on the intarea mailing list on August 21, 2015. The latest 2825 versions are available at: http://linkupnetworks.net/aero. 2827 12. IANA Considerations 2829 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 2830 AERO in the "enterprise-numbers" registry. 2832 The IANA has assigned the UDP port number "8060" for an earlier 2833 experimental version of AERO [RFC6706]. This document obsoletes 2834 [RFC6706] and claims the UDP port number "8060" for all future use. 2836 No further IANA actions are required. 2838 13. Security Considerations 2840 AERO Relays configure secured tunnels with AERO Servers and Proxys 2841 within their local SPAN segments. Applicable secured tunnel 2842 alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 2843 [RFC6347], etc. The AERO Relays of all SPAN segments in turn 2844 configure secured tunnels for their neighboring AERO Relays across 2845 the SPAN. Therefore, packets that traverse the SPAN between any pair 2846 of AERO link neighbors are already secured. 2848 AERO Servers, Gateways and Proxys targeted by a route optimization 2849 may also receive packets directly from the INET partitions instead of 2850 via the SPAN. For INET partitions that apply effective ingress 2851 filtering to defeat source address spoofing, the simple data origin 2852 authentication procedures in Section 3.11 can be applied. This 2853 implies that the ROS list must be maintained consistently by all 2854 route optimization targets within the same INET partition, and that 2855 the ROS list must be securely managed by the partition's 2856 administrative authority. 2858 For INET partitions that cannot apply effective ingress filtering, 2859 the two options for securing communications include 1) disable route 2860 optimization so that all traffic is conveyed over secured tunnels via 2861 the SPAN, or 2) enable on-demand secure tunnel creation between INET 2862 partition neighbors. Option 1) would result in longer routes than 2863 necessary and traffic concentration on overburdened critical 2864 infrastructure elements. Option 2) could be coordinated by 2865 establishing a secured tunnel on-demand instead of performing an NS/ 2866 NA exchange in the route optimization procedures. Procedures for 2867 establishing on-demand secured tunnels are out of scope. 2869 AERO Clients that connect to secured enclaves need not apply security 2870 to their ND messages, since the messages will be intercepted by a 2871 perimeter Proxy that applies security on its outward-facing 2872 interface. AERO Clients located outside of secured enclaves SHOULD 2873 use symmetric network and/or transport layer security services, but 2874 when there are many prospective neighbors with dynamically changing 2875 connectivity an asymmetric security service such as SEND may be 2876 needed (see: Section 7). 2878 Application endpoints SHOULD use application-layer security services 2879 such as TLS/SSL, DTLS or SSH [RFC4251] to assure the same level of 2880 protection as for critical secured Internet services. AERO Clients 2881 that require host-based VPN services SHOULD use symmetric network 2882 and/or transport layer security services such as IPsec, TLS/SSL, 2883 DTLS, etc. AERO Proxys and Servers can also provide a network-based 2884 VPN service on behalf of the Client, e.g., if the Client is located 2885 within a secured enclave and cannot establish a VPN on its own 2886 behalf. 2888 AERO Servers and Relays present targets for traffic amplification 2889 Denial of Service (DoS) attacks. This concern is no different than 2890 for widely-deployed VPN security gateways in the Internet, where 2891 attackers could send spoofed packets to the gateways at high data 2892 rates. This can be mitigated by connecting Servers and Relays over 2893 dedicated links with no connections to the Internet and/or when 2894 connections to the Internet are only permitted through well-managed 2895 firewalls. Traffic amplification DoS attacks can also target an AERO 2896 Client's low data rate links. This is a concern not only for Clients 2897 located on the open Internet but also for Clients in secured 2898 enclaves. AERO Servers and Proxys can institute rate limits that 2899 protect Clients from receiving packet floods that could DoS low data 2900 rate links. 2902 AERO Gateways must implement ingress filtering to avoid a spoofing 2903 attack in which spurious SPAN messages are injected into an AERO link 2904 from an outside attacker. AERO Clients MUST ensure that their 2905 connectivity is not used by unauthorized nodes on their EUNs to gain 2906 access to a protected network, i.e., AERO Clients that act as routers 2907 MUST NOT provide routing services for unauthorized nodes. (This 2908 concern is no different than for ordinary hosts that receive an IP 2909 address delegation but then "share" the address with other nodes via 2910 some form of Internet connection sharing such as tethering.) 2912 The MAP list and ROS lists MUST be well-managed and secured from 2913 unauthorized tampering, even though the list contains only public 2914 information. The MAP list can be conveyed to the Client, e.g., 2915 through secure upload of a static file, through DNS lookups, etc. 2916 The ROS list can be conveyed to Servers and Proxys through 2917 administrative action, secured file distribution, etc. 2919 Although public domain and commercial SEND implementations exist, 2920 concerns regarding the strength of the cryptographic hash algorithm 2921 have been documented [RFC6273] [RFC4982]. 2923 Security considerations for accepting link-layer ICMP messages and 2924 reflected packets are discussed throughout the document. 2926 14. Acknowledgements 2928 Discussions in the IETF, aviation standards communities and private 2929 exchanges helped shape some of the concepts in this work. 2930 Individuals who contributed insights include Mikael Abrahamsson, Mark 2931 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 2932 Wojciech Dec, Ralph Droms, Adrian Farrel, Nick Green, Sri Gundavelli, 2933 Brian Haberman, Bernhard Haindl, Joel Halpern, Tom Herbert, Sascha 2934 Hlusiak, Lee Howard, Andre Kostur, Hubert Kuenig, Ted Lemon, Andy 2935 Malis, Satoru Matsushima, Tomek Mrugalski, Madhu Niraula, Alexandru 2936 Petrescu, Behcet Saikaya, Michal Skorepa, Joe Touch, Bernie Volz, 2937 Ryuji Wakikawa, Tony Whyman, Lloyd Wood and James Woodyatt. Members 2938 of the IESG also provided valuable input during their review process 2939 that greatly improved the document. Special thanks go to Stewart 2940 Bryant, Joel Halpern and Brian Haberman for their shepherding 2941 guidance during the publication of the AERO first edition. 2943 This work has further been encouraged and supported by Boeing 2944 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 2945 Brodie, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, Claudiu 2946 Danilov, Wen Fang, Anthony Gregory, Jeff Holland, Seth Jahne, Ed 2947 King, Gene MacLean III, Rob Muszkiewicz, Sean O'Sullivan, Greg 2948 Saccone, Kent Shuey, Brian Skeen, Mike Slane, Carrie Spiker, Brendan 2949 Williams, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 2950 BR&T and BIT mobile networking teams. Kyle Bae, Wayne Benson and 2951 Eric Yeh are especially acknowledged for implementing the AERO 2952 functions as extensions to the public domain OpenVPN distribution. 2954 Earlier works on NBMA tunneling approaches are found in 2955 [RFC2529][RFC5214][RFC5569]. 2957 Many of the constructs presented in this second edition of AERO are 2958 based on the author's earlier works, including: 2960 o The Internet Routing Overlay Network (IRON) 2961 [RFC6179][I-D.templin-ironbis] 2963 o Virtual Enterprise Traversal (VET) 2964 [RFC5558][I-D.templin-intarea-vet] 2966 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 2967 [RFC5320][I-D.templin-intarea-seal] 2969 o AERO, First Edition [RFC6706] 2971 Note that these works cite numerous earlier efforts that are not also 2972 cited here due to space limitations. The authors of those earlier 2973 works are acknowledged for their insights. 2975 This work is aligned with the NASA Safe Autonomous Systems Operation 2976 (SASO) program under NASA contract number NNA16BD84C. 2978 This work is aligned with the FAA as per the SE2025 contract number 2979 DTFAWA-15-D-00030. 2981 This work is aligned with the Boeing Information Technology (BIT) 2982 MobileNet program. 2984 This work is aligned with the Boeing autonomy program. 2986 15. References 2988 15.1. Normative References 2990 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 2991 DOI 10.17487/RFC0791, September 1981, 2992 . 2994 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 2995 RFC 792, DOI 10.17487/RFC0792, September 1981, 2996 . 2998 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2999 Requirement Levels", BCP 14, RFC 2119, 3000 DOI 10.17487/RFC2119, March 1997, 3001 . 3003 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 3004 "Definition of the Differentiated Services Field (DS 3005 Field) in the IPv4 and IPv6 Headers", RFC 2474, 3006 DOI 10.17487/RFC2474, December 1998, 3007 . 3009 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3010 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3011 DOI 10.17487/RFC3971, March 2005, 3012 . 3014 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3015 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3016 . 3018 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3019 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3020 November 2005, . 3022 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3023 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3024 DOI 10.17487/RFC4861, September 2007, 3025 . 3027 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3028 Address Autoconfiguration", RFC 4862, 3029 DOI 10.17487/RFC4862, September 2007, 3030 . 3032 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3033 (IPv6) Specification", STD 86, RFC 8200, 3034 DOI 10.17487/RFC8200, July 2017, 3035 . 3037 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3038 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3039 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3040 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3041 . 3043 15.2. Informative References 3045 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3046 2016. 3048 [I-D.ietf-6man-segment-routing-header] 3049 Filsfils, C., Dukes, D., Previdi, S., Leddy, J., 3050 Matsushima, S., and d. daniel.voyer@bell.ca, "IPv6 Segment 3051 Routing Header (SRH)", draft-ietf-6man-segment-routing- 3052 header-19 (work in progress), May 2019. 3054 [I-D.ietf-dmm-distributed-mobility-anchoring] 3055 Chan, A., Wei, X., Lee, J., Jeon, S., and C. Bernardos, 3056 "Distributed Mobility Anchoring", draft-ietf-dmm- 3057 distributed-mobility-anchoring-13 (work in progress), 3058 March 2019. 3060 [I-D.ietf-intarea-gue] 3061 Herbert, T., Yong, L., and O. Zia, "Generic UDP 3062 Encapsulation", draft-ietf-intarea-gue-07 (work in 3063 progress), March 2019. 3065 [I-D.ietf-intarea-gue-extensions] 3066 Herbert, T., Yong, L., and F. Templin, "Extensions for 3067 Generic UDP Encapsulation", draft-ietf-intarea-gue- 3068 extensions-06 (work in progress), March 2019. 3070 [I-D.ietf-intarea-tunnels] 3071 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3072 Architecture", draft-ietf-intarea-tunnels-09 (work in 3073 progress), July 2018. 3075 [I-D.ietf-rtgwg-atn-bgp] 3076 Templin, F., Saccone, G., Dawra, G., Lindem, A., and V. 3077 Moreno, "A Simple BGP-based Mobile Routing System for the 3078 Aeronautical Telecommunications Network", draft-ietf- 3079 rtgwg-atn-bgp-02 (work in progress), May 2019. 3081 [I-D.templin-6man-dhcpv6-ndopt] 3082 Templin, F., "A Unified Stateful/Stateless Configuration 3083 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-07 3084 (work in progress), December 2018. 3086 [I-D.templin-intarea-grefrag] 3087 Templin, F., "GRE Tunnel Level Fragmentation", draft- 3088 templin-intarea-grefrag-04 (work in progress), July 2016. 3090 [I-D.templin-intarea-seal] 3091 Templin, F., "The Subnetwork Encapsulation and Adaptation 3092 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 3093 progress), January 2014. 3095 [I-D.templin-intarea-vet] 3096 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 3097 templin-intarea-vet-40 (work in progress), May 2013. 3099 [I-D.templin-ironbis] 3100 Templin, F., "The Interior Routing Overlay Network 3101 (IRON)", draft-templin-ironbis-16 (work in progress), 3102 March 2014. 3104 [I-D.templin-v6ops-pdhost] 3105 Templin, F., "IPv6 Prefix Delegation and Multi-Addressing 3106 Models", draft-templin-v6ops-pdhost-23 (work in progress), 3107 December 2018. 3109 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3111 [RFC1035] Mockapetris, P., "Domain names - implementation and 3112 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3113 November 1987, . 3115 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 3116 Communication Layers", STD 3, RFC 1122, 3117 DOI 10.17487/RFC1122, October 1989, 3118 . 3120 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 3121 DOI 10.17487/RFC1191, November 1990, 3122 . 3124 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3125 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3126 . 3128 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3129 DOI 10.17487/RFC2003, October 1996, 3130 . 3132 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3133 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3134 . 3136 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3137 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3138 December 1998, . 3140 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3141 Domains without Explicit Tunnels", RFC 2529, 3142 DOI 10.17487/RFC2529, March 1999, 3143 . 3145 [RFC2764] Gleeson, B., Lin, A., Heinanen, J., Armitage, G., and A. 3146 Malis, "A Framework for IP Based Virtual Private 3147 Networks", RFC 2764, DOI 10.17487/RFC2764, February 2000, 3148 . 3150 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 3151 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 3152 DOI 10.17487/RFC2784, March 2000, 3153 . 3155 [RFC2890] Dommety, G., "Key and Sequence Number Extensions to GRE", 3156 RFC 2890, DOI 10.17487/RFC2890, September 2000, 3157 . 3159 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", 3160 RFC 2923, DOI 10.17487/RFC2923, September 2000, 3161 . 3163 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3164 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3165 . 3167 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3168 of Explicit Congestion Notification (ECN) to IP", 3169 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3170 . 3172 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3173 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3174 DOI 10.17487/RFC3810, June 2004, 3175 . 3177 [RFC3819] Karn, P., Ed., Bormann, C., Fairhurst, G., Grossman, D., 3178 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 3179 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 3180 RFC 3819, DOI 10.17487/RFC3819, July 2004, 3181 . 3183 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 3184 for IPv6 Hosts and Routers", RFC 4213, 3185 DOI 10.17487/RFC4213, October 2005, 3186 . 3188 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3189 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3190 January 2006, . 3192 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3193 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3194 DOI 10.17487/RFC4271, January 2006, 3195 . 3197 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3198 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3199 2006, . 3201 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3202 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3203 December 2005, . 3205 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3206 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3207 2006, . 3209 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3210 Control Message Protocol (ICMPv6) for the Internet 3211 Protocol Version 6 (IPv6) Specification", STD 89, 3212 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3213 . 3215 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3216 Protocol (LDAP): The Protocol", RFC 4511, 3217 DOI 10.17487/RFC4511, June 2006, 3218 . 3220 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3221 "Considerations for Internet Group Management Protocol 3222 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3223 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3224 . 3226 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3227 "Internet Group Management Protocol (IGMP) / Multicast 3228 Listener Discovery (MLD)-Based Multicast Forwarding 3229 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3230 August 2006, . 3232 [RFC4607] Holbrook, H. and B. Cain, "Source-Specific Multicast for 3233 IP", RFC 4607, DOI 10.17487/RFC4607, August 2006, 3234 . 3236 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 3237 Errors at High Data Rates", RFC 4963, 3238 DOI 10.17487/RFC4963, July 2007, 3239 . 3241 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3242 Algorithms in Cryptographically Generated Addresses 3243 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3244 . 3246 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3247 "Bidirectional Protocol Independent Multicast (BIDIR- 3248 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3249 . 3251 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3252 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3253 DOI 10.17487/RFC5214, March 2008, 3254 . 3256 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3257 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3258 February 2010, . 3260 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3261 Route Optimization Requirements for Operational Use in 3262 Aeronautics and Space Exploration Mobile Networks", 3263 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3264 . 3266 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3267 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3268 . 3270 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3271 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3272 January 2010, . 3274 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3275 "IPv6 Router Advertisement Options for DNS Configuration", 3276 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3277 . 3279 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3280 NAT64: Network Address and Protocol Translation from IPv6 3281 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3282 April 2011, . 3284 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3285 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3286 . 3288 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3289 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3290 DOI 10.17487/RFC6221, May 2011, 3291 . 3293 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3294 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3295 DOI 10.17487/RFC6273, June 2011, 3296 . 3298 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3299 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3300 January 2012, . 3302 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3303 for Equal Cost Multipath Routing and Link Aggregation in 3304 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3305 . 3307 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3308 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3309 . 3311 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID Field", 3312 RFC 6864, DOI 10.17487/RFC6864, February 2013, 3313 . 3315 [RFC7269] Chen, G., Cao, Z., Xie, C., and D. Binet, "NAT64 3316 Deployment Options and Experience", RFC 7269, 3317 DOI 10.17487/RFC7269, June 2014, 3318 . 3320 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3321 Korhonen, "Requirements for Distributed Mobility 3322 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3323 . 3325 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3326 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3327 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3328 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3329 2016, . 3331 [RFC8086] Yong, L., Ed., Crabbe, E., Xu, X., and T. Herbert, "GRE- 3332 in-UDP Encapsulation", RFC 8086, DOI 10.17487/RFC8086, 3333 March 2017, . 3335 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 3336 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 3337 DOI 10.17487/RFC8201, July 2017, 3338 . 3340 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3341 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3342 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3343 July 2018, . 3345 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3346 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3347 . 3349 Appendix A. AERO Alternate Encapsulations 3351 When GUE encapsulation is not needed, AERO can use common 3352 encapsulations such as IP-in-IP [RFC2003][RFC2473][RFC4213], Generic 3353 Routing Encapsulation (GRE) [RFC2784][RFC2890] and others. The 3354 encapsulation is therefore only differentiated from non-AERO tunnels 3355 through the application of AERO control messaging and not through, 3356 e.g., a well-known UDP port number. 3358 As for GUE encapsulation, alternate AERO encapsulation formats may 3359 require encapsulation layer fragmentation. For simple IP-in-IP 3360 encapsulation, an IPv6 fragment header is inserted directly between 3361 the inner and outer IP headers when needed, i.e., even if the outer 3362 header is IPv4. The IPv6 Fragment Header is identified to the outer 3363 IP layer by its IP protocol number, and the Next Header field in the 3364 IPv6 Fragment Header identifies the inner IP header version. For GRE 3365 encapsulation, a GRE fragment header is inserted within the GRE 3366 header [I-D.templin-intarea-grefrag]. 3368 Figure 6 shows the AERO IP-in-IP encapsulation format before any 3369 fragmentation is applied: 3371 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3372 | Outer IPv4 Header | | Outer IPv6 Header | 3373 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3374 |IPv6 Frag Header (optional)| |IPv6 Frag Header (optional)| 3375 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3376 | Inner IP Header | | Inner IP Header | 3377 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3378 | | | | 3379 ~ ~ ~ ~ 3380 ~ Inner Packet Body ~ ~ Inner Packet Body ~ 3381 ~ ~ ~ ~ 3382 | | | | 3383 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3385 Minimal Encapsulation in IPv4 Minimal Encapsulation in IPv6 3387 Figure 6: Minimal Encapsulation Format using IP-in-IP 3389 Figure 7 shows the AERO GRE encapsulation format before any 3390 fragmentation is applied: 3392 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3393 | Outer IP Header | 3394 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3395 | GRE Header | 3396 | (with checksum, key, etc..) | 3397 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3398 | GRE Fragment Header (optional)| 3399 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3400 | Inner IP Header | 3401 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3402 | | 3403 ~ ~ 3404 ~ Inner Packet Body ~ 3405 ~ ~ 3406 | | 3407 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3409 Figure 7: Minimal Encapsulation Using GRE 3411 Alternate encapsulation may be preferred in environments where GUE 3412 encapsulation would add unnecessary overhead. For example, certain 3413 low-bandwidth wireless data links may benefit from a reduced 3414 encapsulation overhead. 3416 GUE encapsulation can traverse network paths that are inaccessible to 3417 non-UDP encapsulations, e.g., for crossing Network Address 3418 Translators (NATs). More and more, network middleboxes are also 3419 being configured to discard packets that include anything other than 3420 a well-known IP protocol such as UDP and TCP. It may therefore be 3421 necessary to determine the potential for middlebox filtering before 3422 enabling alternate encapsulation in a given environment. 3424 In addition to IP-in-IP, GRE and GUE, AERO can also use security 3425 encapsulations such as IPsec, TLS/SSL, DTLS, etc. In that case, AERO 3426 control messaging and route determination occur before security 3427 encapsulation is applied for outgoing packets and after security 3428 decapsulation is applied for incoming packets. 3430 AERO is especially well suited for use with VPN system encapsulations 3431 such as OpenVPN [OVPN]. 3433 Appendix B. S/TLLAO Extensions for Special-Purpose Links 3435 The AERO S/TLLAO format specified in Section 3.6 includes a Length 3436 value of 5 (i.e., 5 units of 8 octets). However, special-purpose 3437 links may extend the basic format to include additional fields and a 3438 Length value larger than 5. 3440 For example, adaptation of AERO to the Aeronautical 3441 Telecommunications Network with Internet Protocol Services (ATN/IPS) 3442 includes link selection preferences based on transport port numbers 3443 in addition to the existing DSCP-based preferences. ATN/IPS nodes 3444 maintain a map of transport port numbers to 64 possible preference 3445 fields, e.g., TCP port 22 maps to preference field 8, TCP port 443 3446 maps to preference field 20, UDP port 8060 maps to preference field 3447 34, etc. The extended S/TLLAO format for ATN/IPS is shown in 3448 Figure 8, where the Length value is 7 and the 'Q(i)' fields provide 3449 link preferences for the corresponding transport port number. 3451 0 1 2 3 3452 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 3453 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3454 | Type | Length = 7 | Prefix Length | Reserved | 3455 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3456 | Interface ID | Port Number | 3457 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3458 | | 3459 + + 3460 | | 3461 + Link-Layer Address + 3462 | | 3463 + + 3464 | | 3465 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3466 |P00|P01|P02|P03|P04|P05|P06|P07|P08|P09|P10|P11|P12|P13|P14|P15| 3467 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3468 |P16|P17|P18|P19|P20|P21|P22|P23|P24|P25|P26|P27|P28|P29|P30|P31| 3469 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3470 |P32|P33|P34|P35|P36|P37|P38|P39|P40|P41|P42|P43|P44|P45|P46|P47| 3471 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3472 |P48|P49|P50|P51|P52|P53|P54|P55|P56|P57|P58|P59|P60|P61|P62|P63| 3473 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3474 |Q00|Q01|Q02|Q03|Q04|Q05|Q06|Q07|Q08|Q09|Q10|Q11|Q12|Q13|Q14|Q15| 3475 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3476 |Q16|Q17|Q18|Q19|Q20|Q21|Q22|Q23|Q24|Q25|Q26|Q27|Q28|Q29|Q30|Q31| 3477 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3478 |Q32|Q33|Q34|Q35|Q36|Q37|Q38|Q39|Q40|Q41|Q42|Q43|Q44|Q45|Q46|Q47| 3479 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3480 |Q48|Q49|Q50|Q51|Q52|Q53|Q54|Q55|Q56|Q57|Q58|Q59|Q60|Q61|Q62|Q63| 3481 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3483 Figure 8: ATN/IPS Extended S/TLLAO Format 3485 Appendix C. Implicit Mobility Management 3487 AERO interface neighbors MAY provide a configuration option that 3488 allows them to perform implicit mobility management in which no ND 3489 messaging is used. In that case, the Client only transmits packets 3490 over a single interface at a time, and the neighbor always observes 3491 packets arriving from the Client from the same link-layer source 3492 address. 3494 If the Client's ANET interface address changes (either due to a 3495 readdressing of the original interface or switching to a new 3496 interface) the neighbor immediately updates the neighbor cache entry 3497 for the Client and begins accepting and sending packets according to 3498 the Client's new ANET address. This implicit mobility method applies 3499 to use cases such as cellphones with both WiFi and Cellular 3500 interfaces where only one of the interfaces is active at a given 3501 time, and the Client automatically switches over to the backup 3502 interface if the primary interface fails. 3504 Appendix D. Implementation Strategies for Route Optimization 3506 Route optimization as discussed in Section 3.17 results in the route 3507 optimization source (ROS) creating an asymmetric neighbor cache entry 3508 for the target neighbor. The neighbor cache entry is maintained for 3509 at most REACHABLE_TIME seconds and then deleted unless updated. In 3510 order to refresh the neighbor cache entry lifetime before the 3511 ReachableTime timer expires, the specification requires 3512 implementations to issue a new NS/NA exchange to reset ReachableTime 3513 to REACHABLE_TIME seconds while data packets are still flowing. 3514 However, the decision of when to initiate a new NS/NA exchange and to 3515 perpetuate the process is left as an implementation detail. 3517 One possible strategy may be to monitor the neighbor cache entry 3518 watching for data packets for (REACHABLE_TIME - 5) seconds. If any 3519 data packets have been sent to the neighbor within this timeframe, 3520 then send an NS to receive a new NA. If no data packets have been 3521 sent, wait for 5 additional seconds and send an immediate NS if any 3522 data packets are sent within this "expiration pending" 5 second 3523 window. If no additional data packets are sent within the 5 second 3524 window, delete the neighbor cache entry. 3526 The monitoring of the neighbor data packet traffic therefore becomes 3527 an asymmetric ongoing process during the neighbor cache entry 3528 lifetime. If the neighbor cache entry expires, future data packets 3529 will trigger a new NS/NA exchange while the packets themselves are 3530 delivered over a longer path until route optimization state is re- 3531 established. 3533 Appendix E. Change Log 3535 << RFC Editor - remove prior to publication >> 3537 Changes from draft-templin-intarea-6706bis-13 to draft-templin- 3538 intrea-6706bis-14: 3540 o Security based on secured tunnels, ingress filtering, MAP list and 3541 ROS list 3543 Changes from draft-templin-intarea-6706bis-12 to draft-templin- 3544 intrea-6706bis-13: 3546 o New paragraph in Section 3.6 on AERO interface layering over 3547 secured tunnels 3549 o Removed extraneous text in Section 3.7 3551 o Added new detail to the forwarding algorithm in Section 3.9 3553 o Clarified use of fragmentation 3555 o Route optimization now supported for both MNP and non-MNP-based 3556 prefixes 3558 o Relays are now seen as link-layer elements in the architecture. 3560 o Built out multicast section in detail. 3562 o New Appendix on implementation considerations for route 3563 optimization. 3565 Changes from draft-templin-intarea-6706bis-11 to draft-templin- 3566 intrea-6706bis-12: 3568 o Introduced Gateways as a new AERO element for connecting 3569 Correspondent Nodes on INET links 3571 o Introduced terms "Access Network (ANET)" and "Internetwork (INET)" 3573 o Changed "ASP" to "MSP", and "ACP" to "MNP" 3575 o New figure on the relation of Segments to the SPAN and AERO link 3577 o New "S" bit in S/TLLAO to indicate the "Source" S/TLLAO as opposed 3578 to additional S/TLLAOs 3580 o Changed Interface ID for Servers from 255 to 0xffff 3581 o Significant updates to Route Optimization, NUD, and Mobility 3582 Management 3584 o New Section on Multicast 3586 o New Section on AERO Clients in the open Internetwork 3588 o New Section on Operation over multiple AERO links (VLANs over the 3589 SPAN) 3591 o New Sections on DNS considerations and Transition considerations 3593 o 3595 Changes from draft-templin-intarea-6706bis-10 to draft-templin- 3596 intrea-6706bis-11: 3598 o Added The SPAN 3600 Changes from draft-templin-intarea-6706bis-09 to draft-templin- 3601 intrea-6706bis-10: 3603 o Orphaned packets in flight (e.g., when a neighbor cache entry is 3604 in the DEPARTED state) are now forwarded at the link layer instead 3605 of at the network layer. Forwarding at the network layer can 3606 result in routing loops and/or excessive delays of forwarded 3607 packets while the routing system is still reconverging. 3609 o Update route optimization to clarify the unsecured nature of the 3610 first NS used for route discovery 3612 o Many cleanups and clarifications on ND messaging parameters 3614 Changes from draft-templin-intarea-6706bis-08 to draft-templin- 3615 intrea-6706bis-09: 3617 o Changed PRL to "MAP list" 3619 o For neighbor cache entries, changed "static" to "symmetric", and 3620 "dynamic" to "asymmetric" 3622 o Specified Proxy RS/RA exchanges with Servers on behalf of Clients 3624 o Added discussion of unsolicited NAs in Section 3.16, and included 3625 forward reference to Section 3.18 3627 o Added discussion of AERO Clients used as critical infrastructure 3628 elements to connect fixed networks. 3630 o Added network-based VPN under security considerations 3632 Changes from draft-templin-intarea-6706bis-07 to draft-templin- 3633 intrea-6706bis-08: 3635 o New section on AERO-Aware Access Router 3637 Changes from draft-templin-intarea-6706bis-06 to draft-templin- 3638 intrea-6706bis-07: 3640 o Added "R" bit for release of PDs. Now have a full RS/RA service 3641 that can do PD without requiring DHCPv6 messaging over-the-air 3643 o Clarifications on solicited vs unsolicited NAs 3645 o Clarified use of MAX_NEIGHBOR_ADVERTISEMENTS for the purpose of 3646 increase reliability 3648 Changes from draft-templin-intarea-6706bis-05 to draft-templin- 3649 intrea-6706bis-06: 3651 o Major re-work and simplification of Route Optimization function 3653 o Added Distributed Mobility Management (DMM) and Mobility Anchor 3654 Point (MAP) terminology 3656 o New section on "AERO Critical Infrastructure Element 3657 Considerations" demonstrating low overall cost for the service 3659 o minor text revisions and deletions 3661 o removed extraneous appendices 3663 Changes from draft-templin-intarea-6706bis-04 to draft-templin- 3664 intrea-6706bis-05: 3666 o New Appendix E on S/TLLAO Extensions for special-purpose links. 3667 Discussed ATN/IPS as example. 3669 o New sentence in introduction to declare appendices as non- 3670 normative. 3672 Changes from draft-templin-intarea-6706bis-03 to draft-templin- 3673 intrea-6706bis-04: 3675 o Added definitions for Potential Router List (PRL) and secure 3676 enclave 3678 o Included text on mapping transport layer port numbers to network 3679 layer DSCP values 3681 o Added reference to DTLS and DMM Distributed Mobility Anchoring 3682 working group document 3684 o Reworked Security Considerations 3686 o Updated references. 3688 Changes from draft-templin-intarea-6706bis-02 to draft-templin- 3689 intrea-6706bis-03: 3691 o Added new section on SEND. 3693 o Clarifications on "AERO Address" section. 3695 o Updated references and added new reference for RFC8086. 3697 o Security considerations updates. 3699 o General text clarifications and cleanup. 3701 Changes from draft-templin-intarea-6706bis-01 to draft-templin- 3702 intrea-6706bis-02: 3704 o Note on encapsulation avoidance in Section 4. 3706 Changes from draft-templin-intarea-6706bis-00 to draft-templin- 3707 intrea-6706bis-01: 3709 o Remove DHCPv6 Server Release procedures that leveraged the old way 3710 Relays used to "route" between Server link-local addresses 3712 o Remove all text relating to Relays needing to do any AERO-specific 3713 operations 3715 o Proxy sends RS and receives RA from Server using SEND. Use CGAs 3716 as source addresses, and destination address of RA reply is to the 3717 AERO address corresponding to the Client's ACP. 3719 o Proxy uses SEND to protect RS and authenticate RA (Client does not 3720 use SEND, but rather relies on subnetwork security. When the 3721 Proxy receives an RS from the Client, it creates a new RS using 3722 its own addresses as the source and uses SEND with CGAs to send a 3723 new RS to the Server. 3725 o Emphasize distributed mobility management 3726 o AERO address-based RS injection of ACP into underlying routing 3727 system. 3729 Changes from draft-templin-aerolink-82 to draft-templin-intarea- 3730 6706bis-00: 3732 o Document use of NUD (NS/NA) for reliable link-layer address 3733 updates as an alternative to unreliable unsolicited NA. 3734 Consistent with Section 7.2.6 of RFC4861. 3736 o Server adds additional layer of encapsulation between outer and 3737 inner headers of NS/NA messages for transmission through Relays 3738 that act as vanilla IPv6 routers. The messages include the AERO 3739 Server Subnet Router Anycast address as the source and the Subnet 3740 Router Anycast address corresponding to the Client's ACP as the 3741 destination. 3743 o Clients use Subnet Router Anycast address as the encapsulation 3744 source address when the access network does not provide a 3745 topologically-fixed address. 3747 Author's Address 3749 Fred L. Templin (editor) 3750 Boeing Research & Technology 3751 P.O. Box 3707 3752 Seattle, WA 98124 3753 USA 3755 Email: fltemplin@acm.org