idnits 2.17.1 draft-templin-intarea-6706bis-17.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 (August 12, 2019) is 1712 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) -- Looks like a reference, but probably isn't: '0' on line 958 -- Looks like a reference, but probably isn't: '1' on line 973 -- Looks like a reference, but probably isn't: '2' on line 977 == Missing Reference: 'N' is mentioned on line 985, but not defined == Unused Reference: 'I-D.ietf-intarea-gue-extensions' is defined on line 3036, but no explicit reference was found in the text == Unused Reference: 'RFC2764' is defined on line 3112, but no explicit reference was found in the text == Unused Reference: 'RFC4291' is defined on line 3164, but no explicit reference was found in the text == Unused Reference: 'RFC4607' is defined on line 3199, but no explicit reference was found in the text == Unused Reference: 'RFC7269' is defined on line 3282, but no explicit reference was found in the text == Unused Reference: 'RFC8086' is defined on line 3298, but no explicit reference was found in the text == Outdated reference: A later version (-26) exists of draft-ietf-6man-segment-routing-header-22 == Outdated reference: A later version (-15) exists of draft-ietf-dmm-distributed-mobility-anchoring-13 == Outdated reference: A later version (-09) exists of draft-ietf-intarea-gue-07 == Outdated reference: A later version (-13) exists of draft-ietf-intarea-tunnels-09 == Outdated reference: A later version (-26) exists of draft-ietf-rtgwg-atn-bgp-02 ==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-08 == Outdated reference: A later version (-27) exists of draft-templin-v6ops-pdhost-24 -- 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 (~~), 16 warnings (==), 6 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group F. Templin, Ed. 3 Internet-Draft Boeing Research & Technology 4 Obsoletes: rfc5320, rfc5558, rfc5720, August 12, 2019 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: February 13, 2020 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-intarea-6706bis-17.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/registration services are employed for network 20 admission and to manage the routing system. Multilink operation, 21 mobility management, quality of service (QoS) signaling and route 22 optimization are naturally supported through dynamic neighbor cache 23 updates. Standard IP multicasting services are also supported. AERO 24 is a widely-applicable mobile internetworking service especially 25 well-suited to aviation services, mobile Virtual Private Networks 26 (VPNs) and many other 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 February 13, 2020. 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 . . . . . . . . . . . . . . . . . . . . . . . . 4 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 3.3.1. IPv4 Compatibility Routing . . . . . . . . . . . . . 15 69 3.4. AERO Addresses . . . . . . . . . . . . . . . . . . . . . 15 70 3.5. Spanning Partitioned AERO Networks (SPAN) . . . . . . . . 17 71 3.5.1. SPAN Compatibility Addressing . . . . . . . . . . . . 21 72 3.6. AERO Interface Characteristics . . . . . . . . . . . . . 21 73 3.7. AERO Interface Initialization . . . . . . . . . . . . . . 25 74 3.7.1. AERO Server/Gateway Behavior . . . . . . . . . . . . 26 75 3.7.2. AERO Proxy Behavior . . . . . . . . . . . . . . . . . 26 76 3.7.3. AERO Client Behavior . . . . . . . . . . . . . . . . 26 77 3.7.4. AERO Relay Behavior . . . . . . . . . . . . . . . . . 27 78 3.8. AERO Interface Neighbor Cache Maintenance . . . . . . . . 27 79 3.9. AERO Interface Encapsulation and Re-encapsulation . . . . 29 80 3.10. AERO Interface Decapsulation . . . . . . . . . . . . . . 30 81 3.11. AERO Interface Data Origin Authentication . . . . . . . . 30 82 3.12. AERO Interface Forwarding Algorithm . . . . . . . . . . . 31 83 3.12.1. Client Forwarding Algorithm . . . . . . . . . . . . 32 84 3.12.2. Proxy Forwarding Algorithm . . . . . . . . . . . . . 32 85 3.12.3. Server/Gateway Forwarding Algorithm . . . . . . . . 33 86 3.12.4. Relay Forwarding Algorithm . . . . . . . . . . . . . 35 87 3.13. AERO Interface MTU and Fragmentation . . . . . . . . . . 35 88 3.13.1. AERO MTU Requirements . . . . . . . . . . . . . . . 38 89 3.14. AERO Interface Error Handling . . . . . . . . . . . . . . 38 90 3.15. AERO Router Discovery, Prefix Delegation and 91 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 41 92 3.15.1. AERO ND/PD Service Model . . . . . . . . . . . . . . 41 93 3.15.2. AERO Client Behavior . . . . . . . . . . . . . . . . 42 94 3.15.3. AERO Server Behavior . . . . . . . . . . . . . . . . 44 95 3.16. The AERO Proxy . . . . . . . . . . . . . . . . . . . . . 46 96 3.17. AERO Route Optimization . . . . . . . . . . . . . . . . . 48 97 3.17.1. Route Optimization Initiation . . . . . . . . . . . 49 98 3.17.2. Relaying the NS . . . . . . . . . . . . . . . . . . 49 99 3.17.3. Processing the NS and Sending the NA . . . . . . . . 49 100 3.17.4. Relaying the NA . . . . . . . . . . . . . . . . . . 50 101 3.17.5. Processing the NA . . . . . . . . . . . . . . . . . 50 102 3.17.6. Route Optimization Maintenance . . . . . . . . . . . 50 103 3.18. Neighbor Unreachability Detection (NUD) . . . . . . . . . 51 104 3.19. Mobility Management and Quality of Service (QoS) . . . . 52 105 3.19.1. Mobility Update Messaging . . . . . . . . . . . . . 53 106 3.19.2. Announcing Link-Layer Address and/or QoS Preference 107 Changes . . . . . . . . . . . . . . . . . . . . . . 54 108 3.19.3. Bringing New Links Into Service . . . . . . . . . . 54 109 3.19.4. Removing Existing Links from Service . . . . . . . . 54 110 3.19.5. Moving to a New Server . . . . . . . . . . . . . . . 54 111 3.20. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 56 112 3.20.1. Source-Specific Multicast (SSM) . . . . . . . . . . 56 113 3.20.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 57 114 3.20.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 58 115 3.21. Operation over Multiple AERO Links (VLANs) . . . . . . . 58 116 3.22. DNS Considerations . . . . . . . . . . . . . . . . . . . 59 117 3.23. Transition Considerations . . . . . . . . . . . . . . . . 60 118 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 60 119 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 60 120 6. Security Considerations . . . . . . . . . . . . . . . . . . . 61 121 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 62 122 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 64 123 8.1. Normative References . . . . . . . . . . . . . . . . . . 64 124 8.2. Informative References . . . . . . . . . . . . . . . . . 65 125 Appendix A. AERO Alternate Encapsulations . . . . . . . . . . . 71 126 Appendix B. S/TLLAO Extensions for Special-Purpose Links . . . . 73 127 Appendix C. Non-Normative Considerations . . . . . . . . . . . . 75 128 C.1. Implementation Strategies for Route Optimization . . . . 75 129 C.2. Implicit Mobility Management . . . . . . . . . . . . . . 75 130 C.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 76 131 C.4. AERO Clients on the Open Internetwork . . . . . . . . . . 76 132 C.5. Operation on AERO Links with /64 ASPs . . . . . . . . . . 76 133 C.6. AERO Adaptations for SEcure Neighbor Discovery (SEND) . . 77 134 C.7. AERO Critical Infrastructure Considerations . . . . . . . 77 135 Appendix D. Change Log . . . . . . . . . . . . . . . . . . . . . 78 136 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 83 138 1. Introduction 140 Asymmetric Extended Route Optimization (AERO) fulfills the 141 requirements of Distributed Mobility Management (DMM) [RFC7333] and 142 route optimization [RFC5522] for aeronautical networking and other 143 network mobility use cases. AERO is based on a Non-Broadcast, 144 Multiple Access (NBMA) virtual link model known as the AERO link. 145 The AERO link is configured over one or more underlying 146 Internetworks, and nodes on the link can exchange IP packets via 147 tunneling. Multilink operation allows for increased reliability, 148 bandwidth optimization and traffic path diversity. 150 The AERO service comprises Clients, Proxys, Servers, and Gateways 151 that are seen as AERO link neighbors. Each node's AERO interface 152 uses an IPv6 link-local address format (known as the AERO address) 153 that supports operation of the IPv6 Neighbor Discovery (ND) protocol 154 [RFC4861] and links ND to IP forwarding. A node's AERO interface can 155 be configured over multiple underlying interfaces, and may therefore 156 may appear as a single interface with multiple link-layer addresses. 157 Each link-layer address is subject to change due to mobility and/or 158 QoS fluctuations, and link-layer address changes are signaled by ND 159 messaging the same as for any IPv6 link. 161 AERO links provide a cloud-based service where mobile nodes may use 162 any Server acting as a Mobility Anchor Point (MAP) and fixed nodes 163 may use any Gateway on the link for efficient communications. Fixed 164 nodes forward packets destined to other AERO nodes to the nearest 165 Gateway, which forwards them through the cloud. A mobile node's 166 initial packets are forwarded through the MAP, while direct routing 167 is supported through asymmetric route optimization while data packets 168 are flowing. Both unicast and multicast communications are 169 supported, and mobile nodes may efficiently move between locations 170 while maintaining continuous communications with correspondents and 171 without changing their IP Address. 173 AERO Relays are interconnected in a secured private BGP overlay 174 routing instance known as the "SPAN". The SPAN provides a hybrid 175 routing/bridging service to join the underlying Internetworks of 176 multiple disjoint administrative domains into a single unified AERO 177 link. Each AERO link instance is characterized by the set of 178 Mobility Service Prefixes (MSPs) common to all mobile nodes. The 179 link should extend to the point where a Gateway/MAP is on the optimal 180 route from any correspondent node on the link, and provides a gateway 181 between the underlying Internetwork and the SPAN. To the underlying 182 Internetwork, the Gateway/MAP is the source of a route to its MSP, 183 and hence uplink traffic to the mobile node is naturally routed to 184 the nearest Gateway/MAP. 186 AERO assumes the use of PIM Sparse Mode in support of multicast 187 communication. In support of Source Specific Multicast (SSM) when a 188 Mobile Node is the source, AERO route optimization ensures that a 189 shortest-path multicast tree is established with provisions for 190 mobility and multilink operation. In all other multicast scenarios 191 there are no AERO dependencies. 193 AERO was designed for aeronautical networking for both manned and 194 unmanned aircraft, where the aircraft is treated as a mobile node 195 that can connect an Internet of Things (IoT). AERO is also 196 applicable to a wide variety of other use cases. For example, it can 197 be used to coordinate the Virtual Private Network (VPN) links of 198 mobile nodes (e.g., cellphones, tablets, laptop computers, etc.) that 199 connect into a home enterprise network via public access networks 200 using services such as OpenVPN [OVPN]. Other applicable use cases 201 are also in scope. 203 The following numbered sections present the AERO specification. The 204 appendices at the end of the document are non-normative. 206 2. Terminology 208 The terminology in the normative references applies; the following 209 terms are defined within the scope of this document: 211 IPv6 Neighbor Discovery (ND) 212 an IPv6 control message service for coordinating neighbor 213 relationships between nodes connected to a common link. AERO 214 interfaces use the ND service specified in [RFC4861]. 216 IPv6 Prefix Delegation (PD) 217 a networking service for delegating IPv6 prefixes to nodes on the 218 link. The nominal PD service is DHCPv6 [RFC8415], however 219 alternate services (e.g., based on ND messaging) are also in scope 220 [I-D.templin-v6ops-pdhost][I-D.templin-6man-dhcpv6-ndopt]. Most 221 notably, a form of PD known as "prefix registration" can be used 222 if the Client knows its prefix in advance and can represent it in 223 the IPv6 source address of an ND message. 225 Access Network (ANET) 226 a node's first-hop data link service network, e.g., a radio access 227 network, cellular service provider network, corporate enterprise 228 network, or the public Internet itself. For secured ANETs, link- 229 layer security services such as IEEE 802.1X and physical-layer 230 security prevent unauthorized access internally while border 231 network-layer security services such as firewalls and proxies 232 prevent unauthorized outside access. 234 ANET interface 235 a node's attachment to a link in an ANET. 237 ANET address 238 an IP address assigned to a node's interface connection to an 239 ANET. 241 Internetwork (INET) 242 a connected IP network topology with a coherent routing and 243 addressing plan and that provides a transit backbone service for 244 ANET end systems. INETs also provide an underlay service over 245 which the AERO virtual link is configured. Example INETs include 246 corporate enterprise networks, aviation networks, and the public 247 Internet itself. When there is no administrative boundary between 248 an ANET and the INET, the ANET and INET are one and the same. 250 INET Partition 251 frequently, INETs such as large corporate enterprise networks are 252 sub-divided internally into separate isolated partitions. Each 253 partition is fully connected internally but disconnected from 254 other partitions, and there is no requirement that separate 255 partitions maintain consistent Internet Protocol and/or addressing 256 plans. (An INET partition is the same as a SPAN segment discussed 257 below.) 259 INET interface 260 a node's attachment to a link in an INET. 262 INET address 263 an IP address assigned to a node's interface connection to an 264 INET. 266 AERO link 267 a Non-Broadcast, Multiple Access (NBMA) tunnel virtual overlay 268 configured over one or more underlying INETs. Nodes on the AERO 269 link appear as single-hop neighbors from the perspective of the 270 virtual overlay even though they may be separated by many 271 underlying INET hops. AERO links may be configured over multiple 272 underlying SPAN segments (see below). 274 AERO interface 275 a node's attachment to an AERO link. Since the addresses assigned 276 to an AERO interface are managed for uniqueness, AERO interfaces 277 do not require Duplicate Address Detection (DAD) and therefore set 278 the administrative variable 'DupAddrDetectTransmits' to zero 279 [RFC4862]. 281 AERO address 282 an IPv6 link-local address assigned to an AERO interface and 283 constructed as specified in Section 3.4. 285 base AERO address 286 the lowest-numbered AERO address aggregated by the MNP (see 287 Section 3.4). 289 Mobility Service Prefix (MSP) 290 an IP prefix assigned to the AERO link and from which more- 291 specific Mobile Network Prefixes (MNPs) are derived. 293 Mobile Network Prefix (MNP) 294 an IP prefix allocated from an MSP and delegated to an AERO Client 295 or Gateway. 297 AERO node 298 a node that is connected to an AERO link, or that provides 299 services to other nodes on an AERO link. 301 AERO Client ("Client") 302 an AERO node that connects to one or more ANETs and requests MNP 303 PDs from AERO Servers. The Client assigns a Client AERO address 304 to the AERO interface for use in ND exchanges with other AERO 305 nodes and forwards packets to correspondents according to AERO 306 interface neighbor cache state. 308 AERO Server ("Server") 309 an INET node that configures an AERO interface to provide default 310 forwarding services and a Mobility Anchor Point (MAP) for AERO 311 Clients. The Server assigns an administratively-provisioned AERO 312 address to the AERO interface to support the operation of the ND/ 313 PD services, and advertises all of its associated MNPs via BGP 314 peerings with Relays. 316 AERO Gateway ("Gateway") 317 an AERO Server that also provides forwarding services between 318 nodes reached via the AERO link and correspondents on other links. 319 AERO Gateways are provisioned with MNPs (i.e., the same as for an 320 AERO Client) and run a dynamic routing protocol to discover any 321 non-MNP IP routes. In both cases, the Gateway advertises the 322 MSP(s) over INET interfaces, and distributes all of its associated 323 MNPs and non-MNP IP routes via BGP peerings with Relays (i.e., the 324 same as for an AERO Server). 326 AERO Relay ("Relay") 327 a node that provides hybrid routing/bridging services (as well as 328 a security trust anchor) for nodes on an AERO link. As a router, 329 the Relay forwards packets using standard IP forwarding. As a 330 bridge, the Relay forwards packets over the AERO link without 331 decrementing the IPv6 Hop Limit. AERO Relays peer with Servers 332 and other Relays to discover the full set of MNPs for the link as 333 well as any non-MNPs that are reachable via Gateways. 335 AERO Proxy ("Proxy") 336 a node that provides proxying services between Clients in an ANET 337 and Servers in external INETs. The AERO Proxy is a conduit 338 between the ANET and external INETs in the same manner as for 339 common web proxies, and behaves in a similar fashion as for ND 340 proxies [RFC4389]. 342 Spanning Partitioned AERO Networks (SPAN) 343 a means for bridging disjoint INET partitions as segments of a 344 unified AERO link the same as for a bridged campus LAN. The SPAN 345 is a mid-layer IPv6 encapsulation service in the AERO routing 346 system that supports a unified AERO link view for all segments. 347 Each segment in the SPAN is a distinct INET partition. 349 SPAN Service Prefix (SSP) 350 a global or unique local /96 IPv6 prefix assigned to the AERO link 351 to support SPAN services. 353 SPAN Partition Prefix (SPP) 354 a sub-prefix of the SPAN Service Prefix uniquely assigned to a 355 single SPAN segment. 357 SPAN Address 358 a global or unique local IPv6 address taken from a SPAN Partition 359 Prefix and constructed as specified in Section 3.5. SPAN 360 addresses are statelessly derived from AERO addresses, and vice- 361 versa. 363 ingress tunnel endpoint (ITE) 364 an AERO interface endpoint that injects encapsulated packets into 365 an AERO link. 367 egress tunnel endpoint (ETE) 368 an AERO interface endpoint that receives encapsulated packets from 369 an AERO link. 371 link-layer address 372 an IP address used as an encapsulation header source or 373 destination address from the perspective of the AERO interface. 374 When UDP encapsulation is used, the UDP port number is also 375 considered as part of the link-layer address. From the 376 perspective of the AERO interface, the link-layer address is 377 either an INET address for intra-segment encapsulation or a SPAN 378 address for inter-segment encapsulation. 380 network layer address 381 the source or destination address of an encapsulated IP packet 382 presented to the AERO interface. 384 end user network (EUN) 385 an internal virtual or external edge IP network that an AERO 386 Client or Gateway connects to the rest of the network via the AERO 387 interface. The Client/Gateway sees each EUN as a "downstream" 388 network, and sees the AERO interface as the point of attachment to 389 the "upstream" network. 391 Mobile Node (MN) 392 an AERO Client and all of its downstream-attached networks that 393 move together as a single unit, i.e., an end system that connects 394 an Internet of Things. 396 Mobile Router (MR) 397 a MN's on-board router that forwards packets between any 398 downstream-attached networks and the AERO link. 400 Mobility Anchor Point (MAP) 401 an AERO Server that is currently tracking and reporting the 402 mobility events of its associated Mobile Node Clients. 404 Route Optimization Source (ROS) 405 the AERO node nearest the source that initiates route 406 optimization. The ROS may be a Server or Proxy acting on behalf 407 of the source Client. 409 Route Optimization responder (ROR) 410 the AERO node nearest the target destination that responds to 411 route optimization requests. The ROR may be a Server acting as a 412 MAP on behalf of a target MNP Client, or a Gateway for a non-MNP 413 destination. 415 MAP List 416 a geographically and/or topologically referenced list of AERO 417 addresses of all MAPs within the same AERO link. There is a 418 single MAP list for the entire AERO link. 420 ROS List 421 a list of AERO/SPAN-to-INET address mappings of all ROSes within 422 the same SPAN segment. There is a distinct ROS list for each 423 segment. 425 Distributed Mobility Management (DMM) 426 a BGP-based overlay routing service coordinated by Servers and 427 Relays that tracks all MAP-to-Client associations. 429 Throughout the document, the simple terms "Client", "Server", 430 "Relay", "Proxy" and "Gateway" refer to "AERO Client", "AERO Server", 431 "AERO Relay", "AERO Proxy" and "AERO Gateway", respectively. 432 Capitalization is used to distinguish these terms from other common 433 Internetworking uses in which they appear without capitalization. 435 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 436 the names of node variables, messages and protocol constants) is used 437 throughout this document. Also, the term "IP" is used to generically 438 refer to either Internet Protocol version, i.e., IPv4 [RFC0791] or 439 IPv6 [RFC8200]. 441 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 442 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 443 document are to be interpreted as described in [RFC2119]. Lower case 444 uses of these words are not to be interpreted as carrying RFC2119 445 significance. 447 3. Asymmetric Extended Route Optimization (AERO) 449 The following sections specify the operation of IP over Asymmetric 450 Extended Route Optimization (AERO) links: 452 3.1. AERO Link Reference Model 453 +----------------+ 454 | AERO Relay R1 | 455 | Nbr: S1, S2, P1| 456 |(X1->S1; X2->S2)| 457 | MSP M1 | 458 +-+---------+--+-+ 459 +--------------+ | Secured | | +--------------+ 460 |AERO Server S1| | tunnels | | |AERO Server S2| 461 | Nbr: C1, R1 +-----+ | +-----+ Nbr: C2, R1 | 462 | default->R1 | | | default->R1 | 463 | X1->C1 | | | X2->C2 | 464 +-------+------+ | +------+-------+ 465 | AERO Link | | 466 X===+===+===================+==)===============+===+===X 467 | | | | 468 +-----+--------+ +--------+--+-----+ +--------+-----+ 469 |AERO Client C1| | AERO Proxy P1 | |AERO Client C2| 470 | Nbr: S1 | |(Proxy Nbr Cache)| | Nbr: S2 | 471 | default->S1 | +--------+--------+ | default->S2 | 472 | MNP X1 | | | MNP X2 | 473 +------+-------+ .--------+------. +-----+--------+ 474 | (- Proxyed Clients -) | 475 .-. `---------------' .-. 476 ,-( _)-. ,-( _)-. 477 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 478 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 479 `-(______)-' +-------+ +-------+ `-(______)-' 481 Figure 1: AERO Link Reference Model 483 Figure 1 presents the AERO link reference model. In this model: 485 o the AERO link is an overlay network service configured over one or 486 more underlying INET partitions which may be managed by different 487 administrative authorities and have incompatible protocols and/or 488 addressing plans. 490 o AERO Relay R1 aggregates Mobility Service Prefix (MSP) M1, 491 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 492 via BGP peerings over secured tunnels to Servers (S1, S2). Relays 493 use the SPAN service to bridge disjoint segments of a partitioned 494 AERO link. 496 o AERO Servers S1 and S2 configure secured tunnels with Relay R1 and 497 also act as Mobility Anchor Points (MAPs) and default routers for 498 their associated Clients C1 and C2. 500 o AERO Clients C1 and C2 associate with Servers S1 and S2, 501 respectively. They receive Mobile Network Prefix (MNP) 502 delegations X1 and X2, and also act as default routers for their 503 associated physical or internal virtual EUNs. Simple hosts H1 and 504 H2 attach to the EUNs served by Clients C1 and C2, respectively. 506 o AERO Proxy P1 configures a secured tunnel with Relay R1 and 507 provides proxy services for AERO Clients in secured enclaves that 508 cannot associate directly with other AERO link neighbors. 510 Each node on the AERO link maintains an AERO interface neighbor cache 511 and an IP forwarding table the same as for any link. Although the 512 figure shows a limited deployment, in common operational practice 513 there will normally be many additional Relays, Servers, Clients and 514 Proxys. 516 3.2. AERO Node Types 518 AERO Relays provide hybrid routing/bridging services (as well as a 519 security trust anchor) for nodes on an AERO link. Relays use 520 standard IPv6 routing to forward packets both within the same INET 521 partitions and between disjoint INET partitions based on a mid-layer 522 IPv6 encapsulation known as the SPAN header. The inner IP layer 523 experiences a virtual bridging service since the inner IP TTL/Hop 524 Limit is not decremented during forwarding. Each Relay also peers 525 with Servers and other Relays in a dynamic routing protocol instance 526 to provide a Distributed Mobility Management (DMM) service for the 527 list of active MNPs (see Section 3.3). Relays present the AERO link 528 as a set of one or more Mobility Service Prefixes (MSPs) but as link- 529 layer devices need not connect directly to the AERO link themselves 530 unless an administrative interface is desired. Relays configure 531 secured tunnels with Servers, Proxys and other Relays; they further 532 maintain IP forwarding table entries for each Mobile Network Prefix 533 (MNP) and any other reachable non-MNP prefixes. 535 AERO Servers provide default forwarding services and a Mobility 536 Anchor Point (MAP) for AERO Client Mobile Nodes (MNs). Each Server 537 also peers with Relays in a dynamic routing protocol instance to 538 advertise its list of associated MNPs (see Section 3.3). Servers 539 facilitate PD exchanges with Clients, where each delegated prefix 540 becomes an MNP taken from an MSP. Servers forward packets between 541 AERO interface neighbors and track each Client's mobility profiles. 543 AERO Clients receive MNPs through PD exchanges with AERO Servers over 544 the AERO link, and distribute the MNPs to nodes on EUNs. Each Client 545 can associate with a single Server or with multiple Servers (e.g., 546 for fault tolerance, load balancing, etc). A Client may also be co- 547 resident on the same physical or virtual platform as a Server; in 548 that case, the Client and Server behave as a single functional unit 549 and without the need for any Client/Server control messaging. 551 AERO Proxys provide a conduit for AERO Clients in ANETs to associate 552 with AERO Servers in external INETs. Client and Servers exchange 553 control plane messages via the Proxy acting as a bridge between the 554 ANET/INET boundary. The Proxy forwards data packets to and from 555 Clients according to forwarding information in the neighbor cache. 556 The Proxy function is specified in Section 3.16. 558 AERO Gateways are Servers that provide forwarding services between 559 the AERO interface and INET/EUN interfaces. Gateways are provisioned 560 with MNPs the same as for an AERO Client, and also run a dynamic 561 routing protocol to discover any non-MNP IP routes. The Gateway 562 advertises the MSP(s) to INETs, and distributes all of its associated 563 MNPs and non-MNP IP routes via BGP peerings with Relays. 565 AERO Relays, Servers, Proxys and Gateways are critical infrastructure 566 elements in fixed (i.e., non-mobile) INET deployments and hence have 567 permanent and unchanging INET addresses. AERO Clients are MNs that 568 connect via ANET interfaces, i.e., their ANET addresses may change 569 when the Client moves to a new ANET connection. 571 3.3. AERO Routing System 573 The AERO routing system comprises a private instance of the Border 574 Gateway Protocol (BGP) [RFC4271] that is coordinated between Relays 575 and Servers and does not interact with either the public Internet BGP 576 routing system or any underlying INET routing systems. 578 In a reference deployment, each Server is configured as an Autonomous 579 System Border Router (ASBR) for a stub Autonomous System (AS) using 580 an AS Number (ASN) that is unique within the BGP instance, and each 581 Server further uses eBGP to peer with one or more Relays but does not 582 peer with other Servers. Each INET of a multi-segment AERO link must 583 include one or more Relays, which peer with the Servers and Proxys 584 within that INET. All Relays within the same INET are members of the 585 same hub AS using a common ASN, and use iBGP to maintain a consistent 586 view of all active MNPs currently in service. The Relays of 587 different INETs peer with one another using eBGP. 589 Relays advertise the AERO link's MSPs and any non-MNP routes to each 590 of their Servers. This means that any aggregated non-MNPs (including 591 "default") are advertised to all Servers. Each Relay configures a 592 black-hole route for each of its MSPs. By black-holing the MSPs, the 593 Relay will maintain forwarding table entries only for the MNPs that 594 are currently active, and packets destined to all other MNPs will 595 correctly incur Destination Unreachable messages due to the black- 596 hole route. In this way, Servers have only partial topology 597 knowledge (i.e., they know only about the MNPs of their directly 598 associated Clients) and they forward all other packets to Relays 599 which have full topology knowledge. 601 Servers maintain a working set of associated MNPs, and dynamically 602 announce new MNPs and withdraw departed MNPs in eBGP updates to 603 Relays. Servers that are configured as Gateways also redistribute 604 non-MNP routes learned from non-AERO interfaces via their eBGP Relay 605 peerings. 607 Clients are expected to remain associated with their current Servers 608 for extended timeframes, however Servers SHOULD selectively suppress 609 updates for impatient Clients that repeatedly associate and 610 disassociate with them in order to dampen routing churn. Servers 611 that are configured as Gateways advertise the MSPs via INET/EUN 612 interfaces, and forward packets between INET/EUN interfaces and the 613 AERO interface using standard IP forwarding. 615 Scaling properties of the AERO routing system are limited by the 616 number of BGP routes that can be carried by Relays. As of 2015, the 617 global public Internet BGP routing system manages more than 500K 618 routes with linear growth and no signs of router resource exhaustion 619 [BGP]. More recent network emulation studies have also shown that a 620 single Relay can accommodate at least 1M dynamically changing BGP 621 routes even on a lightweight virtual machine, i.e., and without 622 requiring high-end dedicated router hardware. 624 Therefore, assuming each Relay can carry 1M or more routes, this 625 means that at least 1M Clients can be serviced by a single set of 626 Relays. A means of increasing scaling would be to assign a different 627 set of Relays for each set of MSPs. In that case, each Server still 628 peers with one or more Relays, but institutes route filters so that 629 BGP updates are only sent to the specific set of Relays that 630 aggregate the MSP. For example, if the MSP for the AERO link is 631 2001:db8::/32, a first set of Relays could service the MSP 632 2001:db8::/40, a second set of Relays could service 633 2001:db8:0100::/40, a third set could service 2001:db8:0200::/40, 634 etc. 636 Assuming up to 1K sets of Relays, the AERO routing system can then 637 accommodate 1B or more MNPs with no additional overhead (for example, 638 it should be possible to service 1B /64 MNPs taken from a /34 MSP and 639 even more for shorter prefixes). In this way, each set of Relays 640 services a specific set of MSPs that they advertise to the native 641 Internetwork routing system, and each Server configures MSP-specific 642 routes that list the correct set of Relays as next hops. This 643 arrangement also allows for natural incremental deployment, and can 644 support small scale initial deployments followed by dynamic 645 deployment of additional Clients, Servers and Relays without 646 disturbing the already-deployed base. 648 A full discussion of the BGP-based routing system used by AERO is 649 found in [I-D.ietf-rtgwg-atn-bgp]. The system provides for 650 Distributed Mobility Management (DMM) per the distributed mobility 651 anchoring architecture [I-D.ietf-dmm-distributed-mobility-anchoring]. 653 3.3.1. IPv4 Compatibility Routing 655 For IPv6 MNPs, the AERO routing system includes only IPv6 routes. 656 For IPv4 MNPs, the AERO routing system includes both IPv4 routes and 657 also IPv6 routes based on an IPv4-embedded IPv6 address format 658 discussed in Section 3.5.1. 660 3.4. AERO Addresses 662 A Client's AERO address is an IPv6 link-local address with an 663 interface identifier based on the Client's delegated MNP. Relay, 664 Server and Proxy AERO addresses are assigned from the range fe80::/96 665 and include an administratively-provisioned value in the lower 32 666 bits. 668 For IPv6, Client AERO addresses begin with the prefix fe80::/64 and 669 include in the interface identifier (i.e., the lower 64 bits) a 670 64-bit prefix taken from one of the Client's IPv6 MNPs. For example, 671 if the AERO Client receives the IPv6 MNP: 673 2001:db8:1000:2000::/56 675 it constructs its corresponding AERO addresses as: 677 fe80::2001:db8:1000:2000 679 fe80::2001:db8:1000:2001 681 fe80::2001:db8:1000:2002 683 ... etc. ... 685 fe80::2001:db8:1000:20ff 687 For IPv4, Client AERO addresses are based on an IPv4-mapped IPv6 688 address formed from an IPv4 MNP and with a Prefix Length of 96 plus 689 the MNP prefix length. For example, for the IPv4 MNP 192.0.2.32/28 690 the IPv4-mapped IPv6 MNP is: 692 0:0:0:0:0:FFFF:192.0.2.16/124 694 The Client then constructs its AERO addresses with the prefix 695 fe80::/64 and with the lower 64 bits of the IPv4-mapped IPv6 address 696 in the interface identifier as: 698 fe80::FFFF:192.0.2.16 700 fe80::FFFF:192.0.2.17 702 fe80::FFFF:192.0.2.18 704 ... etc. ... 706 fe80:FFFF:192.0.2.31 708 Relay, Server and Proxy AERO addresses are allocated from the range 709 fe80::/96, and MUST be managed for uniqueness. The lower 32 bits of 710 the AERO address includes a unique integer value (e.g., fe80::1, 711 fe80::2, fe80::3, etc.) as assigned by the administrative authority 712 for the link. If the link spans multiple SPAN segments, the AERO 713 addresses are assigned to each segment in 1x1 correspondence with 714 SPAN addresses (see: Section 3.5). The address fe80:: is the IPv6 715 link-local Subnet Router Anycast address, and the address 716 fe80::ffff:ffff is reserved as the unspecified AERO address. 718 The lowest-numbered AERO address from a Client's MNP delegation 719 serves as the "base" AERO address (for example, for the MNP 720 2001:db8:1000:2000::/56 the base AERO address is 721 fe80::2001:db8:1000:2000). The Client then assigns the base AERO 722 address to the AERO interface and uses it for the purpose of 723 maintaining the neighbor cache entry. The Server likewise uses the 724 AERO address as its index into the neighbor cache for this Client. 726 If the Client has multiple AERO addresses (i.e., when there are 727 multiple MNPs and/or MNPs with prefix lengths shorter than /64), the 728 Client originates ND messages using the base AERO address as the 729 source address and accepts and responds to ND messages destined to 730 any of its AERO addresses as equivalent to the base AERO address. In 731 this way, the Client maintains a single neighbor cache entry that may 732 be indexed by multiple AERO addresses. 734 The Client's Subnet Router Anycast address can be statelessly 735 determined from its AERO address by simply transposing the AERO 736 address into the upper N bits of the Anycast address followed by 737 128-N bits of zeros. For example, for the AERO address 738 fe80::2001:db8:1:2 the subnet router anycast address is 739 2001:db8:1:2::/64. 741 AERO addresses for mobile node Clients embed a MNP as discussed 742 above, while AERO addresses for non-MNP destinations are constructed 743 in exactly the same way. A Client AERO address therefore encodes 744 either an MNP if the prefix is reached via the SPAN or a non-MNP if 745 the prefix is reached via a Gateway. 747 3.5. Spanning Partitioned AERO Networks (SPAN) 749 In the simplest case, an AERO link configured over a single INET 750 appears as a single unified link with a consistent underlying network 751 addressing plan. In that case, all nodes on the link can exchange 752 packets via encapsulation with INET addresses, since the underlying 753 INET is connected. In common practice, however, an AERO link may be 754 partitioned into multiple "segments", where each segment is a 755 distinct INET potentially managed under a different administrative 756 authority (e.g., as for worldwide aviation service providers such as 757 ARINC, SITA, Inmarsat, etc.). Individual INETs may themselves be 758 partitioned internally, in which case each internal partition is seen 759 as a separate segment. 761 The addressing plan of each segment is consistent internally but will 762 often bear no relation to the addressing plans of other segments. 763 Each segment is also likely to be separated from others by network 764 security devices (e.g., firewalls, proxies, packet filtering 765 gateways, etc.), and in many cases disjoint segments may not even 766 have any common physical link connections at all. Therefore, nodes 767 can only be assured of exchanging packets directly with 768 correspondents in the same segment, and not with those in other 769 segments. The only means for joining the segments therefore is 770 through inter-domain peerings between AERO Relays. 772 The same as for traditional campus LANs, multiple AERO link segments 773 can be joined into a single unified link via a virtual bridging 774 service termed the "SPAN". The SPAN performs link-layer packet 775 forwarding between segments (i.e., bridging) without decrementing the 776 network-layer TTL/Hop Limit. The SPAN model is depicted in Figure 2: 778 . . . . . . . . . . . . . . . . . . . . . . . 779 . . 780 . .-(::::::::) . 781 . .-(::::::::::::)-. +-+ . 782 . (:::: Segment A :::)--|R|---+ . 783 . `-(::::::::::::)-' +-+ | . 784 . `-(::::::)-' | . 785 . | . 786 . .-(::::::::) | . 787 . .-(::::::::::::)-. +-+ | . 788 . (:::: Segment B :::)--|R|---+ . 789 . `-(::::::::::::)-' +-+ | . 790 . `-(::::::)-' | . 791 . | . 792 . .-(::::::::) | . 793 . .-(::::::::::::)-. +-+ | . 794 . (:::: Segment C :::)--|R|---+ . 795 . `-(::::::::::::)-' +-+ | . 796 . `-(::::::)-' | . 797 . | . 798 . ..(etc).. x . 799 . . 800 . . 801 . <- AERO Link Bridged by the SPAN -> . 802 . . . . . . . . . . . . . .. . . . . . . . . 804 Figure 2: The SPAN 806 To support the SPAN, AERO links require a reserved /64 IPv6 "SPAN 807 Service Prefix (SSP)". Although any routable IPv6 prefix can be 808 used, a Unique Local Address (ULA) prefix (e.g., fd00::/64) [RFC4389] 809 is recommended since border routers are commonly configured to 810 prevent packets with ULAs from being injected into the AERO link by 811 an external IPv6 node and from leaking out of the AERO link to the 812 outside world. 814 Each segment in the SPAN assigns a unique sub-prefix of SSP::/96 815 termed a "SPAN Partition Prefix (SPP)". For example, a first segment 816 could assign fd00::1000/116, a second could assign fd00::2000/116, a 817 third could assign fd00::3000/116, etc. The administrative 818 authorities for each segment must therefore coordinate to assure 819 mutually-exclusive SPP assignments, but internal provisioning of the 820 SPP is a local consideration for each administrative authority. 822 A "SPAN address" is an address taken from a SPP and assigned to a 823 Relay, Server or Proxy interface. SPAN addresses are formed by 824 simply replacing the upper portion of an administratively-assigned 825 AERO address with the SPP. For example, if the SPP is 826 fd00::1000/116, the SPAN address formed from the AERO address 827 fe80::1001 is simply fd00::1001. 829 An "INET address" is an address of a node's interface connection to 830 an INET. AERO/SPAN/INET address mappings are maintained as permanent 831 neighbor cache entires as discussed in Section 3.8. 833 AERO Relays serve as bridges to join multiple segments into a unified 834 AERO link over multiple diverse administrative domains. They support 835 the bridging function by first establishing forwarding table entries 836 for their SPPs either via standard BGP routing or static routes. For 837 example, if three Relays ('A', 'B' and 'C') from different segments 838 serviced the SPPs fd00::1000/116, fd00::2000/116 and fd00::3000/116 839 respectively, then the forwarding tables in each Relay are as 840 follows: 842 A: fd00::1000/116->local, fd00::2000/116->B, fd00::3000/116->C 844 B: fd00::1000/116->A, fd00::2000/116->local, fd00::3000/116->C 846 C: fd00::1000/116->A, fd00::2000/116->B, fd00::3000/116->local 848 These forwarding table entries are permanent and never change, since 849 they correspond to fixed infrastructure elements in their respective 850 segments. This provides the basis for a link-layer forwarding 851 service that cannot be disrupted by routing updates due to node 852 mobility. 854 With the SPPs in place in each Relay's forwarding table, control and 855 data packets sent between AERO nodes in different segments can 856 therefore be carried over the SPAN via encapsulation. For example, 857 when a source node in segment A forwards a packet with IPv6 address 858 2001:db8:1:2::1 to a destination node in segment C with IPv6 address 859 2001:db8:1000:2000::1, it first encapsulates the packet in a SPAN 860 header with source SPAN address taken from fd00::1000/116 (e.g., 861 fd00::1001) and destination SPAN address taken from fd00::3000/116 862 (e.g., fd00::3001). Next, it encapsulates the SPAN message in an 863 INET header with source address set to its own INET address (e.g., 864 192.0.2.100) and destination set to the INET address of a Relay 865 (e.g., 192.0.2.1). 867 SPAN encapsulation is based on Generic Packet Tunneling in IPv6 868 [RFC2473]; the encapsulation format in the above example is shown 869 inFigure 3: 871 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 872 | INET Header | 873 | src = 192.0.2.100 | 874 | dst = 192.0.2.1 | 875 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 876 | SPAN Header | 877 | src = fd00::1001 | 878 | dst = fd00::3001 | 879 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 880 | Inner IP Header | 881 | src = 2001:db8:1:2::1 | 882 | dst = 2001:db8:1000:2000::1 | 883 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 884 | | 885 ~ ~ 886 ~ Inner Packet Body ~ 887 ~ ~ 888 | | 889 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 891 Figure 3: SPAN Encapsulation 893 In this format, the inner IP header and packet body are the original 894 IP packet, the SPAN header is an IPv6 header prepared according to 895 [RFC2473], and the INET header is prepared according to Section 3.9. 896 A packet is said to be "forwarded/sent into the SPAN" when it is 897 encapsulated as described above then forwarded via a secured tunnel 898 to a neighboring Relay. 900 This gives rise to a routing system that contains both MNP routes 901 that may change dynamically due to regional node mobility and SPAN 902 routes that never change. The Relays can therefore provide link- 903 layer bridging by sending packets into the SPAN instead of network- 904 layer routing according to MNP routes. As a result, opportunities 905 for packet loss due to node mobility between different segments are 906 mitigated. 908 With reference to Figure 3, for a Client's AERO address the SPAN 909 address is simply set to the Subnet Router Anycast address. For non- 910 link-local addresses, the destination SPAN address may not be known 911 in advance for the first few packets of a flow sent via the SPAN. In 912 that case, the SPAN destination address is set to the original 913 packet's destination, and the SPAN routing system will direct the 914 packet to the correct SPAN egress node. (In the above example, the 915 SPAN destination address is simply 2001:db8:1000:2000::1.) 917 3.5.1. SPAN Compatibility Addressing 919 For IPv4 MNPs, Servers inject the MNP itself into the BGP routing 920 system and also inject a "SPAN Compatibility Prefix (SCP)" that 921 embeds the MNP. The SCP begins with the upper 64 bits of the SSP, 922 followed by the constant string "0000:FFFF" followed by the IPv4 MNP. 923 For example, if the SSP is fd00::/64 and the MNP is 192.0.2.0/24 then 924 the SCP is fd00::FFFF:192.0.2.0/120. 926 This allows for encapuslation of IPv4 packets in IPv6 headers with 927 "SPAN Compatibility Addresses (SCAs)". In this example, the SCA 928 corresponding to the SCP is simply fd00::FFFF:192.0.2.0, and can be 929 used as the SPAN destination address for packets forwarded via the 930 SPAN. This allows for forwarding of initial IPv4 packets over IPv6 931 SPAN routes, followed by route optimization for direct 932 communications. 934 3.6. AERO Interface Characteristics 936 AERO interfaces use encapsulation (see: Section 3.9) to exchange 937 packets with neighbors attached to the AERO link. 939 AERO interfaces maintain a neighbor cache for tracking per-neighbor 940 state the same as for any interface. AERO interfaces use ND messages 941 including Router Solicitation (RS), Router Advertisement (RA), 942 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 943 neighbor cache management. 945 AERO interface ND messages include Source/Target Link-Layer Address 946 Options (S/TLLAOs) formatted as shown in Figure 4: 948 0 1 2 3 949 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 1 950 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 951 | Type | Length | Prefix Length |R|X|N| Reserved| 952 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 953 | ifIndex [0] | Port Number [0] | 954 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 955 | | 956 + + 957 | | 958 + Link Layer Address [0] + 959 | | 960 + + 961 | | 962 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 963 |P00|P01|P02|P03|P04|P05|P06|P07|P08|P09|P10|P11|P12|P13|P14|P15| 964 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 965 |P16|P17|P18|P19|P20|P21|P22|P23|P24|P25|P26|P27|P28|P29|P30|P31| 966 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 967 |P32|P33|P34|P35|P36|P37|P38|P39|P40|P41|P42|P43|P44|P45|P46|P47| 968 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 969 |P48|P49|P50|P51|P52|P53|P54|P55|P56|P57|P58|P59|P60|P61|P62|P63| 970 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 971 ~ ifIndex [1] (16 bits) Port Number [1] (16 bits) ~ 972 ~ Link Layer Address [1] (128 bits) ~ 973 ~ P00 - P63 [1] (128 bits) ~ 974 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 975 ~ ifIndex [2] (16 bits) Port Number [2] (16 bits) ~ 976 ~ Link Layer Address [2] (128 bits) ~ 977 ~ P00 - P63 [2] (128 bits) ~ 978 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 979 . . 980 . . 981 . . 982 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 983 ~ ifIndex [N] (16 bits) Port Number [N] (16 bits) ~ 984 ~ Link Layer Address [N] (128 bits) ~ 985 ~ P00 - P63 [N] (128 bits) ~ 986 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 987 | Trailing zero padding (0 / 4 octets) | 988 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 990 Figure 4: AERO Source/Target Link-Layer Address Option (S/TLLAO) 991 Format 993 In this format: 995 o Type is set to '1' for SLLAO or '2' for TLLAO. 997 o Length is set to the number of 8 octet blocks in the option (with 998 trailing zero padding added if necessary to produce an integral 999 number of 8 octet blocks). 1001 o Prefix Length is set to the MNP prefix length in an ND message for 1002 the Client AERO address found in the source (RS), destination (RA) 1003 or target (NA) address; otherwise set to 0. 1005 o R (the "Release" bit) is set to '1' in an S/TLLAO in an RS/NA sent 1006 for the purpose of departing from a Server; otherwise, set to '0'. 1007 The recipient places the corresponding neighbor cache entry in the 1008 DEPARTED state. For RS messages, the recipient then releases the 1009 corresponding PD and returns an RA with Router Lifetime set to 1010 '0'. 1012 o X (the "proXy" bit) is set to '1' in the SLLAO of an RS/RA message 1013 by the Proxy when there is a Proxy in the path; otherwise, set to 1014 '0'. 1016 o N (the "(Network Address) Translator (NAT)" bit) is set to '1' in 1017 the SLLAO of an RA message by the Server if there is a translator 1018 in the path; otherwise, set to '0'. 1020 o Reserved is set to the value '0' on transmission and ignored on 1021 receipt. 1023 o A list of N (ifIndex[i], Port Number [i], Link Layer Address [i], 1024 P[i])-tuples are included as follows: 1026 * ifIndex [i] is set to a 16-bit integer value corresponding to 1027 an AERO node's ANET/INET interface. Once the node has assigned 1028 an ifIndex value to an interface, the assignment must remain 1029 unchanged until the node fully detaches from the AERO link. 1030 The value 0 is reserved as the Server's ifIndex, i.e., Servers 1031 MUST use ifIndex 0, and Clients MUST number their ifIndex 1032 values in the range of 1-0xffff. 1034 * Port Number [i] and Link Layer Address [i] are set to the 1035 encapsulation addresses required to send packets via the target 1036 node (or to '0' when the addresses are left unspecified). When 1037 UDP is not used as part of the encapsulation, Port Number is 1038 set to '0'. 1040 * P00 - P63 [i] are a set of Preferences that correspond to the 1041 64 Differentiated Service Code Point (DSCP) values [RFC2474]. 1042 Each P value is set to '0' ("disabled"), '1' ("low"), '2' 1043 ("medium") or '3' ("high") to indicate a QoS preference level 1044 for packet forwarding purposes. 1046 A Client's AERO interface may be configured over multiple ANET 1047 interface connections. For example, common mobile handheld devices 1048 have both wireless local area network ("WLAN") and cellular wireless 1049 links. These links are typically used "one at a time" with low-cost 1050 WLAN preferred and highly-available cellular wireless as a standby. 1051 In a more complex example, aircraft frequently have many wireless 1052 data link types (e.g. satellite-based, cellular, terrestrial, air-to- 1053 air directional, etc.) with diverse performance and cost properties. 1055 A Client's ANET interfaces are classified as follows: 1057 o Native interfaces connect to the open INET, and have a global IP 1058 address that is reachable from any INET correspondent. 1060 o NATed interfaces connect to an ANET behind a Network Address 1061 Translator (NAT). The NAT does not participate in any AERO 1062 control message signaling, but the Server can issue control 1063 messages on behalf of the Client. Clients that are behind a NAT 1064 are required to send periodic keepalive messages to keep NAT state 1065 alive when there are no data packets flowing. If no other 1066 periodic messaging service is available, the Client can send RS 1067 messages to receive RA replies from its Server(s). 1069 o VPNed interfaces use security encapsulation to a Virtual Private 1070 Network (VPN) server that also acts as an AERO Server. As with 1071 NATed links, the Server can issue control messages on behalf of 1072 the Client, but the Client need not send periodic keepalives in 1073 addition to those already used to maintain the VPN connection. 1075 o Proxyed interfaces connect to an ANET that is separated from the 1076 open INET by an AERO Proxy. Unlike NATed and VPNed interfaces, 1077 the Proxy can actively issue control messages on behalf of the 1078 Client. 1080 o Direct interfaces connect the Client directly to a neighbor 1081 without crossing any ANET/INET paths. An example is a line-of- 1082 sight link between a remote pilot and an unmanned aircraft. 1084 If a Client's multiple ANET interfaces are used "one at a time" 1085 (i.e., all other interfaces are in standby mode while one interface 1086 is active), then ND messages include only a single ifIndex in the S/ 1087 TLLAO and set to a constant value. In that case, the Client would 1088 appear to have a single ANET interface but with a dynamically 1089 changing Link Layer address. 1091 If the Client has multiple active ANET interfaces, then from the 1092 perspective of ND it would appear to have multiple link-layer 1093 addresses. In that case, ND messages S/TLLAOs MAY include multiple 1094 ifIndexes - each with a value that corresponds to a specific ANET 1095 interface. The first ifIndex in the S/TLLAO MUST correspond to the 1096 ANET interface over which the ND message is sent. Every ND message 1097 need not include all ifIndexes; for any ifIndex not included, the 1098 neighbor considers the status of that ifIndex as unchanged. 1100 When the Client includes an S/TLLAO for an ANET interface for which 1101 it is aware that there is a NAT on the path to the Server, or when a 1102 node includes an S/TLLAO solely for the purpose of announcing new QoS 1103 preferences, the node sets both Port Number and Link-Layer Address to 1104 0 to indicate that the addresses are unspecified at the network layer 1105 and must instead be derived from the link-layer encapsulation 1106 headers. 1108 Relay, Server and Proxy AERO interfaces may be configured over one or 1109 more secured tunnel interfaces. The AERO interface configures both 1110 an AERO address and its corresponding SPAN address, while the 1111 underlying secured tunnel interfaces are either unnumbered or 1112 configure the same SPAN address. The AERO interface encapsulates 1113 each IP packet in a SPAN header and presents the packet to the 1114 underlying secured tunnel interface. For Relays that do not 1115 configure an AERO interface, the secured tunnel interfaces themselves 1116 are exposed to the IP layer with each interface configuring the 1117 Relay's SPAN address. Routing protocols such as BGP therefore run 1118 directly over the Relay's secured tunnel interfaces. For nodes that 1119 configure an AERO interface, routing protocols such as BGP run over 1120 the AERO interface but do not employ SPAN encapsulation. Instead, 1121 the AERO interface presents the routing protocol messages directly to 1122 the underlying secured tunnels without applying encapsulation and 1123 while using the SPAN address as the source address. This distinction 1124 must be honored consistently according to each node's configuration 1125 so that the IP forwarding table will associate discovered IP routes 1126 with the correct interface. 1128 3.7. AERO Interface Initialization 1130 AERO Servers, Proxys and Clients configure AERO interfaces as their 1131 point of attachment to the AERO link. AERO nodes assign the MSPs for 1132 the link to their AERO interfaces (i.e., as a "route-to-interface") 1133 to ensure that packets with destination addresses covered by an MNP 1134 not explicitly assigned to a non-AERO interface are directed to the 1135 AERO interface. 1137 AERO interface initialization procedures for Servers, Proxys, Clients 1138 and Relays are discussed in the following sections. 1140 3.7.1. AERO Server/Gateway Behavior 1142 When a Server enables an AERO interface, it assigns AERO/SPAN 1143 addresses and configures permanent neighbor cache entries for 1144 neighbors in the same SPAN segment by consulting the ROS list for the 1145 segment. The Server also configures secured tunnels with one or more 1146 neighboring Relays and engages in a BGP routing protocol session with 1147 each Relay. 1149 The AERO interface provides a single interface abstraction to the IP 1150 layer, but internally comprises multiple secured tunnels as well as 1151 an NBMA nexus for sending encapsulated data packets to AERO interface 1152 neighbors. The Server further configures a service to facilitate ND/ 1153 PD exchanges with AERO Clients and manages per-Client neighbor cache 1154 entries and IP forwarding table entries based on control message 1155 exchanges. 1157 Gateways are simply Servers that run a dynamic routing protocol 1158 between the AERO interface and INET/EUN interfaces (see: 1159 Section 3.3). The Gateway provisions MNPs to networks on the INET/ 1160 EUN interfaces (i.e., the same as a Client would do) and advertises 1161 the MSP(s) for the AERO link over the INET/EUN interfaces. The 1162 Gateway further provides an attachment point of the AERO link to the 1163 non-MNP-based global topology. 1165 3.7.2. AERO Proxy Behavior 1167 When a Proxy enables an AERO interface, it assigns AERO/SPAN 1168 addresses and configures permanent neighbor cache entries the same as 1169 for Servers. The Proxy also configues secured tunnels with one or 1170 more neighboring Relays and maintains per-Client neighbor cache 1171 entries based on control message exchanges. 1173 3.7.3. AERO Client Behavior 1175 When a Client enables an AERO interface, it sends RS messages with 1176 ND/PD parameters over an ANET interface to one or more Servers in the 1177 MAP list, which return RA messages with corresponding PD parameters. 1178 (The RS/RA messages may pass through a Proxy in the case of a 1179 Client's Proxyed interface.) 1181 After the initial ND/PD message exchange, the Client assigns AERO 1182 addresses to the AERO interface based on the delegated prefix(es). 1183 The Client can then register additional ANET interfaces with the 1184 Server by sending an RS message over each ANET interface. 1186 3.7.4. AERO Relay Behavior 1188 AERO Relays need not connect directly to the AERO link, since they 1189 operate as link-layer forwarding devices instead of network layer 1190 routers. Configuration of AERO interfaces on Relays is therefore 1191 OPTIONAL, e.g., if an administrative interface is needed. Relays 1192 configure secured tunnels with Servers, Proxys and other Relays; they 1193 also configure AERO/SPAN addresses and permanent neighbor cache 1194 entries the same as Servers. Relays engage in a BGP routing protocol 1195 session with a subset of the Servers on the local SPAN segment, and 1196 with other Relays on the SPAN (see: Section 3.3). 1198 3.8. AERO Interface Neighbor Cache Maintenance 1200 Each AERO interface maintains a conceptual neighbor cache that 1201 includes an entry for each neighbor it communicates with on the AERO 1202 link per [RFC4861]. AERO interface neighbor cache entries are said 1203 to be one of "permanent", "symmetric", "asymmetric" or "proxy". 1205 Permanent neighbor cache entries are created through explicit 1206 administrative action; they have no timeout values and remain in 1207 place until explicitly deleted. AERO Servers and Proxys maintain 1208 permanent neighbor cache entries for all other Servers and Proxys 1209 within the same SPAN segment. Each entry maintains the mapping 1210 between the neighbor's network-layer AERO address and corresponding 1211 INET address. The list of all permanent neighbor cache entries for 1212 the SPAN segment is maintained in the segment's ROS list. 1214 Symmetric neighbor cache entries are created and maintained through 1215 RS/RA exchanges as specified in Section 3.15, and remain in place for 1216 durations bounded by ND/PD lifetimes. AERO Servers maintain 1217 symmetric neighbor cache entries for each of their associated 1218 Clients, and AERO Clients maintain symmetric neighbor cache entries 1219 for each of their associated Servers. The list of all Servers on the 1220 AERO link is maintained in the link's MAP list. 1222 Asymmetric neighbor cache entries are created or updated based on 1223 route optimization messaging as specified in Section 3.17, and are 1224 garbage-collected when keepalive timers expire. AERO route 1225 optimization sources (ROSs) maintain asymmetric neighbor cache 1226 entries for active targets with lifetimes based on ND messaging 1227 constants. Asymmetric neighbor cache entries are unidirectional 1228 since only the ROS and not the target (e.g., a Client's MAP) creates 1229 an entry. 1231 Proxy neighbor cache entries are created and maintained by AERO 1232 Proxys when they process Client/Server ND/PD exchanges, and remain in 1233 place for durations bounded by ND/PD lifetimes. AERO Proxys maintain 1234 proxy neighbor cache entries for each of their associated Clients. 1235 Proxy neighbor cache entries track the Client state and the address 1236 of the Client's associated Server. 1238 To the list of neighbor cache entry states in Section 7.3.2 of 1239 [RFC4861], AERO interfaces add an additional state DEPARTED that 1240 applies to symmetric and proxy neighbor cache entries for Clients 1241 that have recently departed. The interface sets a "DepartTime" 1242 variable for the neighbor cache entry to "DEPARTTIME" seconds. 1243 DepartTime is decremented unless a new ND message causes the state to 1244 return to REACHABLE. While a neighbor cache entry is in the DEPARTED 1245 state, packets destined to the target Client are forwarded to the 1246 Client's new location instead of being dropped. When DepartTime 1247 decrements to 0, the neighbor cache entry is deleted. It is 1248 RECOMMENDED that DEPARTTIME be set to the default constant value 40 1249 seconds to allow for packets in flight to be delivered while stale 1250 route optimization state may be present. 1252 When a target Server (acting as a Mobility Anchor Point (MAP)) 1253 receives a valid NS message used for route optimization, it searches 1254 for a symmetric neighbor cache entry for the target Client. The MAP 1255 then returns a solicited NA message without creating a neighbor cache 1256 entry for the ROS, but creates a target Client "Report List" entry 1257 for the ROS and sets a "ReportTime" variable for the entry to 1258 REPORTTIME seconds. The MAP resets ReportTime when it receives a new 1259 authentic NS message, and otherwise decrements ReportTime while no NS 1260 messages have been received. It is RECOMMENDED that REPORTTIME be 1261 set to the default constant value 40 seconds to allow a 10 second 1262 window so that route optimization can converge before ReportTime 1263 decrements below REACHABLETIME. 1265 When the ROS receives a solicited NA message response to its NS 1266 message, it creates or updates an asymmetric neighbor cache entry for 1267 the target network-layer and link-layer addresses. The ROS then 1268 (re)sets ReachableTime for the neighbor cache entry to REACHABLETIME 1269 seconds and uses this value to determine whether packets can be 1270 forwarded directly to the target, i.e., instead of via a default 1271 route. The ROS otherwise decrements ReachableTime while no further 1272 solicited NA messages arrive. It is RECOMMENDED that REACHABLETIME 1273 be set to the default constant value 30 seconds as specified in 1274 [RFC4861]. 1276 The ROS also uses the value MAX_UNICAST_SOLICIT to limit the number 1277 of NS keepalives sent when a correspondent may have gone unreachable, 1278 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1279 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1280 to limit the number of unsolicited NAs that can be sent based on a 1281 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1282 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1283 same as specified in [RFC4861]. 1285 Different values for DEPARTTIME, REPORTTIME, REACHABLETIME, 1286 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1287 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1288 different values are chosen, all nodes on the link MUST consistently 1289 configure the same values. Most importantly, DEPARTTIME and 1290 REPORTTIME SHOULD be set to a value that is sufficiently longer than 1291 REACHABLETIME to avoid packet loss due to stale route optimization 1292 state. 1294 3.9. AERO Interface Encapsulation and Re-encapsulation 1296 AERO interfaces encapsulate packets according to whether they are 1297 entering the AERO interface from the network layer or if they are 1298 being re-admitted into the same AERO link they arrived on. This 1299 latter form of encapsulation is known as "re-encapsulation". 1301 For packets entering the AERO interface from the network layer, the 1302 AERO interface copies the "TTL/Hop Limit", "Type of Service/Traffic 1303 Class" [RFC2983], "Flow Label"[RFC6438] (for IPv6) and "Congestion 1304 Experienced" [RFC3168] values in the packet's IP header into the 1305 corresponding fields in the encapsulation header(s). 1307 For packets undergoing re-encapsulation, the AERO interface instead 1308 copies these values from the original encapsulation header into the 1309 new encapsulation header, i.e., the values are transferred between 1310 encapsulation headers and *not* copied from the encapsulated packet's 1311 network-layer header. (Note especially that by copying the TTL/Hop 1312 Limit between encapsulation headers the value will eventually 1313 decrement to 0 if there is a (temporary) routing loop.) For IPv4 1314 encapsulation/re-encapsulation, the AERO interface sets the DF bit as 1315 discussed in Section 3.13. 1317 Proxy and Relay AERO interfaces encapsulate each packet in a SPAN 1318 header, then encapsulate the resulting SPAN packet in an INET header 1319 according to the next hop determined in the forwarding algorithm in 1320 Section 3.12. If the next hop is reached via a secured tunnel, the 1321 AERO interface uses an INET encapsulation format specific to the 1322 secured tunnel type (see: Section 6). If the next hop is reached via 1323 an unsecured underlying interface, the AERO interface instead uses 1324 Generic UDP Encapsulation (GUE) [I-D.ietf-intarea-gue] or an 1325 alternate minimal encapsulation format Appendix A. 1327 Client AERO interfaces can avoid encapsulation over underlying 1328 interfaces for which the first-hop access router is AERO-aware. For 1329 other underlying interfaces, Client interfaces use INET encapsulation 1330 the same as for Relays/Proxys but do not use SPAN encapsulation 1331 unless the underlying interface itself connects to the SPAN. 1333 When GUE encapsulation is used, the AERO interface next sets the UDP 1334 source port to a constant value that it will use in each successive 1335 packet it sends, and sets the UDP length field to the length of the 1336 SPAN packet plus 8 bytes for the UDP header itself plus the length of 1337 the GUE header (or 0 if GUE direct IP encapsulation is used). For 1338 packets sent to a Server or Relay, the AERO interface sets the UDP 1339 destination port to 8060, i.e., the IANA-registered port number for 1340 AERO. For packets sent to a Client, the AERO interface sets the UDP 1341 destination port to the port value stored in the neighbor cache entry 1342 for this Client. The AERO interface then either includes or omits 1343 the UDP checksum according to the GUE specification. 1345 AERO interfaces observes the packet sizing and fragmentation 1346 considerations found in Section 3.13. 1348 3.10. AERO Interface Decapsulation 1350 AERO interfaces decapsulate packets destined either to the AERO node 1351 itself or to a destination reached via an interface other than the 1352 AERO interface the packet was received on. When the encapsulated 1353 packet arrives in multiple fragments, the AERO interface reassembles 1354 as discussed in Section 3.13. Further decapsulation steps are 1355 performed according to the appropriate encapsulation format 1356 specification. 1358 3.11. AERO Interface Data Origin Authentication 1360 AERO nodes employ simple data origin authentication procedures. In 1361 particular: 1363 o AERO Relays, Servers and Proxys accept encapsulated data packets 1364 and control messages received from secured tunnels. 1366 o AERO Servers and Proxys accept encapsulated data packets and NS 1367 messages used for Neighbor Unreachability Detection (NUD) received 1368 from a source found in the ROS list. 1370 o AERO Proxys and Clients accept packets that originate from within 1371 the same secured ANET. 1373 o AERO Clients and Gateways accept packets from downstream network 1374 correspondents based on ingress filtering. 1376 AERO nodes silently drop any packets that do not satisfy the above 1377 data origin authentication procedures. Further security 1378 considerations are discussed Section 6. 1380 3.12. AERO Interface Forwarding Algorithm 1382 IP packets enter a node's AERO interface either from the network 1383 layer (i.e., from a local application or the IP forwarding system) or 1384 from the link layer (i.e., from an AERO interface neighbor). All 1385 packets entering a node's AERO interface first undergo data origin 1386 authentication as discussed in Section 3.11. Packets that satisfy 1387 data origin authentication are processed further, while all others 1388 are dropped silently. 1390 Packets that enter the AERO interface from the network layer are 1391 forwarded to an AERO interface neighbor. Packets that enter the AERO 1392 interface from the link layer are either re-admitted into the AERO 1393 link or forwarded to the network layer where they are subject to 1394 either local delivery or IP forwarding. In all cases, the AERO 1395 interface itself MUST NOT decrement the network layer TTL/Hop-count 1396 since its forwarding actions occur below the network layer. 1398 AERO interfaces may have multiple underlying ANET/INET interfaces 1399 and/or neighbor cache entries for neighbors with multiple ifIndex 1400 registrations (see Section 3.6). The AERO interface uses each 1401 packet's DSCP value (and/or port number) to select an outgoing ANET/ 1402 INET interface based on the node's own QoS preferences, and also to 1403 select a destination link-layer address based on the neighbor's ANET/ 1404 INET interface with the highest preference. AERO implementations 1405 SHOULD allow for QoS preference values to be modified at runtime 1406 through network management. 1408 If multiple outgoing interfaces and/or neighbor interfaces have a 1409 preference of "high", the AERO node replicates the packet and sends 1410 one copy via each of the (outgoing / neighbor) interface pairs; 1411 otherwise, the node sends a single copy of the packet via the 1412 interface with the highest preference. AERO nodes keep track of 1413 which ANET/INET interfaces are currently "reachable" or 1414 "unreachable", and only use "reachable" interfaces for forwarding 1415 purposes. 1417 The following sections discuss the AERO interface forwarding 1418 algorithms for Clients, Proxys, Servers and Relays. In the following 1419 discussion, a packet's destination address is said to "match" if it 1420 is the same as a cached address, or if it is covered by a cached 1421 prefix (which may be encoded in an AERO address). 1423 3.12.1. Client Forwarding Algorithm 1425 When an IP packet enters a Client's AERO interface from the network 1426 layer the Client searches for an asymmetric neighbor cache entry that 1427 matches the destination. If there is a match, the Client uses one or 1428 more "reachable" neighbor interfaces in the entry for packet 1429 forwarding. If there is no asymmetric neighbor cache entry, the 1430 Client instead forwards the packet toward a Server (the packet is 1431 intercepted by a Proxy if there is a Proxy on the path). 1433 When an IP packet enters a Client's AERO interface from the link- 1434 layer, if the destination matches one of the Client's MNPs or link- 1435 local addresses the Client decapsulates the packet (if necessary) and 1436 delivers it to the network layer. Otherwise, the Client drops the 1437 packet and MAY return a network-layer ICMP Destination Unreachable 1438 message subject to rate limiting (see: Section 3.14). 1440 3.12.2. Proxy Forwarding Algorithm 1442 For control messages originating from or destined to a Client, the 1443 Proxy intercepts the message and updates its proxy neighbor cache 1444 entry for the Client. The Proxy then forwards a (proxyed) copy of 1445 the control message. (For example, the Proxy forwards a proxied 1446 version of a Client's NS/RS message to the target neighbor, and 1447 forwards a proxied version of the NA/RA reply to the Client.) 1449 When the Proxy receives a data packet from a Client within the ANET, 1450 the Proxy searches for an asymmetric neighbor cache entry that 1451 matches the destination and forwards the packet as follows: 1453 o if the destination matches an asymmetric neighbor cache entry, the 1454 Proxy uses one or more "reachable" neighbor interfaces in the 1455 entry for packet forwarding via encapsulation. If the neighbor 1456 interface is in the same SPAN segment, the Proxy forwards the 1457 packet directly to the neighbor; otherwise, it forwards the packet 1458 to a Relay. 1460 o else, the Proxy encapsulates and forwards the packet to a Relay 1461 while using the packet's destination address as the SPAN 1462 destination address. (If the destination is an AERO address, the 1463 Proxy instead uses the corresponding Subnet Router Anycast address 1464 for Client AERO addresses and the SPAN address for 1465 administratively-provisioned AERO addresses.). 1467 When the Proxy receives an encapsulated data packet from an INET 1468 neighbor or from a secured tunnel, it accepts the packet only if data 1469 origin authentication succeeds and the SPAN destination address is 1470 its own adddress. If the packet is a SPAN fragment, the Proxy then 1471 adds the fragment to the reassembly buffer and returns if the 1472 reassembly is still incomplete. Otherwise, the Proxy reassembles the 1473 packet (if necessary) and continues processing. 1475 Next, the Proxy searches for a proxy neighbor cache entry that 1476 matches the destination. If there is a proxy neighbor cache entry in 1477 the REACHABLE state, the Proxy decapsulates and forwards the packet 1478 to the Client. If the neighbor cache entry is in the DEPARTED state, 1479 the Proxy instead re-encpasulates the message with the address of the 1480 Client's Server as the SPAN destination address and forwards the 1481 packet to a Relay. The Proxy then returns an unsolicited NA message 1482 as discussed in Section 3.19. If there is no neighbor cache entry, 1483 the Proxy instead discards the packet. 1485 3.12.3. Server/Gateway Forwarding Algorithm 1487 For control messages destined to a target Client's AERO address that 1488 are received from a secured tunnel, the Server (acting as a MAP) 1489 intercepts the message and sends an appropriate response on behalf of 1490 the Client. (For example, the Server sends an NA message reply in 1491 response to an NS message directed to one of its associated Clients.) 1492 If the Client's neighbor cache entry is in the DEPARTED state, 1493 however, the Server instead forwards the packet to the Client's new 1494 Server as discussed in Section 3.19. 1496 When the Server receives an encapsulated data packet from an INET 1497 neighbor or from a secured tunnel, it accepts the packet only if data 1498 origin authentication succeeds. If the SPAN destination address is 1499 its own adddress, the Server reassembles if necessary and discards 1500 the SPAN header (if the reassembly is incomplete, the Server instead 1501 adds the fragment to the reassembly buffer and returns). The Server 1502 then continues processing as follows: 1504 o if the destination matches a symmetric neighbor cache entry in the 1505 REACHABLE state the Server prepares the packet for forwarding to 1506 the destination Client. If the current header is a SPAN header, 1507 the Server reassembles if necessary and discards the SPAN header 1508 (if the reassembly is incomplete, the Server instead adds the 1509 fragment to the reassembly buffer and returns). The Server then 1510 forwards the packet according to the cached link-layer 1511 information, while using SPAN encapsulation for the Client's 1512 Proxyed/Native interfaces, simple INET encapsulation for NATed/ 1513 VPNed interfaces, or no encapsulation for Direct interfaces. If 1514 the packet is destined to the same Client from which it arrived 1515 (i.e., if the packet was forwarded by one of the Client's Proxys), 1516 the Server forwards the packet via a different "reachable" 1517 neighbor interface than the one the packet arrived on. If there 1518 are no "reachable" neighbor interfaces, the Server drops the 1519 packet. 1521 o else, if the destination matches a symmetric neighbor cache entry 1522 in the DEPARETED state the Server encapsulates the packet in a new 1523 SPAN header and forwards it to the Client's new Server (noting 1524 that the encapsulation may result in the addition of a second SPAN 1525 header). The Server uses its own SPAN address as the source and 1526 the SPAN address of the new Server as the destination. 1528 o else, if the destination matches an asymmetric neighbor cache 1529 entry, the Server uses one or more "reachable" neighbor interfaces 1530 in the entry for packet forwarding via the local INET if the 1531 neighbor is in the same SPAN segment or via a Relay otherwise. 1533 o else, if the destination is an AERO address that is not assigned 1534 on the AERO interface the Server drops the packet. 1536 o else, the Server (acting as a Gateway) releases the packet to the 1537 network layer for local delivery or IP forwarding. Based on the 1538 information in the forwarding table, the network layer may return 1539 the packet to the same AERO interface in which case further 1540 processing occurs as below. (Note that this arrangement 1541 accommodates common implementations in which the IP forwarding 1542 table is not accessible from within the AERO interface. If the 1543 AERO interface can directly access the IP forwarding table, the 1544 forwarding table lookup can instead be performed internally from 1545 within the AERO interface itself.) 1547 When the Server's AERO interface receives a data packet from the 1548 network layer or from a NATed/VPNed/Direct Client, it processes the 1549 packet according to the network-layer destination address as follows: 1551 o if the destination matches a symmetric or asymmetric neighbor 1552 cache entry the Server processes the packet as above. 1554 o else, the Server encapsulates the packet and forwards it to a 1555 Relay. For administratively-assigned AERO address destinations, 1556 the Server uses the SPAN address corresponding to the destination 1557 as the SPAN destination address. For Client AERO address 1558 destinations, the Server uses the Subnet Router Anycast address 1559 corresponding to the destination as the SPAN destination address. 1560 For all others, the Server uses the packet's destination IP 1561 address as the SPAN destination address. 1563 3.12.4. Relay Forwarding Algorithm 1565 Relays forward packets over secured tunnels the same as any IP 1566 router. When the Relay receives an encapsulated packet via a secured 1567 tunnel, it removes the INET header and searches for a forwarding 1568 table entry that matches the destination address in the next header. 1569 The Relay then processes the packet as follows: 1571 o if the destination matches one of the Relay's own addresses, the 1572 Relay submits the packet for local delivery. 1574 o else, if the destination matches a forwarding table entry the 1575 Relay forwards the packet via a secured tunnel to the next hop. 1576 If the destination matches an MSP without matching an MNP, 1577 however, the Relay instead drops the packet and returns an ICMP 1578 Destination Unreachable message subject to rate limiting (see: 1579 Section 3.14). 1581 o else, the Relay drops the packet and returns an ICMP Destination 1582 Unreachable as above. 1584 As for any IP router, the Relay decrements the TTL/Hop Limit when it 1585 forwards the packet. If the packet is encapsulated in a SPAN header, 1586 only the Hop Limit in the SPAN header is decremented, and not the 1587 TTL/Hop Limit in the inner packet header. 1589 3.13. AERO Interface MTU and Fragmentation 1591 The AERO interface is the node's attachment to the AERO link. For 1592 AERO link neighbor underlying interface paths that do not require 1593 encapsulation, the AERO interface sends unencapsulated IP packets. 1594 For other paths, the AERO interface acts as a tunnel ingress when it 1595 sends packets to the neighbor and as a tunnel egress when it receives 1596 packets from the neighbor. 1598 AERO interfaces configure an MTU the same as for any IP interface, 1599 however the MTU does not reflect the physical size of any links in 1600 the path but rather determines the maximum size for reassembly. AERO 1601 interfaces observe the packet sizing considerations for tunnels 1602 discussed in [I-D.ietf-intarea-tunnels] and as specified below. 1604 The Internet Protocol expects that IP packets will either be 1605 delivered to the destination or a suitable Packet Too Big (PTB) 1606 message returned to support the process known as IP Path MTU 1607 Discovery (PMTUD) [RFC1191][RFC8201]. However, PTB messages may be 1608 crafted for malicious purposes or lost in the network [RFC2923]. 1609 This can be especially problematic for tunnels, where a condition 1610 known as a PMTUD "black hole" can result. For these reasons, AERO 1611 interfaces employ operational procedures that avoid interactions with 1612 PMTUD, including the use of fragmentation when necessary. 1614 AERO interfaces observe three different types of fragmentation. 1615 Source fragmentation occurs when the AERO interface (acting as a 1616 tunnel ingress) fragments the encapsulated packet into multiple 1617 fragments before admitting each fragment into the tunnel. Network 1618 fragmentation occurs when an encapsulated packet admitted into the 1619 tunnel by the ingress is fragmented by an IPv4 router on the path to 1620 the egress. Finally, link-layer fragmentation (aka link adaptation) 1621 occurs at a layer below IP and is coordinated between underlying data 1622 link endpoints. 1624 IPv6 specifies a minimum link Maximum Transmission Unit (MTU) of 1280 1625 bytes [RFC8200]. Although IPv4 specifies a smaller minimum link MTU 1626 of 68 bytes [RFC0791], AERO interfaces also observe the IPv6 minimum 1627 for IPv4 even if encapsulated packets may incur network 1628 fragmentation. 1630 IPv6 specifies a minimum Maximum Reassembly Unit (MRU) of 1500 bytes 1631 [RFC8200], while the minimum MRU for IPv4 is only 576 bytes [RFC1122] 1632 (but, note that many standard IPv6 over IPv4 tunnel types already 1633 assume a larger MRU than the IPv4 minimum). 1635 AERO interfaces therefore configure an MTU that MUST NOT be smaller 1636 than 1280 bytes, MUST NOT be larger than the minimum MRU among all 1637 nodes on the AERO link minus the encapsulation overhead ("ENCAPS"), 1638 and SHOULD NOT be smaller than 1500 bytes. AERO interfaces also 1639 configure a Maximum Segment Unit (MSU) as the maximum-sized 1640 encapsulated packet that the ingress can inject into the tunnel 1641 without source fragmentation. The MSU value MUST NOT be larger than 1642 1280 bytes unless there is operational assurance that a larger size 1643 can traverse the link along all paths. 1645 All AERO interfaces on the link MUST configure the same MTU value for 1646 reasons cited in [RFC3819][RFC4861]; in particular, multicast support 1647 requires a common MTU value among all nodes on the link. All AERO 1648 interfaces MUST configure an MRU large enough to reassemble packets 1649 up to (MTU+ENCAPS) bytes in length; nodes that cannot configure a 1650 large-enough MRU MUST NOT enable an AERO interface. For example, for 1651 an MTU of 1500 bytes an appropriate MRU might be 2KB. 1653 The network layer proceeds as follows when it forwards an IP packet 1654 to the AERO interface. For each IPv4 packet that is larger than the 1655 AERO interface MTU and with DF set to 0, the network layer uses IPv4 1656 fragmentation to break the packet into a minimum number of non- 1657 overlapping fragments where the first fragment is no larger than the 1658 MTU and the remaining fragments are no larger than the first. For 1659 all other IP packets, if the packet is larger than the AERO interface 1660 MTU, the network layer drops the packet and returns a PTB message to 1661 the original source. Otherwise, the network layer admits each IP 1662 packet or fragment into the AERO interface. 1664 For each IP packet admitted into the AERO interface, if the neighbor 1665 is reached via an underlying interface that does not require 1666 encapsulation the AERO interface proceeds according to the underlying 1667 interface MTU. If the packet is no larger than the underlying 1668 interface MTU, the AERO interface presents the packet to the 1669 underlying interface. Otherwise, for IPv4 packets with DF set to 0 1670 the AERO interface uses IPv4 fragmentation to break the packet into 1671 fragments no larger than the underlying interface MTU. For other 1672 packets, the AERO interface either performs link adaptation or drops 1673 the packet and returns a PTB message to the original source. (If the 1674 original source corresponds to a local application, the PTB would 1675 appear to have originated from a router on the path when in fact it 1676 was locally generated from within the AERO interface.) In the same 1677 way, when a packet that has been admitted into the AERO link reaches 1678 a target neighbor but is too large to be delivered over the final-hop 1679 underlying interface, the target either performs link adaptation or 1680 drops the packet and returns a PTB. Link adaptation is preferred in 1681 both cases when possible to avoid packet loss. 1683 For underlying interfaces that require encapsulation, the AERO 1684 interface (acting as a tunnel ingress) instead encapsulates the 1685 packet in a SPAN header. If the SPAN packet is larger than the MSU, 1686 the ingress source fragments the SPAN packet into a minimum number of 1687 non-overlapping fragments where the first fragment is no larger than 1688 the MSU and the remaining fragments are no larger than the first. 1689 The ingress then encapsulates each SPAN packet/fragment in an INET 1690 header and admits them into the tunnel. For IPv4, the ingress sets 1691 the DF bit to 0 in the INET header in case any network fragmentation 1692 is necessary. The encapsulated packets will be delivered to the 1693 egress, which reassembles them into a whole packet if necessary. 1695 By fragmenting at the SPAN layer instead of lower layers, standard 1696 IPv6 fragmentation and reassembly [RFC8200] ensures that IPv4 issues 1697 such as data corruption due to reassembly misassociations will not 1698 occur [RFC6864][RFC4963]. Incomplete reassemblies are discarded when 1699 the corresponding neighbor cache entry is deleted. 1701 Since the SPAN header and IPv6 fragment extension header reduces the 1702 room available for packet data, but the original source has no way to 1703 control its insertion, the ingress MUST include their lengths in 1704 ENCAPS even for packets in which the header is absent. 1706 3.13.1. AERO MTU Requirements 1708 In light of the above considerations, AERO interfaces configure an 1709 MTU of 9180 bytes. This means that the AERO interface MUST be 1710 capable of reassembling original IP packets up to 9180 bytes in 1711 length. When an IP packet is admitted into an AERO interface, the 1712 interface encapsulates the packet using SPAN encapsulation and 1713 fragments the encapsulated packet into fragments that are no larger 1714 than 1280 bytes. The fragments will be reassembled by the tunnel 1715 egress that services the final destination. 1717 AERO Clients behind Proxys MAY configure an MTU smaller than 9180 1718 (but no smaller than IP minimum link MTU). If Clients configure a 1719 diversity of MTUs (e.g., 1280, 1500, 4KB, 8KB, etc.) then neighbors 1720 on the link would appear to have multiple MTUs. IPv6 Path MTU 1721 Discovery [RFC8201] accounts for this possibility since MTU discovery 1722 must be performed even between nodes that appear to be connected to 1723 the same link. 1725 Applications that cannot tolerate loss in the network due to MTU 1726 restrictions should restrict themselves to sending packets no larger 1727 than the IP minimum link MTU, i.e., even if the current path MTU 1728 would appear to support a larger size. This is due to the fact that 1729 routing changes could cause the path to traverse links with smaller 1730 MTUs at any given point in time. 1732 3.14. AERO Interface Error Handling 1734 When an AERO node admits encapsulated packets into the AERO 1735 interface, it may receive link-layer or network-layer error 1736 indications. 1738 A link-layer error indication is an ICMP error message generated by a 1739 router in the INET on the path to the neighbor or by the neighbor 1740 itself. The message includes an IP header with the address of the 1741 node that generated the error as the source address and with the 1742 link-layer address of the AERO node as the destination address. 1744 The IP header is followed by an ICMP header that includes an error 1745 Type, Code and Checksum. Valid type values include "Destination 1746 Unreachable", "Time Exceeded" and "Parameter Problem" 1747 [RFC0792][RFC4443]. (AERO interfaces ignore all link-layer IPv4 1748 "Fragmentation Needed" and IPv6 "Packet Too Big" messages since they 1749 only emit packets that are guaranteed to be no larger than the IP 1750 minimum link MTU as discussed in Section 3.13.) 1752 The ICMP header is followed by the leading portion of the packet that 1753 generated the error, also known as the "packet-in-error". For 1754 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1755 much of invoking packet as possible without the ICMPv6 packet 1756 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1757 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1758 "Internet Header + 64 bits of Original Data Datagram", however 1759 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1760 ICMP datagram SHOULD contain as much of the original datagram as 1761 possible without the length of the ICMP datagram exceeding 576 1762 bytes". 1764 The link-layer error message format is shown in Figure 5 (where, "L2" 1765 and "L3" refer to link-layer and network-layer, respectively): 1767 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1768 ~ ~ 1769 | L2 IP Header of | 1770 | error message | 1771 ~ ~ 1772 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1773 | L2 ICMP Header | 1774 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1775 ~ ~ P 1776 | IP and other encapsulation | a 1777 | headers of original L3 packet | c 1778 ~ ~ k 1779 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1780 ~ ~ t 1781 | IP header of | 1782 | original L3 packet | i 1783 ~ ~ n 1784 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1785 ~ ~ e 1786 | Upper layer headers and | r 1787 | leading portion of body | r 1788 | of the original L3 packet | o 1789 ~ ~ r 1790 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1792 Figure 5: AERO Interface Link-Layer Error Message Format 1794 The AERO node rules for processing these link-layer error messages 1795 are as follows: 1797 o When an AERO node receives a link-layer Parameter Problem message, 1798 it processes the message the same as described as for ordinary 1799 ICMP errors in the normative references [RFC0792][RFC4443]. 1801 o When an AERO node receives persistent link-layer Time Exceeded 1802 messages, the IP ID field may be wrapping before earlier fragments 1803 awaiting reassembly have been processed. In that case, the node 1804 SHOULD begin including integrity checks and/or institute rate 1805 limits for subsequent packets. 1807 o When an AERO node receives persistent link-layer Destination 1808 Unreachable messages in response to encapsulated packets that it 1809 sends to one of its asymmetric neighbor correspondents, the node 1810 SHOULD process the message as an indication that a path may be 1811 failing, and MAY initiate NUD over that path. If it receives 1812 Destination Unreachable messages on many or all paths, the node 1813 SHOULD set ReachableTime for the corresponding asymmetric neighbor 1814 cache entry to 0 and allow future packets destined to the 1815 correspondent to flow through a default route. 1817 o When an AERO Client receives persistent link-layer Destination 1818 Unreachable messages in response to encapsulated packets that it 1819 sends to one of its symmetric neighbor Servers, the Client SHOULD 1820 mark the path as unusable and use another path. If it receives 1821 Destination Unreachable messages on many or all paths, the Client 1822 SHOULD associate with a new Server and release its association 1823 with the old Server as specified in Section 3.19.5. 1825 o When an AERO Server receives persistent link-layer Destination 1826 Unreachable messages in response to encapsulated packets that it 1827 sends to one of its symmetric neighbor Clients, the Server SHOULD 1828 mark the underlying path as unusable and use another underlying 1829 path. If it receives Destination Unreachable messages on multiple 1830 paths, the Server should take no further actions unless it 1831 receives an explicit ND/PD release message or if the PD lifetime 1832 expires. In that case, the Server MUST release the Client's 1833 delegated MNP, withdraw the MNP from the AERO routing system and 1834 delete the neighbor cache entry. 1836 o When an AERO Server or Proxy receives link-layer Destination 1837 Unreachable messages in response to an encapsulated packet that it 1838 sends to one of its permanent neighbors, it treats the messages as 1839 an indication that the path to the neighbor may be failing. 1840 However, the dynamic routing protocol should soon reconverge and 1841 correct the temporary outage. 1843 When an AERO Relay receives a packet for which the network-layer 1844 destination address is covered by an MSP, if there is no more- 1845 specific routing information for the destination the Relay drops the 1846 packet and returns a network-layer Destination Unreachable message 1847 subject to rate limiting. The Relay writes the network-layer source 1848 address of the original packet as the destination address and uses 1849 one of its non link-local addresses as the source address of the 1850 message. 1852 When an AERO node receives an encapsulated packet for which the 1853 reassembly buffer it too small, it drops the packet and returns a 1854 network-layer Packet Too Big (PTB) message. The node first writes 1855 the MRU value into the PTB message MTU field, writes the network- 1856 layer source address of the original packet as the destination 1857 address and writes one of its non link-local addresses as the source 1858 address. 1860 3.15. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1862 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1863 coordinated as discussed in the following Sections. 1865 3.15.1. AERO ND/PD Service Model 1867 Each AERO Server on the link configures a PD service to facilitate 1868 Client requests. Each Server is provisioned with a database of MNP- 1869 to-Client ID mappings for all Clients enrolled in the AERO service, 1870 as well as any information necessary to authenticate each Client. 1871 The Client database is maintained by a central administrative 1872 authority for the AERO link and securely distributed to all Servers, 1873 e.g., via the Lightweight Directory Access Protocol (LDAP) [RFC4511], 1874 via static configuration, etc. Clients can receive new PDs from new 1875 Servers before releasing PDs received from existing Servers for 1876 service continuity. Clients receive the same service regardless of 1877 the Servers they select, although selecting Servers that are 1878 topologically nearby may provide better routing. 1880 AERO Clients and Servers use ND messages to maintain neighbor cache 1881 entries. AERO Servers configure their AERO interfaces as advertising 1882 interfaces, and therefore send unicast RA messages with configuration 1883 information in response to a Client's RS message. Thereafter, 1884 Clients send additional RS messages to refresh prefix and/or router 1885 lifetimes. 1887 AERO Clients and Servers include PD parameters in RS/RA messages (see 1888 [I-D.templin-6man-dhcpv6-ndopt] for ND/PD alternatives). The unified 1889 ND/PD messages are exchanged between Client and Server according to 1890 the prefix management schedule required by the PD service. If the 1891 Client knows its MNP in advance, it can include its AERO address as 1892 the source address of an RS message and with an SLLAO with a valid 1893 Prefix Length for the MNP. If the Server (and Proxy) accept the 1894 Client's MNP assertion, they inject the prefix into the routing 1895 system and establish the necessary neighbor cache state. 1897 The following sections specify the Client and Server behavior. 1899 3.15.2. AERO Client Behavior 1901 AERO Clients can discover the addresses of Servers in the MAP list 1902 via static configuration (e.g., from a flat-file map of Server 1903 addresses and locations), or through an automated means such as 1904 Domain Name System (DNS) name resolution [RFC1035]. In the absence 1905 of other information, the Client can resolve the DNS Fully-Qualified 1906 Domain Name (FQDN) "linkupnetworks.[domainname]" where 1907 "linkupnetworks" is a constant text string and "[domainname]" is a 1908 DNS suffix for the AERO link (e.g., "example.com"). Alternatively, 1909 the Client can discover the Server's address through a multicast RS 1910 as described below. 1912 To associate with a Server, the Client acts as a requesting router to 1913 request MNPs. The Client prepares an RS message with PD parameters 1914 (e.g., with an SLLAO with non-zero Prefix Length) and SHOULD include 1915 a Nonce and Timestamp option if the Client needs to correlate RA 1916 replies. If the Client already knows the Server's AERO address, it 1917 includes the AERO address as the network-layer destination address; 1918 otherwise, it includes all-routers multicast (ff02::2) as the 1919 network-layer destination address. If the Client already knows its 1920 own AERO address, it uses the AERO address as the network-layer 1921 source address; otherwise, it uses the unspecified AERO address 1922 (fe80::ffff:ffff) as the network-layer source address. 1924 The Client next includes an SLLAO in the RS message formatted as 1925 described in Section 3.6 to register its link-layer information with 1926 the Server. The first ifIndex in the SLLAO MUST correspond to the 1927 underlying interface over which the Client will send the RS message. 1928 The Client MAY include additional ifIndexes specific to other 1929 underlying interfaces, but if so it MUST set their Port Number and 1930 Link Layer Address fields to 0. If the Client is connected to an 1931 ANET for which encapsulation is required, the Client finally 1932 encapsulates the RS message in an ANET header with its own ANET 1933 address as the source address and the INET address of the Server as 1934 the destination. 1936 The Client then sends the RS message (either directly via Direct 1937 interfaces, via INET encapsulation for NATed interfaces, via a VPN 1938 for VPNed interfaces, via a Proxy for proxyed interfaces or via a 1939 Relay for native interfaces) and waits for an RA message reply (see 1940 Section 3.15.3). The Client retries up to MAX_RTR_SOLICITATIONS 1941 times until an RA is received. If the Client receives no RAs, or if 1942 it receives an RA with Router Lifetime set to 0, the Client SHOULD 1943 abandon this Server and try another Server. Otherwise, the Client 1944 processes the PD information found in the RA message. 1946 Next, the Client creates a symmetric neighbor cache entry with the 1947 Server's AERO address as the network-layer address and the address in 1948 the SLLAO for ifIndex 0 as the Server's INET address. The Client 1949 records the RA Router Lifetime field value in the neighbor cache 1950 entry as the time for which the Server has committed to maintaining 1951 the MNP in the routing system. The Client then autoconfigures AERO 1952 addresses for each of the delegated MNPs and assigns them to the AERO 1953 interface. The Client also caches any MSPs included in Route 1954 Information Options (RIOs) [RFC4191] as MSPs to associate with the 1955 AERO link, and assigns the MTU value in the MTU option to its AERO 1956 interface while configuring an appropriate MRU. 1958 The Client then registers additional ANET interfaces with the Server 1959 by sending RS messages via each additional ANET interface. The RS 1960 messages include the same parameters as for the initial RS/RA 1961 exchange, but with destination address set to the Server's AERO 1962 address and with an ifIndex in the SLLAO specific to the ANET 1963 interface. 1965 The Client examines the X and N bits in the SLLAO with S set to 1 in 1966 each RA message it receives. If X is 1 the Client infers that there 1967 is a Proxy on the path, and if N is 1 the Client infers that there is 1968 a NAT on the path. If N is 1, the Client SHOULD set Port Number and 1969 Link-Layer Address to 0 of any subsequent ND messages it sends to the 1970 Server over that link. 1972 Following autoconfiguration, the Client sub-delegates the MNPs to its 1973 attached EUNs and/or the Client's own internal virtual interfaces as 1974 described in [I-D.templin-v6ops-pdhost] to support the Client's 1975 downstream attached "Internet of Things (IoT)". The Client 1976 subsequently maintains its MNP delegations through each of its 1977 Servers by sending additional RS messages before Router Lifetime 1978 expires. 1980 After the Client registers its ANET interfaces, it may wish to change 1981 one or more registrations, e.g., if an ANET interface changes address 1982 or becomes unavailable, if QoS preferences change, etc. To do so, 1983 the Client prepares an RS message to send over any available ANET 1984 interface. The RS MUST include an ifIndex for the selected ANET 1985 interface as the first ifIndex in the SLLAO and MAY include any 1986 additional ifIndexes specific to other ANET interfaces. The Client 1987 includes fresh P(i) values for each ifIndex to update the Server's 1988 neighbor cache entry. If the Client wishes to update only the P(i) 1989 values, it sets the Port Number and Link-Layer Address fields to 0. 1990 If the Client wishes to disable the underlying interface, it sets all 1991 P(i) values to 0. When the Client receives the Server's RA response, 1992 it has assurance that the Server has been updated with the new 1993 information. 1995 If the Client wishes to discontinue use of a Server it issues an RS 1996 message over any underlying interface with an SLLAO with R set to 1 . 1997 When the Server processes the message, it releases the MNP, sets the 1998 symmetric neighbor cache entry state for the Client to DEPARTED, 1999 withdraws the IP route from the routing system and returns an RA 2000 reply with Router Lifetime set to 0. 2002 Clients SHOULD NOT remain associated with multiple Servers for long 2003 durations, since routing inconsistencies could cause some fragments 2004 of a fragmented packet to be delivered to Server A and others to 2005 Server B. Clients SHOULD therefore associate with a single primary 2006 Server, and send a departure message to the former Server soon after 2007 moving to a new Server. 2009 3.15.3. AERO Server Behavior 2011 AERO Servers act as IP routers and support a PD service for Clients. 2012 Servers arrange to add their AERO and INET addresses to a static map 2013 of Server addresses for the link and/or the DNS resource records for 2014 the FQDN "linkupnetworks.[domainname]" before entering service. The 2015 list of Server addresses should be geographically and/or 2016 topologically referenced, and forms the MAP list for the AERO link. 2018 When a Server receives a prospective Client's RS message on its AERO 2019 interface, it SHOULD return an immediate RA reply with Router 2020 Lifetime set to 0 if it is currently too busy or otherwise unable to 2021 service the Client. Otherwise, the Server authenticates the RS 2022 message and processes the PD parameters. The Server first determines 2023 the correct MNPs to delegate to the Client by searching the Client 2024 database. When the Server delegates the MNPs, it also creates an IP 2025 forwarding table entry for each MNP so that the MNPs are propagated 2026 into the routing system (see: Section 3.3). For IPv6, the Server 2027 creates a single IPv6 forwarding table entry for each MNP. For IPv4, 2028 the Server creates both an IPv4 forwarding table entry and an IPv6 2029 forwarding table entry with the IPv4-mapped IPv6 address 2030 corresponding to the IPv4 address. 2032 The Server next creates a symmetric neighbor cache entry for the 2033 Client using the base AERO address as the network-layer address and 2034 with lifetime set to no more than the smallest PD lifetime. Next, 2035 the Server updates the neighbor cache entry by recording the 2036 information in each ifIndex in the RS SLLAO including the Port 2037 Number, Link Layer Address and P(i) values. For the first ifIndex, 2038 however, the Server records the actual INET header source addresses 2039 instead of those that appear in the SLLAO in case there was a NAT in 2040 the path. The Server also records the value of the X bit to indicate 2041 whether there is a Proxy on the path. 2043 Next, the Server prepares an RA message using its AERO address as the 2044 network-layer source address and the network-layer source address of 2045 the RS message as the network-layer destination address. The Server 2046 includes the delegated MNPs, any other PD parameters and an SLLAO 2047 with the Link Layer Address set to the Server's SPAN address and with 2048 ifIndex set to 0. The Server then includes one or more RIOs that 2049 encode the MSPs for the AERO link, plus an MTU option for the link 2050 MTU (see Section 3.13). The Server finally forwards the message to 2051 the Client using SPAN, INET or NULL encapsulation according to the 2052 Client interface type. (For Proxy/Native interfaces, the Server 2053 encapsulates the message in a SPAN header with source address set to 2054 its own SPAN address and destination address set to the Proxy's (or 2055 Client's) SPAN address, then forwards the message into the SPAN.) 2057 After the initial RS/RA exchange, the Server maintains the symmetric 2058 neighbor cache entry for the Client. If the Client (or Proxy) issues 2059 additional NS/RS messages, the Server resets ReachableTime. If the 2060 Client (or Proxy) issues an RS with PD release parameters (e.g., by 2061 including an SLLAO with R set to 1), or if the Client becomes 2062 unreachable, the Server sets the Client's symmetric neighbor cache 2063 entry to the DEPARTED state and withdraws the IP routes from the AERO 2064 routing system. 2066 The Server processes these and any other Client ND/PD messages, and 2067 returns an NA/RA reply. The Server may also issue unsolicited RA 2068 messages, e.g., with PD reconfigure parameters to cause the Client to 2069 renegotiate its PDs, with Router Lifetime set to 0 if it can no 2070 longer service this Client, etc. Finally, If the symmetric neighbor 2071 cache entry is in the DEPARTED state, the Server deletes the entry 2072 after DepartTime expires. 2074 3.15.3.1. Lightweight DHCPv6 Relay Agent (LDRA) 2076 When DHCPv6 is used as the ND/PD service back end, AERO Clients and 2077 Servers are always on the same link (i.e., the AERO link) from the 2078 perspective of DHCPv6. However, in some implementations the DHCPv6 2079 server and ND function may be located in separate modules. In that 2080 case, the Server's AERO interface module can act as a Lightweight 2081 DHCPv6 Relay Agent (LDRA)[RFC6221] to relay PD messages to and from 2082 the DHCPv6 server module. 2084 When the LDRA receives an authentic RS message, it extracts the PD 2085 message parameters and uses them to construct an IPv6/UDP/DHCPv6 2086 message. It sets the IPv6 source address to the source address of 2087 the RS message, sets the IPv6 destination address to 2088 'All_DHCP_Relay_Agents_and_Servers' and sets the UDP fields to values 2089 that will be understood by the DHCPv6 server. 2091 The LDRA then wraps the message in a DHCPv6 'Relay-Forward' message 2092 header and includes an 'Interface-Id' option that includes enough 2093 information to allow the LDRA to forward the resulting Reply message 2094 back to the Client (e.g., the Client's link-layer addresses, a 2095 security association identifier, etc.). The LDRA also wraps the 2096 SLLAO into the Interface-Id option, then forwards the message to the 2097 DHCPv6 server. 2099 When the DHCPv6 server prepares a Reply message, it wraps the message 2100 in a 'Relay-Reply' message and echoes the Interface-Id option. The 2101 DHCPv6 server then delivers the Relay-Reply message to the LDRA, 2102 which discards the Relay-Reply wrapper and IPv6/UDP headers, then 2103 uses the DHCPv6 message to construct an RA response to the Client. 2104 The Server uses the information in the Interface-Id option to prepare 2105 the RA message and to cache the link-layer addresses taken from the 2106 SLLAOs echoed in the Interface-Id option. 2108 3.16. The AERO Proxy 2110 Clients may connect to ANETs that require a perimeter security 2111 gateway to enable communications to Servers in outside INETs. In 2112 that case, the ANET can employ an AERO Proxy. The Proxy is located 2113 at the ANET/INET border and listens for RS messages originating from 2114 or RA messages destined to ANET Clients. The Proxy acts on these 2115 control messages as follows: 2117 o when the Proxy receives an RS message from a new ANET Client, it 2118 first authenticates the message then examines the network-layer 2119 destination address. If the destination address is a Server's 2120 AERO address, the Proxy proceeds to the next step. Otherwise, if 2121 the destination is all-routers multicast the Proxy selects a 2122 "nearby" Server that is likely to be a good candidate to serve the 2123 Client and replaces the destination address with the Server's AERO 2124 address. Next, the Proxy creates a proxy neighbor cache entry and 2125 caches the Client and Server addresses along with any identifying 2126 information including Transaction IDs, Client Identifiers, Nonce 2127 values, etc. The Proxy then examines the address in the first 2128 ifIndex in the RS message SLLAO. If the address is different than 2129 the Client's ANET address, the Proxy notes that the Client is 2130 behind a NAT. The Proxy then sets X to 1 and changes the Link 2131 Layer Address to its own SPAN address. The Proxy finally 2132 encapsulates the RS message in a SPAN header with destination set 2133 to the Server's SPAN address then forwards the message into the 2134 SPAN. 2136 o when the Server receives the RS message, it authenticates the 2137 message then creates or updates a symmetric neighbor cache entry 2138 for the Client with the Proxy's SPAN address as the link-layer 2139 address. The Server then sends an RA message back to the Proxy 2140 via the SPAN. 2142 o when the Proxy receives the RA message, it matches the message 2143 with the RS that created the proxy neighbor cache entry. The 2144 Proxy then caches the PD route information as a mapping from the 2145 Client's MNPs to the Client's ANET address, and sets the neighbor 2146 cache entry state to REACHABLE. The Proxy then changes the SLLAO 2147 Link Layer Address to its own ANET address, sets X to 1, sets N to 2148 1 if the Client is behind a NAT, then re-encapsulates the RA 2149 message in an ANET header and forwards it to the Client. 2151 After the initial RS/RA exchange, the Proxy forwards any Client data 2152 packets for which there is no matching asymmetric neighbor cache 2153 entry to the Server via the SPAN. Finally, the Proxy forwards any 2154 Client data destined to an asymmetric neighbor cache target directly 2155 to the target according to the link-layer information - the process 2156 of establishing asymmetric neighbor cache entries is specified in 2157 Section 3.17. 2159 While the Client is still attached to the ANET, the Proxy 2160 periodically sends NS/RS messages to update the Server's symmetric 2161 neighbor cache entries on behalf of the Client and/or to convey QoS 2162 updates. If the Server ceases to send solicited NA/RA responses, the 2163 Proxy marks the Server as unreachable and sends an unsolicited RA 2164 with Router Lifetime set to zero to inform the Client that this 2165 Server is no longer able to provide Service. If the Client becomes 2166 unreachable, the Proxy sets the neighbor cache entry state to 2167 DEPARTED and sends an RS message to the Server with an SLLAO with the 2168 Client's ifIndex and with all P values set to 0 so that the Server 2169 will de-register this interface. Although the Proxy engages in these 2170 ND exchanges on behalf of the Client, the Client can also send ND 2171 messages on its own behalf, e.g., if it is in a better position than 2172 the Proxy to convey QoS changes, etc. 2174 In some ANETs that employ a Proxy, the Client's MNP can be injected 2175 into the ANET routing system. In that case, the Client can send data 2176 messages without encapsulation so that the ANET native routing system 2177 transports the unencapsulated packets to the Proxy. This can be very 2178 beneficial, e.g., if the Client connects to the ANET via low-end data 2179 links such as some aviation wireless links. 2181 If the first-hop ANET access router is AERO-aware, the Client can 2182 avoid encapsulation for both its control and data messages. When the 2183 Client connects to the link, it can send an unencapsulated RS message 2184 with source address set to its AERO address and with destination 2185 address set to the AERO address of the Client's selected Server or to 2186 all-routers multicast. The Client includes an SLLAO with ifIndex, 2187 Prefix Length and P(i) information but with Port Number and Link- 2188 Layer Address set to 0. 2190 The Client then sends the unencapsulated RS message, which will be 2191 intercepted by the AERO-Aware access router. The access router then 2192 encapsulates the RS message in an ANET header with its own address as 2193 the source address and the address of a Proxy as the destination 2194 address. The access router further remembers the address of the 2195 Proxy so that it can encapsulate future data packets from the Client 2196 via the same Proxy. If the access router needs to change to a new 2197 Proxy, it simply sends another RS message toward the Server via the 2198 new Proxy on behalf of the Client. 2200 In some cases, the access router and Proxy may be one and the same 2201 node. In that case, the node would be located on the same physical 2202 link as the Client, but its message exchanges with the Server would 2203 need to pass through a security gateway at the ANET/INET border. The 2204 method for deploying access routers and Proxys (i.e. as a single node 2205 or multiple nodes) is an ANET-local administrative consideration. 2207 3.17. AERO Route Optimization 2209 While data packets are flowing between a source and target node, 2210 route optimization SHOULD be used. Route optimization is initiated 2211 by the first eligible Route Optimization Source (ROS) closest to the 2212 source as follows: 2214 o For Clients on VPNed, NATed and Direct interfaces, the Server is 2215 the ROS. 2217 o For Clients on Proxyed interfaces, the Proxy is the ROS. 2219 o For Clients on native interfaces, the Client itself is the ROS. 2221 o For correspondent nodes on INET/EUN interfaces serviced by a 2222 Gateway, the Gateway is the ROS. 2224 The route optimization procedure is conducted between the ROS and the 2225 target Server/Gateway acting as a Route Optimization Responder (ROR) 2226 in the same manner as for IPv6 ND Address Resolution and using the 2227 same NS/NA messaging. The target may either be a MNP Client serviced 2228 by a Server, or a non-MNP correspondent reachable via a Gateway. 2230 The procedures are specified in the following sections. 2232 3.17.1. Route Optimization Initiation 2234 While data packets are flowing from the source node toward a target 2235 node, the ROS performs address resolution by sending an NS message to 2236 receive a solicited NA message from the ROR. 2238 When the ROS sends an NS, it includes the AERO address of the ROS as 2239 the source address (e.g., fe80::1) and the AERO address corresponding 2240 to the data packet's destination address as the destination address 2241 (e.g., if the destination address is 2001:db8:1:2::1 then the 2242 corresponding AERO address is fe80::2001:db8:1:2). The NS message 2243 includes an SLLAO with Link Layer Address set to the SPAN address of 2244 the ROS and with all other fields set to 0. The message SHOULD also 2245 include a Nonce and Timestamp option if the ROS needs to correlate NA 2246 replies. 2248 The ROS then encapsulates the NS message in a SPAN header with source 2249 set to its own SPAN address and destination set to the data packet's 2250 destination address, then sends it into the SPAN without decrementing 2251 the network-layer TTL/Hop Limit field. 2253 3.17.2. Relaying the NS 2255 When the Relay receives the NS message from the ROS, it discards the 2256 INET header and determines that the ROR is the next hop by consulting 2257 its standard IPv6 forwarding table for the SPAN header destination 2258 address. The Relay then forwards the SPAN message toward the ROR the 2259 same as for any IPv6 router. The final-hop Relay in the SPAN will 2260 deliver the message via a secured tunnel to the ROR. 2262 3.17.3. Processing the NS and Sending the NA 2264 When the ROR receives the NS message, it examines the AERO 2265 destination address to determine whether it has a neighbor cache 2266 entry and/or route that matches the target; if not, it drops the NS 2267 message and returns from processing. Next, if the target belongs to 2268 an MNP Client neighbor in the DEPARTED state the ROR changes the NS 2269 message SPAN destination address to the address of the Client's new 2270 Server, forwards the message into the SPAN and returns from 2271 processing. If the target belongs to an MNP Client neighbor in the 2272 REACHABLE state, the ROR instead adds the AERO source address to the 2273 target Client's Report List with time set to ReportTime. If the 2274 target belongs to a non-MNP route, the ROR continues processing 2275 without adding an entry to the Report List. 2277 The ROR then prepares a solicited NA message to send back to the ROS 2278 but does not create a neighbor cache entry. The ROR sets the NA 2279 source address to the destination AERO address of the NS, and 2280 includes the Nonce value received in the NS plus the current 2281 Timestamp. The ROR next includes a TLLAO with ifIndex set to 0, with 2282 all P(i) values set to "low", and with Link Layer Address set to the 2283 ROR's SPAN address. If the target belongs to an MNP Client, the ROR 2284 sets the Prefix Length to the MNP prefix length; otherwise, it sets 2285 Prefix Length to the maximum of the non-MNP prefix length and 64. 2286 (Note that a /64 limit is imposed to avoid causing the ROS to set 2287 short prefixes (e.g., "default") that would match destinations for 2288 which the routing system includes more-specific prefixes. Note also 2289 that prefix lengths longer than /64 are out of scope for this 2290 specification.) 2292 If the target belongs to an MNP Client, the ROR next includes 2293 additional ifIndexes for all of the target Client's ANET interfaces. 2294 For NATed, VPNed and Direct interfaces, the Link Layer Addresses are 2295 the SPAN address of the ROR. For Proxyed and native interfaces, the 2296 Link Layer Addresses are the SPAN addresses of the Proxys and the 2297 Client's native interfaces. The ROR finally encapsulates the NA 2298 message in a SPAN header with source set to its own SPAN address and 2299 destination set to the source SPAN address of the NS message, then 2300 forwards the message into the SPAN without decrementing the network- 2301 layer TTL/Hop Limit field. 2303 3.17.4. Relaying the NA 2305 When the Relay receives the NA message from the ROR, it discards the 2306 INET header and determines that the ROS is the next hop by consulting 2307 its standard IPv6 forwarding table for the SPAN header destination 2308 address. The Relay then forwards the SPAN-encapsulated NA message 2309 toward the ROS the same as for any IPv6 router. The final-hop Relay 2310 in the SPAN will deliver the message via a secured tunnel to the ROS. 2312 3.17.5. Processing the NA 2314 When the ROS receives the solicited NA message, it discards the INET 2315 and SPAN headers. The ROS next verifies the Nonce and Timestamp 2316 values, then creates an asymmetric neighbor cache entry for the ROR 2317 and caches all information found in the solicited NA TLLAOs. The ROS 2318 finally sets the asymmetric neighbor cache entry lifetime to 2319 ReachableTime seconds. 2321 3.17.6. Route Optimization Maintenance 2323 Following route optimization, the ROS forwards future data packets 2324 destined to the target via the addresses found in the cached link- 2325 layer information. The route optimization is shared by all sources 2326 that send packets to the target via the ROS, i.e., and not just the 2327 source on behalf of which the route optimization was initiated. 2329 While new data packets destined to the target are flowing through the 2330 ROS, it sends additional NS messages to the ROR before ReachableTime 2331 expires to receive a fresh solicited NA message the same as described 2332 in the previous sections. (Route optimization refreshment strategies 2333 are an implementation matter, with a non-normative example given in 2334 Appendix C.1). 2336 The ROS then updates the asymmetric neighbor cache entry to refresh 2337 ReachableTime, while (for MNP destinations) the ROR adds or updates 2338 the ROS address to the target Client's Report List and with time set 2339 to ReportTime. While no data packets are flowing, the ROS instead 2340 allows ReachableTime for the asymmetric neighbor cache entry to 2341 expire. When ReachableTime expires, the ROS deletes the asymmetric 2342 neighbor cache entry. Future data packets flowing through the ROS 2343 will again trigger a new route optimization exchange while initial 2344 data packets travel over a suboptimal route. 2346 The ROS may also receive unsolicited NA messages from the ROR at any 2347 time. If there is an asymmetric neighbor cache entry for the target, 2348 the ROS updates the link-layer information but does not update 2349 ReachableTime since the receipt of an unsolicited NA does not confirm 2350 that the forward path is still working. If there is no asymmetric 2351 neighbor cache entry, the route optimization source simply discards 2352 the unsolicited NA. Cases in which unsolicited NA messages are 2353 generated are specified in Section 3.19. 2355 In this arrangement, the ROS holds an asymmetric neighbor cache entry 2356 for the ROR, but the ROR does not hold an asymmetric neighbor cache 2357 entry for the ROS. The route optimization neighbor relationship is 2358 therefore asymmetric and unidirectional. If the target node also has 2359 packets to send back to the source node, then a separate route 2360 optimization procedure is performed in the reverse direction. But, 2361 there is no requirement that the forward and reverse paths be 2362 symmetric. 2364 3.18. Neighbor Unreachability Detection (NUD) 2366 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2367 [RFC4861]. NUD is performed either reactively in response to 2368 persistent link-layer errors (see Section 3.14) or proactively to 2369 confirm reachability. The NUD algorithm may further be seeded by ND 2370 hints of forward progress, but care must be taken to avoid inferring 2371 reachability based on spoofed information. 2373 When an ROR directs an ROS to a neighbor with one or more target 2374 link-layer addresses, the ROS can proactively test each direct path 2375 by sending an initial NS message to elicit a solicited NA response. 2376 While testing the paths, the ROS can optionally continue sending 2377 packets via the SPAN, maintain a small queue of packets until target 2378 reachability is confirmed, or (optimistically) allow packets to flow 2379 via the direct paths. In any case, the ROS should only consider the 2380 neighbor unreachable if NUD fails over multiple target link-layer 2381 address paths. 2383 When a ROS sends an NS message used for NUD, it uses its AERO 2384 addresses as the IPv6 source address and the AERO address 2385 corresponding to a target link-layer address as the destination. For 2386 each target link-layer address, the source node encapsulates the NS 2387 message in SPAN/INET headers with its own SPAN address as the source 2388 and the SPAN address of the target as the destination, If the target 2389 is located within the same SPAN segment, the source sets the INET 2390 address of the target as the destination; otherwise, it sets the INET 2391 address of a Relay as the destination. The source then forwards the 2392 message into the SPAN. 2394 Paths that pass NUD tests are marked as "reachable", while those that 2395 do not are marked as "unreachable". These markings inform the AERO 2396 interface forwarding algorithm specified in Section 3.12. 2398 Proxys can perform NUD to verify Server reachability on behalf of 2399 their proxyed Clients so that the Clients need not engage in NUD 2400 messaging themselves. 2402 3.19. Mobility Management and Quality of Service (QoS) 2404 AERO is a Distributed Mobility Management (DMM) service. Each Server 2405 is responsible for only a subset of the Clients on the AERO link, as 2406 opposed to a Centralized Mobility Management (CMM) service where 2407 there is a single network mobility service for all Clients. Clients 2408 coordinate with their associated Servers via RS/RA exchanges to 2409 maintain the DMM profile, and the AERO routing system tracks all 2410 current Client/Server peering relationships. 2412 Servers provide a Mobility Anchor Point (MAP) for their dependent 2413 Clients. Clients are responsible for maintaining neighbor 2414 relationships with their Servers through periodic RS/RA exchanges, 2415 which also serves to confirm neighbor reachability. When a Client's 2416 underlying interface address and/or QoS information changes, the 2417 Client is responsible for updating the Server with this new 2418 information. Note that for Proxyed interfaces, however, the Proxy 2419 can perform the RS/RA exchanges on the Client's behalf. 2421 Mobility management considerations are specified in the following 2422 sections. 2424 3.19.1. Mobility Update Messaging 2426 Servers acting as MAPs accommodate Client mobility and/or QoS change 2427 events by sending unsolicited NA messages to each ROS in the target 2428 Client's Report List. When a MAP sends an unsolicited NA message, it 2429 sets the IPv6 source address to the Client's AERO address and sets 2430 the IPv6 destination address to all-nodes multicast (ff02::1). The 2431 MAP also includes a TLLAO with a first ifIndex set to 0 and Link 2432 Layer address set to the MAP's SPAN address, and includes additional 2433 ifIndexes for all of the target Client's interfaces with Link Layer 2434 Addresses set to the corresponding SPAN addresses. The MAP finally 2435 encapsulates the message in a SPAN header with source set to its own 2436 SPAN address and destination set to the SPAN address of the ROS, then 2437 sends the message to a Relay in the SPAN. 2439 As for the hot-swap of interface cards discussed in Section 7.2.6 of 2440 [RFC4861], the transmission and reception of unsolicited NA messages 2441 is unreliable but provides a useful optimization. In well-connected 2442 Internetworks with robust data links unsolicited NA messages will be 2443 delivered with high probability, but in any case the MAP can 2444 optionally send up to MAX_NEIGHBOR_ADVERTISEMENT unsolicited NAs to 2445 each ROS to increase the likelihood that at least one will be 2446 received. 2448 When an ROS receives an unsolicited NA message, it ignores the 2449 message if there is no existing neighbor cache entry for the Client. 2450 Otherwise, it uses the included TLLAO to update the Link Layer 2451 Address and QoS information in the neighbor cache entry, but does not 2452 reset ReachableTime since the receipt of an unsolicited NA message 2453 from the target Server does not provide confirmation that any forward 2454 paths to the target Client are working. 2456 If unsolicited NA messages are lost, the ROS may be left with stale 2457 address and/or QoS information for the Client for up to ReachableTime 2458 seconds. During this time, the ROS can continue sending packets to 2459 the target Client according to its current neighbor cache information 2460 but may receive persistent unsolicited NA messages as discussed in 2461 Section 3.19.5.1. 2463 In addition to sending unsolicited NA messages to the current set of 2464 ROSs for the Client, the MAP also sends unsolicited NAs to the former 2465 Link-Layer address for any ifIndex for which the Link-Layer address 2466 has changed. This may be beneficial for ANET links that cannot 2467 easily detect (e.g., without active probing) when a formerly-active 2468 Client has departed. 2470 3.19.2. Announcing Link-Layer Address and/or QoS Preference Changes 2472 When a Client needs to change its ANET addresses and/or QoS 2473 preferences (e.g., due to a mobility event), either the Client or its 2474 Proxys send RS messages to the Server via the SPAN with SLLAOs that 2475 include the new Client Port Number, Link Layer Address and P(i) 2476 values. If the RS messages are sent solely for the purpose of 2477 updating QoS preferences, Port Number and Link-Layer Address are set 2478 to 0. 2480 Up to MAX_RTR_SOLICITATION RS messages MAY be sent in parallel with 2481 sending actual data packets in case one or more RAs are lost. If all 2482 RAs are lost, the Client SHOULD re-associate with a new Server. 2484 When the Server receives the Client's changes, it sends unsolicited 2485 NA messages to all nodes in the Report List the same as described in 2486 the previous section. 2488 3.19.3. Bringing New Links Into Service 2490 When a Client needs to bring new ANET interfaces into service (e.g., 2491 when it activates a new data link), it sends an RS message to its 2492 Server via the ANET interface with SLLAOs that include the new Client 2493 Link Layer Address information. 2495 3.19.4. Removing Existing Links from Service 2497 When a Client needs to remove existing ANET interfaces from service 2498 (e.g., when it de-activates an existing data link), it sends an RS 2499 message to its Server with SLLAO ifIndexes with P(i) values set to 0. 2501 If the Client needs to send RS messages over an ANET interface other 2502 than the one being removed from service, it MUST include an SLLAO 2503 ifIndex for the sending interface as the first ifIndex and include 2504 additional ifIndexes with P(i) values set to 0 for any ANET 2505 interfaces being removed from service. 2507 3.19.5. Moving to a New Server 2509 When a Client associates with a new Server, it performs the Client 2510 procedures specified in Section 3.15.2. The Client then sends an RS 2511 message over any working ANET interface with destination set to the 2512 old Server's AERO address and with an SLLAO with R set to 1 to fully 2513 release itself from the old Server. The SLLAO also includes the SPAN 2514 address of the new Server in the Link Layer Address. If the Client 2515 does not receive an RA reply after MAX_RTR_SOLICITATIONS attempts 2516 over multiple ANET interfaces, the old Server may have failed and the 2517 Client should discontinue its release attempts. 2519 When the old Server processes the RS, it sends unsolicited NA 2520 messages with a TLLAO with ifIndex set to 0 and with R and S set to 1 2521 to all ROSs in the Client's Report List. The Server also changes the 2522 symmetric neighbor cache entry state to DEPARTED, sets the link-layer 2523 address of the Client to the address found in the RS SLLAO (i.e., the 2524 SPAN address of the new Server), and sets a timer to DepartTime 2525 seconds. The old Server then returns an immediate RA message to the 2526 Client with Router Lifetime set to 0. After a short delay (e.g., 2 2527 seconds), the old Server withdraws the Client's MNP from the routing 2528 system. After DepartTime expires, the old Server deletes the 2529 symmetric neighbor cache entry. 2531 Clients SHOULD NOT move rapidly between Servers in order to avoid 2532 causing excessive oscillations in the AERO routing system. Examples 2533 of when a Client might wish to change to a different Server include a 2534 Server that has gone unreachable, topological movements of 2535 significant distance, movement to a new geographic region, movement 2536 to a new SPAN segment, etc. 2538 3.19.5.1. Accommodating Orphaned Fragments 2540 When a Client moves to a new Server, there is a chance that some of 2541 the fragments of a multiple fragment SPAN packet have already arrived 2542 at the old Server, while others are en route to the new Server. The 2543 old Server forwards these orphaned fragments to the new Server over 2544 the SPAN by encapsulating them in a second SPAN header with the old 2545 Server's SPAN address as the source and the SPAN address of the new 2546 Server as the destination. 2548 The new Server will then receive a packet that has an outer SPAN 2549 header inserted by the old Server and an inner SPAN header inserted 2550 by the original ROS. It then discards the outer SPAN header and adds 2551 the enclosed fragment to the reassembly buffer for this SPAN packet. 2552 If the Client has already departed from this new Server, the new 2553 Server instead re-encapsulates the fragment by replacing the outer 2554 SPAN header and forwarding the packet to the next Server the same as 2555 described above. In this way, wayward fragments will follow the 2556 trail of departed Servers until they eventually reach the current 2557 Server. 2559 Note that in some cases reassemblies may be left incomplete after the 2560 Client has departed. Proxys and Servers discard any incomplete 2561 reassemblies when the DEPARTED neighbor cache entry is deleted. 2563 3.20. Multicast 2565 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 2566 [RFC3810] proxy service for its EUNs and/or hosted applications 2567 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 2568 ANET interfaces for which group membership is required. The IGMP/MLD 2569 messages may be further forwarded by a first-hop ANET access router 2570 acting as an IGMP/MLD-snooping switch [RFC4541], then ultimately 2571 delivered to an AERO Proxy/Server acting as a Protocol Independent 2572 Multicast - Sparse-Mode (PIM-SM, or simply "PIM") Designated Router 2573 (DR) [RFC7761]. AERO Gateways also act as PIM routers (i.e., the 2574 same as AERO Proxys/Servers) on behalf of nodes on INET/EUN networks. 2575 The behaviors identified in the following sections correspond to 2576 Source-Specific Multicast (SSM) and Any-Source Multicast (ASM) 2577 operational modes. 2579 3.20.1. Source-Specific Multicast (SSM) 2581 When an ROS (i.e., an AERO Proxy/Server/Gateway) "X" acting as PIM 2582 router receives a Join/Prune message from a node on its downstream 2583 interfaces containing one or more ((S)ource, (G)roup) pairs, it 2584 updates its Multicast Routing Information Base (MRIB) accordingly. 2585 For each S belonging to a prefix reachable via X's non-AERO 2586 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 2587 on those interfaces per [RFC7761]. 2589 For each S belonging to a prefix reachable via X's AERO interface, X 2590 originates a separate copy of the Join/Prune for each (S,G) in the 2591 message using its own AERO address as the source address and ALL-PIM- 2592 ROUTERS as the destination address. X then encapsulates each message 2593 in a SPAN header with source address set to the SPAN address of X and 2594 destination address set to S then forwards the message into the SPAN. 2595 The SPAN in turn forwards the message to AERO Server/Gateway "Y" that 2596 services S. At the same time, if the message was a Join, X sends a 2597 route-optimization NS message toward each S the same as discussed in 2598 Section 3.17. The resulting NAs will return the AERO address for the 2599 prefix that matches S as the network-layer source address and TLLAOs 2600 with the SPAN addresses corresponding to any ifIndexes that are 2601 currently servicing S. 2603 When Y processes the Join/Prune message, if S located behind any 2604 Native, Direct, VPNed or NATed interfaces Y acts as a PIM router and 2605 updates its MRIB to list X as the next hop in the reverse path. If S 2606 is located behind any Proxys "Z"*, Y also forwards the message to 2607 each Z* over the SPAN while continuing to use the AERO address of X 2608 as the source address. Each Z* then updates its MRIB accordingly and 2609 maintains the AERO address of X as the next hop in the reverse path. 2610 Since the Relays in the SPAN do not examine network layer control 2611 messages, this means that the (reverse) multicast tree path is simply 2612 from each Z* (and/or Y) to X with no other multicast-aware routers in 2613 the path. If any Z* (and/or Y) is located on the same SPAN segment 2614 as X, the multicast data traffic sent to X directly using SPAN/INET 2615 encapsulation instead of via a Relay. 2617 Following the initial Join/Prune and NS/NA messaging, X maintains an 2618 asymmetric neighbor cache entry for each S the same as if X was 2619 sending unicast data traffic to S. In particular, X performs 2620 additional NS/NA exchanges to keep the neighbor cache entry alive for 2621 up to t_periodic seconds [RFC7761]. If no new Joins are received 2622 within t_periodic seconds, X allows the neighbor cache entry to 2623 expire. Finally, if X receives any additional Join/Prune messages 2624 for (S,G) it forwards the messages to each Y and Z* in the neighbor 2625 cache entry over the SPAN. 2627 At some later time, Client C that holds an MNP for source S may 2628 depart from a first Proxy Z1 and/or connect via a new Proxy Z2. In 2629 that case, Y sends an unsolicited NA message to X the same as 2630 specified for unicast mobility in Section 3.19. When X receives the 2631 unsolicited NA message, it updates its asymmetric neighbor cache 2632 entry for the AERO address for source S and sends new Join messages 2633 to any new Proxys Z2. There is no requirement to send any Prune 2634 messages to old Proxys Z1 since source S will no longer source any 2635 multicast data traffic via Z1. Instead, the multicast state for 2636 (S,G) in Proxy Z1 will soon time out since no new Joins will arrive. 2638 After some later time, C may move to a new Server Y2 and depart from 2639 old Sever Y1. In that case, Y1 sends Join messages for any of C's 2640 active (S,G) groups to Y2 while including its own AERO address as the 2641 source address. This causes Y2 to include Y1 in the multicast 2642 forwarding tree during the interim time that Y1's symmetric neighbor 2643 cache entry for C is in the DEPARTED state. At the same time, Y1 2644 sends an unsolicited NA message to X with a TLLAO with ifIndex set to 2645 0 and R set to 1 to cause X to release its asymmetric neighbor cache 2646 entry. X then sends a new Join message to S via the SPAN and re- 2647 initiates route optimization the same as if it were receiving a fresh 2648 Join message from a node on a downstream link. 2650 3.20.2. Any-Source Multicast (ASM) 2652 When an ROS X acting as a PIM router receives a Join/Prune from a 2653 node on its downstream interfaces containing one or more (*,G) pairs, 2654 it updates its Multicast Routing Information Base (MRIB) accordingly. 2655 X then forwards a copy of the message to the Rendezvous Point (RP) R 2656 for each G over the SPAN. X uses its own AERO address as the source 2657 address and ALL-PIM-ROUTERS as the destination address, then 2658 encapsulates each message in a SPAN header with source address set to 2659 the SPAN address of X and destination address set to R, then sends 2660 the message into the SPAN. At the same time, if the message was a 2661 Join X initiates NS/NA route optimization the same as for the SSM 2662 case discussed in Section 3.20.1. 2664 For each source S that sends multicast traffic to group G via R, the 2665 Proxy/Server Z* for the Client that aggregates S encapsulates the 2666 packets in PIM Register messages and forwards them to R via the SPAN. 2667 R may then elect to send a PIM Join to Z* over the SPAN. This will 2668 result in an (S,G) tree rooted at Z* with R as the next hop so that R 2669 will begin to receive two copies of the packet; one native copy from 2670 the (S, G) tree and a second copy from the pre-existing (*, G) tree 2671 that still uses PIM Register encapsulation. R can then issue a PIM 2672 Register-stop message to suppress the Register-encapsulated stream. 2673 At some later time, if C moves to a new Proxy/Server Z*, it resumes 2674 sending packets via PIM Register encapsulation via the new Z*. 2676 At the same time, as multicast listeners discover individual S's for 2677 a given G, they can initiate an (S,G) Join for each S under the same 2678 procedures discussed in Section 3.20.1. Once the (S,G) tree is 2679 established, the listeners can send (S, G) Prune messages to R so 2680 that multicast packets for group G sourced by S will only be 2681 delivered via the (S, G) tree and not from the (*, G) tree rooted at 2682 R. All mobility considerations discussed for SSM apply. 2684 3.20.3. Bi-Directional PIM (BIDIR-PIM) 2686 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 2687 approach to ASM that treats the Rendezvous Point (RP) as a Designated 2688 Forwarder (DF). Further considerations for BIDIR-PIM are out of 2689 scope. 2691 3.21. Operation over Multiple AERO Links (VLANs) 2693 An AERO Client can connect to multiple AERO links the same as for any 2694 data link service. In that case, the Client maintains a distinct 2695 AERO interface for each link, e.g., 'aero0' for the first link, 2696 'aero1' for the second, 'aero2' for the third, etc. Each AERO link 2697 would include its own distinct set of Relays, Servers and Proxys, 2698 thereby providing redundancy in case of failures. 2700 The Relays, Servers and Proxys on each AERO link can assign AERO and 2701 SPAN addresses that use the same or different numberings from those 2702 on other links. Since the links are mutually independent there is no 2703 requirement for avoiding inter-link address duplication, e.g., the 2704 same AERO address such as fe80::1000 could be used to number distinct 2705 nodes that connect to different links. 2707 Each AERO link could utilize the same or different ANET connections. 2708 The links can be distinguished at the link-layer via Virtual Local 2709 Area Network (VLAN) tagging (e.g., IEEE 802.1Q) and/or through 2710 assignment of distinct sets of MSPs on each link. This gives rise to 2711 the opportunity for supporting multiple redundant networked paths, 2712 where each VLAN is distinguished by a different label (e.g., colors 2713 such as Red, Green, Blue, etc.). In particular, the Client can tag 2714 its RS messages with the appropriate label to cause the network to 2715 select the desired VLAN. 2717 Clients that connect to multiple AERO interfaces can select the 2718 outgoing interface appropriate for a given Red/Blue/Green/etc. 2719 traffic profile while (in the reverse direction) correspondent nodes 2720 must have some way of steering their packets destined to a target via 2721 the correct AERO link. 2723 In a first alternative, if each AERO link services different MSPs, 2724 then the Client can receive a distinct MNP from each of the links. 2725 IP routing will therefore assure that the correct Red/Green/Blue/etc. 2726 network is used for both outbound and inbound traffic. This can be 2727 accomplished using existing technologies and approaches, and without 2728 requiring any special supporting code in correspondent nodes or 2729 Relays. 2731 In a second alternative, if each AERO link services the same MSP(s) 2732 then each link could assign a distinct "AERO Link Anycast" address 2733 that is configured by all Relays on the link. Correspondent nodes 2734 then include a "type 4" routing header with the Anycast address for 2735 the AERO link as the IPv6 destination and with the address of the 2736 target encoded as the "next segment" in the routing header 2737 [RFC8402][I-D.ietf-6man-segment-routing-header]. Standard IP routing 2738 will then direct the packet to the nearest Relay for the correct AERO 2739 link, which will replace the destination address with the target 2740 address then forward the packet to the target. 2742 3.22. DNS Considerations 2744 AERO Client MNs and INET correspondent nodes consult the Domain Name 2745 System (DNS) the same as for any Internetworking node. When 2746 correspondent nodes and Client MNs use different IP protocol versions 2747 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 2748 A records for IPv4 address mappings to MNs which must then be 2749 populated in Gateway NAT64 mapping caches. In that way, an IPv4 2750 correspondent node can send packets to the IPv4 address mapping of 2751 the target MN, and the Gateway will translate the IPv4 header and 2752 destination address into an IPv6 header and IPv6 destination address 2753 of the MN. 2755 When an AERO Client registers with an AERO Server, the Server returns 2756 the address(es) of DNS servers in RDNSS options [RFC6106]. The DNS 2757 server provides the IP addresses of other MNs and correspondent nodes 2758 in AAAA records for IPv6 or A records for IPv4. 2760 3.23. Transition Considerations 2762 The SPAN ensures that dissimilar INET partitions can be joined into a 2763 single unified AERO link, even though the partitions themselves may 2764 have differing protocol versions and/or incompatible addressing 2765 plans. However, a commonality can be achieved by incrementally 2766 distributing globally routable (i.e., native) IP prefixes to 2767 eventually reach all nodes (both mobile and fixed) in all SPAN 2768 segments. This can be accomplished by incrementally deploying AERO 2769 Gateways on each INET partition, with each Gateway distributing its 2770 MNPs and/or discovering non-MNP prefixes on its INET links. 2772 This gives rise to the opportunity to eventually distribute native IP 2773 addresses to all nodes, and to present a unified AERO link view 2774 (bridged by the SPAN) even if the INET partitions remain in their 2775 current protocol and addressing plans. In that way, the AERO link 2776 can serve the dual purpose of providing a mobility service and a 2777 transition service. Or, if an INET partition is transitioned to a 2778 native IP protocol version and addressing scheme that is compatible 2779 with the AERO link MNP-based addressing scheme, the partition and 2780 AERO link can be joined by Gateways. 2782 Gateways that connect INETs/EUNs with dissimilar IP protocol versions 2783 must employ a network address and protocol translation function such 2784 as NAT64[RFC6146]. 2786 4. Implementation Status 2788 An AERO implementation based on OpenVPN (https://openvpn.net/) was 2789 announced on the v6ops mailing list on January 10, 2018 and an 2790 initial public release of the AERO proof-of-concept source code was 2791 announced on the intarea mailing list on August 21, 2015. The latest 2792 versions are available at: http://linkupnetworks.net/aero. 2794 5. IANA Considerations 2796 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 2797 AERO in the "enterprise-numbers" registry. 2799 The IANA has assigned the UDP port number "8060" for an earlier 2800 experimental version of AERO [RFC6706]. This document obsoletes 2801 [RFC6706] and claims the UDP port number "8060" for all future use. 2803 No further IANA actions are required. 2805 6. Security Considerations 2807 AERO Relays configure secured tunnels with AERO Servers and Proxys 2808 within their local SPAN segments. Applicable secured tunnel 2809 alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 2810 [RFC6347], etc. The AERO Relays of all SPAN segments in turn 2811 configure secured tunnels for their neighboring AERO Relays across 2812 the SPAN. Therefore, packets that traverse the SPAN between any pair 2813 of AERO link neighbors are already secured. 2815 AERO Servers, Gateways and Proxys targeted by a route optimization 2816 may also receive packets directly from the INET partitions instead of 2817 via the SPAN. For INET partitions that apply effective ingress 2818 filtering to defeat source address spoofing, the simple data origin 2819 authentication procedures in Section 3.11 can be applied. This 2820 implies that the ROS list must be maintained consistently by all 2821 route optimization targets within the same INET partition, and that 2822 the ROS list must be securely managed by the partition's 2823 administrative authority. 2825 For INET partitions that cannot apply effective ingress filtering, 2826 the two options for securing communications include 1) disable route 2827 optimization so that all traffic is conveyed over secured tunnels via 2828 the SPAN, or 2) enable on-demand secure tunnel creation between INET 2829 partition neighbors. Option 1) would result in longer routes than 2830 necessary and traffic concentration on critical infrastructure 2831 elements. Option 2) could be coordinated by establishing a secured 2832 tunnel on-demand instead of performing an NS/NA exchange in the route 2833 optimization procedures. Procedures for establishing on-demand 2834 secured tunnels are out of scope. 2836 AERO Clients that connect to secured enclaves need not apply security 2837 to their ND messages, since the messages will be intercepted by a 2838 perimeter Proxy that applies security on its outward-facing 2839 interface. AERO Clients located outside of secured enclaves SHOULD 2840 use symmetric network and/or transport layer security services, but 2841 when there are many prospective neighbors with dynamically changing 2842 connectivity an asymmetric security service such as SEND may be 2843 needed (see: Appendix C.6). 2845 Application endpoints SHOULD use application-layer security services 2846 such as TLS/SSL, DTLS or SSH [RFC4251] to assure the same level of 2847 protection as for critical secured Internet services. AERO Clients 2848 that require host-based VPN services SHOULD use symmetric network 2849 and/or transport layer security services such as IPsec, TLS/SSL, 2850 DTLS, etc. AERO Proxys and Servers can also provide a network-based 2851 VPN service on behalf of the Client, e.g., if the Client is located 2852 within a secured enclave and cannot establish a VPN on its own 2853 behalf. 2855 AERO Servers and Relays present targets for traffic amplification 2856 Denial of Service (DoS) attacks. This concern is no different than 2857 for widely-deployed VPN security gateways in the Internet, where 2858 attackers could send spoofed packets to the gateways at high data 2859 rates. This can be mitigated by connecting Servers and Relays over 2860 dedicated links with no connections to the Internet and/or when 2861 connections to the Internet are only permitted through well-managed 2862 firewalls. Traffic amplification DoS attacks can also target an AERO 2863 Client's low data rate links. This is a concern not only for Clients 2864 located on the open Internet but also for Clients in secured 2865 enclaves. AERO Servers and Proxys can institute rate limits that 2866 protect Clients from receiving packet floods that could DoS low data 2867 rate links. 2869 AERO Gateways must implement ingress filtering to avoid a spoofing 2870 attack in which spurious SPAN messages are injected into an AERO link 2871 from an outside attacker. AERO Clients MUST ensure that their 2872 connectivity is not used by unauthorized nodes on their EUNs to gain 2873 access to a protected network, i.e., AERO Clients that act as routers 2874 MUST NOT provide routing services for unauthorized nodes. (This 2875 concern is no different than for ordinary hosts that receive an IP 2876 address delegation but then "share" the address with other nodes via 2877 some form of Internet connection sharing such as tethering.) 2879 The MAP list and ROS lists MUST be well-managed and secured from 2880 unauthorized tampering, even though the list contains only public 2881 information. The MAP list can be conveyed to the Client, e.g., 2882 through secure upload of a static file, through DNS lookups, etc. 2883 The ROS list can be conveyed to Servers and Proxys through 2884 administrative action, secured file distribution, etc. 2886 Although public domain and commercial SEND implementations exist, 2887 concerns regarding the strength of the cryptographic hash algorithm 2888 have been documented [RFC6273] [RFC4982]. 2890 Security considerations for accepting link-layer ICMP messages and 2891 reflected packets are discussed throughout the document. 2893 7. Acknowledgements 2895 Discussions in the IETF, aviation standards communities and private 2896 exchanges helped shape some of the concepts in this work. 2897 Individuals who contributed insights include Mikael Abrahamsson, Mark 2898 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 2899 Wojciech Dec, Ralph Droms, Adrian Farrel, Nick Green, Sri Gundavelli, 2900 Brian Haberman, Bernhard Haindl, Joel Halpern, Tom Herbert, Sascha 2901 Hlusiak, Lee Howard, Andre Kostur, Hubert Kuenig, Ted Lemon, Andy 2902 Malis, Satoru Matsushima, Tomek Mrugalski, Madhu Niraula, Alexandru 2903 Petrescu, Behcet Saikaya, Michal Skorepa, Joe Touch, Bernie Volz, 2904 Ryuji Wakikawa, Tony Whyman, Lloyd Wood and James Woodyatt. Members 2905 of the IESG also provided valuable input during their review process 2906 that greatly improved the document. Special thanks go to Stewart 2907 Bryant, Joel Halpern and Brian Haberman for their shepherding 2908 guidance during the publication of the AERO first edition. 2910 This work has further been encouraged and supported by Boeing 2911 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 2912 Brodie, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, Claudiu 2913 Danilov, Wen Fang, Anthony Gregory, Jeff Holland, Seth Jahne, Ed 2914 King, Gene MacLean III, Rob Muszkiewicz, Sean O'Sullivan, Greg 2915 Saccone, Kent Shuey, Brian Skeen, Mike Slane, Carrie Spiker, Brendan 2916 Williams, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 2917 BR&T and BIT mobile networking teams. Kyle Bae, Wayne Benson and 2918 Eric Yeh are especially acknowledged for implementing the AERO 2919 functions as extensions to the public domain OpenVPN distribution. 2921 Earlier works on NBMA tunneling approaches are found in 2922 [RFC2529][RFC5214][RFC5569]. 2924 Many of the constructs presented in this second edition of AERO are 2925 based on the author's earlier works, including: 2927 o The Internet Routing Overlay Network (IRON) 2928 [RFC6179][I-D.templin-ironbis] 2930 o Virtual Enterprise Traversal (VET) 2931 [RFC5558][I-D.templin-intarea-vet] 2933 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 2934 [RFC5320][I-D.templin-intarea-seal] 2936 o AERO, First Edition [RFC6706] 2938 Note that these works cite numerous earlier efforts that are not also 2939 cited here due to space limitations. The authors of those earlier 2940 works are acknowledged for their insights. 2942 This work is aligned with the NASA Safe Autonomous Systems Operation 2943 (SASO) program under NASA contract number NNA16BD84C. 2945 This work is aligned with the FAA as per the SE2025 contract number 2946 DTFAWA-15-D-00030. 2948 This work is aligned with the Boeing Information Technology (BIT) 2949 MobileNet program. 2951 This work is aligned with the Boeing autonomy program. 2953 8. References 2955 8.1. Normative References 2957 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 2958 DOI 10.17487/RFC0791, September 1981, 2959 . 2961 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 2962 RFC 792, DOI 10.17487/RFC0792, September 1981, 2963 . 2965 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2966 Requirement Levels", BCP 14, RFC 2119, 2967 DOI 10.17487/RFC2119, March 1997, 2968 . 2970 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 2971 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 2972 December 1998, . 2974 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 2975 "Definition of the Differentiated Services Field (DS 2976 Field) in the IPv4 and IPv6 Headers", RFC 2474, 2977 DOI 10.17487/RFC2474, December 1998, 2978 . 2980 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 2981 "SEcure Neighbor Discovery (SEND)", RFC 3971, 2982 DOI 10.17487/RFC3971, March 2005, 2983 . 2985 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 2986 RFC 3972, DOI 10.17487/RFC3972, March 2005, 2987 . 2989 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 2990 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 2991 November 2005, . 2993 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 2994 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 2995 DOI 10.17487/RFC4861, September 2007, 2996 . 2998 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 2999 Address Autoconfiguration", RFC 4862, 3000 DOI 10.17487/RFC4862, September 2007, 3001 . 3003 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3004 (IPv6) Specification", STD 86, RFC 8200, 3005 DOI 10.17487/RFC8200, July 2017, 3006 . 3008 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3009 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3010 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3011 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3012 . 3014 8.2. Informative References 3016 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3017 2016. 3019 [I-D.ietf-6man-segment-routing-header] 3020 Filsfils, C., Dukes, D., Previdi, S., Leddy, J., 3021 Matsushima, S., and d. daniel.voyer@bell.ca, "IPv6 Segment 3022 Routing Header (SRH)", draft-ietf-6man-segment-routing- 3023 header-22 (work in progress), August 2019. 3025 [I-D.ietf-dmm-distributed-mobility-anchoring] 3026 Chan, A., Wei, X., Lee, J., Jeon, S., and C. Bernardos, 3027 "Distributed Mobility Anchoring", draft-ietf-dmm- 3028 distributed-mobility-anchoring-13 (work in progress), 3029 March 2019. 3031 [I-D.ietf-intarea-gue] 3032 Herbert, T., Yong, L., and O. Zia, "Generic UDP 3033 Encapsulation", draft-ietf-intarea-gue-07 (work in 3034 progress), March 2019. 3036 [I-D.ietf-intarea-gue-extensions] 3037 Herbert, T., Yong, L., and F. Templin, "Extensions for 3038 Generic UDP Encapsulation", draft-ietf-intarea-gue- 3039 extensions-06 (work in progress), March 2019. 3041 [I-D.ietf-intarea-tunnels] 3042 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3043 Architecture", draft-ietf-intarea-tunnels-09 (work in 3044 progress), July 2018. 3046 [I-D.ietf-rtgwg-atn-bgp] 3047 Templin, F., Saccone, G., Dawra, G., Lindem, A., and V. 3048 Moreno, "A Simple BGP-based Mobile Routing System for the 3049 Aeronautical Telecommunications Network", draft-ietf- 3050 rtgwg-atn-bgp-02 (work in progress), May 2019. 3052 [I-D.templin-6man-dhcpv6-ndopt] 3053 Templin, F., "A Unified Stateful/Stateless Configuration 3054 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-08 3055 (work in progress), June 2019. 3057 [I-D.templin-intarea-grefrag] 3058 Templin, F., "GRE Tunnel Level Fragmentation", draft- 3059 templin-intarea-grefrag-04 (work in progress), July 2016. 3061 [I-D.templin-intarea-seal] 3062 Templin, F., "The Subnetwork Encapsulation and Adaptation 3063 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 3064 progress), January 2014. 3066 [I-D.templin-intarea-vet] 3067 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 3068 templin-intarea-vet-40 (work in progress), May 2013. 3070 [I-D.templin-ironbis] 3071 Templin, F., "The Interior Routing Overlay Network 3072 (IRON)", draft-templin-ironbis-16 (work in progress), 3073 March 2014. 3075 [I-D.templin-v6ops-pdhost] 3076 Templin, F., "IPv6 Prefix Delegation and Multi-Addressing 3077 Models", draft-templin-v6ops-pdhost-24 (work in progress), 3078 June 2019. 3080 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3082 [RFC1035] Mockapetris, P., "Domain names - implementation and 3083 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3084 November 1987, . 3086 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 3087 Communication Layers", STD 3, RFC 1122, 3088 DOI 10.17487/RFC1122, October 1989, 3089 . 3091 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 3092 DOI 10.17487/RFC1191, November 1990, 3093 . 3095 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3096 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3097 . 3099 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3100 DOI 10.17487/RFC2003, October 1996, 3101 . 3103 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3104 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3105 . 3107 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3108 Domains without Explicit Tunnels", RFC 2529, 3109 DOI 10.17487/RFC2529, March 1999, 3110 . 3112 [RFC2764] Gleeson, B., Lin, A., Heinanen, J., Armitage, G., and A. 3113 Malis, "A Framework for IP Based Virtual Private 3114 Networks", RFC 2764, DOI 10.17487/RFC2764, February 2000, 3115 . 3117 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 3118 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 3119 DOI 10.17487/RFC2784, March 2000, 3120 . 3122 [RFC2890] Dommety, G., "Key and Sequence Number Extensions to GRE", 3123 RFC 2890, DOI 10.17487/RFC2890, September 2000, 3124 . 3126 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", 3127 RFC 2923, DOI 10.17487/RFC2923, September 2000, 3128 . 3130 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3131 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3132 . 3134 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3135 of Explicit Congestion Notification (ECN) to IP", 3136 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3137 . 3139 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3140 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3141 DOI 10.17487/RFC3810, June 2004, 3142 . 3144 [RFC3819] Karn, P., Ed., Bormann, C., Fairhurst, G., Grossman, D., 3145 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 3146 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 3147 RFC 3819, DOI 10.17487/RFC3819, July 2004, 3148 . 3150 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 3151 for IPv6 Hosts and Routers", RFC 4213, 3152 DOI 10.17487/RFC4213, October 2005, 3153 . 3155 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3156 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3157 January 2006, . 3159 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3160 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3161 DOI 10.17487/RFC4271, January 2006, 3162 . 3164 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3165 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3166 2006, . 3168 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3169 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3170 December 2005, . 3172 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3173 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3174 2006, . 3176 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3177 Control Message Protocol (ICMPv6) for the Internet 3178 Protocol Version 6 (IPv6) Specification", STD 89, 3179 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3180 . 3182 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3183 Protocol (LDAP): The Protocol", RFC 4511, 3184 DOI 10.17487/RFC4511, June 2006, 3185 . 3187 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3188 "Considerations for Internet Group Management Protocol 3189 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3190 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3191 . 3193 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3194 "Internet Group Management Protocol (IGMP) / Multicast 3195 Listener Discovery (MLD)-Based Multicast Forwarding 3196 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3197 August 2006, . 3199 [RFC4607] Holbrook, H. and B. Cain, "Source-Specific Multicast for 3200 IP", RFC 4607, DOI 10.17487/RFC4607, August 2006, 3201 . 3203 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 3204 Errors at High Data Rates", RFC 4963, 3205 DOI 10.17487/RFC4963, July 2007, 3206 . 3208 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3209 Algorithms in Cryptographically Generated Addresses 3210 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3211 . 3213 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3214 "Bidirectional Protocol Independent Multicast (BIDIR- 3215 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3216 . 3218 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3219 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3220 DOI 10.17487/RFC5214, March 2008, 3221 . 3223 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3224 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3225 February 2010, . 3227 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3228 Route Optimization Requirements for Operational Use in 3229 Aeronautics and Space Exploration Mobile Networks", 3230 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3231 . 3233 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3234 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3235 . 3237 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3238 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3239 January 2010, . 3241 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3242 "IPv6 Router Advertisement Options for DNS Configuration", 3243 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3244 . 3246 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3247 NAT64: Network Address and Protocol Translation from IPv6 3248 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3249 April 2011, . 3251 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3252 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3253 . 3255 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3256 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3257 DOI 10.17487/RFC6221, May 2011, 3258 . 3260 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3261 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3262 DOI 10.17487/RFC6273, June 2011, 3263 . 3265 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3266 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3267 January 2012, . 3269 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3270 for Equal Cost Multipath Routing and Link Aggregation in 3271 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3272 . 3274 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3275 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3276 . 3278 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID Field", 3279 RFC 6864, DOI 10.17487/RFC6864, February 2013, 3280 . 3282 [RFC7269] Chen, G., Cao, Z., Xie, C., and D. Binet, "NAT64 3283 Deployment Options and Experience", RFC 7269, 3284 DOI 10.17487/RFC7269, June 2014, 3285 . 3287 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3288 Korhonen, "Requirements for Distributed Mobility 3289 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3290 . 3292 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3293 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3294 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3295 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3296 2016, . 3298 [RFC8086] Yong, L., Ed., Crabbe, E., Xu, X., and T. Herbert, "GRE- 3299 in-UDP Encapsulation", RFC 8086, DOI 10.17487/RFC8086, 3300 March 2017, . 3302 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 3303 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 3304 DOI 10.17487/RFC8201, July 2017, 3305 . 3307 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3308 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3309 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3310 July 2018, . 3312 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3313 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3314 . 3316 Appendix A. AERO Alternate Encapsulations 3318 When GUE encapsulation is not needed, AERO can use common 3319 encapsulations such as IP-in-IP [RFC2003][RFC2473][RFC4213], Generic 3320 Routing Encapsulation (GRE) [RFC2784][RFC2890] and others. The 3321 encapsulation is therefore only differentiated from non-AERO tunnels 3322 through the application of AERO control messaging and not through, 3323 e.g., a well-known UDP port number. 3325 As for GUE encapsulation, alternate AERO encapsulation formats may 3326 require encapsulation layer fragmentation. For simple IP-in-IP 3327 encapsulation, an IPv6 fragment header is inserted directly between 3328 the inner and outer IP headers when needed, i.e., even if the outer 3329 header is IPv4. The IPv6 Fragment Header is identified to the outer 3330 IP layer by its IP protocol number, and the Next Header field in the 3331 IPv6 Fragment Header identifies the inner IP header version. For GRE 3332 encapsulation, a GRE fragment header is inserted within the GRE 3333 header [I-D.templin-intarea-grefrag]. 3335 Figure 6 shows the AERO IP-in-IP encapsulation format before any 3336 fragmentation is applied: 3338 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3339 | Outer IPv4 Header | | Outer IPv6 Header | 3340 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3341 |IPv6 Frag Header (optional)| |IPv6 Frag Header (optional)| 3342 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3343 | Inner IP Header | | Inner IP Header | 3344 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3345 | | | | 3346 ~ ~ ~ ~ 3347 ~ Inner Packet Body ~ ~ Inner Packet Body ~ 3348 ~ ~ ~ ~ 3349 | | | | 3350 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3352 Minimal Encapsulation in IPv4 Minimal Encapsulation in IPv6 3354 Figure 6: Minimal Encapsulation Format using IP-in-IP 3356 Figure 7 shows the AERO GRE encapsulation format before any 3357 fragmentation is applied: 3359 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3360 | Outer IP Header | 3361 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3362 | GRE Header | 3363 | (with checksum, key, etc..) | 3364 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3365 | GRE Fragment Header (optional)| 3366 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3367 | Inner IP Header | 3368 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3369 | | 3370 ~ ~ 3371 ~ Inner Packet Body ~ 3372 ~ ~ 3373 | | 3374 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3376 Figure 7: Minimal Encapsulation Using GRE 3378 Alternate encapsulation may be preferred in environments where GUE 3379 encapsulation would add unnecessary overhead. For example, certain 3380 low-bandwidth wireless data links may benefit from a reduced 3381 encapsulation overhead. 3383 GUE encapsulation can traverse network paths that are inaccessible to 3384 non-UDP encapsulations, e.g., for crossing Network Address 3385 Translators (NATs). More and more, network middleboxes are also 3386 being configured to discard packets that include anything other than 3387 a well-known IP protocol such as UDP and TCP. It may therefore be 3388 necessary to determine the potential for middlebox filtering before 3389 enabling alternate encapsulation in a given environment. 3391 In addition to IP-in-IP, GRE and GUE, AERO can also use security 3392 encapsulations such as IPsec, TLS/SSL, DTLS, etc. In that case, AERO 3393 control messaging and route determination occur before security 3394 encapsulation is applied for outgoing packets and after security 3395 decapsulation is applied for incoming packets. 3397 AERO is especially well suited for use with VPN system encapsulations 3398 such as OpenVPN [OVPN]. 3400 Appendix B. S/TLLAO Extensions for Special-Purpose Links 3402 The AERO S/TLLAO format specified in Section 3.6 includes a Length 3403 value of 5 (i.e., 5 units of 8 octets). However, special-purpose 3404 links may extend the basic format to include additional fields and a 3405 Length value larger than 5. 3407 For example, adaptation of AERO to the Aeronautical 3408 Telecommunications Network with Internet Protocol Services (ATN/IPS) 3409 includes link selection preferences based on transport port numbers 3410 in addition to the existing DSCP-based preferences. ATN/IPS nodes 3411 maintain a map of transport port numbers to 64 possible preference 3412 fields, e.g., TCP port 22 maps to preference field 8, TCP port 443 3413 maps to preference field 20, UDP port 8060 maps to preference field 3414 34, etc. The extended S/TLLAO format for ATN/IPS is shown in 3415 Figure 8, where the Length value is 7 and the 'Q(i)' fields provide 3416 link preferences for the corresponding transport port number. 3418 0 1 2 3 3419 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 3420 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3421 | Type | Length | Prefix Length | Reserved | 3422 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3423 | ifIndex | Port Number | 3424 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3425 | | 3426 + + 3427 | | 3428 + Link-Layer Address + 3429 | | 3430 + + 3431 | | 3432 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3433 |P00|P01|P02|P03|P04|P05|P06|P07|P08|P09|P10|P11|P12|P13|P14|P15| 3434 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3435 |P16|P17|P18|P19|P20|P21|P22|P23|P24|P25|P26|P27|P28|P29|P30|P31| 3436 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3437 |P32|P33|P34|P35|P36|P37|P38|P39|P40|P41|P42|P43|P44|P45|P46|P47| 3438 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3439 |P48|P49|P50|P51|P52|P53|P54|P55|P56|P57|P58|P59|P60|P61|P62|P63| 3440 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3441 |Q00|Q01|Q02|Q03|Q04|Q05|Q06|Q07|Q08|Q09|Q10|Q11|Q12|Q13|Q14|Q15| 3442 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3443 |Q16|Q17|Q18|Q19|Q20|Q21|Q22|Q23|Q24|Q25|Q26|Q27|Q28|Q29|Q30|Q31| 3444 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3445 |Q32|Q33|Q34|Q35|Q36|Q37|Q38|Q39|Q40|Q41|Q42|Q43|Q44|Q45|Q46|Q47| 3446 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3447 |Q48|Q49|Q50|Q51|Q52|Q53|Q54|Q55|Q56|Q57|Q58|Q59|Q60|Q61|Q62|Q63| 3448 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3450 Figure 8: ATN/IPS Extended S/TLLAO Format 3452 Appendix C. Non-Normative Considerations 3454 AERO can be applied to a multitude of Internetworking scenarios, with 3455 each having its own adapations. The following considerations are 3456 provided as non-normative guidance: 3458 C.1. Implementation Strategies for Route Optimization 3460 Route optimization as discussed in Section 3.17 results in the route 3461 optimization source (ROS) creating an asymmetric neighbor cache entry 3462 for the target neighbor. The neighbor cache entry is maintained for 3463 at most REACHABLE_TIME seconds and then deleted unless updated. In 3464 order to refresh the neighbor cache entry lifetime before the 3465 ReachableTime timer expires, the specification requires 3466 implementations to issue a new NS/NA exchange to reset ReachableTime 3467 to REACHABLE_TIME seconds while data packets are still flowing. 3468 However, the decision of when to initiate a new NS/NA exchange and to 3469 perpetuate the process is left as an implementation detail. 3471 One possible strategy may be to monitor the neighbor cache entry 3472 watching for data packets for (REACHABLE_TIME - 5) seconds. If any 3473 data packets have been sent to the neighbor within this timeframe, 3474 then send an NS to receive a new NA. If no data packets have been 3475 sent, wait for 5 additional seconds and send an immediate NS if any 3476 data packets are sent within this "expiration pending" 5 second 3477 window. If no additional data packets are sent within the 5 second 3478 window, delete the neighbor cache entry. 3480 The monitoring of the neighbor data packet traffic therefore becomes 3481 an asymmetric ongoing process during the neighbor cache entry 3482 lifetime. If the neighbor cache entry expires, future data packets 3483 will trigger a new NS/NA exchange while the packets themselves are 3484 delivered over a longer path until route optimization state is re- 3485 established. 3487 C.2. Implicit Mobility Management 3489 AERO interface neighbors MAY provide a configuration option that 3490 allows them to perform implicit mobility management in which no ND 3491 messaging is used. In that case, the Client only transmits packets 3492 over a single interface at a time, and the neighbor always observes 3493 packets arriving from the Client from the same link-layer source 3494 address. 3496 If the Client's ANET interface address changes (either due to a 3497 readdressing of the original interface or switching to a new 3498 interface) the neighbor immediately updates the neighbor cache entry 3499 for the Client and begins accepting and sending packets according to 3500 the Client's new ANET address. This implicit mobility method applies 3501 to use cases such as cellphones with both WiFi and Cellular 3502 interfaces where only one of the interfaces is active at a given 3503 time, and the Client automatically switches over to the backup 3504 interface if the primary interface fails. 3506 C.3. Direct Underlying Interfaces 3508 When a Client's AERO interface is configured over a Direct interface, 3509 the neighbor at the other end of the Direct link can receive packets 3510 without any encapsulation. In that case, the Client sends packets 3511 over the Direct link according to QoS preferences. If the Direct 3512 interface has the highest QoS preference, then the Client's IP 3513 packets are transmitted directly to the peer without going through an 3514 ANET/INET. If other interfaces have higher QoS preferences, then the 3515 Client's IP packets are transmitted via a different interface, which 3516 may result in the inclusion of Proxys, Servers and Relays in the 3517 communications path. Direct interfaces must be tested periodically 3518 for reachability, e.g., via NUD. 3520 C.4. AERO Clients on the Open Internetwork 3522 AERO Clients that connect to the open Internetwork via either a 3523 native or NATed interface can establish a VPN to securely connect to 3524 a Server. Alternatively, the Client can exchange ND messages 3525 directly with other AERO nodes on the same SPAN segment using INET 3526 encapsulation only and without joining the SPAN. In that case, 3527 however, the Client must apply asymmetric security for ND messages to 3528 ensure routing and neighbor cache integrity (see: Section 6). 3530 C.5. Operation on AERO Links with /64 ASPs 3532 IPv6 AERO links typically have MSPs that aggregate many candidate 3533 MNPs of length /64 or shorter. However, in some cases it may be 3534 desirable to use AERO over links that have only a /64 MSP. This can 3535 be accommodated by treating all Clients on the AERO link as simple 3536 hosts that receive /128 prefix delegations. 3538 In that case, the Client sends an RS message to the Server the same 3539 as for ordinary AERO links. The Server responds with an RA message 3540 that includes one or more /128 prefixes (i.e., singleton addresses) 3541 that include the /64 MSP prefix along with an interface identifier 3542 portion to be assigned to the Client. The Client and Server then 3543 configure their AERO addresses based on the interface identifier 3544 portions of the /128s (i.e., the lower 64 bits) and not based on the 3545 /64 prefix (i.e., the upper 64 bits). 3547 For example, if the MSP for the host-only IPv6 AERO link is 3548 2001:db8:1000:2000::/64, each Client will receive one or more /128 3549 IPv6 prefix delegations such as 2001:db8:1000:2000::1/128, 3550 2001:db8:1000:2000::2/128, etc. When the Client receives the prefix 3551 delegations, it assigns the AERO addresses fe80::1, fe80::2, etc. to 3552 the AERO interface, and assigns the global IPv6 addresses (i.e., the 3553 /128s) to either the AERO interface or an internal virtual interface 3554 such as a loopback. In this arrangement, the Client conducts route 3555 optimization in the same sense as discussed in Section 3.17. 3557 This specification has applicability for nodes that act as a Client 3558 on an "upstream" AERO link, but also act as a Server on "downstream" 3559 AERO links. More specifically, if the node acts as a Client to 3560 receive a /64 prefix from the upstream AERO link it can then act as a 3561 Server to provision /128s to Clients on downstream AERO links. 3563 C.6. AERO Adaptations for SEcure Neighbor Discovery (SEND) 3565 SEcure Neighbor Discovery (SEND) [RFC3971] and Cryptographically 3566 Generated Addresses (CGAs) [RFC3972] were designed to secure IPv6 ND 3567 messaging in environments where symmetric network and/or transport- 3568 layer security services are impractical (see: Section 6). AERO nodes 3569 that use SEND/CGA employ the following adaptations. 3571 When a source AERO node prepares a SEND-protected ND message, it uses 3572 a link-local CGA as the IPv6 source address and writes the prefix 3573 embedded in its AERO address (i.e., instead of fe80::/64) in the CGA 3574 parameters Subnet Prefix field. When the neighbor receives the ND 3575 message, it first verifies the message checksum and SEND/CGA 3576 parameters while using the link-local prefix fe80::/64 (i.e., instead 3577 of the value in the Subnet Prefix field) to match against the IPv6 3578 source address of the ND message. 3580 The neighbor then derives the AERO address of the source by using the 3581 value in the Subnet Prefix field as the interface identifier of an 3582 AERO address. For example, if the Subnet Prefix field contains 3583 2001:db8:1:2, the neighbor constructs the AERO address as 3584 fe80::2001:db8:1:2. The neighbor then caches the AERO address in the 3585 neighbor cache entry it creates for the source, and uses the AERO 3586 address as the IPv6 destination address of any ND message replies. 3588 C.7. AERO Critical Infrastructure Considerations 3590 AERO Relays can be either Commercial off-the Shelf (COTS) standard IP 3591 routers or virtual machines in the cloud. Relays must be 3592 provisioned, supported and managed by the INET administrative 3593 authority, and connected to the Relays of other INETs via inter- 3594 domain peerings. Cost for purchasing, configuring and managing 3595 Relays is nominal even for very large AERO links. 3597 AERO Servers can be standard dedicated server platforms, but most 3598 often will be deployed as virtual machines in the cloud. The only 3599 requirements for Servers are that they can run the AERO user-level 3600 code and have at least one network interface connection to the INET. 3601 As with Relays, Servers must be provisioned, supported and managed by 3602 the INET administrative authority. Cost for purchasing, configuring 3603 and managing Servers is nominal especially for virtual Servers hosted 3604 in the cloud. 3606 AERO Proxys are most often standard dedicated server platforms with 3607 one network interface connected to the ANET and a second interface 3608 connected to an INET. As with Servers, the only requirements are 3609 that they can run the AERO user-level code and have at least one 3610 interface connection to the INET. Proxys must be provisioned, 3611 supported and managed by the ANET administrative authority. Cost for 3612 purchasing, configuring and managing Proxys is nominal, and borne by 3613 the ANET administrative authority. 3615 AERO Gateways can be any dedicated server or COTS router platform 3616 connected to INETs and/or EUNs. The Gateway joins the SPAN and 3617 engages in eBGP peering with one or more Relays as a stub AS. The 3618 Gateway then injects its MNPs and/or non-MNP prefixes into the BGP 3619 routing system, and provisions the prefixes to its downstream- 3620 attached networks. The Gateway can perform ROS and MAP services the 3621 same as for any Server, and can route between the MNP and non-MNP 3622 address spaces. 3624 Appendix D. Change Log 3626 << RFC Editor - remove prior to publication >> 3628 Changes from draft-templin-intarea-6706bis-15 to draft-templin- 3629 intrea-6706bis-17: 3631 o S/TLLAO now includes multiple link-layer addresses within a single 3632 option instead of requiring multiple options 3634 o New unsolicited NA message to inform the old link that a Client 3635 has moved to a new link 3637 Changes from draft-templin-intarea-6706bis-14 to draft-templin- 3638 intrea-6706bis-15: 3640 o MTU and fragmentation 3641 o New details in movement to new Server 3643 Changes from draft-templin-intarea-6706bis-13 to draft-templin- 3644 intrea-6706bis-14: 3646 o Security based on secured tunnels, ingress filtering, MAP list and 3647 ROS list 3649 Changes from draft-templin-intarea-6706bis-12 to draft-templin- 3650 intrea-6706bis-13: 3652 o New paragraph in Section 3.6 on AERO interface layering over 3653 secured tunnels 3655 o Removed extraneous text in Section 3.7 3657 o Added new detail to the forwarding algorithm in Section 3.9 3659 o Clarified use of fragmentation 3661 o Route optimization now supported for both MNP and non-MNP-based 3662 prefixes 3664 o Relays are now seen as link-layer elements in the architecture. 3666 o Built out multicast section in detail. 3668 o New Appendix on implementation considerations for route 3669 optimization. 3671 Changes from draft-templin-intarea-6706bis-11 to draft-templin- 3672 intrea-6706bis-12: 3674 o Introduced Gateways as a new AERO element for connecting 3675 Correspondent Nodes on INET links 3677 o Introduced terms "Access Network (ANET)" and "Internetwork (INET)" 3679 o Changed "ASP" to "MSP", and "ACP" to "MNP" 3681 o New figure on the relation of Segments to the SPAN and AERO link 3683 o New "S" bit in S/TLLAO to indicate the "Source" S/TLLAO as opposed 3684 to additional S/TLLAOs 3686 o Changed Interface ID for Servers from 255 to 0xffff 3687 o Significant updates to Route Optimization, NUD, and Mobility 3688 Management 3690 o New Section on Multicast 3692 o New Section on AERO Clients in the open Internetwork 3694 o New Section on Operation over multiple AERO links (VLANs over the 3695 SPAN) 3697 o New Sections on DNS considerations and Transition considerations 3699 o 3701 Changes from draft-templin-intarea-6706bis-10 to draft-templin- 3702 intrea-6706bis-11: 3704 o Added The SPAN 3706 Changes from draft-templin-intarea-6706bis-09 to draft-templin- 3707 intrea-6706bis-10: 3709 o Orphaned packets in flight (e.g., when a neighbor cache entry is 3710 in the DEPARTED state) are now forwarded at the link layer instead 3711 of at the network layer. Forwarding at the network layer can 3712 result in routing loops and/or excessive delays of forwarded 3713 packets while the routing system is still reconverging. 3715 o Update route optimization to clarify the unsecured nature of the 3716 first NS used for route discovery 3718 o Many cleanups and clarifications on ND messaging parameters 3720 Changes from draft-templin-intarea-6706bis-08 to draft-templin- 3721 intrea-6706bis-09: 3723 o Changed PRL to "MAP list" 3725 o For neighbor cache entries, changed "static" to "symmetric", and 3726 "dynamic" to "asymmetric" 3728 o Specified Proxy RS/RA exchanges with Servers on behalf of Clients 3730 o Added discussion of unsolicited NAs in Section 3.16, and included 3731 forward reference to Section 3.18 3733 o Added discussion of AERO Clients used as critical infrastructure 3734 elements to connect fixed networks. 3736 o Added network-based VPN under security considerations 3738 Changes from draft-templin-intarea-6706bis-07 to draft-templin- 3739 intrea-6706bis-08: 3741 o New section on AERO-Aware Access Router 3743 Changes from draft-templin-intarea-6706bis-06 to draft-templin- 3744 intrea-6706bis-07: 3746 o Added "R" bit for release of PDs. Now have a full RS/RA service 3747 that can do PD without requiring DHCPv6 messaging over-the-air 3749 o Clarifications on solicited vs unsolicited NAs 3751 o Clarified use of MAX_NEIGHBOR_ADVERTISEMENTS for the purpose of 3752 increase reliability 3754 Changes from draft-templin-intarea-6706bis-05 to draft-templin- 3755 intrea-6706bis-06: 3757 o Major re-work and simplification of Route Optimization function 3759 o Added Distributed Mobility Management (DMM) and Mobility Anchor 3760 Point (MAP) terminology 3762 o New section on "AERO Critical Infrastructure Element 3763 Considerations" demonstrating low overall cost for the service 3765 o minor text revisions and deletions 3767 o removed extraneous appendices 3769 Changes from draft-templin-intarea-6706bis-04 to draft-templin- 3770 intrea-6706bis-05: 3772 o New Appendix E on S/TLLAO Extensions for special-purpose links. 3773 Discussed ATN/IPS as example. 3775 o New sentence in introduction to declare appendices as non- 3776 normative. 3778 Changes from draft-templin-intarea-6706bis-03 to draft-templin- 3779 intrea-6706bis-04: 3781 o Added definitions for Potential Router List (PRL) and secure 3782 enclave 3784 o Included text on mapping transport layer port numbers to network 3785 layer DSCP values 3787 o Added reference to DTLS and DMM Distributed Mobility Anchoring 3788 working group document 3790 o Reworked Security Considerations 3792 o Updated references. 3794 Changes from draft-templin-intarea-6706bis-02 to draft-templin- 3795 intrea-6706bis-03: 3797 o Added new section on SEND. 3799 o Clarifications on "AERO Address" section. 3801 o Updated references and added new reference for RFC8086. 3803 o Security considerations updates. 3805 o General text clarifications and cleanup. 3807 Changes from draft-templin-intarea-6706bis-01 to draft-templin- 3808 intrea-6706bis-02: 3810 o Note on encapsulation avoidance in Section 4. 3812 Changes from draft-templin-intarea-6706bis-00 to draft-templin- 3813 intrea-6706bis-01: 3815 o Remove DHCPv6 Server Release procedures that leveraged the old way 3816 Relays used to "route" between Server link-local addresses 3818 o Remove all text relating to Relays needing to do any AERO-specific 3819 operations 3821 o Proxy sends RS and receives RA from Server using SEND. Use CGAs 3822 as source addresses, and destination address of RA reply is to the 3823 AERO address corresponding to the Client's ACP. 3825 o Proxy uses SEND to protect RS and authenticate RA (Client does not 3826 use SEND, but rather relies on subnetwork security. When the 3827 Proxy receives an RS from the Client, it creates a new RS using 3828 its own addresses as the source and uses SEND with CGAs to send a 3829 new RS to the Server. 3831 o Emphasize distributed mobility management 3832 o AERO address-based RS injection of ACP into underlying routing 3833 system. 3835 Changes from draft-templin-aerolink-82 to draft-templin-intarea- 3836 6706bis-00: 3838 o Document use of NUD (NS/NA) for reliable link-layer address 3839 updates as an alternative to unreliable unsolicited NA. 3840 Consistent with Section 7.2.6 of RFC4861. 3842 o Server adds additional layer of encapsulation between outer and 3843 inner headers of NS/NA messages for transmission through Relays 3844 that act as vanilla IPv6 routers. The messages include the AERO 3845 Server Subnet Router Anycast address as the source and the Subnet 3846 Router Anycast address corresponding to the Client's ACP as the 3847 destination. 3849 o Clients use Subnet Router Anycast address as the encapsulation 3850 source address when the access network does not provide a 3851 topologically-fixed address. 3853 Author's Address 3855 Fred L. Templin (editor) 3856 Boeing Research & Technology 3857 P.O. Box 3707 3858 Seattle, WA 98124 3859 USA 3861 Email: fltemplin@acm.org