idnits 2.17.1 draft-templin-intarea-6706bis-18.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 (January 1, 2020) is 1574 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: '1' on line 1024 -- Looks like a reference, but probably isn't: '2' on line 1029 -- Looks like a reference, but probably isn't: '3' on line 1035 == Missing Reference: 'N' is mentioned on line 1042, but not defined == Unused Reference: 'RFC2474' is defined on line 2997, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-gue-extensions' is defined on line 3059, but no explicit reference was found in the text == Unused Reference: 'RFC2764' is defined on line 3140, but no explicit reference was found in the text == Unused Reference: 'RFC4607' is defined on line 3227, but no explicit reference was found in the text == Unused Reference: 'RFC7269' is defined on line 3314, but no explicit reference was found in the text == Unused Reference: 'RFC8086' is defined on line 3330, but no explicit reference was found in the text == Outdated reference: A later version (-15) exists of draft-ietf-dmm-distributed-mobility-anchoring-14 == Outdated reference: A later version (-13) exists of draft-ietf-intarea-tunnels-10 == Outdated reference: A later version (-26) exists of draft-ietf-rtgwg-atn-bgp-04 == Outdated reference: A later version (-11) exists of draft-templin-6man-dhcpv6-ndopt-08 == Outdated reference: A later version (-21) exists of draft-templin-atn-aero-interface-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 (~~), 15 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, January 1, 2020 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: July 4, 2020 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-intarea-6706bis-18 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 July 4, 2020. 45 Copyright Notice 47 Copyright (c) 2020 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 . . . . . . . . . . . . . . 24 74 3.7.1. AERO Server/Gateway Behavior . . . . . . . . . . . . 25 75 3.7.2. AERO Proxy Behavior . . . . . . . . . . . . . . . . . 25 76 3.7.3. AERO Client Behavior . . . . . . . . . . . . . . . . 25 77 3.7.4. AERO Relay Behavior . . . . . . . . . . . . . . . . . 26 78 3.8. AERO Interface Neighbor Cache Maintenance . . . . . . . . 26 79 3.9. AERO Interface Encapsulation and Re-encapsulation . . . . 28 80 3.10. AERO Interface Decapsulation . . . . . . . . . . . . . . 29 81 3.11. AERO Interface Data Origin Authentication . . . . . . . . 29 82 3.12. AERO Interface Forwarding Algorithm . . . . . . . . . . . 30 83 3.12.1. Client Forwarding Algorithm . . . . . . . . . . . . 30 84 3.12.2. Proxy Forwarding Algorithm . . . . . . . . . . . . . 31 85 3.12.3. Server/Gateway Forwarding Algorithm . . . . . . . . 32 86 3.12.4. Relay Forwarding Algorithm . . . . . . . . . . . . . 33 87 3.13. AERO Interface MTU and Fragmentation . . . . . . . . . . 34 88 3.13.1. AERO MTU Requirements . . . . . . . . . . . . . . . 36 89 3.14. AERO Interface Error Handling . . . . . . . . . . . . . . 37 90 3.15. AERO Router Discovery, Prefix Delegation and 91 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 40 92 3.15.1. AERO ND/PD Service Model . . . . . . . . . . . . . . 40 93 3.15.2. AERO Client Behavior . . . . . . . . . . . . . . . . 40 94 3.15.3. AERO Server Behavior . . . . . . . . . . . . . . . . 43 95 3.16. The AERO Proxy . . . . . . . . . . . . . . . . . . . . . 45 96 3.16.1. Detecting and Responding to Server Failures . . . . 47 97 3.17. AERO Route Optimization . . . . . . . . . . . . . . . . . 47 98 3.17.1. Route Optimization Initiation . . . . . . . . . . . 48 99 3.17.2. Relaying the NS . . . . . . . . . . . . . . . . . . 48 100 3.17.3. Processing the NS and Sending the NA . . . . . . . . 49 101 3.17.4. Relaying the NA . . . . . . . . . . . . . . . . . . 49 102 3.17.5. Processing the NA . . . . . . . . . . . . . . . . . 50 103 3.17.6. Route Optimization Maintenance . . . . . . . . . . . 50 104 3.18. Neighbor Unreachability Detection (NUD) . . . . . . . . . 51 105 3.19. Mobility Management and Quality of Service (QoS) . . . . 52 106 3.19.1. Mobility Update Messaging . . . . . . . . . . . . . 52 107 3.19.2. Announcing Link-Layer Address and/or QoS Preference 108 Changes . . . . . . . . . . . . . . . . . . . . . . 53 109 3.19.3. Bringing New Links Into Service . . . . . . . . . . 53 110 3.19.4. Removing Existing Links from Service . . . . . . . . 54 111 3.19.5. Moving to a New Server . . . . . . . . . . . . . . . 54 112 3.20. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 55 113 3.20.1. Source-Specific Multicast (SSM) . . . . . . . . . . 55 114 3.20.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 56 115 3.20.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 57 116 3.21. Operation over Multiple AERO Links (VLANs) . . . . . . . 57 117 3.22. DNS Considerations . . . . . . . . . . . . . . . . . . . 58 118 3.23. Transition Considerations . . . . . . . . . . . . . . . . 59 119 3.24. Detecting and Reacting to Proxy, Server and Relay 120 Failures . . . . . . . . . . . . . . . . . . . . . . . . 59 121 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 60 122 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 60 123 6. Security Considerations . . . . . . . . . . . . . . . . . . . 60 124 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 62 125 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 63 126 8.1. Normative References . . . . . . . . . . . . . . . . . . 63 127 8.2. Informative References . . . . . . . . . . . . . . . . . 65 128 Appendix A. P[i] Preference Value Extensions in S/TLLAOs . . . . 71 129 Appendix B. AERO Alternate Encapsulations . . . . . . . . . . . 72 130 Appendix C. Non-Normative Considerations . . . . . . . . . . . . 74 131 C.1. Implementation Strategies for Route Optimization . . . . 75 132 C.2. Implicit Mobility Management . . . . . . . . . . . . . . 75 133 C.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 76 134 C.4. AERO Clients on the Open Internetwork . . . . . . . . . . 76 135 C.5. Operation on AERO Links with /64 ASPs . . . . . . . . . . 76 136 C.6. AERO Adaptations for SEcure Neighbor Discovery (SEND) . . 77 137 C.7. AERO Critical Infrastructure Considerations . . . . . . . 77 138 C.8. AERO Server Failure Implications . . . . . . . . . . . . 78 139 Appendix D. Change Log . . . . . . . . . . . . . . . . . . . . . 79 140 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 84 142 1. Introduction 144 Asymmetric Extended Route Optimization (AERO) fulfills the 145 requirements of Distributed Mobility Management (DMM) [RFC7333] and 146 route optimization [RFC5522] for aeronautical networking and other 147 network mobility use cases. AERO is based on a Non-Broadcast, 148 Multiple Access (NBMA) virtual link model known as the AERO link. 149 The AERO link is a virtual overlay configured over one or more 150 underlying Internetworks, and nodes on the link can exchange IP 151 packets via tunneling. Multilink operation allows for increased 152 reliability, bandwidth optimization and traffic path diversity. 154 The AERO service comprises Clients, Proxys, Servers, and Gateways 155 that are seen as AERO link neighbors. Each node's AERO interface 156 uses an IPv6 link-local address format (known as the AERO address) 157 that supports operation of the IPv6 Neighbor Discovery (ND) protocol 158 [RFC4861] and links ND to IP forwarding. A node's AERO interface can 159 be configured over multiple underlying interfaces, and may therefore 160 may appear as a single interface with multiple link-layer addresses. 161 Each link-layer address is subject to change due to mobility and/or 162 QoS fluctuations, and link-layer address changes are signaled by ND 163 messaging the same as for any IPv6 link. 165 AERO links provide a cloud-based service where mobile nodes may use 166 any Server acting as a Mobility Anchor Point (MAP) and fixed nodes 167 may use any Gateway on the link for efficient communications. Fixed 168 nodes forward packets destined to other AERO nodes to the nearest 169 Gateway, which forwards them through the cloud. A mobile node's 170 initial packets are forwarded through the MAP, while direct routing 171 is supported through asymmetric extended route optimization while 172 data packets are flowing. Both unicast and multicast communications 173 are supported, and mobile nodes may efficiently move between 174 locations while maintaining continuous communications with 175 correspondents and without changing their IP Address. 177 AERO Relays are interconnected in a secured private BGP overlay 178 routing instance known as the "SPAN". The SPAN provides a hybrid 179 routing/bridging service to join the underlying Internetworks of 180 multiple disjoint administrative domains into a single unified AERO 181 link. Each AERO link instance is characterized by the set of 182 Mobility Service Prefixes (MSPs) common to all mobile nodes. The 183 link should extend to the point where a Gateway/MAP is on the optimal 184 route from any correspondent node on the link, and provides a gateway 185 between the underlying Internetwork and the SPAN. To the underlying 186 Internetwork, the Gateway/MAP is the source of a route to its MSP, 187 and hence uplink traffic to the mobile node is naturally routed to 188 the nearest Gateway/MAP. 190 AERO assumes the use of PIM Sparse Mode in support of multicast 191 communication. In support of Source Specific Multicast (SSM) when a 192 Mobile Node is the source, AERO route optimization ensures that a 193 shortest-path multicast tree is established with provisions for 194 mobility and multilink operation. In all other multicast scenarios 195 there are no AERO dependencies. 197 AERO was designed for aeronautical networking for both manned and 198 unmanned aircraft, where the aircraft is treated as a mobile node 199 that can connect an Internet of Things (IoT). AERO is also 200 applicable to a wide variety of other use cases. For example, it can 201 be used to coordinate the Virtual Private Network (VPN) links of 202 mobile nodes (e.g., cellphones, tablets, laptop computers, etc.) that 203 connect into a home enterprise network via public access networks 204 using services such as OpenVPN [OVPN]. Other applicable use cases 205 are also in scope. 207 The following numbered sections present the AERO specification. The 208 appendices at the end of the document are non-normative. 210 2. Terminology 212 The terminology in the normative references applies; the following 213 terms are defined within the scope of this document: 215 IPv6 Neighbor Discovery (ND) 216 an IPv6 control message service for coordinating neighbor 217 relationships between nodes connected to a common link. AERO 218 interfaces use the ND service specified in [RFC4861]. 220 IPv6 Prefix Delegation (PD) 221 a networking service for delegating IPv6 prefixes to nodes on the 222 link. The nominal PD service is DHCPv6 [RFC8415], however 223 alternate services (e.g., based on ND messaging) are also in scope 224 [I-D.templin-v6ops-pdhost][I-D.templin-6man-dhcpv6-ndopt]. Most 225 notably, a form of PD known as "prefix registration" can be used 226 if the Client knows its prefix in advance and can represent it in 227 the IPv6 source address of an ND message. 229 Access Network (ANET) 230 a node's first-hop data link service network, e.g., a radio access 231 network, cellular service provider network, corporate enterprise 232 network, or the public Internet itself. For secured ANETs, link- 233 layer security services such as IEEE 802.1X and physical-layer 234 security prevent unauthorized access internally while border 235 network-layer security services such as firewalls and proxies 236 prevent unauthorized outside access. 238 ANET interface 239 a node's attachment to a link in an ANET. 241 ANET address 242 an IP address assigned to a node's interface connection to an 243 ANET. 245 Internetwork (INET) 246 a connected IP network topology with a coherent routing and 247 addressing plan and that provides a transit backbone service for 248 ANET end systems. INETs also provide an underlay service over 249 which the AERO virtual link is configured. Example INETs include 250 corporate enterprise networks, aviation networks, and the public 251 Internet itself. When there is no administrative boundary between 252 an ANET and the INET, the ANET and INET are one and the same. 254 INET Partition 255 frequently, INETs such as large corporate enterprise networks are 256 sub-divided internally into separate isolated partitions. Each 257 partition is fully connected internally but disconnected from 258 other partitions, and there is no requirement that separate 259 partitions maintain consistent Internet Protocol and/or addressing 260 plans. (An INET partition is the same as a SPAN segment discussed 261 below.) 263 INET interface 264 a node's attachment to a link in an INET. 266 INET address 267 an IP address assigned to a node's interface connection to an 268 INET. 270 AERO link 271 a Non-Broadcast, Multiple Access (NBMA) tunnel virtual overlay 272 configured over one or more underlying INETs. Nodes on the AERO 273 link appear as single-hop neighbors from the perspective of the 274 virtual overlay even though they may be separated by many 275 underlying INET hops. AERO links may be configured over multiple 276 underlying SPAN segments (see below). 278 AERO interface 279 a node's attachment to an AERO link. Since the addresses assigned 280 to an AERO interface are managed for uniqueness, AERO interfaces 281 do not require Duplicate Address Detection (DAD) and therefore set 282 the administrative variable 'DupAddrDetectTransmits' to zero 283 [RFC4862]. 285 underlying interface 286 an ANET or INET interface over which an AERO interface is 287 configured. 289 AERO address 290 an IPv6 link-local address assigned to an AERO interface and 291 constructed as specified in Section 3.4. 293 base AERO address 294 the lowest-numbered AERO address aggregated by the MNP (see 295 Section 3.4). 297 Mobility Service Prefix (MSP) 298 an IP prefix assigned to the AERO link and from which more- 299 specific Mobile Network Prefixes (MNPs) are derived. 301 Mobile Network Prefix (MNP) 302 an IP prefix allocated from an MSP and delegated to an AERO Client 303 or Gateway. 305 AERO node 306 a node that is connected to an AERO link, or that provides 307 services to other nodes on an AERO link. 309 AERO Client ("Client") 310 an AERO node that connects to one or more ANETs and requests MNP 311 PDs from AERO Servers. The Client assigns a Client AERO address 312 to the AERO interface for use in ND exchanges with other AERO 313 nodes and forwards packets to correspondents according to AERO 314 interface neighbor cache state. 316 AERO Server ("Server") 317 an INET node that configures an AERO interface to provide default 318 forwarding services and a Mobility Anchor Point (MAP) for AERO 319 Clients. The Server assigns an administratively-provisioned AERO 320 address to its AERO interface to support the operation of the ND/ 321 PD services, and advertises all of its associated MNPs via BGP 322 peerings with Relays. 324 AERO Gateway ("Gateway") 325 an AERO Server that also provides forwarding services between 326 nodes reached via the AERO link and correspondents on other links. 327 AERO Gateways are provisioned with MNPs (i.e., the same as for an 328 AERO Client) and run a dynamic routing protocol to discover any 329 non-MNP IP routes. In both cases, the Gateway advertises the 330 MSP(s) over INET interfaces, and distributes all of its associated 331 MNPs and non-MNP IP routes via BGP peerings with Relays (i.e., the 332 same as for an AERO Server). 334 AERO Relay ("Relay") 335 a node that provides hybrid routing/bridging services (as well as 336 a security trust anchor) for nodes on an AERO link. As a router, 337 the Relay forwards packets using standard IP forwarding. As a 338 bridge, the Relay forwards packets over the AERO link without 339 decrementing the IPv6 Hop Limit. AERO Relays peer with Servers 340 and other Relays to discover the full set of MNPs for the link as 341 well as any non-MNPs that are reachable via Gateways. 343 AERO Proxy ("Proxy") 344 a node that provides proxying services between Clients in an ANET 345 and Servers in external INETs. The AERO Proxy is a conduit 346 between the ANET and external INETs in the same manner as for 347 common web proxies, and behaves in a similar fashion as for ND 348 proxies [RFC4389]. 350 Spanning Partitioned AERO Networks (SPAN) 351 a means for bridging disjoint INET partitions as segments of a 352 unified AERO link the same as for a bridged campus LAN. The SPAN 353 is a mid-layer IPv6 encapsulation service in the AERO routing 354 system that supports a unified AERO link view for all segments. 355 Each segment in the SPAN is a distinct INET partition. 357 SPAN Service Prefix (SSP) 358 a global or unique local /96 IPv6 prefix assigned to the AERO link 359 to support SPAN services. 361 SPAN Partition Prefix (SPP) 362 a sub-prefix of the SPAN Service Prefix uniquely assigned to a 363 single SPAN segment. 365 SPAN Address 366 a global or unique local IPv6 address taken from a SPAN Partition 367 Prefix and constructed as specified in Section 3.5. SPAN 368 addresses are statelessly derived from AERO addresses, and vice- 369 versa. 371 ingress tunnel endpoint (ITE) 372 an AERO interface endpoint that injects encapsulated packets into 373 an AERO link. 375 egress tunnel endpoint (ETE) 376 an AERO interface endpoint that receives encapsulated packets from 377 an AERO link. 379 link-layer address 380 an IP address used as an encapsulation header source or 381 destination address from the perspective of the AERO interface. 383 When UDP encapsulation is used, the UDP port number is also 384 considered as part of the link-layer address. From the 385 perspective of the AERO interface, the link-layer address is 386 either an INET address for intra-segment encapsulation or a SPAN 387 address for inter-segment encapsulation. 389 network layer address 390 the source or destination address of an encapsulated IP packet 391 presented to the AERO interface. 393 end user network (EUN) 394 an internal virtual or external edge IP network that an AERO 395 Client or Gateway connects to the rest of the network via the AERO 396 interface. The Client/Gateway sees each EUN as a "downstream" 397 network, and sees the AERO interface as the point of attachment to 398 the "upstream" network. 400 Mobile Node (MN) 401 an AERO Client and all of its downstream-attached networks that 402 move together as a single unit, i.e., an end system that connects 403 an Internet of Things. 405 Mobile Router (MR) 406 a MN's on-board router that forwards packets between any 407 downstream-attached networks and the AERO link. 409 Mobility Anchor Point (MAP) 410 an AERO Server that is currently tracking and reporting the 411 mobility events of its associated Mobile Node Clients. 413 Route Optimization Source (ROS) 414 the AERO node nearest the source that initiates route 415 optimization. The ROS may be a Server or Proxy acting on behalf 416 of the source Client. 418 Route Optimization responder (ROR) 419 the AERO node nearest the target destination that responds to 420 route optimization requests. The ROR may be a Server acting as a 421 MAP on behalf of a target MNP Client, or a Gateway for a non-MNP 422 destination. 424 MAP List 425 a geographically and/or topologically referenced list of AERO 426 addresses of all MAPs within the same AERO link. There is a 427 single MAP list for the entire AERO link. 429 ROS List 430 a list of AERO/SPAN-to-INET address mappings of all ROSes within 431 the same SPAN segment. There is a distinct ROS list for each 432 segment. 434 Distributed Mobility Management (DMM) 435 a BGP-based overlay routing service coordinated by Servers and 436 Relays that tracks all MAP-to-Client associations. 438 Throughout the document, the simple terms "Client", "Server", 439 "Relay", "Proxy" and "Gateway" refer to "AERO Client", "AERO Server", 440 "AERO Relay", "AERO Proxy" and "AERO Gateway", respectively. 441 Capitalization is used to distinguish these terms from other common 442 Internetworking uses in which they appear without capitalization. 444 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 445 the names of node variables, messages and protocol constants) is used 446 throughout this document. Also, the term "IP" is used to generically 447 refer to either Internet Protocol version, i.e., IPv4 [RFC0791] or 448 IPv6 [RFC8200]. 450 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 451 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 452 document are to be interpreted as described in [RFC2119]. Lower case 453 uses of these words are not to be interpreted as carrying RFC2119 454 significance. 456 3. Asymmetric Extended Route Optimization (AERO) 458 The following sections specify the operation of IP over Asymmetric 459 Extended Route Optimization (AERO) links: 461 3.1. AERO Link Reference Model 462 +----------------+ 463 | AERO Relay R1 | 464 | Nbr: S1, S2, P1| 465 |(X1->S1; X2->S2)| 466 | MSP M1 | 467 +-+---------+--+-+ 468 +--------------+ | Secured | | +--------------+ 469 |AERO Server S1| | tunnels | | |AERO Server S2| 470 | Nbr: C1, R1 +-----+ | +-----+ Nbr: C2, R1 | 471 | default->R1 | | | default->R1 | 472 | X1->C1 | | | X2->C2 | 473 +-------+------+ | +------+-------+ 474 | AERO Link | | 475 X===+===+===================+==)===============+===+===X 476 | | | | 477 +-----+--------+ +--------+--+-----+ +--------+-----+ 478 |AERO Client C1| | AERO Proxy P1 | |AERO Client C2| 479 | Nbr: S1 | |(Proxy Nbr Cache)| | Nbr: S2 | 480 | default->S1 | +--------+--------+ | default->S2 | 481 | MNP X1 | | | MNP X2 | 482 +------+-------+ .--------+------. +-----+--------+ 483 | (- Proxyed Clients -) | 484 .-. `---------------' .-. 485 ,-( _)-. ,-( _)-. 486 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 487 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 488 `-(______)-' +-------+ +-------+ `-(______)-' 490 Figure 1: AERO Link Reference Model 492 Figure 1 presents the AERO link reference model. In this model: 494 o the AERO link is an overlay network service configured over one or 495 more underlying INET partitions which may be managed by different 496 administrative authorities and have incompatible protocols and/or 497 addressing plans. 499 o AERO Relay R1 aggregates Mobility Service Prefix (MSP) M1, 500 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 501 via BGP peerings over secured tunnels to Servers (S1, S2). Relays 502 use the SPAN service to bridge disjoint segments of a partitioned 503 AERO link. 505 o AERO Servers S1 and S2 configure secured tunnels with Relay R1 and 506 also act as Mobility Anchor Points (MAPs) and default routers for 507 their associated Clients C1 and C2. 509 o AERO Clients C1 and C2 associate with Servers S1 and S2, 510 respectively. They receive Mobile Network Prefix (MNP) 511 delegations X1 and X2, and also act as default routers for their 512 associated physical or internal virtual EUNs. Simple hosts H1 and 513 H2 attach to the EUNs served by Clients C1 and C2, respectively. 515 o AERO Proxy P1 configures a secured tunnel with Relay R1 and 516 provides proxy services for AERO Clients in secured enclaves that 517 cannot associate directly with other AERO link neighbors. 519 Each node on the AERO link maintains an AERO interface neighbor cache 520 and an IP forwarding table the same as for any link. Although the 521 figure shows a limited deployment, in common operational practice 522 there will normally be many additional Relays, Servers, Clients and 523 Proxys. 525 3.2. AERO Node Types 527 AERO Relays provide hybrid routing/bridging services (as well as a 528 security trust anchor) for nodes on an AERO link. Relays use 529 standard IPv6 routing to forward packets both within the same INET 530 partitions and between disjoint INET partitions based on a mid-layer 531 IPv6 encapsulation known as the SPAN header. The inner IP layer 532 experiences a virtual bridging service since the inner IP TTL/Hop 533 Limit is not decremented during forwarding. Each Relay also peers 534 with Servers and other Relays in a dynamic routing protocol instance 535 to provide a Distributed Mobility Management (DMM) service for the 536 list of active MNPs (see Section 3.3). Relays present the AERO link 537 as a set of one or more Mobility Service Prefixes (MSPs) but as link- 538 layer devices need not connect directly to the AERO link themselves 539 unless an administrative interface is desired. Relays configure 540 secured tunnels with Servers, Proxys and other Relays; they further 541 maintain IP forwarding table entries for each Mobile Network Prefix 542 (MNP) and any other reachable non-MNP prefixes. 544 AERO Servers provide default forwarding services and a Mobility 545 Anchor Point (MAP) for AERO Client Mobile Nodes (MNs). Each Server 546 also peers with Relays in a dynamic routing protocol instance to 547 advertise its list of associated MNPs (see Section 3.3). Servers 548 facilitate PD exchanges with Clients, where each delegated prefix 549 becomes an MNP taken from an MSP. Servers forward packets between 550 AERO interface neighbors and track each Client's mobility profiles. 552 AERO Clients register their MNPs through PD exchanges with AERO 553 Servers over the AERO link, and distribute the MNPs to nodes on EUNs. 554 A Client may also be co-resident on the same physical or virtual 555 platform as a Server; in that case, the Client and Server behave as a 556 single functional unit and without the need for any Client/Server 557 control messaging. 559 AERO Proxys provide a conduit for ANET AERO Clients to associate with 560 AERO Servers in external INETs. Client and Servers exchange control 561 plane messages via the Proxy acting as a bridge between the ANET/INET 562 boundary. The Proxy forwards data packets to and from Clients 563 according to forwarding information in the neighbor cache. The Proxy 564 function is specified in Section 3.16. 566 AERO Gateways are Servers that provide forwarding services between 567 the AERO interface and INET/EUN interfaces. Gateways are provisioned 568 with MNPs the same as for an AERO Client, and also run a dynamic 569 routing protocol to discover any non-MNP IP routes. The Gateway 570 advertises the MSP(s) to INETs, and distributes all of its associated 571 MNPs and non-MNP IP routes via BGP peerings with Relays. 573 AERO Relays, Servers, Proxys and Gateways are critical infrastructure 574 elements in fixed (i.e., non-mobile) INET deployments and hence have 575 permanent and unchanging INET addresses. AERO Clients are MNs that 576 connect via ANET interfaces, i.e., their ANET addresses may change 577 when the Client moves to a new ANET connection. 579 3.3. AERO Routing System 581 The AERO routing system comprises a private instance of the Border 582 Gateway Protocol (BGP) [RFC4271] that is coordinated between Relays 583 and Servers and does not interact with either the public Internet BGP 584 routing system or any underlying INET routing systems. 586 In a reference deployment, each Server is configured as an Autonomous 587 System Border Router (ASBR) for a stub Autonomous System (AS) using 588 an AS Number (ASN) that is unique within the BGP instance, and each 589 Server further uses eBGP to peer with one or more Relays but does not 590 peer with other Servers. Each INET of a multi-segment AERO link must 591 include one or more Relays, which peer with the Servers and Proxys 592 within that INET. All Relays within the same INET are members of the 593 same hub AS using a common ASN, and use iBGP to maintain a consistent 594 view of all active MNPs currently in service. The Relays of 595 different INETs peer with one another using eBGP. 597 Relays advertise the AERO link's MSPs and any non-MNP routes to each 598 of their Servers. This means that any aggregated non-MNPs (including 599 "default") are advertised to all Servers. Each Relay configures a 600 black-hole route for each of its MSPs. By black-holing the MSPs, the 601 Relay will maintain forwarding table entries only for the MNPs that 602 are currently active, and packets destined to all other MNPs will 603 correctly incur Destination Unreachable messages due to the black- 604 hole route. In this way, Servers have only partial topology 605 knowledge (i.e., they know only about the MNPs of their directly 606 associated Clients) and they forward all other packets to Relays 607 which have full topology knowledge. 609 Servers maintain a working set of associated MNPs, and dynamically 610 announce new MNPs and withdraw departed MNPs in eBGP updates to 611 Relays. Servers that are configured as Gateways also redistribute 612 non-MNP routes learned from non-AERO interfaces via their eBGP Relay 613 peerings. 615 Clients are expected to remain associated with their current Servers 616 for extended timeframes, however Servers SHOULD selectively suppress 617 updates for impatient Clients that repeatedly associate and 618 disassociate with them in order to dampen routing churn. Servers 619 that are configured as Gateways advertise the MSPs via INET/EUN 620 interfaces, and forward packets between INET/EUN interfaces and the 621 AERO interface using standard IP forwarding. 623 Scaling properties of the AERO routing system are limited by the 624 number of BGP routes that can be carried by Relays. As of 2015, the 625 global public Internet BGP routing system manages more than 500K 626 routes with linear growth and no signs of router resource exhaustion 627 [BGP]. More recent network emulation studies have also shown that a 628 single Relay can accommodate at least 1M dynamically changing BGP 629 routes even on a lightweight virtual machine, i.e., and without 630 requiring high-end dedicated router hardware. 632 Therefore, assuming each Relay can carry 1M or more routes, this 633 means that at least 1M Clients can be serviced by a single set of 634 Relays. A means of increasing scaling would be to assign a different 635 set of Relays for each set of MSPs. In that case, each Server still 636 peers with one or more Relays, but institutes route filters so that 637 BGP updates are only sent to the specific set of Relays that 638 aggregate the MSP. For example, if the MSP for the AERO link is 639 2001:db8::/32, a first set of Relays could service the MSP 640 2001:db8::/40, a second set of Relays could service 641 2001:db8:0100::/40, a third set could service 2001:db8:0200::/40, 642 etc. 644 Assuming up to 1K sets of Relays, the AERO routing system can then 645 accommodate 1B or more MNPs with no additional overhead (for example, 646 it should be possible to service 1B /64 MNPs taken from a /34 MSP and 647 even more for shorter prefixes). In this way, each set of Relays 648 services a specific set of MSPs that they advertise to the native 649 Internetwork routing system, and each Server configures MSP-specific 650 routes that list the correct set of Relays as next hops. This 651 arrangement also allows for natural incremental deployment, and can 652 support small scale initial deployments followed by dynamic 653 deployment of additional Clients, Servers and Relays without 654 disturbing the already-deployed base. 656 Server and Relays can use the Bidirectional Forwarding Detection 657 (BFD) protocol [RFC5880] to quickly detect link failures that don't 658 result in interface state changes, BGP peer failures, and 659 administrative state changes. BFD is important in environments where 660 rapid response to failures is required for routing reconvergence and, 661 hence, communications continuity. 663 A full discussion of the BGP-based routing system used by AERO is 664 found in [I-D.ietf-rtgwg-atn-bgp]. The system provides for 665 Distributed Mobility Management (DMM) per the distributed mobility 666 anchoring architecture [I-D.ietf-dmm-distributed-mobility-anchoring]. 668 3.3.1. IPv4 Compatibility Routing 670 For IPv6 MNPs, the AERO routing system includes ordinary IPv6 routes. 671 For IPv4 MNPs, the AERO routing system includes IPv6 routes based on 672 an IPv4-embedded IPv6 address format discussed in Section 3.5.1. 674 3.4. AERO Addresses 676 A Client's AERO address is an IPv6 link-local address with an 677 interface identifier based on the Client's delegated MNP. Relay, 678 Server and Proxy AERO addresses are assigned from the range fe80::/96 679 and include an administratively-provisioned value in the lower 32 680 bits. 682 For IPv6, Client AERO addresses begin with the prefix fe80::/64 and 683 include in the interface identifier (i.e., the lower 64 bits) a 684 64-bit prefix taken from one of the Client's IPv6 MNPs. For example, 685 if the AERO Client receives the IPv6 MNP: 687 2001:db8:1000:2000::/56 689 it constructs its corresponding AERO addresses as: 691 fe80::2001:db8:1000:2000 693 fe80::2001:db8:1000:2001 695 fe80::2001:db8:1000:2002 697 ... etc. ... 699 fe80::2001:db8:1000:20ff 701 For IPv4, Client AERO addresses are based on an IPv4-mapped IPv6 702 address [RFC4291] formed from an IPv4 MNP and with a Prefix Length of 703 96 plus the MNP prefix length. For example, for the IPv4 MNP 704 192.0.2.32/28 the IPv4-mapped IPv6 MNP is: 706 0:0:0:0:0:FFFF:192.0.2.16/124 (also written as 707 0:0:0:0:0:FFFF:c000:0210/124) 709 The Client then constructs its AERO addresses with the prefix 710 fe80::/64 and with the lower 64 bits of the IPv4-mapped IPv6 address 711 in the interface identifier as: 713 fe80::FFFF:192.0.2.16 715 fe80::FFFF:192.0.2.17 717 fe80::FFFF:192.0.2.18 719 ... etc. ... 721 fe80:FFFF:192.0.2.31 723 Relay, Server and Proxy AERO addresses are allocated from the range 724 fe80::/96, and MUST be managed for uniqueness. The lower 32 bits of 725 the AERO address includes a unique integer value between 1 and 0xfffe 726 (e.g., fe80::1, fe80::2, fe80::3, etc.) as assigned by the 727 administrative authority for the link. If the link spans multiple 728 SPAN segments, the AERO addresses are assigned to each segment in 1x1 729 correspondence with SPAN addresses (see: Section 3.5). The address 730 fe80:: is the IPv6 link-local Subnet Router Anycast address, and the 731 address fe80::ffff:ffff is reserved as the unspecified AERO address. 733 The lowest-numbered AERO address from a Client's MNP delegation 734 serves as the "base" AERO address (for example, for the MNP 735 2001:db8:1000:2000::/56 the base AERO address is 736 fe80::2001:db8:1000:2000). The Client then assigns the base AERO 737 address to the AERO interface and uses it for the purpose of 738 maintaining the neighbor cache entry. The Server likewise uses the 739 AERO address as its index into the neighbor cache for this Client. 741 If the Client has multiple AERO addresses (i.e., when there are 742 multiple MNPs and/or MNPs with prefix lengths shorter than /64), the 743 Client originates ND messages using the base AERO address as the 744 source address and accepts and responds to ND messages destined to 745 any of its AERO addresses as equivalent to the base AERO address. In 746 this way, the Client maintains a single neighbor cache entry that may 747 be indexed by multiple AERO addresses. 749 The Client's Subnet Router Anycast address can be statelessly 750 determined from its AERO address by simply transposing the AERO 751 address into the upper N bits of the Anycast address followed by 752 128-N bits of zeroes. For example, for the AERO address 753 fe80::2001:db8:1:2 the subnet router anycast address is 754 2001:db8:1:2::. 756 AERO addresses for mobile node Clients embed a MNP as discussed 757 above, while AERO addresses for non-MNP destinations are constructed 758 in exactly the same way. A Client AERO address therefore encodes 759 either an MNP if the prefix is reached via the SPAN or a non-MNP if 760 the prefix is reached via a Gateway. 762 3.5. Spanning Partitioned AERO Networks (SPAN) 764 An AERO link configured over a single INET appears as a single 765 unified link with a consistent underlying network addressing plan. 766 In that case, all nodes on the link can exchange packets via 767 encapsulation with INET addresses, since the underlying INET is 768 connected. In common practice, however, an AERO link may be 769 partitioned into multiple "segments", where each segment is a 770 distinct INET potentially managed under a different administrative 771 authority (e.g., as for worldwide aviation service providers such as 772 ARINC, SITA, Inmarsat, etc.). Individual INETs may themselves be 773 partitioned internally, in which case each internal partition is seen 774 as a separate segment. 776 The addressing plan of each segment is consistent internally but will 777 often bear no relation to the addressing plans of other segments. 778 Each segment is also likely to be separated from others by network 779 security devices (e.g., firewalls, proxies, packet filtering 780 gateways, etc.), and in many cases disjoint segments may not even 781 have any common physical link connections at all. Therefore, nodes 782 can only be assured of exchanging packets directly with 783 correspondents in the same segment, and not with those in other 784 segments. The only means for joining the segments therefore is 785 through inter-domain peerings between AERO Relays. 787 The same as for traditional campus LANs, multiple AERO link segments 788 can be joined into a single unified link via a virtual bridging 789 service termed the "SPAN". The SPAN performs link-layer packet 790 forwarding between segments (i.e., bridging) without decrementing the 791 network-layer TTL/Hop Limit. The SPAN model is depicted in Figure 2: 793 . . . . . . . . . . . . . . . . . . . . . . . 794 . . 795 . .-(::::::::) . 796 . .-(::::::::::::)-. +-+ . 797 . (:::: Segment A :::)--|R|---+ . 798 . `-(::::::::::::)-' +-+ | . 799 . `-(::::::)-' | . 800 . | . 801 . .-(::::::::) | . 802 . .-(::::::::::::)-. +-+ | . 803 . (:::: Segment B :::)--|R|---+ . 804 . `-(::::::::::::)-' +-+ | . 805 . `-(::::::)-' | . 806 . | . 807 . .-(::::::::) | . 808 . .-(::::::::::::)-. +-+ | . 809 . (:::: Segment C :::)--|R|---+ . 810 . `-(::::::::::::)-' +-+ | . 811 . `-(::::::)-' | . 812 . | . 813 . ..(etc).. x . 814 . . 815 . . 816 . <- AERO Link Bridged by the SPAN -> . 817 . . . . . . . . . . . . . .. . . . . . . . . 819 Figure 2: The SPAN 821 To support the SPAN, AERO links require a reserved /64 IPv6 "SPAN 822 Service Prefix (SSP)". Although any routable IPv6 prefix can be 823 used, a Unique Local Address (ULA) prefix (e.g., fd00::/64) [RFC4389] 824 is recommended since border routers are commonly configured to 825 prevent packets with ULAs from being injected into the AERO link by 826 an external IPv6 node and from leaking out of the AERO link to the 827 outside world. 829 Each segment in the SPAN assigns a unique sub-prefix of SSP::/96 830 termed a "SPAN Partition Prefix (SPP)". For example, a first segment 831 could assign fd00::1000/116, a second could assign fd00::2000/116, a 832 third could assign fd00::3000/116, etc. The administrative 833 authorities for each segment must therefore coordinate to assure 834 mutually-exclusive SPP assignments, but internal provisioning of the 835 SPP is an independent local consideration for each administrative 836 authority. 838 A "SPAN address" is an address taken from a SPP and assigned to a 839 Relay, Server or Proxy interface. SPAN addresses are formed by 840 simply replacing the upper portion of an administratively-assigned 841 AERO address with the SPP. For example, if the SPP is 842 fd00::1000/116, the SPAN address formed from the AERO address 843 fe80::1001 is simply fd00::1001. 845 An "INET address" is an address of a node's interface connection to 846 an INET. AERO/SPAN/INET address mappings are maintained as permanent 847 neighbor cache entires as discussed in Section 3.8. 849 AERO Relays serve as bridges to join multiple segments into a unified 850 AERO link over multiple diverse administrative domains. They support 851 the bridging function by first establishing forwarding table entries 852 for their SPPs either via standard BGP routing or static routes. For 853 example, if three Relays ('A', 'B' and 'C') from different segments 854 serviced the SPPs fd00::1000/116, fd00::2000/116 and fd00::3000/116 855 respectively, then the forwarding tables in each Relay are as 856 follows: 858 A: fd00::1000/116->local, fd00::2000/116->B, fd00::3000/116->C 860 B: fd00::1000/116->A, fd00::2000/116->local, fd00::3000/116->C 862 C: fd00::1000/116->A, fd00::2000/116->B, fd00::3000/116->local 864 These forwarding table entries are permanent and never change, since 865 they correspond to fixed infrastructure elements in their respective 866 segments. This provides the basis for a link-layer forwarding 867 service that cannot be disrupted by routing updates due to node 868 mobility. 870 With the SPPs in place in each Relay's forwarding table, control and 871 data packets sent between AERO nodes in different segments can 872 therefore be carried over the SPAN via encapsulation. For example, 873 when a source node in segment A forwards a packet with IPv6 address 874 2001:db8:1:2::1 to a destination node in segment C with IPv6 address 875 2001:db8:1000:2000::1, it first encapsulates the packet in a SPAN 876 header with source SPAN address taken from fd00::1000/116 (e.g., 877 fd00::1001) and destination SPAN address taken from fd00::3000/116 878 (e.g., fd00::3001). Next, it encapsulates the SPAN message in an 879 INET header with source address set to its own INET address (e.g., 880 192.0.2.100) and destination set to the INET address of a Relay 881 (e.g., 192.0.2.1). 883 SPAN encapsulation is based on Generic Packet Tunneling in IPv6 884 [RFC2473]; the encapsulation format in the above example is shown in 885 Figure 3: 887 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 888 | INET Header | 889 | src = 192.0.2.100 | 890 | dst = 192.0.2.1 | 891 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 892 | SPAN Header | 893 | src = fd00::1001 | 894 | dst = fd00::3001 | 895 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 896 | Inner IP Header | 897 | src = 2001:db8:1:2::1 | 898 | dst = 2001:db8:1000:2000::1 | 899 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 900 | | 901 ~ ~ 902 ~ Inner Packet Body ~ 903 ~ ~ 904 | | 905 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 907 Figure 3: SPAN Encapsulation 909 In this format, the inner IP header and packet body are the original 910 IP packet, the SPAN header is an IPv6 header prepared according to 911 [RFC2473], and the INET header is prepared according to Section 3.9. 912 A packet is said to be "forwarded/sent into the SPAN" when it is 913 encapsulated as described above then forwarded via a secured tunnel 914 to a neighboring Relay. 916 This gives rise to a routing system that contains both MNP routes 917 that may change dynamically due to regional node mobility and SPAN 918 routes that never change. The Relays can therefore provide link- 919 layer bridging by sending packets into the SPAN instead of network- 920 layer routing according to MNP routes. As a result, opportunities 921 for packet loss due to node mobility between different segments are 922 mitigated. 924 With reference to Figure 3, for a Client's AERO address the SPAN 925 address is simply set to the Subnet Router Anycast address. For non- 926 link-local addresses, the destination SPAN address may not be known 927 in advance for the first few packets of a flow sent via the SPAN. In 928 that case, the SPAN destination address is set to the original 929 packet's destination, and the SPAN routing system will direct the 930 packet to the correct SPAN egress node. (In the above example, the 931 SPAN destination address is simply 2001:db8:1000:2000::1.) 933 3.5.1. SPAN Compatibility Addressing 935 For IPv4 MNPs, Servers injects a "SPAN Compatibility Prefix (SCP)" 936 that embeds the MNP into the BGP routing system. The SCP begins with 937 the upper 64 bits of the SSP, followed by the constant string 938 "0000:FFFF" followed by the IPv4 MNP. For example, if the SSP is 939 fd00::/64 and the MNP is 192.0.2.0/24 then the SCP is 940 fd00::FFFF:192.0.2.0/120. 942 This allows for encapsulation of IPv4 packets in IPv6 headers with 943 "SPAN Compatibility Addresses (SCAs)". In this example, the SCA 944 corresponding to the SCP is simply fd00::FFFF:192.0.2.0, and can be 945 used as the SPAN destination address for packets forwarded via the 946 SPAN. This allows for forwarding of initial IPv4 packets over IPv6 947 SPAN routes, followed by route optimization for direct 948 communications. 950 3.6. AERO Interface Characteristics 952 AERO interfaces are virtual interfaces configured over one or more 953 underlying interfaces classified as follows: 955 o Native interfaces have global IP addresses that are reachable from 956 any INET correspondent. All Server and Relay interfaces are 957 native interfaces, as are INET-facing interfaces of Proxys. 959 o NATed interfaces connect to a private network behind a Network 960 Address Translator (NAT). The NAT does not participate in any 961 AERO control message signaling, but the Server can issue control 962 messages on behalf of the Client. Clients that are behind a NAT 963 are required to send periodic keepalive messages to keep NAT state 964 alive when there are no data packets flowing. If no other 965 periodic messaging service is available, the Client can send RS 966 messages to receive RA replies from its Server(s). 968 o VPNed interfaces use security encapsulation to a Virtual Private 969 Network (VPN) server that also acts as an AERO Server. As with 970 NATed links, the Server can issue control messages on behalf of 971 the Client, but the Client need not send periodic keepalives in 972 addition to those already used to maintain the VPN connection. 974 o Proxyed interfaces connect to an ANET that is separated from the 975 open INET by an AERO Proxy. Unlike NATed and VPNed interfaces, 976 the Proxy can actively issue control messages on behalf of the 977 Client. 979 o Direct interfaces connect a Client directly to a neighbor without 980 crossing any ANET/INET paths. An example is a line-of-sight link 981 between a remote pilot and an unmanned aircraft. 983 AERO interfaces use encapsulation (see: Section 3.9) to exchange 984 packets with AERO link neighbors over Native, NATed or VPNed 985 interfaces. AERO interfaces avoid encapsulation on Proxyed and 986 Direct underlying interfaces. 988 AERO interfaces maintain a neighbor cache for tracking per-neighbor 989 state the same as for any interface. AERO interfaces use ND messages 990 including Router Solicitation (RS), Router Advertisement (RA), 991 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 992 neighbor cache management. 994 AERO interfaces send ND messages over Proxyed and Direct interfaces 995 with an AERO option formatted as specified in 996 [I-D.templin-atn-aero-interface]. For native, NATed and VPNed 997 interfaces (i.e., for those that use encapsulation), the AERO 998 interface omits the AERO option and instead includes a Source/Target 999 Link-Layer Address Options (S/TLLAO) formatted as shown in Figure 4: 1001 0 1 2 3 1002 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 1003 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1004 | Type | Length | Prefix Length |R|X|N| Reserved| 1005 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1006 | ifIndex[1] | ifType[1] | Flags [1] |Link[1]|QoS[1] | 1007 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1008 |P00|P01|P02|P03|P04|P05|P06|P07|P08|P09|P10|P11|P12|P13|P14|P15| 1009 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1010 |P16|P17|P18|P19|P20|P21|P22|P23|P24|P25|P26|P27|P28|P29|P30|P31| 1011 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1012 |P32|P33|P34|P35|P36|P37|P38|P39|P40|P41|P42|P43|P44|P45|P46|P47| 1013 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1014 |P48|P49|P50|P51|P52|P53|P54|P55|P56|P57|P58|P59|P60|P61|P62|P63| 1015 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1016 | | 1017 + + 1018 | | 1019 + Link Layer Address [1] + 1020 | | 1021 + + 1022 | | 1023 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1024 | Port Number [1] | ifIndex[2] | ifType[2] | 1025 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1026 | Flags [2] |Link[2]|QoS[2] | ~ 1027 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 1028 ~ ~ 1029 ~ P00 - P63, Link Layer Address, Port Number for ifIndex[2] ~ 1030 ~ ~ 1031 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1032 | ifIndex[3] | ifType[3] | Flags [3] |Link[3]|QoS[3] | 1033 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1034 ~ ~ 1035 ~ P00 - P63, Link Layer Address, Port Number for ifIndex[3] ~ 1036 ~ ~ 1037 ... ... 1038 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1039 | ifIndex[N] | ifType[N] | Flags [N] |Link[N]|QoS[N] | 1040 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1041 ~ ~ 1042 ~ P00 - P63, Link Layer Address, Port Number for ifIndex[N] ~ 1043 ~ ~ 1044 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1045 | Trailing zero padding | 1046 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1048 Figure 4: AERO Source/Target Link-Layer Address Option (S/TLLAO) 1049 Format 1051 In this format, the ifIndex-tuple field values have the same format 1052 and meaning as for the AERO option specified in 1053 [I-D.templin-atn-aero-interface] except that a 16-byte Link Layer 1054 address and 2-bye Port Number follows immediately after the P[i] 1055 fields. These values represent the link-layer addresses for this 1056 ifIndex-tuple. 1058 A Client's AERO interface may be configured over multiple underlying 1059 interface connections. For example, common mobile handheld devices 1060 have both wireless local area network ("WLAN") and cellular wireless 1061 links. These links are typically used "one at a time" with low-cost 1062 WLAN preferred and highly-available cellular wireless as a standby. 1063 In a more complex example, aircraft frequently have many wireless 1064 data link types (e.g. satellite-based, cellular, terrestrial, air-to- 1065 air directional, etc.) with diverse performance and cost properties. 1067 If a Client's multiple underlying interfaces are used "one at a time" 1068 (i.e., all other interfaces are in standby mode while one interface 1069 is active), then ND messages include only a single ifIndex-tuple in 1070 the option and set to a constant value. In that case, the Client 1071 would appear to have a single interface but with a dynamically 1072 changing Link Layer address. 1074 If the Client has multiple active underlying interfaces, then from 1075 the perspective of ND it would appear to have multiple link-layer 1076 addresses. In that case, ND messages options MAY include multiple 1077 ifIndex-tuples - each with a value that corresponds to a specific 1078 interface. The first ifIndex in the option MUST correspond to the 1079 interface over which the ND message is sent. Every ND message need 1080 not include all ifIndexes; for any ifIndex not included, the neighbor 1081 considers the status of that ifIndex as unchanged. 1083 When the Client includes an option for an underlying interface for 1084 which it is aware that there is a NAT on the path to the Server, or 1085 when a node includes an option solely for the purpose of announcing 1086 new QoS preferences, the node sets both Port Number and Link-Layer 1087 Address to 0 to indicate that the addresses are unspecified at the 1088 network layer and must instead be derived from the link-layer 1089 encapsulation headers. 1091 Relay, Server and Proxy AERO interfaces may be configured over one or 1092 more secured tunnel interfaces. The AERO interface configures both 1093 an AERO address and its corresponding SPAN address, while the 1094 underlying secured tunnel interfaces are either unnumbered or 1095 configure the same SPAN address. The AERO interface encapsulates 1096 each IP packet in a SPAN header and presents the packet to the 1097 underlying secured tunnel interface. For Relays that do not 1098 configure an AERO interface, the secured tunnel interfaces themselves 1099 are exposed to the IP layer with each interface configuring the 1100 Relay's SPAN address. Routing protocols such as BGP therefore run 1101 directly over the Relay's secured tunnel interfaces. For nodes that 1102 configure an AERO interface, routing protocols such as BGP run over 1103 the AERO interface but do not employ SPAN encapsulation. Instead, 1104 the AERO interface presents the routing protocol messages directly to 1105 the underlying secured tunnels without applying encapsulation and 1106 while using the SPAN address as the source address. This distinction 1107 must be honored consistently according to each node's configuration 1108 so that the IP forwarding table will associate discovered IP routes 1109 with the correct interface. 1111 3.7. AERO Interface Initialization 1113 AERO Servers, Proxys and Clients configure AERO interfaces as their 1114 point of attachment to the AERO link. AERO nodes assign the MSPs for 1115 the link to their AERO interfaces (i.e., as a "route-to-interface") 1116 to ensure that packets with destination addresses covered by an MNP 1117 not explicitly assigned to a non-AERO interface are directed to the 1118 AERO interface. 1120 AERO interface initialization procedures for Servers, Proxys, Clients 1121 and Relays are discussed in the following sections. 1123 3.7.1. AERO Server/Gateway Behavior 1125 When a Server enables an AERO interface, it assigns AERO/SPAN 1126 addresses and configures permanent neighbor cache entries for 1127 neighbors in the same SPAN segment by consulting the ROS list for the 1128 segment. The Server also configures secured tunnels with one or more 1129 neighboring Relays and engages in a BGP routing protocol session with 1130 each Relay. 1132 The AERO interface provides a single interface abstraction to the IP 1133 layer, but internally comprises multiple secured tunnels as well as 1134 an NBMA nexus for sending encapsulated data packets to AERO interface 1135 neighbors. The Server further configures a service to facilitate ND/ 1136 PD exchanges with AERO Clients and manages per-Client neighbor cache 1137 entries and IP forwarding table entries based on control message 1138 exchanges. 1140 Gateways are simply Servers that run a dynamic routing protocol 1141 between the AERO interface and INET/EUN interfaces (see: 1142 Section 3.3). The Gateway provisions MNPs to networks on the INET/ 1143 EUN interfaces (i.e., the same as a Client would do) and advertises 1144 the MSP(s) for the AERO link over the INET/EUN interfaces. The 1145 Gateway further provides an attachment point of the AERO link to the 1146 non-MNP-based global topology. 1148 3.7.2. AERO Proxy Behavior 1150 When a Proxy enables an AERO interface, it assigns AERO/SPAN 1151 addresses and configures permanent neighbor cache entries the same as 1152 for Servers. The Proxy also configures secured tunnels with one or 1153 more neighboring Relays and maintains per-Client neighbor cache 1154 entries based on control message exchanges. 1156 3.7.3. AERO Client Behavior 1158 When a Client enables an AERO interface, it sends RS messages with 1159 ND/PD parameters over an ANET interface to one or more Servers in the 1160 MAP list, which return RA messages with corresponding PD parameters. 1161 (The RS/RA messages may pass through a Proxy in the case of a 1162 Client's Proxyed interface.) 1164 After the initial ND/PD message exchange, the Client assigns AERO 1165 addresses to the AERO interface based on the delegated prefix(es). 1166 The Client can then register additional ANET interfaces with the 1167 Server by sending an RS message over each ANET interface. 1169 3.7.4. AERO Relay Behavior 1171 AERO Relays need not connect directly to the AERO link, since they 1172 operate as link-layer forwarding devices instead of network layer 1173 routers. Configuration of AERO interfaces on Relays is therefore 1174 OPTIONAL, e.g., if an administrative interface is needed. Relays 1175 configure secured tunnels with Servers, Proxys and other Relays; they 1176 also configure AERO/SPAN addresses and permanent neighbor cache 1177 entries the same as Servers. Relays engage in a BGP routing protocol 1178 session with a subset of the Servers on the local SPAN segment, and 1179 with other Relays on the SPAN (see: Section 3.3). 1181 3.8. AERO Interface Neighbor Cache Maintenance 1183 Each AERO interface maintains a conceptual neighbor cache that 1184 includes an entry for each neighbor it communicates with on the AERO 1185 link per [RFC4861]. AERO interface neighbor cache entries are said 1186 to be one of "permanent", "symmetric", "asymmetric" or "proxy". 1188 Permanent neighbor cache entries are created through explicit 1189 administrative action; they have no timeout values and remain in 1190 place until explicitly deleted. AERO Servers and Proxys maintain 1191 permanent neighbor cache entries for all other Servers and Proxys 1192 within the same SPAN segment. Each entry maintains the mapping 1193 between the neighbor's network-layer AERO address and corresponding 1194 INET address. The list of all permanent neighbor cache entries for 1195 the SPAN segment is maintained in the segment's ROS list. 1197 Symmetric neighbor cache entries are created and maintained through 1198 RS/RA exchanges as specified in Section 3.15, and remain in place for 1199 durations bounded by ND/PD lifetimes. AERO Servers maintain 1200 symmetric neighbor cache entries for each of their associated 1201 Clients, and AERO Clients maintain symmetric neighbor cache entries 1202 for each of their associated Servers. The list of all Servers on the 1203 AERO link is maintained in the link's MAP list. 1205 Asymmetric neighbor cache entries are created or updated based on 1206 route optimization messaging as specified in Section 3.17, and are 1207 garbage-collected when keepalive timers expire. AERO route 1208 optimization sources (ROSs) maintain asymmetric neighbor cache 1209 entries for active targets with lifetimes based on ND messaging 1210 constants. Asymmetric neighbor cache entries are unidirectional 1211 since only the ROS and not the target (e.g., a Client's MAP) creates 1212 an entry. 1214 Proxy neighbor cache entries are created and maintained by AERO 1215 Proxys when they process Client/Server ND/PD exchanges, and remain in 1216 place for durations bounded by ND/PD lifetimes. AERO Proxys maintain 1217 proxy neighbor cache entries for each of their associated Clients. 1218 Proxy neighbor cache entries track the Client state and the address 1219 of the Client's associated Server. 1221 To the list of neighbor cache entry states in Section 7.3.2 of 1222 [RFC4861], AERO interfaces add an additional state DEPARTED that 1223 applies to symmetric and proxy neighbor cache entries for Clients 1224 that have recently departed. The interface sets a "DepartTime" 1225 variable for the neighbor cache entry to "DEPARTTIME" seconds. 1226 DepartTime is decremented unless a new ND message causes the state to 1227 return to REACHABLE. While a neighbor cache entry is in the DEPARTED 1228 state, packets destined to the target Client are forwarded to the 1229 Client's new location instead of being dropped. When DepartTime 1230 decrements to 0, the neighbor cache entry is deleted. It is 1231 RECOMMENDED that DEPARTTIME be set to the default constant value 40 1232 seconds to allow for packets in flight to be delivered while stale 1233 route optimization state may be present. 1235 When a target Server (acting as a Mobility Anchor Point (MAP)) 1236 receives a valid NS message used for route optimization, it searches 1237 for a symmetric neighbor cache entry for the target Client. The MAP 1238 then returns a solicited NA message without creating a neighbor cache 1239 entry for the ROS, but creates a target Client "Report List" entry 1240 for the ROS and sets a "ReportTime" variable for the entry to 1241 REPORTTIME seconds. The MAP resets ReportTime when it receives a new 1242 authentic NS message, and otherwise decrements ReportTime while no NS 1243 messages have been received. It is RECOMMENDED that REPORTTIME be 1244 set to the default constant value 40 seconds to allow a 10 second 1245 window so that route optimization can converge before ReportTime 1246 decrements below REACHABLETIME. 1248 When the ROS receives a solicited NA message response to its NS 1249 message, it creates or updates an asymmetric neighbor cache entry for 1250 the target network-layer and link-layer addresses. The ROS then 1251 (re)sets ReachableTime for the neighbor cache entry to REACHABLETIME 1252 seconds and uses this value to determine whether packets can be 1253 forwarded directly to the target, i.e., instead of via a default 1254 route. The ROS otherwise decrements ReachableTime while no further 1255 solicited NA messages arrive. It is RECOMMENDED that REACHABLETIME 1256 be set to the default constant value 30 seconds as specified in 1257 [RFC4861]. 1259 The ROS also uses the value MAX_UNICAST_SOLICIT to limit the number 1260 of NS keepalives sent when a correspondent may have gone unreachable, 1261 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1262 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1263 to limit the number of unsolicited NAs that can be sent based on a 1264 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1265 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1266 same as specified in [RFC4861]. 1268 Different values for DEPARTTIME, REPORTTIME, REACHABLETIME, 1269 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1270 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1271 different values are chosen, all nodes on the link MUST consistently 1272 configure the same values. Most importantly, DEPARTTIME and 1273 REPORTTIME SHOULD be set to a value that is sufficiently longer than 1274 REACHABLETIME to avoid packet loss due to stale route optimization 1275 state. 1277 3.9. AERO Interface Encapsulation and Re-encapsulation 1279 AERO interfaces encapsulate packets according to whether they are 1280 entering the AERO interface from the network layer or if they are 1281 being re-admitted into the same AERO link they arrived on. This 1282 latter form of encapsulation is known as "re-encapsulation". 1284 For packets entering the AERO interface from the network layer, the 1285 AERO interface copies the "TTL/Hop Limit", "Type of Service/Traffic 1286 Class" [RFC2983], "Flow Label"[RFC6438] (for IPv6) and "Congestion 1287 Experienced" [RFC3168] values in the packet's IP header into the 1288 corresponding fields in the encapsulation header(s). 1290 For packets undergoing re-encapsulation, the AERO interface instead 1291 copies these values from the original encapsulation header into the 1292 new encapsulation header, i.e., the values are transferred between 1293 encapsulation headers and *not* copied from the encapsulated packet's 1294 network-layer header. (Note especially that by copying the TTL/Hop 1295 Limit between encapsulation headers the value will eventually 1296 decrement to 0 if there is a (temporary) routing loop.) For IPv4 1297 encapsulation/re-encapsulation, the AERO interface sets the DF bit as 1298 discussed in Section 3.13. 1300 AERO interfaces configured over INET underlying interfaces 1301 encapsulate each packet in a SPAN header, then encapsulate the 1302 resulting SPAN packet in an INET header according to the next hop 1303 determined in the forwarding algorithm in Section 3.12. If the next 1304 hop is reached via a secured tunnel, the AERO interface uses an INET 1305 encapsulation format specific to the secured tunnel type (see: 1306 Section 6). If the next hop is reached via an unsecured underlying 1307 interface, the AERO interface instead uses Generic UDP Encapsulation 1308 (GUE) [I-D.ietf-intarea-gue] or an alternate minimal encapsulation 1309 format Appendix B. 1311 When GUE encapsulation is used, the AERO interface next sets the UDP 1312 source port to a constant value that it will use in each successive 1313 packet it sends, and sets the UDP length field to the length of the 1314 SPAN packet plus 8 bytes for the UDP header itself plus the length of 1315 the GUE header (or 0 if GUE direct IP encapsulation is used). For 1316 packets sent to a Server or Relay, the AERO interface sets the UDP 1317 destination port to 8060, i.e., the IANA-registered port number for 1318 AERO. For packets sent to a Client, the AERO interface sets the UDP 1319 destination port to the port value stored in the neighbor cache entry 1320 for this Client. The AERO interface then either includes or omits 1321 the UDP checksum according to the GUE specification. 1323 Client AERO interfaces can avoid encapsulation over Direct underlying 1324 interface and Proxyed underlying interfaces for which the first-hop 1325 access router is AERO-aware. 1327 AERO interfaces observes the packet sizing and fragmentation 1328 considerations found in Section 3.13. 1330 3.10. AERO Interface Decapsulation 1332 AERO interfaces decapsulate packets destined either to the AERO node 1333 itself or to a destination reached via an interface other than the 1334 AERO interface the packet was received on. When the encapsulated 1335 packet arrives in multiple fragments, the AERO interface reassembles 1336 as discussed in Section 3.13. Further decapsulation steps are 1337 performed according to the appropriate encapsulation format 1338 specification. 1340 3.11. AERO Interface Data Origin Authentication 1342 AERO nodes employ simple data origin authentication procedures. In 1343 particular: 1345 o AERO Relays, Servers and Proxys accept encapsulated data packets 1346 and control messages received from secured tunnels. 1348 o AERO Servers and Proxys accept encapsulated data packets and NS 1349 messages used for Neighbor Unreachability Detection (NUD) received 1350 from a source found in the ROS list. 1352 o AERO Proxys and Clients accept packets that originate from within 1353 the same secured ANET. 1355 o AERO Clients and Gateways accept packets from downstream network 1356 correspondents based on ingress filtering. 1358 AERO nodes silently drop any packets that do not satisfy the above 1359 data origin authentication procedures. Further security 1360 considerations are discussed Section 6. 1362 3.12. AERO Interface Forwarding Algorithm 1364 IP packets enter a node's AERO interface either from the network 1365 layer (i.e., from a local application or the IP forwarding system) or 1366 from the link layer (i.e., from an AERO interface neighbor). All 1367 packets entering a node's AERO interface first undergo data origin 1368 authentication as discussed in Section 3.11. Packets that satisfy 1369 data origin authentication are processed further, while all others 1370 are dropped silently. 1372 Packets that enter the AERO interface from the network layer are 1373 forwarded to an AERO interface neighbor. Packets that enter the AERO 1374 interface from the link layer are either re-admitted into the AERO 1375 link or forwarded to the network layer where they are subject to 1376 either local delivery or IP forwarding. In all cases, the AERO 1377 interface itself MUST NOT decrement the network layer TTL/Hop-count 1378 since its forwarding actions occur below the network layer. 1380 AERO interfaces may have multiple underlying interfaces and/or 1381 neighbor cache entries for neighbors with multiple ifIndex 1382 registrations (see Section 3.6). The AERO interface uses each 1383 packet's DSCP value (and/or port number) to select an outgoing 1384 underlying interface based on the node's own QoS preferences, and 1385 also to select a destination link-layer address based on the 1386 neighbor's underlying interface with the highest preference. AERO 1387 implementations SHOULD allow for QoS preference values to be modified 1388 at runtime through network management. 1390 If multiple outgoing interfaces and/or neighbor interfaces have a 1391 preference of "high", the AERO node replicates the packet and sends 1392 one copy via each of the (outgoing / neighbor) interface pairs; 1393 otherwise, the node sends a single copy of the packet via the 1394 interface with the highest preference. AERO nodes keep track of 1395 which underlying interfaces are currently "reachable" or 1396 "unreachable", and only use "reachable" interfaces for forwarding 1397 purposes. 1399 The following sections discuss the AERO interface forwarding 1400 algorithms for Clients, Proxys, Servers and Relays. In the following 1401 discussion, a packet's destination address is said to "match" if it 1402 is the same as a cached address, or if it is covered by a cached 1403 prefix (which may be encoded in an AERO address). 1405 3.12.1. Client Forwarding Algorithm 1407 When an IP packet enters a Client's AERO interface from the network 1408 layer the Client searches for an asymmetric neighbor cache entry that 1409 matches the destination. If there is a match, the Client uses one or 1410 more "reachable" neighbor interfaces in the entry for packet 1411 forwarding. If there is no asymmetric neighbor cache entry, the 1412 Client instead forwards the packet toward a Server (the packet is 1413 intercepted by a Proxy if there is a Proxy on the path). 1415 When an IP packet enters a Client's AERO interface from the link- 1416 layer, if the destination matches one of the Client's MNPs or link- 1417 local addresses the Client decapsulates the packet (if necessary) and 1418 delivers it to the network layer. Otherwise, the Client drops the 1419 packet and MAY return a network-layer ICMP Destination Unreachable 1420 message subject to rate limiting (see: Section 3.14). 1422 3.12.2. Proxy Forwarding Algorithm 1424 For control messages originating from or destined to a Client, the 1425 Proxy intercepts the message and updates its proxy neighbor cache 1426 entry for the Client. The Proxy then forwards a (proxyed) copy of 1427 the control message. (For example, the Proxy forwards a proxied 1428 version of a Client's NS/RS message to the target neighbor, and 1429 forwards a proxied version of the NA/RA reply to the Client.) 1431 When the Proxy receives a data packet from a Client within the ANET, 1432 the Proxy searches for an asymmetric neighbor cache entry that 1433 matches the destination and forwards the packet as follows: 1435 o if the destination matches an asymmetric neighbor cache entry, the 1436 Proxy uses one or more "reachable" neighbor interfaces in the 1437 entry for packet forwarding via encapsulation. If the neighbor 1438 interface is in the same SPAN segment, the Proxy forwards the 1439 packet directly to the neighbor; otherwise, it forwards the packet 1440 to a Relay. 1442 o else, the Proxy encapsulates and forwards the packet to a Relay 1443 while using the packet's destination address as the SPAN 1444 destination address. (If the destination is an AERO address, the 1445 Proxy instead uses the corresponding Subnet Router Anycast address 1446 for Client AERO addresses and the SPAN address for 1447 administratively-provisioned AERO addresses.). 1449 When the Proxy receives an encapsulated data packet from an INET 1450 neighbor or from a secured tunnel, it accepts the packet only if data 1451 origin authentication succeeds and the SPAN destination address is 1452 its own address. If the packet is a SPAN fragment, the Proxy then 1453 adds the fragment to the reassembly buffer and returns if the 1454 reassembly is still incomplete. Otherwise, the Proxy reassembles the 1455 packet (if necessary) and continues processing. 1457 Next, the Proxy searches for a proxy neighbor cache entry that 1458 matches the destination. If there is a proxy neighbor cache entry in 1459 the REACHABLE state, the Proxy decapsulates and forwards the packet 1460 to the Client. If the neighbor cache entry is in the DEPARTED state, 1461 the Proxy instead re-encapsulates the message and forwards it to a 1462 Relay. If there is no neighbor cache entry, the Proxy instead 1463 discards the packet. 1465 3.12.3. Server/Gateway Forwarding Algorithm 1467 For control messages destined to a target Client's AERO address that 1468 are received from a secured tunnel, the Server (acting as a MAP) 1469 intercepts the message and sends an appropriate response on behalf of 1470 the Client. (For example, the Server sends an NA message reply in 1471 response to an NS message directed to one of its associated Clients.) 1472 If the Client's neighbor cache entry is in the DEPARTED state, 1473 however, the Server instead forwards the packet to the Client's new 1474 Server as discussed in Section 3.19. 1476 When the Server receives an encapsulated data packet from an INET 1477 neighbor or from a secured tunnel, it accepts the packet only if data 1478 origin authentication succeeds. If the SPAN destination address is 1479 its own address, the Server reassembles if necessary and discards the 1480 SPAN header (if the reassembly is incomplete, the Server instead adds 1481 the fragment to the reassembly buffer and returns). The Server then 1482 continues processing as follows: 1484 o if the destination matches a symmetric neighbor cache entry in the 1485 REACHABLE state the Server prepares the packet for forwarding to 1486 the destination Client. If the current header is a SPAN header, 1487 the Server reassembles if necessary and discards the SPAN header 1488 (if the reassembly is incomplete, the Server instead adds the 1489 fragment to the reassembly buffer and returns). The Server then 1490 forwards the packet according to the cached link-layer 1491 information, while using SPAN encapsulation for the Client's 1492 Proxyed/Native interfaces, simple INET encapsulation for NATed/ 1493 VPNed interfaces, or no encapsulation for Direct interfaces. If 1494 the packet is destined to the same Client from which it arrived 1495 (i.e., if the packet was forwarded by one of the Client's Proxys), 1496 the Server forwards the packet via a different "reachable" 1497 neighbor interface than the one the packet arrived on. If there 1498 are no "reachable" neighbor interfaces, the Server drops the 1499 packet. 1501 o else, if the destination matches a symmetric neighbor cache entry 1502 in the DEPARETED state the Server encapsulates the packet in a new 1503 SPAN header and forwards it to the Client's new Server (noting 1504 that the encapsulation may result in the addition of a second SPAN 1505 header). The Server uses its own SPAN address as the source and 1506 the SPAN address of the new Server as the destination. 1508 o else, if the destination matches an asymmetric neighbor cache 1509 entry, the Server uses one or more "reachable" neighbor interfaces 1510 in the entry for packet forwarding via the local INET if the 1511 neighbor is in the same SPAN segment or via a Relay otherwise. 1513 o else, if the destination is an AERO address that is not assigned 1514 on the AERO interface the Server drops the packet. 1516 o else, the Server (acting as a Gateway) releases the packet to the 1517 network layer for local delivery or IP forwarding. Based on the 1518 information in the forwarding table, the network layer may return 1519 the packet to the same AERO interface in which case further 1520 processing occurs as below. (Note that this arrangement 1521 accommodates common implementations in which the IP forwarding 1522 table is not accessible from within the AERO interface. If the 1523 AERO interface can directly access the IP forwarding table, the 1524 forwarding table lookup can instead be performed internally from 1525 within the AERO interface itself.) 1527 When the Server's AERO interface receives a data packet from the 1528 network layer or from a NATed/VPNed/Direct Client, it processes the 1529 packet according to the network-layer destination address as follows: 1531 o if the destination matches a symmetric or asymmetric neighbor 1532 cache entry the Server processes the packet as above. 1534 o else, the Server encapsulates the packet and forwards it to a 1535 Relay. For administratively-assigned AERO address destinations, 1536 the Server uses the SPAN address corresponding to the destination 1537 as the SPAN destination address. For Client AERO address 1538 destinations, the Server uses the Subnet Router Anycast address 1539 corresponding to the destination as the SPAN destination address. 1540 For all others, the Server uses the packet's destination IP 1541 address as the SPAN destination address. 1543 3.12.4. Relay Forwarding Algorithm 1545 Relays forward packets over secured tunnels the same as any IP 1546 router. When the Relay receives an encapsulated packet via a secured 1547 tunnel, it removes the INET header and searches for a forwarding 1548 table entry that matches the destination address in the next header. 1549 The Relay then processes the packet as follows: 1551 o if the destination matches one of the Relay's own addresses, the 1552 Relay submits the packet for local delivery. 1554 o else, if the destination matches a forwarding table entry the 1555 Relay forwards the packet via a secured tunnel to the next hop. 1556 If the destination matches an MSP without matching an MNP, 1557 however, the Relay instead drops the packet and returns an ICMP 1558 Destination Unreachable message subject to rate limiting (see: 1559 Section 3.14). 1561 o else, the Relay drops the packet and returns an ICMP Destination 1562 Unreachable as above. 1564 As for any IP router, the Relay decrements the TTL/Hop Limit when it 1565 forwards the packet. If the packet is encapsulated in a SPAN header, 1566 only the Hop Limit in the SPAN header is decremented, and not the 1567 TTL/Hop Limit in the inner packet header. 1569 3.13. AERO Interface MTU and Fragmentation 1571 The AERO interface is the node's attachment to the AERO link. For 1572 AERO link neighbor underlying interface paths that do not require 1573 encapsulation, the AERO interface sends unencapsulated IP packets. 1574 For other paths, the AERO interface acts as a tunnel ingress when it 1575 sends packets to the neighbor and as a tunnel egress when it receives 1576 packets from the neighbor. 1578 AERO interfaces configure an MTU the same as for any IP interface, 1579 however the MTU does not reflect the physical size of any links in 1580 the path but rather determines the maximum size for reassembly. AERO 1581 interfaces observe the packet sizing considerations for tunnels 1582 discussed in [I-D.ietf-intarea-tunnels] and as specified below. 1584 The Internet Protocol expects that IP packets will either be 1585 delivered to the destination or a suitable Packet Too Big (PTB) 1586 message returned to support the process known as IP Path MTU 1587 Discovery (PMTUD) [RFC1191][RFC8201]. However, PTB messages may be 1588 crafted for malicious purposes or lost in the network [RFC2923]. 1589 This can be especially problematic for tunnels, where a condition 1590 known as a PMTUD "black hole" can result. For these reasons, AERO 1591 interfaces employ operational procedures that avoid interactions with 1592 PMTUD, including the use of fragmentation when necessary. 1594 AERO interfaces observe three different types of fragmentation. 1595 Source fragmentation occurs when the AERO interface (acting as a 1596 tunnel ingress) fragments the encapsulated packet into multiple 1597 fragments before admitting each fragment into the tunnel. Network 1598 fragmentation occurs when an encapsulated packet admitted into the 1599 tunnel by the ingress is fragmented by an IPv4 router on the path to 1600 the egress. Finally, link-layer fragmentation (aka link adaptation) 1601 occurs at a layer below IP and is coordinated between underlying data 1602 link endpoints. 1604 IPv6 specifies a minimum link Maximum Transmission Unit (MTU) of 1280 1605 bytes [RFC8200]. Although IPv4 specifies a smaller minimum link MTU 1606 of 68 bytes [RFC0791], AERO interfaces also observe the IPv6 minimum 1607 for IPv4 even if encapsulated packets may incur network 1608 fragmentation. 1610 IPv6 specifies a minimum Maximum Reassembly Unit (MRU) of 1500 bytes 1611 [RFC8200], while the minimum MRU for IPv4 is only 576 bytes [RFC1122] 1612 (but, note that many standard IPv6 over IPv4 tunnel types already 1613 assume a larger MRU than the IPv4 minimum). 1615 AERO interfaces therefore configure an MTU that MUST NOT be smaller 1616 than 1280 bytes, MUST NOT be larger than the minimum MRU among all 1617 nodes on the AERO link minus the encapsulation overhead ("ENCAPS"), 1618 and SHOULD NOT be smaller than 1500 bytes. AERO interfaces also 1619 configure a Maximum Segment Unit (MSU) as the maximum-sized 1620 encapsulated packet that the ingress can inject into the tunnel 1621 without source fragmentation. The MSU value MUST NOT be larger than 1622 1280 bytes unless there is operational assurance that a larger size 1623 can traverse the link along all paths. 1625 All AERO interfaces on the link MUST configure the same MTU value for 1626 reasons cited in [RFC3819][RFC4861]; in particular, multicast support 1627 requires a common MTU value among all nodes on the link. All AERO 1628 interfaces MUST configure an MRU large enough to reassemble packets 1629 up to (MTU+ENCAPS) bytes in length; nodes that cannot configure a 1630 large-enough MRU MUST NOT enable an AERO interface. For example, for 1631 an MTU of 1500 bytes an appropriate MRU might be 2KB. 1633 The network layer proceeds as follows when it forwards an IP packet 1634 to the AERO interface. For each IPv4 packet that is larger than the 1635 AERO interface MTU and with DF set to 0, the network layer uses IPv4 1636 fragmentation to break the packet into a minimum number of non- 1637 overlapping fragments where the first fragment is no larger than the 1638 MTU and the remaining fragments are no larger than the first. For 1639 all other IP packets, if the packet is larger than the AERO interface 1640 MTU, the network layer drops the packet and returns a PTB message to 1641 the original source. Otherwise, the network layer admits each IP 1642 packet or fragment into the AERO interface. 1644 For each IP packet admitted into AERO interface, if the neighbor is 1645 reached via an underlying interface that does not require 1646 encapsulation the AERO interface proceeds according to the underlying 1647 interface MTU. If the packet is no larger than the underlying 1648 interface MTU, the AERO interface presents the packet to the 1649 underlying interface. Otherwise, for IPv4 packets with DF set to 0 1650 the AERO interface uses IPv4 fragmentation to break the packet into 1651 fragments no larger than the underlying interface MTU. For other 1652 packets, the AERO interface either performs link adaptation or drops 1653 the packet and returns a PTB message to the original source. (If the 1654 original source corresponds to a local application, the PTB would 1655 appear to have originated from a router on the path when in fact it 1656 was locally generated from within the AERO interface.) In the same 1657 way, when a packet that has been admitted into the AERO link reaches 1658 a target neighbor but is too large to be delivered over the final-hop 1659 underlying interface, the target either performs link adaptation or 1660 drops the packet and returns a PTB. Link adaptation is preferred in 1661 both cases when possible to avoid packet loss. 1663 For underlying interfaces that require encapsulation, the AERO 1664 interface (acting as a tunnel ingress) instead encapsulates the 1665 packet and performs path MTU procedures according to the specific 1666 encapsulation format. For INET interfaces, the ingress encapsulates 1667 the packet in a SPAN header. If the SPAN packet is larger than the 1668 MSU, the ingress source fragments the SPAN packet into a minimum 1669 number of non-overlapping fragments where the first fragment is no 1670 larger than the MSU and the remaining fragments are no larger than 1671 the first. The ingress then encapsulates each SPAN packet/fragment 1672 in an INET header and admits them into the tunnel. For IPv4, the 1673 ingress sets the DF bit to 0 in the INET header in case any network 1674 fragmentation is necessary. The encapsulated packets will be 1675 delivered to the egress, which reassembles them into a whole packet 1676 if necessary. 1678 By fragmenting at the SPAN layer instead of lower layers, standard 1679 IPv6 fragmentation and reassembly [RFC8200] ensures that IPv4 issues 1680 such as data corruption due to reassembly misassociations will not 1681 occur [RFC6864][RFC4963]. The ingress sends each fragment with 1682 minimal delay so that individual fragments are unlikely to be 1683 diverted to different destinations due to routing fluctuations. 1685 Since the SPAN header and IPv6 fragment extension header reduces the 1686 room available for packet data, but the original source has no way to 1687 control its insertion, the ingress MUST include their lengths in 1688 ENCAPS even for packets in which the header is absent. 1690 3.13.1. AERO MTU Requirements 1692 In light of the above considerations, AERO interfaces configure an 1693 MTU of 9180 bytes. This means that the AERO interface MUST be 1694 capable of reassembling original IP packets up to 9180 bytes in 1695 length. When an IP packet is admitted into an AERO interface, the 1696 interface encapsulates the packet using SPAN encapsulation and 1697 fragments the encapsulated packet into fragments that are no larger 1698 than 1280 bytes. The fragments will be reassembled by the tunnel 1699 egress that services the final destination. 1701 AERO Clients behind Proxys MAY configure an MTU smaller than 9180 1702 (but no smaller than IP minimum link MTU). If Clients configure a 1703 diversity of MTUs (e.g., 1280, 1500, 4KB, 8KB, etc.) then neighbors 1704 on the link would appear to have multiple MTUs. IPv6 Path MTU 1705 Discovery [RFC8201] accounts for this possibility since MTU discovery 1706 must be performed even between nodes that appear to be connected to 1707 the same link. 1709 Applications that cannot tolerate loss in the network due to MTU 1710 restrictions should restrict themselves to sending packets no larger 1711 than the IP minimum link MTU, i.e., even if the current path MTU 1712 would appear to support a larger size. This is due to the fact that 1713 routing changes could cause the path to traverse links with smaller 1714 MTUs at any given point in time. 1716 3.14. AERO Interface Error Handling 1718 When an AERO node admits encapsulated packets into the AERO 1719 interface, it may receive link-layer or network-layer error 1720 indications. 1722 A link-layer error indication is an ICMP error message generated by a 1723 router in the INET on the path to the neighbor or by the neighbor 1724 itself. The message includes an IP header with the address of the 1725 node that generated the error as the source address and with the 1726 link-layer address of the AERO node as the destination address. 1728 The IP header is followed by an ICMP header that includes an error 1729 Type, Code and Checksum. Valid type values include "Destination 1730 Unreachable", "Time Exceeded" and "Parameter Problem" 1731 [RFC0792][RFC4443]. (AERO interfaces ignore all link-layer IPv4 1732 "Fragmentation Needed" and IPv6 "Packet Too Big" messages since they 1733 only emit packets that are guaranteed to be no larger than the IP 1734 minimum link MTU as discussed in Section 3.13.) 1736 The ICMP header is followed by the leading portion of the packet that 1737 generated the error, also known as the "packet-in-error". For 1738 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1739 much of invoking packet as possible without the ICMPv6 packet 1740 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1741 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1742 "Internet Header + 64 bits of Original Data Datagram", however 1743 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1744 ICMP datagram SHOULD contain as much of the original datagram as 1745 possible without the length of the ICMP datagram exceeding 576 1746 bytes". 1748 The link-layer error message format is shown in Figure 5 (where, "L2" 1749 and "L3" refer to link-layer and network-layer, respectively): 1751 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1752 ~ ~ 1753 | L2 IP Header of | 1754 | error message | 1755 ~ ~ 1756 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1757 | L2 ICMP Header | 1758 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1759 ~ ~ P 1760 | IP and other encapsulation | a 1761 | headers of original L3 packet | c 1762 ~ ~ k 1763 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1764 ~ ~ t 1765 | IP header of | 1766 | original L3 packet | i 1767 ~ ~ n 1768 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1769 ~ ~ e 1770 | Upper layer headers and | r 1771 | leading portion of body | r 1772 | of the original L3 packet | o 1773 ~ ~ r 1774 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1776 Figure 5: AERO Interface Link-Layer Error Message Format 1778 The AERO node rules for processing these link-layer error messages 1779 are as follows: 1781 o When an AERO node receives a link-layer Parameter Problem message, 1782 it processes the message the same as described as for ordinary 1783 ICMP errors in the normative references [RFC0792][RFC4443]. 1785 o When an AERO node receives persistent link-layer Time Exceeded 1786 messages, the IP ID field may be wrapping before earlier fragments 1787 awaiting reassembly have been processed. In that case, the node 1788 SHOULD begin including integrity checks and/or institute rate 1789 limits for subsequent packets. 1791 o When an AERO node receives persistent link-layer Destination 1792 Unreachable messages in response to encapsulated packets that it 1793 sends to one of its asymmetric neighbor correspondents, the node 1794 SHOULD process the message as an indication that a path may be 1795 failing, and MAY initiate NUD over that path. If it receives 1796 Destination Unreachable messages on many or all paths, the node 1797 SHOULD set ReachableTime for the corresponding asymmetric neighbor 1798 cache entry to 0 and allow future packets destined to the 1799 correspondent to flow through a default route. 1801 o When an AERO Client receives persistent link-layer Destination 1802 Unreachable messages in response to encapsulated packets that it 1803 sends to one of its symmetric neighbor Servers, the Client SHOULD 1804 mark the path as unusable and use another path. If it receives 1805 Destination Unreachable messages on many or all paths, the Client 1806 SHOULD associate with a new Server and release its association 1807 with the old Server as specified in Section 3.19.5. 1809 o When an AERO Server receives persistent link-layer Destination 1810 Unreachable messages in response to encapsulated packets that it 1811 sends to one of its symmetric neighbor Clients, the Server SHOULD 1812 mark the underlying path as unusable and use another underlying 1813 path. If it receives Destination Unreachable messages on multiple 1814 paths, the Server should take no further actions unless it 1815 receives an explicit ND/PD release message or if the PD lifetime 1816 expires. In that case, the Server MUST release the Client's 1817 delegated MNP, withdraw the MNP from the AERO routing system and 1818 delete the neighbor cache entry. 1820 o When an AERO Server or Proxy receives link-layer Destination 1821 Unreachable messages in response to an encapsulated packet that it 1822 sends to one of its permanent neighbors, it treats the messages as 1823 an indication that the path to the neighbor may be failing. 1824 However, the dynamic routing protocol should soon reconverge and 1825 correct the temporary outage. 1827 When an AERO Relay receives a packet for which the network-layer 1828 destination address is covered by an MSP, if there is no more- 1829 specific routing information for the destination the Relay drops the 1830 packet and returns a network-layer Destination Unreachable message 1831 subject to rate limiting. The Relay writes the network-layer source 1832 address of the original packet as the destination address and uses 1833 one of its non link-local addresses as the source address of the 1834 message. 1836 When an AERO node receives an encapsulated packet for which the 1837 reassembly buffer it too small, it drops the packet and returns a 1838 network-layer Packet Too Big (PTB) message. The node first writes 1839 the MRU value into the PTB message MTU field, writes the network- 1840 layer source address of the original packet as the destination 1841 address and writes one of its non link-local addresses as the source 1842 address. 1844 3.15. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1846 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1847 coordinated as discussed in the following Sections. 1849 3.15.1. AERO ND/PD Service Model 1851 Each AERO Server on the link configures a PD service to facilitate 1852 Client requests. Each Server is provisioned with a database of MNP- 1853 to-Client ID mappings for all Clients enrolled in the AERO service, 1854 as well as any information necessary to authenticate each Client. 1855 The Client database is maintained by a central administrative 1856 authority for the AERO link and securely distributed to all Servers, 1857 e.g., via the Lightweight Directory Access Protocol (LDAP) [RFC4511], 1858 via static configuration, etc. Clients can receive new PDs from new 1859 Servers before releasing PDs received from existing Servers for 1860 service continuity. Clients receive the same service regardless of 1861 the Servers they select. 1863 AERO Clients and Servers use ND messages to maintain neighbor cache 1864 entries. AERO Servers configure their AERO interfaces as advertising 1865 interfaces, and therefore send unicast RA messages with configuration 1866 information in response to a Client's RS message. Thereafter, 1867 Clients send additional RS messages to refresh prefix and/or router 1868 lifetimes. 1870 AERO Clients and Servers include PD parameters in RS/RA messages (see 1871 [I-D.templin-6man-dhcpv6-ndopt] for ND/PD alternatives). The unified 1872 ND/PD messages are exchanged between Client and Server according to 1873 the prefix management schedule required by the PD service. If the 1874 Client knows its MNP in advance, it can include its AERO address as 1875 the source address of an RS message and with an SLLAO with a valid 1876 Prefix Length for the MNP. If the Server (and Proxy) accept the 1877 Client's MNP assertion, they inject the prefix into the routing 1878 system and establish the necessary neighbor cache state. 1880 The following sections specify the Client and Server behavior. 1882 3.15.2. AERO Client Behavior 1884 AERO Clients can discover the addresses of Servers in the same manner 1885 described in [RFC5214]. Discovery methods include static 1886 configuration (e.g., from a flat-file map of Server addresses and 1887 locations), or through an automated means such as Domain Name System 1888 (DNS) name resolution [RFC1035]. In the absence of other 1889 information, the Client can resolve the DNS Fully-Qualified Domain 1890 Name (FQDN) "linkupnetworks.[domainname]" where "linkupnetworks" is a 1891 constant text string and "[domainname]" is a DNS suffix for the AERO 1892 link (e.g., "example.com"). Alternatively, the Client can discover 1893 Server addresses through a layer 2 data link login exchange, or 1894 through a multicast or anycast RS as described below. 1896 To associate with a Server, the Client acts as a requesting router to 1897 request MNPs. The Client prepares an RS message with PD parameters 1898 and includes a Nonce and Timestamp option if the Client needs to 1899 correlate RA replies. If the Client already knows the Server's AERO 1900 address, it includes the AERO address as the network-layer 1901 destination address; otherwise, it includes all-routers multicast 1902 (ff02::2) or subnet routers anycast (fe80::) as the network-layer 1903 destination address. If the Client already knows its own AERO 1904 address, it uses the AERO address as the network-layer source 1905 address; otherwise, it uses the unspecified AERO address 1906 (fe80::ffff:ffff) as the network-layer source address. 1908 The Client next includes an SLLAO in the RS message formatted as 1909 described in Section 3.6 to register its link-layer information with 1910 the Server. The first ifIndex in the SLLAO MUST correspond to the 1911 underlying interface over which the Client will send the RS message. 1912 The Client MAY include additional ifIndexes specific to other 1913 underlying interfaces, but if so it MUST set their Port Number and 1914 Link Layer Address fields to 0. If the Client is connected to an 1915 ANET for which encapsulation is required, the Client finally 1916 encapsulates the RS message in an ANET header with its own ANET 1917 address as the source address and the INET address of the Server as 1918 the destination. 1920 The Client then sends the RS message (either directly via Direct 1921 interfaces, via INET encapsulation for NATed interfaces, via a VPN 1922 for VPNed interfaces, via a Proxy for proxyed interfaces or via a 1923 Relay for native interfaces) and waits for an RA message reply (see 1924 Section 3.15.3). The Client retries up to MAX_RTR_SOLICITATIONS 1925 times until an RA is received. If the Client receives no RAs, or if 1926 it receives an RA with Router Lifetime set to 0, the Client SHOULD 1927 abandon this Server and try another Server. Otherwise, the Client 1928 processes the PD information found in the RA message. 1930 Next, the Client creates a symmetric neighbor cache entry with the 1931 Server's AERO address as the network-layer address and the address in 1932 the SLLAO for ifIndex 0 as the Server's INET address. The Client 1933 records the RA Router Lifetime field value in the neighbor cache 1934 entry as the time for which the Server has committed to maintaining 1935 the MNP in the routing system. The Client then autoconfigures AERO 1936 addresses for each of the delegated MNPs and assigns them to the AERO 1937 interface. The Client also caches any MSPs included in Route 1938 Information Options (RIOs) [RFC4191] as MSPs to associate with the 1939 AERO link, and assigns the MTU value in the MTU option to its AERO 1940 interface while configuring an appropriate MRU. 1942 The Client then registers additional underlying interfaces with the 1943 Server by sending RS messages via each additional interface. The RS 1944 messages include the same parameters as for the initial RS/RA 1945 exchange, but with destination address set to the Server's AERO 1946 address and with an ifIndex in the SLLAO specific to the underlying 1947 interface. 1949 The Client examines the X and N bits in the SLLAO with S set to 1 in 1950 each RA message it receives. If X is 1 the Client infers that there 1951 is a Proxy on the path, and if N is 1 the Client infers that there is 1952 a NAT on the path. If N is 1, the Client SHOULD set Port Number and 1953 Link-Layer Address to 0 of any subsequent ND messages it sends to the 1954 Server over that link. 1956 Following autoconfiguration, the Client sub-delegates the MNPs to its 1957 attached EUNs and/or the Client's own internal virtual interfaces as 1958 described in [I-D.templin-v6ops-pdhost] to support the Client's 1959 downstream attached "Internet of Things (IoT)". The Client 1960 subsequently maintains its MNP delegations through each of its 1961 Servers by sending additional RS messages before Router Lifetime 1962 expires. 1964 After the Client registers its underlying interfaces, it may wish to 1965 change one or more registrations, e.g., if an interface changes 1966 address or becomes unavailable, if QoS preferences change, etc. To 1967 do so, the Client prepares an RS message to send over any available 1968 underlying interface. The RS MUST include an ifIndex for the 1969 selected interface as the first ifIndex in the SLLAO and MAY include 1970 any additional ifIndexes specific to other underlying interfaces. 1971 The Client includes fresh P(i) values for each ifIndex to update the 1972 Server's neighbor cache entry. If the Client wishes to update only 1973 the P(i) values, it sets the Port Number and Link-Layer Address 1974 fields to 0. If the Client wishes to disable the underlying 1975 interface, it sets all P(i) values to 0. When the Client receives 1976 the Server's RA response, it has assurance that the Server has been 1977 updated with the new information. 1979 If the Client wishes to discontinue use of a Server it issues an RS 1980 message over any underlying interface with an SLLAO with R set to 0. 1981 When the Server processes the message, it releases the MNP, sets the 1982 symmetric neighbor cache entry state for the Client to DEPARTED, 1983 withdraws the IP route from the routing system and returns an RA 1984 reply with Router Lifetime set to 0. 1986 3.15.3. AERO Server Behavior 1988 AERO Servers act as IP routers and support a PD service for Clients. 1989 Servers arrange to add their AERO and INET addresses to a static map 1990 of Server addresses for the link and/or the DNS resource records for 1991 the FQDN "linkupnetworks.[domainname]" before entering service. 1992 Server addresses should be geographically and/or topologically 1993 referenced, and made available for discovery by Clients on AERO link. 1995 When a Server receives a prospective Client's RS message on its AERO 1996 interface, it SHOULD return an immediate RA reply with Router 1997 Lifetime set to 0 if it is currently too busy or otherwise unable to 1998 service the Client. Otherwise, the Server authenticates the RS 1999 message and processes the PD parameters. The Server first determines 2000 the correct MNPs to delegate to the Client by searching the Client 2001 database. When the Server delegates the MNPs, it also creates an IP 2002 forwarding table entry for each MNP so that the MNPs are propagated 2003 into the routing system (see: Section 3.3). For IPv6, the Server 2004 creates a single IPv6 forwarding table entry for each MNP. For IPv4, 2005 the Server creates an IPv6 forwarding table entry with the 2006 IPv4-mapped IPv6 address corresponding to the IPv4 address. 2008 The Server next creates a symmetric neighbor cache entry for the 2009 Client using the base AERO address as the network-layer address and 2010 with lifetime set to no more than the smallest PD lifetime. Next, 2011 the Server updates the neighbor cache entry by recording the 2012 information in each ifIndex in the RS SLLAO including the Port 2013 Number, Link Layer Address and P(i) values. For the first ifIndex, 2014 however, the Server records the actual INET header source addresses 2015 instead of those that appear in the SLLAO in case there was a NAT in 2016 the path. The Server also records the value of the X bit to indicate 2017 whether there is a Proxy on the path. 2019 Next, the Server prepares an RA message using its AERO address as the 2020 network-layer source address and the network-layer source address of 2021 the RS message as the network-layer destination address. The Server 2022 includes the delegated MNPs, any other PD parameters and an SLLAO 2023 with the Link Layer Address set to the Server's SPAN address and with 2024 ifIndex set to 0. The Server then includes one or more RIOs that 2025 encode the MSPs for the AERO link, plus an MTU option for the link 2026 MTU (see Section 3.13). The Server finally forwards the message to 2027 the Client using SPAN, INET or NULL encapsulation according to the 2028 Client interface type. (For Proxy/Native interfaces, the Server 2029 encapsulates the message in a SPAN header with source address set to 2030 its own SPAN address and destination address set to the Proxy's (or 2031 Client's) SPAN address, then forwards the message into the SPAN.) 2032 After the initial RS/RA exchange, the Server maintains the symmetric 2033 neighbor cache entry for the Client. If the Client (or Proxy) issues 2034 additional NS/RS messages, the Server resets ReachableTime. If the 2035 Client (or Proxy) issues an RS with PD release parameters (e.g., by 2036 including an SLLAO with R set to 0), or if the Client becomes 2037 unreachable, the Server sets the Client's symmetric neighbor cache 2038 entry to the DEPARTED state and withdraws the IP routes from the AERO 2039 routing system. 2041 The Server processes these and any other Client ND/PD messages, and 2042 returns an NA/RA reply. The Server may also issue unsolicited RA 2043 messages, e.g., with PD reconfigure parameters to cause the Client to 2044 renegotiate its PDs, with Router Lifetime set to 0 if it can no 2045 longer service this Client, etc. Finally, If the symmetric neighbor 2046 cache entry is in the DEPARTED state, the Server deletes the entry 2047 after DepartTime expires. 2049 3.15.3.1. Lightweight DHCPv6 Relay Agent (LDRA) 2051 When DHCPv6 is used as the ND/PD service back end, AERO Clients and 2052 Servers are always on the same link (i.e., the AERO link) from the 2053 perspective of DHCPv6. However, in some implementations the DHCPv6 2054 server and ND function may be located in separate modules. In that 2055 case, the Server's AERO interface module can act as a Lightweight 2056 DHCPv6 Relay Agent (LDRA)[RFC6221] to relay PD messages to and from 2057 the DHCPv6 server module. 2059 When the LDRA receives an authentic RS message, it extracts the PD 2060 message parameters and uses them to construct an IPv6/UDP/DHCPv6 2061 message. It sets the IPv6 source address to the source address of 2062 the RS message, sets the IPv6 destination address to 2063 'All_DHCP_Relay_Agents_and_Servers' and sets the UDP fields to values 2064 that will be understood by the DHCPv6 server. 2066 The LDRA then wraps the message in a DHCPv6 'Relay-Forward' message 2067 header and includes an 'Interface-Id' option that includes enough 2068 information to allow the LDRA to forward the resulting Reply message 2069 back to the Client (e.g., the Client's link-layer addresses, a 2070 security association identifier, etc.). The LDRA also wraps the 2071 SLLAO into the Interface-Id option, then forwards the message to the 2072 DHCPv6 server. 2074 When the DHCPv6 server prepares a Reply message, it wraps the message 2075 in a 'Relay-Reply' message and echoes the Interface-Id option. The 2076 DHCPv6 server then delivers the Relay-Reply message to the LDRA, 2077 which discards the Relay-Reply wrapper and IPv6/UDP headers, then 2078 uses the DHCPv6 message to construct an RA response to the Client. 2079 The Server uses the information in the Interface-Id option to prepare 2080 the RA message and to cache the link-layer addresses taken from the 2081 SLLAOs echoed in the Interface-Id option. 2083 3.16. The AERO Proxy 2085 Clients may connect to ANETs that require a perimeter security 2086 gateway to enable communications to Servers in outside INETs. In 2087 that case, the ANET can employ an AERO Proxy. The Proxy is located 2088 at the ANET/INET border and listens for RS messages originating from 2089 or RA messages destined to ANET Clients. The Proxy acts on these 2090 control messages as follows: 2092 o when the Proxy receives an RS message from a new ANET Client, it 2093 first authenticates the message then examines the network-layer 2094 destination address. If the destination address is a Server's 2095 AERO address, the Proxy proceeds to the next step. Otherwise, if 2096 the destination is all-routers multicast or subnet routers 2097 anycast, the Proxy selects a "nearby" Server that is likely to be 2098 a good candidate to serve the Client and replaces the destination 2099 address with the Server's AERO address. Next, the Proxy creates a 2100 proxy neighbor cache entry and caches the Client and Server 2101 addresses along with any identifying information including 2102 Transaction IDs, Client Identifiers, Nonce values, etc. The Proxy 2103 then examines the address in the first ifIndex in the RS message 2104 AERO option. The Proxy then converts the AERO option into an 2105 SLLAO, sets X to 1 and changes the Link Layer Address to its own 2106 SPAN address. The Proxy finally encapsulates the (proxyed) RS 2107 message in a SPAN header with destination set to the Server's SPAN 2108 address then forwards the message into the SPAN. 2110 o when the Server receives the RS message, it authenticates the 2111 message then creates or updates a symmetric neighbor cache entry 2112 for the Client with the Proxy's SPAN address as the link-layer 2113 address. The Server then sends an RA message back to the Proxy 2114 via the SPAN. 2116 o when the Proxy receives the RA message, it matches the message 2117 with the RS that created the proxy neighbor cache entry. The 2118 Proxy then caches the PD route information as a mapping from the 2119 Client's MNPs to the Client's ANET address, and sets the neighbor 2120 cache entry state to REACHABLE. The Proxy then converts the SLLAO 2121 in the RA message to an AERO option and forwards the (proxyed) 2122 message to the Client. 2124 After the initial RS/RA exchange, the Proxy forwards any Client data 2125 packets for which there is no matching asymmetric neighbor cache 2126 entry to a Relay via the SPAN. Finally, the Proxy forwards any 2127 Client data destined to an asymmetric neighbor cache target directly 2128 to the target according to the link-layer information - the process 2129 of establishing asymmetric neighbor cache entries is specified in 2130 Section 3.17. 2132 While the Client is still attached to the ANET, the Proxy send RS or 2133 unsolicited NA messages to update the Server's symmetric neighbor 2134 cache entries on behalf of the Client and/or to convey QoS updates. 2135 If the Server ceases to send solicited RA responses, the Proxy marks 2136 the Server as unreachable and sends an unsolicited RA with Router 2137 Lifetime set to zero to inform Clients that this Server is no longer 2138 able to provide service. Although the Proxy engages in ND exchanges 2139 on behalf of the Client, the Client can also send ND messages on its 2140 own behalf, e.g., if it is in a better position than the Proxy to 2141 convey QoS changes, etc. 2143 If the Client becomes unreachable, the Proxy sets the neighbor cache 2144 entry state to DEPARTED and retains the entry for DEPARTTIME seconds. 2145 While the state is DEPARTED, the Proxy forwards any packets destined 2146 to the Client to a Relay. The Relay in turn forwards the packets to 2147 the Client's current Server. When DepartTime expires, the Proxy 2148 deletes the neighbor cache entry and discards any further packets 2149 destined to this (now forgotten) Client. 2151 When a neighbor cache entry transitions to DEPARTED, some of the 2152 fragments of a multiple fragment packet may have already arrived at 2153 the Proxy while others are en route to the Client's new location. 2154 However, no special attention in the reassembly algorithm is 2155 necessary when re-routed packets are simply treated as loss. Since 2156 the fragments of a multiple-fragment packet are sent with minimal 2157 inter-packet delay, such occasions will be rare. 2159 In some ANETs that employ a Proxy, the Client's MNP can be injected 2160 into the ANET routing system. In that case, the Client can send data 2161 messages without encapsulation so that the ANET native routing system 2162 transports the unencapsulated packets to the Proxy. This can be very 2163 beneficial, e.g., if the Client connects to the ANET via low-end data 2164 links such as some aviation wireless links. 2166 If the first-hop ANET access router is AERO-aware, the Client can 2167 avoid encapsulation for both its control and data messages. When the 2168 Client connects to the link, it can send an unencapsulated RS message 2169 with source address set to its AERO address and with destination 2170 address set to the AERO address of the Client's selected Server or to 2171 all-routers multicast or subnet router anycast. The Client includes 2172 an AERO option formatted as specified in 2173 [I-D.templin-atn-aero-interface]. 2175 The Client then sends the unencapsulated RS message, which will be 2176 intercepted by the AERO-Aware access router. The access router then 2177 encapsulates the RS message in an ANET header with its own address as 2178 the source address and the address of a Proxy as the destination 2179 address. The access router further remembers the address of the 2180 Proxy so that it can encapsulate future data packets from the Client 2181 via the same Proxy. If the access router needs to change to a new 2182 Proxy, it simply sends another RS message toward the Server via the 2183 new Proxy on behalf of the Client. 2185 In some cases, the access router and Proxy may be one and the same 2186 node. In that case, the node would be located on the same physical 2187 link as the Client, but its message exchanges with the Server would 2188 need to pass through a security gateway at the ANET/INET border. The 2189 method for deploying access routers and Proxys (i.e. as a single node 2190 or multiple nodes) is an ANET-local administrative consideration. 2192 3.16.1. Detecting and Responding to Server Failures 2194 In environments where fast recovery from Server failure is required, 2195 Proxys SHOULD use Bidirectional Forwarding Detection (BFD) [RFC5880] 2196 to track Server reachability. Nodes that use BFD can quickly detect 2197 and react to failures so that cached information is re-established 2198 through alternate paths. BFD control messaging is carried only over 2199 well-connected ground domain networks (i.e., and not low-end 2200 aeronautical radio links) and can therefore be tuned for rapid 2201 response. 2203 Proxys establish BFD sessions with Servers for which there are 2204 currently active ANET Clients. If a Server fails, Proxys can quickly 2205 inform Clients by sending RA messages on the ANET interface. The 2206 Proxy sends RA messages with source address set to the Server's 2207 address, destination address set to all-nodes multicast, and Router 2208 Lifetime set to 0. 2210 The Proxy SHOULD send MAX_FINAL_RTR_ADVERTISEMENTS RA messages 2211 separated by small delays [RFC4861]. Any Clients on the ANET that 2212 have been using the (now defunct) Server will receive the RA messages 2213 and associate with a new Server. 2215 3.17. AERO Route Optimization 2217 While data packets are flowing between a source and target node, 2218 route optimization SHOULD be used. Route optimization is initiated 2219 by the first eligible Route Optimization Source (ROS) closest to the 2220 source as follows: 2222 o For Clients on VPNed, NATed and Direct interfaces, the Server is 2223 the ROS. 2225 o For Clients on Proxyed interfaces, the Proxy is the ROS. 2227 o For Clients on native interfaces, the Client itself is the ROS. 2229 o For correspondent nodes on INET/EUN interfaces serviced by a 2230 Gateway, the Gateway is the ROS. 2232 The route optimization procedure is conducted between the ROS and the 2233 target Server/Gateway acting as a Route Optimization Responder (ROR) 2234 in the same manner as for IPv6 ND Address Resolution and using the 2235 same NS/NA messaging. The target may either be a MNP Client serviced 2236 by a Server, or a non-MNP correspondent reachable via a Gateway. 2238 The procedures are specified in the following sections. 2240 3.17.1. Route Optimization Initiation 2242 While data packets are flowing from the source node toward a target 2243 node, the ROS performs address resolution by sending an NS message to 2244 receive a solicited NA message from the ROR. 2246 When the ROS sends an NS, it includes the AERO address of the ROS as 2247 the source address (e.g., fe80::1) and the AERO address corresponding 2248 to the data packet's destination address as the destination address 2249 (e.g., if the destination address is 2001:db8:1:2::1 then the 2250 corresponding AERO address is fe80::2001:db8:1:2). The NS message 2251 includes an SLLAO with Link Layer Address set to the SPAN address of 2252 the ROS and with all other fields set to 0. The message includes a 2253 Nonce and Timestamp option if the ROS needs to correlate NA replies. 2255 The ROS then encapsulates the NS message in a SPAN header with source 2256 set to its own SPAN address and destination set to the data packet's 2257 destination address, then sends it into the SPAN without decrementing 2258 the network-layer TTL/Hop Limit field. 2260 3.17.2. Relaying the NS 2262 When the Relay receives the NS message from the ROS, it discards the 2263 INET header and determines that the ROR is the next hop by consulting 2264 its standard IPv6 forwarding table for the SPAN header destination 2265 address. The Relay then forwards the SPAN message toward the ROR the 2266 same as for any IPv6 router. The final-hop Relay in the SPAN will 2267 deliver the message via a secured tunnel to the ROR. 2269 3.17.3. Processing the NS and Sending the NA 2271 When the ROR receives the NS message, it examines the AERO 2272 destination address to determine whether it has a neighbor cache 2273 entry and/or route that matches the target; if not, it drops the NS 2274 message and returns from processing. Next, if the target belongs to 2275 an MNP Client neighbor in the DEPARTED state the ROR changes the NS 2276 message SPAN destination address to the address of the Client's new 2277 Server, forwards the message into the SPAN and returns from 2278 processing. If the target belongs to an MNP Client neighbor in the 2279 REACHABLE state, the ROR instead adds the AERO source address to the 2280 target Client's Report List with time set to ReportTime. If the 2281 target belongs to a non-MNP route, the ROR continues processing 2282 without adding an entry to the Report List. 2284 The ROR then prepares a solicited NA message to send back to the ROS 2285 but does not create a neighbor cache entry. The ROR sets the NA 2286 source address to the destination AERO address of the NS, and 2287 includes the Nonce value received in the NS plus the current 2288 Timestamp. The ROR next includes a TLLAO with ifIndex set to 0, with 2289 all P(i) values set to "low", and with Link Layer Address set to the 2290 ROR's SPAN address. If the target belongs to an MNP Client, the ROR 2291 sets the Prefix Length to the MNP prefix length; otherwise, it sets 2292 Prefix Length to the maximum of the non-MNP prefix length and 64. 2293 (Note that a /64 limit is imposed to avoid causing the ROS to set 2294 short prefixes (e.g., "default") that would match destinations for 2295 which the routing system includes more-specific prefixes. Note also 2296 that prefix lengths longer than /64 are out of scope for this 2297 specification.) 2299 If the target belongs to an MNP Client, the ROR next includes 2300 additional ifIndexes for all of the target Client's underlying 2301 interfaces. For NATed, VPNed and Direct interfaces, the Link Layer 2302 Addresses are the SPAN address of the ROR. For Proxyed and native 2303 interfaces, the Link Layer Addresses are the SPAN addresses of the 2304 Proxys and the Client's native interfaces. The ROR finally 2305 encapsulates the NA message in a SPAN header with source set to its 2306 own SPAN address and destination set to the source SPAN address of 2307 the NS message, then forwards the message into the SPAN without 2308 decrementing the network-layer TTL/Hop Limit field. 2310 3.17.4. Relaying the NA 2312 When the Relay receives the NA message from the ROR, it discards the 2313 INET header and determines that the ROS is the next hop by consulting 2314 its standard IPv6 forwarding table for the SPAN header destination 2315 address. The Relay then forwards the SPAN-encapsulated NA message 2316 toward the ROS the same as for any IPv6 router. The final-hop Relay 2317 in the SPAN will deliver the message via a secured tunnel to the ROS. 2319 3.17.5. Processing the NA 2321 When the ROS receives the solicited NA message, it discards the INET 2322 and SPAN headers. The ROS next verifies the Nonce and Timestamp 2323 values, then creates an asymmetric neighbor cache entry for the ROR 2324 and caches all information found in the solicited NA TLLAOs. The ROS 2325 finally sets the asymmetric neighbor cache entry lifetime to 2326 ReachableTime seconds. 2328 3.17.6. Route Optimization Maintenance 2330 Following route optimization, the ROS forwards future data packets 2331 destined to the target via the addresses found in the cached link- 2332 layer information. The route optimization is shared by all sources 2333 that send packets to the target via the ROS, i.e., and not just the 2334 source on behalf of which the route optimization was initiated. 2336 While new data packets destined to the target are flowing through the 2337 ROS, it sends additional NS messages to the ROR before ReachableTime 2338 expires to receive a fresh solicited NA message the same as described 2339 in the previous sections. (Route optimization refreshment strategies 2340 are an implementation matter, with a non-normative example given in 2341 Appendix C.1). 2343 The ROS then updates the asymmetric neighbor cache entry to refresh 2344 ReachableTime, while (for MNP destinations) the ROR adds or updates 2345 the ROS address to the target Client's Report List and with time set 2346 to ReportTime. While no data packets are flowing, the ROS instead 2347 allows ReachableTime for the asymmetric neighbor cache entry to 2348 expire. When ReachableTime expires, the ROS deletes the asymmetric 2349 neighbor cache entry. Future data packets flowing through the ROS 2350 will again trigger a new route optimization exchange while initial 2351 data packets travel over a suboptimal route. 2353 The ROS may also receive unsolicited NA messages from the ROR at any 2354 time. If there is an asymmetric neighbor cache entry for the target, 2355 the ROS updates the link-layer information but does not update 2356 ReachableTime since the receipt of an unsolicited NA does not confirm 2357 that the forward path is still working. If there is no asymmetric 2358 neighbor cache entry, the ROS simply discards the unsolicited NA. 2359 Cases in which unsolicited NA messages are generated are specified in 2360 Section 3.19. 2362 In this arrangement, the ROS holds an asymmetric neighbor cache entry 2363 for the ROR, but the ROR does not hold an asymmetric neighbor cache 2364 entry for the ROS. The route optimization neighbor relationship is 2365 therefore asymmetric and unidirectional. If the target node also has 2366 packets to send back to the source node, then a separate route 2367 optimization procedure is performed in the reverse direction. But, 2368 there is no requirement that the forward and reverse paths be 2369 symmetric. 2371 3.18. Neighbor Unreachability Detection (NUD) 2373 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2374 [RFC4861]. NUD is performed either reactively in response to 2375 persistent link-layer errors (see Section 3.14) or proactively to 2376 confirm reachability. The NUD algorithm may further be seeded by ND 2377 hints of forward progress, but care must be taken to avoid inferring 2378 reachability based on spoofed information. 2380 When an ROR directs an ROS to a neighbor with one or more target 2381 link-layer addresses, the ROS can proactively test each direct path 2382 by sending an initial NS message to elicit a solicited NA response. 2383 While testing the paths, the ROS can optionally continue sending 2384 packets via the SPAN, maintain a small queue of packets until target 2385 reachability is confirmed, or (optimistically) allow packets to flow 2386 via the direct paths. In any case, the ROS should only consider the 2387 neighbor unreachable if NUD fails over multiple target link-layer 2388 address paths. 2390 When a ROS sends an NS message used for NUD, it uses its AERO 2391 addresses as the IPv6 source address and the AERO address 2392 corresponding to a target link-layer address as the destination. For 2393 each target link-layer address, the source node encapsulates the NS 2394 message in SPAN/INET headers with its own SPAN address as the source 2395 and the SPAN address of the target as the destination, If the target 2396 is located within the same SPAN segment, the source sets the INET 2397 address of the target as the destination; otherwise, it sets the INET 2398 address of a Relay as the destination. The source then forwards the 2399 message into the SPAN. 2401 Paths that pass NUD tests are marked as "reachable", while those that 2402 do not are marked as "unreachable". These markings inform the AERO 2403 interface forwarding algorithm specified in Section 3.12. 2405 Proxys can perform NUD to verify Server reachability on behalf of 2406 their proxyed Clients so that the Clients need not engage in NUD 2407 messaging themselves. 2409 3.19. Mobility Management and Quality of Service (QoS) 2411 AERO is a Distributed Mobility Management (DMM) service. Each Server 2412 is responsible for only a subset of the Clients on the AERO link, as 2413 opposed to a Centralized Mobility Management (CMM) service where 2414 there is a single network mobility service for all Clients. Clients 2415 coordinate with their associated Servers via RS/RA exchanges to 2416 maintain the DMM profile, and the AERO routing system tracks all 2417 current Client/Server peering relationships. 2419 Servers provide a Mobility Anchor Point (MAP) for their dependent 2420 Clients. Clients are responsible for maintaining neighbor 2421 relationships with their Servers through periodic RS/RA exchanges, 2422 which also serves to confirm neighbor reachability. When a Client's 2423 underlying interface address and/or QoS information changes, the 2424 Client is responsible for updating the Server with this new 2425 information. Note that for Proxyed interfaces, however, the Proxy 2426 can perform the RS/RA exchanges on the Client's behalf. 2428 Mobility management considerations are specified in the following 2429 sections. 2431 3.19.1. Mobility Update Messaging 2433 Servers acting as MAPs accommodate Client mobility and/or QoS change 2434 events by sending unsolicited NA messages to each ROS in the target 2435 Client's Report List. When a MAP sends an unsolicited NA message, it 2436 sets the IPv6 source address to the Client's AERO address and sets 2437 the IPv6 destination address to all-nodes multicast (ff02::1). The 2438 MAP also includes a TLLAO with a first ifIndex set to 0 and Link 2439 Layer address set to the MAP's SPAN address, and includes additional 2440 ifIndexes for all of the target Client's interfaces with Link Layer 2441 Addresses set to the corresponding SPAN addresses. The MAP finally 2442 encapsulates the message in a SPAN header with source set to its own 2443 SPAN address and destination set to the SPAN address of the ROS, then 2444 sends the message to a Relay in the SPAN. 2446 As for the hot-swap of interface cards discussed in Section 7.2.6 of 2447 [RFC4861], the transmission and reception of unsolicited NA messages 2448 is unreliable but provides a useful optimization. In well-connected 2449 Internetworks with robust data links unsolicited NA messages will be 2450 delivered with high probability, but in any case the MAP can 2451 optionally send up to MAX_NEIGHBOR_ADVERTISEMENT unsolicited NAs to 2452 each ROS to increase the likelihood that at least one will be 2453 received. 2455 When an ROS receives an unsolicited NA message, it ignores the 2456 message if there is no existing neighbor cache entry for the Client. 2458 Otherwise, it uses the included TLLAO to update the Link Layer 2459 Address and QoS information in the neighbor cache entry, but does not 2460 reset ReachableTime since the receipt of an unsolicited NA message 2461 from the target Server does not provide confirmation that any forward 2462 paths to the target Client are working. 2464 If unsolicited NA messages are lost, the ROS may be left with stale 2465 address and/or QoS information for the Client for up to ReachableTime 2466 seconds. During this time, the ROS can continue sending packets 2467 according to its stale neighbor cache information. When 2468 ReachableTime is close to expiring, the ROS will re-initiate route 2469 optimization and receive fresh state information. 2471 In addition to sending unsolicited NA messages to the current set of 2472 ROSs for the Client, the MAP also sends unsolicited NAs to the former 2473 Link-Layer address for any ifIndex for which the Link-Layer address 2474 has changed. The NA messages update Proxys or Servers that cannot 2475 easily detect (e.g., without active probing) when a formerly-active 2476 Client has departed. 2478 3.19.2. Announcing Link-Layer Address and/or QoS Preference Changes 2480 When a Client needs to change its ANET addresses and/or QoS 2481 preferences (e.g., due to a mobility event), either the Client or its 2482 Proxys send RS messages to the Server via the SPAN with SLLAOs that 2483 include the new Client Port Number, Link Layer Address and P(i) 2484 values. If the RS messages are sent solely for the purpose of 2485 updating QoS preferences, Port Number and Link-Layer Address are set 2486 to 0. 2488 Up to MAX_RTR_SOLICITATION RS messages MAY be sent in parallel with 2489 sending actual data packets in case one or more RAs are lost. If all 2490 RAs are lost, the Client SHOULD re-associate with a new Server. 2492 When the Server receives the Client's changes, it sends unsolicited 2493 NA messages to all nodes in the Report List the same as described in 2494 the previous section. 2496 3.19.3. Bringing New Links Into Service 2498 When a Client needs to bring new underlying interfaces into service 2499 (e.g., when it activates a new data link), it sends an RS message to 2500 its Server via the underlying interface with SLLAOs that include the 2501 new Client Link Layer Address information. 2503 3.19.4. Removing Existing Links from Service 2505 When a Client needs to remove existing underlying interfaces from 2506 service (e.g., when it de-activates an existing data link), it sends 2507 an RS message to its Server with SLLAO ifIndexes with Link values set 2508 to 0. 2510 If the Client needs to send RS messages over an underlying interface 2511 other than the one being removed from service, it MUST include an 2512 SLLAO ifIndex for the sending interface as the first ifIndex and 2513 include additional ifIndexes with Link values set to 0 for any 2514 underlying interfaces being removed from service. 2516 3.19.5. Moving to a New Server 2518 When a Client associates with a new Server, it performs the Client 2519 procedures specified in Section 3.15.2. The Client then sends an RS 2520 message over any working underlying interface with destination set to 2521 the old Server's AERO address and with an SLLAO with R set to 0 to 2522 fully release itself from the old Server. The SLLAO also includes 2523 the SPAN address of the new Server in the Link Layer Address. If the 2524 Client does not receive an RA reply after MAX_RTR_SOLICITATIONS 2525 attempts over multiple underlying interfaces, the old Server may have 2526 failed and the Client should discontinue its release attempts. 2528 When the old Server processes the RS, it sends unsolicited NA 2529 messages with a TLLAO with ifIndex set to 0 and with R set to 0 to 2530 all ROSs in the Client's Report List. The Server also changes the 2531 symmetric neighbor cache entry state to DEPARTED, sets the link-layer 2532 address of the Client to the address found in the RS SLLAO (i.e., the 2533 SPAN address of the new Server), and sets a timer to DEPARTTIME 2534 seconds. The old Server then returns an immediate RA message to the 2535 Client with Router Lifetime set to 0 and withdraws the Client's MNP 2536 from the routing system. After DepartTime expires, the old Server 2537 deletes the symmetric neighbor cache entry. 2539 Clients SHOULD NOT move rapidly between Servers in order to avoid 2540 causing excessive oscillations in the AERO routing system. Examples 2541 of when a Client might wish to change to a different Server include a 2542 Server that has gone unreachable, topological movements of 2543 significant distance, movement to a new geographic region, movement 2544 to a new SPAN segment, etc. 2546 When a Client moves to a new Server, some of the fragments of a 2547 multiple fragment packet may have already arrived at the old Server 2548 while others are en route to the new Server. However, no special 2549 attention in the reassembly algorithm is necessary when re-routed 2550 packets are simply treated as loss. Since the fragments of a 2551 multiple-fragment packet are sent with minimal inter-packet delay, 2552 such occasions will be rare. 2554 3.20. Multicast 2556 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 2557 [RFC3810] proxy service for its EUNs and/or hosted applications 2558 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 2559 underlying interfaces for which group membership is required. The 2560 IGMP/MLD messages may be further forwarded by a first-hop ANET access 2561 router acting as an IGMP/MLD-snooping switch [RFC4541], then 2562 ultimately delivered to an AERO Proxy/Server acting as a Protocol 2563 Independent Multicast - Sparse-Mode (PIM-SM, or simply "PIM") 2564 Designated Router (DR) [RFC7761]. AERO Gateways also act as PIM 2565 routers (i.e., the same as AERO Proxys/Servers) on behalf of nodes on 2566 INET/EUN networks. The behaviors identified in the following 2567 sections correspond to Source-Specific Multicast (SSM) and Any-Source 2568 Multicast (ASM) operational modes. 2570 3.20.1. Source-Specific Multicast (SSM) 2572 When an ROS (i.e., an AERO Proxy/Server/Gateway) "X" acting as PIM 2573 router receives a Join/Prune message from a node on its downstream 2574 interfaces containing one or more ((S)ource, (G)roup) pairs, it 2575 updates its Multicast Routing Information Base (MRIB) accordingly. 2576 For each S belonging to a prefix reachable via X's non-AERO 2577 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 2578 on those interfaces per [RFC7761]. 2580 For each S belonging to a prefix reachable via X's AERO interface, X 2581 originates a separate copy of the Join/Prune for each (S,G) in the 2582 message using its own AERO address as the source address and ALL-PIM- 2583 ROUTERS as the destination address. X then encapsulates each message 2584 in a SPAN header with source address set to the SPAN address of X and 2585 destination address set to S then forwards the message into the SPAN. 2586 The SPAN in turn forwards the message to AERO Server/Gateway "Y" that 2587 services S. At the same time, if the message was a Join, X sends a 2588 route-optimization NS message toward each S the same as discussed in 2589 Section 3.17. The resulting NAs will return the AERO address for the 2590 prefix that matches S as the network-layer source address and TLLAOs 2591 with the SPAN addresses corresponding to any ifIndexes that are 2592 currently servicing S. 2594 When Y processes the Join/Prune message, if S located behind any 2595 Native, Direct, VPNed or NATed interfaces Y acts as a PIM router and 2596 updates its MRIB to list X as the next hop in the reverse path. If S 2597 is located behind any Proxys "Z"*, Y also forwards the message to 2598 each Z* over the SPAN while continuing to use the AERO address of X 2599 as the source address. Each Z* then updates its MRIB accordingly and 2600 maintains the AERO address of X as the next hop in the reverse path. 2601 Since the Relays in the SPAN do not examine network layer control 2602 messages, this means that the (reverse) multicast tree path is simply 2603 from each Z* (and/or Y) to X with no other multicast-aware routers in 2604 the path. If any Z* (and/or Y) is located on the same SPAN segment 2605 as X, the multicast data traffic sent to X directly using SPAN/INET 2606 encapsulation instead of via a Relay. 2608 Following the initial Join/Prune and NS/NA messaging, X maintains an 2609 asymmetric neighbor cache entry for each S the same as if X was 2610 sending unicast data traffic to S. In particular, X performs 2611 additional NS/NA exchanges to keep the neighbor cache entry alive for 2612 up to t_periodic seconds [RFC7761]. If no new Joins are received 2613 within t_periodic seconds, X allows the neighbor cache entry to 2614 expire. Finally, if X receives any additional Join/Prune messages 2615 for (S,G) it forwards the messages to each Y and Z* in the neighbor 2616 cache entry over the SPAN. 2618 At some later time, Client C that holds an MNP for source S may 2619 depart from a first Proxy Z1 and/or connect via a new Proxy Z2. In 2620 that case, Y sends an unsolicited NA message to X the same as 2621 specified for unicast mobility in Section 3.19. When X receives the 2622 unsolicited NA message, it updates its asymmetric neighbor cache 2623 entry for the AERO address for source S and sends new Join messages 2624 to any new Proxys Z2. There is no requirement to send any Prune 2625 messages to old Proxys Z1 since source S will no longer source any 2626 multicast data traffic via Z1. Instead, the multicast state for 2627 (S,G) in Proxy Z1 will soon time out since no new Joins will arrive. 2629 After some later time, C may move to a new Server Y2 and depart from 2630 old Sever Y1. In that case, Y1 sends Join messages for any of C's 2631 active (S,G) groups to Y2 while including its own AERO address as the 2632 source address. This causes Y2 to include Y1 in the multicast 2633 forwarding tree during the interim time that Y1's symmetric neighbor 2634 cache entry for C is in the DEPARTED state. At the same time, Y1 2635 sends an unsolicited NA message to X with a TLLAO with ifIndex set to 2636 0 and R set to 1 to cause X to release its asymmetric neighbor cache 2637 entry. X then sends a new Join message to S via the SPAN and re- 2638 initiates route optimization the same as if it were receiving a fresh 2639 Join message from a node on a downstream link. 2641 3.20.2. Any-Source Multicast (ASM) 2643 When an ROS X acting as a PIM router receives a Join/Prune from a 2644 node on its downstream interfaces containing one or more (*,G) pairs, 2645 it updates its Multicast Routing Information Base (MRIB) accordingly. 2646 X then forwards a copy of the message to the Rendezvous Point (RP) R 2647 for each G over the SPAN. X uses its own AERO address as the source 2648 address and ALL-PIM-ROUTERS as the destination address, then 2649 encapsulates each message in a SPAN header with source address set to 2650 the SPAN address of X and destination address set to R, then sends 2651 the message into the SPAN. At the same time, if the message was a 2652 Join X initiates NS/NA route optimization the same as for the SSM 2653 case discussed in Section 3.20.1. 2655 For each source S that sends multicast traffic to group G via R, the 2656 Proxy/Server Z* for the Client that aggregates S encapsulates the 2657 packets in PIM Register messages and forwards them to R via the SPAN. 2658 R may then elect to send a PIM Join to Z* over the SPAN. This will 2659 result in an (S,G) tree rooted at Z* with R as the next hop so that R 2660 will begin to receive two copies of the packet; one native copy from 2661 the (S, G) tree and a second copy from the pre-existing (*, G) tree 2662 that still uses PIM Register encapsulation. R can then issue a PIM 2663 Register-stop message to suppress the Register-encapsulated stream. 2664 At some later time, if C moves to a new Proxy/Server Z*, it resumes 2665 sending packets via PIM Register encapsulation via the new Z*. 2667 At the same time, as multicast listeners discover individual S's for 2668 a given G, they can initiate an (S,G) Join for each S under the same 2669 procedures discussed in Section 3.20.1. Once the (S,G) tree is 2670 established, the listeners can send (S, G) Prune messages to R so 2671 that multicast packets for group G sourced by S will only be 2672 delivered via the (S, G) tree and not from the (*, G) tree rooted at 2673 R. All mobility considerations discussed for SSM apply. 2675 3.20.3. Bi-Directional PIM (BIDIR-PIM) 2677 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 2678 approach to ASM that treats the Rendezvous Point (RP) as a Designated 2679 Forwarder (DF). Further considerations for BIDIR-PIM are out of 2680 scope. 2682 3.21. Operation over Multiple AERO Links (VLANs) 2684 An AERO Client can connect to multiple AERO links the same as for any 2685 data link service. In that case, the Client maintains a distinct 2686 AERO interface for each link, e.g., 'aero0' for the first link, 2687 'aero1' for the second, 'aero2' for the third, etc. Each AERO link 2688 would include its own distinct set of Relays, Servers and Proxys, 2689 thereby providing redundancy in case of failures. 2691 The Relays, Servers and Proxys on each AERO link can assign AERO and 2692 SPAN addresses that use the same or different numberings from those 2693 on other links. Since the links are mutually independent there is no 2694 requirement for avoiding inter-link address duplication, e.g., the 2695 same AERO address such as fe80::1000 could be used to number distinct 2696 nodes that connect to different links. 2698 Each AERO link could utilize the same or different ANET connections. 2699 The links can be distinguished at the link-layer via Virtual Local 2700 Area Network (VLAN) tagging (e.g., IEEE 802.1Q) and/or through 2701 assignment of distinct sets of MSPs on each link. This gives rise to 2702 the opportunity for supporting multiple redundant networked paths, 2703 where each VLAN is distinguished by a different label (e.g., colors 2704 such as Red, Green, Blue, etc.). In particular, the Client can tag 2705 its RS messages with the appropriate label to cause the network to 2706 select the desired VLAN. 2708 Clients that connect to multiple AERO interfaces can select the 2709 outgoing interface appropriate for a given Red/Blue/Green/etc. 2710 traffic profile while (in the reverse direction) correspondent nodes 2711 must have some way of steering their packets destined to a target via 2712 the correct AERO link. 2714 In a first alternative, if each AERO link services different MSPs, 2715 then the Client can receive a distinct MNP from each of the links. 2716 IP routing will therefore assure that the correct Red/Green/Blue/etc. 2717 network is used for both outbound and inbound traffic. This can be 2718 accomplished using existing technologies and approaches, and without 2719 requiring any special supporting code in correspondent nodes or 2720 Relays. 2722 In a second alternative, if each AERO link services the same MSP(s) 2723 then each link could assign a distinct "AERO Link Anycast" address 2724 that is configured by all Relays on the link. Correspondent nodes 2725 then include a "type 4" routing header with the Anycast address for 2726 the AERO link as the IPv6 destination and with the address of the 2727 target encoded as the "next segment" in the routing header 2728 [RFC8402][I-D.ietf-6man-segment-routing-header]. Standard IP routing 2729 will then direct the packet to the nearest Relay for the correct AERO 2730 link, which will replace the destination address with the target 2731 address then forward the packet to the target. 2733 3.22. DNS Considerations 2735 AERO Client MNs and INET correspondent nodes consult the Domain Name 2736 System (DNS) the same as for any Internetworking node. When 2737 correspondent nodes and Client MNs use different IP protocol versions 2738 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 2739 A records for IPv4 address mappings to MNs which must then be 2740 populated in Gateway NAT64 mapping caches. In that way, an IPv4 2741 correspondent node can send packets to the IPv4 address mapping of 2742 the target MN, and the Gateway will translate the IPv4 header and 2743 destination address into an IPv6 header and IPv6 destination address 2744 of the MN. 2746 When an AERO Client registers with an AERO Server, the Server returns 2747 the address(es) of DNS servers in RDNSS options [RFC6106]. The DNS 2748 server provides the IP addresses of other MNs and correspondent nodes 2749 in AAAA records for IPv6 or A records for IPv4. 2751 3.23. Transition Considerations 2753 The SPAN ensures that dissimilar INET partitions can be joined into a 2754 single unified AERO link, even though the partitions themselves may 2755 have differing protocol versions and/or incompatible addressing 2756 plans. However, a commonality can be achieved by incrementally 2757 distributing globally routable (i.e., native) IP prefixes to 2758 eventually reach all nodes (both mobile and fixed) in all SPAN 2759 segments. This can be accomplished by incrementally deploying AERO 2760 Gateways on each INET partition, with each Gateway distributing its 2761 MNPs and/or discovering non-MNP prefixes on its INET links. 2763 This gives rise to the opportunity to eventually distribute native IP 2764 addresses to all nodes, and to present a unified AERO link view 2765 (bridged by the SPAN) even if the INET partitions remain in their 2766 current protocol and addressing plans. In that way, the AERO link 2767 can serve the dual purpose of providing a mobility service and a 2768 transition service. Or, if an INET partition is transitioned to a 2769 native IP protocol version and addressing scheme that is compatible 2770 with the AERO link MNP-based addressing scheme, the partition and 2771 AERO link can be joined by Gateways. 2773 Gateways that connect INETs/EUNs with dissimilar IP protocol versions 2774 must employ a network address and protocol translation function such 2775 as NAT64[RFC6146]. 2777 3.24. Detecting and Reacting to Proxy, Server and Relay Failures 2779 In environments where rapid failure recovery is required, Proxys, 2780 Servers and Relays SHOULD use Bidirectional Forwarding Detection 2781 (BFD) [RFC5880]. Nodes that use BFD can quickly detect and react to 2782 failures so that cached information is re-established through 2783 alternate nodes. BFD control messaging is carried only over well- 2784 connected ground domain networks (i.e., and not low-end radio links) 2785 and can therefore be tuned for rapid response. 2787 Servers and Relays maintain BFD sessions in parallel with their BGP 2788 peerings. If a Server or Relay fails, BGP peers will quickly re- 2789 establish routes through alternate paths the same as for common BGP 2790 deployments. 2792 Proxys establish BFD sessions with Servers for which there are 2793 currently active ANET Clients. If a Proxy fails, Servers MAY clear 2794 their cached Client link-layer information associated with the Proxy. 2795 If a Server fails, Proxys quickly inform Clients of the outage by 2796 sending RA messages on the ANET interface. 2798 The Proxy sends RA messages with source address set to the Server's 2799 address, destination address set to all-nodes multicast, and Router 2800 Lifetime set to 0. The Proxy SHOULD send 2801 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2802 [RFC4861]. Any Clients on the ANET interface that have been using 2803 the (now defunct) Server will receive the RA messages and associate 2804 with a new Server. 2806 4. Implementation Status 2808 An AERO implementation based on OpenVPN (https://openvpn.net/) was 2809 announced on the v6ops mailing list on January 10, 2018 and an 2810 initial public release of the AERO proof-of-concept source code was 2811 announced on the intarea mailing list on August 21, 2015. The latest 2812 versions are available at: http://linkupnetworks.net/aero. 2814 5. IANA Considerations 2816 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 2817 AERO in the "enterprise-numbers" registry. 2819 The IANA has assigned the UDP port number "8060" for an earlier 2820 experimental version of AERO [RFC6706]. This document obsoletes 2821 [RFC6706] and claims the UDP port number "8060" for all future use. 2823 No further IANA actions are required. 2825 6. Security Considerations 2827 AERO Relays configure secured tunnels with AERO Servers and Proxys 2828 within their local SPAN segments. Applicable secured tunnel 2829 alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 2830 [RFC6347], etc. The AERO Relays of all SPAN segments in turn 2831 configure secured tunnels for their neighboring AERO Relays across 2832 the SPAN. Therefore, packets that traverse the SPAN between any pair 2833 of AERO link neighbors are already secured. 2835 AERO Servers, Gateways and Proxys targeted by a route optimization 2836 may also receive packets directly from the INET partitions instead of 2837 via the SPAN. For INET partitions that apply effective ingress 2838 filtering to defeat source address spoofing, the simple data origin 2839 authentication procedures in Section 3.11 can be applied. This 2840 implies that the ROS list must be maintained consistently by all 2841 route optimization targets within the same INET partition, and that 2842 the ROS list must be securely managed by the partition's 2843 administrative authority. 2845 For INET partitions that cannot apply effective ingress filtering, 2846 the two options for securing communications include 1) disable route 2847 optimization so that all traffic is conveyed over secured tunnels via 2848 the SPAN, or 2) enable on-demand secure tunnel creation between INET 2849 partition neighbors. Option 1) would result in longer routes than 2850 necessary and traffic concentration on critical infrastructure 2851 elements. Option 2) could be coordinated by establishing a secured 2852 tunnel on-demand instead of performing an NS/NA exchange in the route 2853 optimization procedures. Procedures for establishing on-demand 2854 secured tunnels are out of scope. 2856 AERO Clients that connect to secured enclaves need not apply security 2857 to their ND messages, since the messages will be intercepted by a 2858 perimeter Proxy that applies security on its outward-facing 2859 interface. AERO Clients located outside of secured enclaves SHOULD 2860 use symmetric network and/or transport layer security services, but 2861 when there are many prospective neighbors with dynamically changing 2862 connectivity an asymmetric security service such as SEND may be 2863 needed (see: Appendix C.6). 2865 Application endpoints SHOULD use application-layer security services 2866 such as TLS/SSL, DTLS or SSH [RFC4251] to assure the same level of 2867 protection as for critical secured Internet services. AERO Clients 2868 that require host-based VPN services SHOULD use symmetric network 2869 and/or transport layer security services such as IPsec, TLS/SSL, 2870 DTLS, etc. AERO Proxys and Servers can also provide a network-based 2871 VPN service on behalf of the Client, e.g., if the Client is located 2872 within a secured enclave and cannot establish a VPN on its own 2873 behalf. 2875 AERO Servers and Relays present targets for traffic amplification 2876 Denial of Service (DoS) attacks. This concern is no different than 2877 for widely-deployed VPN security gateways in the Internet, where 2878 attackers could send spoofed packets to the gateways at high data 2879 rates. This can be mitigated by connecting Servers and Relays over 2880 dedicated links with no connections to the Internet and/or when 2881 connections to the Internet are only permitted through well-managed 2882 firewalls. Traffic amplification DoS attacks can also target an AERO 2883 Client's low data rate links. This is a concern not only for Clients 2884 located on the open Internet but also for Clients in secured 2885 enclaves. AERO Servers and Proxys can institute rate limits that 2886 protect Clients from receiving packet floods that could DoS low data 2887 rate links. 2889 AERO Gateways must implement ingress filtering to avoid a spoofing 2890 attack in which spurious SPAN messages are injected into an AERO link 2891 from an outside attacker. AERO Clients MUST ensure that their 2892 connectivity is not used by unauthorized nodes on their EUNs to gain 2893 access to a protected network, i.e., AERO Clients that act as routers 2894 MUST NOT provide routing services for unauthorized nodes. (This 2895 concern is no different than for ordinary hosts that receive an IP 2896 address delegation but then "share" the address with other nodes via 2897 some form of Internet connection sharing such as tethering.) 2899 The MAP list and ROS lists MUST be well-managed and secured from 2900 unauthorized tampering, even though the list contains only public 2901 information. The MAP list can be conveyed to the Client in a similar 2902 fashion as in [RFC5214] (e.g., through layer 2 data link login 2903 messaging, secure upload of a static file, DNS lookups, etc.). The 2904 ROS list can be conveyed to Servers and Proxys through administrative 2905 action, secured file distribution, etc. 2907 Although public domain and commercial SEND implementations exist, 2908 concerns regarding the strength of the cryptographic hash algorithm 2909 have been documented [RFC6273] [RFC4982]. 2911 Security considerations for accepting link-layer ICMP messages and 2912 reflected packets are discussed throughout the document. 2914 7. Acknowledgements 2916 Discussions in the IETF, aviation standards communities and private 2917 exchanges helped shape some of the concepts in this work. 2918 Individuals who contributed insights include Mikael Abrahamsson, Mark 2919 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 2920 Wojciech Dec, Ralph Droms, Adrian Farrel, Nick Green, Sri Gundavelli, 2921 Brian Haberman, Bernhard Haindl, Joel Halpern, Tom Herbert, Sascha 2922 Hlusiak, Lee Howard, Andre Kostur, Hubert Kuenig, Ted Lemon, Andy 2923 Malis, Satoru Matsushima, Tomek Mrugalski, Madhu Niraula, Alexandru 2924 Petrescu, Behcet Saikaya, Michal Skorepa, Joe Touch, Bernie Volz, 2925 Ryuji Wakikawa, Tony Whyman, Lloyd Wood and James Woodyatt. Members 2926 of the IESG also provided valuable input during their review process 2927 that greatly improved the document. Special thanks go to Stewart 2928 Bryant, Joel Halpern and Brian Haberman for their shepherding 2929 guidance during the publication of the AERO first edition. 2931 This work has further been encouraged and supported by Boeing 2932 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 2933 Brodie, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, Claudiu 2934 Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Anthony Gregory, 2935 Jeff Holland, Seth Jahne, Ed King, Laurel Matthew, Gene MacLean III, 2936 Rob Muszkiewicz, Sean O'Sullivan, Greg Saccone, Kent Shuey, Brian 2937 Skeen, Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Julie 2938 Wulff, Yueli Yang, Eric Yeh and other members of the BR&T and BIT 2939 mobile networking teams. Kyle Bae, Wayne Benson, Katie Tran and Eric 2940 Yeh are especially acknowledged for implementing the AERO functions 2941 as extensions to the public domain OpenVPN distribution. 2943 Earlier works on NBMA tunneling approaches are found in 2944 [RFC2529][RFC5214][RFC5569]. 2946 Many of the constructs presented in this second edition of AERO are 2947 based on the author's earlier works, including: 2949 o The Internet Routing Overlay Network (IRON) 2950 [RFC6179][I-D.templin-ironbis] 2952 o Virtual Enterprise Traversal (VET) 2953 [RFC5558][I-D.templin-intarea-vet] 2955 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 2956 [RFC5320][I-D.templin-intarea-seal] 2958 o AERO, First Edition [RFC6706] 2960 Note that these works cite numerous earlier efforts that are not also 2961 cited here due to space limitations. The authors of those earlier 2962 works are acknowledged for their insights. 2964 This work is aligned with the NASA Safe Autonomous Systems Operation 2965 (SASO) program under NASA contract number NNA16BD84C. 2967 This work is aligned with the FAA as per the SE2025 contract number 2968 DTFAWA-15-D-00030. 2970 This work is aligned with the Boeing Commercial Airplanes (BCA) 2971 Internet of Things (IoT) and autonomy programs. 2973 This work is aligned with the Boeing Information Technology (BIT) 2974 MobileNet program. 2976 8. References 2978 8.1. Normative References 2980 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 2981 DOI 10.17487/RFC0791, September 1981, 2982 . 2984 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 2985 RFC 792, DOI 10.17487/RFC0792, September 1981, 2986 . 2988 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2989 Requirement Levels", BCP 14, RFC 2119, 2990 DOI 10.17487/RFC2119, March 1997, 2991 . 2993 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 2994 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 2995 December 1998, . 2997 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 2998 "Definition of the Differentiated Services Field (DS 2999 Field) in the IPv4 and IPv6 Headers", RFC 2474, 3000 DOI 10.17487/RFC2474, December 1998, 3001 . 3003 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3004 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3005 DOI 10.17487/RFC3971, March 2005, 3006 . 3008 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3009 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3010 . 3012 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3013 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3014 November 2005, . 3016 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3017 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3018 DOI 10.17487/RFC4861, September 2007, 3019 . 3021 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3022 Address Autoconfiguration", RFC 4862, 3023 DOI 10.17487/RFC4862, September 2007, 3024 . 3026 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3027 (IPv6) Specification", STD 86, RFC 8200, 3028 DOI 10.17487/RFC8200, July 2017, 3029 . 3031 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3032 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3033 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3034 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3035 . 3037 8.2. Informative References 3039 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3040 2016. 3042 [I-D.ietf-6man-segment-routing-header] 3043 Filsfils, C., Dukes, D., Previdi, S., Leddy, J., 3044 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3045 (SRH)", draft-ietf-6man-segment-routing-header-26 (work in 3046 progress), October 2019. 3048 [I-D.ietf-dmm-distributed-mobility-anchoring] 3049 Chan, A., Wei, X., Lee, J., Jeon, S., and C. Bernardos, 3050 "Distributed Mobility Anchoring", draft-ietf-dmm- 3051 distributed-mobility-anchoring-14 (work in progress), 3052 November 2019. 3054 [I-D.ietf-intarea-gue] 3055 Herbert, T., Yong, L., and O. Zia, "Generic UDP 3056 Encapsulation", draft-ietf-intarea-gue-09 (work in 3057 progress), October 2019. 3059 [I-D.ietf-intarea-gue-extensions] 3060 Herbert, T., Yong, L., and F. Templin, "Extensions for 3061 Generic UDP Encapsulation", draft-ietf-intarea-gue- 3062 extensions-06 (work in progress), March 2019. 3064 [I-D.ietf-intarea-tunnels] 3065 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3066 Architecture", draft-ietf-intarea-tunnels-10 (work in 3067 progress), September 2019. 3069 [I-D.ietf-rtgwg-atn-bgp] 3070 Templin, F., Saccone, G., Dawra, G., Lindem, A., and V. 3071 Moreno, "A Simple BGP-based Mobile Routing System for the 3072 Aeronautical Telecommunications Network", draft-ietf- 3073 rtgwg-atn-bgp-04 (work in progress), December 2019. 3075 [I-D.templin-6man-dhcpv6-ndopt] 3076 Templin, F., "A Unified Stateful/Stateless Configuration 3077 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-08 3078 (work in progress), June 2019. 3080 [I-D.templin-atn-aero-interface] 3081 Templin, F. and T. Whyman, "Transmission of IPv6 Packets 3082 over Aeronautical ("aero") Interfaces", draft-templin-atn- 3083 aero-interface-08 (work in progress), December 2019. 3085 [I-D.templin-intarea-grefrag] 3086 Templin, F., "GRE Tunnel Level Fragmentation", draft- 3087 templin-intarea-grefrag-04 (work in progress), July 2016. 3089 [I-D.templin-intarea-seal] 3090 Templin, F., "The Subnetwork Encapsulation and Adaptation 3091 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 3092 progress), January 2014. 3094 [I-D.templin-intarea-vet] 3095 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 3096 templin-intarea-vet-40 (work in progress), May 2013. 3098 [I-D.templin-ironbis] 3099 Templin, F., "The Interior Routing Overlay Network 3100 (IRON)", draft-templin-ironbis-16 (work in progress), 3101 March 2014. 3103 [I-D.templin-v6ops-pdhost] 3104 Templin, F., "IPv6 Prefix Delegation and Multi-Addressing 3105 Models", draft-templin-v6ops-pdhost-24 (work in progress), 3106 June 2019. 3108 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3110 [RFC1035] Mockapetris, P., "Domain names - implementation and 3111 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3112 November 1987, . 3114 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 3115 Communication Layers", STD 3, RFC 1122, 3116 DOI 10.17487/RFC1122, October 1989, 3117 . 3119 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 3120 DOI 10.17487/RFC1191, November 1990, 3121 . 3123 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3124 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3125 . 3127 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3128 DOI 10.17487/RFC2003, October 1996, 3129 . 3131 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3132 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3133 . 3135 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3136 Domains without Explicit Tunnels", RFC 2529, 3137 DOI 10.17487/RFC2529, March 1999, 3138 . 3140 [RFC2764] Gleeson, B., Lin, A., Heinanen, J., Armitage, G., and A. 3141 Malis, "A Framework for IP Based Virtual Private 3142 Networks", RFC 2764, DOI 10.17487/RFC2764, February 2000, 3143 . 3145 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 3146 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 3147 DOI 10.17487/RFC2784, March 2000, 3148 . 3150 [RFC2890] Dommety, G., "Key and Sequence Number Extensions to GRE", 3151 RFC 2890, DOI 10.17487/RFC2890, September 2000, 3152 . 3154 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", 3155 RFC 2923, DOI 10.17487/RFC2923, September 2000, 3156 . 3158 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3159 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3160 . 3162 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3163 of Explicit Congestion Notification (ECN) to IP", 3164 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3165 . 3167 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3168 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3169 DOI 10.17487/RFC3810, June 2004, 3170 . 3172 [RFC3819] Karn, P., Ed., Bormann, C., Fairhurst, G., Grossman, D., 3173 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 3174 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 3175 RFC 3819, DOI 10.17487/RFC3819, July 2004, 3176 . 3178 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 3179 for IPv6 Hosts and Routers", RFC 4213, 3180 DOI 10.17487/RFC4213, October 2005, 3181 . 3183 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3184 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3185 January 2006, . 3187 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3188 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3189 DOI 10.17487/RFC4271, January 2006, 3190 . 3192 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3193 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3194 2006, . 3196 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3197 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3198 December 2005, . 3200 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3201 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3202 2006, . 3204 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3205 Control Message Protocol (ICMPv6) for the Internet 3206 Protocol Version 6 (IPv6) Specification", STD 89, 3207 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3208 . 3210 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3211 Protocol (LDAP): The Protocol", RFC 4511, 3212 DOI 10.17487/RFC4511, June 2006, 3213 . 3215 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3216 "Considerations for Internet Group Management Protocol 3217 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3218 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3219 . 3221 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3222 "Internet Group Management Protocol (IGMP) / Multicast 3223 Listener Discovery (MLD)-Based Multicast Forwarding 3224 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3225 August 2006, . 3227 [RFC4607] Holbrook, H. and B. Cain, "Source-Specific Multicast for 3228 IP", RFC 4607, DOI 10.17487/RFC4607, August 2006, 3229 . 3231 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 3232 Errors at High Data Rates", RFC 4963, 3233 DOI 10.17487/RFC4963, July 2007, 3234 . 3236 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3237 Algorithms in Cryptographically Generated Addresses 3238 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3239 . 3241 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3242 "Bidirectional Protocol Independent Multicast (BIDIR- 3243 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3244 . 3246 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3247 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3248 DOI 10.17487/RFC5214, March 2008, 3249 . 3251 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3252 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3253 February 2010, . 3255 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3256 Route Optimization Requirements for Operational Use in 3257 Aeronautics and Space Exploration Mobile Networks", 3258 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3259 . 3261 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3262 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3263 . 3265 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3266 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3267 January 2010, . 3269 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 3270 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 3271 . 3273 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3274 "IPv6 Router Advertisement Options for DNS Configuration", 3275 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3276 . 3278 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3279 NAT64: Network Address and Protocol Translation from IPv6 3280 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3281 April 2011, . 3283 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3284 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3285 . 3287 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3288 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3289 DOI 10.17487/RFC6221, May 2011, 3290 . 3292 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3293 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3294 DOI 10.17487/RFC6273, June 2011, 3295 . 3297 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3298 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3299 January 2012, . 3301 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3302 for Equal Cost Multipath Routing and Link Aggregation in 3303 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3304 . 3306 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3307 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3308 . 3310 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID Field", 3311 RFC 6864, DOI 10.17487/RFC6864, February 2013, 3312 . 3314 [RFC7269] Chen, G., Cao, Z., Xie, C., and D. Binet, "NAT64 3315 Deployment Options and Experience", RFC 7269, 3316 DOI 10.17487/RFC7269, June 2014, 3317 . 3319 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3320 Korhonen, "Requirements for Distributed Mobility 3321 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3322 . 3324 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3325 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3326 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3327 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3328 2016, . 3330 [RFC8086] Yong, L., Ed., Crabbe, E., Xu, X., and T. Herbert, "GRE- 3331 in-UDP Encapsulation", RFC 8086, DOI 10.17487/RFC8086, 3332 March 2017, . 3334 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 3335 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 3336 DOI 10.17487/RFC8201, July 2017, 3337 . 3339 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3340 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3341 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3342 July 2018, . 3344 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3345 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3346 . 3348 Appendix A. P[i] Preference Value Extensions in S/TLLAOs 3350 The AERO S/TLLAO format specified in Section 3.6 includes up to 64 3351 P[i] values which represent preferences for the 64 DSCP values. 3352 However, some AERO link types may require additional preference 3353 fields for "pseudo-DSCP" values P4, P65, P66, etc. 3355 For example, adaptation of AERO to the Aeronautical 3356 Telecommunications Network with Internet Protocol Services (ATN/IPS) 3357 includes link selection preferences based on transport port numbers 3358 in addition to the existing DSCP-based preferences. ATN/IPS nodes 3359 maintain a map of transport port numbers to additional P[i] 3360 preference fields, e.g., TCP port 22 maps to preference field P68, 3361 TCP port 443 maps to preference field P73, UDP port 8060 maps to 3362 preference field P89, etc. 3364 Figure 6 shows a format where two additional 4-byte blocks of P[i] 3365 preferences (i.e., P64 - P95) are included by setting the QoS field 3366 value to 6. 3368 0 1 2 3 3369 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 3370 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3371 | Type | Length | Prefix Length |R|X|N| Reserved| 3372 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3373 | ifIndex | ifType | Flags | Link | QoS=6 | 3374 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3375 |P00|P01|P02|P03|P04|P05|P06|P07|P08|P09|P10|P11|P12|P13|P14|P15| 3376 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3377 |P16|P17|P18|P19|P20|P21|P22|P23|P24|P25|P26|P27|P28|P29|P30|P31| 3378 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3379 |P32|P33|P34|P35|P36|P37|P38|P39|P40|P41|P42|P43|P44|P45|P46|P47| 3380 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3381 |P48|P49|P50|P51|P52|P53|P54|P55|P56|P57|P58|P59|P60|P61|P62|P63| 3382 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3383 |P64|P65|P66|P67|P68|P69|P70|P71|P72|P73|P74|P75|P76|P77|P78|P79| 3384 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3385 |P80|P81|P82|P83|P84|P85|P86|P87|P88|P89|P90|P91|P92|P93|P94|P95| 3386 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3387 | | 3388 + + 3389 | | 3390 + Link-Layer Address + 3391 | | 3392 + + 3393 | | 3394 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3395 | Port Number | ... 3396 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3398 Figure 6: P[i]-Extended S/TLLAO Format 3400 Appendix B. AERO Alternate Encapsulations 3402 When GUE encapsulation is not needed, AERO can use common 3403 encapsulations such as IP-in-IP [RFC2003][RFC2473][RFC4213], Generic 3404 Routing Encapsulation (GRE) [RFC2784][RFC2890] and others. The 3405 encapsulation is therefore only differentiated from non-AERO tunnels 3406 through the application of AERO control messaging and not through, 3407 e.g., a well-known UDP port number. 3409 As for GUE encapsulation, alternate AERO encapsulation formats may 3410 require encapsulation layer fragmentation. For simple IP-in-IP 3411 encapsulation, an IPv6 fragment header is inserted directly between 3412 the inner and outer IP headers when needed, i.e., even if the outer 3413 header is IPv4. The IPv6 Fragment Header is identified to the outer 3414 IP layer by its IP protocol number, and the Next Header field in the 3415 IPv6 Fragment Header identifies the inner IP header version. For GRE 3416 encapsulation, a GRE fragment header is inserted within the GRE 3417 header [I-D.templin-intarea-grefrag]. 3419 Figure 7 shows the AERO IP-in-IP encapsulation format before any 3420 fragmentation is applied: 3422 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3423 | Outer IPv4 Header | | Outer IPv6 Header | 3424 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3425 |IPv6 Frag Header (optional)| |IPv6 Frag Header (optional)| 3426 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3427 | Inner IP Header | | Inner IP Header | 3428 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3429 | | | | 3430 ~ ~ ~ ~ 3431 ~ Inner Packet Body ~ ~ Inner Packet Body ~ 3432 ~ ~ ~ ~ 3433 | | | | 3434 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3436 Minimal Encapsulation in IPv4 Minimal Encapsulation in IPv6 3438 Figure 7: Minimal Encapsulation Format using IP-in-IP 3440 Figure 8 shows the AERO GRE encapsulation format before any 3441 fragmentation is applied: 3443 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3444 | Outer IP Header | 3445 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3446 | GRE Header | 3447 | (with checksum, key, etc..) | 3448 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3449 | GRE Fragment Header (optional)| 3450 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3451 | Inner IP Header | 3452 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3453 | | 3454 ~ ~ 3455 ~ Inner Packet Body ~ 3456 ~ ~ 3457 | | 3458 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3460 Figure 8: Minimal Encapsulation Using GRE 3462 Alternate encapsulation may be preferred in environments where GUE 3463 encapsulation would add unnecessary overhead. For example, certain 3464 low-bandwidth wireless data links may benefit from a reduced 3465 encapsulation overhead. 3467 GUE encapsulation can traverse network paths that are inaccessible to 3468 non-UDP encapsulations, e.g., for crossing Network Address 3469 Translators (NATs). More and more, network middleboxes are also 3470 being configured to discard packets that include anything other than 3471 a well-known IP protocol such as UDP and TCP. It may therefore be 3472 necessary to determine the potential for middlebox filtering before 3473 enabling alternate encapsulation in a given environment. 3475 In addition to IP-in-IP, GRE and GUE, AERO can also use security 3476 encapsulations such as IPsec, TLS/SSL, DTLS, etc. In that case, AERO 3477 control messaging and route determination occur before security 3478 encapsulation is applied for outgoing packets and after security 3479 decapsulation is applied for incoming packets. 3481 AERO is especially well suited for use with VPN system encapsulations 3482 such as OpenVPN [OVPN]. 3484 Appendix C. Non-Normative Considerations 3486 AERO can be applied to a multitude of Internetworking scenarios, with 3487 each having its own adaptations. The following considerations are 3488 provided as non-normative guidance: 3490 C.1. Implementation Strategies for Route Optimization 3492 Route optimization as discussed in Section 3.17 results in the route 3493 optimization source (ROS) creating an asymmetric neighbor cache entry 3494 for the target neighbor. The neighbor cache entry is maintained for 3495 at most REACHABLETIME seconds and then deleted unless updated. In 3496 order to refresh the neighbor cache entry lifetime before the 3497 ReachableTime timer expires, the specification requires 3498 implementations to issue a new NS/NA exchange to reset ReachableTime 3499 to REACHABLETIME seconds while data packets are still flowing. 3500 However, the decision of when to initiate a new NS/NA exchange and to 3501 perpetuate the process is left as an implementation detail. 3503 One possible strategy may be to monitor the neighbor cache entry 3504 watching for data packets for (REACHABLETIME - 5) seconds. If any 3505 data packets have been sent to the neighbor within this timeframe, 3506 then send an NS to receive a new NA. If no data packets have been 3507 sent, wait for 5 additional seconds and send an immediate NS if any 3508 data packets are sent within this "expiration pending" 5 second 3509 window. If no additional data packets are sent within the 5 second 3510 window, delete the neighbor cache entry. 3512 The monitoring of the neighbor data packet traffic therefore becomes 3513 an asymmetric ongoing process during the neighbor cache entry 3514 lifetime. If the neighbor cache entry expires, future data packets 3515 will trigger a new NS/NA exchange while the packets themselves are 3516 delivered over a longer path until route optimization state is re- 3517 established. 3519 C.2. Implicit Mobility Management 3521 AERO interface neighbors MAY provide a configuration option that 3522 allows them to perform implicit mobility management in which no ND 3523 messaging is used. In that case, the Client only transmits packets 3524 over a single interface at a time, and the neighbor always observes 3525 packets arriving from the Client from the same link-layer source 3526 address. 3528 If the Client's underlying interface address changes (either due to a 3529 readdressing of the original interface or switching to a new 3530 interface) the neighbor immediately updates the neighbor cache entry 3531 for the Client and begins accepting and sending packets according to 3532 the Client's new address. This implicit mobility method applies to 3533 use cases such as cellphones with both WiFi and Cellular interfaces 3534 where only one of the interfaces is active at a given time, and the 3535 Client automatically switches over to the backup interface if the 3536 primary interface fails. 3538 C.3. Direct Underlying Interfaces 3540 When a Client's AERO interface is configured over a Direct interface, 3541 the neighbor at the other end of the Direct link can receive packets 3542 without any encapsulation. In that case, the Client sends packets 3543 over the Direct link according to QoS preferences. If the Direct 3544 interface has the highest QoS preference, then the Client's IP 3545 packets are transmitted directly to the peer without going through an 3546 ANET/INET. If other interfaces have higher QoS preferences, then the 3547 Client's IP packets are transmitted via a different interface, which 3548 may result in the inclusion of Proxys, Servers and Relays in the 3549 communications path. Direct interfaces must be tested periodically 3550 for reachability, e.g., via NUD. 3552 C.4. AERO Clients on the Open Internetwork 3554 AERO Clients that connect to the open Internetwork via either a 3555 native or NATed interface can establish a VPN to securely connect to 3556 a Server. Alternatively, the Client can exchange ND messages 3557 directly with other AERO nodes on the same SPAN segment using INET 3558 encapsulation only and without joining the SPAN. In that case, 3559 however, the Client must apply asymmetric security for ND messages to 3560 ensure routing and neighbor cache integrity (see: Section 6). 3562 C.5. Operation on AERO Links with /64 ASPs 3564 IPv6 AERO links typically have MSPs that aggregate many candidate 3565 MNPs of length /64 or shorter. However, in some cases it may be 3566 desirable to use AERO over links that have only a /64 MSP. This can 3567 be accommodated by treating all Clients on the AERO link as simple 3568 hosts that receive /128 prefix delegations. 3570 In that case, the Client sends an RS message to the Server the same 3571 as for ordinary AERO links. The Server responds with an RA message 3572 that includes one or more /128 prefixes (i.e., singleton addresses) 3573 that include the /64 MSP prefix along with an interface identifier 3574 portion to be assigned to the Client. The Client and Server then 3575 configure their AERO addresses based on the interface identifier 3576 portions of the /128s (i.e., the lower 64 bits) and not based on the 3577 /64 prefix (i.e., the upper 64 bits). 3579 For example, if the MSP for the host-only IPv6 AERO link is 3580 2001:db8:1000:2000::/64, each Client will receive one or more /128 3581 IPv6 prefix delegations such as 2001:db8:1000:2000::1/128, 3582 2001:db8:1000:2000::2/128, etc. When the Client receives the prefix 3583 delegations, it assigns the AERO addresses fe80::1, fe80::2, etc. to 3584 the AERO interface, and assigns the global IPv6 addresses (i.e., the 3585 /128s) to either the AERO interface or an internal virtual interface 3586 such as a loopback. In this arrangement, the Client conducts route 3587 optimization in the same sense as discussed in Section 3.17. 3589 This specification has applicability for nodes that act as a Client 3590 on an "upstream" AERO link, but also act as a Server on "downstream" 3591 AERO links. More specifically, if the node acts as a Client to 3592 receive a /64 prefix from the upstream AERO link it can then act as a 3593 Server to provision /128s to Clients on downstream AERO links. 3595 C.6. AERO Adaptations for SEcure Neighbor Discovery (SEND) 3597 SEcure Neighbor Discovery (SEND) [RFC3971] and Cryptographically 3598 Generated Addresses (CGAs) [RFC3972] were designed to secure IPv6 ND 3599 messaging in environments where symmetric network and/or transport- 3600 layer security services are impractical (see: Section 6). AERO nodes 3601 that use SEND/CGA employ the following adaptations. 3603 When a source AERO node prepares a SEND-protected ND message, it uses 3604 a link-local CGA as the IPv6 source address and writes the prefix 3605 embedded in its AERO address (i.e., instead of fe80::/64) in the CGA 3606 parameters Subnet Prefix field. When the neighbor receives the ND 3607 message, it first verifies the message checksum and SEND/CGA 3608 parameters while using the link-local prefix fe80::/64 (i.e., instead 3609 of the value in the Subnet Prefix field) to match against the IPv6 3610 source address of the ND message. 3612 The neighbor then derives the AERO address of the source by using the 3613 value in the Subnet Prefix field as the interface identifier of an 3614 AERO address. For example, if the Subnet Prefix field contains 3615 2001:db8:1:2, the neighbor constructs the AERO address as 3616 fe80::2001:db8:1:2. The neighbor then caches the AERO address in the 3617 neighbor cache entry it creates for the source, and uses the AERO 3618 address as the IPv6 destination address of any ND message replies. 3620 C.7. AERO Critical Infrastructure Considerations 3622 AERO Relays can be either Commercial off-the Shelf (COTS) standard IP 3623 routers or virtual machines in the cloud. Relays must be 3624 provisioned, supported and managed by the INET administrative 3625 authority, and connected to the Relays of other INETs via inter- 3626 domain peerings. Cost for purchasing, configuring and managing 3627 Relays is nominal even for very large AERO links. 3629 AERO Servers can be standard dedicated server platforms, but most 3630 often will be deployed as virtual machines in the cloud. The only 3631 requirements for Servers are that they can run the AERO user-level 3632 code and have at least one network interface connection to the INET. 3633 As with Relays, Servers must be provisioned, supported and managed by 3634 the INET administrative authority. Cost for purchasing, configuring 3635 and managing Servers is nominal especially for virtual Servers hosted 3636 in the cloud. 3638 AERO Proxys are most often standard dedicated server platforms with 3639 one network interface connected to the ANET and a second interface 3640 connected to an INET. As with Servers, the only requirements are 3641 that they can run the AERO user-level code and have at least one 3642 interface connection to the INET. Proxys must be provisioned, 3643 supported and managed by the ANET administrative authority. Cost for 3644 purchasing, configuring and managing Proxys is nominal, and borne by 3645 the ANET administrative authority. 3647 AERO Gateways can be any dedicated server or COTS router platform 3648 connected to INETs and/or EUNs. The Gateway joins the SPAN and 3649 engages in eBGP peering with one or more Relays as a stub AS. The 3650 Gateway then injects its MNPs and/or non-MNP prefixes into the BGP 3651 routing system, and provisions the prefixes to its downstream- 3652 attached networks. The Gateway can perform ROS and MAP services the 3653 same as for any Server, and can route between the MNP and non-MNP 3654 address spaces. 3656 C.8. AERO Server Failure Implications 3658 AERO Servers may appear as a single point of failure in the 3659 architecture, but such is not the case since all Servers on the link 3660 provide identical services and loss of a Server does not imply 3661 immediate and/or comprehensive communication failures. Although 3662 Clients typically associate with a single Server at a time, Server 3663 failure is quickly detected and conveyed by Bidirectional Forward 3664 Detection (BFD) allowing Clients to migrate to new Servers. 3666 If a Server fails, ongoing packet forwarding to Clients will continue 3667 by virtue of the asymmetric neighbor cache entries that have already 3668 been established in route optimization sources (ROSs). If a Client 3669 also experiences mobility events at roughly the same time the Server 3670 fails, unsolicited NA messages may be lost but proxy neighbor cache 3671 entries in the DEPARTED state will ensure that packet forwarding to 3672 the Client's new locations will continue for up to DEPARTTIME 3673 seconds. 3675 If a Client is left without a Server for an extended timeframe (e.g., 3676 greater than REACHABLETIIME seconds) then existing asymmetric 3677 neighbor cache entries will eventually expire and both ongoing and 3678 new communications will fail. The original source will continue to 3679 retransmit until the Client has established a new Server 3680 relationship, after which time continuous communications will resume. 3682 Therefore, providing many Servers on the link with high availability 3683 profiles provides resilience against loss of individual Servers and 3684 assurance that Clients can establish new Server relationships quickly 3685 in event of a Server failure. 3687 Appendix D. Change Log 3689 << RFC Editor - remove prior to publication >> 3691 Changes from draft-templin-intarea-6706bis-17 to draft-templin- 3692 intrea-6706bis-18: 3694 o Discuss how AERO option is used in relation to S/TLLAOs 3696 o New text on Bidirectional Forwarding Detection (BFD) 3698 o Cleaned up usage (and non-usage) of unsolicited NAs 3700 o New appendix on Server failures 3702 Changes from draft-templin-intarea-6706bis-15 to draft-templin- 3703 intrea-6706bis-17: 3705 o S/TLLAO now includes multiple link-layer addresses within a single 3706 option instead of requiring multiple options 3708 o New unsolicited NA message to inform the old link that a Client 3709 has moved to a new link 3711 Changes from draft-templin-intarea-6706bis-14 to draft-templin- 3712 intrea-6706bis-15: 3714 o MTU and fragmentation 3716 o New details in movement to new Server 3718 Changes from draft-templin-intarea-6706bis-13 to draft-templin- 3719 intrea-6706bis-14: 3721 o Security based on secured tunnels, ingress filtering, MAP list and 3722 ROS list 3724 Changes from draft-templin-intarea-6706bis-12 to draft-templin- 3725 intrea-6706bis-13: 3727 o New paragraph in Section 3.6 on AERO interface layering over 3728 secured tunnels 3730 o Removed extraneous text in Section 3.7 3732 o Added new detail to the forwarding algorithm in Section 3.9 3734 o Clarified use of fragmentation 3736 o Route optimization now supported for both MNP and non-MNP-based 3737 prefixes 3739 o Relays are now seen as link-layer elements in the architecture. 3741 o Built out multicast section in detail. 3743 o New Appendix on implementation considerations for route 3744 optimization. 3746 Changes from draft-templin-intarea-6706bis-11 to draft-templin- 3747 intrea-6706bis-12: 3749 o Introduced Gateways as a new AERO element for connecting 3750 Correspondent Nodes on INET links 3752 o Introduced terms "Access Network (ANET)" and "Internetwork (INET)" 3754 o Changed "ASP" to "MSP", and "ACP" to "MNP" 3756 o New figure on the relation of Segments to the SPAN and AERO link 3758 o New "S" bit in S/TLLAO to indicate the "Source" S/TLLAO as opposed 3759 to additional S/TLLAOs 3761 o Changed Interface ID for Servers from 255 to 0xffff 3763 o Significant updates to Route Optimization, NUD, and Mobility 3764 Management 3766 o New Section on Multicast 3768 o New Section on AERO Clients in the open Internetwork 3770 o New Section on Operation over multiple AERO links (VLANs over the 3771 SPAN) 3773 o New Sections on DNS considerations and Transition considerations 3775 o 3776 Changes from draft-templin-intarea-6706bis-10 to draft-templin- 3777 intrea-6706bis-11: 3779 o Added The SPAN 3781 Changes from draft-templin-intarea-6706bis-09 to draft-templin- 3782 intrea-6706bis-10: 3784 o Orphaned packets in flight (e.g., when a neighbor cache entry is 3785 in the DEPARTED state) are now forwarded at the link layer instead 3786 of at the network layer. Forwarding at the network layer can 3787 result in routing loops and/or excessive delays of forwarded 3788 packets while the routing system is still reconverging. 3790 o Update route optimization to clarify the unsecured nature of the 3791 first NS used for route discovery 3793 o Many cleanups and clarifications on ND messaging parameters 3795 Changes from draft-templin-intarea-6706bis-08 to draft-templin- 3796 intrea-6706bis-09: 3798 o Changed PRL to "MAP list" 3800 o For neighbor cache entries, changed "static" to "symmetric", and 3801 "dynamic" to "asymmetric" 3803 o Specified Proxy RS/RA exchanges with Servers on behalf of Clients 3805 o Added discussion of unsolicited NAs in Section 3.16, and included 3806 forward reference to Section 3.18 3808 o Added discussion of AERO Clients used as critical infrastructure 3809 elements to connect fixed networks. 3811 o Added network-based VPN under security considerations 3813 Changes from draft-templin-intarea-6706bis-07 to draft-templin- 3814 intrea-6706bis-08: 3816 o New section on AERO-Aware Access Router 3818 Changes from draft-templin-intarea-6706bis-06 to draft-templin- 3819 intrea-6706bis-07: 3821 o Added "R" bit for release of PDs. Now have a full RS/RA service 3822 that can do PD without requiring DHCPv6 messaging over-the-air 3824 o Clarifications on solicited vs unsolicited NAs 3826 o Clarified use of MAX_NEIGHBOR_ADVERTISEMENTS for the purpose of 3827 increase reliability 3829 Changes from draft-templin-intarea-6706bis-05 to draft-templin- 3830 intrea-6706bis-06: 3832 o Major re-work and simplification of Route Optimization function 3834 o Added Distributed Mobility Management (DMM) and Mobility Anchor 3835 Point (MAP) terminology 3837 o New section on "AERO Critical Infrastructure Element 3838 Considerations" demonstrating low overall cost for the service 3840 o minor text revisions and deletions 3842 o removed extraneous appendices 3844 Changes from draft-templin-intarea-6706bis-04 to draft-templin- 3845 intrea-6706bis-05: 3847 o New Appendix E on S/TLLAO Extensions for special-purpose links. 3848 Discussed ATN/IPS as example. 3850 o New sentence in introduction to declare appendices as non- 3851 normative. 3853 Changes from draft-templin-intarea-6706bis-03 to draft-templin- 3854 intrea-6706bis-04: 3856 o Added definitions for Potential Router List (PRL) and secure 3857 enclave 3859 o Included text on mapping transport layer port numbers to network 3860 layer DSCP values 3862 o Added reference to DTLS and DMM Distributed Mobility Anchoring 3863 working group document 3865 o Reworked Security Considerations 3867 o Updated references. 3869 Changes from draft-templin-intarea-6706bis-02 to draft-templin- 3870 intrea-6706bis-03: 3872 o Added new section on SEND. 3874 o Clarifications on "AERO Address" section. 3876 o Updated references and added new reference for RFC8086. 3878 o Security considerations updates. 3880 o General text clarifications and cleanup. 3882 Changes from draft-templin-intarea-6706bis-01 to draft-templin- 3883 intrea-6706bis-02: 3885 o Note on encapsulation avoidance in Section 4. 3887 Changes from draft-templin-intarea-6706bis-00 to draft-templin- 3888 intrea-6706bis-01: 3890 o Remove DHCPv6 Server Release procedures that leveraged the old way 3891 Relays used to "route" between Server link-local addresses 3893 o Remove all text relating to Relays needing to do any AERO-specific 3894 operations 3896 o Proxy sends RS and receives RA from Server using SEND. Use CGAs 3897 as source addresses, and destination address of RA reply is to the 3898 AERO address corresponding to the Client's ACP. 3900 o Proxy uses SEND to protect RS and authenticate RA (Client does not 3901 use SEND, but rather relies on subnetwork security. When the 3902 Proxy receives an RS from the Client, it creates a new RS using 3903 its own addresses as the source and uses SEND with CGAs to send a 3904 new RS to the Server. 3906 o Emphasize distributed mobility management 3908 o AERO address-based RS injection of ACP into underlying routing 3909 system. 3911 Changes from draft-templin-aerolink-82 to draft-templin-intarea- 3912 6706bis-00: 3914 o Document use of NUD (NS/NA) for reliable link-layer address 3915 updates as an alternative to unreliable unsolicited NA. 3916 Consistent with Section 7.2.6 of RFC4861. 3918 o Server adds additional layer of encapsulation between outer and 3919 inner headers of NS/NA messages for transmission through Relays 3920 that act as vanilla IPv6 routers. The messages include the AERO 3921 Server Subnet Router Anycast address as the source and the Subnet 3922 Router Anycast address corresponding to the Client's ACP as the 3923 destination. 3925 o Clients use Subnet Router Anycast address as the encapsulation 3926 source address when the access network does not provide a 3927 topologically-fixed address. 3929 Author's Address 3931 Fred L. Templin (editor) 3932 Boeing Research & Technology 3933 P.O. Box 3707 3934 Seattle, WA 98124 3935 USA 3937 Email: fltemplin@acm.org