idnits 2.17.1 draft-templin-intarea-6706bis-47.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 (April 29, 2020) is 1458 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 1031 -- Looks like a reference, but probably isn't: '2' on line 1037 == Missing Reference: 'N' is mentioned on line 1047, but not defined == Unused Reference: 'RFC2474' is defined on line 3229, but no explicit reference was found in the text == Unused Reference: 'RFC5175' is defined on line 3267, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-6man-segment-routing-header' is defined on line 3296, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-gue' is defined on line 3308, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-gue-extensions' is defined on line 3313, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-tunnels' is defined on line 3318, but no explicit reference was found in the text == Unused Reference: 'RFC1191' is defined on line 3368, but no explicit reference was found in the text == Unused Reference: 'RFC2764' is defined on line 3393, but no explicit reference was found in the text == Unused Reference: 'RFC2923' is defined on line 3407, but no explicit reference was found in the text == Unused Reference: 'RFC3819' is defined on line 3425, but no explicit reference was found in the text == Unused Reference: 'RFC4607' is defined on line 3480, but no explicit reference was found in the text == Unused Reference: 'RFC4963' is defined on line 3484, but no explicit reference was found in the text == Unused Reference: 'RFC5991' is defined on line 3526, but no explicit reference was found in the text == Unused Reference: 'RFC6864' is defined on line 3567, but no explicit reference was found in the text == Unused Reference: 'RFC7269' is defined on line 3581, but no explicit reference was found in the text == Unused Reference: 'RFC7421' is defined on line 3591, but no explicit reference was found in the text == Unused Reference: 'RFC8086' is defined on line 3603, but no explicit reference was found in the text == Outdated reference: A later version (-99) exists of draft-templin-6man-omni-interface-17 == 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-05 == Outdated reference: A later version (-11) exists of draft-templin-6man-dhcpv6-ndopt-09 == Outdated reference: A later version (-27) exists of draft-templin-v6ops-pdhost-25 -- 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 (~~), 25 warnings (==), 5 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, April 29, 2020 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: October 31, 2020 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-intarea-6706bis-47 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, intelligent transportation systems, 26 mobile Virtual Private Networks (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 October 31, 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) . . . . . . . . 11 65 3.1. AERO Link Reference Model . . . . . . . . . . . . . . . . 11 66 3.2. AERO Node Types . . . . . . . . . . . . . . . . . . . . . 12 67 3.3. AERO Routing System . . . . . . . . . . . . . . . . . . . 13 68 3.4. AERO Addresses . . . . . . . . . . . . . . . . . . . . . 15 69 3.5. Spanning Partitioned AERO Networks (SPAN) . . . . . . . . 16 70 3.5.1. SPAN Routing Topologies . . . . . . . . . . . . . . . 20 71 3.5.2. Segment Routing Over the SPAN . . . . . . . . . . . . 20 72 3.6. AERO Interface Characteristics . . . . . . . . . . . . . 21 73 3.7. AERO Interface Initialization . . . . . . . . . . . . . . 26 74 3.7.1. AERO Server/Relay Behavior . . . . . . . . . . . . . 26 75 3.7.2. AERO Proxy Behavior . . . . . . . . . . . . . . . . . 26 76 3.7.3. AERO Client Behavior . . . . . . . . . . . . . . . . 26 77 3.7.4. AERO Bridge Behavior . . . . . . . . . . . . . . . . 27 78 3.8. AERO Interface Neighbor Cache Maintenance . . . . . . . . 27 79 3.9. AERO Interface Encapsulation and Re-encapsulation . . . . 29 80 3.10. AERO Interface Decapsulation . . . . . . . . . . . . . . 30 81 3.11. AERO Interface Data Origin Authentication . . . . . . . . 30 82 3.12. AERO Interface MTU and Fragmentation . . . . . . . . . . 31 83 3.13. AERO Interface Forwarding Algorithm . . . . . . . . . . . 33 84 3.13.1. Client Forwarding Algorithm . . . . . . . . . . . . 33 85 3.13.2. Proxy Forwarding Algorithm . . . . . . . . . . . . . 34 86 3.13.3. Server/Relay Forwarding Algorithm . . . . . . . . . 35 87 3.13.4. Bridge Forwarding Algorithm . . . . . . . . . . . . 36 88 3.14. AERO Interface Error Handling . . . . . . . . . . . . . . 37 89 3.15. AERO Router Discovery, Prefix Delegation and 90 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 39 91 3.15.1. AERO ND/PD Service Model . . . . . . . . . . . . . . 39 92 3.15.2. AERO Client Behavior . . . . . . . . . . . . . . . . 40 93 3.15.3. AERO Server Behavior . . . . . . . . . . . . . . . . 42 94 3.16. The AERO Proxy . . . . . . . . . . . . . . . . . . . . . 45 95 3.16.1. Detecting and Responding to Server Failures . . . . 47 96 3.16.2. Point-to-Multipoint Server Coordination . . . . . . 48 97 3.17. AERO Route Optimization . . . . . . . . . . . . . . . . . 48 98 3.17.1. Route Optimization Initiation . . . . . . . . . . . 49 99 3.17.2. Relaying the NS . . . . . . . . . . . . . . . . . . 49 100 3.17.3. Processing the NS and Sending the NA . . . . . . . . 49 101 3.17.4. Relaying the NA . . . . . . . . . . . . . . . . . . 51 102 3.17.5. Processing the NA . . . . . . . . . . . . . . . . . 51 103 3.17.6. Route Optimization Maintenance . . . . . . . . . . . 51 104 3.18. Neighbor Unreachability Detection (NUD) . . . . . . . . . 52 105 3.19. Mobility Management and Quality of Service (QoS) . . . . 53 106 3.19.1. Mobility Update Messaging . . . . . . . . . . . . . 54 107 3.19.2. Announcing Link-Layer Address and/or QoS Preference 108 Changes . . . . . . . . . . . . . . . . . . . . . . 55 109 3.19.3. Bringing New Links Into Service . . . . . . . . . . 55 110 3.19.4. Removing Existing Links from Service . . . . . . . . 55 111 3.19.5. Moving to a New Server . . . . . . . . . . . . . . . 56 112 3.20. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 57 113 3.20.1. Source-Specific Multicast (SSM) . . . . . . . . . . 57 114 3.20.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 58 115 3.20.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 59 116 3.21. Operation over Multiple AERO Links (VLANs) . . . . . . . 59 117 3.22. DNS Considerations . . . . . . . . . . . . . . . . . . . 60 118 3.23. Transition Considerations . . . . . . . . . . . . . . . . 61 119 3.24. Detecting and Reacting to Server and Bridge Failures . . 61 120 3.25. AERO Clients on the Open Internet . . . . . . . . . . . . 62 121 3.26. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . 64 122 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 65 123 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 65 124 6. Security Considerations . . . . . . . . . . . . . . . . . . . 65 125 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 67 126 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 68 127 8.1. Normative References . . . . . . . . . . . . . . . . . . 68 128 8.2. Informative References . . . . . . . . . . . . . . . . . 70 129 Appendix A. AERO Alternate Encapsulations . . . . . . . . . . . 77 130 Appendix B. Non-Normative Considerations . . . . . . . . . . . . 79 131 B.1. Implementation Strategies for Route Optimization . . . . 80 132 B.2. Implicit Mobility Management . . . . . . . . . . . . . . 80 133 B.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 81 134 B.4. Operation on AERO Links with /64 ASPs . . . . . . . . . . 81 135 B.5. AERO Critical Infrastructure Considerations . . . . . . . 82 136 B.6. AERO Server Failure Implications . . . . . . . . . . . . 82 137 B.7. AERO Client / Server Architecture . . . . . . . . . . . . 83 138 Appendix C. Change Log . . . . . . . . . . . . . . . . . . . . . 85 139 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 94 141 1. Introduction 143 Asymmetric Extended Route Optimization (AERO) fulfills the 144 requirements of Distributed Mobility Management (DMM) [RFC7333] and 145 route optimization [RFC5522] for aeronautical networking and other 146 network mobility use cases such as intelligent transportation 147 systems. AERO is based on a Non-Broadcast, Multiple Access (NBMA) 148 virtual link model known as the AERO link. The AERO link is a 149 virtual overlay configured over one or more underlying Internetworks, 150 and nodes on the link can exchange IP packets via tunneling. 151 Multilink operation allows for increased reliability, bandwidth 152 optimization and traffic path diversity. 154 The AERO service comprises Clients, Proxys, Servers and Relays that 155 are seen as AERO link neighbors as well as Bridges that interconnect 156 AERO link segments. Each node's AERO interface uses an IPv6 link- 157 local address format (known as the AERO address) that supports 158 operation of the IPv6 Neighbor Discovery (ND) protocol [RFC4861] and 159 links ND to IP forwarding. A node's AERO interface can be configured 160 over multiple underlying interfaces, and may therefore appear as a 161 single interface with multiple link-layer addresses. Each link-layer 162 address is subject to change due to mobility and/or QoS fluctuations, 163 and link-layer address changes are signaled by ND messaging the same 164 as for any IPv6 link. 166 AERO links provide a cloud-based service where mobile nodes may use 167 any Server acting as a Mobility Anchor Point (MAP) and fixed nodes 168 may use any Relay on the link for efficient communications. Fixed 169 nodes forward packets destined to other AERO nodes to the nearest 170 Relay, which forwards them through the cloud. A mobile node's 171 initial packets are forwarded through the Server, while direct 172 routing is supported through asymmetric extended route optimization 173 while data packets are flowing. Both unicast and multicast 174 communications are supported, and mobile nodes may efficiently move 175 between locations while maintaining continuous communications with 176 correspondents and without changing their IP Address. 178 AERO Bridges are interconnected in a secured private BGP overlay 179 routing instance known as "The SPAN". The SPAN provides a hybrid 180 routing/bridging service to join the underlying Internetworks of 181 multiple disjoint administrative domains into a single unified AERO 182 link. Each AERO link instance is characterized by the set of 183 Mobility Service Prefixes (MSPs) common to all mobile nodes. The 184 link extends to the point where a Relay/Server is on the optimal 185 route from any correspondent node on the link, and provides a conduit 186 between the underlying Internetwork and the SPAN. To the underlying 187 Internetwork, the Relay/Server is the source of a route to the MSP, 188 and hence uplink traffic to the mobile node is naturally routed to 189 the nearest Relay/Server. 191 AERO assumes the use of PIM Sparse Mode in support of multicast 192 communication. In support of Source Specific Multicast (SSM) when a 193 Mobile Node is the source, AERO route optimization ensures that a 194 shortest-path multicast tree is established with provisions for 195 mobility and multilink operation. In all other multicast scenarios 196 there are no AERO dependencies. 198 AERO was designed for aeronautical networking for both manned and 199 unmanned aircraft, where the aircraft is treated as a mobile node 200 that can connect an Internet of Things (IoT). AERO is also 201 applicable to a wide variety of other use cases. For example, it can 202 be used to coordinate the Virtual Private Network (VPN) links of 203 mobile nodes (e.g., cellphones, tablets, laptop computers, etc.) that 204 connect into a home enterprise network via public access networks 205 using services such as OpenVPN [OVPN]. It can also be used to 206 facilitate vehicular and pedestrian communications services for 207 intelligent transportation systems. Other applicable use cases are 208 also in scope. 210 The following numbered sections present the AERO specification. The 211 appendices at the end of the document are non-normative. 213 2. Terminology 215 The terminology in the normative references applies; the following 216 terms are defined within the scope of this document: 218 IPv6 Neighbor Discovery (ND) 219 an IPv6 control message service for coordinating neighbor 220 relationships between nodes connected to a common link. AERO 221 interfaces use the ND service specified in [RFC4861]. 223 IPv6 Prefix Delegation (PD) 224 a networking service for delegating IPv6 prefixes to nodes on the 225 link. The nominal PD service is DHCPv6 [RFC8415], however 226 alternate services (e.g., based on ND messaging) are also in scope 227 [I-D.templin-v6ops-pdhost][I-D.templin-6man-dhcpv6-ndopt]. Most 228 notably, a minimal form of PD known as "prefix registration" can 229 be used if the Client knows its prefix in advance and can 230 represent it in the IPv6 source address of an ND message. 232 Access Network (ANET) 233 a node's first-hop data link service network, e.g., a radio access 234 network, cellular service provider network, corporate enterprise 235 network, or the public Internet itself. For secured ANETs, link- 236 layer security services such as IEEE 802.1X and physical-layer 237 security prevent unauthorized access internally while border 238 network-layer security services such as firewalls and proxies 239 prevent unauthorized outside access. 241 ANET interface 242 a node's attachment to a link in an ANET. 244 ANET address 245 an IP address assigned to a node's interface connection to an 246 ANET. 248 Internetwork (INET) 249 a connected IP network topology with a coherent routing and 250 addressing plan and that provides a transit backbone service for 251 ANET end systems. INETs also provide an underlay service over 252 which the AERO virtual link is configured. Example INETs include 253 corporate enterprise networks, aviation networks, and the public 254 Internet itself. When there is no administrative boundary between 255 an ANET and the INET, the ANET and INET are one and the same. 257 INET Partition 258 frequently, INETs such as large corporate enterprise networks are 259 sub-divided internally into separate isolated partitions. Each 260 partition is fully connected internally but disconnected from 261 other partitions, and there is no requirement that separate 262 partitions maintain consistent Internet Protocol and/or addressing 263 plans. (Each INET partition is seen as a separate SPAN segment as 264 discussed below.) 266 INET interface 267 a node's attachment to a link in an INET. 269 INET address 270 an IP address assigned to a node's interface connection to an 271 INET. 273 INET encapsulation 274 the encapsulation of a packet in an outer header or headers that 275 can be routed within the scope of the local INET partition. 277 AERO link 278 a Non-Broadcast, Multiple Access (NBMA) tunnel virtual overlay 279 configured over one or more underlying INETs. Nodes on the AERO 280 link appear as single-hop neighbors from the perspective of the 281 virtual overlay even though they may be separated by many 282 underlying INET hops. AERO links may be configured over multiple 283 underlying SPAN segments (see below). 285 AERO interface 286 a node's attachment to an AERO link. Since the addresses assigned 287 to an AERO interface are managed for uniqueness, AERO interfaces 288 do not require Duplicate Address Detection (DAD) and therefore set 289 the administrative variable 'DupAddrDetectTransmits' to zero 290 [RFC4862]. 292 underlying interface 293 an ANET or INET interface over which an AERO interface is 294 configured. 296 AERO address 297 an IPv6 link-local address assigned to an AERO interface and 298 constructed as specified in Section 3.4. 300 base AERO address 301 the lowest-numbered AERO address aggregated by the MNP (see 302 Section 3.4). 304 Mobility Service Prefix (MSP) 305 an IP prefix assigned to the AERO link and from which more- 306 specific Mobile Network Prefixes (MNPs) are derived. 308 Mobile Network Prefix (MNP) 309 an IP prefix allocated from an MSP and delegated to an AERO Client 310 or Relay. 312 AERO node 313 a node that is connected to an AERO link, or that provides 314 services to other nodes on an AERO link. 316 AERO Client ("Client") 317 an AERO node that connects over one or more underlying interfaces 318 and requests MNP PDs from AERO Servers. The Client assigns a 319 Client AERO address to the AERO interface for use in ND exchanges 320 with other AERO nodes and forwards packets to correspondents 321 according to AERO interface neighbor cache state. 323 AERO Server ("Server") 324 an INET node that configures an AERO interface to provide default 325 forwarding and mobility/multilink services for AERO Clients. The 326 Server assigns an administratively-provisioned AERO address to its 327 AERO interface to support the operation of the ND/PD services, and 328 advertises all of its associated MNPs via BGP peerings with 329 Bridges. 331 AERO Relay ("Relay") 332 an AERO Server that also provides forwarding services between 333 nodes reached via the AERO link and correspondents on other links. 334 AERO Relays are provisioned with MNPs (i.e., the same as for an 335 AERO Client) and run a dynamic routing protocol to discover any 336 non-MNP IP routes. In both cases, the Relay advertises the MSP(s) 337 to its downstream networks, and distributes all of its associated 338 MNPs and non-MNP IP routes via BGP peerings with Bridges (i.e., 339 the same as for an AERO Server). 341 AERO Bridge ("Bridge") 342 a node that provides hybrid routing/bridging services (as well as 343 a security trust anchor) for nodes on an AERO link. As a router, 344 the Bridge forwards packets using standard IP forwarding. As a 345 bridge, the Bridge forwards packets over the AERO link without 346 decrementing the IPv6 Hop Limit. AERO Bridges peer with Servers 347 and other Bridges to discover the full set of MNPs for the link as 348 well as any non-MNPs that are reachable via Relays. 350 AERO Proxy ("Proxy") 351 a node that provides proxying services between Clients in an ANET 352 and Servers in external INETs. The AERO Proxy is a conduit 353 between the ANET and external INETs in the same manner as for 354 common web proxies, and behaves in a similar fashion as for ND 355 proxies [RFC4389]. 357 Spanning Partitioned AERO Networks (SPAN) 358 a means for bridging disjoint INET partitions as segments of a 359 unified AERO link the same as for a bridged campus LAN. The SPAN 360 is a mid-layer IPv6 encapsulation service in the AERO routing 361 system that supports a unified AERO link view for all segments. 362 Each segment in the SPAN is a distinct INET partition, and 363 individual segments are joined by AERO Bridges. Segment Routing 364 [RFC8402][RFC8754] can be used to cause packets to visit selected 365 hops on the SPAN. 367 SPAN Service Prefix (SSP) 368 a global or unique local /96 IPv6 prefix assigned to the AERO link 369 to support SPAN services. 371 SPAN Partition Prefix (SPP) 372 a sub-prefix of the SPAN Service Prefix uniquely assigned to a 373 single SPAN segment. 375 SPAN Client Prefix (SCP) 376 a SPAN prefix formed from an AERO Client address. 378 SPAN Address 379 a unique local IPv6 address taken from a SPAN Client/Partition 380 Prefix and constructed as specified in Section 3.5. SPAN 381 addresses are statelessly derived from AERO addresses, and vice- 382 versa. 384 SPAN encapsulation 385 the addition of an IPv6 header with SPAN source and destinations 386 per [RFC2473]. SPAN encapsulation is often used in conjunction 387 with INET encapsulation, or in raw form over ANET interfaces. 389 ingress tunnel endpoint (ITE) 390 an AERO interface endpoint that injects encapsulated packets into 391 an AERO link. 393 egress tunnel endpoint (ETE) 394 an AERO interface endpoint that receives encapsulated packets from 395 an AERO link. 397 link-layer address 398 an IP address used as an encapsulation header source or 399 destination address from the perspective of the AERO interface. 400 When an upper layer protocol (e.g., UDP) is used as part of the 401 encapsulation, the port number is also considered as part of the 402 link-layer address. From the perspective of the AERO interface, 403 the link-layer address is either an INET address for intra-segment 404 encapsulation or a SPAN address for inter-segment encapsulation. 406 network layer address 407 the source or destination address of an encapsulated IP packet 408 presented to the AERO interface. 410 end user network (EUN) 411 an internal virtual or external edge IP network that an AERO 412 Client or Relay connects to the rest of the network via the AERO 413 interface. The Client/Relay sees each EUN as a "downstream" 414 network, and sees the AERO interface as the point of attachment to 415 the "upstream" network. 417 Mobile Node (MN) 418 an AERO Client and all of its downstream-attached networks that 419 move together as a single unit, i.e., an end system that connects 420 an Internet of Things. 422 Mobile Router (MR) 423 a MN's on-board router that forwards packets between any 424 downstream-attached networks and the AERO link. 426 Route Optimization Source (ROS) 427 the AERO node nearest the source that initiates route 428 optimization. The ROS may be a Server or Proxy acting on behalf 429 of the source Client. 431 Route Optimization responder (ROR) 432 the AERO node nearest the target destination that responds to 433 route optimization requests. The ROR may be a Server acting on 434 behalf of a target MNP Client, or a Relay for a non-MNP 435 destination. 437 MAP List 438 a geographically and/or topologically referenced list of AERO 439 addresses of all Servers within the same AERO link. There is a 440 single MAP list for the entire AERO link. 442 Distributed Mobility Management (DMM) 443 a BGP-based overlay routing service coordinated by Servers and 444 Bridges that tracks all Server-to-Client associations. 446 Mobility Service (MS) 447 the collective set of all Servers, Proxys, Bridges and Relays that 448 provide the AERO Service to Clients. 450 Mobility Service Endpoint MSE) 451 an individual Server, Proxy, Bridge or Relay in the Mobility 452 Service. 454 Throughout the document, the simple terms "Client", "Server", 455 "Bridge", "Proxy" and "Relay" refer to "AERO Client", "AERO Server", 456 "AERO Bridge", "AERO Proxy" and "AERO Relay", respectively. 457 Capitalization is used to distinguish these terms from other common 458 Internetworking uses in which they appear without capitalization. 460 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 461 the names of node variables, messages and protocol constants) is used 462 throughout this document. The terms "All-Routers multicast", "All- 463 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 464 anycast" are defined in [RFC4291] (with Link-Local scope assumed). 465 Also, the term "IP" is used to generically refer to either Internet 466 Protocol version, i.e., IPv4 [RFC0791] or IPv6 [RFC8200]. 468 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 469 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 470 "OPTIONAL" in this document are to be interpreted as described in BCP 471 14 [RFC2119][RFC8174] when, and only when, they appear in all 472 capitals, as shown here. 474 3. Asymmetric Extended Route Optimization (AERO) 476 The following sections specify the operation of IP over Asymmetric 477 Extended Route Optimization (AERO) links: 479 3.1. AERO Link Reference Model 481 +----------------+ 482 | AERO Bridge B1 | 483 | Nbr: S1, S2, P1| 484 |(X1->S1; X2->S2)| 485 | MSP M1 | 486 +-+---------+--+-+ 487 +--------------+ | Secured | | +--------------+ 488 |AERO Server S1| | tunnels | | |AERO Server S2| 489 | Nbr: C1, B1 +-----+ | +-----+ Nbr: C2, B1 | 490 | default->B1 | | | default->B1 | 491 | X1->C1 | | | X2->C2 | 492 +-------+------+ | +------+-------+ 493 | AERO Link | | 494 X===+===+===================+==)===============+===+===X 495 | | | | 496 +-----+--------+ +--------+--+-----+ +--------+-----+ 497 |AERO Client C1| | AERO Proxy P1 | |AERO Client C2| 498 | Nbr: S1 | |(Proxy Nbr Cache)| | Nbr: S2 | 499 | default->S1 | +--------+--------+ | default->S2 | 500 | MNP X1 | | | MNP X2 | 501 +------+-------+ .--------+------. +-----+--------+ 502 | (- Proxyed Clients -) | 503 .-. `---------------' .-. 504 ,-( _)-. ,-( _)-. 505 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 506 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 507 `-(______)-' +-------+ +-------+ `-(______)-' 509 Figure 1: AERO Link Reference Model 511 Figure 1 presents the AERO link reference model. In this model: 513 o the AERO link is an overlay network service configured over one or 514 more underlying INET partitions which may be managed by different 515 administrative authorities and have incompatible protocols and/or 516 addressing plans. 518 o AERO Bridge B1 aggregates Mobility Service Prefix (MSP) M1, 519 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 520 via BGP peerings over secured tunnels to Servers (S1, S2). 522 Bridges use the SPAN service to bridge disjoint segments of a 523 partitioned AERO link. 525 o AERO Servers S1 and S2 configure secured tunnels with Bridge B1 526 and also provide mobility, multilink and default router services 527 for their associated Clients C1 and C2. 529 o AERO Clients C1 and C2 associate with Servers S1 and S2, 530 respectively. They receive Mobile Network Prefix (MNP) 531 delegations X1 and X2, and also act as default routers for their 532 associated physical or internal virtual EUNs. Simple hosts H1 and 533 H2 attach to the EUNs served by Clients C1 and C2, respectively. 535 o AERO Proxy P1 configures a secured tunnel with Bridge B1 and 536 provides proxy services for AERO Clients in secured enclaves that 537 cannot associate directly with other AERO link neighbors. 539 Each node on the AERO link maintains an AERO interface neighbor cache 540 and an IP forwarding table the same as for any link. Although the 541 figure shows a limited deployment, in common operational practice 542 there will normally be many additional Bridges, Servers, Clients and 543 Proxys. 545 3.2. AERO Node Types 547 AERO Bridges provide hybrid routing/bridging services (as well as a 548 security trust anchor) for nodes on an AERO link. Bridges use 549 standard IPv6 routing to forward packets both within the same INET 550 partitions and between disjoint INET partitions based on a mid-layer 551 IPv6 encapsulation known as the SPAN header. The inner IP layer 552 experiences a virtual bridging service since the inner IP TTL/Hop 553 Limit is not decremented during forwarding. Each Bridge also peers 554 with Servers and other Bridges in a dynamic routing protocol instance 555 to provide a Distributed Mobility Management (DMM) service for the 556 list of active MNPs (see Section 3.3). Bridges present the AERO link 557 as a set of one or more Mobility Service Prefixes (MSPs) but as link- 558 layer devices need not connect directly to the AERO link themselves 559 unless an administrative interface is desired. Bridges configure 560 secured tunnels with Servers, Proxys and other Bridges; they further 561 maintain IP forwarding table entries for each Mobile Network Prefix 562 (MNP) and any other reachable non-MNP prefixes. 564 AERO Servers provide default forwarding and mobility/multilink 565 services for AERO Client Mobile Nodes (MNs). Each Server also peers 566 with Bridges in a dynamic routing protocol instance to advertise its 567 list of associated MNPs (see Section 3.3). Servers facilitate PD 568 exchanges with Clients, where each delegated prefix becomes an MNP 569 taken from an MSP. Servers forward packets between AERO interface 570 neighbors and track each Client's mobility profiles. 572 AERO Clients register their MNPs through PD exchanges with AERO 573 Servers over the AERO link, and distribute the MNPs to nodes on EUNs. 574 A Client may also be co-resident on the same physical or virtual 575 platform as a Server; in that case, the Client and Server behave as a 576 single functional unit. 578 AERO Proxys provide a conduit for ANET Clients to associate with 579 Servers in external INETs. Client and Servers exchange control plane 580 messages via the Proxy acting as a bridge between the ANET/INET 581 boundary. The Proxy forwards data packets between Clients and the 582 AERO link according to forwarding information in the neighbor cache. 583 The Proxy function is specified in Section 3.16. 585 AERO Relays are Servers that provide forwarding services between the 586 AERO interface and INET/EUN interfaces. Relays are provisioned with 587 MNPs the same as for an AERO Client, and also run a dynamic routing 588 protocol to discover any non-MNP IP routes. The Relay advertises the 589 MSP(s) to its connected networks, and distributes all of its 590 associated MNPs and non-MNP IP routes via BGP peerings with Bridges. 592 AERO Bridges, Servers, Proxys and Relays are critical infrastructure 593 elements in fixed (i.e., non-mobile) INET deployments and hence have 594 permanent and unchanging INET addresses. AERO Clients are MNs that 595 connect via underlying interfaces with addresses that may change when 596 the Client moves to a new network connection point. 598 3.3. AERO Routing System 600 The AERO routing system comprises a private instance of the Border 601 Gateway Protocol (BGP) [RFC4271] that is coordinated between Bridges 602 and Servers and does not interact with either the public Internet BGP 603 routing system or any underlying INET routing systems. 605 In a reference deployment, each Server is configured as an Autonomous 606 System Border Router (ASBR) for a stub Autonomous System (AS) using 607 an AS Number (ASN) that is unique within the BGP instance, and each 608 Server further uses eBGP to peer with one or more Bridges but does 609 not peer with other Servers. Each INET of a multi-segment AERO link 610 must include one or more Bridges, which peer with the Servers and 611 Proxys within that INET. All Bridges within the same INET are 612 members of the same hub AS using a common ASN, and use iBGP to 613 maintain a consistent view of all active MNPs currently in service. 614 The Bridges of different INETs peer with one another using eBGP. 616 Bridges advertise the AERO link's MSPs and any non-MNP routes to each 617 of their Servers. This means that any aggregated non-MNPs (including 618 "default") are advertised to all Servers. Each Bridge configures a 619 black-hole route for each of its MSPs. By black-holing the MSPs, the 620 Bridge will maintain forwarding table entries only for the MNPs that 621 are currently active, and packets destined to all other MNPs will 622 correctly incur Destination Unreachable messages due to the black- 623 hole route. In this way, Servers have only partial topology 624 knowledge (i.e., they know only about the MNPs of their directly 625 associated Clients) and they forward all other packets to Bridges 626 which have full topology knowledge. 628 Servers maintain a working set of associated MNPs, and dynamically 629 announce new MNPs and withdraw departed MNPs in eBGP updates to 630 Bridges. Servers that are configured as Relays also redistribute 631 non-MNP routes learned from non-AERO interfaces via their eBGP Bridge 632 peerings. 634 Clients are expected to remain associated with their current Servers 635 for extended timeframes, however Servers SHOULD selectively suppress 636 updates for impatient Clients that repeatedly associate and 637 disassociate with them in order to dampen routing churn. Servers 638 that are configured as Relays advertise the MSPs via INET/EUN 639 interfaces, and forward packets between INET/EUN interfaces and the 640 AERO interface using standard IP forwarding. 642 Scaling properties of the AERO routing system are limited by the 643 number of BGP routes that can be carried by Bridges. As of 2015, the 644 global public Internet BGP routing system manages more than 500K 645 routes with linear growth and no signs of router resource exhaustion 646 [BGP]. More recent network emulation studies have also shown that a 647 single Bridge can accommodate at least 1M dynamically changing BGP 648 routes even on a lightweight virtual machine, i.e., and without 649 requiring high-end dedicated router hardware. 651 Therefore, assuming each Bridge can carry 1M or more routes, this 652 means that at least 1M Clients can be serviced by a single set of 653 Bridges. A means of increasing scaling would be to assign a 654 different set of Bridges for each set of MSPs. In that case, each 655 Server still peers with one or more Bridges, but institutes route 656 filters so that BGP updates are only sent to the specific set of 657 Bridges that aggregate the MSP. For example, if the MSP for the AERO 658 link is 2001:db8::/32, a first set of Bridges could service the MSP 659 2001:db8::/40, a second set of Bridges could service 660 2001:db8:0100::/40, a third set could service 2001:db8:0200::/40, 661 etc. 663 Assuming up to 1K sets of Bridges, the AERO routing system can then 664 accommodate 1B or more MNPs with no additional overhead (for example, 665 it should be possible to service 1B /64 MNPs taken from a /34 MSP and 666 even more for shorter prefixes). In this way, each set of Bridges 667 services a specific set of MSPs that they advertise to the native 668 Internetwork routing system, and each Server configures MSP-specific 669 routes that list the correct set of Bridges as next hops. This 670 arrangement also allows for natural incremental deployment, and can 671 support small scale initial deployments followed by dynamic 672 deployment of additional Clients, Servers and Bridges without 673 disturbing the already-deployed base. 675 Server and Bridges can use the Bidirectional Forwarding Detection 676 (BFD) protocol [RFC5880] to quickly detect link failures that don't 677 result in interface state changes, BGP peer failures, and 678 administrative state changes. BFD is important in environments where 679 rapid response to failures is required for routing reconvergence and, 680 hence, communications continuity. 682 A full discussion of the BGP-based routing system used by AERO is 683 found in [I-D.ietf-rtgwg-atn-bgp]. The system provides for 684 Distributed Mobility Management (DMM) per the distributed mobility 685 anchoring architecture [I-D.ietf-dmm-distributed-mobility-anchoring]. 687 3.4. AERO Addresses 689 A Client's AERO address is an IPv6 link-local address formed from the 690 Client's delegated MNP. Bridge, Server, Relay and Proxy AERO 691 addresses are assigned from the range fe80::/96 and include an 692 administratively-provisioned value in the lower 32 bits. 694 IPv6 Client AERO addresses encode the Subnet-Router anycast address 695 of a MNP (or non-MNP globally routable prefix) within the least- 696 significant 112 bits of the IPv6 link-local prefix fe80::/16. For 697 example, for the MNP 2001:db8:1000:2000::/56 the corresponding AERO 698 address is fe80:2001:db8:1000:2000::/72. 700 IPv4-compatible Client AERO addresses are based on an IPv4-mapped 701 IPv6 address [RFC4291] formed from an IPv4 MNP and with a prefix 702 length of 96 plus the MNP prefix length. For example, for the IPv4 703 MNP 192.0.2.16/28 the IPv4-mapped IPv6 MNP is: 705 0:0:0:0:0:ffff:192.0.2.16/124 (also written as 706 0:0:0:0:0:ffff:c000:0210/124) 708 The Client then constructs its AERO address with the prefix fe80::/64 709 and with the lower 64 bits of the IPv4-mapped IPv6 address in the 710 interface identifier as: fe80::ffff:192.0.2.16. 712 Mobility Service (MS) AERO addresses (used by Bridges, Servers, 713 Relays and Proxys) are allocated from the range fe80::/96, and MUST 714 be managed for uniqueness. The lower 32 bits of the AERO address 715 includes a unique integer value between 1 and 0xfeffffff (e.g., 716 fe80::1, fe80::2, fe80::3, etc., fe80::feff:ffff) as assigned by the 717 administrative authority for the link. The address fe80:: is the 718 IPv6 link-local Subnet-Router anycast address, and the address range 719 fe80::ff00:0000/104 is reserved for future use. 721 Finally, the address range fe80::/32 is used as the Teredo service 722 prefix for AERO according to the format in Section 4 of [RFC4380] 723 (see Section 3.25 for further discussion). 725 For a full discussion of the above address format and implications 726 for the /64 boundary, see: [I-D.templin-6man-omni-interface]. 728 3.5. Spanning Partitioned AERO Networks (SPAN) 730 An AERO link configured over a single INET appears as a single 731 unified link with a consistent underlying network addressing plan. 732 In that case, all nodes on the link can exchange packets via simple 733 INET encapsulation, since the underlying INET is connected. In 734 common practice, however, an AERO link may be partitioned into 735 multiple "segments", where each segment is a distinct INET 736 potentially managed under a different administrative authority (e.g., 737 as for worldwide aviation service providers such as ARINC, SITA, 738 Inmarsat, etc.). Individual INETs may also themselves be partitioned 739 internally, in which case each internal partition is seen as a 740 separate segment. 742 The addressing plan of each segment is consistent internally but will 743 often bear no relation to the addressing plans of other segments. 744 Each segment is also likely to be separated from others by network 745 security devices (e.g., firewalls, proxies, packet filtering 746 gateways, etc.), and in many cases disjoint segments may not even 747 have any common physical link connections. Therefore, nodes can only 748 be assured of exchanging packets directly with correspondents in the 749 same segment, and not with those in other segments. The only means 750 for joining the segments therefore is through inter-domain peerings 751 between AERO Bridges. 753 The same as for traditional campus LANs, multiple AERO link segments 754 can be joined into a single unified link via a virtual bridging 755 service termed "The SPAN". The SPAN performs link-layer packet 756 forwarding between segments (i.e., bridging) without decrementing the 757 network-layer TTL/Hop Limit. The SPAN model is depicted in Figure 2: 759 . . . . . . . . . . . . . . . . . . . . . . . 760 . . 761 . .-(::::::::) . 762 . .-(::::::::::::)-. +-+ . 763 . (:::: Segment A :::)--|B|---+ . 764 . `-(::::::::::::)-' +-+ | . 765 . `-(::::::)-' | . 766 . | . 767 . .-(::::::::) | . 768 . .-(::::::::::::)-. +-+ | . 769 . (:::: Segment B :::)--|B|---+ . 770 . `-(::::::::::::)-' +-+ | . 771 . `-(::::::)-' | . 772 . | . 773 . .-(::::::::) | . 774 . .-(::::::::::::)-. +-+ | . 775 . (:::: Segment C :::)--|B|---+ . 776 . `-(::::::::::::)-' +-+ | . 777 . `-(::::::)-' | . 778 . | . 779 . ..(etc).. x . 780 . . 781 . . 782 . <- AERO Link Bridged by the SPAN -> . 783 . . . . . . . . . . . . . .. . . . . . . . . 785 Figure 2: The SPAN 787 To support the SPAN, AERO links use the Unique Local Address (ULA) 788 prefix fd80::/10 [RFC4193] as the SPAN Service Prefix (SSP). The 789 prefix length intentionally matches the IPv6 link-local prefix 790 (fe80::/10), and enables a simple stateless translation between AERO 791 and SPAN addresses. Additional SSPs can be employed to identify 792 distinct SPAN Routing Topologies (SRTs) (see: Section 3.5.1). 794 Each segment in the SPAN assigns a unique sub-prefix of SSP::/96 795 termed the "SPAN Partition Prefix (SPP)". For example, for fd80::/96 796 a first segment could assign fd80::1000/116, a second could assign 797 fd80::2000/116, a third could assign fd80::3000/116, etc. The 798 administrative authorities for each segment must therefore coordinate 799 to assure mutually-exclusive SPP assignments, but internal 800 provisioning of the SPP is an independent local consideration for 801 each administrative authority. 803 SPAN addresses are formed by simply rewriting the upper 16 bits of 804 the corresponding AERO address. For example: 806 o the SPAN address formed from the IPv6 Client AERO address 807 fe80:2001:db8:1000:2000:: is simply fd80:2001:db8:1000:2000:: 809 o the SPAN address formed from the IPv4-compatible Client AERO 810 address fe80::ffff:192.0.2.1 is simply fd80::ffff:192.0.2.1 812 o the SPAN address formed from the administrative AERO address 813 fe80::1001 is simply fd80::1001. 815 AERO Bridges join multiple segments into a unified AERO link over 816 multiple diverse administrative domains. They support a bridging 817 function by first establishing forwarding table entries for their 818 SPPs either via standard BGP routing or static routes. For example, 819 if three Bridges ('A', 'B' and 'C') from different segments serviced 820 the SPPs fd80::1000/116, fd80::2000/116 and fd80::3000/116 821 respectively, then the forwarding tables in each Bridge are as 822 follows: 824 A: fd80::1000/116->local, fd80::2000/116->B, fd80::3000/116->C 826 B: fd80::1000/116->A, fd80::2000/116->local, fd80::3000/116->C 828 C: fd80::1000/116->A, fd80::2000/116->B, fd80::3000/116->local 830 These forwarding table entries are permanent and never change, since 831 they correspond to fixed infrastructure elements in their respective 832 segments. 834 SPAN Client Prefixes (SCPs) are instead dynamically advertised in the 835 AERO link routing system by Servers and Relays that provide service 836 for their corresponding MNPs. For example, if three Servers ('D', 837 'E' and 'F') service the MNPs 2001:db8:1000:2000::/56, 838 2001:db8:3000:4000::/56 and 2001:db8:5000:6000::/56 then the routing 839 system would include: 841 D: fd80:2001:db8:1000:2000::/72 843 E: fd80:2001:db8:3000:4000::/72 845 F: fd80:2001:db8:5000:6000::/72 847 With the SCPs and SPPs in place in each Bridge's forwarding table, 848 control and data packets sent between AERO nodes in different 849 segments can therefore be carried over the via encapsulation in a 850 mid-layer IPv6 header known as the "SPAN header". For example, when 851 a source AERO node forwards a packet with IPv6 address 852 2001:db8:1:2::1 to a target AERO node with IPv6 address 853 2001:db8:1000:2000::1, it first encapsulates the packet in a SPAN 854 header with source address set to fd80:2001:db8:1:2:: and destination 855 address set to fd80:2001:db8:1000:2000::. Next, it encapsulates the 856 resulting SPAN packet in an INET header with source address set to 857 its own INET address (e.g., 192.0.2.100) and destination set to the 858 INET address of a Bridge (e.g., 192.0.2.1). 860 SPAN encapsulation is based on Generic Packet Tunneling in IPv6 861 [RFC2473]; the encapsulation format in the above example is shown in 862 Figure 3: 864 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 865 | INET Header | 866 | src = 192.0.2.100 | 867 | dst = 192.0.2.1 | 868 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 869 | SPAN Header | 870 | src = fd80:2001:db8:1:2:: | 871 | dst=fd80:2001:db8:1000:2000:: | 872 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 873 | Inner IP Header | 874 | src = 2001:db8:1:2::1 | 875 | dst = 2001:db8:1000:2000::1 | 876 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 877 | | 878 ~ ~ 879 ~ Inner Packet Body ~ 880 ~ ~ 881 | | 882 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 884 Figure 3: SPAN Encapsulation 886 In this format, the inner IP header and packet body are the original 887 IP packet, the SPAN header is an IPv6 header prepared according to 888 [RFC2473], and the INET header is prepared as discussed in 889 Section 3.9. A packet is said to be "forwarded/sent into the SPAN" 890 when it is encapsulated as described above then forwarded via a 891 secured tunnel to a neighboring Bridge. 893 This gives rise to a routing system that contains both SCP routes 894 that may change dynamically due to regional node mobility and SPP 895 routes that never change. The Bridges can therefore provide link- 896 layer bridging by sending packets into the SPAN instead of network- 897 layer routing according to MNP routes. As a result, opportunities 898 for packet loss due to node mobility between different segments are 899 mitigated. 901 In normal operations, IPv6 ND messages are conveyed to SPP addresses 902 over the SPAN so that specific Proxys, Servers or Relays can be 903 addressed without being subject to mobility events. Conversely, only 904 the first few packets destined to SCP addresses traverse the SPAN 905 until route optimization can determine a more direct path. 907 3.5.1. SPAN Routing Topologies 909 In some cases (e.g., when redundant topologies are needed for fault 910 tolerance and reliability) it may be beneficial to deploy multiple 911 SPAN Routing Topologies (SRTs) that act as independent overlay 912 instances. A communication failure in one instance therefore will 913 not affect communications in other instances. 915 This document asserts that up to four SRTs provide a level of safety 916 sufficient for critical communications such as civil aviation. Each 917 SRT is designated with a color that identifies a different SSP as 918 follows: 920 o Red (default) - corresponds to the SSP fd80::/16 922 o Green - corresponds to the SSP fd81::/16 924 o Blue-1 - corresponds to the SSP fd82::/16 926 o Blue-2 - corresponds to SSP fd83::/16 928 o SSPs fd84::/16 through fdbf::/16 are reserved for future use. 930 3.5.2. Segment Routing Over the SPAN 932 As discussed in the following sections, Segment Routing is used over 933 the SPAN to influence the path of packets destined to Clients on INET 934 interfaces without causing all packets to traverse the Client's 935 Server. When a Client, Proxy or Server has a packet to send to a 936 target discovered through route optimization located in the same SPAN 937 segment, it encapsulates the packet in a SPAN header with the SPAN 938 address of the target as the destination address, then uses the 939 target's Link Layer Address information for INET encapsulation. 941 When a Client, Proxy or Server has a packet to send to a route 942 optimization target located in a different SPAN segment, it 943 encapsulates the packet in a SPAN header with the SPAN address of the 944 target's Server as the destination. The node also includes a Segment 945 Routing Header (SRH) [RFC8754] with the SPAN address of the target as 946 the penultimate address and with the IP encapsulation address of the 947 target as the ultimate address. (When the encapsulation address is 948 an IPv6 address and a port number is included, the port number is 949 written into the SRH Tag field.) The node then forwards the packet 950 into the SPAN, which will eventually direct it to a Bridge on the 951 same segment as the target's Server. 953 When a Bridge on the same segment as the target's Server receives a 954 SPAN-encapsulated packet destined to the target Server, it looks 955 ahead into the Segment Routing List to determine that the penultimate 956 destination is set to the target's SPAN address and the ultimate 957 destination is set to the target's Link Layer Address. The Bridge 958 then advances the SPAN destination address to the target's SPAN 959 address and encapsulates the SPAN packet in an INET header based on 960 the target's Link Layer Address, then forwards the packet to the 961 target directly while bypassing the target's Server. In this way, 962 the Bridge participates in route optimization to greatly reduce 963 traffic load and suboptimal routing through the target's Server. 965 3.6. AERO Interface Characteristics 967 AERO interfaces are virtual interfaces configured over one or more 968 underlying interfaces classified as follows: 970 o INET interfaces connect to an INET either natively or through one 971 or several IPv4 Network Address Translators (NATs). Native INET 972 interfaces have global IP addresses that are reachable from any 973 INET correspondent. All Server, Relay and Bridge interfaces are 974 native interfaces, as are INET-facing interfaces of Proxys. NATed 975 INET interfaces connect to a private network behind one or more 976 NATs that provide INET access. Clients that are behind a NAT are 977 required to send periodic keepalive messages to keep NAT state 978 alive when there are no data packets flowing. 980 o Proxyed interfaces connect to an ANET that is separated from the 981 open INET by an AERO Proxy. Proxys can actively issue control 982 messages over the INET on behalf of the Client to reduce ANET 983 congestion. Clients connected to Proxyed interfaces receive RAs 984 with the P flag set to 1. 986 o VPNed interfaces use security encapsulation over the INET to a 987 Virtual Private Network (VPN) server that also acts as an AERO 988 Server. Clients connected to VPNed interfaces receive RAs with 989 the P flag set to 1 the same as for Proxyed interfaces. Other 990 than the link-layer encapsulation format, VPNed interfaces behave 991 the same as Direct interfaces. 993 o Direct interfaces connect a Client directly to a Server without 994 crossing any ANET/INET paths. An example is a line-of-sight link 995 between a remote pilot and an unmanned aircraft. The same Client 996 considerations apply as for VPNed interfaces above, and the Client 997 receives RA messages with the P flag set to 1. 999 AERO interfaces use SPAN-layer encapsulation as necessary as 1000 discussed in Section 3.5. AERO interfaces use link-layer 1001 encapsulation (see: Section 3.9) to exchange packets with AERO link 1002 neighbors over INET or VPNed interfaces. AERO interfaces do not use 1003 encapsulation over Proxyed and Direct underlying interfaces. 1005 AERO interfaces maintain a neighbor cache for tracking per-neighbor 1006 state the same as for any interface. AERO interfaces use ND messages 1007 including Router Solicitation (RS), Router Advertisement (RA), 1008 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 1009 neighbor cache management. 1011 AERO interfaces send ND messages with an Overlay Multilink Network 1012 Interface (OMNI) option formatted as specified in 1013 [I-D.templin-6man-omni-interface]. The OMNI option includes prefix 1014 registration information and "ifIndex-tuples" containing link 1015 information parameters for the AERO interface's underlying 1016 interfaces. 1018 When encapsulation is used, AERO interface ND messages MAY also 1019 include an AERO Source/Target Link-Layer Address Option (S/TLLAO) 1020 formatted as shown in Figure 4: 1022 0 1 2 3 1023 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 1024 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1025 | Type | Length | ifIndex[1] | SRT | LHS |FMT| 1026 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1027 ~ Segment Routing List [1] ~ 1028 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1029 ~ Link Layer Address [1] ~ 1030 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1031 | Port Number [1] | ifIndex[2] | SRT | LHS |FMT| 1032 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1033 ~ Segment Routing List [2] ~ 1034 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1035 ~ Link Layer Address [2] ~ 1036 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1037 | Port Number [2] | .... ~ 1038 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 1039 ~ ... ~ 1040 ~ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1041 ~ | ifIndex[N] | SRT | LHS |FMT| 1042 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1043 ~ Segment Routing List [N] ~ 1044 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1045 ~ Link Layer Address [N] ~ 1046 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1047 | Port Number [N] | Zero Padding (if necessary) ... 1048 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+--+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1050 Figure 4: AERO Source/Target Link-Layer Address Option (S/TLLAO) 1051 Format 1053 In this format, Type and Length are set the same as specified for S/ 1054 TLLAOs in [RFC4861], with trailing zero padding octets added as 1055 necessary to produce an integral number of 8 octet blocks. The S/ 1056 TLLAO includes N ifIndex-tuples in correspondence to ifIndex-tuples 1057 that appear in the OMNI option. Each ifIndex-tuple includes the 1058 following information: 1060 o ifIndex[i] - the same value as in the corresponding ifIndex-tuple 1061 included in the OMNI option. 1063 o SRT[i] - a 2-bit "SPAN Routing Topology" value (see: 1064 Section 3.5.1) coded as follows: 1066 * 000 - Red 1068 * 001 - Green 1069 * 010 - Blue-1 1071 * 011 - Blue-2 1073 * 100 - 111 - Reserved 1075 o LHS[i] - a 3-bit "LookaHead Segments" value that encodes the 1076 number (from 0 to 7) of entries in Segment Routing List [i]. 1078 o FMT[i] - a 2-bit "Format" code. Determines the format of the Link 1079 Layer Address [i] field as follows: 1081 * 00 - Link Layer Address [i] encodes a Teredo-format AERO 1082 address for a node behind a NAT. 1084 * 01 - Link Layer Address [i] encodes a Teredo-format AERO 1085 address for a node on the open INET. 1087 * 10 - Link Layer Address [i] encodes a native IPv6 address. 1089 * 11 - Link Layer Address [i] encodes a native IPv6 address with 1090 Port Number [i] field included. 1092 o Segment Routing List [i] - Includes LHS[i]-many 16 byte SPAN 1093 addresses corresponding to the Segment IDs (SIDs) that must be 1094 visited prior to forwarding to Link Layer Address [i]. The 1095 ultimate SID appears first, followed by the penultimate SID 1096 second, etc. 1098 o Link Layer Address [i] - Included according to FMT[i], and 1099 identifies the link-layer address of the source/target. 1101 o Port Number [i] - Present only when FMT[i] is 11. When present, 1102 the field is 2 bytes in length and immediately follows Link Layer 1103 Address [i]. Encodes the upper layer protocol port number to be 1104 used as the encapsulation source port. 1106 If an S/TLLAO is included, any ifIndex-tuples correspond to a proper 1107 subset of the OMNI option ifIndex-tuples. Any S/TLLAO ifIndex-tuple 1108 having an ifIndex value that does not appear in an OMNI option 1109 ifindex-tuple is ignored. If the same ifIndex value appears in 1110 multiple ifIndex-tuples, the first tuple is processed and the 1111 remaining tuples are ignored. Any S/TLLAO ifIndex-tuples can 1112 therefore be viewed as inter-dependent extensions of their 1113 corresponding OMNI option ifIndex-tuples, i.e., the OMNI option and 1114 S/TLLAO are companions that are interpreted in conjunction with each 1115 other. 1117 A Client's AERO interface may be configured over multiple underlying 1118 interface connections. For example, common mobile handheld devices 1119 have both wireless local area network ("WLAN") and cellular wireless 1120 links. These links are often used "one at a time" with low-cost WLAN 1121 preferred and highly-available cellular wireless as a standby, but a 1122 simultaneous-use capability could provide benefits. In a more 1123 complex example, aircraft frequently have many wireless data link 1124 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1125 directional, etc.) with diverse performance and cost properties. 1127 If a Client's multiple underlying interfaces are used "one at a time" 1128 (i.e., all other interfaces are in standby mode while one interface 1129 is active), then ND message OMNI options include only a single 1130 ifIndex-tuple set to constant values. In that case, the Client would 1131 appear to have a single interface but with a dynamically changing 1132 link-layer address. 1134 If the Client has multiple active underlying interfaces, then from 1135 the perspective of ND it would appear to have multiple link-layer 1136 addresses. In that case, ND message OMNI options MAY include 1137 multiple ifIndex-tuples - each with values that correspond to a 1138 specific interface. Every ND message need not include all OMNI and/ 1139 or S/TLLAO ifIndex-tuples; for any ifIndex-tuple not included, the 1140 neighbor considers the status as unchanged. 1142 Bridge, Server and Proxy AERO interfaces may be configured over one 1143 or more secured tunnel interfaces. The AERO interface configures 1144 both an AERO address and its corresponding SPAN address, while the 1145 underlying secured tunnel interfaces are either unnumbered or 1146 configure the same SPAN address. The AERO interface encapsulates 1147 each IP packet in a SPAN header and presents the packet to the 1148 underlying secured tunnel interface. For Bridges that do not 1149 configure an AERO interface, the secured tunnel interfaces themselves 1150 are exposed to the IP layer with each interface configuring the 1151 Bridge's SPAN address. Routing protocols such as BGP therefore run 1152 directly over the Bridge's secured tunnel interfaces. For nodes that 1153 configure an AERO interface, routing protocols such as BGP run over 1154 the AERO interface but do not employ SPAN encapsulation. Instead, 1155 the AERO interface presents the routing protocol messages directly to 1156 the underlying secured tunnels without applying encapsulation and 1157 while using the SPAN address as the source address. This distinction 1158 must be honored consistently according to each node's configuration 1159 so that the IP forwarding table will associate discovered IP routes 1160 with the correct interface. 1162 3.7. AERO Interface Initialization 1164 AERO Servers, Proxys and Clients configure AERO interfaces as their 1165 point of attachment to the AERO link. AERO nodes assign the MSPs for 1166 the link to their AERO interfaces (i.e., as a "route-to-interface") 1167 to ensure that packets with destination addresses covered by an MNP 1168 not explicitly assigned to a non-AERO interface are directed to the 1169 AERO interface. 1171 AERO interface initialization procedures for Servers, Proxys, Clients 1172 and Bridges are discussed in the following sections. 1174 3.7.1. AERO Server/Relay Behavior 1176 When a Server enables an AERO interface, it assigns AERO/SPAN 1177 addresses appropriate for the given SPAN segment. The Server also 1178 configures secured tunnels with one or more neighboring Bridges and 1179 engages in a BGP routing protocol session with each Bridge. 1181 The AERO interface provides a single interface abstraction to the IP 1182 layer, but internally comprises multiple secured tunnels as well as 1183 an NBMA nexus for sending encapsulated data packets to AERO interface 1184 neighbors. The Server further configures a service to facilitate ND/ 1185 PD exchanges with AERO Clients and manages per-Client neighbor cache 1186 entries and IP forwarding table entries based on control message 1187 exchanges. 1189 Relays are simply Servers that run a dynamic routing protocol to 1190 redistribute routes between the AERO interface and INET/EUN 1191 interfaces (see: Section 3.3). The Relay provisions MNPs to networks 1192 on the INET/EUN interfaces (i.e., the same as a Client would do) and 1193 advertises the MSP(s) for the AERO link over the INET/EUN interfaces. 1194 The Relay further provides an attachment point of the AERO link to a 1195 non-MNP-based global topology. 1197 3.7.2. AERO Proxy Behavior 1199 When a Proxy enables an AERO interface, it assigns AERO/SPAN 1200 addresses and configures permanent neighbor cache entries the same as 1201 for Servers. The Proxy also configures secured tunnels with one or 1202 more neighboring Bridges and maintains per-Client neighbor cache 1203 entries based on control message exchanges. 1205 3.7.3. AERO Client Behavior 1207 When a Client enables an AERO interface, it sends RS messages with 1208 ND/PD parameters over its underlying interfaces to a Server in the 1209 MAP list, which returns an RA message with corresponding parameters. 1211 (The RS/RA messages may pass through a Proxy in the case of a 1212 Client's Proxyed interface, or through one or more NATs in the case 1213 of a Client's INET interface.) 1215 3.7.4. AERO Bridge Behavior 1217 AERO Bridges need not connect directly to the AERO link, since they 1218 operate as link-layer forwarding devices instead of network layer 1219 routers. Configuration of AERO interfaces on Bridges is therefore 1220 OPTIONAL, e.g., if an administrative interface is needed. Bridges 1221 configure secured tunnels with Servers, Proxys and other Bridges; 1222 they also configure AERO/SPAN addresses and permanent neighbor cache 1223 entries the same as Servers. Bridges engage in a BGP routing 1224 protocol session with a subset of the Servers on the local SPAN 1225 segment, and with other Bridges on the SPAN (see: Section 3.3). 1227 3.8. AERO Interface Neighbor Cache Maintenance 1229 Each AERO interface maintains a conceptual neighbor cache that 1230 includes an entry for each neighbor it communicates with on the AERO 1231 link per [RFC4861]. AERO interface neighbor cache entries are said 1232 to be one of "permanent", "symmetric", "asymmetric" or "proxy". 1234 Permanent neighbor cache entries are created through explicit 1235 administrative action; they have no timeout values and remain in 1236 place until explicitly deleted. AERO Bridges maintain permanent 1237 neighbor cache entries for their associated Proxys and Servers (and 1238 vice-versa). Each entry maintains the mapping between the neighbor's 1239 network-layer AERO address and corresponding INET address. 1241 Symmetric neighbor cache entries are created and maintained through 1242 RS/RA exchanges as specified in Section 3.15, and remain in place for 1243 durations bounded by ND/PD lifetimes. AERO Servers maintain 1244 symmetric neighbor cache entries for each of their associated 1245 Clients, and AERO Clients maintain symmetric neighbor cache entries 1246 for each of their associated Servers. The list of all Servers on the 1247 AERO link is maintained in the link's MAP list. 1249 Asymmetric neighbor cache entries are created or updated based on 1250 route optimization messaging as specified in Section 3.17, and are 1251 garbage-collected when keepalive timers expire. AERO ROSs maintain 1252 asymmetric neighbor cache entries for active targets with lifetimes 1253 based on ND messaging constants. Asymmetric neighbor cache entries 1254 are unidirectional since only the ROS (and not the ROR) creates an 1255 entry. 1257 Proxy neighbor cache entries are created and maintained by AERO 1258 Proxys when they process Client/Server ND/PD exchanges, and remain in 1259 place for durations bounded by ND/PD lifetimes. AERO Proxys maintain 1260 proxy neighbor cache entries for each of their associated Clients. 1261 Proxy neighbor cache entries track the Client state and the address 1262 of the Client's associated Server(s). 1264 To the list of neighbor cache entry states in Section 7.3.2 of 1265 [RFC4861], Proxy and Server AERO interfaces add an additional state 1266 DEPARTED that applies to symmetric and proxy neighbor cache entries 1267 for Clients that have recently departed. The interface sets a 1268 "DepartTime" variable for the neighbor cache entry to "DEPART_TIME" 1269 seconds. DepartTime is decremented unless a new ND message causes 1270 the state to return to REACHABLE. While a neighbor cache entry is in 1271 the DEPARTED state, packets destined to the target Client are 1272 forwarded to the Client's new location instead of being dropped. 1273 When DepartTime decrements to 0, the neighbor cache entry is deleted. 1274 It is RECOMMENDED that DEPART_TIME be set to the default constant 1275 value REACHABLE_TIME plus 10 seconds (40 seconds by default) to allow 1276 a window for packets in flight to be delivered while stale route 1277 optimization state may be present. 1279 When an ROR receives an authentic NS message used for route 1280 optimization, it searches for a symmetric neighbor cache entry for 1281 the target Client. The ROR then returns a solicited NA message 1282 without creating a neighbor cache entry for the ROS, but creates or 1283 updates a target Client "Report List" entry for the ROS and sets a 1284 "ReportTime" variable for the entry to REPORT_TIME seconds. The ROR 1285 resets ReportTime when it receives a new authentic NS message, and 1286 otherwise decrements ReportTime while no authentic NS messages have 1287 been received. It is RECOMMENDED that REPORT_TIME be set to the 1288 default constant value REACHABLE_TIME plus 10 seconds (40 seconds by 1289 default) to allow a window for route optimization to converge before 1290 ReportTime decrements below REACHABLE_TIME. 1292 When the ROS receives a solicited NA message response to its NS 1293 message used for route optimization, it creates or updates an 1294 asymmetric neighbor cache entry for the target network-layer and 1295 link-layer addresses. The ROS then (re)sets ReachableTime for the 1296 neighbor cache entry to REACHABLE_TIME seconds and uses this value to 1297 determine whether packets can be forwarded directly to the target, 1298 i.e., instead of via a default route. The ROS otherwise decrements 1299 ReachableTime while no further solicited NA messages arrive. It is 1300 RECOMMENDED that REACHABLE_TIME be set to the default constant value 1301 30 seconds as specified in [RFC4861]. 1303 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1304 of NS keepalives sent when a correspondent may have gone unreachable, 1305 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1306 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1307 to limit the number of unsolicited NAs that can be sent based on a 1308 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1309 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1310 same as specified in [RFC4861]. 1312 Different values for DEPART_TIME, REPORT_TIME, REACHABLE_TIME, 1313 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1314 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1315 different values are chosen, all nodes on the link MUST consistently 1316 configure the same values. Most importantly, DEPART_TIME and 1317 REPORT_TIME SHOULD be set to a value that is sufficiently longer than 1318 REACHABLE_TIME to avoid packet loss due to stale route optimization 1319 state. 1321 3.9. AERO Interface Encapsulation and Re-encapsulation 1323 In some instances, AERO interfaces insert a mid-layer IPv6 header 1324 known as the SPAN header as discussed in the following sections. 1325 After either inserting or omitting the SPAN header, the AERO 1326 interface inserts an outer encapsulation header as discussed below. 1328 AERO interfaces avoid outer encapsulation over Direct underlying 1329 interfaces and Proxyed underlying interfaces for which the first-hop 1330 access router is AERO-aware. Other AERO interfaces encapsulate 1331 packets according to whether they are entering the AERO interface 1332 from the network layer or if they are being re-admitted into the same 1333 AERO link they arrived on. This latter form of encapsulation is 1334 known as "re-encapsulation". 1336 For packets entering the AERO interface from the network layer, the 1337 AERO interface copies the "TTL/Hop Limit", "Type of Service/Traffic 1338 Class" [RFC2983], "Flow Label"[RFC6438] (for IPv6) and "Congestion 1339 Experienced" [RFC3168] values in the inner packet's IP header into 1340 the corresponding fields in the SPAN and outer encapsulation 1341 header(s). 1343 For packets undergoing re-encapsulation, the AERO interface instead 1344 copies these values from the original encapsulation header into the 1345 new encapsulation header, i.e., the values are transferred between 1346 encapsulation headers and *not* copied from the encapsulated packet's 1347 network-layer header. (Note especially that by copying the TTL/Hop 1348 Limit between encapsulation headers the value will eventually 1349 decrement to 0 if there is a (temporary) routing loop.) For IPv4 1350 encapsulation/re-encapsulation, the AERO interface sets the DF bit as 1351 discussed in Section 3.12. 1353 AERO interfaces configured over INET underlying interfaces 1354 encapsulate packets in INET headers according to the next hop 1355 determined in the forwarding algorithm in Section 3.13. If the next 1356 hop is reached via a secured tunnel, the AERO interface uses an 1357 encapsulation format specific to the secured tunnel type (see: 1358 Section 6). If the next hop is reached via an unsecured underlying 1359 interface, the AERO interface instead uses UDP/IP encapsulation 1360 according to the Teredo format specified in[RFC4380] and as extended 1361 in [RFC6081]. 1363 When Teredo encapsulation is used, the AERO interface next sets the 1364 UDP source port to a constant value that it will use in each 1365 successive packet it sends, and sets the UDP length field to the 1366 length of the encapsulated packet plus 8 bytes for the UDP header 1367 itself plus the length of any included Teredo extension headers or 1368 trailers. For packets sent to a Server or Bridge, the AERO interface 1369 sets the UDP destination port to 8060, i.e., the IANA-registered port 1370 number for AERO. For packets sent to a Client, the AERO interface 1371 sets the UDP destination port to the port value stored in the 1372 neighbor cache entry for this Client. The AERO interface finally 1373 includes/omits the UDP checksum according to [RFC6935][RFC6936]. 1375 AERO interfaces observe the packet sizing and fragmentation 1376 considerations found in Section 3.12. 1378 3.10. AERO Interface Decapsulation 1380 AERO interfaces decapsulate packets destined either to the AERO node 1381 itself or to a destination reached via an interface other than the 1382 AERO interface the packet was received on. When the encapsulated 1383 packet arrives in multiple SPAN fragments, the AERO interface 1384 reassembles as discussed in Section 3.12. Further decapsulation 1385 steps are performed according to the appropriate encapsulation format 1386 specification. 1388 3.11. AERO Interface Data Origin Authentication 1390 AERO nodes employ simple data origin authentication procedures. In 1391 particular: 1393 o AERO Bridges, Servers and Proxys accept encapsulated data packets 1394 and control messages received from secured tunnels via the SPAN. 1396 o AERO Proxys and Clients accept packets that originate from within 1397 the same secured ANET. 1399 o AERO Clients and Relays accept packets from downstream network 1400 correspondents based on ingress filtering. 1402 o AERO Clients, Relays and Servers verify the outer UDP/IP 1403 encapsulation addresses according to the Teredo specification 1404 [RFC4380]. 1406 AERO nodes silently drop any packets that do not satisfy the above 1407 data origin authentication procedures. Further security 1408 considerations are discussed in Section 6. 1410 3.12. AERO Interface MTU and Fragmentation 1412 IPv6 underlying interfaces are REQUIRED to configure a minimum 1413 Maximum Transmission Unit (MTU) of 1280 bytes [RFC8200]. The minimum 1414 MTU for IPv4 underlying interfaces is only 68 bytes [RFC1122], 1415 meaning that a packet smaller than the IPv6 MTU may require 1416 fragmentation when IPv4 encapsulation is used. Therefore, the Don't 1417 Fragment (DF) bit in the IPv4 encapsulation header MUST be set to 0. 1419 The AERO interface configures an MTU of 9180 bytes [RFC2492]; the 1420 size is therefore not a reflection of the underlying interface MTUs, 1421 but rather determines the largest packet the AERO interface can 1422 forward or reassemble. The AERO interface therefore accommodates IP 1423 packets up to 9180 bytes while generating IPv6 Path MTU Discovery 1424 (PMTUD) Packet Too Big (PTB) messages [RFC8201] as necessary (see 1425 below). 1427 AERO interfaces employ mid-layer IPv6 encapsulation and 1428 fragmentation/reassembly per [RFC2473] (aka "SPAN encapsulation") to 1429 accommodate the 9180 byte MTU. The AERO interface returns 1430 internally-generated PTB messages for packets admitted into the 1431 interface that it deems too large (e.g., according to link 1432 performance characteristics, reassembly cost, etc.) while either 1433 dropping or forwarding the packet as necessary. The AERO interface 1434 performs PMTUD even if the destination appears to be on the same link 1435 since intermediate AERO link nodes may return a PTB. This ensures 1436 that the path MTU is adaptive and reflects the current path used for 1437 a given data flow. 1439 AERO nodes perform SPAN encapsulation and fragmentation/reassembly as 1440 follows: 1442 o When a node's AERO interface sends a packet over a Proxyed, VPNed 1443 or Direct underlying interface, it sends without SPAN 1444 encapsulation if the packet is no larger than the underlying 1445 interface MTU. Otherwise, it inserts a SPAN header with source 1446 address set to the node's own SPAN address and destination set to 1447 the SPAN address of the link-layer peer Proxy, Server or Client on 1448 the underlying interface. The AERO interface then uses IPv6 1449 fragmentation to break the packet into a minimum number of non- 1450 overlapping fragments, where the largest fragment size is 1451 determined by the underlying interface MTU and the smallest 1452 fragment is no smaller than 640 bytes. The AERO interface then 1453 sends the fragments to the link-layer peer, which reassembles 1454 before forwarding toward the final destination. 1456 o When a node's AERO interface sends a packet over an INET 1457 underlying interface, it sends encapsulated packets no larger than 1458 1280 bytes without a SPAN header if the destination is reached via 1459 an INET address within the same SPAN segment. Otherwise, it 1460 inserts a SPAN header with source address set to the node's SPAN 1461 address, destination set to the SPAN address of the next hop AERO 1462 node toward the final destination and (if necessary) with a SRH 1463 with the remaining Segment IDs on the path to the final 1464 destination. The AERO interface then uses IPv6 fragmentation to 1465 break the encapsulated packet into a minimum number of non- 1466 overlapping fragments, where the largest fragment size (including 1467 both SPAN and INET encapsulation) is 1280 bytes and the smallest 1468 fragment is no smaller than 640 bytes. The AERO interface then 1469 encapsulates the SPAN fragments in INET headers and sends them to 1470 the SPAN destination, which reassembles before forwarding toward 1471 the final destination. 1473 In order to avoid a "tiny fragment" attack, AERO interfaces 1474 unconditionally drop all SPAN fragments smaller than 640 bytes. In 1475 order to set the correct context for reassembly, the AERO interface 1476 that inserts a SPAN header MUST also be the one that inserts the IPv6 1477 Fragment Header Identification value. Although all fragments of the 1478 same fragmented SPAN packet are typically sent via the same 1479 underlying interface, this is not strictly required since all 1480 fragments will arrive at the AERO interface that performs reassembly 1481 even if they travel over different paths. 1483 Note that the AERO interface can forward large packets via 1484 encapsulation and fragmentation while at the same time returning 1485 advisory PTB messages, e.g., subject to rate limiting. The receiving 1486 node that performs reassembly can also send advisory PTB messages if 1487 reassembly conditions become unfavorable. The AERO interface can 1488 therefore continuously forward large packets without loss while 1489 returning advisory messages recommending a smaller size (but no 1490 smaller than 1280). Advisory PTB messages are differentiated from 1491 PTB messages that report loss by setting the Code field in the ICMPv6 1492 message header to the value 1. This document therefore updates 1493 [RFC4443] and [RFC8201]. 1495 3.13. AERO Interface Forwarding Algorithm 1497 IP packets enter a node's AERO interface either from the network 1498 layer (i.e., from a local application or the IP forwarding system) or 1499 from the link layer (i.e., from an AERO interface neighbor). All 1500 packets entering a node's AERO interface first undergo data origin 1501 authentication as discussed in Section 3.11. Packets that satisfy 1502 data origin authentication are processed further, while all others 1503 are dropped silently. 1505 Packets that enter the AERO interface from the network layer are 1506 forwarded to an AERO interface neighbor. Packets that enter the AERO 1507 interface from the link layer are either re-admitted into the AERO 1508 link or forwarded to the network layer where they are subject to 1509 either local delivery or IP forwarding. In all cases, the AERO 1510 interface itself MUST NOT decrement the network layer TTL/Hop-count 1511 since its forwarding actions occur below the network layer. 1513 AERO interfaces may have multiple underlying interfaces and/or 1514 neighbor cache entries for neighbors with multiple ifIndex-tuple 1515 registrations (see Section 3.6). The AERO interface uses traffic 1516 classifiers (e.g., DSCP value, port number, etc.) to select an 1517 outgoing underlying interface for each packet based on the node's own 1518 QoS preferences, and also to select a destination link-layer address 1519 based on the neighbor's underlying interface with the highest 1520 preference. AERO implementations SHOULD allow for QoS preference 1521 values to be modified at runtime through network management. 1523 If multiple outgoing interfaces and/or neighbor interfaces have a 1524 preference of "high", the AERO node replicates the packet and sends 1525 one copy via each of the (outgoing / neighbor) interface pairs; 1526 otherwise, the node sends a single copy of the packet via an 1527 interface with the highest preference. AERO nodes keep track of 1528 which underlying interfaces are currently "reachable" or 1529 "unreachable", and only use "reachable" interfaces for forwarding 1530 purposes. 1532 The following sections discuss the AERO interface forwarding 1533 algorithms for Clients, Proxys, Servers and Bridges. In the 1534 following discussion, a packet's destination address is said to 1535 "match" if it is the same as a cached address, or if it is covered by 1536 a cached prefix (which may be encoded in an AERO address). 1538 3.13.1. Client Forwarding Algorithm 1540 When an IP packet enters a Client's AERO interface from the network 1541 layer the Client searches for an asymmetric neighbor cache entry that 1542 matches the destination. If there is a match, the Client uses one or 1543 more "reachable" neighbor interfaces in the entry for packet 1544 forwarding. If there is no asymmetric neighbor cache entry, the 1545 Client instead forwards the packet toward a Server (the packet is 1546 intercepted by a Proxy if there is a Proxy on the path). The Client 1547 encapsulates the packet in a SPAN header and fragments if necessary 1548 according to MTU requirements (see: Section 3.12). 1550 When an IP packet enters a Client's AERO interface from the link- 1551 layer, if the destination matches one of the Client's MNPs or link- 1552 local addresses the Client reassembles and decapsulates as necessary 1553 and delivers the inner packet to the network layer. Otherwise, the 1554 Client drops the packet and MAY return a network-layer ICMP 1555 Destination Unreachable message subject to rate limiting (see: 1556 Section 3.14). 1558 3.13.2. Proxy Forwarding Algorithm 1560 For control messages originating from or destined to a Client, the 1561 Proxy intercepts the message and updates its proxy neighbor cache 1562 entry for the Client. The Proxy then forwards a (proxyed) copy of 1563 the control message. (For example, the Proxy forwards a proxied 1564 version of a Client's NS/RS message to the target neighbor, and 1565 forwards a proxied version of the NA/RA reply to the Client.) 1567 When the Proxy receives a data packet from a Client within the ANET, 1568 the Proxy reassembles and re-fragments if necessary then searches for 1569 an asymmetric neighbor cache entry that matches the destination and 1570 forwards as follows: 1572 o if the destination matches an asymmetric neighbor cache entry, the 1573 Proxy uses one or more "reachable" neighbor interfaces in the 1574 entry for packet forwarding using SPAN encapsulation and including 1575 a SRH if necessary according to the cached TLLAO information. If 1576 the neighbor interface is in the same SPAN segment, the Proxy 1577 forwards the packet directly to the neighbor; otherwise, it 1578 forwards the packet to a Bridge. 1580 o else, the Proxy uses SPAN encapsulation and forwards the packet to 1581 a Bridge while using the SPAN address corresponding to the 1582 packet's destination as the SPAN destination address. 1584 When the Proxy receives an encapsulated data packet from an INET 1585 neighbor or from a secured tunnel from a Bridge, it accepts the 1586 packet only if data origin authentication succeeds and if there is a 1587 proxy neighbor cache entry that matches the inner destination. Next, 1588 the Proxy reassembles the packet (if necessary) and continues 1589 processing. 1591 Next if reassembly is complete and the neighbor cache state is 1592 REACHABLE, the Proxy returns a PTB if necessary (see: Section 3.12) 1593 then either drops or forwards the packet to the Client while 1594 performing SPAN encapsulation and re-fragmentation to the ANET MTU 1595 size if necessary. If the neighbor cache entry state is DEPARTED, 1596 the Proxy instead changes the SPAN destination address to the address 1597 of the new Server and forwards it to a Bridge while performing re- 1598 fragmentation to 1280 bytes if necessary. 1600 3.13.3. Server/Relay Forwarding Algorithm 1602 For control messages destined to a target Client's AERO address that 1603 are received from a secured tunnel, the Server intercepts the message 1604 and sends an appropriate response on behalf of the Client. (For 1605 example, the Server sends an NA message reply in response to an NS 1606 message directed to one of its associated Clients.) If the Client's 1607 neighbor cache entry is in the DEPARTED state, however, the Server 1608 instead forwards the packet to the Client's new Server as discussed 1609 in Section 3.19. 1611 When the Server receives an encapsulated data packet from an INET 1612 neighbor or from a secured tunnel, it accepts the packet only if data 1613 origin authentication succeeds. If the SPAN destination address is 1614 its own address, the Server continues processing as follows: 1616 o if the destination matches a symmetric neighbor cache entry in the 1617 REACHABLE state the Server prepares the packet for forwarding to 1618 the destination Client. The Server first reassembles (if 1619 necessary) and forwards the packet (while re-fragmenting if 1620 necessary) as specified inSection 3.12. 1622 o else, if the destination matches a symmetric neighbor cache entry 1623 in the DEPARETED state the Server re-encapsulates the packet and 1624 forwards it using the SPAN address of the Client's new Server as 1625 the destination. 1627 o else, if the destination matches an asymmetric neighbor cache 1628 entry, the Server uses one or more "reachable" neighbor interfaces 1629 in the entry for packet forwarding via the local INET if the 1630 neighbor is in the same SPAN segment or using SPAN encapsulation 1631 and Segment Routing if necessary with the final destination set to 1632 the neighbor's SPAN address otherwise. 1634 o else, if the destination is an AERO address that is not assigned 1635 on the AERO interface the Server drops the packet. 1637 o else, the Server (acting as a Relay) reassembles if necessary, 1638 decapsulates the packet and releases it to the network layer for 1639 local delivery or IP forwarding. Based on the information in the 1640 forwarding table, the network layer may return the packet to the 1641 same AERO interface in which case further processing occurs as 1642 below. (Note that this arrangement accommodates common 1643 implementations in which the IP forwarding table is not accessible 1644 from within the AERO interface. If the AERO interface can 1645 directly access the IP forwarding table (such as for in-kernel 1646 implementations) the forwarding table lookup can instead be 1647 performed internally from within the AERO interface itself.) 1649 When the Server's AERO interface receives a data packet from the 1650 network layer or from a VPNed or Direct Client, it performs SPAN 1651 encapsulation and fragmentation if necessary, then processes the 1652 packet according to the network-layer destination address as follows: 1654 o if the destination matches a symmetric or asymmetric neighbor 1655 cache entry the Server processes the packet as above. 1657 o else, the Server encapsulates the packet and forwards it to a 1658 Bridge using its own SPAN address as the source and the SPAN 1659 address corresponding to the destination as the destination. 1661 3.13.4. Bridge Forwarding Algorithm 1663 Bridges forward SPAN-encapsulated packets over secured tunnels the 1664 same as any IP router. When the Bridge receives a SPAN-encapsulated 1665 packet via a secured tunnel, it removes the outer INET header and 1666 searches for a forwarding table entry that matches the SPAN 1667 destination address. The Bridge then processes the packet as 1668 follows: 1670 o if the destination is the SPAN address of a Server located in the 1671 local SPAN partition, the Bridge checks for a SRH. If there is a 1672 SRH with the SPAN address of the final destination as the 1673 penultimate ID and with a Link Layer Address of the final 1674 destination as the ultimate ID, the Bridge copies the SPAN address 1675 of the final destination into the destination address. If the 1676 Link Layer Address does not indicate the presence of a NAT, the 1677 Bridge then forwards the packet directly to the Link Layer Address 1678 using link-layer (UDP/IP) encapsulation. Otherwise, the Bridge 1679 forwards the packet directly to the Server. 1681 o if the destination matches one of the Bridge's own addresses, the 1682 Bridge submits the packet for local delivery. 1684 o else, if the destination matches a forwarding table entry the 1685 Bridge forwards the packet via a secured tunnel to the next hop. 1686 If the destination matches an MSP without matching an MNP, 1687 however, the Bridge instead drops the packet and returns an ICMP 1688 Destination Unreachable message subject to rate limiting (see: 1689 Section 3.14). 1691 o else, the Bridge drops the packet and returns an ICMP Destination 1692 Unreachable as above. 1694 As for any IP router, the Bridge decrements the TTL/Hop Limit when it 1695 forwards the packet. Therefore, only the Hop Limit in the SPAN 1696 header is decremented, and not the TTL/Hop Limit in the inner packet 1697 header. 1699 3.14. AERO Interface Error Handling 1701 When an AERO node admits a packet into the AERO interface, it may 1702 receive link-layer or network-layer error indications. 1704 A link-layer error indication is an ICMP error message generated by a 1705 router in the INET on the path to the neighbor or by the neighbor 1706 itself. The message includes an IP header with the address of the 1707 node that generated the error as the source address and with the 1708 link-layer address of the AERO node as the destination address. 1710 The IP header is followed by an ICMP header that includes an error 1711 Type, Code and Checksum. Valid type values include "Destination 1712 Unreachable", "Time Exceeded" and "Parameter Problem" 1713 [RFC0792][RFC4443]. (AERO interfaces ignore all link-layer IPv4 1714 "Fragmentation Needed" and IPv6 "Packet Too Big" messages since they 1715 only emit packets that are guaranteed to be no larger than the IP 1716 minimum link MTU as discussed in Section 3.12.) 1718 The ICMP header is followed by the leading portion of the packet that 1719 generated the error, also known as the "packet-in-error". For 1720 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1721 much of invoking packet as possible without the ICMPv6 packet 1722 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1723 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1724 "Internet Header + 64 bits of Original Data Datagram", however 1725 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1726 ICMP datagram SHOULD contain as much of the original datagram as 1727 possible without the length of the ICMP datagram exceeding 576 1728 bytes". 1730 The link-layer error message format is shown in Figure 5 (where, "L2" 1731 and "L3" refer to link-layer and network-layer, respectively): 1733 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1734 ~ ~ 1735 | L2 IP Header of | 1736 | error message | 1737 ~ ~ 1738 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1739 | L2 ICMP Header | 1740 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1741 ~ ~ P 1742 | IP and other encapsulation | a 1743 | headers of original L3 packet | c 1744 ~ ~ k 1745 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1746 ~ ~ t 1747 | IP header of | 1748 | original L3 packet | i 1749 ~ ~ n 1750 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1751 ~ ~ e 1752 | Upper layer headers and | r 1753 | leading portion of body | r 1754 | of the original L3 packet | o 1755 ~ ~ r 1756 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1758 Figure 5: AERO Interface Link-Layer Error Message Format 1760 The AERO node rules for processing these link-layer error messages 1761 are as follows: 1763 o When an AERO node receives a link-layer Parameter Problem message, 1764 it processes the message the same as described as for ordinary 1765 ICMP errors in the normative references [RFC0792][RFC4443]. 1767 o When an AERO node receives persistent link-layer Time Exceeded 1768 messages, the IP ID field may be wrapping before earlier fragments 1769 awaiting reassembly have been processed. In that case, the node 1770 should begin including integrity checks and/or institute rate 1771 limits for subsequent packets. 1773 o When an AERO node receives persistent link-layer Destination 1774 Unreachable messages in response to encapsulated packets that it 1775 sends to one of its asymmetric neighbor correspondents, the node 1776 should process the message as an indication that a path may be 1777 failing, and optionally initiate NUD over that path. If it 1778 receives Destination Unreachable messages over multiple paths, the 1779 node should allow future packets destined to the correspondent to 1780 flow through a default route and re-initiate route optimization. 1782 o When an AERO Client receives persistent link-layer Destination 1783 Unreachable messages in response to encapsulated packets that it 1784 sends to one of its symmetric neighbor Servers, the Client should 1785 mark the path as unusable and use another path. If it receives 1786 Destination Unreachable messages on many or all paths, the Client 1787 should associate with a new Server and release its association 1788 with the old Server as specified in Section 3.19.5. 1790 o When an AERO Server receives persistent link-layer Destination 1791 Unreachable messages in response to encapsulated packets that it 1792 sends to one of its symmetric neighbor Clients, the Server should 1793 mark the underlying path as unusable and use another underlying 1794 path. 1796 o When an AERO Server or Proxy receives link-layer Destination 1797 Unreachable messages in response to an encapsulated packet that it 1798 sends to one of its permanent neighbors, it treats the messages as 1799 an indication that the path to the neighbor may be failing. 1800 However, the dynamic routing protocol should soon reconverge and 1801 correct the temporary outage. 1803 When an AERO Bridge receives a packet for which the network-layer 1804 destination address is covered by an MSP, if there is no more- 1805 specific routing information for the destination the Bridge drops the 1806 packet and returns a network-layer Destination Unreachable message 1807 subject to rate limiting. The Bridge writes the network-layer source 1808 address of the original packet as the destination address and uses 1809 one of its non link-local addresses as the source address of the 1810 message. 1812 When an AERO node receives an encapsulated packet for which the 1813 reassembly buffer it too small, it drops the packet and returns a 1814 network-layer Packet Too Big (PTB) message. The node first writes 1815 the MRU value into the PTB message MTU field, writes the network- 1816 layer source address of the original packet as the destination 1817 address and writes one of its non link-local addresses as the source 1818 address. 1820 3.15. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1822 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1823 coordinated as discussed in the following Sections. 1825 3.15.1. AERO ND/PD Service Model 1827 Each AERO Server on the link configures a PD service to facilitate 1828 Client requests. Each Server is provisioned with a database of MNP- 1829 to-Client ID mappings for all Clients enrolled in the AERO service, 1830 as well as any information necessary to authenticate each Client. 1831 The Client database is maintained by a central administrative 1832 authority for the AERO link and securely distributed to all Servers, 1833 e.g., via the Lightweight Directory Access Protocol (LDAP) [RFC4511], 1834 via static configuration, etc. Clients receive the same service 1835 regardless of the Servers they select. 1837 AERO Clients and Servers use ND messages to maintain neighbor cache 1838 entries. AERO Servers configure their AERO interfaces as advertising 1839 NBMA interfaces, and therefore send unicast RA messages with a short 1840 Router Lifetime value (e.g., REACHABLE_TIME seconds) in response to a 1841 Client's RS message. Thereafter, Clients send additional RS messages 1842 to keep Server state alive. 1844 AERO Clients and Servers include PD parameters in RS/RA messages (see 1845 [I-D.templin-6man-dhcpv6-ndopt] for ND/PD alternatives). The unified 1846 ND/PD messages are exchanged between Client and Server according to 1847 the prefix management schedule required by the PD service. If the 1848 Client knows its MNP in advance, it can instead employ prefix 1849 registration by including its AERO address as the source address of 1850 an RS message and with an OMNI option with valid prefix registration 1851 information for the MNP. If the Server (and Proxy) accept the 1852 Client's MNP assertion, they inject the prefix into the routing 1853 system and establish the necessary neighbor cache state. 1855 The following sections specify the Client and Server behavior. 1857 3.15.2. AERO Client Behavior 1859 AERO Clients discover the addresses of Servers in a similar manner as 1860 described in [RFC5214]. Discovery methods include static 1861 configuration (e.g., from a flat-file map of Server addresses and 1862 locations), or through an automated means such as Domain Name System 1863 (DNS) name resolution [RFC1035]. Alternatively, the Client can 1864 discover Server addresses through a layer 2 data link login exchange, 1865 or through a unicast RA response to a multicast/anycast RS as 1866 described below. In the absence of other information, the Client can 1867 resolve the DNS Fully-Qualified Domain Name (FQDN) 1868 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 1869 text string and "[domainname]" is a DNS suffix for the AERO link 1870 (e.g., "example.com"). 1872 To associate with a Server, the Client acts as a requesting router to 1873 request MNPs. The Client prepares an RS message with PD parameters 1874 and includes a Nonce and Timestamp option if the Client needs to 1875 correlate RA replies. If the Client already knows the Server's AERO 1876 address, it includes the AERO address as the network-layer 1877 destination address; otherwise, it includes the link-scoped All- 1878 Routers multicast (ff02::2) or Subnet-Router anycast (fe80::) address 1879 as the network-layer destination. If the Client already knows its 1880 own AERO address, it uses the AERO address as the network-layer 1881 source address; otherwise, it uses the unspecified IPv6 address 1882 (::/128) as the network-layer source address. 1884 The Client next includes an OMNI option in the RS message to register 1885 its link-layer information with the Server. The Client sets the OMNI 1886 option prefix registration information according to the MNP, and 1887 includes an ifIndex-tuple with S set to '1' corresponding to the 1888 underlying interface over which the Client will send the RS message. 1889 The Client MAY include additional ifIndex-tuples specific to other 1890 underlying interfaces. The Client MAY also include an SLLAO 1891 corresponding to the OMNI option ifIndex-tuple with S set to '1'. 1893 The Client then sends the RS message (either directly via Direct 1894 interfaces, via a VPN for VPNed interfaces, via a Proxy for proxyed 1895 interfaces or via INET encapsulation for INET interfaces) and waits 1896 for an RA message reply (see Section 3.15.3). The Client retries up 1897 to MAX_RTR_SOLICITATIONS times until an RA is received. If the 1898 Client receives no RAs, or if it receives an RA with Router Lifetime 1899 set to 0, the Client SHOULD abandon this Server and try another 1900 Server. Otherwise, the Client processes the PD information found in 1901 the RA message. 1903 Next, the Client creates a symmetric neighbor cache entry with the 1904 Server's AERO address as the network-layer address and the Server's 1905 encapsulation and/or link-layer addresses as the link-layer address. 1906 The Client records the RA Router Lifetime field value in the neighbor 1907 cache entry as the time for which the Server has committed to 1908 maintaining the MNP in the routing system via this underlying 1909 interface, and caches the other RA configuration information 1910 including Cur Hop Limit, M and O flags, Reachable Time and Retrans 1911 Timer. The Client then autoconfigures AERO addresses for each of the 1912 delegated MNPs and assigns them to the AERO interface. The Client 1913 also caches any MSPs included in Route Information Options (RIOs) 1914 [RFC4191] as MSPs to associate with the AERO link, and assigns the 1915 MTU value in the MTU option to the underlying interface. 1917 The Client then registers additional underlying interfaces with the 1918 Server by sending RS messages via each additional interface. The RS 1919 messages include the same parameters as for the initial RS/RA 1920 exchange, but with destination address set to the Server's AERO 1921 address. 1923 Following autoconfiguration, the Client sub-delegates the MNPs to its 1924 attached EUNs and/or the Client's own internal virtual interfaces as 1925 described in [I-D.templin-v6ops-pdhost] to support the Client's 1926 downstream attached "Internet of Things (IoT)". The Client 1927 subsequently sends additional RS messages over each underlying 1928 interface before the Router Lifetime received for that interface 1929 expires. 1931 After the Client registers its underlying interfaces, it may wish to 1932 change one or more registrations, e.g., if an interface changes 1933 address or becomes unavailable, if QoS preferences change, etc. To 1934 do so, the Client prepares an RS message to send over any available 1935 underlying interface. The RS includes an OMNI option with prefix 1936 registration information specific to its MNP, with an ifIndex-tuple 1937 specific to the selected underlying interface with S set to '1', and 1938 with any additional ifIndex-tuples specific to other underlying 1939 interfaces. The Client includes fresh ifIndex-tuple values to update 1940 the Server's neighbor cache entry. When the Client receives the 1941 Server's RA response, it has assurance that the Server has been 1942 updated with the new information. 1944 If the Client wishes to discontinue use of a Server it issues an RS 1945 message over any underlying interface with an OMNI option with a 1946 prefix release indication. When the Server processes the message, it 1947 releases the MNP, sets the symmetric neighbor cache entry state for 1948 the Client to DEPARTED and returns an RA reply with Router Lifetime 1949 set to 0. After a short delay (e.g., 2 seconds), the Server 1950 withdraws the MNP from the routing system. 1952 3.15.3. AERO Server Behavior 1954 AERO Servers act as IP routers and support a PD service for Clients. 1955 Servers arrange to add their AERO addresses to a static map of Server 1956 addresses for the link and/or the DNS resource records for the FQDN 1957 "linkupnetworks.[domainname]" before entering service. Server 1958 addresses should be geographically and/or topologically referenced, 1959 and made available for discovery by Clients on the AERO link. 1961 When a Server receives a prospective Client's RS message on its AERO 1962 interface, it SHOULD return an immediate RA reply with Router 1963 Lifetime set to 0 if it is currently too busy or otherwise unable to 1964 service the Client. Otherwise, the Server authenticates the RS 1965 message and processes the PD parameters. The Server first determines 1966 the correct MNPs to delegate to the Client by searching the Client 1967 database. When the Server delegates the MNPs, it also creates a 1968 forwarding table entry for each MNP so that the MNPs are propagated 1969 into the routing system (see: Section 3.3). For IPv6, the Server 1970 creates an IPv6 forwarding table entry for each MNP. For IPv4, the 1971 Server creates an IPv6 forwarding table entry with the SPAN 1972 Compatibility Prefix (SCP) corresponding to the IPv4 address. 1974 The Server next creates a symmetric neighbor cache entry for the 1975 Client using the base AERO address as the network-layer address and 1976 with lifetime set to no more than the smallest PD lifetime. Next, 1977 the Server updates the neighbor cache entry by recording the 1978 information in each ifIndex-tuple in the RS OMNI option. The Server 1979 also records the actual SPAN/INET addresses in the neighbor cache 1980 entry. 1982 Next, the Server prepares an RA message using its AERO address as the 1983 network-layer source address and the network-layer source address of 1984 the RS message as the network-layer destination address. The Server 1985 sets the Router Lifetime to the time for which it will maintain both 1986 this underlying interface individually and the symmetric neighbor 1987 cache entry as a whole. The Server also sets Cur Hop Limit, M and O 1988 flags, Reachable Time and Retrans Timer to values appropriate for the 1989 AERO link. The Server includes the delegated MNPs, any other PD 1990 parameters and an OMNI option with no ifIndex-tuples. The Server 1991 then includes one or more RIOs that encode the MSPs for the AERO 1992 link, plus an MTU option (see Section 3.12). The Server finally 1993 forwards the message to the Client using SPAN/INET, INET, or NULL 1994 encapsulation as necessary. 1996 After the initial RS/RA exchange, the Server maintains a 1997 ReachableTime timer for each of the Client's underlying interfaces 1998 individually (and for the Client's symmetric neighbor cache entry 1999 collectively) set to expire after Router Lifetime seconds. If the 2000 Client (or Proxy) issues additional RS messages, the Server sends an 2001 RA response and resets ReachableTime. If the Server receives an ND 2002 message with PD release indication it sets the Client's symmetric 2003 neighbor cache entry to the DEPARTED state and withdraws the MNP from 2004 the routing system after a short delay (e.g., 2 seconds). If 2005 ReachableTime expires before a new RS is received on an individual 2006 underlying interface, the Server marks the interface as DOWN. If 2007 ReachableTime expires before any new RS is received on any individual 2008 underlying interface, the Server deletes the neighbor cache entry and 2009 withdraws the MNP without delay. 2011 The Server processes any ND/PD messages pertaining to the Client and 2012 returns an NA/RA reply in response to solicitations. The Server may 2013 also issue unsolicited RA messages, e.g., with PD reconfigure 2014 parameters to cause the Client to renegotiate its PDs, with Router 2015 Lifetime set to 0 if it can no longer service this Client, etc. 2016 Finally, If the symmetric neighbor cache entry is in the DEPARTED 2017 state, the Server deletes the entry after DepartTime expires. 2019 Note: Clients SHOULD notify former Servers of their departures, but 2020 Servers are responsible for expiring neighbor cache entries and 2021 withdrawing routes even if no departure notification is received 2022 (e.g., if the Client leaves the network unexpectedly). Servers 2023 SHOULD therefore set Router Lifetime to REACHABLE_TIME seconds in 2024 solicited RA messages to minimize persistent stale cache information 2025 in the absence of Client departure notifications. A short Router 2026 Lifetime also ensures that proactive Client/Server RS/RA messaging 2027 will keep any NAT state alive (see above). 2029 Note: All Servers on an AERO link MUST advertise consistent values in 2030 the RA Cur Hop Limit, M and O flags, Reachable Time and Retrans Timer 2031 fields the same as for any link, since unpredictable behavior could 2032 result if different Servers on the same link advertised different 2033 values. 2035 3.15.3.1. Lightweight DHCPv6 Relay Agent (LDRA) 2037 When DHCPv6 is used as the ND/PD service back end, AERO Clients and 2038 Servers are always on the same link (i.e., the AERO link) from the 2039 perspective of DHCPv6. However, in some implementations the DHCPv6 2040 server and ND function may be located in separate modules. In that 2041 case, the Server's AERO interface module can act as a Lightweight 2042 DHCPv6 Relay Agent (LDRA)[RFC6221] to relay PD messages to and from 2043 the DHCPv6 server module. 2045 When the LDRA receives an authentic RS message, it extracts the PD 2046 message parameters and uses them to construct an IPv6/UDP/DHCPv6 2047 message. It sets the IPv6 source address to the source address of 2048 the RS message, sets the IPv6 destination address to 2049 'All_DHCP_Relay_Agents_and_Servers' and sets the UDP fields to values 2050 that will be understood by the DHCPv6 server. 2052 The LDRA then wraps the message in a DHCPv6 'Relay-Forward' message 2053 header and includes an 'Interface-Id' option that includes enough 2054 information to allow the LDRA to forward the resulting Reply message 2055 back to the Client (e.g., the Client's link-layer addresses, a 2056 security association identifier, etc.). The LDRA also wraps the OMNI 2057 option and SLLAO into the Interface-Id option, then forwards the 2058 message to the DHCPv6 server. 2060 When the DHCPv6 server prepares a Reply message, it wraps the message 2061 in a 'Relay-Reply' message and echoes the Interface-Id option. The 2062 DHCPv6 server then delivers the Relay-Reply message to the LDRA, 2063 which discards the Relay-Reply wrapper and IPv6/UDP headers, then 2064 uses the DHCPv6 message to construct an RA response to the Client. 2065 The Server uses the information in the Interface-Id option to prepare 2066 the RA message and to cache the link-layer addresses taken from the 2067 OMNI option and SLLAO echoed in the Interface-Id option. 2069 3.16. The AERO Proxy 2071 Clients may connect to ANETs that deploy perimeter security services 2072 to facilitate communications to Servers in outside INETs. In that 2073 case, the ANET can employ an AERO Proxy. The Proxy is located at the 2074 ANET/INET border and listens for RS messages originating from or RA 2075 messages destined to ANET Clients. The Proxy acts on these control 2076 messages as follows: 2078 o when the Proxy receives an RS message from a new ANET Client, it 2079 first authenticates the message then examines the network-layer 2080 destination address. If the destination address is a Server's 2081 AERO address, the Proxy proceeds to the next step. Otherwise, if 2082 the destination is All-Routers multicast or Subnet-Router anycast, 2083 the Proxy selects a "nearby" Server that is likely to be a good 2084 candidate to serve the Client and replaces the destination address 2085 with the Server's AERO address. Next, the Proxy creates a proxy 2086 neighbor cache entry and caches the Client and Server link-layer 2087 addresses along with the OMNI option information and any other 2088 identifying information including Transaction IDs, Client 2089 Identifiers, Nonce values, etc. The Proxy finally encapsulates 2090 the (proxyed) RS message in a SPAN header with source set to the 2091 Proxy's SPAN address and destination set to the Server's SPAN 2092 address then forwards the message into the SPAN. 2094 o when the Server receives the RS, it authenticates the message then 2095 creates or updates a symmetric neighbor cache entry for the Client 2096 with the Proxy's SPAN address as the link-layer address. The 2097 Server then sends an RA message back to the Proxy via the SPAN. 2099 o when the Proxy receives the RA, it authenticates the message and 2100 matches it with the proxy neighbor cache entry created by the RS. 2101 The Proxy then caches the PD route information as a mapping from 2102 the Client's MNPs to the Client's ANET address, caches the 2103 Server's advertised Router Lifetime and sets the neighbor cache 2104 entry state to REACHABLE. The Proxy then sets the P bit in the RA 2105 flags field, optionally rewrites the Router Lifetime and forwards 2106 the (proxyed) message to the Client. The Proxy finally includes 2107 an MTU option (if necessary) with an MTU to use for the underlying 2108 ANET interface. 2110 After the initial RS/RA exchange, the Proxy forwards any Client data 2111 packets for which there is no matching asymmetric neighbor cache 2112 entry to a Bridge using SPAN encapsulation with its own SPAN address 2113 as the source and the SPAN address corresponding to the Client as the 2114 destination. The Proxy instead forwards any Client data destined to 2115 an asymmetric neighbor cache target directly to the target according 2116 to the SPAN/link-layer information - the process of establishing 2117 asymmetric neighbor cache entries is specified in Section 3.17. 2119 While the Client is still attached to the ANET, the Proxy sends NS, 2120 RS and/or unsolicited NA messages to update the Server's symmetric 2121 neighbor cache entries on behalf of the Client and/or to convey QoS 2122 updates. This allows for higher-frequency Proxy-initiated RS/RA 2123 messaging over well-connected INET infrastructure supplemented by 2124 lower-frequency Client-initiated RS/RA messaging over constrained 2125 ANET data links. 2127 If the Server ceases to send solicited advertisements, the Proxy 2128 sends unsolicited RAs on the ANET interface with destination set to 2129 All-Nodes multicast (ff02::1) and with Router Lifetime set to zero to 2130 inform Clients that the Server has failed. Although the Proxy 2131 engages in ND exchanges on behalf of the Client, the Client can also 2132 send ND messages on its own behalf, e.g., if it is in a better 2133 position than the Proxy to convey QoS changes, etc. For this reason, 2134 the Proxy marks any Client-originated solicitation messages (e.g. by 2135 inserting a Nonce option) so that it can return the solicited 2136 advertisement to the Client instead of processing it locally. 2138 If the Client becomes unreachable, the Proxy sets the neighbor cache 2139 entry state to DEPARTED and retains the entry for DEPART_TIME 2140 seconds. While the state is DEPARTED, the Proxy forwards any packets 2141 destined to the Client to a Bridge via SPAN encapsulation with the 2142 Client's current Server as the destination. The Bridge in turn 2143 forwards the packets to the Client's current Server. When DepartTime 2144 expires, the Proxy deletes the neighbor cache entry and discards any 2145 further packets destined to this (now forgotten) Client. 2147 In some ANETs that employ a Proxy, the Client's MNP can be injected 2148 into the ANET routing system. In that case, the Client can send data 2149 messages without encapsulation so that the ANET routing system 2150 transports the unencapsulated packets to the Proxy. This can be very 2151 beneficial, e.g., if the Client connects to the ANET via low-end data 2152 links such as some aviation wireless links. 2154 If the first-hop ANET access router is AERO-aware, the Client can 2155 avoid encapsulation for both its control and data messages. When the 2156 Client connects to the link, it can send an unencapsulated RS message 2157 with source address set to its AERO address and with destination 2158 address set to the AERO address of the Client's selected Server or to 2159 All-Routers multicast or Subnet-Router anycast. The Client includes 2160 an OMNI option formatted as specified in 2161 [I-D.templin-6man-omni-interface]. 2163 The Client then sends the unencapsulated RS message, which will be 2164 intercepted by the AERO-Aware access router. The access router then 2165 encapsulates the RS message in an ANET header with its own address as 2166 the source address and the address of a Proxy as the destination 2167 address. The access router further remembers the address of the 2168 Proxy so that it can encapsulate future data packets from the Client 2169 via the same Proxy. If the access router needs to change to a new 2170 Proxy, it simply sends another RS message toward the Server via the 2171 new Proxy on behalf of the Client. 2173 In some cases, the access router and Proxy may be one and the same 2174 node. In that case, the node would be located on the same physical 2175 link as the Client, but its message exchanges with the Server would 2176 need to pass through a security gateway at the ANET/INET border. The 2177 method for deploying access routers and Proxys (i.e. as a single node 2178 or multiple nodes) is an ANET-local administrative consideration. 2180 3.16.1. Detecting and Responding to Server Failures 2182 In environments where fast recovery from Server failure is required, 2183 Proxys SHOULD use proactive Neighbor Unreachability Detection (NUD) 2184 to track Server reachability in a similar fashion as for 2185 Bidirectional Forwarding Detection (BFD) [RFC5880]. Proxys can then 2186 quickly detect and react to failures so that cached information is 2187 re-established through alternate paths. The NUD control messaging is 2188 carried only over well-connected ground domain networks (i.e., and 2189 not low-end aeronautical radio links) and can therefore be tuned for 2190 rapid response. 2192 Proxys perform proactive NUD with Servers for which there are 2193 currently active ANET Clients by sending continuous NS messages in 2194 rapid succession, e.g., one message per second. The Proxy sends the 2195 NS message via the SPAN with the Proxy's AERO address as the source 2196 and the AERO address of the Server as the destination. When the 2197 Proxy is also sending RS messages to the Server on behalf of ANET 2198 Clients, the resulting RA responses can be considered as equivalent 2199 hints of forward progress. This means that the Proxy need not also 2200 send a periodic NS if it has already sent an RS within the same 2201 period. If the Server fails (i.e., if the Proxy ceases to receive 2202 advertisements), the Proxy can quickly inform Clients by sending 2203 multicast RA messages on the ANET interface. 2205 The Proxy sends RA messages on the ANET interface with source address 2206 set to the Server's address, destination address set to All-Nodes 2207 multicast, and Router Lifetime set to 0. The Proxy SHOULD send 2208 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2209 [RFC4861]. Any Clients on the ANET that had been using the failed 2210 Server will receive the RA messages and associate with a new Server. 2212 3.16.2. Point-to-Multipoint Server Coordination 2214 In environments where Client messaging over ANETs is bandwidth- 2215 limited and/or expensive, Clients can enlist the services of the 2216 Proxy to coordinate with multiple Servers in a single RS/RA message 2217 exchange. The Client can send a single RS message to All-Routers 2218 multicast that includes the ID's of multiple Servers in MS-Register 2219 sub-options of the OMNI option,. 2221 When the Proxy receives the RS and processes the OMNI option, it 2222 performs a separate RS/RA exchange with each MS-Register Server. 2223 When it has received the RA messages, it creates an "aggregate" RA 2224 message to return to the Client with an OMNI option with each 2225 responding Server's ID recorded in an MS-Register sub-option. 2227 Client's can thereafter employ efficient point-to-multipoint Server 2228 coordination under the assistance of the Proxy to dramatically reduce 2229 the number of messages sent over the ANET while enlisting the support 2230 of multiple Servers for fault tolerance. Clients can further include 2231 MS-Release suboptions in RS messages to request the Proxy to release 2232 from former Servers via the procedures discussed in Section 3.19.5. 2234 The OMNI interface specification [I-D.templin-6man-omni-interface] 2235 provides further discussion of the Client/Proxy RS/RA messaging 2236 involved in point-to-multipoint coordination. 2238 3.17. AERO Route Optimization 2240 While data packets are flowing between a source and target node, 2241 route optimization SHOULD be used. Route optimization is initiated 2242 by the first eligible Route Optimization Source (ROS) closest to the 2243 source as follows: 2245 o For Clients on VPNed and Direct interfaces, the Server is the ROS. 2247 o For Clients on Proxyed interfaces, the Proxy is the ROS. 2249 o For Clients on INET interfaces, the Client itself is the ROS. 2251 o For correspondent nodes on INET/EUN interfaces serviced by a 2252 Relay, the Relay is the ROS. 2254 The route optimization procedure is conducted between the ROS and the 2255 target Server/Relay acting as a Route Optimization Responder (ROR) in 2256 the same manner as for IPv6 ND Address Resolution and using the same 2257 NS/NA messaging. The target may either be a MNP Client serviced by a 2258 Server, or a non-MNP correspondent reachable via a Relay. 2260 The procedures are specified in the following sections. 2262 3.17.1. Route Optimization Initiation 2264 While data packets are flowing from the source node toward a target 2265 node, the ROS performs address resolution by sending an NS message 2266 for Address Resolution (NS(AR)) to receive a solicited NA message 2267 from the ROR. When the ROS sends an NS(AR), it includes: 2269 o the AERO address of the ROS as the source address. 2271 o the data packet's destination as the Target Address. 2273 o the Solicited-Node multicast address [RFC4291] formed from the 2274 lower 24 bits of the data packet's destination as the destination 2275 address, e.g., for 2001:db8:1:2::10:2000 the NS destination 2276 address is ff02:0:0:0:0:1:ff10:2000. 2278 The NS(AR) message includes an OMNI option with no ifIndex-tuples and 2279 no SLLAO, such that the target will not create a neighbor cache 2280 entry. 2282 The ROS then encapsulates the NS(AR) message in a SPAN header with 2283 source set to its own SPAN address and destination set to the SPAN 2284 address corresponding to the packet's final destination, then sends 2285 the message into the SPAN without decrementing the network-layer TTL/ 2286 Hop Limit field. 2288 3.17.2. Relaying the NS 2290 When the Bridge receives the NS(AR) message from the ROS, it discards 2291 the INET header and determines that the ROR is the next hop by 2292 consulting its standard IPv6 forwarding table for the SPAN header 2293 destination address. The Bridge then forwards the message toward the 2294 ROR via the SPAN the same as for any IPv6 router. The final-hop 2295 Bridge in the SPAN will deliver the message via a secured tunnel to 2296 the ROR. 2298 3.17.3. Processing the NS and Sending the NA 2300 When the ROR receives the NS(AR) message, it examines the Target 2301 Address to determine whether it has a neighbor cache entry and/or 2302 route that matches the target. If there is no match, the ROR drops 2303 the message. Otherwise, the ROR continues processing as follows: 2305 o if the target belongs to an MNP Client neighbor in the DEPARTED 2306 state the ROR changes the NS(AR) message SPAN destination address 2307 to the SPAN address of the Client's new Server, forwards the 2308 message into the SPAN and returns from processing. 2310 o If the target belongs to an MNP Client neighbor in the REACHABLE 2311 state, the ROR instead adds the AERO source address to the target 2312 Client's Report List with time set to ReportTime. 2314 o If the target belongs to a non-MNP route, the ROR continues 2315 processing without adding an entry to the Report List. 2317 The ROR then prepares a solicited NA message to send back to the ROS 2318 but does not create a neighbor cache entry. The ROR sets the NA 2319 source address to the AERO address corresponding to the target, sets 2320 the Target Address to the target of the solicitation, and sets the 2321 destination address to the source of the solicitation. 2323 The ROR then includes an OMNI option with prefix registration length 2324 set to the length of the MNP if the target is an MNP Client; 2325 otherwise, set to the maximum of the non-MNP prefix length and 64. 2326 (Note that a /64 limit is imposed to avoid causing the ROS to set 2327 short prefixes (e.g., "default") that would match destinations for 2328 which the routing system includes more-specific prefixes.) 2330 If the target is an MNP Client, the ROR next includes ifIndex-tuples 2331 in the OMNI option for each of the target Client's underlying 2332 interfaces with current information for each interface and with the S 2333 flag set to 0. The ROR then includes a TLLAO with ifIndex-tuples in 2334 one-to-one correspondence with the tuples that appear in the OMNI 2335 option. 2337 The ROR sets the Link Layer Address and Port Number (if necessary) to 2338 its own INET address for VPNed and Direct interfaces or to the INET 2339 address of the Proxy for Proxyed interface, then includes its own 2340 SPAN address or the SPAN address of the Proxy as the ultimate Segment 2341 Routing List entry. For INET interfaces, the ROR instead sets the 2342 Link Layer Address and Port Number (if necessary) to the Client's 2343 INET address (while using Teredo format for IPv4 - see Section 3.25) 2344 then sets its own SPAN address in the penultimate Segment Routing 2345 List entry and sets the target's SPAN address in the ultimate Segment 2346 Routing List entry. 2348 The ROR then sets the NA message R flag to 1 (as a router), S flag to 2349 1 (as a response to a solicitation), and O flag to 0 (as a proxy). 2350 The ROR finally encapsulates the NA message in a SPAN header with 2351 source set to its own SPAN address and destination set to the source 2352 SPAN address of the NS(AR) message, then forwards the message into 2353 the SPAN without decrementing the network-layer TTL/Hop Limit field. 2355 3.17.4. Relaying the NA 2357 When the Bridge receives the NA message from the ROR, it discards the 2358 INET header and determines that the ROS is the next hop by consulting 2359 its standard IPv6 forwarding table for the SPAN header destination 2360 address. The Bridge then forwards the SPAN-encapsulated NA message 2361 toward the ROS the same as for any IPv6 router. The final-hop Bridge 2362 in the SPAN will deliver the message via a secured tunnel to the ROS. 2364 3.17.5. Processing the NA 2366 When the ROS receives the solicited NA message, it processes the 2367 message the same as for standard IPv6 Address Resolution [RFC4861]. 2368 In the process, it caches the source SPAN address then creates an 2369 asymmetric neighbor cache entry for the ROR and caches all 2370 information found in the OMNI and TLLAO options. The ROS finally 2371 sets the asymmetric neighbor cache entry lifetime to REACHABLE_TIME 2372 seconds. 2374 3.17.6. Route Optimization Maintenance 2376 Following route optimization, the ROS forwards future data packets 2377 destined to the target via the addresses found in the cached link- 2378 layer information. The route optimization is shared by all sources 2379 that send packets to the target via the ROS, i.e., and not just the 2380 source on behalf of which the route optimization was initiated. 2382 While new data packets destined to the target are flowing through the 2383 ROS, it sends additional NS(AR) messages to the ROR before 2384 ReachableTime expires to receive a fresh solicited NA message the 2385 same as described in the previous sections (route optimization 2386 refreshment strategies are an implementation matter, with a non- 2387 normative example given in Appendix B.1). The ROS uses the cached 2388 SPAN address of the ROR as the NS(AR) SPAN destination address, and 2389 sends up to MAX_MULTICAST_SOLICIT NS(AR) messages separated by 1 2390 second until an NA is received. If no NA is received, the ROS 2391 assumes that the current ROR has become unreachable and deletes the 2392 neighbor cache entry. Subsequent data packets will trigger a new 2393 route optimization per Section 3.17.1 to discover a new ROR while 2394 initial data packets travel over a suboptimal route. 2396 If an NA is received, the ROS then updates the asymmetric neighbor 2397 cache entry to refresh ReachableTime, while (for MNP destinations) 2398 the ROR adds or updates the ROS address to the target Client's Report 2399 List and with time set to ReportTime. While no data packets are 2400 flowing, the ROS instead allows ReachableTime for the asymmetric 2401 neighbor cache entry to expire. When ReachableTime expires, the ROS 2402 deletes the asymmetric neighbor cache entry. Any future data packets 2403 flowing through the ROS will again trigger a new route optimization. 2405 The ROS may also receive unsolicited NA messages from the ROR at any 2406 time (see: Section 3.19). If there is an asymmetric neighbor cache 2407 entry for the target, the ROS updates the link-layer information but 2408 does not update ReachableTime since the receipt of an unsolicited NA 2409 does not confirm that any forward paths are working. If there is no 2410 asymmetric neighbor cache entry, the ROS simply discards the 2411 unsolicited NA. 2413 In this arrangement, the ROS holds an asymmetric neighbor cache entry 2414 for the ROR, but the ROR does not hold an asymmetric neighbor cache 2415 entry for the ROS. The route optimization neighbor relationship is 2416 therefore asymmetric and unidirectional. If the target node also has 2417 packets to send back to the source node, then a separate route 2418 optimization procedure is performed in the reverse direction. But, 2419 there is no requirement that the forward and reverse paths be 2420 symmetric. 2422 3.18. Neighbor Unreachability Detection (NUD) 2424 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2425 [RFC4861] either reactively in response to persistent link-layer 2426 errors (see Section 3.14) or proactively to confirm reachability. 2427 The NUD algorithm is based on periodic control message exchanges. 2428 The algorithm may further be seeded by ND hints of forward progress, 2429 but care must be taken to avoid inferring reachability based on 2430 spoofed information. For example, authentic IPv6 ND message 2431 exchanges may be considered as acceptable hints of forward progress, 2432 while spurious data packets should not be. 2434 AERO Servers, Proxys and Relays can use standard NS/NA NUD exchanges 2435 sent over the SPAN to securely test reachability without risk of DoS 2436 attacks from nodes pretending to be a neighbor; Proxys can further 2437 perform NUD to securely verify Server reachability on behalf of their 2438 proxyed Clients. However, a means for a ROS to test the unsecured 2439 forward directions of target route optimized paths is also necessary. 2441 When an ROR directs an ROS to a neighbor with one or more target 2442 link-layer addresses, the ROS can proactively test each such 2443 unsecured route optimized path by sending "loopback" NS(NUD) 2444 messages. While testing the paths, the ROS can optionally continue 2445 to send packets via the SPAN, maintain a small queue of packets until 2446 target reachability is confirmed, or (optimistically) allow packets 2447 to flow via the route optimized paths. 2449 When the ROS sends a loopback NS(NUD) message, it uses its AERO 2450 address as both the IPv6 source and destination address, and the MNP 2451 Subnet-Router anycast address as the Target Address. The ROS 2452 includes a Nonce and Timestamp option, then encapsulates the message 2453 in SPAN/INET headers with its own SPAN address as the source and the 2454 SPAN address of the route optimization target as the destination. 2455 The ROS then forwards the message to the target (either directly to 2456 the link layer address of the target if the target is in the same 2457 SPAN segment, or via a Bridge if the target is in a different SPAN 2458 segment). 2460 When the route optimization target receives the NS(NUD) message, it 2461 notices that the IPv6 destination address is the same as the source 2462 address. It then reverses the SPAN source and destination addresses 2463 and returns the message to the ROS (either directly or via the SPAN). 2464 The route optimization target does not decrement the NS(NUD) message 2465 IPv6 Hop-Limit in the process, since the message has not exited the 2466 SPAN. 2468 When the ROS receives the NS(NUD) message, it can determine from the 2469 Nonce, Timestamp and Target Address that the message originated from 2470 itself and that it transited the forward path. The ROS need not 2471 prepare an NA response, since the destination of the response would 2472 be itself and testing the route optimization path again would be 2473 redundant. 2475 The ROS marks route optimization target paths that pass these NUD 2476 tests as "reachable", and those that do not as "unreachable". These 2477 markings inform the AERO interface forwarding algorithm specified in 2478 Section 3.13. 2480 Note that to avoid a DoS vector nodes MUST NOT return loopback 2481 NS(NUD) messages received from an unsecured link-layer source via a 2482 secured SPAN path. 2484 3.19. Mobility Management and Quality of Service (QoS) 2486 AERO is a Distributed Mobility Management (DMM) service. Each Server 2487 is responsible for only a subset of the Clients on the AERO link, as 2488 opposed to a Centralized Mobility Management (CMM) service where 2489 there is a single network mobility collective entity for all Clients. 2490 Clients coordinate with their associated Servers via RS/RA exchanges 2491 to maintain the DMM profile, and the AERO routing system tracks all 2492 current Client/Server peering relationships. 2494 Servers provide default routing and mobility/multilink services for 2495 their dependent Clients. Clients are responsible for maintaining 2496 neighbor relationships with their Servers through periodic RS/RA 2497 exchanges, which also serves to confirm neighbor reachability. When 2498 a Client's underlying interface address and/or QoS information 2499 changes, the Client is responsible for updating the Server with this 2500 new information. Note that for Proxyed interfaces, however, the 2501 Proxy can also perform some RS/RA exchanges on the Client's behalf. 2503 Mobility management considerations are specified in the following 2504 sections. 2506 3.19.1. Mobility Update Messaging 2508 Servers accommodate Client mobility/multilink and/or QoS change 2509 events by sending unsolicited NA (uNA) messages to each ROS in the 2510 target Client's Report List. When a Server sends a uNA message, it 2511 sets the IPv6 source address to the Client's AERO address, sets the 2512 destination address to All-Nodes multicast and sets the Target 2513 Address to the Client's Subnet-Router anycast address. The Server 2514 also includes an OMNI option with prefix registration information and 2515 with ifIndex-tuples for the target Client's remaining interfaces with 2516 S set to 0. The Server then includes a TLLAO with corresponding 2517 ifIndex-tuples prepared the same as for the initial route 2518 optimization event. The Server sets the NA R flag to 1, the S flag 2519 to 0 and the O flag to 0, then encapsulates the message in a SPAN 2520 header with source set to its own SPAN address and destination set to 2521 the SPAN address of the ROS and sends the message into the SPAN. 2523 As discussed in Section 7.2.6 of [RFC4861], the transmission and 2524 reception of uNA messages is unreliable but provides a useful 2525 optimization. In well-connected Internetworks with robust data links 2526 uNA messages will be delivered with high probability, but in any case 2527 the Server can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT uNAs 2528 to each ROS to increase the likelihood that at least one will be 2529 received. 2531 When the ROS receives a uNA message, it ignores the message if there 2532 is no existing neighbor cache entry for the Client. Otherwise, it 2533 uses the included OMNI option and TLLAO information to update the 2534 neighbor cache entry, but does not reset ReachableTime since the 2535 receipt of an unsolicited NA message from the target Server does not 2536 provide confirmation that any forward paths to the target Client are 2537 working. 2539 If uNA messages are lost, the ROS may be left with stale address and/ 2540 or QoS information for the Client for up to REACHABLE_TIME seconds. 2541 During this time, the ROS can continue sending packets according to 2542 its stale neighbor cache information. When ReachableTime is close to 2543 expiring, the ROS will re-initiate route optimization and receive 2544 fresh link-layer address information. 2546 In addition to sending uNA messages to the current set of ROSs for 2547 the Client, the Server also sends uNAs to the former link-layer 2548 address for any ifIndex-tuple for which the link-layer address has 2549 changed. The uNA messages update Proxys that cannot easily detect 2550 (e.g., without active probing) when a formerly-active Client has 2551 departed. 2553 3.19.2. Announcing Link-Layer Address and/or QoS Preference Changes 2555 When a Client needs to change its underlying interface addresses and/ 2556 or QoS preferences (e.g., due to a mobility event), either the Client 2557 or its Proxys send RS messages to the Server via the SPAN with an 2558 OMNI option that includes an ifIndex-tuple with S set to 1 and with 2559 the new link quality and address information. 2561 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 2562 sending actual data packets in case one or more RAs are lost. If all 2563 RAs are lost, the Client SHOULD re-associate with a new Server. 2565 When the Server receives the Client's changes, it sends uNA messages 2566 to all nodes in the Report List the same as described in the previous 2567 section. 2569 3.19.3. Bringing New Links Into Service 2571 When a Client needs to bring new underlying interfaces into service 2572 (e.g., when it activates a new data link), it sends an RS message to 2573 the Server via the underlying interface with an OMNI option that 2574 includes an ifIndex-tuple with S set to 1 and appropriate link 2575 quality values and with link-layer address information for the new 2576 link. 2578 3.19.4. Removing Existing Links from Service 2580 When a Client needs to remove existing underlying interfaces from 2581 service (e.g., when it de-activates an existing data link), it sends 2582 an RS or uNA message to its Server with an OMNI option with 2583 appropriate link quality values. 2585 If the Client needs to send RS/uNA messages over an underlying 2586 interface other than the one being removed from service, it MUST 2587 include ifIndex-tuples with appropriate link quality values for any 2588 underlying interfaces being removed from service. 2590 3.19.5. Moving to a New Server 2592 When a Client associates with a new Server, it performs the Client 2593 procedures specified in Section 3.15.2. The Client also includes MS- 2594 Release identifiers in the RS message OMNI option per 2595 [I-D.templin-6man-omni-interface] if it wants the new Server to 2596 notify any old Servers from which the Client is departing. 2598 When the new Server receives the Client's RS message, it returns an 2599 RA as specified in Section 3.15.3 and sends up to 2600 MAX_NEIGHBOR_ADVERTIISEMENT uNA messages to any old Servers listed in 2601 OMNI option MS-Release identifiers. Each uNA message includes the 2602 Client's AERO address as the source address, the old Server's AERO 2603 address as the destination address, and an OMNI option with the 2604 Register/Release bit set to 0. The new Server wraps the uNA in a 2605 SPAN header with its own SPAN address as the source and the old 2606 Server's SPAN address as the destination, then sends the message into 2607 the SPAN. 2609 When an old Server receives the uNA, it changes the Client's neighbor 2610 cache entry state to DEPARTED, sets the link-layer address of the 2611 Client to the new Server's SPAN address, and sets DepartTime to 2612 DEPART_TIME seconds. After a short delay (e.g., 2 seconds) the old 2613 Server withdraws the Client's MNP from the routing system. After 2614 DepartTime expires, the old Server deletes the Client's neighbor 2615 cache entry. 2617 The old Server also sends unsolicited NA messages to all ROSs in the 2618 Client's Report List with an OMNI option with a single ifIndex-tuple 2619 with ifIndex set to 0 and S set to '1', and with the SPAN address of 2620 the new Server in a companion TLLAO. When the ROS receives the NA, 2621 it caches the address of the new Server in the existing asymmetric 2622 neighbor cache entry and marks the entry as STALE. Subsequent data 2623 packets will then flow according to any existing cached link-layer 2624 information and trigger a new NS(AR)/NA exchange via the new Server. 2626 Clients SHOULD NOT move rapidly between Servers in order to avoid 2627 causing excessive oscillations in the AERO routing system. Examples 2628 of when a Client might wish to change to a different Server include a 2629 Server that has gone unreachable, topological movements of 2630 significant distance, movement to a new geographic region, movement 2631 to a new SPAN segment, etc. 2633 When a Client moves to a new Server, some of the fragments of a 2634 multiple fragment packet may have already arrived at the old Server 2635 while others are en route to the new Server, however no special 2636 attention in the reassembly algorithm is necessary when re-routed 2637 fragments are simply treated as loss. 2639 3.20. Multicast 2641 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 2642 [RFC3810] proxy service for its EUNs and/or hosted applications 2643 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 2644 underlying interfaces for which group membership is required. The 2645 IGMP/MLD messages may be further forwarded by a first-hop ANET access 2646 router acting as an IGMP/MLD-snooping switch [RFC4541], then 2647 ultimately delivered to an AERO Proxy/Server acting as a Protocol 2648 Independent Multicast - Sparse-Mode (PIM-SM, or simply "PIM") 2649 Designated Router (DR) [RFC7761]. AERO Relays also act as PIM 2650 routers (i.e., the same as AERO Proxys/Servers) on behalf of nodes on 2651 INET/EUN networks. The behaviors identified in the following 2652 sections correspond to Source-Specific Multicast (SSM) and Any-Source 2653 Multicast (ASM) operational modes. 2655 3.20.1. Source-Specific Multicast (SSM) 2657 When an ROS (i.e., an AERO Proxy/Server/Relay) "X" acting as PIM 2658 router receives a Join/Prune message from a node on its downstream 2659 interfaces containing one or more ((S)ource, (G)roup) pairs, it 2660 updates its Multicast Routing Information Base (MRIB) accordingly. 2661 For each S belonging to a prefix reachable via X's non-AERO 2662 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 2663 on those interfaces per [RFC7761]. 2665 For each S belonging to a prefix reachable via X's AERO interface, X 2666 originates a separate copy of the Join/Prune for each (S,G) in the 2667 message using its own AERO address as the source address and ALL-PIM- 2668 ROUTERS as the destination address. X then encapsulates each message 2669 in a SPAN header with source address set to the SPAN address of X and 2670 destination address set to S then forwards the message into the SPAN. 2671 The SPAN in turn forwards the message to AERO Server/Relay "Y" that 2672 services S. At the same time, if the message was a Join, X sends a 2673 route-optimization NS message toward each S the same as discussed in 2674 Section 3.17. The resulting NAs will return the AERO address for the 2675 prefix that matches S as the network-layer source address and TLLAOs 2676 with the SPAN addresses corresponding to any ifIndex-tuples that are 2677 currently servicing S. 2679 When Y processes the Join/Prune message, if S located behind any 2680 INET, Direct, or VPNed interfaces Y acts as a PIM router and updates 2681 its MRIB to list X as the next hop in the reverse path. If S is 2682 located behind any Proxys "Z"*, Y also forwards the message to each 2683 Z* over the SPAN while continuing to use the AERO address of X as the 2684 source address. Each Z* then updates its MRIB accordingly and 2685 maintains the AERO address of X as the next hop in the reverse path. 2686 Since the Bridges in the SPAN do not examine network layer control 2687 messages, this means that the (reverse) multicast tree path is simply 2688 from each Z* (and/or Y) to X with no other multicast-aware routers in 2689 the path. If any Z* (and/or Y) is located on the same SPAN segment 2690 as X, the multicast data traffic sent to X directly using SPAN/INET 2691 encapsulation instead of via a Bridge. 2693 Following the initial Join/Prune and NS/NA messaging, X maintains an 2694 asymmetric neighbor cache entry for each S the same as if X was 2695 sending unicast data traffic to S. In particular, X performs 2696 additional NS/NA exchanges to keep the neighbor cache entry alive for 2697 up to t_periodic seconds [RFC7761]. If no new Joins are received 2698 within t_periodic seconds, X allows the neighbor cache entry to 2699 expire. Finally, if X receives any additional Join/Prune messages 2700 for (S,G) it forwards the messages to each Y and Z* in the neighbor 2701 cache entry over the SPAN. 2703 At some later time, Client C that holds an MNP for source S may 2704 depart from a first Proxy Z1 and/or connect via a new Proxy Z2. In 2705 that case, Y sends an unsolicited NA message to X the same as 2706 specified for unicast mobility in Section 3.19. When X receives the 2707 unsolicited NA message, it updates its asymmetric neighbor cache 2708 entry for the AERO address for source S and sends new Join messages 2709 to any new Proxys Z2. There is no requirement to send any Prune 2710 messages to old Proxys Z1 since source S will no longer source any 2711 multicast data traffic via Z1. Instead, the multicast state for 2712 (S,G) in Proxy Z1 will soon time out since no new Joins will arrive. 2714 After some later time, C may move to a new Server Y2 and depart from 2715 old Sever Y1. In that case, Y1 sends Join messages for any of C's 2716 active (S,G) groups to Y2 while including its own AERO address as the 2717 source address. This causes Y2 to include Y1 in the multicast 2718 forwarding tree during the interim time that Y1's symmetric neighbor 2719 cache entry for C is in the DEPARTED state. At the same time, Y1 2720 sends an unsolicited NA message to X with an OMNI option and TLLAO 2721 with ifIndex-tuple set to 0 and a release indication to cause X to 2722 release its asymmetric neighbor cache entry. X then sends a new Join 2723 message to S via the SPAN and re-initiates route optimization the 2724 same as if it were receiving a fresh Join message from a node on a 2725 downstream link. 2727 3.20.2. Any-Source Multicast (ASM) 2729 When an ROS X acting as a PIM router receives a Join/Prune from a 2730 node on its downstream interfaces containing one or more (*,G) pairs, 2731 it updates its Multicast Routing Information Base (MRIB) accordingly. 2732 X then forwards a copy of the message to the Rendezvous Point (RP) R 2733 for each G over the SPAN. X uses its own AERO address as the source 2734 address and ALL-PIM-ROUTERS as the destination address, then 2735 encapsulates each message in a SPAN header with source address set to 2736 the SPAN address of X and destination address set to R, then sends 2737 the message into the SPAN. At the same time, if the message was a 2738 Join X initiates NS/NA route optimization the same as for the SSM 2739 case discussed in Section 3.20.1. 2741 For each source S that sends multicast traffic to group G via R, the 2742 Proxy/Server Z* for the Client that aggregates S encapsulates the 2743 packets in PIM Register messages and forwards them to R via the SPAN. 2744 R may then elect to send a PIM Join to Z* over the SPAN. This will 2745 result in an (S,G) tree rooted at Z* with R as the next hop so that R 2746 will begin to receive two copies of the packet; one native copy from 2747 the (S, G) tree and a second copy from the pre-existing (*, G) tree 2748 that still uses PIM Register encapsulation. R can then issue a PIM 2749 Register-stop message to suppress the Register-encapsulated stream. 2750 At some later time, if C moves to a new Proxy/Server Z*, it resumes 2751 sending packets via PIM Register encapsulation via the new Z*. 2753 At the same time, as multicast listeners discover individual S's for 2754 a given G, they can initiate an (S,G) Join for each S under the same 2755 procedures discussed in Section 3.20.1. Once the (S,G) tree is 2756 established, the listeners can send (S, G) Prune messages to R so 2757 that multicast packets for group G sourced by S will only be 2758 delivered via the (S, G) tree and not from the (*, G) tree rooted at 2759 R. All mobility considerations discussed for SSM apply. 2761 3.20.3. Bi-Directional PIM (BIDIR-PIM) 2763 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 2764 approach to ASM that treats the Rendezvous Point (RP) as a Designated 2765 Forwarder (DF). Further considerations for BIDIR-PIM are out of 2766 scope. 2768 3.21. Operation over Multiple AERO Links (VLANs) 2770 An AERO Client can connect to multiple AERO links the same as for any 2771 data link service. In that case, the Client maintains a distinct 2772 AERO interface for each link, e.g., 'aero0' for the first link, 2773 'aero1' for the second, 'aero2' for the third, etc. Each AERO link 2774 would include its own distinct set of Bridges, Servers and Proxys, 2775 thereby providing redundancy in case of failures. 2777 The Bridges, Servers and Proxys on each AERO link can assign AERO and 2778 SPAN addresses that use the same or different numberings from those 2779 on other links. Since the links are mutually independent there is no 2780 requirement for avoiding inter-link address duplication, e.g., the 2781 same AERO address such as fe80::1000 could be used to number distinct 2782 nodes that connect to different AERO links. 2784 Each AERO link could utilize the same or different ANET connections. 2785 The links can be distinguished at the link-layer via the SSP in a 2786 similar fashion as for Virtual Local Area Network (VLAN) tagging 2787 (e.g., IEEE 802.1Q) and/or through assignment of distinct sets of 2788 MSPs on each link. This gives rise to the opportunity for supporting 2789 multiple redundant networked paths, where each VLAN is distinguished 2790 by a different SRT color (see: Section 3.5.1). In particular, the 2791 Client can tag its RS messages with the appropriate label to cause 2792 the network to select the desired VLAN. 2794 The Client's IP layer can select the outgoing AERO interface 2795 appropriate for a given traffic profile while (in the reverse 2796 direction) correspondent nodes must have some way of steering their 2797 packets destined to a target via the correct AERO link. 2799 In a first alternative, if each AERO link services different MSPs, 2800 then the Client can receive a distinct MNP from each of the links. 2801 IP routing will therefore assure that the correct Red/Green/Blue/etc. 2802 network is used for both outbound and inbound traffic. This can be 2803 accomplished using existing technologies and approaches, and without 2804 requiring any special supporting code in correspondent nodes or 2805 Bridges. 2807 In a second alternative, if each AERO link services the same MSP(s) 2808 then each link could assign a distinct "AERO Link Anycast" address 2809 that is configured by all Bridges on the link. Correspondent nodes 2810 can then perform segment routing at the SPAN layer 2811 [RFC8402][RFC8754]. Segment Routing in the correct SRT will then 2812 direct the packet over multiple hops to the target. 2814 3.22. DNS Considerations 2816 AERO Client MNs and INET correspondent nodes consult the Domain Name 2817 System (DNS) the same as for any Internetworking node. When 2818 correspondent nodes and Client MNs use different IP protocol versions 2819 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 2820 A records for IPv4 address mappings to MNs which must then be 2821 populated in Relay NAT64 mapping caches. In that way, an IPv4 2822 correspondent node can send packets to the IPv4 address mapping of 2823 the target MN, and the Relay will translate the IPv4 header and 2824 destination address into an IPv6 header and IPv6 destination address 2825 of the MN. 2827 When an AERO Client registers with an AERO Server, the Server can 2828 return the address(es) of DNS servers in RDNSS options [RFC6106]. 2829 The DNS server provides the IP addresses of other MNs and 2830 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 2832 3.23. Transition Considerations 2834 The SPAN ensures that dissimilar INET partitions can be joined into a 2835 single unified AERO link, even though the partitions themselves may 2836 have differing protocol versions and/or incompatible addressing 2837 plans. However, a commonality can be achieved by incrementally 2838 distributing globally routable (i.e., native) IP prefixes to 2839 eventually reach all nodes (both mobile and fixed) in all SPAN 2840 segments. This can be accomplished by incrementally deploying AERO 2841 Relays on each INET partition, with each Relay distributing its MNPs 2842 and/or discovering non-MNP prefixes on its INET links. 2844 This gives rise to the opportunity to eventually distribute native IP 2845 addresses to all nodes, and to present a unified AERO link view 2846 (bridged by the SPAN) even if the INET partitions remain in their 2847 current protocol and addressing plans. In that way, the AERO link 2848 can serve the dual purpose of providing a mobility/multilink service 2849 and a transition service. Or, if an INET partition is transitioned 2850 to a native IP protocol version and addressing scheme that is 2851 compatible with the AERO link MNP-based addressing scheme, the 2852 partition and AERO link can be joined by Relays. 2854 Relays that connect INETs/EUNs with dissimilar IP protocol versions 2855 may need to employ a network address and protocol translation 2856 function such as NAT64[RFC6146]. 2858 3.24. Detecting and Reacting to Server and Bridge Failures 2860 In environments where rapid failure recovery is required, Servers and 2861 Bridges SHOULD use Bidirectional Forwarding Detection (BFD) 2862 [RFC5880]. Nodes that use BFD can quickly detect and react to 2863 failures so that cached information is re-established through 2864 alternate nodes. BFD control messaging is carried only over well- 2865 connected ground domain networks (i.e., and not low-end radio links) 2866 and can therefore be tuned for rapid response. 2868 Servers and Bridges maintain BFD sessions in parallel with their BGP 2869 peerings. If a Server or Bridge fails, BGP peers will quickly re- 2870 establish routes through alternate paths the same as for common BGP 2871 deployments. Similarly, Proxys maintain BFD sessions with their 2872 associated Bridges even though they do not establish BGP peerings 2873 with them. 2875 Proxys SHOULD use proactive NUD for Servers for which there are 2876 currently active ANET Clients in a manner that parallels BFD, i.e., 2877 by sending unicast NS messages in rapid succession to receive 2878 solicited NA messages. When the Proxy is also sending RS messages on 2879 behalf of ANET Clients, the RS/RA messaging can be considered as 2880 equivalent hints of forward progress. This means that the Proxy need 2881 not also send a periodic NS if it has already sent an RS within the 2882 same period. If a Server fails, the Proxy will cease to receive 2883 advertisements and can quickly inform Clients of the outage by 2884 sending multicast RA messages on the ANET interface. 2886 The Proxy sends multicast RA messages with source address set to the 2887 Server's address, destination address set to All-Nodes multicast, and 2888 Router Lifetime set to 0. The Proxy SHOULD send 2889 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2890 [RFC4861]. Any Clients on the ANET interface that have been using 2891 the (now defunct) Server will receive the RA messages and associate 2892 with a new Server. 2894 3.25. AERO Clients on the Open Internet 2896 AERO Clients that connect to the open Internet via INET interfaces 2897 can establish a VPN to securely connect to a Server in a "tethered" 2898 arrangement with all of the Client's traffic transiting the Server. 2899 In that case, the Client/Server relationship is the same as for 2900 Direct interfaces. Alternatively, the Client can associate with an 2901 INET Server using the Teredo protocol [RFC4380] as discussed in the 2902 following sections. 2904 When a Client's AERO interface enables INET underlying interfaces, it 2905 sends a Teredo UDP/IP-encapsulated RS message with IPv6 source 2906 address set to its AERO address, with IPv6 destination set to All- 2907 Routers multicast, with an OMNI option and with a Teredo 2908 Authentication option to provide message authentication. The Client 2909 also includes an SLLAO with Link Layer Address set according to the 2910 address of the underlying interface used for INET encapsulation. If 2911 the underlying interface address is IPv6, the Client sets the FMT 2912 according to whether the Port Number also must be included as 2913 discussed in Section 3.6. If the underlying address is IPv4, the 2914 Client sets the FMT to Teredo with NAT and includes a Teredo address 2915 [RFC4380] using the prefix fe80::/32 with the Server's IPv4 address, 2916 and with the IP address and Port Number used for INET encapsulation 2917 written in obfuscated form. The Client then sets the UDP/IP source 2918 to its INET address and UDP port, and sets the destination to the 2919 Server's INET address and the AERO service port number (8060), then 2920 sends the message to the Server. 2922 When the Server receives the RS message, it authenticates the message 2923 and registers the Client's MNP and INET interface information 2924 according to the OMNI option parameters. The Server then returns an 2925 RA message with IPv6 source and destination set corresponding to the 2926 addresses in the RS, and with a Teredo Authentication option. For 2927 IPv4 INET interfaces, the Server also includes a Teredo Origin option 2928 with the mapped and obfuscated Client observed IP address and port 2929 number for the Client, and with a Teredo Authentication option. The 2930 Server then sends the message to the Client and records the Client's 2931 IPv6 address (for IPv6 INET interfaces) or fe80:: Teredo address (for 2932 IPv4 INET interfaces) as the link layer address in the neighbor 2933 cache. For IPv4, if the Client's IPv4 address and port from the 2934 SLLAO match the UDP/IPv4 header information (and if the IPv4 address 2935 is global unicast) the Server notes the FMT for this Client as Teredo 2936 on the open INET. 2938 When the Client receives the RA message, for IPv6 INET interfaces the 2939 Client proceeds under the assumption that there are no NATs on the 2940 path to the Server. For IPv4 INET interfaces, the Client instead 2941 compares the mapped IP address and port from the Teredo Origin option 2942 with its own address. If the addresses are the same, the Client 2943 assumes there are no NATs on the path; if the addresses are 2944 different, the Client instead assumes that further Server 2945 qualification procedures are necessary to detect the type of NAT and 2946 proceeds according to standard Teredo procedures. 2948 After the Client has registered its INET interfaces in such RS/RA 2949 exchanges it sends periodic RS messages to receive fresh RA messages 2950 before the Router Lifetime received on each INET interface expires. 2951 The Client also maintains default routes via its Servers, i.e., the 2952 same as described in earlier sections. 2954 When the Client sends messages to target IP addresses, it also 2955 invokes route optimization per Section 3.17 using IPv6 ND address 2956 resolution messaging. The Client sends the NS(AR) message to the 2957 Server wrapped in a UDP/IP header with a Teredo Authentication option 2958 with the NS source address set to the Client's AERO address and 2959 destination address set to the target solicited node multicast 2960 address. The Server authenticates the message and sends a 2961 corresponding NS(AR) message over the SPAN the same as if it were the 2962 ROS, but with the SPAN source address set to the Server's SPAN 2963 address and destination set to the SPAN address of the target. When 2964 the ROR receives the NS(AR), it adds the Server's SPAN address and 2965 Client's AERO address to the target's Report List, and returns an NA 2966 with OMNI and TLLAO information for the target. The Server then 2967 returns a UDP/IP encapsulated NA message with a Teredo Authentication 2968 option to the Client. 2970 Following route optimization, for targets in the same SPAN segment if 2971 the target's Link Layer Address is native IPv6 or a Teredo address on 2972 the open INET, the Client forwards data packets directly to the 2973 target INET address. If the Link Layer Address is a Teredo address 2974 for a peer behind a NAT, the Client first establishes NAT state for 2975 the Link Layer Address using the "bubble" mechanisms specified in 2977 [RFC6081][RFC4380]. The Client continues to send data packets via 2978 its Server until NAT state is populated, then begins forwarding 2979 packets via the direct path through the NAT to the target. For 2980 targets in different SPAN segments, the Client inserts a Segment 2981 Routing header and forwards data packets to the Bridge that returned 2982 the NA message. 2984 The ROR may return uNAs via the Server if the target moves, and the 2985 Server will send corresponding Teredo Authentication-protected uNAs 2986 to the Client. The Client can also send "loopback" NS(NUD) messages 2987 to test forward path reachability even though there is no security 2988 association between the Client and the target. 2990 The Client sends Teredo UDP/IP encapsulated IPv6 packets no larger 2991 than 1280 bytes in one piece. In order to accommodate larger IPv6 2992 packets (up to the AERO interface 9180 MTU), the Client inserts a 2993 SPAN header with source set to its own SPAN address and destination 2994 set to the SPAN address of the target and uses IPv6 fragmentation 2995 according to Section 3.12. The Client then encapsulates each 2996 fragment in a UDP/IP header and sends the fragments to the next hop. 2998 Note that by adopting the Teredo encapsulation format and NAT 2999 traversal mechanisms, the combined AERO/Teredo service fulfills the 3000 [RFC4380], Section 6 long-term goal of implementing a (general- 3001 purpose) tunneling service. 3003 3.26. Time-Varying MNPs 3005 In some use cases, it is desirable, beneficial and efficient for the 3006 Client to receive a constant MNP that travels with the Client 3007 wherever it moves. For example, this would allow air traffic 3008 controllers to easily track aircraft, etc. In other cases, however 3009 (e.g., intelligent transportation systems), the MN may be willing to 3010 sacrifice a modicum of efficiency in order to have time-varying MNPs 3011 that can be changed every so often to defeat adversarial tracking. 3013 The DHCPv6-PD service offers a way for Clients that desire time- 3014 varying MNPs to obtain short-lived prefixes (e.g., on the order of a 3015 small number of minutes). In that case, the identity of the Client 3016 would not be bound to the MNP but rather the Client's identity would 3017 be bound to the DHCPv6 Device Unique Identifier (DUID) and used as 3018 the seed for Prefix Delegation. The Client would then be obligated 3019 to renumber its internal networks whenever its MNP (and therefore 3020 also its AERO address) changes. This should not present a challenge 3021 for Clients with automated network renumbering services, however 3022 presents limits for the durations of ongoing sessions that would 3023 prefer to use a constant address. 3025 4. Implementation Status 3027 An AERO implementation based on OpenVPN (https://openvpn.net/) was 3028 announced on the v6ops mailing list on January 10, 2018 and an 3029 initial public release of the AERO proof-of-concept source code was 3030 announced on the intarea mailing list on August 21, 2015. 3032 As of 4/1/2020, more recent updated implementations are under 3033 internal development and testing with plans to release in the near 3034 future. 3036 5. IANA Considerations 3038 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 3039 AERO in the "enterprise-numbers" registry. 3041 The IANA has assigned the UDP port number "8060" for an earlier 3042 experimental version of AERO [RFC6706]. This document obsoletes 3043 [RFC6706] and claims the UDP port number "8060" for all future use. 3045 No further IANA actions are required. 3047 6. Security Considerations 3049 AERO Bridges configure secured tunnels with AERO Servers and Proxys 3050 within their local SPAN segments. Applicable secured tunnel 3051 alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 3052 [RFC6347], WireGuard, etc. The AERO Bridges of all SPAN segments in 3053 turn configure secured tunnels for their neighboring AERO Bridges 3054 across the SPAN. Therefore, control messages that traverse the SPAN 3055 between any pair of AERO link neighbors are already secured. 3057 AERO Servers, Relays and Proxys targeted by a route optimization may 3058 also receive packets directly from the INET partitions instead of via 3059 the SPAN. For INET partitions that apply effective ingress filtering 3060 to defeat source address spoofing, the simple data origin 3061 authentication procedures in Section 3.11 can be applied. 3063 For INET partitions that cannot apply effective ingress filtering, 3064 the two options for securing communications include 1) disable route 3065 optimization so that all traffic is conveyed over secured tunnels via 3066 the SPAN, or 2) enable on-demand secure tunnel creation between INET 3067 partition neighbors. Option 1) would result in longer routes than 3068 necessary and traffic concentration on critical infrastructure 3069 elements. Option 2) could be coordinated by establishing a secured 3070 tunnel on-demand instead of performing an NS/NA exchange in the route 3071 optimization procedures. Procedures for establishing on-demand 3072 secured tunnels are out of scope. 3074 AERO Clients that connect to secured ANET enclaves need not apply 3075 security to their ND messages, since the messages will be intercepted 3076 by a perimeter Proxy that applies security on its outward-facing 3077 interface. AERO Clients connected to the open INET can use symmetric 3078 network and/or transport layer security services such as VPNs, but 3079 when there are many prospective neighbors with dynamically changing 3080 connectivity an asymmetric security service such as SEcure Neighbor 3081 Discovery (SEND) [RFC3971] and Cryptographically Generated Addresses 3082 (CGAs) [RFC3972] may be more appropriate. However, the Teredo 3083 Authentication option [RFC4380] already provides sufficient security 3084 for protecting IPv6 ND messages on the open INET without the need for 3085 also employing SEND/CGA. 3087 Application endpoints SHOULD use application-layer security services 3088 such as TLS/SSL, DTLS or SSH [RFC4251] to assure the same level of 3089 protection as for critical secured Internet services. AERO Clients 3090 that require host-based VPN services SHOULD use symmetric network 3091 and/or transport layer security services such as IPsec, TLS/SSL, 3092 DTLS, etc. AERO Proxys and Servers can also provide a network-based 3093 VPN service on behalf of the Client, e.g., if the Client is located 3094 within a secured enclave and cannot establish a VPN on its own 3095 behalf. 3097 AERO Servers and Bridges present targets for traffic amplification 3098 Denial of Service (DoS) attacks. This concern is no different than 3099 for widely-deployed VPN security gateways in the Internet, where 3100 attackers could send spoofed packets to the gateways at high data 3101 rates. This can be mitigated by connecting Servers and Bridges over 3102 dedicated links with no connections to the Internet and/or when 3103 connections to the Internet are only permitted through well-managed 3104 firewalls. Traffic amplification DoS attacks can also target an AERO 3105 Client's low data rate links. This is a concern not only for Clients 3106 located on the open Internet but also for Clients in secured 3107 enclaves. AERO Servers and Proxys can institute rate limits that 3108 protect Clients from receiving packet floods that could DoS low data 3109 rate links. 3111 AERO Relays must implement ingress filtering to avoid a spoofing 3112 attack in which spurious SPAN messages are injected into an AERO link 3113 from an outside attacker. AERO Clients MUST ensure that their 3114 connectivity is not used by unauthorized nodes on their EUNs to gain 3115 access to a protected network, i.e., AERO Clients that act as routers 3116 MUST NOT provide routing services for unauthorized nodes. (This 3117 concern is no different than for ordinary hosts that receive an IP 3118 address delegation but then "share" the address with other nodes via 3119 some form of Internet connection sharing such as tethering.) 3120 The MAP list MUST be well-managed and secured from unauthorized 3121 tampering, even though the list contains only public information. 3122 The MAP list can be conveyed to the Client in a similar fashion as in 3123 [RFC5214] (e.g., through layer 2 data link login messaging, secure 3124 upload of a static file, DNS lookups, etc.). 3126 Although public domain and commercial SEND implementations exist, 3127 concerns regarding the strength of the cryptographic hash algorithm 3128 have been documented [RFC6273] [RFC4982]. 3130 Segment routing provides authentication facilities that can be used 3131 to authenticate the information in the SRH [RFC8754]. 3133 Security considerations for accepting link-layer ICMP messages and 3134 reflected packets are discussed throughout the document. 3136 7. Acknowledgements 3138 Discussions in the IETF, aviation standards communities and private 3139 exchanges helped shape some of the concepts in this work. 3140 Individuals who contributed insights include Mikael Abrahamsson, Mark 3141 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 3142 Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian Farrel, Nick Green, 3143 Sri Gundavelli, Brian Haberman, Bernhard Haindl, Joel Halpern, Tom 3144 Herbert, Sascha Hlusiak, Lee Howard, Zdenek Jaron, Andre Kostur, 3145 Hubert Kuenig, Ted Lemon, Andy Malis, Satoru Matsushima, Tomek 3146 Mrugalski, Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal 3147 Skorepa, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, Lloyd 3148 Wood and James Woodyatt. Members of the IESG also provided valuable 3149 input during their review process that greatly improved the document. 3150 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 3151 for their shepherding guidance during the publication of the AERO 3152 first edition. 3154 This work has further been encouraged and supported by Boeing 3155 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 3156 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 3157 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 3158 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 3159 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 3160 Gene MacLean III, Rob Muszkiewicz, Sean O'Sullivan, Vijay 3161 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 3162 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 3163 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 3164 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 3165 Benson, Katie Tran and Eric Yeh are especially acknowledged for 3166 implementing the AERO functions as extensions to the public domain 3167 OpenVPN distribution. 3169 Earlier works on NBMA tunneling approaches are found in 3170 [RFC2529][RFC5214][RFC5569]. 3172 Many of the constructs presented in this second edition of AERO are 3173 based on the author's earlier works, including: 3175 o The Internet Routing Overlay Network (IRON) 3176 [RFC6179][I-D.templin-ironbis] 3178 o Virtual Enterprise Traversal (VET) 3179 [RFC5558][I-D.templin-intarea-vet] 3181 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 3182 [RFC5320][I-D.templin-intarea-seal] 3184 o AERO, First Edition [RFC6706] 3186 Note that these works cite numerous earlier efforts that are not also 3187 cited here due to space limitations. The authors of those earlier 3188 works are acknowledged for their insights. 3190 This work is aligned with the NASA Safe Autonomous Systems Operation 3191 (SASO) program under NASA contract number NNA16BD84C. 3193 This work is aligned with the FAA as per the SE2025 contract number 3194 DTFAWA-15-D-00030. 3196 This work is aligned with the Boeing Commercial Airplanes (BCA) 3197 Internet of Things (IoT) and autonomy programs. 3199 This work is aligned with the Boeing Information Technology (BIT) 3200 MobileNet program. 3202 8. References 3204 8.1. Normative References 3206 [I-D.templin-6man-omni-interface] 3207 Templin, F. and T. Whyman, "Transmission of IPv6 Packets 3208 over Overlay Multilink Network (OMNI) Interfaces", draft- 3209 templin-6man-omni-interface-17 (work in progress), April 3210 2020. 3212 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3213 DOI 10.17487/RFC0791, September 1981, 3214 . 3216 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 3217 RFC 792, DOI 10.17487/RFC0792, September 1981, 3218 . 3220 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3221 Requirement Levels", BCP 14, RFC 2119, 3222 DOI 10.17487/RFC2119, March 1997, 3223 . 3225 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3226 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3227 December 1998, . 3229 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 3230 "Definition of the Differentiated Services Field (DS 3231 Field) in the IPv4 and IPv6 Headers", RFC 2474, 3232 DOI 10.17487/RFC2474, December 1998, 3233 . 3235 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3236 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3237 DOI 10.17487/RFC3971, March 2005, 3238 . 3240 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3241 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3242 . 3244 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3245 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3246 November 2005, . 3248 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3249 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3250 . 3252 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 3253 Network Address Translations (NATs)", RFC 4380, 3254 DOI 10.17487/RFC4380, February 2006, 3255 . 3257 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3258 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3259 DOI 10.17487/RFC4861, September 2007, 3260 . 3262 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3263 Address Autoconfiguration", RFC 4862, 3264 DOI 10.17487/RFC4862, September 2007, 3265 . 3267 [RFC5175] Haberman, B., Ed. and R. Hinden, "IPv6 Router 3268 Advertisement Flags Option", RFC 5175, 3269 DOI 10.17487/RFC5175, March 2008, 3270 . 3272 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 3273 DOI 10.17487/RFC6081, January 2011, 3274 . 3276 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3277 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3278 May 2017, . 3280 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3281 (IPv6) Specification", STD 86, RFC 8200, 3282 DOI 10.17487/RFC8200, July 2017, 3283 . 3285 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3286 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3287 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3288 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3289 . 3291 8.2. Informative References 3293 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3294 2016. 3296 [I-D.ietf-6man-segment-routing-header] 3297 Filsfils, C., Dukes, D., Previdi, S., Leddy, J., 3298 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3299 (SRH)", draft-ietf-6man-segment-routing-header-26 (work in 3300 progress), October 2019. 3302 [I-D.ietf-dmm-distributed-mobility-anchoring] 3303 Chan, A., Wei, X., Lee, J., Jeon, S., and C. Bernardos, 3304 "Distributed Mobility Anchoring", draft-ietf-dmm- 3305 distributed-mobility-anchoring-15 (work in progress), 3306 March 2020. 3308 [I-D.ietf-intarea-gue] 3309 Herbert, T., Yong, L., and O. Zia, "Generic UDP 3310 Encapsulation", draft-ietf-intarea-gue-09 (work in 3311 progress), October 2019. 3313 [I-D.ietf-intarea-gue-extensions] 3314 Herbert, T., Yong, L., and F. Templin, "Extensions for 3315 Generic UDP Encapsulation", draft-ietf-intarea-gue- 3316 extensions-06 (work in progress), March 2019. 3318 [I-D.ietf-intarea-tunnels] 3319 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3320 Architecture", draft-ietf-intarea-tunnels-10 (work in 3321 progress), September 2019. 3323 [I-D.ietf-rtgwg-atn-bgp] 3324 Templin, F., Saccone, G., Dawra, G., Lindem, A., and V. 3325 Moreno, "A Simple BGP-based Mobile Routing System for the 3326 Aeronautical Telecommunications Network", draft-ietf- 3327 rtgwg-atn-bgp-05 (work in progress), January 2020. 3329 [I-D.templin-6man-dhcpv6-ndopt] 3330 Templin, F., "A Unified Stateful/Stateless Configuration 3331 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-09 3332 (work in progress), January 2020. 3334 [I-D.templin-intarea-grefrag] 3335 Templin, F., "GRE Tunnel Level Fragmentation", draft- 3336 templin-intarea-grefrag-04 (work in progress), July 2016. 3338 [I-D.templin-intarea-seal] 3339 Templin, F., "The Subnetwork Encapsulation and Adaptation 3340 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 3341 progress), January 2014. 3343 [I-D.templin-intarea-vet] 3344 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 3345 templin-intarea-vet-40 (work in progress), May 2013. 3347 [I-D.templin-ironbis] 3348 Templin, F., "The Interior Routing Overlay Network 3349 (IRON)", draft-templin-ironbis-16 (work in progress), 3350 March 2014. 3352 [I-D.templin-v6ops-pdhost] 3353 Templin, F., "IPv6 Prefix Delegation and Multi-Addressing 3354 Models", draft-templin-v6ops-pdhost-25 (work in progress), 3355 January 2020. 3357 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3359 [RFC1035] Mockapetris, P., "Domain names - implementation and 3360 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3361 November 1987, . 3363 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 3364 Communication Layers", STD 3, RFC 1122, 3365 DOI 10.17487/RFC1122, October 1989, 3366 . 3368 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 3369 DOI 10.17487/RFC1191, November 1990, 3370 . 3372 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3373 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3374 . 3376 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3377 DOI 10.17487/RFC2003, October 1996, 3378 . 3380 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3381 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3382 . 3384 [RFC2492] Armitage, G., Schulter, P., and M. Jork, "IPv6 over ATM 3385 Networks", RFC 2492, DOI 10.17487/RFC2492, January 1999, 3386 . 3388 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3389 Domains without Explicit Tunnels", RFC 2529, 3390 DOI 10.17487/RFC2529, March 1999, 3391 . 3393 [RFC2764] Gleeson, B., Lin, A., Heinanen, J., Armitage, G., and A. 3394 Malis, "A Framework for IP Based Virtual Private 3395 Networks", RFC 2764, DOI 10.17487/RFC2764, February 2000, 3396 . 3398 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 3399 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 3400 DOI 10.17487/RFC2784, March 2000, 3401 . 3403 [RFC2890] Dommety, G., "Key and Sequence Number Extensions to GRE", 3404 RFC 2890, DOI 10.17487/RFC2890, September 2000, 3405 . 3407 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", 3408 RFC 2923, DOI 10.17487/RFC2923, September 2000, 3409 . 3411 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3412 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3413 . 3415 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3416 of Explicit Congestion Notification (ECN) to IP", 3417 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3418 . 3420 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3421 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3422 DOI 10.17487/RFC3810, June 2004, 3423 . 3425 [RFC3819] Karn, P., Ed., Bormann, C., Fairhurst, G., Grossman, D., 3426 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 3427 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 3428 RFC 3819, DOI 10.17487/RFC3819, July 2004, 3429 . 3431 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 3432 for IPv6 Hosts and Routers", RFC 4213, 3433 DOI 10.17487/RFC4213, October 2005, 3434 . 3436 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3437 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3438 January 2006, . 3440 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3441 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3442 DOI 10.17487/RFC4271, January 2006, 3443 . 3445 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3446 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3447 2006, . 3449 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3450 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3451 December 2005, . 3453 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3454 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3455 2006, . 3457 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3458 Control Message Protocol (ICMPv6) for the Internet 3459 Protocol Version 6 (IPv6) Specification", STD 89, 3460 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3461 . 3463 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3464 Protocol (LDAP): The Protocol", RFC 4511, 3465 DOI 10.17487/RFC4511, June 2006, 3466 . 3468 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3469 "Considerations for Internet Group Management Protocol 3470 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3471 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3472 . 3474 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3475 "Internet Group Management Protocol (IGMP) / Multicast 3476 Listener Discovery (MLD)-Based Multicast Forwarding 3477 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3478 August 2006, . 3480 [RFC4607] Holbrook, H. and B. Cain, "Source-Specific Multicast for 3481 IP", RFC 4607, DOI 10.17487/RFC4607, August 2006, 3482 . 3484 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 3485 Errors at High Data Rates", RFC 4963, 3486 DOI 10.17487/RFC4963, July 2007, 3487 . 3489 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3490 Algorithms in Cryptographically Generated Addresses 3491 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3492 . 3494 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3495 "Bidirectional Protocol Independent Multicast (BIDIR- 3496 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3497 . 3499 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3500 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3501 DOI 10.17487/RFC5214, March 2008, 3502 . 3504 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3505 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3506 February 2010, . 3508 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3509 Route Optimization Requirements for Operational Use in 3510 Aeronautics and Space Exploration Mobile Networks", 3511 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3512 . 3514 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3515 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3516 . 3518 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3519 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3520 January 2010, . 3522 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 3523 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 3524 . 3526 [RFC5991] Thaler, D., Krishnan, S., and J. Hoagland, "Teredo 3527 Security Updates", RFC 5991, DOI 10.17487/RFC5991, 3528 September 2010, . 3530 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3531 "IPv6 Router Advertisement Options for DNS Configuration", 3532 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3533 . 3535 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3536 NAT64: Network Address and Protocol Translation from IPv6 3537 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3538 April 2011, . 3540 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3541 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3542 . 3544 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3545 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3546 DOI 10.17487/RFC6221, May 2011, 3547 . 3549 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3550 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3551 DOI 10.17487/RFC6273, June 2011, 3552 . 3554 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3555 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3556 January 2012, . 3558 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3559 for Equal Cost Multipath Routing and Link Aggregation in 3560 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3561 . 3563 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3564 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3565 . 3567 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID Field", 3568 RFC 6864, DOI 10.17487/RFC6864, February 2013, 3569 . 3571 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 3572 UDP Checksums for Tunneled Packets", RFC 6935, 3573 DOI 10.17487/RFC6935, April 2013, 3574 . 3576 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 3577 for the Use of IPv6 UDP Datagrams with Zero Checksums", 3578 RFC 6936, DOI 10.17487/RFC6936, April 2013, 3579 . 3581 [RFC7269] Chen, G., Cao, Z., Xie, C., and D. Binet, "NAT64 3582 Deployment Options and Experience", RFC 7269, 3583 DOI 10.17487/RFC7269, June 2014, 3584 . 3586 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3587 Korhonen, "Requirements for Distributed Mobility 3588 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3589 . 3591 [RFC7421] Carpenter, B., Ed., Chown, T., Gont, F., Jiang, S., 3592 Petrescu, A., and A. Yourtchenko, "Analysis of the 64-bit 3593 Boundary in IPv6 Addressing", RFC 7421, 3594 DOI 10.17487/RFC7421, January 2015, 3595 . 3597 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3598 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3599 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3600 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3601 2016, . 3603 [RFC8086] Yong, L., Ed., Crabbe, E., Xu, X., and T. Herbert, "GRE- 3604 in-UDP Encapsulation", RFC 8086, DOI 10.17487/RFC8086, 3605 March 2017, . 3607 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 3608 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 3609 DOI 10.17487/RFC8201, July 2017, 3610 . 3612 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3613 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3614 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3615 July 2018, . 3617 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3618 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3619 . 3621 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 3622 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3623 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 3624 . 3626 Appendix A. AERO Alternate Encapsulations 3628 When GUE encapsulation is not needed, AERO can use common 3629 encapsulations such as IP-in-IP [RFC2003][RFC2473][RFC4213], Generic 3630 Routing Encapsulation (GRE) [RFC2784][RFC2890] and others. The 3631 encapsulation is therefore only differentiated from non-AERO tunnels 3632 through the application of AERO control messaging and not through, 3633 e.g., a well-known UDP port number. 3635 As for GUE encapsulation, alternate AERO encapsulation formats may 3636 require encapsulation layer fragmentation. For simple IP-in-IP 3637 encapsulation, an IPv6 fragment header is inserted directly between 3638 the inner and outer IP headers when needed, i.e., even if the outer 3639 header is IPv4. The IPv6 Fragment Header is identified to the outer 3640 IP layer by its IP protocol number, and the Next Header field in the 3641 IPv6 Fragment Header identifies the inner IP header version. For GRE 3642 encapsulation, a GRE fragment header is inserted within the GRE 3643 header [I-D.templin-intarea-grefrag]. 3645 Figure 6 shows the AERO IP-in-IP encapsulation format before any 3646 fragmentation is applied: 3648 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3649 | Outer IPv4 Header | | Outer IPv6 Header | 3650 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3651 |IPv6 Frag Header (optional)| |IPv6 Frag Header (optional)| 3652 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3653 | Inner IP Header | | Inner IP Header | 3654 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3655 | | | | 3656 ~ ~ ~ ~ 3657 ~ Inner Packet Body ~ ~ Inner Packet Body ~ 3658 ~ ~ ~ ~ 3659 | | | | 3660 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3662 Minimal Encapsulation in IPv4 Minimal Encapsulation in IPv6 3664 Figure 6: Minimal Encapsulation Format using IP-in-IP 3666 Figure 7 shows the AERO GRE encapsulation format before any 3667 fragmentation is applied: 3669 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3670 | Outer IP Header | 3671 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3672 | GRE Header | 3673 | (with checksum, key, etc..) | 3674 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3675 | GRE Fragment Header (optional)| 3676 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3677 | Inner IP Header | 3678 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3679 | | 3680 ~ ~ 3681 ~ Inner Packet Body ~ 3682 ~ ~ 3683 | | 3684 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3686 Figure 7: Minimal Encapsulation Using GRE 3688 Alternate encapsulation may be preferred in environments where GUE 3689 encapsulation would add unnecessary overhead. For example, certain 3690 low-bandwidth wireless data links may benefit from a reduced 3691 encapsulation overhead. 3693 GUE encapsulation can traverse network paths that are inaccessible to 3694 non-UDP encapsulations, e.g., for crossing Network Address 3695 Translators (NATs). More and more, network middleboxes are also 3696 being configured to discard packets that include anything other than 3697 a well-known IP protocol such as UDP and TCP. It may therefore be 3698 necessary to determine the potential for middlebox filtering before 3699 enabling alternate encapsulation in a given environment. 3701 In addition to IP-in-IP, GRE and GUE, AERO can also use security 3702 encapsulations such as IPsec, TLS/SSL, DTLS, etc. In that case, AERO 3703 control messaging and route determination occur before security 3704 encapsulation is applied for outgoing packets and after security 3705 decapsulation is applied for incoming packets. 3707 AERO is especially well suited for use with VPN system encapsulations 3708 such as OpenVPN [OVPN]. 3710 Appendix B. Non-Normative Considerations 3712 AERO can be applied to a multitude of Internetworking scenarios, with 3713 each having its own adaptations. The following considerations are 3714 provided as non-normative guidance: 3716 B.1. Implementation Strategies for Route Optimization 3718 Route optimization as discussed in Section 3.17 results in the route 3719 optimization source (ROS) creating an asymmetric neighbor cache entry 3720 for the target neighbor. The neighbor cache entry is maintained for 3721 at most REACHABLE_TIME seconds and then deleted unless updated. In 3722 order to refresh the neighbor cache entry lifetime before the 3723 ReachableTime timer expires, the specification requires 3724 implementations to issue a new NS/NA exchange to reset ReachableTime 3725 to REACHABLE_TIME seconds while data packets are still flowing. 3726 However, the decision of when to initiate a new NS/NA exchange and to 3727 perpetuate the process is left as an implementation detail. 3729 One possible strategy may be to monitor the neighbor cache entry 3730 watching for data packets for (REACHABLE_TIME - 5) seconds. If any 3731 data packets have been sent to the neighbor within this timeframe, 3732 then send an NS to receive a new NA. If no data packets have been 3733 sent, wait for 5 additional seconds and send an immediate NS if any 3734 data packets are sent within this "expiration pending" 5 second 3735 window. If no additional data packets are sent within the 5 second 3736 window, delete the neighbor cache entry. 3738 The monitoring of the neighbor data packet traffic therefore becomes 3739 an asymmetric ongoing process during the neighbor cache entry 3740 lifetime. If the neighbor cache entry expires, future data packets 3741 will trigger a new NS/NA exchange while the packets themselves are 3742 delivered over a longer path until route optimization state is re- 3743 established. 3745 B.2. Implicit Mobility Management 3747 AERO interface neighbors MAY provide a configuration option that 3748 allows them to perform implicit mobility management in which no ND 3749 messaging is used. In that case, the Client only transmits packets 3750 over a single interface at a time, and the neighbor always observes 3751 packets arriving from the Client from the same link-layer source 3752 address. 3754 If the Client's underlying interface address changes (either due to a 3755 readdressing of the original interface or switching to a new 3756 interface) the neighbor immediately updates the neighbor cache entry 3757 for the Client and begins accepting and sending packets according to 3758 the Client's new address. This implicit mobility method applies to 3759 use cases such as cellphones with both WiFi and Cellular interfaces 3760 where only one of the interfaces is active at a given time, and the 3761 Client automatically switches over to the backup interface if the 3762 primary interface fails. 3764 B.3. Direct Underlying Interfaces 3766 When a Client's AERO interface is configured over a Direct interface, 3767 the neighbor at the other end of the Direct link can receive packets 3768 without any encapsulation. In that case, the Client sends packets 3769 over the Direct link according to QoS preferences. If the Direct 3770 interface has the highest QoS preference, then the Client's IP 3771 packets are transmitted directly to the peer without going through an 3772 ANET/INET. If other interfaces have higher QoS preferences, then the 3773 Client's IP packets are transmitted via a different interface, which 3774 may result in the inclusion of Proxys, Servers and Bridges in the 3775 communications path. Direct interfaces must be tested periodically 3776 for reachability, e.g., via NUD. 3778 B.4. Operation on AERO Links with /64 ASPs 3780 IPv6 AERO links typically have MSPs that aggregate many candidate 3781 MNPs of length /64 or shorter. However, in some cases it may be 3782 desirable to use AERO over links that have only a /64 MSP. This can 3783 be accommodated by treating all Clients on the AERO link as simple 3784 hosts that receive /128 prefix delegations. 3786 In that case, the Client sends an RS message to the Server the same 3787 as for ordinary AERO links. The Server responds with an RA message 3788 that includes one or more /128 prefixes (i.e., singleton addresses) 3789 that include the /64 MSP prefix along with an interface identifier 3790 portion to be assigned to the Client. The Client and Server then 3791 configure their AERO addresses based on the interface identifier 3792 portions of the /128s (i.e., the lower 64 bits) and not based on the 3793 /64 prefix (i.e., the upper 64 bits). 3795 For example, if the MSP for the host-only IPv6 AERO link is 3796 2001:db8:1000:2000::/64, each Client will receive one or more /128 3797 IPv6 prefix delegations such as 2001:db8:1000:2000::1/128, 3798 2001:db8:1000:2000::2/128, etc. When the Client receives the prefix 3799 delegations, it assigns the AERO addresses fe80::1, fe80::2, etc. to 3800 the AERO interface, and assigns the global IPv6 addresses (i.e., the 3801 /128s) to either the AERO interface or an internal virtual interface 3802 such as a loopback. In this arrangement, the Client conducts route 3803 optimization in the same sense as discussed in Section 3.17. 3805 This specification has applicability for nodes that act as a Client 3806 on an "upstream" AERO link, but also act as a Server on "downstream" 3807 AERO links. More specifically, if the node acts as a Client to 3808 receive a /64 prefix from the upstream AERO link it can then act as a 3809 Server to provision /128s to Clients on downstream AERO links. 3811 B.5. AERO Critical Infrastructure Considerations 3813 AERO Bridges can be either Commercial off-the Shelf (COTS) standard 3814 IP routers or virtual machines in the cloud. Bridges must be 3815 provisioned, supported and managed by the INET administrative 3816 authority, and connected to the Bridges of other INETs via inter- 3817 domain peerings. Cost for purchasing, configuring and managing 3818 Bridges is nominal even for very large AERO links. 3820 AERO Servers can be standard dedicated server platforms, but most 3821 often will be deployed as virtual machines in the cloud. The only 3822 requirements for Servers are that they can run the AERO user-level 3823 code and have at least one network interface connection to the INET. 3824 As with Bridges, Servers must be provisioned, supported and managed 3825 by the INET administrative authority. Cost for purchasing, 3826 configuring and managing Servers is nominal especially for virtual 3827 Servers hosted in the cloud. 3829 AERO Proxys are most often standard dedicated server platforms with 3830 one network interface connected to the ANET and a second interface 3831 connected to an INET. As with Servers, the only requirements are 3832 that they can run the AERO user-level code and have at least one 3833 interface connection to the INET. Proxys must be provisioned, 3834 supported and managed by the ANET administrative authority. Cost for 3835 purchasing, configuring and managing Proxys is nominal, and borne by 3836 the ANET administrative authority. 3838 AERO Relays can be any dedicated server or COTS router platform 3839 connected to INETs and/or EUNs. The Relay joins the SPAN and engages 3840 in eBGP peering with one or more Bridges as a stub AS. The Relay 3841 then injects its MNPs and/or non-MNP prefixes into the BGP routing 3842 system, and provisions the prefixes to its downstream-attached 3843 networks. The Relay can perform ROS/ROR services the same as for any 3844 Server, and can route between the MNP and non-MNP address spaces. 3846 B.6. AERO Server Failure Implications 3848 AERO Servers may appear as a single point of failure in the 3849 architecture, but such is not the case since all Servers on the link 3850 provide identical services and loss of a Server does not imply 3851 immediate and/or comprehensive communication failures. Although 3852 Clients typically associate with a single Server at a time, Server 3853 failure is quickly detected and conveyed by Bidirectional Forward 3854 Detection (BFD) and/or proactive NUD allowing Clients to migrate to 3855 new Servers. 3857 If a Server fails, ongoing packet forwarding to Clients will continue 3858 by virtue of the asymmetric neighbor cache entries that have already 3859 been established in route optimization sources (ROSs). If a Client 3860 also experiences mobility events at roughly the same time the Server 3861 fails, unsolicited NA messages may be lost but proxy neighbor cache 3862 entries in the DEPARTED state will ensure that packet forwarding to 3863 the Client's new locations will continue for up to DEPART_TIME 3864 seconds. 3866 If a Client is left without a Server for an extended timeframe (e.g., 3867 greater than REACHABLETIIME seconds) then existing asymmetric 3868 neighbor cache entries will eventually expire and both ongoing and 3869 new communications will fail. The original source will continue to 3870 retransmit until the Client has established a new Server 3871 relationship, after which time continuous communications will resume. 3873 Therefore, providing many Servers on the link with high availability 3874 profiles provides resilience against loss of individual Servers and 3875 assurance that Clients can establish new Server relationships quickly 3876 in event of a Server failure. 3878 B.7. AERO Client / Server Architecture 3880 The AERO architectural model is client / server in the control plane, 3881 with route optimization in the data plane. The same as for common 3882 Internet services, the AERO Client discovers the addresses of AERO 3883 Servers and selects one Server to connect to. The AERO service is 3884 analogous to common Internet services such as google.com, yahoo.com, 3885 cnn.com, etc. However, there is only one AERO service for the link 3886 and all Servers provide identical services. 3888 Common Internet services provide differing strategies for advertising 3889 server addresses to clients. The strategy is conveyed through the 3890 DNS resource records returned in response to name resolution queries. 3891 As of January 2020 Internet-based 'nslookup' services were used to 3892 determine the following: 3894 o When a client resolves the domainname "google.com", the DNS always 3895 returns one A record (i.e., an IPv4 address) and one AAAA record 3896 (i.e., an IPv6 address). The client receives the same addresses 3897 each time it resolves the domainname via the same DNS resolver, 3898 but may receive different addresses when it resolves the 3899 domainname via different DNS resolvers. But, in each case, 3900 exactly one A and one AAAA record are returned. 3902 o When a client resolves the domainname "ietf.org", the DNS always 3903 returns one A record and one AAAA record with the same addresses 3904 regardless of which DNS resolver is used. 3906 o When a client resolves the domainname "yahoo.com", the DNS always 3907 returns a list of 4 A records and 4 AAAA records. Each time the 3908 client resolves the domainname via the same DNS resolver, the same 3909 list of addresses are returned but in randomized order (i.e., 3910 consistent with a DNS round-robin strategy). But, interestingly, 3911 the same addresses are returned (albeit in randomized order) when 3912 the domainname is resolved via different DNS resolvers. 3914 o When a client resolves the domainname "amazon.com", the DNS always 3915 returns a list of 3 A records and no AAAA records. As with 3916 "yahoo.com", the same three A records are returned from any 3917 worldwide Internet connection point in randomized order. 3919 The above example strategies show differing approaches to Internet 3920 resilience and service distribution offered by major Internet 3921 services. The Google approach exposes only a single IPv4 and a 3922 single IPv6 address to clients. Clients can then select whichever IP 3923 protocol version offers the best response, but will always use the 3924 same IP address according to the current Internet connection point. 3925 This means that the IP address offered by the network must lead to a 3926 highly-available server and/or service distribution point. In other 3927 words, resilience is predicated on high availability within the 3928 network and with no client-initiated failovers expected (i.e., it is 3929 all-or-nothing from the client's perspective). However, Google does 3930 provide for worldwide distributed service distribution by virtue of 3931 the fact that each Internet connection point responds with a 3932 different IPv6 and IPv4 address. The IETF approach is like google 3933 (all-or-nothing from the client's perspective), but provides only a 3934 single IPv4 or IPv6 address on a worldwide basis. This means that 3935 the addresses must be made highly-available at the network level with 3936 no client failover possibility, and if there is any worldwide service 3937 distribution it would need to be conducted by a network element that 3938 is reached via the IP address acting as a service distribution point. 3940 In contrast to the Google and IETF philosophies, Yahoo and Amazon 3941 both provide clients with a (short) list of IP addresses with Yahoo 3942 providing both IP protocol versions and Amazon as IPv4-only. The 3943 order of the list is randomized with each name service query 3944 response, with the effect of round-robin load balancing for service 3945 distribution. With a short list of addresses, there is still 3946 expectation that the network will implement high availability for 3947 each address but in case any single address fails the client can 3948 switch over to using a different address. The balance then becomes 3949 one of function in the network vs function in the end system. 3951 The same implications observed for common highly-available services 3952 in the Internet apply also to the AERO client/server architecture. 3953 When an AERO Client connects to one or more ANETs, it discovers one 3954 or more AERO Server addresses through the mechanisms discussed in 3955 earlier sections. Each Server address presumably leads to a fault- 3956 tolerant clustering arrangement such as supported by Linux-HA, 3957 Extended Virtual Synchrony or Paxos. Such an arrangement has 3958 precedence in common Internet service deployments in lightweight 3959 virtual machines without requiring expensive hardware deployment. 3960 Similarly, common Internet service deployments set service IP 3961 addresses on service distribution points that may relay requests to 3962 many different servers. 3964 For AERO, the expectation is that a combination of the Google/IETF 3965 and Yahoo/Amazon philosophies would be employed. The AERO Client 3966 connects to different ANET access points and can receive 1-2 Server 3967 AERO addresses at each point. It then selects one AERO Server 3968 address, and engages in RS/RA exchanges with the same Server from all 3969 ANET connections. The Client remains with this Server unless or 3970 until the Server fails, in which case it can switch over to an 3971 alternate Server. The Client can likewise switch over to a different 3972 Server at any time if there is some reason for it to do so. So, the 3973 AERO expectation is for a balance of function in the network and end 3974 system, with fault tolerance and resilience at both levels. 3976 Appendix C. Change Log 3978 << RFC Editor - remove prior to publication >> 3980 Changes from draft-templin-intarea-6706bis-46 to draft-templin- 3981 intrea-6706bis-47: 3983 o Major changes to align with Teredo, including changed AERO "Relay" 3984 to "Bridge", and changed AERO "Gateway" to "Relay". The term 3985 "[Rr]elay" now refers to exactly the same thing in both AERO and 3986 Teredo. 3988 o Changed to use Teredo message authentication instead of SEND. 3990 Changes from draft-templin-intarea-6706bis-42 to draft-templin- 3991 intrea-6706bis-43: 3993 o Segment Routing. 3995 Changes from draft-templin-intarea-6706bis-39 to draft-templin- 3996 intrea-6706bis-40: 3998 o Teredo. 4000 Changes from draft-templin-intarea-6706bis-38 to draft-templin- 4001 intrea-6706bis-39: 4003 o Major clarifications and simplifications of SPAN fragmentation/ 4004 reassembly. 4006 o Revised AERO address format to support prefix lengths up to 112. 4008 o New method for forming SPAN Client Prefixes and population in the 4009 routing system. 4011 o Updates RFC4443 to set a new value in the ICMP PTB Code field. 4013 Changes from draft-templin-intarea-6706bis-35 to draft-templin- 4014 intrea-6706bis-36: 4016 o Clients in the open Internet secured using SEND/CGA. 4018 Changes from draft-templin-intarea-6706bis-32 to draft-templin- 4019 intrea-6706bis-33: 4021 o Updated Proxy discussion with "point-to-multipoint" server 4022 coordination 4024 o Significant updates to Address Resolution and NUD to include 4025 correct addresses in messages 4027 o Differentiate between NS(AR) and NS(NUD) as their addresses and 4028 use cases differ. 4030 Changes from draft-templin-intarea-6706bis-30 to draft-templin- 4031 intrea-6706bis-31: 4033 o Added "advisory PTB messages" under FAA SE2025 contract number 4034 DTFAWA-15-D-00030. 4036 Changes from draft-templin-intarea-6706bis-29 to draft-templin- 4037 intrea-6706bis-30: 4039 o Deprecate "primary" concept. Now, RS/RA keepalives are maintained 4040 over *all* underlying interfaces (i.e., and not just one primary). 4042 Changes from draft-templin-intarea-6706bis-28 to draft-templin- 4043 intrea-6706bis-29: 4045 o Changed OMNI interface citation to "draft-templin-6man-omni- 4046 interface" 4048 o Changed SPAN Service Prefix to fd80::/10. 4050 o Changed S/TLLAO format to include 'S' bit for ifIndex 4051 corresponding to the underlying interface that is Source of ND 4052 message. 4054 o Updated Path MTU 4056 Changes from draft-templin-intarea-6706bis-27 to draft-templin- 4057 intrea-6706bis-28: 4059 o MTU and fragmentation. 4061 Changes from draft-templin-intarea-6706bis-26 to draft-templin- 4062 intrea-6706bis-27: 4064 o MTU and fragmentation. 4066 o SPAN Service Prefix set to fd00::/10 4068 o Client SPAN addresses defined. 4070 Changes from draft-templin-intarea-6706bis-25 to draft-templin- 4071 intrea-6706bis-26: 4073 o MTU and RA configuration information updated. 4075 Changes from draft-templin-intarea-6706bis-24 to draft-templin- 4076 intrea-6706bis-25: 4078 o Added concept of "primary" to allow for proxyed RS/RA over only 4079 selected underlying interfaces. 4081 o General Cleanup. 4083 Changes from draft-templin-intarea-6706bis-23 to draft-templin- 4084 intrea-6706bis-24: 4086 o OMNI interface spec now a normative reference. 4088 o Use REACHABLE_TIME as the nominal Router Lifetime to return in 4089 RAs. 4091 o General cleanup. 4093 Changes from draft-templin-intarea-6706bis-22 to draft-templin- 4094 intrea-6706bis-23: 4096 o Choice of using either RS/RA or unsolicited NA for old Server 4097 notification. 4099 o General cleanup. 4101 Changes from draft-templin-intarea-6706bis-21 to draft-templin- 4102 intrea-6706bis-22: 4104 o Tightened up text on Proxy. 4106 o Removed unnecessarily restrictive texts. 4108 o General cleanup. 4110 Changes from draft-templin-intarea-6706bis-20 to draft-templin- 4111 intrea-6706bis-21: 4113 o Clarified relationship between OMNI and S/TLLAO ifIndex-tuples. 4115 o Important text in Section 13.15.3 on Servers timing out Clients 4116 that have gone silent without sending a departure notification. 4118 o New text on RS/RA as "hints of forward progress" for proactive 4119 NUD. 4121 Changes from draft-templin-intarea-6706bis-19 to draft-templin- 4122 intrea-6706bis-20: 4124 o Included new route optimization source and destination addressing 4125 strategy. Now, route optimization maintenance uses the address of 4126 the existing Server instead of the data packet destination address 4127 so that less pressure is placed on the BGP routing system 4128 convergence time and Server constancy is supported. 4130 o Included new method for releasing from old MSE without requiring 4131 Client messaging. 4133 o Included references to new OMNI interface spec (including the OMNI 4134 option). 4136 o New appendix on AERO Client/Server architecture. 4138 Changes from draft-templin-intarea-6706bis-18 to draft-templin- 4139 intrea-6706bis-19: 4141 o Changed Proxy/Server keepalives to use "proactive NUD" in a manner 4142 that parallels BFD 4144 Changes from draft-templin-intarea-6706bis-17 to draft-templin- 4145 intrea-6706bis-18: 4147 o Discuss how AERO option is used in relation to S/TLLAOs 4149 o New text on Bidirectional Forwarding Detection (BFD) 4151 o Cleaned up usage (and non-usage) of unsolicited NAs 4153 o New appendix on Server failures 4155 Changes from draft-templin-intarea-6706bis-15 to draft-templin- 4156 intrea-6706bis-17: 4158 o S/TLLAO now includes multiple link-layer addresses within a single 4159 option instead of requiring multiple options 4161 o New unsolicited NA message to inform the old link that a Client 4162 has moved to a new link 4164 Changes from draft-templin-intarea-6706bis-14 to draft-templin- 4165 intrea-6706bis-15: 4167 o MTU and fragmentation 4169 o New details in movement to new Server 4171 Changes from draft-templin-intarea-6706bis-13 to draft-templin- 4172 intrea-6706bis-14: 4174 o Security based on secured tunnels, ingress filtering, MAP list and 4175 ROS list 4177 Changes from draft-templin-intarea-6706bis-12 to draft-templin- 4178 intrea-6706bis-13: 4180 o New paragraph in Section 3.6 on AERO interface layering over 4181 secured tunnels 4183 o Removed extraneous text in Section 3.7 4185 o Added new detail to the forwarding algorithm in Section 3.9 4187 o Clarified use of fragmentation 4189 o Route optimization now supported for both MNP and non-MNP-based 4190 prefixes 4192 o Relays are now seen as link-layer elements in the architecture. 4194 o Built out multicast section in detail. 4196 o New Appendix on implementation considerations for route 4197 optimization. 4199 Changes from draft-templin-intarea-6706bis-11 to draft-templin- 4200 intrea-6706bis-12: 4202 o Introduced Gateways as a new AERO element for connecting 4203 Correspondent Nodes on INET links 4205 o Introduced terms "Access Network (ANET)" and "Internetwork (INET)" 4207 o Changed "ASP" to "MSP", and "ACP" to "MNP" 4209 o New figure on the relation of Segments to the SPAN and AERO link 4211 o New "S" bit in S/TLLAO to indicate the "Source" S/TLLAO as opposed 4212 to additional S/TLLAOs 4214 o Changed Interface ID for Servers from 255 to 0xffff 4216 o Significant updates to Route Optimization, NUD, and Mobility 4217 Management 4219 o New Section on Multicast 4221 o New Section on AERO Clients in the open Internetwork 4223 o New Section on Operation over multiple AERO links (VLANs over the 4224 SPAN) 4226 o New Sections on DNS considerations and Transition considerations 4228 o 4230 Changes from draft-templin-intarea-6706bis-10 to draft-templin- 4231 intrea-6706bis-11: 4233 o Added The SPAN 4235 Changes from draft-templin-intarea-6706bis-09 to draft-templin- 4236 intrea-6706bis-10: 4238 o Orphaned packets in flight (e.g., when a neighbor cache entry is 4239 in the DEPARTED state) are now forwarded at the link layer instead 4240 of at the network layer. Forwarding at the network layer can 4241 result in routing loops and/or excessive delays of forwarded 4242 packets while the routing system is still reconverging. 4244 o Update route optimization to clarify the unsecured nature of the 4245 first NS used for route discovery 4247 o Many cleanups and clarifications on ND messaging parameters 4249 Changes from draft-templin-intarea-6706bis-08 to draft-templin- 4250 intrea-6706bis-09: 4252 o Changed PRL to "MAP list" 4254 o For neighbor cache entries, changed "static" to "symmetric", and 4255 "dynamic" to "asymmetric" 4257 o Specified Proxy RS/RA exchanges with Servers on behalf of Clients 4259 o Added discussion of unsolicited NAs in Section 3.16, and included 4260 forward reference to Section 3.18 4262 o Added discussion of AERO Clients used as critical infrastructure 4263 elements to connect fixed networks. 4265 o Added network-based VPN under security considerations 4267 Changes from draft-templin-intarea-6706bis-07 to draft-templin- 4268 intrea-6706bis-08: 4270 o New section on AERO-Aware Access Router 4272 Changes from draft-templin-intarea-6706bis-06 to draft-templin- 4273 intrea-6706bis-07: 4275 o Added "R" bit for release of PDs. Now have a full RS/RA service 4276 that can do PD without requiring DHCPv6 messaging over-the-air 4278 o Clarifications on solicited vs unsolicited NAs 4280 o Clarified use of MAX_NEIGHBOR_ADVERTISEMENT for the purpose of 4281 increase reliability 4283 Changes from draft-templin-intarea-6706bis-05 to draft-templin- 4284 intrea-6706bis-06: 4286 o Major re-work and simplification of Route Optimization function 4288 o Added Distributed Mobility Management (DMM) and Mobility Anchor 4289 Point (MAP) terminology 4291 o New section on "AERO Critical Infrastructure Element 4292 Considerations" demonstrating low overall cost for the service 4294 o minor text revisions and deletions 4296 o removed extraneous appendices 4298 Changes from draft-templin-intarea-6706bis-04 to draft-templin- 4299 intrea-6706bis-05: 4301 o New Appendix E on S/TLLAO Extensions for special-purpose links. 4302 Discussed ATN/IPS as example. 4304 o New sentence in introduction to declare appendices as non- 4305 normative. 4307 Changes from draft-templin-intarea-6706bis-03 to draft-templin- 4308 intrea-6706bis-04: 4310 o Added definitions for Potential Router List (PRL) and secure 4311 enclave 4313 o Included text on mapping transport layer port numbers to network 4314 layer DSCP values 4316 o Added reference to DTLS and DMM Distributed Mobility Anchoring 4317 working group document 4319 o Reworked Security Considerations 4321 o Updated references. 4323 Changes from draft-templin-intarea-6706bis-02 to draft-templin- 4324 intrea-6706bis-03: 4326 o Added new section on SEND. 4328 o Clarifications on "AERO Address" section. 4330 o Updated references and added new reference for RFC8086. 4332 o Security considerations updates. 4334 o General text clarifications and cleanup. 4336 Changes from draft-templin-intarea-6706bis-01 to draft-templin- 4337 intrea-6706bis-02: 4339 o Note on encapsulation avoidance in Section 4. 4341 Changes from draft-templin-intarea-6706bis-00 to draft-templin- 4342 intrea-6706bis-01: 4344 o Remove DHCPv6 Server Release procedures that leveraged the old way 4345 Relays used to "route" between Server link-local addresses 4347 o Remove all text relating to Relays needing to do any AERO-specific 4348 operations 4350 o Proxy sends RS and receives RA from Server using SEND. Use CGAs 4351 as source addresses, and destination address of RA reply is to the 4352 AERO address corresponding to the Client's ACP. 4354 o Proxy uses SEND to protect RS and authenticate RA (Client does not 4355 use SEND, but rather relies on subnetwork security. When the 4356 Proxy receives an RS from the Client, it creates a new RS using 4357 its own addresses as the source and uses SEND with CGAs to send a 4358 new RS to the Server. 4360 o Emphasize distributed mobility management 4362 o AERO address-based RS injection of ACP into underlying routing 4363 system. 4365 Changes from draft-templin-aerolink-82 to draft-templin-intarea- 4366 6706bis-00: 4368 o Document use of NUD (NS/NA) for reliable link-layer address 4369 updates as an alternative to unreliable unsolicited NA. 4370 Consistent with Section 7.2.6 of RFC4861. 4372 o Server adds additional layer of encapsulation between outer and 4373 inner headers of NS/NA messages for transmission through Relays 4374 that act as vanilla IPv6 routers. The messages include the AERO 4375 Server Subnet Router Anycast address as the source and the Subnet 4376 Router Anycast address corresponding to the Client's ACP as the 4377 destination. 4379 o Clients use Subnet Router Anycast address as the encapsulation 4380 source address when the access network does not provide a 4381 topologically-fixed address. 4383 Author's Address 4385 Fred L. Templin (editor) 4386 Boeing Research & Technology 4387 P.O. Box 3707 4388 Seattle, WA 98124 4389 USA 4391 Email: fltemplin@acm.org