idnits 2.17.1 draft-templin-intarea-6706bis-43.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 15, 2020) is 1469 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 968 -- Looks like a reference, but probably isn't: '2' on line 974 == Missing Reference: 'N' is mentioned on line 984, but not defined == Unused Reference: 'RFC2474' is defined on line 3100, but no explicit reference was found in the text == Unused Reference: 'RFC5175' is defined on line 3138, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-gue-extensions' is defined on line 3184, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-tunnels' is defined on line 3189, but no explicit reference was found in the text == Unused Reference: 'RFC1191' is defined on line 3239, but no explicit reference was found in the text == Unused Reference: 'RFC2764' is defined on line 3264, but no explicit reference was found in the text == Unused Reference: 'RFC2923' is defined on line 3278, but no explicit reference was found in the text == Unused Reference: 'RFC3819' is defined on line 3296, but no explicit reference was found in the text == Unused Reference: 'RFC4607' is defined on line 3351, but no explicit reference was found in the text == Unused Reference: 'RFC4963' is defined on line 3355, but no explicit reference was found in the text == Unused Reference: 'RFC6864' is defined on line 3434, but no explicit reference was found in the text == Unused Reference: 'RFC7269' is defined on line 3438, but no explicit reference was found in the text == Unused Reference: 'RFC7421' is defined on line 3448, but no explicit reference was found in the text == Unused Reference: 'RFC8086' is defined on line 3460, but no explicit reference was found in the text == Outdated reference: A later version (-99) exists of draft-templin-6man-omni-interface-14 == Outdated reference: A later version (-13) exists of draft-ietf-intarea-tunnels-10 == Outdated reference: A later version (-26) exists of draft-ietf-rtgwg-atn-bgp-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 (~~), 22 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 15, 2020 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: October 17, 2020 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-intarea-6706bis-43 13 Abstract 15 This document specifies the operation of IP over tunnel virtual links 16 using Asymmetric Extended Route Optimization (AERO). AERO interfaces 17 use an IPv6 link-local address format that supports operation of the 18 IPv6 Neighbor Discovery (ND) protocol and links ND to IP forwarding. 19 Prefix delegation/registration services are employed for network 20 admission and to manage the routing system. Multilink operation, 21 mobility management, quality of service (QoS) signaling and route 22 optimization are naturally supported through dynamic neighbor cache 23 updates. Standard IP multicasting services are also supported. AERO 24 is a widely-applicable mobile internetworking service especially 25 well-suited to aviation services, mobile Virtual Private Networks 26 (VPNs) and many other applications. 28 Status of This Memo 30 This Internet-Draft is submitted in full conformance with the 31 provisions of BCP 78 and BCP 79. 33 Internet-Drafts are working documents of the Internet Engineering 34 Task Force (IETF). Note that other groups may also distribute 35 working documents as Internet-Drafts. The list of current Internet- 36 Drafts is at https://datatracker.ietf.org/drafts/current/. 38 Internet-Drafts are draft documents valid for a maximum of six months 39 and may be updated, replaced, or obsoleted by other documents at any 40 time. It is inappropriate to use Internet-Drafts as reference 41 material or to cite them other than as "work in progress." 43 This Internet-Draft will expire on October 17, 2020. 45 Copyright Notice 47 Copyright (c) 2020 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents 52 (https://trustee.ietf.org/license-info) in effect on the date of 53 publication of this document. Please review these documents 54 carefully, as they describe your rights and restrictions with respect 55 to this document. Code Components extracted from this document must 56 include Simplified BSD License text as described in Section 4.e of 57 the Trust Legal Provisions and are provided without warranty as 58 described in the Simplified BSD License. 60 Table of Contents 62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 63 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 64 3. Asymmetric Extended Route Optimization (AERO) . . . . . . . . 10 65 3.1. AERO Link Reference Model . . . . . . . . . . . . . . . . 10 66 3.2. AERO Node Types . . . . . . . . . . . . . . . . . . . . . 12 67 3.3. AERO Routing System . . . . . . . . . . . . . . . . . . . 13 68 3.4. AERO Addresses . . . . . . . . . . . . . . . . . . . . . 15 69 3.5. Spanning Partitioned AERO Networks (SPAN) . . . . . . . . 16 70 3.6. AERO Interface Characteristics . . . . . . . . . . . . . 20 71 3.7. AERO Interface Initialization . . . . . . . . . . . . . . 25 72 3.7.1. AERO Server/Gateway Behavior . . . . . . . . . . . . 25 73 3.7.2. AERO Proxy Behavior . . . . . . . . . . . . . . . . . 25 74 3.7.3. AERO Client Behavior . . . . . . . . . . . . . . . . 26 75 3.7.4. AERO Relay Behavior . . . . . . . . . . . . . . . . . 26 76 3.8. AERO Interface Neighbor Cache Maintenance . . . . . . . . 26 77 3.9. AERO Interface Encapsulation and Re-encapsulation . . . . 28 78 3.10. AERO Interface Decapsulation . . . . . . . . . . . . . . 29 79 3.11. AERO Interface Data Origin Authentication . . . . . . . . 29 80 3.12. AERO Interface MTU and Fragmentation . . . . . . . . . . 30 81 3.13. AERO Interface Forwarding Algorithm . . . . . . . . . . . 32 82 3.13.1. Client Forwarding Algorithm . . . . . . . . . . . . 32 83 3.13.2. Proxy Forwarding Algorithm . . . . . . . . . . . . . 33 84 3.13.3. Server/Gateway Forwarding Algorithm . . . . . . . . 34 85 3.13.4. Relay Forwarding Algorithm . . . . . . . . . . . . . 35 86 3.14. AERO Interface Error Handling . . . . . . . . . . . . . . 36 87 3.15. AERO Router Discovery, Prefix Delegation and 88 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 38 89 3.15.1. AERO ND/PD Service Model . . . . . . . . . . . . . . 38 90 3.15.2. AERO Client Behavior . . . . . . . . . . . . . . . . 39 91 3.15.3. AERO Server Behavior . . . . . . . . . . . . . . . . 41 92 3.16. The AERO Proxy . . . . . . . . . . . . . . . . . . . . . 44 93 3.16.1. Detecting and Responding to Server Failures . . . . 46 94 3.16.2. Point-to-Multipoint Server Coordindation . . . . . . 47 95 3.17. AERO Route Optimization . . . . . . . . . . . . . . . . . 47 96 3.17.1. Route Optimization Initiation . . . . . . . . . . . 48 97 3.17.2. Relaying the NS . . . . . . . . . . . . . . . . . . 48 98 3.17.3. Processing the NS and Sending the NA . . . . . . . . 48 99 3.17.4. Relaying the NA . . . . . . . . . . . . . . . . . . 50 100 3.17.5. Processing the NA . . . . . . . . . . . . . . . . . 50 101 3.17.6. Route Optimization Maintenance . . . . . . . . . . . 50 102 3.18. Neighbor Unreachability Detection (NUD) . . . . . . . . . 51 103 3.19. Mobility Management and Quality of Service (QoS) . . . . 52 104 3.19.1. Mobility Update Messaging . . . . . . . . . . . . . 53 105 3.19.2. Announcing Link-Layer Address and/or QoS Preference 106 Changes . . . . . . . . . . . . . . . . . . . . . . 54 107 3.19.3. Bringing New Links Into Service . . . . . . . . . . 54 108 3.19.4. Removing Existing Links from Service . . . . . . . . 54 109 3.19.5. Moving to a New Server . . . . . . . . . . . . . . . 55 110 3.20. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 56 111 3.20.1. Source-Specific Multicast (SSM) . . . . . . . . . . 56 112 3.20.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 57 113 3.20.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 58 114 3.21. Operation over Multiple AERO Links (VLANs) . . . . . . . 58 115 3.22. DNS Considerations . . . . . . . . . . . . . . . . . . . 59 116 3.23. Transition Considerations . . . . . . . . . . . . . . . . 60 117 3.24. Detecting and Reacting to Server and Relay Failures . . . 60 118 3.25. AERO Clients on the Open Internet . . . . . . . . . . . . 61 119 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 63 120 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 63 121 6. Security Considerations . . . . . . . . . . . . . . . . . . . 63 122 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 65 123 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 66 124 8.1. Normative References . . . . . . . . . . . . . . . . . . 66 125 8.2. Informative References . . . . . . . . . . . . . . . . . 68 126 Appendix A. AERO Alternate Encapsulations . . . . . . . . . . . 75 127 Appendix B. Non-Normative Considerations . . . . . . . . . . . . 77 128 B.1. Implementation Strategies for Route Optimization . . . . 77 129 B.2. Implicit Mobility Management . . . . . . . . . . . . . . 78 130 B.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 78 131 B.4. Operation on AERO Links with /64 ASPs . . . . . . . . . . 78 132 B.5. AERO Critical Infrastructure Considerations . . . . . . . 79 133 B.6. AERO Server Failure Implications . . . . . . . . . . . . 80 134 B.7. AERO Client / Server Architecture . . . . . . . . . . . . 80 135 Appendix C. Change Log . . . . . . . . . . . . . . . . . . . . . 82 136 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 91 138 1. Introduction 140 Asymmetric Extended Route Optimization (AERO) fulfills the 141 requirements of Distributed Mobility Management (DMM) [RFC7333] and 142 route optimization [RFC5522] for aeronautical networking and other 143 network mobility use cases. AERO is based on a Non-Broadcast, 144 Multiple Access (NBMA) virtual link model known as the AERO link. 145 The AERO link is a virtual overlay configured over one or more 146 underlying Internetworks, and nodes on the link can exchange IP 147 packets via tunneling. Multilink operation allows for increased 148 reliability, bandwidth optimization and traffic path diversity. 150 The AERO service comprises Clients, Proxys, Servers and Gateways that 151 are seen as AERO link neighbors as well as Relays that bridge AERO 152 link partitions. Each node's AERO interface uses an IPv6 link-local 153 address format (known as the AERO address) that supports operation of 154 the IPv6 Neighbor Discovery (ND) protocol [RFC4861] and links ND to 155 IP forwarding. A node's AERO interface can be configured over 156 multiple underlying interfaces, and may therefore appear as a single 157 interface with multiple link-layer addresses. Each link-layer 158 address is subject to change due to mobility and/or QoS fluctuations, 159 and link-layer address changes are signaled by ND messaging the same 160 as for any IPv6 link. 162 AERO links provide a cloud-based service where mobile nodes may use 163 any Server acting as a Mobility Anchor Point (MAP) and fixed nodes 164 may use any Gateway on the link for efficient communications. Fixed 165 nodes forward packets destined to other AERO nodes to the nearest 166 Gateway, which forwards them through the cloud. A mobile node's 167 initial packets are forwarded through the Server, while direct 168 routing is supported through asymmetric extended route optimization 169 while data packets are flowing. Both unicast and multicast 170 communications are supported, and mobile nodes may efficiently move 171 between locations while maintaining continuous communications with 172 correspondents and without changing their IP Address. 174 AERO Relays are interconnected in a secured private BGP overlay 175 routing instance known as "The SPAN". The SPAN provides a hybrid 176 routing/bridging service to join the underlying Internetworks of 177 multiple disjoint administrative domains into a single unified AERO 178 link. Each AERO link instance is characterized by the set of 179 Mobility Service Prefixes (MSPs) common to all mobile nodes. The 180 link extends to the point where a Gateway/Server is on the optimal 181 route from any correspondent node on the link, and provides a gateway 182 between the underlying Internetwork and the SPAN. To the underlying 183 Internetwork, the Gateway/Server is the source of a route to the MSP, 184 and hence uplink traffic to the mobile node is naturally routed to 185 the nearest Gateway/Server. 187 AERO assumes the use of PIM Sparse Mode in support of multicast 188 communication. In support of Source Specific Multicast (SSM) when a 189 Mobile Node is the source, AERO route optimization ensures that a 190 shortest-path multicast tree is established with provisions for 191 mobility and multilink operation. In all other multicast scenarios 192 there are no AERO dependencies. 194 AERO was designed for aeronautical networking for both manned and 195 unmanned aircraft, where the aircraft is treated as a mobile node 196 that can connect an Internet of Things (IoT). AERO is also 197 applicable to a wide variety of other use cases. For example, it can 198 be used to coordinate the Virtual Private Network (VPN) links of 199 mobile nodes (e.g., cellphones, tablets, laptop computers, etc.) that 200 connect into a home enterprise network via public access networks 201 using services such as OpenVPN [OVPN]. Other applicable use cases 202 are also in scope. 204 The following numbered sections present the AERO specification. The 205 appendices at the end of the document are non-normative. 207 2. Terminology 209 The terminology in the normative references applies; the following 210 terms are defined within the scope of this document: 212 IPv6 Neighbor Discovery (ND) 213 an IPv6 control message service for coordinating neighbor 214 relationships between nodes connected to a common link. AERO 215 interfaces use the ND service specified in [RFC4861]. 217 IPv6 Prefix Delegation (PD) 218 a networking service for delegating IPv6 prefixes to nodes on the 219 link. The nominal PD service is DHCPv6 [RFC8415], however 220 alternate services (e.g., based on ND messaging) are also in scope 221 [I-D.templin-v6ops-pdhost][I-D.templin-6man-dhcpv6-ndopt]. Most 222 notably, a minimal form of PD known as "prefix registration" can 223 be used if the Client knows its prefix in advance and can 224 represent it in the IPv6 source address of an ND message. 226 Access Network (ANET) 227 a node's first-hop data link service network, e.g., a radio access 228 network, cellular service provider network, corporate enterprise 229 network, or the public Internet itself. For secured ANETs, link- 230 layer security services such as IEEE 802.1X and physical-layer 231 security prevent unauthorized access internally while border 232 network-layer security services such as firewalls and proxies 233 prevent unauthorized outside access. 235 ANET interface 236 a node's attachment to a link in an ANET. 238 ANET address 239 an IP address assigned to a node's interface connection to an 240 ANET. 242 Internetwork (INET) 243 a connected IP network topology with a coherent routing and 244 addressing plan and that provides a transit backbone service for 245 ANET end systems. INETs also provide an underlay service over 246 which the AERO virtual link is configured. Example INETs include 247 corporate enterprise networks, aviation networks, and the public 248 Internet itself. When there is no administrative boundary between 249 an ANET and the INET, the ANET and INET are one and the same. 251 INET Partition 252 frequently, INETs such as large corporate enterprise networks are 253 sub-divided internally into separate isolated partitions. Each 254 partition is fully connected internally but disconnected from 255 other partitions, and there is no requirement that separate 256 partitions maintain consistent Internet Protocol and/or addressing 257 plans. (An INET partition is the same as a SPAN segment discussed 258 below.) 260 INET interface 261 a node's attachment to a link in an INET. 263 INET address 264 an IP address assigned to a node's interface connection to an 265 INET. 267 AERO link 268 a Non-Broadcast, Multiple Access (NBMA) tunnel virtual overlay 269 configured over one or more underlying INETs. Nodes on the AERO 270 link appear as single-hop neighbors from the perspective of the 271 virtual overlay even though they may be separated by many 272 underlying INET hops. AERO links may be configured over multiple 273 underlying SPAN segments (see below). 275 AERO interface 276 a node's attachment to an AERO link. Since the addresses assigned 277 to an AERO interface are managed for uniqueness, AERO interfaces 278 do not require Duplicate Address Detection (DAD) and therefore set 279 the administrative variable 'DupAddrDetectTransmits' to zero 280 [RFC4862]. 282 underlying interface 283 an ANET or INET interface over which an AERO interface is 284 configured. 286 AERO address 287 an IPv6 link-local address assigned to an AERO interface and 288 constructed as specified in Section 3.4. 290 base AERO address 291 the lowest-numbered AERO address aggregated by the MNP (see 292 Section 3.4). 294 Mobility Service Prefix (MSP) 295 an IP prefix assigned to the AERO link and from which more- 296 specific Mobile Network Prefixes (MNPs) are derived. 298 Mobile Network Prefix (MNP) 299 an IP prefix allocated from an MSP and delegated to an AERO Client 300 or Gateway. 302 AERO node 303 a node that is connected to an AERO link, or that provides 304 services to other nodes on an AERO link. 306 AERO Client ("Client") 307 an AERO node that connects to one or more ANETs and requests MNP 308 PDs from AERO Servers. The Client assigns a Client AERO address 309 to the AERO interface for use in ND exchanges with other AERO 310 nodes and forwards packets to correspondents according to AERO 311 interface neighbor cache state. 313 AERO Server ("Server") 314 an INET node that configures an AERO interface to provide default 315 forwarding and mobility/multilink services for AERO Clients. The 316 Server assigns an administratively-provisioned AERO address to its 317 AERO interface to support the operation of the ND/PD services, and 318 advertises all of its associated MNPs via BGP peerings with 319 Relays. 321 AERO Gateway ("Gateway") 322 an AERO Server that also provides forwarding services between 323 nodes reached via the AERO link and correspondents on other links. 324 AERO Gateways are provisioned with MNPs (i.e., the same as for an 325 AERO Client) and run a dynamic routing protocol to discover any 326 non-MNP IP routes. In both cases, the Gateway advertises the 327 MSP(s) over INET interfaces, and distributes all of its associated 328 MNPs and non-MNP IP routes via BGP peerings with Relays (i.e., the 329 same as for an AERO Server). 331 AERO Relay ("Relay") 332 a node that provides hybrid routing/bridging services (as well as 333 a security trust anchor) for nodes on an AERO link. As a router, 334 the Relay forwards packets using standard IP forwarding. As a 335 bridge, the Relay forwards packets over the AERO link without 336 decrementing the IPv6 Hop Limit. AERO Relays peer with Servers 337 and other Relays to discover the full set of MNPs for the link as 338 well as any non-MNPs that are reachable via Gateways. 340 AERO Proxy ("Proxy") 341 a node that provides proxying services between Clients in an ANET 342 and Servers in external INETs. The AERO Proxy is a conduit 343 between the ANET and external INETs in the same manner as for 344 common web proxies, and behaves in a similar fashion as for ND 345 proxies [RFC4389]. 347 Spanning Partitioned AERO Networks (SPAN) 348 a means for bridging disjoint INET partitions as segments of a 349 unified AERO link the same as for a bridged campus LAN. The SPAN 350 is a mid-layer IPv6 encapsulation service in the AERO routing 351 system that supports a unified AERO link view for all segments. 352 Each segment in the SPAN is a distinct INET partition. Segment 353 Routing [RFC8754] can be used to cause packets to visit selected 354 hops on the SPAN. 356 SPAN Service Prefix (SSP) 357 a global or unique local /96 IPv6 prefix assigned to the AERO link 358 to support SPAN services. 360 SPAN Partition Prefix (SPP) 361 a sub-prefix of the SPAN Service Prefix uniquely assigned to a 362 single SPAN segment. 364 SPAN Client Prefix (SCP) 365 a SPAN prefix formed from an AERO Client address. 367 SPAN Address 368 a unique local IPv6 address taken from a SPAN Client/Partition 369 Prefix and constructed as specified in Section 3.5. SPAN 370 addresses are statelessly derived from AERO addresses, and vice- 371 versa. 373 ingress tunnel endpoint (ITE) 374 an AERO interface endpoint that injects encapsulated packets into 375 an AERO link. 377 egress tunnel endpoint (ETE) 378 an AERO interface endpoint that receives encapsulated packets from 379 an AERO link. 381 link-layer address 382 an IP address used as an encapsulation header source or 383 destination address from the perspective of the AERO interface. 384 When an upper layer protocol (e.g., UDP) is used as part of the 385 encapsulation, the port number is also considered as part of the 386 link-layer address. From the perspective of the AERO interface, 387 the link-layer address is either an INET address for intra-segment 388 encapsulation or a SPAN address for inter-segment encapsulation. 390 network layer address 391 the source or destination address of an encapsulated IP packet 392 presented to the AERO interface. 394 end user network (EUN) 395 an internal virtual or external edge IP network that an AERO 396 Client or Gateway connects to the rest of the network via the AERO 397 interface. The Client/Gateway sees each EUN as a "downstream" 398 network, and sees the AERO interface as the point of attachment to 399 the "upstream" network. 401 Mobile Node (MN) 402 an AERO Client and all of its downstream-attached networks that 403 move together as a single unit, i.e., an end system that connects 404 an Internet of Things. 406 Mobile Router (MR) 407 a MN's on-board router that forwards packets between any 408 downstream-attached networks and the AERO link. 410 Route Optimization Source (ROS) 411 the AERO node nearest the source that initiates route 412 optimization. The ROS may be a Server or Proxy acting on behalf 413 of the source Client. 415 Route Optimization responder (ROR) 416 the AERO node nearest the target destination that responds to 417 route optimization requests. The ROR may be a Server acting on 418 behalf of a target MNP Client, or a Gateway for a non-MNP 419 destination. 421 MAP List 422 a geographically and/or topologically referenced list of AERO 423 addresses of all Servers within the same AERO link. There is a 424 single MAP list for the entire AERO link. 426 ROS List 427 a list of AERO/SPAN-to-INET address mappings of all ROSes within 428 the same SPAN segment. There is a distinct ROS list for each 429 segment. 431 Distributed Mobility Management (DMM) 432 a BGP-based overlay routing service coordinated by Servers and 433 Relays that tracks all Server-to-Client associations. 435 Mobility Service (MS) 436 the collective set of all Servers, Proxys, Relays and Gateways 437 that provide the AERO Service to Clients. 439 Mobility Service Endpoint MSE) 440 an individual Server, Proxy, Relay or Gateway in the Mobility 441 Service. 443 Throughout the document, the simple terms "Client", "Server", 444 "Relay", "Proxy" and "Gateway" refer to "AERO Client", "AERO Server", 445 "AERO Relay", "AERO Proxy" and "AERO Gateway", respectively. 446 Capitalization is used to distinguish these terms from other common 447 Internetworking uses in which they appear without capitalization. 449 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 450 the names of node variables, messages and protocol constants) is used 451 throughout this document. The terms "All-Routers multicast", "All- 452 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 453 anycast" are defined in [RFC4291] (with Link-Local scope assumed). 454 Also, the term "IP" is used to generically refer to either Internet 455 Protocol version, i.e., IPv4 [RFC0791] or IPv6 [RFC8200]. 457 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 458 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 459 "OPTIONAL" in this document are to be interpreted as described in BCP 460 14 [RFC2119][RFC8174] when, and only when, they appear in all 461 capitals, as shown here. 463 3. Asymmetric Extended Route Optimization (AERO) 465 The following sections specify the operation of IP over Asymmetric 466 Extended Route Optimization (AERO) links: 468 3.1. AERO Link Reference Model 469 +----------------+ 470 | AERO Relay R1 | 471 | Nbr: S1, S2, P1| 472 |(X1->S1; X2->S2)| 473 | MSP M1 | 474 +-+---------+--+-+ 475 +--------------+ | Secured | | +--------------+ 476 |AERO Server S1| | tunnels | | |AERO Server S2| 477 | Nbr: C1, R1 +-----+ | +-----+ Nbr: C2, R1 | 478 | default->R1 | | | default->R1 | 479 | X1->C1 | | | X2->C2 | 480 +-------+------+ | +------+-------+ 481 | AERO Link | | 482 X===+===+===================+==)===============+===+===X 483 | | | | 484 +-----+--------+ +--------+--+-----+ +--------+-----+ 485 |AERO Client C1| | AERO Proxy P1 | |AERO Client C2| 486 | Nbr: S1 | |(Proxy Nbr Cache)| | Nbr: S2 | 487 | default->S1 | +--------+--------+ | default->S2 | 488 | MNP X1 | | | MNP X2 | 489 +------+-------+ .--------+------. +-----+--------+ 490 | (- Proxyed Clients -) | 491 .-. `---------------' .-. 492 ,-( _)-. ,-( _)-. 493 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 494 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 495 `-(______)-' +-------+ +-------+ `-(______)-' 497 Figure 1: AERO Link Reference Model 499 Figure 1 presents the AERO link reference model. In this model: 501 o the AERO link is an overlay network service configured over one or 502 more underlying INET partitions which may be managed by different 503 administrative authorities and have incompatible protocols and/or 504 addressing plans. 506 o AERO Relay R1 aggregates Mobility Service Prefix (MSP) M1, 507 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 508 via BGP peerings over secured tunnels to Servers (S1, S2). Relays 509 use the SPAN service to bridge disjoint segments of a partitioned 510 AERO link. 512 o AERO Servers S1 and S2 configure secured tunnels with Relay R1 and 513 also provide mobility, multilink and default router services for 514 their associated Clients C1 and C2. 516 o AERO Clients C1 and C2 associate with Servers S1 and S2, 517 respectively. They receive Mobile Network Prefix (MNP) 518 delegations X1 and X2, and also act as default routers for their 519 associated physical or internal virtual EUNs. Simple hosts H1 and 520 H2 attach to the EUNs served by Clients C1 and C2, respectively. 522 o AERO Proxy P1 configures a secured tunnel with Relay R1 and 523 provides proxy services for AERO Clients in secured enclaves that 524 cannot associate directly with other AERO link neighbors. 526 Each node on the AERO link maintains an AERO interface neighbor cache 527 and an IP forwarding table the same as for any link. Although the 528 figure shows a limited deployment, in common operational practice 529 there will normally be many additional Relays, Servers, Clients and 530 Proxys. 532 3.2. AERO Node Types 534 AERO Relays provide hybrid routing/bridging services (as well as a 535 security trust anchor) for nodes on an AERO link. Relays use 536 standard IPv6 routing to forward packets both within the same INET 537 partitions and between disjoint INET partitions based on a mid-layer 538 IPv6 encapsulation known as the SPAN header. The inner IP layer 539 experiences a virtual bridging service since the inner IP TTL/Hop 540 Limit is not decremented during forwarding. Each Relay also peers 541 with Servers and other Relays in a dynamic routing protocol instance 542 to provide a Distributed Mobility Management (DMM) service for the 543 list of active MNPs (see Section 3.3). Relays present the AERO link 544 as a set of one or more Mobility Service Prefixes (MSPs) but as link- 545 layer devices need not connect directly to the AERO link themselves 546 unless an administrative interface is desired. Relays configure 547 secured tunnels with Servers, Proxys and other Relays; they further 548 maintain IP forwarding table entries for each Mobile Network Prefix 549 (MNP) and any other reachable non-MNP prefixes. 551 AERO Servers provide default forwarding and mobility/multilink 552 services for AERO Client Mobile Nodes (MNs). Each Server also peers 553 with Relays in a dynamic routing protocol instance to advertise its 554 list of associated MNPs (see Section 3.3). Servers facilitate PD 555 exchanges with Clients, where each delegated prefix becomes an MNP 556 taken from an MSP. Servers forward packets between AERO interface 557 neighbors and track each Client's mobility profiles. 559 AERO Clients register their MNPs through PD exchanges with AERO 560 Servers over the AERO link, and distribute the MNPs to nodes on EUNs. 561 A Client may also be co-resident on the same physical or virtual 562 platform as a Server; in that case, the Client and Server behave as a 563 single functional unit. 565 AERO Proxys provide a conduit for ANET AERO Clients to associate with 566 AERO Servers in external INETs. Client and Servers exchange control 567 plane messages via the Proxy acting as a bridge between the ANET/INET 568 boundary. The Proxy forwards data packets between Clients and the 569 AERO link according to forwarding information in the neighbor cache. 570 The Proxy function is specified in Section 3.16. 572 AERO Gateways are Servers that provide forwarding services between 573 the AERO interface and INET/EUN interfaces. Gateways are provisioned 574 with MNPs the same as for an AERO Client, and also run a dynamic 575 routing protocol to discover any non-MNP IP routes. The Gateway 576 advertises the MSP(s) to INETs, and distributes all of its associated 577 MNPs and non-MNP IP routes via BGP peerings with Relays. 579 AERO Relays, Servers, Proxys and Gateways are critical infrastructure 580 elements in fixed (i.e., non-mobile) INET deployments and hence have 581 permanent and unchanging INET addresses. AERO Clients are MNs that 582 connect via ANET interfaces, i.e., their ANET addresses may change 583 when the Client moves to a new ANET connection. 585 3.3. AERO Routing System 587 The AERO routing system comprises a private instance of the Border 588 Gateway Protocol (BGP) [RFC4271] that is coordinated between Relays 589 and Servers and does not interact with either the public Internet BGP 590 routing system or any underlying INET routing systems. 592 In a reference deployment, each Server is configured as an Autonomous 593 System Border Router (ASBR) for a stub Autonomous System (AS) using 594 an AS Number (ASN) that is unique within the BGP instance, and each 595 Server further uses eBGP to peer with one or more Relays but does not 596 peer with other Servers. Each INET of a multi-segment AERO link must 597 include one or more Relays, which peer with the Servers and Proxys 598 within that INET. All Relays within the same INET are members of the 599 same hub AS using a common ASN, and use iBGP to maintain a consistent 600 view of all active MNPs currently in service. The Relays of 601 different INETs peer with one another using eBGP. 603 Relays advertise the AERO link's MSPs and any non-MNP routes to each 604 of their Servers. This means that any aggregated non-MNPs (including 605 "default") are advertised to all Servers. Each Relay configures a 606 black-hole route for each of its MSPs. By black-holing the MSPs, the 607 Relay will maintain forwarding table entries only for the MNPs that 608 are currently active, and packets destined to all other MNPs will 609 correctly incur Destination Unreachable messages due to the black- 610 hole route. In this way, Servers have only partial topology 611 knowledge (i.e., they know only about the MNPs of their directly 612 associated Clients) and they forward all other packets to Relays 613 which have full topology knowledge. 615 Servers maintain a working set of associated MNPs, and dynamically 616 announce new MNPs and withdraw departed MNPs in eBGP updates to 617 Relays. Servers that are configured as Gateways also redistribute 618 non-MNP routes learned from non-AERO interfaces via their eBGP Relay 619 peerings. 621 Clients are expected to remain associated with their current Servers 622 for extended timeframes, however Servers SHOULD selectively suppress 623 updates for impatient Clients that repeatedly associate and 624 disassociate with them in order to dampen routing churn. Servers 625 that are configured as Gateways advertise the MSPs via INET/EUN 626 interfaces, and forward packets between INET/EUN interfaces and the 627 AERO interface using standard IP forwarding. 629 Scaling properties of the AERO routing system are limited by the 630 number of BGP routes that can be carried by Relays. As of 2015, the 631 global public Internet BGP routing system manages more than 500K 632 routes with linear growth and no signs of router resource exhaustion 633 [BGP]. More recent network emulation studies have also shown that a 634 single Relay can accommodate at least 1M dynamically changing BGP 635 routes even on a lightweight virtual machine, i.e., and without 636 requiring high-end dedicated router hardware. 638 Therefore, assuming each Relay can carry 1M or more routes, this 639 means that at least 1M Clients can be serviced by a single set of 640 Relays. A means of increasing scaling would be to assign a different 641 set of Relays for each set of MSPs. In that case, each Server still 642 peers with one or more Relays, but institutes route filters so that 643 BGP updates are only sent to the specific set of Relays that 644 aggregate the MSP. For example, if the MSP for the AERO link is 645 2001:db8::/32, a first set of Relays could service the MSP 646 2001:db8::/40, a second set of Relays could service 647 2001:db8:0100::/40, a third set could service 2001:db8:0200::/40, 648 etc. 650 Assuming up to 1K sets of Relays, the AERO routing system can then 651 accommodate 1B or more MNPs with no additional overhead (for example, 652 it should be possible to service 1B /64 MNPs taken from a /34 MSP and 653 even more for shorter prefixes). In this way, each set of Relays 654 services a specific set of MSPs that they advertise to the native 655 Internetwork routing system, and each Server configures MSP-specific 656 routes that list the correct set of Relays as next hops. This 657 arrangement also allows for natural incremental deployment, and can 658 support small scale initial deployments followed by dynamic 659 deployment of additional Clients, Servers and Relays without 660 disturbing the already-deployed base. 662 Server and Relays can use the Bidirectional Forwarding Detection 663 (BFD) protocol [RFC5880] to quickly detect link failures that don't 664 result in interface state changes, BGP peer failures, and 665 administrative state changes. BFD is important in environments where 666 rapid response to failures is required for routing reconvergence and, 667 hence, communications continuity. 669 A full discussion of the BGP-based routing system used by AERO is 670 found in [I-D.ietf-rtgwg-atn-bgp]. The system provides for 671 Distributed Mobility Management (DMM) per the distributed mobility 672 anchoring architecture [I-D.ietf-dmm-distributed-mobility-anchoring]. 674 3.4. AERO Addresses 676 A Client's AERO address is an IPv6 link-local address formed from the 677 Client's delegated MNP. Relay, Server and Proxy AERO addresses are 678 assigned from the range fe80::/96 and include an administratively- 679 provisioned value in the lower 32 bits. 681 IPv6 Client AERO addresses encode the Subnet-Router anycast address 682 of a MNP (or non-MNP globally routable prefix) within the least- 683 significant 112 bits of the IPv6 link-local prefix fe80::/16. For 684 example, for the MNP 2001:db8:1000:2000::/56 the corresponding AERO 685 address is fe80:2001:db8:1000:2000::/72. 687 IPv4-compatible Client AERO addresses are based on an IPv4-mapped 688 IPv6 address [RFC4291] formed from an IPv4 MNP and with a prefix 689 length of 96 plus the MNP prefix length. For example, for the IPv4 690 MNP 192.0.2.16/28 the IPv4-mapped IPv6 MNP is: 692 0:0:0:0:0:ffff:192.0.2.16/124 (also written as 693 0:0:0:0:0:ffff:c000:0210/124) 695 The Client then constructs its AERO address with the prefix fe80::/64 696 and with the lower 64 bits of the IPv4-mapped IPv6 address in the 697 interface identifier as: fe80::ffff:192.0.2.16. 699 Mobility Service (MS) AERO addresses (used by Relays, Servers, 700 Gateways and Proxys) are allocated from the range fe80::/96, and MUST 701 be managed for uniqueness. The lower 32 bits of the AERO address 702 includes a unique integer value between 1 and 0xfeffffff (e.g., 703 fe80::1, fe80::2, fe80::3, etc., fe80::feff:ffff) as assigned by the 704 administrative authority for the link. The address fe80:: is the 705 IPv6 link-local Subnet-Router anycast address, and the address range 706 fe80::ff00:0000/104 is reserved for future use. 708 Finally, the address range fe80::/32 is used as the Teredo service 709 prefix for AERO according to the format in Section 4 of [RFC4380] 710 (see Section 3.25 for further discussion). 712 For a full discussion of the above address format and implications 713 for the /64 boundary, see: [I-D.templin-6man-omni-interface]. 715 3.5. Spanning Partitioned AERO Networks (SPAN) 717 An AERO link configured over a single INET appears as a single 718 unified link with a consistent underlying network addressing plan. 719 In that case, all nodes on the link can exchange packets via simple 720 INET encapsulation, since the underlying INET is connected. In 721 common practice, however, an AERO link may be partitioned into 722 multiple "segments", where each segment is a distinct INET 723 potentially managed under a different administrative authority (e.g., 724 as for worldwide aviation service providers such as ARINC, SITA, 725 Inmarsat, etc.). Individual INETs may also themselves be partitioned 726 internally, in which case each internal partition is seen as a 727 separate segment. 729 The addressing plan of each segment is consistent internally but will 730 often bear no relation to the addressing plans of other segments. 731 Each segment is also likely to be separated from others by network 732 security devices (e.g., firewalls, proxies, packet filtering 733 gateways, etc.), and in many cases disjoint segments may not even 734 have any common physical link connections. Therefore, nodes can only 735 be assured of exchanging packets directly with correspondents in the 736 same segment, and not with those in other segments. The only means 737 for joining the segments therefore is through inter-domain peerings 738 between AERO Relays. 740 The same as for traditional campus LANs, multiple AERO link segments 741 can be joined into a single unified link via a virtual bridging 742 service termed "The SPAN". The SPAN performs link-layer packet 743 forwarding between segments (i.e., bridging) without decrementing the 744 network-layer TTL/Hop Limit. The SPAN model is depicted in Figure 2: 746 . . . . . . . . . . . . . . . . . . . . . . . 747 . . 748 . .-(::::::::) . 749 . .-(::::::::::::)-. +-+ . 750 . (:::: Segment A :::)--|R|---+ . 751 . `-(::::::::::::)-' +-+ | . 752 . `-(::::::)-' | . 753 . | . 754 . .-(::::::::) | . 755 . .-(::::::::::::)-. +-+ | . 756 . (:::: Segment B :::)--|R|---+ . 757 . `-(::::::::::::)-' +-+ | . 758 . `-(::::::)-' | . 759 . | . 760 . .-(::::::::) | . 761 . .-(::::::::::::)-. +-+ | . 762 . (:::: Segment C :::)--|R|---+ . 763 . `-(::::::::::::)-' +-+ | . 764 . `-(::::::)-' | . 765 . | . 766 . ..(etc).. x . 767 . . 768 . . 769 . <- AERO Link Bridged by the SPAN -> . 770 . . . . . . . . . . . . . .. . . . . . . . . 772 Figure 2: The SPAN 774 To support the SPAN, AERO links use the Unique Local Address (ULA) 775 prefix fd80::/16 [RFC4193] as the SPAN Service Prefix (SSP). The 776 prefix length intentionally matches the IPv6 link-local prefix 777 (fe80::/16), and enables a simple 1-bit stateless translation between 778 AERO and SPAN addresses (i.e., bit 7 is '1' for AERO addresses or '0' 779 for SPAN addresses). 781 Each segment in the SPAN assigns a unique sub-prefix of fd80::/96 782 termed the "SPAN Partition Prefix (SPP)". For example, a first 783 segment could assign fd80::1000/116, a second could assign 784 fd80::2000/116, a third could assign fd80::3000/116, etc. The 785 administrative authorities for each segment must therefore coordinate 786 to assure mutually-exclusive SPP assignments, but internal 787 provisioning of the SPP is an independent local consideration for 788 each administrative authority. 790 SPAN addresses are formed by simply clearing bit 7 of the 791 corresponding AERO address. For example: 793 o the SPAN address formed from the IPv6 Client AERO address 794 fe80:2001:db8:1000:2000:: is simply fd80:2001:db8:1000:2000:: 796 o the SPAN address formed from the IPv4-compatible Client AERO 797 address fe80::ffff:192.0.2.1 is simply fd80::ffff:192.0.2.1 799 o the SPAN address formed from the administrative AERO address 800 fe80::1001 is simply fd80::1001. 802 AERO Relays serve as bridges to join multiple segments into a unified 803 AERO link over multiple diverse administrative domains. They support 804 the bridging function by first establishing forwarding table entries 805 for their SPPs either via standard BGP routing or static routes. For 806 example, if three Relays ('A', 'B' and 'C') from different segments 807 serviced the SPPs fd80::1000/116, fd80::2000/116 and fd80::3000/116 808 respectively, then the forwarding tables in each Relay are as 809 follows: 811 A: fd80::1000/116->local, fd80::2000/116->B, fd80::3000/116->C 813 B: fd80::1000/116->A, fd80::2000/116->local, fd80::3000/116->C 815 C: fd80::1000/116->A, fd80::2000/116->B, fd80::3000/116->local 817 These forwarding table entries are permanent and never change, since 818 they correspond to fixed infrastructure elements in their respective 819 segments. 821 SPAN Client Prefixes (SCPs) are instead dynamically advertised in the 822 AERO link routing system by Servers and Gateways that aggregate their 823 correspnding MNPs. For example, if three Servers ('D', 'E' and 'F') 824 aggregated the MNPs 2001:db8:1000:2000::/56, 2001:db8:3000:4000::/56 825 and 2001:db8:5000:6000::/56 then the routing system would include: 827 D: fd80:2001:db8:1000:2000::/72 829 E: fd80:2001:db8:3000:4000::/72 831 F: fd80:2001:db8:5000:6000::/72 833 With the SCPs and SPPs in place in each Relay's forwarding table, 834 control and data packets sent between AERO nodes in different 835 segments can therefore be carried over the via encapsulation in a 836 mid-layer IPv6 header known as the "SPAN header". For example, when 837 a source AERO node forwards a packet with IPv6 address 838 2001:db8:1:2::1 to a target AERO node with IPv6 address 839 2001:db8:1000:2000::1, it first encapsulates the packet in a SPAN 840 header with source address set to fd80:2001:db8:1:2:: and destination 841 address set to fd80:2001:db8:1000:2000::. Next, it encapsulates the 842 resulting SPAN packet in an INET header with source address set to 843 its own INET address (e.g., 192.0.2.100) and destination set to the 844 INET address of a Relay (e.g., 192.0.2.1). 846 SPAN encapsulation is based on Generic Packet Tunneling in IPv6 847 [RFC2473]; the encapsulation format in the above example is shown in 848 Figure 3: 850 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 851 | INET Header | 852 | src = 192.0.2.100 | 853 | dst = 192.0.2.1 | 854 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 855 | SPAN Header | 856 | src = fd80:2001:db8:1:2:: | 857 | dst=fd80:2001:db8:1000:2000:: | 858 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 859 | Inner IP Header | 860 | src = 2001:db8:1:2::1 | 861 | dst = 2001:db8:1000:2000::1 | 862 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 863 | | 864 ~ ~ 865 ~ Inner Packet Body ~ 866 ~ ~ 867 | | 868 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 870 Figure 3: SPAN Encapsulation 872 In this format, the inner IP header and packet body are the original 873 IP packet, the SPAN header is an IPv6 header prepared according to 874 [RFC2473], and the INET header is prepared as discussed in 875 Section 3.9. A packet is said to be "forwarded/sent into the SPAN" 876 when it is encapsulated as described above then forwarded via a 877 secured tunnel to a neighboring Relay. 879 This gives rise to a routing system that contains both SCP routes 880 that may change dynamically due to regional node mobility and SPP 881 routes that never change. The Relays can therefore provide link- 882 layer bridging by sending packets into the SPAN instead of network- 883 layer routing according to MNP routes. As a result, opportunities 884 for packet loss due to node mobility between different segments are 885 mitigated. 887 In normal operations, IPv6 ND messages are conveyed to SPP addresses 888 over the SPAN so that specific Proxys, Servers or Gateways can be 889 addressed without being subject to mobility events. Conversely, only 890 the first few packets destined to SCP addresses traverse the SPAN 891 until route optimization can determine a more direct path. 893 3.6. AERO Interface Characteristics 895 AERO interfaces are virtual interfaces configured over one or more 896 underlying interfaces classified as follows: 898 o Native interfaces have global IP addresses that are reachable from 899 any INET correspondent. All Server, Gateway and Relay interfaces 900 are native interfaces, as are INET-facing interfaces of Proxys. 901 Clients connected to native interfaces receive Router 902 Advertisemenets (RAs) with a Client AERO destination address and 903 the P flag set to 0, as discussed in the following sections. 905 o NATed interfaces connect to a private network behind a Network 906 Address Translator (NAT). The NAT does not participate in any 907 AERO control message signaling, but the Server can issue control 908 messages on behalf of the Client. Clients that are behind a NAT 909 are required to send periodic keepalive messages to keep NAT state 910 alive when there are no data packets flowing. If no other 911 periodic messaging service is available, the Client can send 912 Router Solicitation (RS) messages to receive RA replies from its 913 Server(s). Clients connected to NATed interfaces receive RAs with 914 a Teredo-formatted AERO destination address. 916 o Proxyed interfaces connect to an ANET that is separated from the 917 open INET by an AERO Proxy. Unlike NATed and VPNed interfaces, 918 the Proxy can actively issue control messages on behalf of the 919 Client. Clients connected to Proxyed interfaces receive RAs with 920 the P flag set to 1. 922 o VPNed interfaces use security encapsulation to a Virtual Private 923 Network (VPN) server that also acts as an AERO Server. As with 924 NATed links, the Server can issue control messages on behalf of 925 the Client, and the Client can send periodic messages such as RS/ 926 RA to keep the VPN state alive. Clients connected to VPNed 927 interfaces receive RAs with the P flag set to 1 the same as for 928 Proxyed interfaces. 930 o Direct interfaces connect a Client directly to a Server without 931 crossing any ANET/INET paths. An example is a line-of-sight link 932 between a remote pilot and an unmanned aircraft. The same Client 933 considerations apply as for VPNed interfaces above, and the Client 934 receives RA messages with the P flag set to 1. 936 AERO interfaces use SPAN-layer encapsualtion as necessary as 937 discussed in Section 3.5. AERO interfaces use link-layer 938 encapsulation (see: Section 3.9) to exchange packets with AERO link 939 neighbors over Native, NATed or VPNed interfaces. AERO interfaces do 940 not use encapsulation over Proxyed and Direct underlying interfaces. 942 AERO interfaces maintain a neighbor cache for tracking per-neighbor 943 state the same as for any interface. AERO interfaces use ND messages 944 including Router Solicitation (RS), Router Advertisement (RA), 945 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 946 neighbor cache management. 948 AERO interfaces send ND messages with an Overlay Multilink Network 949 Interface (OMNI) option formatted as specified in 950 [I-D.templin-6man-omni-interface]. The OMNI option includes prefix 951 registration information and "ifIndex-tuples" containing link 952 information parameters for the AERO interface's underlying 953 interfaces. 955 When encapsulation is used, AERO interface ND messages MAY also 956 include an AERO Source/Target Link-Layer Address Option (S/TLLAO) 957 formatted as shown in Figure 4: 959 0 1 2 3 960 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 1 961 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 962 | Type | Length | ifIndex[1] |SRT| LHS |FMT|P| 963 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 964 ~ Segment Routing List [1] ~ 965 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 966 ~ Link Layer Address [1] ~ 967 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 968 | Port Number [1] | ifIndex[2] |SRT| LHS |FMT|P| 969 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 970 ~ Segment Routing List [2] ~ 971 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 972 ~ Link Layer Address [2] ~ 973 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 974 | Port Number [2] | .... ~ 975 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 976 ~ ... ~ 977 ~ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 978 ~ | ifIndex[N] |SRT| LHS |FMT|P| 979 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 980 ~ Segment Routing List [N] ~ 981 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 982 ~ Link Layer Address [N] ~ 983 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 984 | Port Number [N] | Zero Padding (if necessary) ... 985 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+--+-+-+-+-+-+-+-+-+-+-+-+-+-+- 987 Figure 4: AERO Source/Target Link-Layer Address Option (S/TLLAO) 988 Format 990 In this format, Type and Length are set the same as specified for S/ 991 TLLAOs in [RFC4861], with trailing zero padding octets added as 992 necessary to produce an integral number of 8 octet blocks. The S/ 993 TLLAO includes N ifIndex-tuples in correspondence to ifIndex-tuples 994 that appear in the OMNI option. Each ifIndex-tuple includes the 995 folllowing information: 997 o ifIndex[i] - the same value as in the corresponding ifIndex-tuple 998 included in the OMNI option. 1000 o SRT[i] - a 2-bit "Segment Routing Tone" value coded as follows: 1002 * 00 - Red 1004 * 01 - Green 1006 * 10 - Blue-1 1007 * 11 - Blue-2 1009 o LHS[i] - a 3-bit "LookaHead Segments" value that encodes the 1010 number (from 0 to 7) of entries in Segment Routing List [i]. 1012 o FMT[i] - a 2-bit "Format" code. Determines the format of the Link 1013 Layer Address [i] field as follows: 1015 * 00 - Link Layer Address [i] is unspecified. 1017 * 01 - Link Layer Address [i] encodes a native IPv4 address. 1019 * 10 - Link Layer Address [i] encodes a NATed IPv4 address. 1021 * 11 - Link Layer Address [i] encodes an IPv6 address. 1023 o P[i] - the "Port" bit. If set to 0 the Port Number [i] field is 1024 omitted; otherwise, the Port Number [i] field is included. 1026 o Segment Routing List [i] - Includes LHS[i]-many 4 byte 1027 administrative SPAN IDs for Segment routing, in the order to be 1028 visited on the path to the source/target (i.e., ID 1 first, then 1029 ID 2, then ID 3, etc). If LHS[i] is 0, no SPAN IDs are present 1030 and the FMT[i] and Link Layer Address [i] fields are consulted. 1032 o Link Layer Address [i] - Included according to FMT[i], and is 0 1033 bytes in length for "unspecified", 4 bytes in length for IPv4 or 1034 16 bytes in length for IPv6. For IP addresses, the value is 1035 written in "obfuscated" form by applying an exclusive-or of all 1036 bits of the address with "all-ones" as discussed in [RFC4380] - 1037 see also Section 3.25. 1039 o Port Number [i] - When present, the field is 2 bytes in length and 1040 immediately follows Link Layer Address [i]. Encodes the upper 1041 layer protocol port number used as the encapsulation source port. 1042 As with the Link-Layer Address, the value is written in 1043 "obfuscated" form. 1045 If an S/TLLAO is included, any ifIndex-tuples correspond to a proper 1046 subset of the OMNI option ifIndex-tuples. Any S/TLLAO ifIndex-tuple 1047 having an ifIndex value that does not appear in an OMNI option 1048 ifindex-tuple is ignored. If the same ifIndex value appears in 1049 multiple ifIndex-tuples, the first tuple is processed and the 1050 remaining tuples are ignored. Any S/TLLAO ifIndex-tuples can 1051 therefore be viewed as inter-dependent extensions of their 1052 corresponidng OMNI option ifIndex-tuples, i.e., the OMNI option and 1053 S/TLLAO are companions that are interpreted in conjunction with each 1054 other. 1056 A Client's AERO interface may be configured over multiple underlying 1057 interface connections. For example, common mobile handheld devices 1058 have both wireless local area network ("WLAN") and cellular wireless 1059 links. These links are often used "one at a time" with low-cost WLAN 1060 preferred and highly-available cellular wireless as a standby, but a 1061 simultaneous-use capability could provide benefits. In a more 1062 complex example, aircraft frequently have many wireless data link 1063 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1064 directional, etc.) with diverse performance and cost properties. 1066 If a Client's multiple underlying interfaces are used "one at a time" 1067 (i.e., all other interfaces are in standby mode while one interface 1068 is active), then ND message OMNI options include only a single 1069 ifIndex-tuple set to constant values. In that case, the Client would 1070 appear to have a single interface but with a dynamically changing 1071 link-layer address. 1073 If the Client has multiple active underlying interfaces, then from 1074 the perspective of ND it would appear to have multiple link-layer 1075 addresses. In that case, ND message OMNI options MAY include 1076 multiple ifIndex-tuples - each with values that correspond to a 1077 specific interface. Every ND message need not include all OMNI and/ 1078 or S/TLLAO ifIndex-tuples; for any ifIndex-tuple not included, the 1079 neighbor considers the status as unchanged. 1081 Relay, Server and Proxy AERO interfaces may be configured over one or 1082 more secured tunnel interfaces. The AERO interface configures both 1083 an AERO address and its corresponding SPAN address, while the 1084 underlying secured tunnel interfaces are either unnumbered or 1085 configure the same SPAN address. The AERO interface encapsulates 1086 each IP packet in a SPAN header and presents the packet to the 1087 underlying secured tunnel interface. For Relays that do not 1088 configure an AERO interface, the secured tunnel interfaces themselves 1089 are exposed to the IP layer with each interface configuring the 1090 Relay's SPAN address. Routing protocols such as BGP therefore run 1091 directly over the Relay's secured tunnel interfaces. For nodes that 1092 configure an AERO interface, routing protocols such as BGP run over 1093 the AERO interface but do not employ SPAN encapsulation. Instead, 1094 the AERO interface presents the routing protocol messages directly to 1095 the underlying secured tunnels without applying encapsulation and 1096 while using the SPAN address as the source address. This distinction 1097 must be honored consistently according to each node's configuration 1098 so that the IP forwarding table will associate discovered IP routes 1099 with the correct interface. 1101 3.7. AERO Interface Initialization 1103 AERO Servers, Proxys and Clients configure AERO interfaces as their 1104 point of attachment to the AERO link. AERO nodes assign the MSPs for 1105 the link to their AERO interfaces (i.e., as a "route-to-interface") 1106 to ensure that packets with destination addresses covered by an MNP 1107 not explicitly assigned to a non-AERO interface are directed to the 1108 AERO interface. 1110 AERO interface initialization procedures for Servers, Proxys, Clients 1111 and Relays are discussed in the following sections. 1113 3.7.1. AERO Server/Gateway Behavior 1115 When a Server enables an AERO interface, it assigns AERO/SPAN 1116 addresses and configures permanent neighbor cache entries for 1117 neighbors in the same SPAN segment by consulting the ROS list for the 1118 segment. The Server also configures secured tunnels with one or more 1119 neighboring Relays and engages in a BGP routing protocol session with 1120 each Relay. 1122 The AERO interface provides a single interface abstraction to the IP 1123 layer, but internally comprises multiple secured tunnels as well as 1124 an NBMA nexus for sending encapsulated data packets to AERO interface 1125 neighbors. The Server further configures a service to facilitate ND/ 1126 PD exchanges with AERO Clients and manages per-Client neighbor cache 1127 entries and IP forwarding table entries based on control message 1128 exchanges. 1130 Gateways are simply Servers that run a dynamic routing protocol to 1131 redistribute routes between the AERO interface and INET/EUN 1132 interfaces (see: Section 3.3). The Gateway provisions MNPs to 1133 networks on the INET/EUN interfaces (i.e., the same as a Client would 1134 do) and advertises the MSP(s) for the AERO link over the INET/EUN 1135 interfaces. The Gateway further provides an attachment point of the 1136 AERO link to the non-MNP-based global topology. 1138 3.7.2. AERO Proxy Behavior 1140 When a Proxy enables an AERO interface, it assigns AERO/SPAN 1141 addresses and configures permanent neighbor cache entries the same as 1142 for Servers. The Proxy also configures secured tunnels with one or 1143 more neighboring Relays and maintains per-Client neighbor cache 1144 entries based on control message exchanges. 1146 3.7.3. AERO Client Behavior 1148 When a Client enables an AERO interface, it sends RS messages with 1149 ND/PD parameters over its underlying interfaces to a Server in the 1150 MAP list, which returns an RA message with corresponding parameters. 1151 (The RS/RA messages may pass through a Proxy in the case of a 1152 Client's Proxyed interface.) 1154 3.7.4. AERO Relay Behavior 1156 AERO Relays need not connect directly to the AERO link, since they 1157 operate as link-layer forwarding devices instead of network layer 1158 routers. Configuration of AERO interfaces on Relays is therefore 1159 OPTIONAL, e.g., if an administrative interface is needed. Relays 1160 configure secured tunnels with Servers, Proxys and other Relays; they 1161 also configure AERO/SPAN addresses and permanent neighbor cache 1162 entries the same as Servers. Relays engage in a BGP routing protocol 1163 session with a subset of the Servers on the local SPAN segment, and 1164 with other Relays on the SPAN (see: Section 3.3). 1166 3.8. AERO Interface Neighbor Cache Maintenance 1168 Each AERO interface maintains a conceptual neighbor cache that 1169 includes an entry for each neighbor it communicates with on the AERO 1170 link per [RFC4861]. AERO interface neighbor cache entries are said 1171 to be one of "permanent", "symmetric", "asymmetric" or "proxy". 1173 Permanent neighbor cache entries are created through explicit 1174 administrative action; they have no timeout values and remain in 1175 place until explicitly deleted. AERO Servers and Proxys maintain 1176 permanent neighbor cache entries for all other Servers and Proxys 1177 within the same SPAN segment. Each entry maintains the mapping 1178 between the neighbor's network-layer AERO address and corresponding 1179 INET address. The list of all permanent neighbor cache entries for 1180 the SPAN segment is maintained in the segment's ROS list. 1182 Symmetric neighbor cache entries are created and maintained through 1183 RS/RA exchanges as specified in Section 3.15, and remain in place for 1184 durations bounded by ND/PD lifetimes. AERO Servers maintain 1185 symmetric neighbor cache entries for each of their associated 1186 Clients, and AERO Clients maintain symmetric neighbor cache entries 1187 for each of their associated Servers. The list of all Servers on the 1188 AERO link is maintained in the link's MAP list. 1190 Asymmetric neighbor cache entries are created or updated based on 1191 route optimization messaging as specified in Section 3.17, and are 1192 garbage-collected when keepalive timers expire. AERO ROSs maintain 1193 asymmetric neighbor cache entries for active targets with lifetimes 1194 based on ND messaging constants. Asymmetric neighbor cache entries 1195 are unidirectional since only the ROS (and not the ROR) creates an 1196 entry. 1198 Proxy neighbor cache entries are created and maintained by AERO 1199 Proxys when they process Client/Server ND/PD exchanges, and remain in 1200 place for durations bounded by ND/PD lifetimes. AERO Proxys maintain 1201 proxy neighbor cache entries for each of their associated Clients. 1202 Proxy neighbor cache entries track the Client state and the address 1203 of the Client's associated Server(s). 1205 To the list of neighbor cache entry states in Section 7.3.2 of 1206 [RFC4861], Proxy and Server AERO interfaces add an additional state 1207 DEPARTED that applies to symmetric and proxy neighbor cache entries 1208 for Clients that have recently departed. The interface sets a 1209 "DepartTime" variable for the neighbor cache entry to "DEPART_TIME" 1210 seconds. DepartTime is decremented unless a new ND message causes 1211 the state to return to REACHABLE. While a neighbor cache entry is in 1212 the DEPARTED state, packets destined to the target Client are 1213 forwarded to the Client's new location instead of being dropped. 1214 When DepartTime decrements to 0, the neighbor cache entry is deleted. 1215 It is RECOMMENDED that DEPART_TIME be set to the default constant 1216 value REACHABLE_TIME plus 10 seconds (40 seconds by default) to allow 1217 a window for packets in flight to be delivered while stale route 1218 optimization state may be present. 1220 When an ROR receives an authentic NS message used for route 1221 optimization, it searches for a symmetric neighbor cache entry for 1222 the target Client. The ROR then returns a solicited NA message 1223 without creating a neighbor cache entry for the ROS, but creates or 1224 updates a target Client "Report List" entry for the ROS and sets a 1225 "ReportTime" variable for the entry to REPORT_TIME seconds. The ROR 1226 resets ReportTime when it receives a new authentic NS message, and 1227 otherwise decrements ReportTime while no authentic NS messages have 1228 been received. It is RECOMMENDED that REPORT_TIME be set to the 1229 default constant value REACHABLE_TIME plus 10 seconds (40 seconds by 1230 default) to allow a window for route optimization to converge before 1231 ReportTime decrements below REACHABLE_TIME. 1233 When the ROS receives a solicited NA message response to its NS 1234 message used for route optimization, it creates or updates an 1235 asymmetric neighbor cache entry for the target network-layer and 1236 link-layer addresses. The ROS then (re)sets ReachableTime for the 1237 neighbor cache entry to REACHABLE_TIME seconds and uses this value to 1238 determine whether packets can be forwarded directly to the target, 1239 i.e., instead of via a default route. The ROS otherwise decrements 1240 ReachableTime while no further solicited NA messages arrive. It is 1241 RECOMMENDED that REACHABLE_TIME be set to the default constant value 1242 30 seconds as specified in [RFC4861]. 1244 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1245 of NS keepalives sent when a correspondent may have gone unreachable, 1246 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1247 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1248 to limit the number of unsolicited NAs that can be sent based on a 1249 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1250 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1251 same as specified in [RFC4861]. 1253 Different values for DEPART_TIME, REPORT_TIME, REACHABLE_TIME, 1254 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1255 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1256 different values are chosen, all nodes on the link MUST consistently 1257 configure the same values. Most importantly, DEPART_TIME and 1258 REPORT_TIME SHOULD be set to a value that is sufficiently longer than 1259 REACHABLE_TIME to avoid packet loss due to stale route optimization 1260 state. 1262 3.9. AERO Interface Encapsulation and Re-encapsulation 1264 In some instances, AERO interfaces insert a mid-layer IPv6 header 1265 known as the SPAN header as discussed in the following sections. 1266 After either inserting or omitting the SPAN header, the AERO 1267 interface inserts an outer encapsulation header as discussed below. 1269 AERO interfaces avoid outer encapsulation over Direct underlying 1270 interfaces and Proxyed underlying interfaces for which the first-hop 1271 access router is AERO-aware. Other AERO interfaces encapsulate 1272 packets according to whether they are entering the AERO interface 1273 from the network layer or if they are being re-admitted into the same 1274 AERO link they arrived on. This latter form of encapsulation is 1275 known as "re-encapsulation". 1277 For packets entering the AERO interface from the network layer, the 1278 AERO interface copies the "TTL/Hop Limit", "Type of Service/Traffic 1279 Class" [RFC2983], "Flow Label"[RFC6438] (for IPv6) and "Congestion 1280 Experienced" [RFC3168] values in the inner packet's IP header into 1281 the corresponding fields in the SPAN and outer encapsulation 1282 header(s). 1284 For packets undergoing re-encapsulation, the AERO interface instead 1285 copies these values from the original encapsulation header into the 1286 new encapsulation header, i.e., the values are transferred between 1287 encapsulation headers and *not* copied from the encapsulated packet's 1288 network-layer header. (Note especially that by copying the TTL/Hop 1289 Limit between encapsulation headers the value will eventually 1290 decrement to 0 if there is a (temporary) routing loop.) For IPv4 1291 encapsulation/re-encapsulation, the AERO interface sets the DF bit as 1292 discussed in Section 3.12. 1294 AERO interfaces configured over native and NATed underlying 1295 interfaces encapsulate packets in INET headers according to the next 1296 hop determined in the forwarding algorithm in Section 3.13. If the 1297 next hop is reached via a secured tunnel, the AERO interface uses an 1298 encapsulation format specific to the secured tunnel type (see: 1299 Section 6). If the next hop is reached via an unsecured underlying 1300 interface, the AERO interface instead uses Generic UDP Encapsulation 1301 (GUE) [I-D.ietf-intarea-gue] or an alternate minimal encapsulation 1302 format Appendix A. 1304 When GUE encapsulation is used, the AERO interface next sets the UDP 1305 source port to a constant value that it will use in each successive 1306 packet it sends, and sets the UDP length field to the length of the 1307 SPAN packet plus 8 bytes for the UDP header itself plus the length of 1308 the GUE header (or 0 if GUE direct IP encapsulation is used). For 1309 packets sent to a Server or Relay, the AERO interface sets the UDP 1310 destination port to 8060, i.e., the IANA-registered port number for 1311 AERO. For packets sent to a Client, the AERO interface sets the UDP 1312 destination port to the port value stored in the neighbor cache entry 1313 for this Client. The AERO interface then either includes or omits 1314 the UDP checksum according to the GUE specification. 1316 AERO interfaces observes the packet sizing and fragmentation 1317 considerations found in Section 3.12. 1319 3.10. AERO Interface Decapsulation 1321 AERO interfaces decapsulate packets destined either to the AERO node 1322 itself or to a destination reached via an interface other than the 1323 AERO interface the packet was received on. When the encapsulated 1324 packet arrives in multiple SPAN fragments, the AERO interface 1325 reassembles as discussed in Section 3.12. Further decapsulation 1326 steps are performed according to the appropriate encapsulation format 1327 specification. 1329 3.11. AERO Interface Data Origin Authentication 1331 AERO nodes employ simple data origin authentication procedures. In 1332 particular: 1334 o AERO Relays, Servers and Proxys accept encapsulated data packets 1335 and control messages received from secured tunnels via the SPAN. 1337 o AERO Servers and Proxys accept encapsulated data packets and NS 1338 messages used for Neighbor Unreachability Detection (NUD) received 1339 from a member of the ROS list. 1341 o AERO Proxys and Clients accept packets that originate from within 1342 the same secured ANET. 1344 o AERO Clients and Gateways accept packets from downstream network 1345 correspondents based on ingress filtering. 1347 AERO nodes silently drop any packets that do not satisfy the above 1348 data origin authentication procedures. Further security 1349 considerations are discussed inSection 6. 1351 3.12. AERO Interface MTU and Fragmentation 1353 IPv6 underlying interfaces are REQUIRED to configure a minimum 1354 Maximum Transmission Unit (MTU) of 1280 bytes [RFC8200]. The minimum 1355 MTU for IPv4 underlying interfaces is only 68 bytes [RFC1122], 1356 meaning that a packet smaller than the IPv6 MTU may require 1357 fragmentation when IPv4 encapsulation is used. Therefore, the Don't 1358 Fragment (DF) bit in the IPv4 encapsulation header MUST be set to 0. 1360 The AERO interface configures an MTU of 9180 bytes [RFC2492]; the 1361 size is therefore not a reflection of the underlying interface MTUs, 1362 but rather determines the largest packet the AERO interface can 1363 forward or reassemble. The AERO interface therefore accommodates IP 1364 packets up to 9180 bytes while generating IPv6 Path MTU Discovery 1365 (PMTUD) Packet Too Big (PTB) messages [RFC8201] as necessary (see 1366 below). 1368 AERO interfaces employ mid-layer IPv6 encapsulation and 1369 fragmentation/reassembly per [RFC2473] (aka "SPAN encapsulation") to 1370 accommodate the 9180 byte MTU. The AERO interface returns 1371 internally-generated PTB messages for packets admitted into the 1372 interface that it deems too large (e.g., according to link 1373 performance characteristics, reassembly cost, etc.) while either 1374 dropping or forwarding the packet as necessary. The AERO interface 1375 performs PMTUD even if the destination appears to be on the same link 1376 since intermediate AERO link nodes may return a PTB. This ensures 1377 that the path MTU is adaptive and reflects the current path used for 1378 a given data flow. 1380 AERO nodes perform SPAN encapsulation and fragmentation/reassembly as 1381 follows: 1383 o When a node's AERO interface sends a packet over a Proxyed, VPNed 1384 or Direct underlying interface, it sends without SPAN 1385 encapsulation if the packet is no larger than the underlying 1386 interface MTU. Otherwise, it encapsulates the packet in a SPAN 1387 header with source address set to the node's own SPAN address and 1388 destination set to the SPAN address of the link-layer peer Proxy, 1389 Server or Client on the underlying interface. The AERO interface 1390 then uses IPv6 fragmentation to break the encapsulated packet into 1391 a minimum number of non-overlapping fragments, where the largest 1392 fragment size is determined by the underlying interface MTU and 1393 the smallest fragment is no smaller than 640 bytes. The AERO 1394 interface then sends the fragments to the link-layer peer, which 1395 reassembles before forwarding toward the final destination. 1397 o When a node's AERO interface sends a packet over a Native or 1398 NATted underlying interface, it sends packets no larger than 1280 1399 bytes without SPAN encapsulation if the destination is reached via 1400 an INET address within the same SPAN segment. Otherwise, it 1401 encapsulates the packet in a SPAN header with source address set 1402 to the node's SPAN address and destination set to the SPAN address 1403 of the next hop AERO node toward the final destination. The AERO 1404 interface then uses IPv6 fragmentation to break the encapsulated 1405 packet into a minimum number of non-overlapping fragments, where 1406 the largest fragment size is 1280 bytes and the smallest fragment 1407 is no smaller than 640 bytes. The AERO interface then sends the 1408 fragments to the SPAN destination, which reassembles before 1409 forwarding toward the final destination. 1411 In order to avoid a "tiny fragment" attack, AERO interfaces 1412 unconditionally drop all SPAN fragments smaller than 640 bytes. In 1413 order to set the correct context for reassembly, the AERO interface 1414 that inserts a SPAN header MUST also be the one that inserts the IPv6 1415 Fragment Header Identification value. Although all fragmnets of the 1416 same fragmented SPAN packet are typically sent via the same 1417 underlying interface, this is not strictly required since all 1418 fragments will arrive at the AERO interface that performs reassembly 1419 even if they travel over different paths. 1421 Note that the AERO interface can forward large packets via 1422 encapsulation and fragmentation while at the same time returning 1423 advisory PTB messages, e.g., subject to rate limiting. The receiving 1424 node that performs reassembly can also send advisory PTB messages if 1425 reassembly conditions become unfavorable. The AERO interface can 1426 therefore continuously forward large packets without loss while 1427 returning advisory messages recommending a smaller size. Advisory 1428 PTB messages are differentiated from PTB messages that report loss by 1429 setting the Code field in the ICMPv6 message header to the value 1. 1430 This document therefore updates [RFC4443] and [RFC8201]. 1432 3.13. AERO Interface Forwarding Algorithm 1434 IP packets enter a node's AERO interface either from the network 1435 layer (i.e., from a local application or the IP forwarding system) or 1436 from the link layer (i.e., from an AERO interface neighbor). All 1437 packets entering a node's AERO interface first undergo data origin 1438 authentication as discussed in Section 3.11. Packets that satisfy 1439 data origin authentication are processed further, while all others 1440 are dropped silently. 1442 Packets that enter the AERO interface from the network layer are 1443 forwarded to an AERO interface neighbor. Packets that enter the AERO 1444 interface from the link layer are either re-admitted into the AERO 1445 link or forwarded to the network layer where they are subject to 1446 either local delivery or IP forwarding. In all cases, the AERO 1447 interface itself MUST NOT decrement the network layer TTL/Hop-count 1448 since its forwarding actions occur below the network layer. 1450 AERO interfaces may have multiple underlying interfaces and/or 1451 neighbor cache entries for neighbors with multiple ifIndex-tuple 1452 registrations (see Section 3.6). The AERO interface uses traffic 1453 classifiers (e.g., DSCP value, port number, etc.) to select an 1454 outgoing underlying interface for each packet based on the node's own 1455 QoS preferences, and also to select a destination link-layer address 1456 based on the neighbor's underlying interface with the highest 1457 preference. AERO implementations SHOULD allow for QoS preference 1458 values to be modified at runtime through network management. 1460 If multiple outgoing interfaces and/or neighbor interfaces have a 1461 preference of "high", the AERO node replicates the packet and sends 1462 one copy via each of the (outgoing / neighbor) interface pairs; 1463 otherwise, the node sends a single copy of the packet via an 1464 interface with the highest preference. AERO nodes keep track of 1465 which underlying interfaces are currently "reachable" or 1466 "unreachable", and only use "reachable" interfaces for forwarding 1467 purposes. 1469 The following sections discuss the AERO interface forwarding 1470 algorithms for Clients, Proxys, Servers and Relays. In the following 1471 discussion, a packet's destination address is said to "match" if it 1472 is the same as a cached address, or if it is covered by a cached 1473 prefix (which may be encoded in an AERO address). 1475 3.13.1. Client Forwarding Algorithm 1477 When an IP packet enters a Client's AERO interface from the network 1478 layer the Client searches for an asymmetric neighbor cache entry that 1479 matches the destination. If there is a match, the Client uses one or 1480 more "reachable" neighbor interfaces in the entry for packet 1481 forwarding. If there is no asymmetric neighbor cache entry, the 1482 Client instead forwards the packet toward a Server (the packet is 1483 intercepted by a Proxy if there is a Proxy on the path). The Client 1484 encapuslates the packet in an IPv6 header and fragments if necessary 1485 according to MTU requirements (see: Section 3.12). 1487 When an IP packet enters a Client's AERO interface from the link- 1488 layer, if the destination matches one of the Client's MNPs or link- 1489 local addresses the Client reassembles and decapsulates as necessary 1490 and delivers the inner packet to the network layer. Otherwise, the 1491 Client drops the packet and MAY return a network-layer ICMP 1492 Destination Unreachable message subject to rate limiting (see: 1493 Section 3.14). 1495 3.13.2. Proxy Forwarding Algorithm 1497 For control messages originating from or destined to a Client, the 1498 Proxy intercepts the message and updates its proxy neighbor cache 1499 entry for the Client. The Proxy then forwards a (proxyed) copy of 1500 the control message. (For example, the Proxy forwards a proxied 1501 version of a Client's NS/RS message to the target neighbor, and 1502 forwards a proxied version of the NA/RA reply to the Client.) 1504 When the Proxy receives a data packet from a Client within the ANET, 1505 ithe Proxy reassembles and re-fragments if necessary then searches 1506 for an asymmetric neighbor cache entry that matches the destination 1507 and forwards as follows: 1509 o if the destination matches an asymmetric neighbor cache entry, the 1510 Proxy uses one or more "reachable" neighbor interfaces in the 1511 entry for packet forwarding using SPAN encapsualtion and Segment 1512 Routing if necessary with the final destination set to the 1513 neighbor's SPAN address. If the neighbor interface is in the same 1514 SPAN segment, the Proxy forwards the packet directly to the 1515 neighbor; otherwise, it forwards the packet to a Relay. 1517 o else, the Proxy encapsulates and forwards the packet to a Relay 1518 while using the SPAN address corresponding to the packet's 1519 destination as the SPAN destination address. 1521 When the Proxy receives an encapsulated data packet from an INET 1522 neighbor or from a secured tunnel from a Relay, it accepts the packet 1523 only if data origin authentication succeeds and if there is a proxy 1524 neighbor cache entry that matches the inner destination. Next, the 1525 Proxy reassembles the packet (if necessary) and continues processing. 1527 Next if reassembly is complete and the neighbor cache state is 1528 REACHABLE, the Proxy either drops and returns a PTB (see: 1529 Section 3.12) or forwards the packet to the Client while performing 1530 SPAN encapsulation and re-fragmentation to the ANET MTU size if 1531 necessary. If the neighbor cache entry state is DEPARTED, the Proxy 1532 instead changes the SPAN destination address to the address of the 1533 new Server and forwards it to a Relay while performing re- 1534 fragmentation to 1280 bytes if necessary. 1536 3.13.3. Server/Gateway Forwarding Algorithm 1538 For control messages destined to a target Client's AERO address that 1539 are received from a secured tunnel, the Server intercepts the message 1540 and sends an appropriate response on behalf of the Client. (For 1541 example, the Server sends an NA message reply in response to an NS 1542 message directed to one of its associated Clients.) If the Client's 1543 neighbor cache entry is in the DEPARTED state, however, the Server 1544 instead forwards the packet to the Client's new Server as discussed 1545 in Section 3.19. 1547 When the Server receives an encapsulated data packet from an INET 1548 neighbor or from a secured tunnel, it accepts the packet only if data 1549 origin authentication succeeds. If the SPAN destination address is 1550 its own address, the Server continues processing as follows: 1552 o if the destination matches a symmetric neighbor cache entry in the 1553 REACHABLE state the Server prepares the packet for forwarding to 1554 the destination Client. The Server first reassembles (if 1555 necessary) and forwards the packet (while re-fragmenting if 1556 necessary) as specified inSection 3.12. 1558 o else, if the destination matches a symmetric neighbor cache entry 1559 in the DEPARETED state the Server re-encapsulates the packet and 1560 forwards it using the SPAN address of the Client's new Server as 1561 the destination. 1563 o else, if the destination matches an asymmetric neighbor cache 1564 entry, the Server uses one or more "reachable" neighbor interfaces 1565 in the entry for packet forwarding via the local INET if the 1566 neighbor is in the same SPAN segment or using SPAN encapsualtion 1567 and Segment Routing if necessary with the final destination set to 1568 the neighbor's SPAN address otherwise. 1570 o else, if the destination is an AERO address that is not assigned 1571 on the AERO interface the Server drops the packet. 1573 o else, the Server (acting as a Gateway) reassembles if necessary, 1574 decapsulates the packet and releases it to the network layer for 1575 local delivery or IP forwarding. Based on the information in the 1576 forwarding table, the network layer may return the packet to the 1577 same AERO interface in which case further processing occurs as 1578 below. (Note that this arrangement accommodates common 1579 implementations in which the IP forwarding table is not accessible 1580 from within the AERO interface. If the AERO interface can 1581 directly access the IP forwarding table (such as for in-kernel 1582 implementations) the forwarding table lookup can instead be 1583 performed internally from within the AERO interface itself.) 1585 When the Server's AERO interface receives a data packet from the 1586 network layer or from a VPNed or Direct Client, it performs SPAN 1587 encapsualtion and fragmentation if necessary, then processes the 1588 packet according to the network-layer destination address as follows: 1590 o if the destination matches a symmetric or asymmetric neighbor 1591 cache entry the Server processes the packet as above. 1593 o else, the Server encapsulates the packet and forwards it to a 1594 Relay using its own SPAN address as the source and the SPAN 1595 address corresponding to the destination as the destination. 1597 3.13.4. Relay Forwarding Algorithm 1599 Relays forward packets over secured tunnels the same as any IP 1600 router. When the Relay receives an encapsulated packet via a secured 1601 tunnel, it removes the INET header and searches for a forwarding 1602 table entry that matches the destination address in the next header. 1603 The Relay then processes the packet as follows: 1605 o if the destination matches one of the Relay's own addresses, the 1606 Relay submits the packet for local delivery. 1608 o else, if the destination matches a forwarding table entry the 1609 Relay forwards the packet via a secured tunnel to the next hop. 1610 If the destination matches an MSP without matching an MNP, 1611 however, the Relay instead drops the packet and returns an ICMP 1612 Destination Unreachable message subject to rate limiting (see: 1613 Section 3.14). 1615 o else, the Relay drops the packet and returns an ICMP Destination 1616 Unreachable as above. 1618 As for any IP router, the Relay decrements the TTL/Hop Limit when it 1619 forwards the packet. Therefore, only the Hop Limit in the SPAN 1620 header is decremented, and not the TTL/Hop Limit in the inner packet 1621 header. 1623 3.14. AERO Interface Error Handling 1625 When an AERO node admits a packet into the AERO interface, it may 1626 receive link-layer or network-layer error indications. 1628 A link-layer error indication is an ICMP error message generated by a 1629 router in the INET on the path to the neighbor or by the neighbor 1630 itself. The message includes an IP header with the address of the 1631 node that generated the error as the source address and with the 1632 link-layer address of the AERO node as the destination address. 1634 The IP header is followed by an ICMP header that includes an error 1635 Type, Code and Checksum. Valid type values include "Destination 1636 Unreachable", "Time Exceeded" and "Parameter Problem" 1637 [RFC0792][RFC4443]. (AERO interfaces ignore all link-layer IPv4 1638 "Fragmentation Needed" and IPv6 "Packet Too Big" messages since they 1639 only emit packets that are guaranteed to be no larger than the IP 1640 minimum link MTU as discussed in Section 3.12.) 1642 The ICMP header is followed by the leading portion of the packet that 1643 generated the error, also known as the "packet-in-error". For 1644 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1645 much of invoking packet as possible without the ICMPv6 packet 1646 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1647 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1648 "Internet Header + 64 bits of Original Data Datagram", however 1649 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1650 ICMP datagram SHOULD contain as much of the original datagram as 1651 possible without the length of the ICMP datagram exceeding 576 1652 bytes". 1654 The link-layer error message format is shown in Figure 5 (where, "L2" 1655 and "L3" refer to link-layer and network-layer, respectively): 1657 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1658 ~ ~ 1659 | L2 IP Header of | 1660 | error message | 1661 ~ ~ 1662 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1663 | L2 ICMP Header | 1664 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1665 ~ ~ P 1666 | IP and other encapsulation | a 1667 | headers of original L3 packet | c 1668 ~ ~ k 1669 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1670 ~ ~ t 1671 | IP header of | 1672 | original L3 packet | i 1673 ~ ~ n 1674 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1675 ~ ~ e 1676 | Upper layer headers and | r 1677 | leading portion of body | r 1678 | of the original L3 packet | o 1679 ~ ~ r 1680 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1682 Figure 5: AERO Interface Link-Layer Error Message Format 1684 The AERO node rules for processing these link-layer error messages 1685 are as follows: 1687 o When an AERO node receives a link-layer Parameter Problem message, 1688 it processes the message the same as described as for ordinary 1689 ICMP errors in the normative references [RFC0792][RFC4443]. 1691 o When an AERO node receives persistent link-layer Time Exceeded 1692 messages, the IP ID field may be wrapping before earlier fragments 1693 awaiting reassembly have been processed. In that case, the node 1694 should begin including integrity checks and/or institute rate 1695 limits for subsequent packets. 1697 o When an AERO node receives persistent link-layer Destination 1698 Unreachable messages in response to encapsulated packets that it 1699 sends to one of its asymmetric neighbor correspondents, the node 1700 should process the message as an indication that a path may be 1701 failing, and optionally initiate NUD over that path. If it 1702 receives Destination Unreachable messages over multiple paths, the 1703 node should allow future packets destined to the correspondent to 1704 flow through a default route and re-initiate route optimization. 1706 o When an AERO Client receives persistent link-layer Destination 1707 Unreachable messages in response to encapsulated packets that it 1708 sends to one of its symmetric neighbor Servers, the Client should 1709 mark the path as unusable and use another path. If it receives 1710 Destination Unreachable messages on many or all paths, the Client 1711 should associate with a new Server and release its association 1712 with the old Server as specified in Section 3.19.5. 1714 o When an AERO Server receives persistent link-layer Destination 1715 Unreachable messages in response to encapsulated packets that it 1716 sends to one of its symmetric neighbor Clients, the Server should 1717 mark the underlying path as unusable and use another underlying 1718 path. 1720 o When an AERO Server or Proxy receives link-layer Destination 1721 Unreachable messages in response to an encapsulated packet that it 1722 sends to one of its permanent neighbors, it treats the messages as 1723 an indication that the path to the neighbor may be failing. 1724 However, the dynamic routing protocol should soon reconverge and 1725 correct the temporary outage. 1727 When an AERO Relay receives a packet for which the network-layer 1728 destination address is covered by an MSP, if there is no more- 1729 specific routing information for the destination the Relay drops the 1730 packet and returns a network-layer Destination Unreachable message 1731 subject to rate limiting. The Relay writes the network-layer source 1732 address of the original packet as the destination address and uses 1733 one of its non link-local addresses as the source address of the 1734 message. 1736 When an AERO node receives an encapsulated packet for which the 1737 reassembly buffer it too small, it drops the packet and returns a 1738 network-layer Packet Too Big (PTB) message. The node first writes 1739 the MRU value into the PTB message MTU field, writes the network- 1740 layer source address of the original packet as the destination 1741 address and writes one of its non link-local addresses as the source 1742 address. 1744 3.15. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1746 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1747 coordinated as discussed in the following Sections. 1749 3.15.1. AERO ND/PD Service Model 1751 Each AERO Server on the link configures a PD service to facilitate 1752 Client requests. Each Server is provisioned with a database of MNP- 1753 to-Client ID mappings for all Clients enrolled in the AERO service, 1754 as well as any information necessary to authenticate each Client. 1755 The Client database is maintained by a central administrative 1756 authority for the AERO link and securely distributed to all Servers, 1757 e.g., via the Lightweight Directory Access Protocol (LDAP) [RFC4511], 1758 via static configuration, etc. Clients receive the same service 1759 regardless of the Servers they select. 1761 AERO Clients and Servers use ND messages to maintain neighbor cache 1762 entries. AERO Servers configure their AERO interfaces as advertising 1763 NBMA interfaces, and therefore send unicast RA messages with a short 1764 Router Lifetime value (e.g., REACHABLE_TIME seconds) in response to a 1765 Client's RS message. Thereafter, Clients send additional RS messages 1766 to keep Server state alive. 1768 AERO Clients and Servers include PD parameters in RS/RA messages (see 1769 [I-D.templin-6man-dhcpv6-ndopt] for ND/PD alternatives). The unified 1770 ND/PD messages are exchanged between Client and Server according to 1771 the prefix management schedule required by the PD service. If the 1772 Client knows its MNP in advance, it can instead employ prefix 1773 registration by including its AERO address as the source address of 1774 an RS message and with an OMNI option with valid prefix registration 1775 information for the MNP. If the Server (and Proxy) accept the 1776 Client's MNP assertion, they inject the prefix into the routing 1777 system and establish the necessary neighbor cache state. 1779 The following sections specify the Client and Server behavior. 1781 3.15.2. AERO Client Behavior 1783 AERO Clients discover the addresses of Servers in a similar manner as 1784 described in [RFC5214]. Discovery methods include static 1785 configuration (e.g., from a flat-file map of Server addresses and 1786 locations), or through an automated means such as Domain Name System 1787 (DNS) name resolution [RFC1035]. Alternatively, the Client can 1788 discover Server addresses through a layer 2 data link login exchange, 1789 or through a unicast RA response to a multicast/anycast RS as 1790 described below. In the absence of other information, the Client can 1791 resolve the DNS Fully-Qualified Domain Name (FQDN) 1792 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 1793 text string and "[domainname]" is a DNS suffix for the AERO link 1794 (e.g., "example.com"). 1796 To associate with a Server, the Client acts as a requesting router to 1797 request MNPs. The Client prepares an RS message with PD parameters 1798 and includes a Nonce and Timestamp option if the Client needs to 1799 correlate RA replies. If the Client already knows the Server's AERO 1800 address, it includes the AERO address as the network-layer 1801 destination address; otherwise, it includes the link-scoped All- 1802 Routers multicast (ff02::2) or Subnet-Router anycast (fe80::) address 1803 as the network-layer destination. If the Client already knows its 1804 own AERO address, it uses the AERO address as the network-layer 1805 source address; otherwise, it uses the unspecified IPv6 address 1806 (::/128) as the network-layer source address. 1808 The Client next includes an OMNI option in the RS message to register 1809 its link-layer information with the Server. The Client sets the OMNI 1810 option prefix registration information according to the MNP, and 1811 includes an ifIndex-tuple with S set to '1' corresponding to the 1812 underlying interface over which the Client will send the RS message. 1813 The Client MAY include additional ifIndex-tuples specific to other 1814 underlying interfaces. The Client MAY also include an SLLAO 1815 corresponding to the OMNI option ifIndex-tuple with S set to '1'. 1817 The Client then sends the RS message (either directly via Direct 1818 interfaces, via INET encapsulation for NATed interfaces, via a VPN 1819 for VPNed interfaces, via a Proxy for proxyed interfaces or via a 1820 Relay for native interfaces) and waits for an RA message reply (see 1821 Section 3.15.3). The Client retries up to MAX_RTR_SOLICITATIONS 1822 times until an RA is received. If the Client receives no RAs, or if 1823 it receives an RA with Router Lifetime set to 0, the Client SHOULD 1824 abandon this Server and try another Server. Otherwise, the Client 1825 processes the PD information found in the RA message. 1827 Next, the Client creates a symmetric neighbor cache entry with the 1828 Server's AERO address as the network-layer address and the Server's 1829 encapsulation and/or link-layer addresses as the link-layer address. 1830 The Client records the RA Router Lifetime field value in the neighbor 1831 cache entry as the time for which the Server has committed to 1832 maintaining the MNP in the routing system via this underlying 1833 interface, and caches the other RA configuration information 1834 including Cur Hop Limit, M and O flags, Reachable Time and Retrans 1835 Timer. The Client then autoconfigures AERO addresses for each of the 1836 delegated MNPs and assigns them to the AERO interface. The Client 1837 also caches any MSPs included in Route Information Options (RIOs) 1838 [RFC4191] as MSPs to associate with the AERO link, and assigns the 1839 MTU value in the MTU option to the underlying interface. 1841 The Client then registers additional underlying interfaces with the 1842 Server by sending RS messages via each additional interface. The RS 1843 messages include the same parameters as for the initial RS/RA 1844 exchange, but with destination address set to the Server's AERO 1845 address. 1847 Following autoconfiguration, the Client sub-delegates the MNPs to its 1848 attached EUNs and/or the Client's own internal virtual interfaces as 1849 described in [I-D.templin-v6ops-pdhost] to support the Client's 1850 downstream attached "Internet of Things (IoT)". The Client 1851 subsequently sends additional RS messages over each underlying 1852 interface before the Router Lifetime received for that interface 1853 expires. 1855 After the Client registers its underlying interfaces, it may wish to 1856 change one or more registrations, e.g., if an interface changes 1857 address or becomes unavailable, if QoS preferences change, etc. To 1858 do so, the Client prepares an RS message to send over any available 1859 underlying interface. The RS includes an OMNI option with prefix 1860 registration information specific to its MNP, with an ifIndex-tuple 1861 specific to the selected underlying interface with S set to '1', and 1862 with any additional ifIndex-tuples specific to other underlying 1863 interfaces. The Client includes fresh ifIndex-tuple values to update 1864 the Server's neighbor cache entry. When the Client receives the 1865 Server's RA response, it has assurance that the Server has been 1866 updated with the new information. 1868 If the Client wishes to discontinue use of a Server it issues an RS 1869 message over any underlying interface with an OMNI option with a 1870 prefix release indication. When the Server processes the message, it 1871 releases the MNP, sets the symmetric neighbor cache entry state for 1872 the Client to DEPARTED and returns an RA reply with Router Lifetime 1873 set to 0. After a short delay (e.g., 2 seconds), the Server 1874 withdraws the MNP from the routing system. 1876 3.15.3. AERO Server Behavior 1878 AERO Servers act as IP routers and support a PD service for Clients. 1879 Servers arrange to add their AERO addresses to a static map of Server 1880 addresses for the link and/or the DNS resource records for the FQDN 1881 "linkupnetworks.[domainname]" before entering service. Server 1882 addresses should be geographically and/or topologically referenced, 1883 and made available for discovery by Clients on the AERO link. 1885 When a Server receives a prospective Client's RS message on its AERO 1886 interface, it SHOULD return an immediate RA reply with Router 1887 Lifetime set to 0 if it is currently too busy or otherwise unable to 1888 service the Client. Otherwise, the Server authenticates the RS 1889 message and processes the PD parameters. The Server first determines 1890 the correct MNPs to delegate to the Client by searching the Client 1891 database. When the Server delegates the MNPs, it also creates a 1892 forwarding table entry for each MNP so that the MNPs are propagated 1893 into the routing system (see: Section 3.3). For IPv6, the Server 1894 creates an IPv6 forwarding table entry for each MNP. For IPv4, the 1895 Server creates an IPv6 forwarding table entry with the SPAN 1896 Compatibility Prefix (SCP) corresponding to the IPv4 address. 1898 The Server next creates a symmetric neighbor cache entry for the 1899 Client using the base AERO address as the network-layer address and 1900 with lifetime set to no more than the smallest PD lifetime. Next, 1901 the Server updates the neighbor cache entry by recording the 1902 information in each ifIndex-tuple in the RS OMNI option. The Server 1903 also records the actual SPAN/INET addresses in the neighbor cache 1904 entry. 1906 Next, the Server prepares an RA message using its AERO address as the 1907 network-layer source address and the network-layer source address of 1908 the RS message as the network-layer destination address. The Server 1909 sets the Router Lifetime to the time for which it will maintain both 1910 this underlying interface individually and the symmetric neighbor 1911 cache entry as a whole. The Server also sets Cur Hop Limit, M and O 1912 flags, Reachable Time and Retrans Timer to values appropriate for the 1913 AERO link. The Server includes the delegated MNPs, any other PD 1914 parameters and an OMNI option with no ifIndex-tuples. The Server 1915 then includes one or more RIOs that encode the MSPs for the AERO 1916 link, plus an MTU option (see Section 3.12). The Server finally 1917 forwards the message to the Client using SPAN/INET, INET, or NULL 1918 encapsulation as necessary. 1920 After the initial RS/RA exchange, the Server maintains a 1921 ReachableTime timer for each of the Client's underlying interfaces 1922 individually (and for the Client's symmetric neighbor cache entry 1923 collectively) set to expire after Router Lifetime seconds. If the 1924 Client (or Proxy) issues additional RS messages, the Server sends an 1925 RA response and resets ReachableTime. If the Server receives an ND 1926 message with PD release indication it sets the Client's symmetric 1927 neighbor cache entry to the DEPARTED state and withdraws the MNP from 1928 the routing system after a short delay (e.g., 2 seconds). If 1929 ReachableTime expires before a new RS is received on an individual 1930 underlying interface, the Server marks the interface as DOWN. If 1931 ReachableTime expires before any new RS is received on any individual 1932 underlying interface, the Server deletes the neighbor cache entry and 1933 withdraws the MNP without delay. 1935 The Server processes any ND/PD messages pertaining to the Client and 1936 returns an NA/RA reply in response to solicitations. The Server may 1937 also issue unsolicited RA messages, e.g., with PD reconfigure 1938 parameters to cause the Client to renegotiate its PDs, with Router 1939 Lifetime set to 0 if it can no longer service this Client, etc. 1940 Finally, If the symmetric neighbor cache entry is in the DEPARTED 1941 state, the Server deletes the entry after DepartTime expires. 1943 Note: Clients SHOULD notify former Servers of their departures, but 1944 Servers are responsible for expiring neighbor cache entries and 1945 withdrawing routes even if no departure notification is received 1946 (e.g., if the Client leaves the network unexpectedly). Servers 1947 SHOULD therefore set Router Lifetime to REACHABLE_TIME seconds in 1948 solicited RA messages to minimize persistent stale cache information 1949 in the absence of Client departure notifications. A short Router 1950 Lifetime also ensures that proactive Client/Server RS/RA messaging 1951 will keep any NAT state alive (see above). 1953 Note: All Servers on an AERO link MUST advertise consistent values in 1954 the RA Cur Hop Limit, M and O flags, Reachable Time and Retrans Timer 1955 fields the same as for any link, since unpredictable behavior could 1956 result if different Servers on the same link advertised different 1957 values. 1959 3.15.3.1. Lightweight DHCPv6 Relay Agent (LDRA) 1961 When DHCPv6 is used as the ND/PD service back end, AERO Clients and 1962 Servers are always on the same link (i.e., the AERO link) from the 1963 perspective of DHCPv6. However, in some implementations the DHCPv6 1964 server and ND function may be located in separate modules. In that 1965 case, the Server's AERO interface module can act as a Lightweight 1966 DHCPv6 Relay Agent (LDRA)[RFC6221] to relay PD messages to and from 1967 the DHCPv6 server module. 1969 When the LDRA receives an authentic RS message, it extracts the PD 1970 message parameters and uses them to construct an IPv6/UDP/DHCPv6 1971 message. It sets the IPv6 source address to the source address of 1972 the RS message, sets the IPv6 destination address to 1973 'All_DHCP_Relay_Agents_and_Servers' and sets the UDP fields to values 1974 that will be understood by the DHCPv6 server. 1976 The LDRA then wraps the message in a DHCPv6 'Relay-Forward' message 1977 header and includes an 'Interface-Id' option that includes enough 1978 information to allow the LDRA to forward the resulting Reply message 1979 back to the Client (e.g., the Client's link-layer addresses, a 1980 security association identifier, etc.). The LDRA also wraps the OMNI 1981 option and SLLAO into the Interface-Id option, then forwards the 1982 message to the DHCPv6 server. 1984 When the DHCPv6 server prepares a Reply message, it wraps the message 1985 in a 'Relay-Reply' message and echoes the Interface-Id option. The 1986 DHCPv6 server then delivers the Relay-Reply message to the LDRA, 1987 which discards the Relay-Reply wrapper and IPv6/UDP headers, then 1988 uses the DHCPv6 message to construct an RA response to the Client. 1989 The Server uses the information in the Interface-Id option to prepare 1990 the RA message and to cache the link-layer addresses taken from the 1991 OMNI option and SLLAO echoed in the Interface-Id option. 1993 3.16. The AERO Proxy 1995 Clients may connect to ANETs that deploy perimeter security services 1996 to facilitate communications to Servers in outside INETs. In that 1997 case, the ANET can employ an AERO Proxy. The Proxy is located at the 1998 ANET/INET border and listens for RS messages originating from or RA 1999 messages destined to ANET Clients. The Proxy acts on these control 2000 messages as follows: 2002 o when the Proxy receives an RS message from a new ANET Client, it 2003 first authenticates the message then examines the network-layer 2004 destination address. If the destination address is a Server's 2005 AERO address, the Proxy proceeds to the next step. Otherwise, if 2006 the destination is All-Routers multicast or Subnet-Router anycast, 2007 the Proxy selects a "nearby" Server that is likely to be a good 2008 candidate to serve the Client and replaces the destination address 2009 with the Server's AERO address. Next, the Proxy creates a proxy 2010 neighbor cache entry and caches the Client and Server link-layer 2011 addresses along with the OMNI option information and any other 2012 identifying information including Transaction IDs, Client 2013 Identifiers, Nonce values, etc. The Proxy finally encapsulates 2014 the (proxyed) RS message in a SPAN header with source set to the 2015 Proxy's SPAN address and destination set to the Server's SPAN 2016 address then forwards the message into the SPAN. 2018 o when the Server receives the RS, it authenticates the message then 2019 creates or updates a symmetric neighbor cache entry for the Client 2020 with the Proxy's SPAN address as the link-layer address. The 2021 Server then sends an RA message back to the Proxy via the SPAN. 2023 o when the Proxy receives the RA, it authenticates the message and 2024 matches it with the proxy neighbor cache entry created by the RS. 2025 The Proxy then caches the PD route information as a mapping from 2026 the Client's MNPs to the Client's ANET address, caches the 2027 Server's advertised Router Lifetime and sets the neighbor cache 2028 entry state to REACHABLE. The Proxy then sets the P bit in the RA 2029 flags field, optionally rewrites the Router Lifetime and forwards 2030 the (proxyed) message to the Client. The Proxy finally includes 2031 an MTU option (if necessary) with an MTU to use for the underlying 2032 ANET interface. 2034 After the initial RS/RA exchange, the Proxy forwards any Client data 2035 packets for which there is no matching asymmetric neighbor cache 2036 entry to a Relay using SPAN encapsulation with its own SPAN address 2037 as the source and the SPAN address corresponding to the Client as the 2038 destination. The Proxy instead forwards any Client data destined to 2039 an asymmetric neighbor cache target directly to the target according 2040 to the SPAN/link-layer information - the process of establishing 2041 asymmetric neighbor cache entries is specified in Section 3.17. 2043 While the Client is still attached to the ANET, the Proxy sends NS, 2044 RS and/or unsolicited NA messages to update the Server's symmetric 2045 neighbor cache entries on behalf of the Client and/or to convey QoS 2046 updates. This allows for higher-frequency Proxy-initiated RS/RA 2047 messaging over well-connected INET infrastructure supplemented by 2048 lower-frequency Client-initiated RS/RA messaging over constrained 2049 ANET data links. 2051 If the Server ceases to send solicited advertisements, the Proxy 2052 sends unsolicited RAs on the ANET interface with destination set to 2053 All-Nodes multicast (ff02::1) and with Router Lifetime set to zero to 2054 inform Clients that the Server has failed. Although the Proxy 2055 engages in ND exchanges on behalf of the Client, the Client can also 2056 send ND messages on its own behalf, e.g., if it is in a better 2057 position than the Proxy to convey QoS changes, etc. For this reason, 2058 the Proxy marks any Client-originated solicitation messages (e.g. by 2059 inserting a Nonce option) so that it can return the solicited 2060 advertisement to the Client instead of processsing it locally. 2062 If the Client becomes unreachable, the Proxy sets the neighbor cache 2063 entry state to DEPARTED and retains the entry for DEPART_TIME 2064 seconds. While the state is DEPARTED, the Proxy forwards any packets 2065 destined to the Client to a Relay via SPAN encapsulation with the 2066 Client's current Server as the destination. The Relay in turn 2067 forwards the packets to the Client's current Server. When DepartTime 2068 expires, the Proxy deletes the neighbor cache entry and discards any 2069 further packets destined to this (now forgotten) Client. 2071 In some ANETs that employ a Proxy, the Client's MNP can be injected 2072 into the ANET routing system. In that case, the Client can send data 2073 messages without encapsulation so that the ANET native routing system 2074 transports the unencapsulated packets to the Proxy. This can be very 2075 beneficial, e.g., if the Client connects to the ANET via low-end data 2076 links such as some aviation wireless links. 2078 If the first-hop ANET access router is AERO-aware, the Client can 2079 avoid encapsulation for both its control and data messages. When the 2080 Client connects to the link, it can send an unencapsulated RS message 2081 with source address set to its AERO address and with destination 2082 address set to the AERO address of the Client's selected Server or to 2083 All-Routers multicast or Subnet-Router anycast. The Client includes 2084 an OMNI option formatted as specified in 2085 [I-D.templin-6man-omni-interface]. 2087 The Client then sends the unencapsulated RS message, which will be 2088 intercepted by the AERO-Aware access router. The access router then 2089 encapsulates the RS message in an ANET header with its own address as 2090 the source address and the address of a Proxy as the destination 2091 address. The access router further remembers the address of the 2092 Proxy so that it can encapsulate future data packets from the Client 2093 via the same Proxy. If the access router needs to change to a new 2094 Proxy, it simply sends another RS message toward the Server via the 2095 new Proxy on behalf of the Client. 2097 In some cases, the access router and Proxy may be one and the same 2098 node. In that case, the node would be located on the same physical 2099 link as the Client, but its message exchanges with the Server would 2100 need to pass through a security gateway at the ANET/INET border. The 2101 method for deploying access routers and Proxys (i.e. as a single node 2102 or multiple nodes) is an ANET-local administrative consideration. 2104 3.16.1. Detecting and Responding to Server Failures 2106 In environments where fast recovery from Server failure is required, 2107 Proxys SHOULD use proactive Neighbor Unreachability Detection (NUD) 2108 to track Server reachability in a similar fashion as for 2109 Bidirectional Forwarding Detection (BFD) [RFC5880]. Proxys can then 2110 quickly detect and react to failures so that cached information is 2111 re-established through alternate paths. The NUD control messaging is 2112 carried only over well-connected ground domain networks (i.e., and 2113 not low-end aeronautical radio links) and can therefore be tuned for 2114 rapid response. 2116 Proxys perform proactive NUD with Servers for which there are 2117 currently active ANET Clients by sending continuous NS messages in 2118 rapid succession, e.g., one message per second. The Proxy sends the 2119 NS message via the SPAN with the Proxy's AERO address as the source 2120 and the AERO address of the Server as the destination. When the 2121 Proxy is also sending RS messages to the Server on behalf of ANET 2122 Clients, the resulting RA responses can be considered as equivalent 2123 hints of forward progress. This means that the Proxy need not also 2124 send a periodic NS if it has already sent an RS within the same 2125 period. If the Server fails (i.e., if the Proxy ceases to receive 2126 advertisements), the Proxy can quickly inform Clients by sending 2127 multicast RA messages on the ANET interface. 2129 The Proxy sends RA messages on the ANET interface with source address 2130 set to the Server's address, destination address set to All-Nodes 2131 multicast, and Router Lifetime set to 0. The Proxy SHOULD send 2132 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2133 [RFC4861]. Any Clients on the ANET that had been using the failed 2134 Server will receive the RA messages and associate with a new Server. 2136 3.16.2. Point-to-Multipoint Server Coordindation 2138 In environments where Client messaging over ANETs is bandwidth- 2139 limited and/or expensive, Clients can enlist the services of the 2140 Proxy to coordinate with multiple Servers in a single RS/RA message 2141 exchange. The Client can send a single RS message to All-Routers 2142 multicast that includes the ID's of multiple Servers in MS-Register 2143 sub-options of the OMNI option,. 2145 When the Proxy receives the RS and processes the OMNI option, it 2146 performs a separate RS/RA exchange with each MS-Register Server. 2147 When it has received the RA messages, it creates an "aggregate" RA 2148 message to return to the Client with an OMNI option with each 2149 responding Server's ID recorded in an MS-Register sub-option. 2151 Client's can thereafter employ efficient point-to-multipoint Server 2152 coordination under the assistance of the Proxy to dramatically reduce 2153 the number of messages sent over the ANET while enlisting the support 2154 of multiple Servers for fault tolerance. Clients can further include 2155 MS-Release suboptions in RS messages to request the Proxy to release 2156 from former Servers via the procedures discussed in Section 3.19.5. 2158 The OMNI interface specification [I-D.templin-6man-omni-interface] 2159 provides further discussion of the Client/Proxy RS/RA messaging 2160 involved in point-to-multipoint coordination. 2162 3.17. AERO Route Optimization 2164 While data packets are flowing between a source and target node, 2165 route optimization SHOULD be used. Route optimization is initiated 2166 by the first eligible Route Optimization Source (ROS) closest to the 2167 source as follows: 2169 o For Clients on VPNed and Direct interfaces, the Server is the ROS. 2171 o For Clients on Proxyed interfaces, the Proxy is the ROS. 2173 o For Clients on native and NATed interfaces, the Client itself is 2174 the ROS. 2176 o For correspondent nodes on INET/EUN interfaces serviced by a 2177 Gateway, the Gateway is the ROS. 2179 The route optimization procedure is conducted between the ROS and the 2180 target Server/Gateway acting as a Route Optimization Responder (ROR) 2181 in the same manner as for IPv6 ND Address Resolution and using the 2182 same NS/NA messaging. The target may either be a MNP Client serviced 2183 by a Server, or a non-MNP correspondent reachable via a Gateway. 2185 The procedures are specified in the following sections. 2187 3.17.1. Route Optimization Initiation 2189 While data packets are flowing from the source node toward a target 2190 node, the ROS performs address resolution by sending an NS message 2191 for Address Resolution (NS(AR)) to receive a solicited NA message 2192 from the ROR. When the ROS sends an NS(AR), it includes: 2194 o the AERO address of the ROS as the source address. 2196 o the data packet's destination as the Target Address. 2198 o the Solicited-Node multicast address [RFC4291] formed from the 2199 lower 24 bits of the data packet's destination as the destination 2200 address, e.g., for 2001:db8:1:2::10:2000 the NS destination 2201 address is ff02:0:0:0:0:1:ff10:2000. 2203 The NS(AR) message includes an OMNI option with no ifIndex-tuples and 2204 no SLLAO, such that the target will not create a neighbor cache 2205 entry. 2207 The ROS then encapsulates the NS(AR) message in a SPAN header with 2208 source set to its own SPAN address and destination set to the SPAN 2209 address corresponding to the packet's final destination, then sends 2210 the message into the SPAN without decrementing the network-layer TTL/ 2211 Hop Limit field. 2213 3.17.2. Relaying the NS 2215 When the Relay receives the NS(AR) message from the ROS, it discards 2216 the INET header and determines that the ROR is the next hop by 2217 consulting its standard IPv6 forwarding table for the SPAN header 2218 destination address. The Relay then forwards the message toward the 2219 ROR via the SPAN the same as for any IPv6 router. The final-hop 2220 Relay in the SPAN will deliver the message via a secured tunnel to 2221 the ROR. 2223 3.17.3. Processing the NS and Sending the NA 2225 When the ROR receives the NS(AR) message, it examines the Target 2226 Address to determine whether it has a neighbor cache entry and/or 2227 route that matches the target. If there is no match, the ROR drops 2228 the NS(AR) message. Otherwise, the ROR continues processing as 2229 follows: 2231 o if the target belongs to an MNP Client neighbor in the DEPARTED 2232 state the ROR changes the NS(AR) message SPAN destination address 2233 to the SPAN address of the Client's new Server, forwards the 2234 message into the SPAN and returns from processing. 2236 o If the target belongs to an MNP Client neighbor in the REACHABLE 2237 state, the ROR instead adds the AERO source address to the target 2238 Client's Report List with time set to ReportTime. 2240 o If the target belongs to a non-MNP route, the ROR continues 2241 processing without adding an entry to the Report List. 2243 The ROR then prepares a solicited NA message to send back to the ROS 2244 but does not create a neighbor cache entry. The ROR sets the NA 2245 source address to the AERO address corresponding to the target, sets 2246 the Target Address to the target of the solicitation, and sets the 2247 destination address to the source of the solicitation. 2249 The ROR then includes an OMNI option with prefix registration length 2250 set to the length of the MNP if the target is an MNP Client; 2251 otherwise, set to the maximum of the non-MNP prefix length and 64. 2252 (Note that a /64 limit is imposed to avoid causing the ROS to set 2253 short prefixes (e.g., "default") that would match destinations for 2254 which the routing system includes more-specific prefixes.) 2256 If the target is an MNP Client, the ROR next includes ifIndex-tuples 2257 in the OMNI option for each of the target Client's underlying 2258 interfaces with current information for each interface and with the S 2259 flag set to 0. The ROR then includes a TLLAO with ifIndex-tuples in 2260 one-to-one correspondence with the tuples that appear in the OMNI 2261 option. 2263 For VPNed, Direct and Proxyed interfaces, the ROR includes the IDs of 2264 all SPAN routers in the path in the Segment Routing List while 2265 setting the final entry to its own SPAN ID for VPNed and Direct 2266 interfaces or to the SPAN ID of the Proxy for Proxyed interfaces. 2267 For NATed and Native interfaces, if the ROS and ROR are located in 2268 the same SPAN partition and Segment Routing is not needed the ROR 2269 sets the Link Layer Address and Port Number (if necessary) to the 2270 Client's INET addresss for that interface. Otherwise, the ROR 2271 includes a Segment Routing List with its own SPAN ID as the final 2272 entry and omits the Link Layer Address. 2274 The ROR then sets the NA message R flag to 1 (as a router), S flag to 2275 1 (as a response to a solicitation), and O flag to 0 (as a proxy). 2276 The ROR finally encapsulates the NA message in a SPAN header with 2277 source set to its own SPAN address and destination set to the source 2278 SPAN address of the NS(AR) message, then forwards the message into 2279 the SPAN without decrementing the network-layer TTL/Hop Limit field. 2281 3.17.4. Relaying the NA 2283 When the Relay receives the NA message from the ROR, it discards the 2284 INET header and determines that the ROS is the next hop by consulting 2285 its standard IPv6 forwarding table for the SPAN header destination 2286 address. The Relay then forwards the SPAN-encapsulated NA message 2287 toward the ROS the same as for any IPv6 router. The final-hop Relay 2288 in the SPAN will deliver the message via a secured tunnel to the ROS. 2290 3.17.5. Processing the NA 2292 When the ROS receives the solicited NA message, it processes the 2293 message the same as for standard IPv6 Address Resolution [RFC4861]. 2294 In the process, it caches the source SPAN address then creates an 2295 asymmetric neighbor cache entry for the ROR and caches all 2296 information found in the OMNI and TLLAO options. The ROS finally 2297 sets the asymmetric neighbor cache entry lifetime to REACHABLE_TIME 2298 seconds. 2300 3.17.6. Route Optimization Maintenance 2302 Following route optimization, the ROS forwards future data packets 2303 destined to the target via the addresses found in the cached link- 2304 layer information. The route optimization is shared by all sources 2305 that send packets to the target via the ROS, i.e., and not just the 2306 source on behalf of which the route optimization was initiated. 2308 While new data packets destined to the target are flowing through the 2309 ROS, it sends additional NS(AR) messages to the ROR before 2310 ReachableTime expires to receive a fresh solicited NA message the 2311 same as described in the previous sections (route optimization 2312 refreshment strategies are an implementation matter, with a non- 2313 normative example given in Appendix B.1). The ROS uses the cached 2314 SPAN address of the ROR as the NS(AR) SPAN destination address, and 2315 sends up to MAX_MULTICAST_SOLICIT NS(AR) messages separated by 1 2316 second until an NA is received. If no NA is received, the ROS 2317 assumes that the current ROR has become unreachable and deletes the 2318 neighbor cache entry. Subsequent data packets will trigger a new 2319 route optimization per Section 3.17.1 to discover a new ROR while 2320 initial data packets travel over a suboptimal route. 2322 If an NA is received, the ROS then updates the asymmetric neighbor 2323 cache entry to refresh ReachableTime, while (for MNP destinations) 2324 the ROR adds or updates the ROS address to the target Client's Report 2325 List and with time set to ReportTime. While no data packets are 2326 flowing, the ROS instead allows ReachableTime for the asymmetric 2327 neighbor cache entry to expire. When ReachableTime expires, the ROS 2328 deletes the asymmetric neighbor cache entry. Any future data packets 2329 flowing through the ROS will again trigger a new route optimization. 2331 The ROS may also receive unsolicited NA messages from the ROR at any 2332 time (see: Section 3.19). If there is an asymmetric neighbor cache 2333 entry for the target, the ROS updates the link-layer information but 2334 does not update ReachableTime since the receipt of an unsolicited NA 2335 does not confirm that any forward paths are working. If there is no 2336 asymmetric neighbor cache entry, the ROS simply discards the 2337 unsolicited NA. 2339 In this arrangement, the ROS holds an asymmetric neighbor cache entry 2340 for the ROR, but the ROR does not hold an asymmetric neighbor cache 2341 entry for the ROS. The route optimization neighbor relationship is 2342 therefore asymmetric and unidirectional. If the target node also has 2343 packets to send back to the source node, then a separate route 2344 optimization procedure is performed in the reverse direction. But, 2345 there is no requirement that the forward and reverse paths be 2346 symmetric. 2348 3.18. Neighbor Unreachability Detection (NUD) 2350 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2351 [RFC4861] either reactively in response to persistent link-layer 2352 errors (see Section 3.14) or proactively to confirm reachability. 2353 The NUD algorithm is based on periodic control message exchanges. 2354 The algorithm may further be seeded by ND hints of forward progress, 2355 but care must be taken to avoid inferring reachability based on 2356 spoofed information. For example, authentic IPv6 ND message 2357 exchanges may be considered as acceptable hints of forward progress, 2358 while spurious data packets should not be. 2360 AERO Servers, Proxys and Gateways can use standard NS/NA NUD 2361 exchanges sent over the SPAN to securely test reachability without 2362 risk of DoS attacks from nodes pretending to be a neighbor; Proxys 2363 can further perform NUD to securely verify Server reachability on 2364 behalf of their proxyed Clients. However, a means for a ROS to test 2365 the unsecured forward directions of target route optimized paths is 2366 also necessary. 2368 When an ROR directs an ROS to a neighbor with one or more target 2369 link-layer addresses, the ROS can proactively test each such 2370 unsecured route optimized path by sending "loopback" NS(NUD) 2371 messages. While testing the paths, the ROS can optionally continue 2372 to send packets via the SPAN, maintain a small queue of packets until 2373 target reachability is confirmed, or (optimistically) allow packets 2374 to flow via the route optimized paths. 2376 When the ROS sends a loopback NS(NUD) message, it uses its AERO 2377 address as both the IPv6 source and destination address, and the MNP 2378 Subnet-Router anycast address as the Target Address. The ROS 2379 includes a Nonce and Timestamp option, then encapsulates the message 2380 in SPAN/INET headers with its own SPAN address as the source and the 2381 SPAN address of the route optimization target as the destination. 2382 The ROS then forwards the message to the target (either directly to 2383 the link layer address of the target if the target is in the same 2384 SPAN segment, or via a Relay if the target is in a different SPAN 2385 segment). 2387 When the route optimization target receives the NS(NUD) message, it 2388 notices that the IPv6 destination address is the same as the source 2389 address. It then reverses the SPAN source and destination addresses 2390 and returns the message to the ROS (either directly or via the SPAN). 2391 The route optimization target does not decrement the NS(NUD) message 2392 IPv6 Hop-Limit in the process, since the message has not exited the 2393 SPAN. 2395 When the ROS receives the NS(NUD) message, it can determine from the 2396 Nonce, Timestamp and Target Address that the message originated from 2397 itself and that it transited the forward path. The ROS need not 2398 prepare an NA response, since the destination of the response would 2399 be itself and testing the route optimization path again would be 2400 redundant. 2402 The ROS marks route optimization target paths that pass these NUD 2403 tests as "reachable", and those that do not as "unreachable". These 2404 markings inform the AERO interface forwarding algorithm specified in 2405 Section 3.13. 2407 Note that to avoid a DoS vector nodes MUST NOT return loopback 2408 NS(NUD) messages received from an unsecured link-layer source via a 2409 secured SPAN path. 2411 3.19. Mobility Management and Quality of Service (QoS) 2413 AERO is a Distributed Mobility Management (DMM) service. Each Server 2414 is responsible for only a subset of the Clients on the AERO link, as 2415 opposed to a Centralized Mobility Management (CMM) service where 2416 there is a single network mobility collective entity for all Clients. 2417 Clients coordinate with their associated Servers via RS/RA exchanges 2418 to maintain the DMM profile, and the AERO routing system tracks all 2419 current Client/Server peering relationships. 2421 Servers provide default routing and mobility/multilink services for 2422 their dependent Clients. Clients are responsible for maintaining 2423 neighbor relationships with their Servers through periodic RS/RA 2424 exchanges, which also serves to confirm neighbor reachability. When 2425 a Client's underlying interface address and/or QoS information 2426 changes, the Client is responsible for updating the Server with this 2427 new information. Note that for Proxyed interfaces, however, the 2428 Proxy can also perform some RS/RA exchanges on the Client's behalf. 2430 Mobility management considerations are specified in the following 2431 sections. 2433 3.19.1. Mobility Update Messaging 2435 Servers accommodate Client mobility/multilink and/or QoS change 2436 events by sending unsolicited NA (uNA) messages to each ROS in the 2437 target Client's Report List. When a Server sends a uNA message, it 2438 sets the IPv6 source address to the Client's AERO address, sets the 2439 destination address to All-Nodes multicast and sets the Target 2440 Address to the Client's Subnet-Router anycast address. The Server 2441 also includes an OMNI option with prefix registration information and 2442 with ifIndex-tuples for the target Client's remaining interfaces with 2443 S set to 0. The Server then includes a TLLAO with corresponding 2444 ifIndex-tuples prepared the same as for the initial route 2445 optimization event. The Server sets the NA R flag to 1, the S flag 2446 to 0 and the O flag to 0, then encapsulates the message in a SPAN 2447 header with source set to its own SPAN address and destination set to 2448 the SPAN address of the ROS and sends the message into the SPAN. 2450 As discussed in Section 7.2.6 of [RFC4861], the transmission and 2451 reception of uNA messages is unreliable but provides a useful 2452 optimization. In well-connected Internetworks with robust data links 2453 uNA messages will be delivered with high probability, but in any case 2454 the Server can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT uNAs 2455 to each ROS to increase the likelihood that at least one will be 2456 received. 2458 When the ROS receives an uNA message, it ignores the message if there 2459 is no existing neighbor cache entry for the Client. Otherwise, it 2460 uses the included OMNI option and TLLAO information to update the 2461 neighbor cache entry, but does not reset ReachableTime since the 2462 receipt of an unsolicited NA message from the target Server does not 2463 provide confirmation that any forward paths to the target Client are 2464 working. 2466 If uNA messages are lost, the ROS may be left with stale address and/ 2467 or QoS information for the Client for up to REACHABLE_TIME seconds. 2468 During this time, the ROS can continue sending packets according to 2469 its stale neighbor cache information. When ReachableTime is close to 2470 expiring, the ROS will re-initiate route optimization and receive 2471 fresh link-layer address information. 2473 In addition to sending uNA messages to the current set of ROSs for 2474 the Client, the Server also sends uNAs to the former link-layer 2475 address for any ifIndex-tuple for which the link-layer address has 2476 changed. The uNA messages update Proxys that cannot easily detect 2477 (e.g., without active probing) when a formerly-active Client has 2478 departed. 2480 3.19.2. Announcing Link-Layer Address and/or QoS Preference Changes 2482 When a Client needs to change its ANET addresses and/or QoS 2483 preferences (e.g., due to a mobility event), either the Client or its 2484 Proxys send RS messages to the Server via the SPAN with an OMNI 2485 option that includes an ifIndex-tuple with S set to 1 and with the 2486 new link quality and address information. 2488 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 2489 sending actual data packets in case one or more RAs are lost. If all 2490 RAs are lost, the Client SHOULD re-associate with a new Server. 2492 When the Server receives the Client's changes, it sends uNA messages 2493 to all nodes in the Report List the same as described in the previous 2494 section. 2496 3.19.3. Bringing New Links Into Service 2498 When a Client needs to bring new underlying interfaces into service 2499 (e.g., when it activates a new data link), it sends an RS message to 2500 the Server via the underlying interface with an OMNI option that 2501 includes an ifIndex-tuple with S set to 1 and appropriate link 2502 quality values and with link-layer address information for the new 2503 link. 2505 3.19.4. Removing Existing Links from Service 2507 When a Client needs to remove existing underlying interfaces from 2508 service (e.g., when it de-activates an existing data link), it sends 2509 an RS or uNA message to its Server with an OMNI option with 2510 appropriate link quality values. 2512 If the Client needs to send RS/uNA messages over an underlying 2513 interface other than the one being removed from service, it MUST 2514 include ifIndex-tuples with appropriate link quality values for any 2515 underlying interfaces being removed from service. 2517 3.19.5. Moving to a New Server 2519 When a Client associates with a new Server, it performs the Client 2520 procedures specified in Section 3.15.2. The Client also includes MS- 2521 Release identifiers in the RS message OMNI option per 2522 [I-D.templin-6man-omni-interface] if it wants the new Server to 2523 notify any old Servers from which the Client is departing. 2525 When the new Server receives the Client's RS message, it returns an 2526 RA as specified in Section 3.15.3 and sends up to 2527 MAX_NEIGHBOR_ADVERTIISEMENT uNA messages to any old Servers listed in 2528 OMNI option MS-Release identifiers. Each uNA message includes the 2529 Client's AERO address as the source address, the old Server's AERO 2530 address as the destination address, and an OMNI option with the 2531 Register/Release bit set to 0. The new Server wraps the uNA in a 2532 SPAN header with its own SPAN address as the source and the old 2533 Server's SPAN address as the destination, then sends the message into 2534 the SPAN. 2536 When an old Server receives the uNA, it changes the Client's neighbor 2537 cache entry state to DEPARTED, sets the link-layer address of the 2538 Client to the new Server's SPAN address, and sets DepartTime to 2539 DEPART_TIME seconds. After a short delay (e.g., 2 seconds) the old 2540 Server withdraws the Client's MNP from the routing system. After 2541 DepartTime expires, the old Server deletes the Client's neighbor 2542 cache entry. 2544 The old Server also sends unsolicited NA messages to all ROSs in the 2545 Client's Report List with an OMNI option with a single ifIndex-tuple 2546 with ifIndex set to 0 and S set to '1', and with the SPAN address of 2547 the new Server in a companion TLLAO. When the ROS receives the NA, 2548 it caches the address of the new Server in the existing asymmetric 2549 neighbor cache entry and marks the entry as STALE. Subsequent data 2550 packets will then flow according to any existing cached link-layer 2551 information and trigger a new NS(AR)/NA exchange via the new Server. 2553 Clients SHOULD NOT move rapidly between Servers in order to avoid 2554 causing excessive oscillations in the AERO routing system. Examples 2555 of when a Client might wish to change to a different Server include a 2556 Server that has gone unreachable, topological movements of 2557 significant distance, movement to a new geographic region, movement 2558 to a new SPAN segment, etc. 2560 When a Client moves to a new Server, some of the fragments of a 2561 multiple fragment packet may have already arrived at the old Server 2562 while others are en route to the new Server, however no special 2563 attention in the reassembly algorithm is necessary when re-routed 2564 fragments are simply treated as loss. 2566 3.20. Multicast 2568 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 2569 [RFC3810] proxy service for its EUNs and/or hosted applications 2570 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 2571 underlying interfaces for which group membership is required. The 2572 IGMP/MLD messages may be further forwarded by a first-hop ANET access 2573 router acting as an IGMP/MLD-snooping switch [RFC4541], then 2574 ultimately delivered to an AERO Proxy/Server acting as a Protocol 2575 Independent Multicast - Sparse-Mode (PIM-SM, or simply "PIM") 2576 Designated Router (DR) [RFC7761]. AERO Gateways also act as PIM 2577 routers (i.e., the same as AERO Proxys/Servers) on behalf of nodes on 2578 INET/EUN networks. The behaviors identified in the following 2579 sections correspond to Source-Specific Multicast (SSM) and Any-Source 2580 Multicast (ASM) operational modes. 2582 3.20.1. Source-Specific Multicast (SSM) 2584 When an ROS (i.e., an AERO Proxy/Server/Gateway) "X" acting as PIM 2585 router receives a Join/Prune message from a node on its downstream 2586 interfaces containing one or more ((S)ource, (G)roup) pairs, it 2587 updates its Multicast Routing Information Base (MRIB) accordingly. 2588 For each S belonging to a prefix reachable via X's non-AERO 2589 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 2590 on those interfaces per [RFC7761]. 2592 For each S belonging to a prefix reachable via X's AERO interface, X 2593 originates a separate copy of the Join/Prune for each (S,G) in the 2594 message using its own AERO address as the source address and ALL-PIM- 2595 ROUTERS as the destination address. X then encapsulates each message 2596 in a SPAN header with source address set to the SPAN address of X and 2597 destination address set to S then forwards the message into the SPAN. 2598 The SPAN in turn forwards the message to AERO Server/Gateway "Y" that 2599 services S. At the same time, if the message was a Join, X sends a 2600 route-optimization NS message toward each S the same as discussed in 2601 Section 3.17. The resulting NAs will return the AERO address for the 2602 prefix that matches S as the network-layer source address and TLLAOs 2603 with the SPAN addresses corresponding to any ifIndex-tuples that are 2604 currently servicing S. 2606 When Y processes the Join/Prune message, if S located behind any 2607 Native, Direct, VPNed or NATed interfaces Y acts as a PIM router and 2608 updates its MRIB to list X as the next hop in the reverse path. If S 2609 is located behind any Proxys "Z"*, Y also forwards the message to 2610 each Z* over the SPAN while continuing to use the AERO address of X 2611 as the source address. Each Z* then updates its MRIB accordingly and 2612 maintains the AERO address of X as the next hop in the reverse path. 2613 Since the Relays in the SPAN do not examine network layer control 2614 messages, this means that the (reverse) multicast tree path is simply 2615 from each Z* (and/or Y) to X with no other multicast-aware routers in 2616 the path. If any Z* (and/or Y) is located on the same SPAN segment 2617 as X, the multicast data traffic sent to X directly using SPAN/INET 2618 encapsulation instead of via a Relay. 2620 Following the initial Join/Prune and NS/NA messaging, X maintains an 2621 asymmetric neighbor cache entry for each S the same as if X was 2622 sending unicast data traffic to S. In particular, X performs 2623 additional NS/NA exchanges to keep the neighbor cache entry alive for 2624 up to t_periodic seconds [RFC7761]. If no new Joins are received 2625 within t_periodic seconds, X allows the neighbor cache entry to 2626 expire. Finally, if X receives any additional Join/Prune messages 2627 for (S,G) it forwards the messages to each Y and Z* in the neighbor 2628 cache entry over the SPAN. 2630 At some later time, Client C that holds an MNP for source S may 2631 depart from a first Proxy Z1 and/or connect via a new Proxy Z2. In 2632 that case, Y sends an unsolicited NA message to X the same as 2633 specified for unicast mobility in Section 3.19. When X receives the 2634 unsolicited NA message, it updates its asymmetric neighbor cache 2635 entry for the AERO address for source S and sends new Join messages 2636 to any new Proxys Z2. There is no requirement to send any Prune 2637 messages to old Proxys Z1 since source S will no longer source any 2638 multicast data traffic via Z1. Instead, the multicast state for 2639 (S,G) in Proxy Z1 will soon time out since no new Joins will arrive. 2641 After some later time, C may move to a new Server Y2 and depart from 2642 old Sever Y1. In that case, Y1 sends Join messages for any of C's 2643 active (S,G) groups to Y2 while including its own AERO address as the 2644 source address. This causes Y2 to include Y1 in the multicast 2645 forwarding tree during the interim time that Y1's symmetric neighbor 2646 cache entry for C is in the DEPARTED state. At the same time, Y1 2647 sends an unsolicited NA message to X with an OMNI option and TLLAO 2648 with ifIndex-tuple set to 0 and a release indication to cause X to 2649 release its asymmetric neighbor cache entry. X then sends a new Join 2650 message to S via the SPAN and re-initiates route optimization the 2651 same as if it were receiving a fresh Join message from a node on a 2652 downstream link. 2654 3.20.2. Any-Source Multicast (ASM) 2656 When an ROS X acting as a PIM router receives a Join/Prune from a 2657 node on its downstream interfaces containing one or more (*,G) pairs, 2658 it updates its Multicast Routing Information Base (MRIB) accordingly. 2659 X then forwards a copy of the message to the Rendezvous Point (RP) R 2660 for each G over the SPAN. X uses its own AERO address as the source 2661 address and ALL-PIM-ROUTERS as the destination address, then 2662 encapsulates each message in a SPAN header with source address set to 2663 the SPAN address of X and destination address set to R, then sends 2664 the message into the SPAN. At the same time, if the message was a 2665 Join X initiates NS/NA route optimization the same as for the SSM 2666 case discussed in Section 3.20.1. 2668 For each source S that sends multicast traffic to group G via R, the 2669 Proxy/Server Z* for the Client that aggregates S encapsulates the 2670 packets in PIM Register messages and forwards them to R via the SPAN. 2671 R may then elect to send a PIM Join to Z* over the SPAN. This will 2672 result in an (S,G) tree rooted at Z* with R as the next hop so that R 2673 will begin to receive two copies of the packet; one native copy from 2674 the (S, G) tree and a second copy from the pre-existing (*, G) tree 2675 that still uses PIM Register encapsulation. R can then issue a PIM 2676 Register-stop message to suppress the Register-encapsulated stream. 2677 At some later time, if C moves to a new Proxy/Server Z*, it resumes 2678 sending packets via PIM Register encapsulation via the new Z*. 2680 At the same time, as multicast listeners discover individual S's for 2681 a given G, they can initiate an (S,G) Join for each S under the same 2682 procedures discussed in Section 3.20.1. Once the (S,G) tree is 2683 established, the listeners can send (S, G) Prune messages to R so 2684 that multicast packets for group G sourced by S will only be 2685 delivered via the (S, G) tree and not from the (*, G) tree rooted at 2686 R. All mobility considerations discussed for SSM apply. 2688 3.20.3. Bi-Directional PIM (BIDIR-PIM) 2690 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 2691 approach to ASM that treats the Rendezvous Point (RP) as a Designated 2692 Forwarder (DF). Further considerations for BIDIR-PIM are out of 2693 scope. 2695 3.21. Operation over Multiple AERO Links (VLANs) 2697 An AERO Client can connect to multiple AERO links the same as for any 2698 data link service. In that case, the Client maintains a distinct 2699 AERO interface for each link, e.g., 'aero0' for the first link, 2700 'aero1' for the second, 'aero2' for the third, etc. Each AERO link 2701 would include its own distinct set of Relays, Servers and Proxys, 2702 thereby providing redundancy in case of failures. 2704 The Relays, Servers and Proxys on each AERO link can assign AERO and 2705 SPAN addresses that use the same or different numberings from those 2706 on other links. Since the links are mutually independent there is no 2707 requirement for avoiding inter-link address duplication, e.g., the 2708 same AERO address such as fe80::1000 could be used to number distinct 2709 nodes that connect to different AERO links. 2711 Each AERO link could utilize the same or different ANET connections. 2712 The links can be distinguished at the link-layer via Virtual Local 2713 Area Network (VLAN) tagging (e.g., IEEE 802.1Q) and/or through 2714 assignment of distinct sets of MSPs on each link. This gives rise to 2715 the opportunity for supporting multiple redundant networked paths, 2716 where each VLAN is distinguished by a different label (e.g., colors 2717 such as Red, Green, Blue, etc.). In particular, the Client can tag 2718 its RS messages with the appropriate label to cause the network to 2719 select the desired VLAN. 2721 Clients that connect to multiple AERO interfaces can select the 2722 outgoing interface appropriate for a given Red/Blue/Green/etc. 2723 traffic profile while (in the reverse direction) correspondent nodes 2724 must have some way of steering their packets destined to a target via 2725 the correct AERO link. 2727 In a first alternative, if each AERO link services different MSPs, 2728 then the Client can receive a distinct MNP from each of the links. 2729 IP routing will therefore assure that the correct Red/Green/Blue/etc. 2730 network is used for both outbound and inbound traffic. This can be 2731 accomplished using existing technologies and approaches, and without 2732 requiring any special supporting code in correspondent nodes or 2733 Relays. 2735 In a second alternative, if each AERO link services the same MSP(s) 2736 then each link could assign a distinct "AERO Link Anycast" address 2737 that is configured by all Relays on the link. Correspondent nodes 2738 then include a "type 4" routing header with the Anycast address for 2739 the AERO link as the IPv6 destination and with the address of the 2740 target encoded as the "next segment" in the routing header 2741 [RFC8402][I-D.ietf-6man-segment-routing-header]. Standard IP routing 2742 will then direct the packet to the nearest Relay for the correct AERO 2743 link, which will replace the destination address with the target 2744 address then forward the packet to the target. 2746 3.22. DNS Considerations 2748 AERO Client MNs and INET correspondent nodes consult the Domain Name 2749 System (DNS) the same as for any Internetworking node. When 2750 correspondent nodes and Client MNs use different IP protocol versions 2751 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 2752 A records for IPv4 address mappings to MNs which must then be 2753 populated in Gateway NAT64 mapping caches. In that way, an IPv4 2754 correspondent node can send packets to the IPv4 address mapping of 2755 the target MN, and the Gateway will translate the IPv4 header and 2756 destination address into an IPv6 header and IPv6 destination address 2757 of the MN. 2759 When an AERO Client registers with an AERO Server, the Server can 2760 return the address(es) of DNS servers in RDNSS options [RFC6106]. 2761 The DNS server provides the IP addresses of other MNs and 2762 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 2764 3.23. Transition Considerations 2766 The SPAN ensures that dissimilar INET partitions can be joined into a 2767 single unified AERO link, even though the partitions themselves may 2768 have differing protocol versions and/or incompatible addressing 2769 plans. However, a commonality can be achieved by incrementally 2770 distributing globally routable (i.e., native) IP prefixes to 2771 eventually reach all nodes (both mobile and fixed) in all SPAN 2772 segments. This can be accomplished by incrementally deploying AERO 2773 Gateways on each INET partition, with each Gateway distributing its 2774 MNPs and/or discovering non-MNP prefixes on its INET links. 2776 This gives rise to the opportunity to eventually distribute native IP 2777 addresses to all nodes, and to present a unified AERO link view 2778 (bridged by the SPAN) even if the INET partitions remain in their 2779 current protocol and addressing plans. In that way, the AERO link 2780 can serve the dual purpose of providing a mobility/multilink service 2781 and a transition service. Or, if an INET partition is transitioned 2782 to a native IP protocol version and addressing scheme that is 2783 compatible with the AERO link MNP-based addressing scheme, the 2784 partition and AERO link can be joined by Gateways. 2786 Gateways that connect INETs/EUNs with dissimilar IP protocol versions 2787 must employ a network address and protocol translation function such 2788 as NAT64[RFC6146]. 2790 3.24. Detecting and Reacting to Server and Relay Failures 2792 In environments where rapid failure recovery is required, Servers and 2793 Relays SHOULD use Bidirectional Forwarding Detection (BFD) [RFC5880]. 2794 Nodes that use BFD can quickly detect and react to failures so that 2795 cached information is re-established through alternate nodes. BFD 2796 control messaging is carried only over well-connected ground domain 2797 networks (i.e., and not low-end radio links) and can therefore be 2798 tuned for rapid response. 2800 Servers and Relays maintain BFD sessions in parallel with their BGP 2801 peerings. If a Server or Relay fails, BGP peers will quickly re- 2802 establish routes through alternate paths the same as for common BGP 2803 deployments. Similarly, Proxys maintain BFD sessions with their 2804 associated Relays even though they do not establish BGP peerings with 2805 them. 2807 Proxys SHOULD use proactive NUD for Servers for which there are 2808 currently active ANET Clients in a manner that parallels BFD, i.e., 2809 by sending unicast NS messages in rapid succession to receive 2810 solicited NA messages. When the Proxy is also sending RS messages on 2811 behalf of ANET Clients, the RS/RA messaging can be considered as 2812 equivalent hints of forward progress. This means that the Proxy need 2813 not also send a periodic NS if it has already sent an RS within the 2814 same period. If a Server fails, the Proxy will cease to receive 2815 advertisements and can quickly inform Clients of the outage by 2816 sending multicast RA messages on the ANET interface. 2818 The Proxy sends multicast RA messages with source address set to the 2819 Server's address, destination address set to All-Nodes multicast, and 2820 Router Lifetime set to 0. The Proxy SHOULD send 2821 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2822 [RFC4861]. Any Clients on the ANET interface that have been using 2823 the (now defunct) Server will receive the RA messages and associate 2824 with a new Server. 2826 3.25. AERO Clients on the Open Internet 2828 AERO Clients that connect to the open Internet via native and/or 2829 NATed interfaces can establish a VPN to securely connect to a Server 2830 in a "tethered" arrangement with all of the Client's traffic 2831 transiting the Server. Alternatively, the Client can exchange ND 2832 messages directly with other AERO nodes using SPAN/INET 2833 encapsulation. In that case, the Client must apply asymmetric 2834 security for IPv6 ND messages to ensure routing and neighbor cache 2835 integrity using SEcure Neighbor Discovery (SEND) [RFC3971] and 2836 Cryptographically Generated Addresses (CGAs) [RFC3972]. 2838 When a Client enables native and/or NATed underlying interfaces, the 2839 AERO interface sends an RS message with IPv6 source address set to a 2840 link-local CGA, with IPv6 destination set to All-Routers multicast, 2841 with an OMNI opton and with SEND/CGA options to provide message 2842 authentication. The Client also includes an SLLAO with the IP 2843 address and Port Number used for INET enacpsulation written in 2844 obfuscated form as discussed in Section 3.6. The Client wraps the 2845 message in a SPAN header with source address set to the Client's SPAN 2846 address and with destination address set to the SPAN address of a 2847 Server. The Client then further wraps the SPAN message in an INET 2848 header with source set to the Client's INET address and destination 2849 set to the Server's INET address, then sends the message to the 2850 Server. 2852 When the Server receives the RS message, it authenticates the message 2853 and registers the Client's MNP and INET interface information 2854 according to the OMNI option parameters. The Server then returns an 2855 RA message with IPv6 source set to its CGA, with IPv6 destination set 2856 to the Client's CGA, with an OMNI option and with SEND/CGA options to 2857 provide message authentication. If the Client is located behind a 2858 NAT, the Server instead sets the IPv6 destination to the Teredo IPv6 2859 address formed using the service prefix fe80::/32 (see Section 4 of 2860 [RFC4380]). The Server then wraps the message in a SPAN header with 2861 source address set to the Server's SPAN address and destination set 2862 to the Client's SPAN address, and further wraps the SPAN message in 2863 an INET header with source set to the Server's INET address and 2864 destination set to the Client's INET address. 2866 After the Client has registered its INET interfaces in such RS/RA 2867 exchanges it sends periodic RS messages to receive fresh RA messages 2868 before the Router Lifetime received on each INET interface expires 2869 (again, using SEND/CGA). The Client also maintains default routes 2870 via its Servers, i.e., the same as described in earlier sections. 2872 When the Client sends messages to target IP addresses, it also 2873 invokes route optimization per Section 3.17 using IPv6 ND address 2874 resolution messaging. The Client sends the (SEND/CGA-protected) 2875 NS(AR) message wrapped in a SPAN header with source set to the 2876 Client's SPAN address and destination set to the SPAN address 2877 corresponding to the target, and with INET destination address set to 2878 the address of the Server. The Server authenticates the message and 2879 sends a corresponding NS(AR) message over the SPAN. When the ROR 2880 receives the NS(AR), it adds the Client's SPAN address and Server's 2881 INET address to the target's Report List, and returns an NA with OMNI 2882 and TLLAO information for the target. The Server then returns a 2883 (SEND/CGA-protected) NA message to the Client. 2885 Following route optimization, for any of the target's Native 2886 addresses the Client forwards data packets directly to the target 2887 INET addresses according to the OMNI/TLLAO information. For any 2888 NATed addresses, the Client first establishes NAT state as specified 2889 in [RFC6081][RFC4380] while using "loopback" NS(NUD) messages as 2890 discussed in Section 3.18 as "bubbles". The Client continues to send 2891 data packets via the SPAN until NAT state is populated, then begins 2892 forwarding packets via the direct path through the NAT to the target. 2894 The ROR may return uNAs via the Server if the target moves, and the 2895 Server will send corresponding (SEND/CGA-protected) uNAs to the 2896 Client. The Client can also send "loopback" NS(NUD) messages to test 2897 forward path reachability even though there is no security 2898 association between the Client and the target. 2900 4. Implementation Status 2902 An AERO implementation based on OpenVPN (https://openvpn.net/) was 2903 announced on the v6ops mailing list on January 10, 2018 and an 2904 initial public release of the AERO proof-of-concept source code was 2905 announced on the intarea mailing list on August 21, 2015. 2907 As of 4/1/2020, more recent updated implementations are under 2908 internal development and testing with plans to release in the near 2909 future. 2911 5. IANA Considerations 2913 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 2914 AERO in the "enterprise-numbers" registry. 2916 The IANA has assigned the UDP port number "8060" for an earlier 2917 experimental version of AERO [RFC6706]. This document obsoletes 2918 [RFC6706] and claims the UDP port number "8060" for all future use. 2920 No further IANA actions are required. 2922 6. Security Considerations 2924 AERO Relays configure secured tunnels with AERO Servers and Proxys 2925 within their local SPAN segments. Applicable secured tunnel 2926 alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 2927 [RFC6347], WireGuard, etc. The AERO Relays of all SPAN segments in 2928 turn configure secured tunnels for their neighboring AERO Relays 2929 across the SPAN. Therefore, control messages that traverse the SPAN 2930 between any pair of AERO link neighbors are already secured. 2932 AERO Servers, Gateways and Proxys targeted by a route optimization 2933 may also receive packets directly from the INET partitions instead of 2934 via the SPAN. For INET partitions that apply effective ingress 2935 filtering to defeat source address spoofing, the simple data origin 2936 authentication procedures in Section 3.11 can be applied. 2938 For INET partitions that cannot apply effective ingress filtering, 2939 the two options for securing communications include 1) disable route 2940 optimization so that all traffic is conveyed over secured tunnels via 2941 the SPAN, or 2) enable on-demand secure tunnel creation between INET 2942 partition neighbors. Option 1) would result in longer routes than 2943 necessary and traffic concentration on critical infrastructure 2944 elements. Option 2) could be coordinated by establishing a secured 2945 tunnel on-demand instead of performing an NS/NA exchange in the route 2946 optimization procedures. Procedures for establishing on-demand 2947 secured tunnels are out of scope. 2949 AERO Clients that connect to secured enclaves need not apply security 2950 to their ND messages, since the messages will be intercepted by a 2951 perimeter Proxy that applies security on its outward-facing 2952 interface. AERO Clients located outside of secured enclaves can use 2953 symmetric network and/or transport layer security services such as 2954 VPNs, but when there are many prospective neighbors with dynamically 2955 changing connectivity an asymmetric security service such as SEcure 2956 Neighbor Discovery (SEND) [RFC3971] and Cryptographically Generated 2957 Addresses (CGAs) [RFC3972] may be more appropriate. 2959 Application endpoints SHOULD use application-layer security services 2960 such as TLS/SSL, DTLS or SSH [RFC4251] to assure the same level of 2961 protection as for critical secured Internet services. AERO Clients 2962 that require host-based VPN services SHOULD use symmetric network 2963 and/or transport layer security services such as IPsec, TLS/SSL, 2964 DTLS, etc. AERO Proxys and Servers can also provide a network-based 2965 VPN service on behalf of the Client, e.g., if the Client is located 2966 within a secured enclave and cannot establish a VPN on its own 2967 behalf. 2969 AERO Servers and Relays present targets for traffic amplification 2970 Denial of Service (DoS) attacks. This concern is no different than 2971 for widely-deployed VPN security gateways in the Internet, where 2972 attackers could send spoofed packets to the gateways at high data 2973 rates. This can be mitigated by connecting Servers and Relays over 2974 dedicated links with no connections to the Internet and/or when 2975 connections to the Internet are only permitted through well-managed 2976 firewalls. Traffic amplification DoS attacks can also target an AERO 2977 Client's low data rate links. This is a concern not only for Clients 2978 located on the open Internet but also for Clients in secured 2979 enclaves. AERO Servers and Proxys can institute rate limits that 2980 protect Clients from receiving packet floods that could DoS low data 2981 rate links. 2983 AERO Gateways must implement ingress filtering to avoid a spoofing 2984 attack in which spurious SPAN messages are injected into an AERO link 2985 from an outside attacker. AERO Clients MUST ensure that their 2986 connectivity is not used by unauthorized nodes on their EUNs to gain 2987 access to a protected network, i.e., AERO Clients that act as routers 2988 MUST NOT provide routing services for unauthorized nodes. (This 2989 concern is no different than for ordinary hosts that receive an IP 2990 address delegation but then "share" the address with other nodes via 2991 some form of Internet connection sharing such as tethering.) 2993 The MAP list MUST be well-managed and secured from unauthorized 2994 tampering, even though the list contains only public information. 2995 The MAP list can be conveyed to the Client in a similar fashion as in 2997 [RFC5214] (e.g., through layer 2 data link login messaging, secure 2998 upload of a static file, DNS lookups, etc.). 3000 Although public domain and commercial SEND implementations exist, 3001 concerns regarding the strength of the cryptographic hash algorithm 3002 have been documented [RFC6273] [RFC4982]. 3004 Security considerations for accepting link-layer ICMP messages and 3005 reflected packets are discussed throughout the document. 3007 7. Acknowledgements 3009 Discussions in the IETF, aviation standards communities and private 3010 exchanges helped shape some of the concepts in this work. 3011 Individuals who contributed insights include Mikael Abrahamsson, Mark 3012 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 3013 Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian Farrel, Nick Green, 3014 Sri Gundavelli, Brian Haberman, Bernhard Haindl, Joel Halpern, Tom 3015 Herbert, Sascha Hlusiak, Lee Howard, Zdenek Jaron, Andre Kostur, 3016 Hubert Kuenig, Ted Lemon, Andy Malis, Satoru Matsushima, Tomek 3017 Mrugalski, Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal 3018 Skorepa, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, Lloyd 3019 Wood and James Woodyatt. Members of the IESG also provided valuable 3020 input during their review process that greatly improved the document. 3021 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 3022 for their shepherding guidance during the publication of the AERO 3023 first edition. 3025 This work has further been encouraged and supported by Boeing 3026 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 3027 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 3028 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 3029 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 3030 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 3031 Gene MacLean III, Rob Muszkiewicz, Sean O'Sullivan, Vijay 3032 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 3033 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 3034 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 3035 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 3036 Benson, Katie Tran and Eric Yeh are especially acknowledged for 3037 implementing the AERO functions as extensions to the public domain 3038 OpenVPN distribution. 3040 Earlier works on NBMA tunneling approaches are found in 3041 [RFC2529][RFC5214][RFC5569]. 3043 Many of the constructs presented in this second edition of AERO are 3044 based on the author's earlier works, including: 3046 o The Internet Routing Overlay Network (IRON) 3047 [RFC6179][I-D.templin-ironbis] 3049 o Virtual Enterprise Traversal (VET) 3050 [RFC5558][I-D.templin-intarea-vet] 3052 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 3053 [RFC5320][I-D.templin-intarea-seal] 3055 o AERO, First Edition [RFC6706] 3057 Note that these works cite numerous earlier efforts that are not also 3058 cited here due to space limitations. The authors of those earlier 3059 works are acknowledged for their insights. 3061 This work is aligned with the NASA Safe Autonomous Systems Operation 3062 (SASO) program under NASA contract number NNA16BD84C. 3064 This work is aligned with the FAA as per the SE2025 contract number 3065 DTFAWA-15-D-00030. 3067 This work is aligned with the Boeing Commercial Airplanes (BCA) 3068 Internet of Things (IoT) and autonomy programs. 3070 This work is aligned with the Boeing Information Technology (BIT) 3071 MobileNet program. 3073 8. References 3075 8.1. Normative References 3077 [I-D.templin-6man-omni-interface] 3078 Templin, F. and T. Whyman, "Transmission of IPv6 Packets 3079 over Overlay Multilink Network (OMNI) Interfaces", draft- 3080 templin-6man-omni-interface-14 (work in progress), April 3081 2020. 3083 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3084 DOI 10.17487/RFC0791, September 1981, 3085 . 3087 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 3088 RFC 792, DOI 10.17487/RFC0792, September 1981, 3089 . 3091 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3092 Requirement Levels", BCP 14, RFC 2119, 3093 DOI 10.17487/RFC2119, March 1997, 3094 . 3096 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3097 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3098 December 1998, . 3100 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 3101 "Definition of the Differentiated Services Field (DS 3102 Field) in the IPv4 and IPv6 Headers", RFC 2474, 3103 DOI 10.17487/RFC2474, December 1998, 3104 . 3106 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3107 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3108 DOI 10.17487/RFC3971, March 2005, 3109 . 3111 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3112 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3113 . 3115 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3116 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3117 November 2005, . 3119 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3120 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3121 . 3123 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 3124 Network Address Translations (NATs)", RFC 4380, 3125 DOI 10.17487/RFC4380, February 2006, 3126 . 3128 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3129 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3130 DOI 10.17487/RFC4861, September 2007, 3131 . 3133 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3134 Address Autoconfiguration", RFC 4862, 3135 DOI 10.17487/RFC4862, September 2007, 3136 . 3138 [RFC5175] Haberman, B., Ed. and R. Hinden, "IPv6 Router 3139 Advertisement Flags Option", RFC 5175, 3140 DOI 10.17487/RFC5175, March 2008, 3141 . 3143 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 3144 DOI 10.17487/RFC6081, January 2011, 3145 . 3147 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3148 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3149 May 2017, . 3151 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3152 (IPv6) Specification", STD 86, RFC 8200, 3153 DOI 10.17487/RFC8200, July 2017, 3154 . 3156 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3157 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3158 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3159 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3160 . 3162 8.2. Informative References 3164 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3165 2016. 3167 [I-D.ietf-6man-segment-routing-header] 3168 Filsfils, C., Dukes, D., Previdi, S., Leddy, J., 3169 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3170 (SRH)", draft-ietf-6man-segment-routing-header-26 (work in 3171 progress), October 2019. 3173 [I-D.ietf-dmm-distributed-mobility-anchoring] 3174 Chan, A., Wei, X., Lee, J., Jeon, S., and C. Bernardos, 3175 "Distributed Mobility Anchoring", draft-ietf-dmm- 3176 distributed-mobility-anchoring-15 (work in progress), 3177 March 2020. 3179 [I-D.ietf-intarea-gue] 3180 Herbert, T., Yong, L., and O. Zia, "Generic UDP 3181 Encapsulation", draft-ietf-intarea-gue-09 (work in 3182 progress), October 2019. 3184 [I-D.ietf-intarea-gue-extensions] 3185 Herbert, T., Yong, L., and F. Templin, "Extensions for 3186 Generic UDP Encapsulation", draft-ietf-intarea-gue- 3187 extensions-06 (work in progress), March 2019. 3189 [I-D.ietf-intarea-tunnels] 3190 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3191 Architecture", draft-ietf-intarea-tunnels-10 (work in 3192 progress), September 2019. 3194 [I-D.ietf-rtgwg-atn-bgp] 3195 Templin, F., Saccone, G., Dawra, G., Lindem, A., and V. 3196 Moreno, "A Simple BGP-based Mobile Routing System for the 3197 Aeronautical Telecommunications Network", draft-ietf- 3198 rtgwg-atn-bgp-05 (work in progress), January 2020. 3200 [I-D.templin-6man-dhcpv6-ndopt] 3201 Templin, F., "A Unified Stateful/Stateless Configuration 3202 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-09 3203 (work in progress), January 2020. 3205 [I-D.templin-intarea-grefrag] 3206 Templin, F., "GRE Tunnel Level Fragmentation", draft- 3207 templin-intarea-grefrag-04 (work in progress), July 2016. 3209 [I-D.templin-intarea-seal] 3210 Templin, F., "The Subnetwork Encapsulation and Adaptation 3211 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 3212 progress), January 2014. 3214 [I-D.templin-intarea-vet] 3215 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 3216 templin-intarea-vet-40 (work in progress), May 2013. 3218 [I-D.templin-ironbis] 3219 Templin, F., "The Interior Routing Overlay Network 3220 (IRON)", draft-templin-ironbis-16 (work in progress), 3221 March 2014. 3223 [I-D.templin-v6ops-pdhost] 3224 Templin, F., "IPv6 Prefix Delegation and Multi-Addressing 3225 Models", draft-templin-v6ops-pdhost-25 (work in progress), 3226 January 2020. 3228 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3230 [RFC1035] Mockapetris, P., "Domain names - implementation and 3231 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3232 November 1987, . 3234 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 3235 Communication Layers", STD 3, RFC 1122, 3236 DOI 10.17487/RFC1122, October 1989, 3237 . 3239 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 3240 DOI 10.17487/RFC1191, November 1990, 3241 . 3243 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3244 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3245 . 3247 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3248 DOI 10.17487/RFC2003, October 1996, 3249 . 3251 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3252 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3253 . 3255 [RFC2492] Armitage, G., Schulter, P., and M. Jork, "IPv6 over ATM 3256 Networks", RFC 2492, DOI 10.17487/RFC2492, January 1999, 3257 . 3259 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3260 Domains without Explicit Tunnels", RFC 2529, 3261 DOI 10.17487/RFC2529, March 1999, 3262 . 3264 [RFC2764] Gleeson, B., Lin, A., Heinanen, J., Armitage, G., and A. 3265 Malis, "A Framework for IP Based Virtual Private 3266 Networks", RFC 2764, DOI 10.17487/RFC2764, February 2000, 3267 . 3269 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 3270 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 3271 DOI 10.17487/RFC2784, March 2000, 3272 . 3274 [RFC2890] Dommety, G., "Key and Sequence Number Extensions to GRE", 3275 RFC 2890, DOI 10.17487/RFC2890, September 2000, 3276 . 3278 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", 3279 RFC 2923, DOI 10.17487/RFC2923, September 2000, 3280 . 3282 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3283 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3284 . 3286 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3287 of Explicit Congestion Notification (ECN) to IP", 3288 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3289 . 3291 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3292 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3293 DOI 10.17487/RFC3810, June 2004, 3294 . 3296 [RFC3819] Karn, P., Ed., Bormann, C., Fairhurst, G., Grossman, D., 3297 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 3298 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 3299 RFC 3819, DOI 10.17487/RFC3819, July 2004, 3300 . 3302 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 3303 for IPv6 Hosts and Routers", RFC 4213, 3304 DOI 10.17487/RFC4213, October 2005, 3305 . 3307 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3308 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3309 January 2006, . 3311 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3312 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3313 DOI 10.17487/RFC4271, January 2006, 3314 . 3316 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3317 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3318 2006, . 3320 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3321 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3322 December 2005, . 3324 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3325 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3326 2006, . 3328 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3329 Control Message Protocol (ICMPv6) for the Internet 3330 Protocol Version 6 (IPv6) Specification", STD 89, 3331 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3332 . 3334 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3335 Protocol (LDAP): The Protocol", RFC 4511, 3336 DOI 10.17487/RFC4511, June 2006, 3337 . 3339 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3340 "Considerations for Internet Group Management Protocol 3341 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3342 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3343 . 3345 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3346 "Internet Group Management Protocol (IGMP) / Multicast 3347 Listener Discovery (MLD)-Based Multicast Forwarding 3348 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3349 August 2006, . 3351 [RFC4607] Holbrook, H. and B. Cain, "Source-Specific Multicast for 3352 IP", RFC 4607, DOI 10.17487/RFC4607, August 2006, 3353 . 3355 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 3356 Errors at High Data Rates", RFC 4963, 3357 DOI 10.17487/RFC4963, July 2007, 3358 . 3360 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3361 Algorithms in Cryptographically Generated Addresses 3362 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3363 . 3365 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3366 "Bidirectional Protocol Independent Multicast (BIDIR- 3367 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3368 . 3370 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3371 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3372 DOI 10.17487/RFC5214, March 2008, 3373 . 3375 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3376 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3377 February 2010, . 3379 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3380 Route Optimization Requirements for Operational Use in 3381 Aeronautics and Space Exploration Mobile Networks", 3382 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3383 . 3385 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3386 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3387 . 3389 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3390 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3391 January 2010, . 3393 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 3394 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 3395 . 3397 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3398 "IPv6 Router Advertisement Options for DNS Configuration", 3399 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3400 . 3402 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3403 NAT64: Network Address and Protocol Translation from IPv6 3404 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3405 April 2011, . 3407 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3408 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3409 . 3411 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3412 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3413 DOI 10.17487/RFC6221, May 2011, 3414 . 3416 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3417 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3418 DOI 10.17487/RFC6273, June 2011, 3419 . 3421 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3422 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3423 January 2012, . 3425 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3426 for Equal Cost Multipath Routing and Link Aggregation in 3427 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3428 . 3430 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3431 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3432 . 3434 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID Field", 3435 RFC 6864, DOI 10.17487/RFC6864, February 2013, 3436 . 3438 [RFC7269] Chen, G., Cao, Z., Xie, C., and D. Binet, "NAT64 3439 Deployment Options and Experience", RFC 7269, 3440 DOI 10.17487/RFC7269, June 2014, 3441 . 3443 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3444 Korhonen, "Requirements for Distributed Mobility 3445 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3446 . 3448 [RFC7421] Carpenter, B., Ed., Chown, T., Gont, F., Jiang, S., 3449 Petrescu, A., and A. Yourtchenko, "Analysis of the 64-bit 3450 Boundary in IPv6 Addressing", RFC 7421, 3451 DOI 10.17487/RFC7421, January 2015, 3452 . 3454 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3455 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3456 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3457 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3458 2016, . 3460 [RFC8086] Yong, L., Ed., Crabbe, E., Xu, X., and T. Herbert, "GRE- 3461 in-UDP Encapsulation", RFC 8086, DOI 10.17487/RFC8086, 3462 March 2017, . 3464 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 3465 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 3466 DOI 10.17487/RFC8201, July 2017, 3467 . 3469 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3470 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3471 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3472 July 2018, . 3474 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3475 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3476 . 3478 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 3479 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3480 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 3481 . 3483 Appendix A. AERO Alternate Encapsulations 3485 When GUE encapsulation is not needed, AERO can use common 3486 encapsulations such as IP-in-IP [RFC2003][RFC2473][RFC4213], Generic 3487 Routing Encapsulation (GRE) [RFC2784][RFC2890] and others. The 3488 encapsulation is therefore only differentiated from non-AERO tunnels 3489 through the application of AERO control messaging and not through, 3490 e.g., a well-known UDP port number. 3492 As for GUE encapsulation, alternate AERO encapsulation formats may 3493 require encapsulation layer fragmentation. For simple IP-in-IP 3494 encapsulation, an IPv6 fragment header is inserted directly between 3495 the inner and outer IP headers when needed, i.e., even if the outer 3496 header is IPv4. The IPv6 Fragment Header is identified to the outer 3497 IP layer by its IP protocol number, and the Next Header field in the 3498 IPv6 Fragment Header identifies the inner IP header version. For GRE 3499 encapsulation, a GRE fragment header is inserted within the GRE 3500 header [I-D.templin-intarea-grefrag]. 3502 Figure 6 shows the AERO IP-in-IP encapsulation format before any 3503 fragmentation is applied: 3505 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3506 | Outer IPv4 Header | | Outer IPv6 Header | 3507 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3508 |IPv6 Frag Header (optional)| |IPv6 Frag Header (optional)| 3509 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3510 | Inner IP Header | | Inner IP Header | 3511 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3512 | | | | 3513 ~ ~ ~ ~ 3514 ~ Inner Packet Body ~ ~ Inner Packet Body ~ 3515 ~ ~ ~ ~ 3516 | | | | 3517 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3519 Minimal Encapsulation in IPv4 Minimal Encapsulation in IPv6 3521 Figure 6: Minimal Encapsulation Format using IP-in-IP 3523 Figure 7 shows the AERO GRE encapsulation format before any 3524 fragmentation is applied: 3526 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3527 | Outer IP Header | 3528 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3529 | GRE Header | 3530 | (with checksum, key, etc..) | 3531 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3532 | GRE Fragment Header (optional)| 3533 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3534 | Inner IP Header | 3535 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3536 | | 3537 ~ ~ 3538 ~ Inner Packet Body ~ 3539 ~ ~ 3540 | | 3541 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3543 Figure 7: Minimal Encapsulation Using GRE 3545 Alternate encapsulation may be preferred in environments where GUE 3546 encapsulation would add unnecessary overhead. For example, certain 3547 low-bandwidth wireless data links may benefit from a reduced 3548 encapsulation overhead. 3550 GUE encapsulation can traverse network paths that are inaccessible to 3551 non-UDP encapsulations, e.g., for crossing Network Address 3552 Translators (NATs). More and more, network middleboxes are also 3553 being configured to discard packets that include anything other than 3554 a well-known IP protocol such as UDP and TCP. It may therefore be 3555 necessary to determine the potential for middlebox filtering before 3556 enabling alternate encapsulation in a given environment. 3558 In addition to IP-in-IP, GRE and GUE, AERO can also use security 3559 encapsulations such as IPsec, TLS/SSL, DTLS, etc. In that case, AERO 3560 control messaging and route determination occur before security 3561 encapsulation is applied for outgoing packets and after security 3562 decapsulation is applied for incoming packets. 3564 AERO is especially well suited for use with VPN system encapsulations 3565 such as OpenVPN [OVPN]. 3567 Appendix B. Non-Normative Considerations 3569 AERO can be applied to a multitude of Internetworking scenarios, with 3570 each having its own adaptations. The following considerations are 3571 provided as non-normative guidance: 3573 B.1. Implementation Strategies for Route Optimization 3575 Route optimization as discussed in Section 3.17 results in the route 3576 optimization source (ROS) creating an asymmetric neighbor cache entry 3577 for the target neighbor. The neighbor cache entry is maintained for 3578 at most REACHABLE_TIME seconds and then deleted unless updated. In 3579 order to refresh the neighbor cache entry lifetime before the 3580 ReachableTime timer expires, the specification requires 3581 implementations to issue a new NS/NA exchange to reset ReachableTime 3582 to REACHABLE_TIME seconds while data packets are still flowing. 3583 However, the decision of when to initiate a new NS/NA exchange and to 3584 perpetuate the process is left as an implementation detail. 3586 One possible strategy may be to monitor the neighbor cache entry 3587 watching for data packets for (REACHABLE_TIME - 5) seconds. If any 3588 data packets have been sent to the neighbor within this timeframe, 3589 then send an NS to receive a new NA. If no data packets have been 3590 sent, wait for 5 additional seconds and send an immediate NS if any 3591 data packets are sent within this "expiration pending" 5 second 3592 window. If no additional data packets are sent within the 5 second 3593 window, delete the neighbor cache entry. 3595 The monitoring of the neighbor data packet traffic therefore becomes 3596 an asymmetric ongoing process during the neighbor cache entry 3597 lifetime. If the neighbor cache entry expires, future data packets 3598 will trigger a new NS/NA exchange while the packets themselves are 3599 delivered over a longer path until route optimization state is re- 3600 established. 3602 B.2. Implicit Mobility Management 3604 AERO interface neighbors MAY provide a configuration option that 3605 allows them to perform implicit mobility management in which no ND 3606 messaging is used. In that case, the Client only transmits packets 3607 over a single interface at a time, and the neighbor always observes 3608 packets arriving from the Client from the same link-layer source 3609 address. 3611 If the Client's underlying interface address changes (either due to a 3612 readdressing of the original interface or switching to a new 3613 interface) the neighbor immediately updates the neighbor cache entry 3614 for the Client and begins accepting and sending packets according to 3615 the Client's new address. This implicit mobility method applies to 3616 use cases such as cellphones with both WiFi and Cellular interfaces 3617 where only one of the interfaces is active at a given time, and the 3618 Client automatically switches over to the backup interface if the 3619 primary interface fails. 3621 B.3. Direct Underlying Interfaces 3623 When a Client's AERO interface is configured over a Direct interface, 3624 the neighbor at the other end of the Direct link can receive packets 3625 without any encapsulation. In that case, the Client sends packets 3626 over the Direct link according to QoS preferences. If the Direct 3627 interface has the highest QoS preference, then the Client's IP 3628 packets are transmitted directly to the peer without going through an 3629 ANET/INET. If other interfaces have higher QoS preferences, then the 3630 Client's IP packets are transmitted via a different interface, which 3631 may result in the inclusion of Proxys, Servers and Relays in the 3632 communications path. Direct interfaces must be tested periodically 3633 for reachability, e.g., via NUD. 3635 B.4. Operation on AERO Links with /64 ASPs 3637 IPv6 AERO links typically have MSPs that aggregate many candidate 3638 MNPs of length /64 or shorter. However, in some cases it may be 3639 desirable to use AERO over links that have only a /64 MSP. This can 3640 be accommodated by treating all Clients on the AERO link as simple 3641 hosts that receive /128 prefix delegations. 3643 In that case, the Client sends an RS message to the Server the same 3644 as for ordinary AERO links. The Server responds with an RA message 3645 that includes one or more /128 prefixes (i.e., singleton addresses) 3646 that include the /64 MSP prefix along with an interface identifier 3647 portion to be assigned to the Client. The Client and Server then 3648 configure their AERO addresses based on the interface identifier 3649 portions of the /128s (i.e., the lower 64 bits) and not based on the 3650 /64 prefix (i.e., the upper 64 bits). 3652 For example, if the MSP for the host-only IPv6 AERO link is 3653 2001:db8:1000:2000::/64, each Client will receive one or more /128 3654 IPv6 prefix delegations such as 2001:db8:1000:2000::1/128, 3655 2001:db8:1000:2000::2/128, etc. When the Client receives the prefix 3656 delegations, it assigns the AERO addresses fe80::1, fe80::2, etc. to 3657 the AERO interface, and assigns the global IPv6 addresses (i.e., the 3658 /128s) to either the AERO interface or an internal virtual interface 3659 such as a loopback. In this arrangement, the Client conducts route 3660 optimization in the same sense as discussed in Section 3.17. 3662 This specification has applicability for nodes that act as a Client 3663 on an "upstream" AERO link, but also act as a Server on "downstream" 3664 AERO links. More specifically, if the node acts as a Client to 3665 receive a /64 prefix from the upstream AERO link it can then act as a 3666 Server to provision /128s to Clients on downstream AERO links. 3668 B.5. AERO Critical Infrastructure Considerations 3670 AERO Relays can be either Commercial off-the Shelf (COTS) standard IP 3671 routers or virtual machines in the cloud. Relays must be 3672 provisioned, supported and managed by the INET administrative 3673 authority, and connected to the Relays of other INETs via inter- 3674 domain peerings. Cost for purchasing, configuring and managing 3675 Relays is nominal even for very large AERO links. 3677 AERO Servers can be standard dedicated server platforms, but most 3678 often will be deployed as virtual machines in the cloud. The only 3679 requirements for Servers are that they can run the AERO user-level 3680 code and have at least one network interface connection to the INET. 3681 As with Relays, Servers must be provisioned, supported and managed by 3682 the INET administrative authority. Cost for purchasing, configuring 3683 and managing Servers is nominal especially for virtual Servers hosted 3684 in the cloud. 3686 AERO Proxys are most often standard dedicated server platforms with 3687 one network interface connected to the ANET and a second interface 3688 connected to an INET. As with Servers, the only requirements are 3689 that they can run the AERO user-level code and have at least one 3690 interface connection to the INET. Proxys must be provisioned, 3691 supported and managed by the ANET administrative authority. Cost for 3692 purchasing, configuring and managing Proxys is nominal, and borne by 3693 the ANET administrative authority. 3695 AERO Gateways can be any dedicated server or COTS router platform 3696 connected to INETs and/or EUNs. The Gateway joins the SPAN and 3697 engages in eBGP peering with one or more Relays as a stub AS. The 3698 Gateway then injects its MNPs and/or non-MNP prefixes into the BGP 3699 routing system, and provisions the prefixes to its downstream- 3700 attached networks. The Gateway can perform ROS/ROR services the same 3701 as for any Server, and can route between the MNP and non-MNP address 3702 spaces. 3704 B.6. AERO Server Failure Implications 3706 AERO Servers may appear as a single point of failure in the 3707 architecture, but such is not the case since all Servers on the link 3708 provide identical services and loss of a Server does not imply 3709 immediate and/or comprehensive communication failures. Although 3710 Clients typically associate with a single Server at a time, Server 3711 failure is quickly detected and conveyed by Bidirectional Forward 3712 Detection (BFD) and/or proactive NUD allowing Clients to migrate to 3713 new Servers. 3715 If a Server fails, ongoing packet forwarding to Clients will continue 3716 by virtue of the asymmetric neighbor cache entries that have already 3717 been established in route optimization sources (ROSs). If a Client 3718 also experiences mobility events at roughly the same time the Server 3719 fails, unsolicited NA messages may be lost but proxy neighbor cache 3720 entries in the DEPARTED state will ensure that packet forwarding to 3721 the Client's new locations will continue for up to DEPART_TIME 3722 seconds. 3724 If a Client is left without a Server for an extended timeframe (e.g., 3725 greater than REACHABLETIIME seconds) then existing asymmetric 3726 neighbor cache entries will eventually expire and both ongoing and 3727 new communications will fail. The original source will continue to 3728 retransmit until the Client has established a new Server 3729 relationship, after which time continuous communications will resume. 3731 Therefore, providing many Servers on the link with high availability 3732 profiles provides resilience against loss of individual Servers and 3733 assurance that Clients can establish new Server relationships quickly 3734 in event of a Server failure. 3736 B.7. AERO Client / Server Architecture 3738 The AERO architectural model is client / server in the control plane, 3739 with route optimization in the data plane. The same as for common 3740 Internet services, the AERO Client discovers the addresses of AERO 3741 Servers and selects one Server to connect to. The AERO service is 3742 analogous to common Internet services such as google.com, yahoo.com, 3743 cnn.com, etc. However, there is only one AERO service for the link 3744 and all Servers provide identical services. 3746 Common Internet services provide differing strategies for advertising 3747 server addresses to clients. The strategy is conveyed through the 3748 DNS resource records returned in response to name resolution queries. 3749 As of January 2020 Internet-based 'nslookup' services were used to 3750 determine the following: 3752 o When a client resolves the domainname "google.com", the DNS always 3753 returns one A record (i.e., an IPv4 address) and one AAAA record 3754 (i.e., an IPv6 address). The client receives the same addresses 3755 each time it resolves the domainname via the same DNS resolver, 3756 but may receive different addresses when it resolves the 3757 domainname via different DNS resolvers. But, in each case, 3758 exactly one A and one AAAA record are returned. 3760 o When a client resolves the domainname "ietf.org", the DNS always 3761 returns one A record and one AAAA record with the same addresses 3762 regardless of which DNS resolver is used. 3764 o When a client resolves the domainname "yahoo.com", the DNS always 3765 returns a list of 4 A records and 4 AAAA records. Each time the 3766 client resolves the domainname via the same DNS resolver, the same 3767 list of addresses are returned but in randomized order (i.e., 3768 consistent with a DNS round-robin strategy). But, interestingly, 3769 the same addresses are returned (albeit in randomized order) when 3770 the domainname is resolved via different DNS resolvers. 3772 o When a client resolves the domainname "amazon.com", the DNS always 3773 returns a list of 3 A records and no AAAA records. As with 3774 "yahoo.com", the same three A records are returned from any 3775 worldwide Internet connection point in randomized order. 3777 The above example strategies show differing approaches to Internet 3778 resilience and service distribution offered by major Internet 3779 services. The Google approach exposes only a single IPv4 and a 3780 single IPv6 address to clients. Clients can then select whichever IP 3781 protocol version offers the best response, but will always use the 3782 same IP address according to the current Internet connection point. 3783 This means that the IP address offered by the network must lead to a 3784 highly-available server and/or service distribution point. In other 3785 words, resilience is predicated on high availability within the 3786 network and with no client-initiated failovers expected (i.e., it is 3787 all-or-nothing from the client's perspective). However, Google does 3788 provide for worldwide distributed service distribution by virtue of 3789 the fact that each Internet connection point responds with a 3790 different IPv6 and IPv4 address. The IETF approach is like google 3791 (all-or-nothing from the client's perspective), but provides only a 3792 single IPv4 or IPv6 address on a worldwide basis. This means that 3793 the addresses must be made highly-available at the network level with 3794 no client failover possibility, and if there is any worldwide service 3795 distribution it would need to be conducted by a network element that 3796 is reached via the IP address acting as a service distribution point. 3798 In contrast to the Google and IETF philosophies, Yahoo and Amazon 3799 both provide clients with a (short) list of IP addresses with Yahoo 3800 providing both IP protocol versions and Amazon as IPv4-only. The 3801 order of the list is randomized with each name service query 3802 response, with the effect of round-robin load balancing for service 3803 distribution. With a short list of addresses, there is still 3804 expectation that the network will implement high availability for 3805 each address but in case any single address fails the client can 3806 switch over to using a different address. The balance then becomes 3807 one of function in the network vs function in the end system. 3809 The same implications observed for common highly-available services 3810 in the Internet apply also to the AERO client/server architecture. 3811 When an AERO Client connects to one or more ANETs, it discovers one 3812 or more AERO Server addresses through the mechanisms discussed in 3813 earlier sections. Each Server address presumably leads to a fault- 3814 tolerant clustering arrangement such as supported by Linux-HA, 3815 Extended Virtual Synchrony or Paxos. Such an arrangement has 3816 precedence in common Internet service deployments in lightweight 3817 virtual machines without requiring expensive hardware deployment. 3818 Similarly, common Internet service deployments set service IP 3819 addresses on service distribution points that may relay requests to 3820 many different servers. 3822 For AERO, the expectation is that a combination of the Google/IETF 3823 and Yahoo/Amazon philosophies would be employed. The AERO Client 3824 connects to different ANET access points and can receive 1-2 Server 3825 AERO addresses at each point. It then selects one AERO Server 3826 address, and engages in RS/RA exchanges with the same Server from all 3827 ANET connections. The Client remains with this Server unless or 3828 until the Server fails, in which case it can switch over to an 3829 alternate Server. The Client can likewise switch over to a different 3830 Server at any time if there is some reason for it to do so. So, the 3831 AERO expectation is for a balance of function in the network and end 3832 system, with fault tolerance and resilience at both levels. 3834 Appendix C. Change Log 3836 << RFC Editor - remove prior to publication >> 3837 Changes from draft-templin-intarea-6706bis-42 to draft-templin- 3838 intrea-6706bis-43: 3840 o Segment Routing. 3842 Changes from draft-templin-intarea-6706bis-39 to draft-templin- 3843 intrea-6706bis-40: 3845 o Teredo. 3847 Changes from draft-templin-intarea-6706bis-38 to draft-templin- 3848 intrea-6706bis-39: 3850 o Major clrifications and simplifications of SPAN fragmentation/ 3851 reassembly. 3853 o Revised AERO address format to support prefix lengths up to 112. 3855 o New method for forming SPAN Client Prefixes and population in the 3856 routing system. 3858 o Updates RFC4443 to set a new value in the ICMP PTB Code field. 3860 Changes from draft-templin-intarea-6706bis-35 to draft-templin- 3861 intrea-6706bis-36: 3863 o Clients in the open Internet secured using SEND/CGA. 3865 Changes from draft-templin-intarea-6706bis-32 to draft-templin- 3866 intrea-6706bis-33: 3868 o Updated Proxy discussion with "point-to-multipoint" server 3869 coordination 3871 o Significant updates to Address Resolution and NUD to include 3872 correct addresses in messages 3874 o Differentiate between NS(AR) and NS(NUD) as their addresses and 3875 use cases differ. 3877 Changes from draft-templin-intarea-6706bis-30 to draft-templin- 3878 intrea-6706bis-31: 3880 o Added "advisory PTB messages" under FAA SE2025 contract number 3881 DTFAWA-15-D-00030. 3883 Changes from draft-templin-intarea-6706bis-29 to draft-templin- 3884 intrea-6706bis-30: 3886 o Deprecate "primary" concept. Now, RS/RA keepalives are 3887 mainatained over *all* underlying interfaces (i.e., and not just 3888 one primary). 3890 Changes from draft-templin-intarea-6706bis-28 to draft-templin- 3891 intrea-6706bis-29: 3893 o Changed OMNI interface citation to "draft-templin-6man-omni- 3894 interface" 3896 o Changed SPAN Service Prefix to fd80::/10. 3898 o Changed S/TLLAO format to include 'S' bit for ifIndex 3899 corresponding to the underlying interface that is Source of ND 3900 message. 3902 o Updated Path MTU 3904 Changes from draft-templin-intarea-6706bis-27 to draft-templin- 3905 intrea-6706bis-28: 3907 o MTU and fragmentation. 3909 Changes from draft-templin-intarea-6706bis-26 to draft-templin- 3910 intrea-6706bis-27: 3912 o MTU and fragmentation. 3914 o SPAN Service Prefix set to fd00::/10 3916 o Client SPAN addresses defined. 3918 Changes from draft-templin-intarea-6706bis-25 to draft-templin- 3919 intrea-6706bis-26: 3921 o MTU and RA configuration information updated. 3923 Changes from draft-templin-intarea-6706bis-24 to draft-templin- 3924 intrea-6706bis-25: 3926 o Added concept of "primary" to allow for proxyed RS/RA over only 3927 selected underlying interfaces. 3929 o General Cleanup. 3931 Changes from draft-templin-intarea-6706bis-23 to draft-templin- 3932 intrea-6706bis-24: 3934 o OMNI interface spec now a normative reference. 3936 o Use REACHABLE_TIME as the nominal Router Lifetime to return in 3937 RAs. 3939 o General cleanup. 3941 Changes from draft-templin-intarea-6706bis-22 to draft-templin- 3942 intrea-6706bis-23: 3944 o Choice of using either RS/RA or unsolicited NA for old Server 3945 notification. 3947 o General cleanup. 3949 Changes from draft-templin-intarea-6706bis-21 to draft-templin- 3950 intrea-6706bis-22: 3952 o Tightened up text on Proxy. 3954 o Removed unnecessarily restrictive texts. 3956 o General cleanup. 3958 Changes from draft-templin-intarea-6706bis-20 to draft-templin- 3959 intrea-6706bis-21: 3961 o Clarified relationship between OMNI and S/TLLAO ifIndex-tuples. 3963 o Important text in Section 13.15.3 on Servers timing out Clients 3964 that have gone silent without sending a departure notification. 3966 o New text on RS/RA as "hints of forward progress" for proactive 3967 NUD. 3969 Changes from draft-templin-intarea-6706bis-19 to draft-templin- 3970 intrea-6706bis-20: 3972 o Included new route optimization source and destination addressing 3973 strategy. Now, route optimization maintenance uses the address of 3974 the existing Server instead of the data packet destination address 3975 so that less pressure is placed on the BGP routing system 3976 convergence time and Server constancy is supported. 3978 o Included new method for releasing from old MSE without requiring 3979 Client messaging. 3981 o Included references to new OMNI interface spec (including the OMNI 3982 option). 3984 o New appendix on AERO Client/Server architecture. 3986 Changes from draft-templin-intarea-6706bis-18 to draft-templin- 3987 intrea-6706bis-19: 3989 o Changed Proxy/Server keepalives to use "proactive NUD" in a manner 3990 tha paralles BFD 3992 Changes from draft-templin-intarea-6706bis-17 to draft-templin- 3993 intrea-6706bis-18: 3995 o Discuss how AERO option is used in relation to S/TLLAOs 3997 o New text on Bidirectional Forwarding Detection (BFD) 3999 o Cleaned up usage (and non-usage) of unsolicited NAs 4001 o New appendix on Server failures 4003 Changes from draft-templin-intarea-6706bis-15 to draft-templin- 4004 intrea-6706bis-17: 4006 o S/TLLAO now includes multiple link-layer addresses within a single 4007 option instead of requiring multiple options 4009 o New unsolicited NA message to inform the old link that a Client 4010 has moved to a new link 4012 Changes from draft-templin-intarea-6706bis-14 to draft-templin- 4013 intrea-6706bis-15: 4015 o MTU and fragmentation 4017 o New details in movement to new Server 4019 Changes from draft-templin-intarea-6706bis-13 to draft-templin- 4020 intrea-6706bis-14: 4022 o Security based on secured tunnels, ingress filtering, MAP list and 4023 ROS list 4025 Changes from draft-templin-intarea-6706bis-12 to draft-templin- 4026 intrea-6706bis-13: 4028 o New paragraph in Section 3.6 on AERO interface layering over 4029 secured tunnels 4031 o Removed extraneous text in Section 3.7 4033 o Added new detail to the forwarding algorithm in Section 3.9 4035 o Clarified use of fragmentation 4037 o Route optimization now supported for both MNP and non-MNP-based 4038 prefixes 4040 o Relays are now seen as link-layer elements in the architecture. 4042 o Built out multicast section in detail. 4044 o New Appendix on implementation considerations for route 4045 optimization. 4047 Changes from draft-templin-intarea-6706bis-11 to draft-templin- 4048 intrea-6706bis-12: 4050 o Introduced Gateways as a new AERO element for connecting 4051 Correspondent Nodes on INET links 4053 o Introduced terms "Access Network (ANET)" and "Internetwork (INET)" 4055 o Changed "ASP" to "MSP", and "ACP" to "MNP" 4057 o New figure on the relation of Segments to the SPAN and AERO link 4059 o New "S" bit in S/TLLAO to indicate the "Source" S/TLLAO as opposed 4060 to additional S/TLLAOs 4062 o Changed Interface ID for Servers from 255 to 0xffff 4064 o Significant updates to Route Optimization, NUD, and Mobility 4065 Management 4067 o New Section on Multicast 4069 o New Section on AERO Clients in the open Internetwork 4071 o New Section on Operation over multiple AERO links (VLANs over the 4072 SPAN) 4074 o New Sections on DNS considerations and Transition considerations 4075 o 4077 Changes from draft-templin-intarea-6706bis-10 to draft-templin- 4078 intrea-6706bis-11: 4080 o Added The SPAN 4082 Changes from draft-templin-intarea-6706bis-09 to draft-templin- 4083 intrea-6706bis-10: 4085 o Orphaned packets in flight (e.g., when a neighbor cache entry is 4086 in the DEPARTED state) are now forwarded at the link layer instead 4087 of at the network layer. Forwarding at the network layer can 4088 result in routing loops and/or excessive delays of forwarded 4089 packets while the routing system is still reconverging. 4091 o Update route optimization to clarify the unsecured nature of the 4092 first NS used for route discovery 4094 o Many cleanups and clarifications on ND messaging parameters 4096 Changes from draft-templin-intarea-6706bis-08 to draft-templin- 4097 intrea-6706bis-09: 4099 o Changed PRL to "MAP list" 4101 o For neighbor cache entries, changed "static" to "symmetric", and 4102 "dynamic" to "asymmetric" 4104 o Specified Proxy RS/RA exchanges with Servers on behalf of Clients 4106 o Added discussion of unsolicited NAs in Section 3.16, and included 4107 forward reference to Section 3.18 4109 o Added discussion of AERO Clients used as critical infrastructure 4110 elements to connect fixed networks. 4112 o Added network-based VPN under security considerations 4114 Changes from draft-templin-intarea-6706bis-07 to draft-templin- 4115 intrea-6706bis-08: 4117 o New section on AERO-Aware Access Router 4119 Changes from draft-templin-intarea-6706bis-06 to draft-templin- 4120 intrea-6706bis-07: 4122 o Added "R" bit for release of PDs. Now have a full RS/RA service 4123 that can do PD without requiring DHCPv6 messaging over-the-air 4125 o Clarifications on solicited vs unsolicited NAs 4127 o Clarified use of MAX_NEIGHBOR_ADVERTISEMENT for the purpose of 4128 increase reliability 4130 Changes from draft-templin-intarea-6706bis-05 to draft-templin- 4131 intrea-6706bis-06: 4133 o Major re-work and simplification of Route Optimization function 4135 o Added Distributed Mobility Management (DMM) and Mobility Anchor 4136 Point (MAP) terminology 4138 o New section on "AERO Critical Infrastructure Element 4139 Considerations" demonstrating low overall cost for the service 4141 o minor text revisions and deletions 4143 o removed extraneous appendices 4145 Changes from draft-templin-intarea-6706bis-04 to draft-templin- 4146 intrea-6706bis-05: 4148 o New Appendix E on S/TLLAO Extensions for special-purpose links. 4149 Discussed ATN/IPS as example. 4151 o New sentence in introduction to declare appendices as non- 4152 normative. 4154 Changes from draft-templin-intarea-6706bis-03 to draft-templin- 4155 intrea-6706bis-04: 4157 o Added definitions for Potential Router List (PRL) and secure 4158 enclave 4160 o Included text on mapping transport layer port numbers to network 4161 layer DSCP values 4163 o Added reference to DTLS and DMM Distributed Mobility Anchoring 4164 working group document 4166 o Reworked Security Considerations 4168 o Updated references. 4170 Changes from draft-templin-intarea-6706bis-02 to draft-templin- 4171 intrea-6706bis-03: 4173 o Added new section on SEND. 4175 o Clarifications on "AERO Address" section. 4177 o Updated references and added new reference for RFC8086. 4179 o Security considerations updates. 4181 o General text clarifications and cleanup. 4183 Changes from draft-templin-intarea-6706bis-01 to draft-templin- 4184 intrea-6706bis-02: 4186 o Note on encapsulation avoidance in Section 4. 4188 Changes from draft-templin-intarea-6706bis-00 to draft-templin- 4189 intrea-6706bis-01: 4191 o Remove DHCPv6 Server Release procedures that leveraged the old way 4192 Relays used to "route" between Server link-local addresses 4194 o Remove all text relating to Relays needing to do any AERO-specific 4195 operations 4197 o Proxy sends RS and receives RA from Server using SEND. Use CGAs 4198 as source addresses, and destination address of RA reply is to the 4199 AERO address corresponding to the Client's ACP. 4201 o Proxy uses SEND to protect RS and authenticate RA (Client does not 4202 use SEND, but rather relies on subnetwork security. When the 4203 Proxy receives an RS from the Client, it creates a new RS using 4204 its own addresses as the source and uses SEND with CGAs to send a 4205 new RS to the Server. 4207 o Emphasize distributed mobility management 4209 o AERO address-based RS injection of ACP into underlying routing 4210 system. 4212 Changes from draft-templin-aerolink-82 to draft-templin-intarea- 4213 6706bis-00: 4215 o Document use of NUD (NS/NA) for reliable link-layer address 4216 updates as an alternative to unreliable unsolicited NA. 4217 Consistent with Section 7.2.6 of RFC4861. 4219 o Server adds additional layer of encapsulation between outer and 4220 inner headers of NS/NA messages for transmission through Relays 4221 that act as vanilla IPv6 routers. The messages include the AERO 4222 Server Subnet Router Anycast address as the source and the Subnet 4223 Router Anycast address corresponding to the Client's ACP as the 4224 destination. 4226 o Clients use Subnet Router Anycast address as the encapsulation 4227 source address when the access network does not provide a 4228 topologically-fixed address. 4230 Author's Address 4232 Fred L. Templin (editor) 4233 Boeing Research & Technology 4234 P.O. Box 3707 4235 Seattle, WA 98124 4236 USA 4238 Email: fltemplin@acm.org