idnits 2.17.1 draft-templin-intarea-6706bis-35.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 (March 30, 2020) is 1487 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 1004 -- Looks like a reference, but probably isn't: '2' on line 1008 == Missing Reference: 'N' is mentioned on line 1018, but not defined == Unused Reference: 'RFC2474' is defined on line 3055, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-gue-extensions' is defined on line 3125, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-tunnels' is defined on line 3130, but no explicit reference was found in the text == Unused Reference: 'RFC1191' is defined on line 3180, but no explicit reference was found in the text == Unused Reference: 'RFC2764' is defined on line 3205, but no explicit reference was found in the text == Unused Reference: 'RFC2923' is defined on line 3219, but no explicit reference was found in the text == Unused Reference: 'RFC3819' is defined on line 3237, but no explicit reference was found in the text == Unused Reference: 'RFC4607' is defined on line 3292, but no explicit reference was found in the text == Unused Reference: 'RFC4963' is defined on line 3296, but no explicit reference was found in the text == Unused Reference: 'RFC6864' is defined on line 3375, but no explicit reference was found in the text == Unused Reference: 'RFC7269' is defined on line 3379, but no explicit reference was found in the text == Unused Reference: 'RFC8086' is defined on line 3401, but no explicit reference was found in the text == Outdated reference: A later version (-99) exists of draft-templin-6man-omni-interface-07 == 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 (~~), 20 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, March 30, 2020 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: October 1, 2020 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-intarea-6706bis-35 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 1, 2020. 45 Copyright Notice 47 Copyright (c) 2020 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents 52 (https://trustee.ietf.org/license-info) in effect on the date of 53 publication of this document. Please review these documents 54 carefully, as they describe your rights and restrictions with respect 55 to this document. Code Components extracted from this document must 56 include Simplified BSD License text as described in Section 4.e of 57 the Trust Legal Provisions and are provided without warranty as 58 described in the Simplified BSD License. 60 Table of Contents 62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 63 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 64 3. Asymmetric Extended Route Optimization (AERO) . . . . . . . . 10 65 3.1. AERO Link Reference Model . . . . . . . . . . . . . . . . 10 66 3.2. AERO Node Types . . . . . . . . . . . . . . . . . . . . . 12 67 3.3. AERO Routing System . . . . . . . . . . . . . . . . . . . 13 68 3.3.1. IPv4 Compatibility Routing . . . . . . . . . . . . . 15 69 3.4. AERO Addresses . . . . . . . . . . . . . . . . . . . . . 15 70 3.5. Spanning Partitioned AERO Networks (SPAN) . . . . . . . . 16 71 3.5.1. SPAN Compatibility Addressing . . . . . . . . . . . . 20 72 3.5.2. Client SPAN Addresses . . . . . . . . . . . . . . . . 20 73 3.6. AERO Interface Characteristics . . . . . . . . . . . . . 21 74 3.7. AERO Interface Initialization . . . . . . . . . . . . . . 24 75 3.7.1. AERO Server/Gateway Behavior . . . . . . . . . . . . 24 76 3.7.2. AERO Proxy Behavior . . . . . . . . . . . . . . . . . 25 77 3.7.3. AERO Client Behavior . . . . . . . . . . . . . . . . 25 78 3.7.4. AERO Relay Behavior . . . . . . . . . . . . . . . . . 25 79 3.8. AERO Interface Neighbor Cache Maintenance . . . . . . . . 26 80 3.9. AERO Interface Encapsulation and Re-encapsulation . . . . 28 81 3.10. AERO Interface Decapsulation . . . . . . . . . . . . . . 29 82 3.11. AERO Interface Data Origin Authentication . . . . . . . . 29 83 3.12. AERO Interface MTU and Fragmentation . . . . . . . . . . 29 84 3.13. AERO Interface Forwarding Algorithm . . . . . . . . . . . 31 85 3.13.1. Client Forwarding Algorithm . . . . . . . . . . . . 32 86 3.13.2. Proxy Forwarding Algorithm . . . . . . . . . . . . . 33 87 3.13.3. Server/Gateway Forwarding Algorithm . . . . . . . . 34 88 3.13.4. Relay Forwarding Algorithm . . . . . . . . . . . . . 35 89 3.14. AERO Interface Error Handling . . . . . . . . . . . . . . 36 90 3.15. AERO Router Discovery, Prefix Delegation and 91 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 38 92 3.15.1. AERO ND/PD Service Model . . . . . . . . . . . . . . 38 93 3.15.2. AERO Client Behavior . . . . . . . . . . . . . . . . 39 94 3.15.3. AERO Server Behavior . . . . . . . . . . . . . . . . 41 95 3.16. The AERO Proxy . . . . . . . . . . . . . . . . . . . . . 44 96 3.16.1. Detecting and Responding to Server Failures . . . . 46 97 3.16.2. Point-to-Multipoint Server Coordindation . . . . . . 47 98 3.17. AERO Route Optimization . . . . . . . . . . . . . . . . . 47 99 3.17.1. Route Optimization Initiation . . . . . . . . . . . 48 100 3.17.2. Relaying the NS . . . . . . . . . . . . . . . . . . 48 101 3.17.3. Processing the NS and Sending the NA . . . . . . . . 48 102 3.17.4. Relaying the NA . . . . . . . . . . . . . . . . . . 49 103 3.17.5. Processing the NA . . . . . . . . . . . . . . . . . 50 104 3.17.6. Route Optimization Maintenance . . . . . . . . . . . 50 105 3.18. Neighbor Unreachability Detection (NUD) . . . . . . . . . 51 106 3.19. Mobility Management and Quality of Service (QoS) . . . . 52 107 3.19.1. Mobility Update Messaging . . . . . . . . . . . . . 53 108 3.19.2. Announcing Link-Layer Address and/or QoS Preference 109 Changes . . . . . . . . . . . . . . . . . . . . . . 54 110 3.19.3. Bringing New Links Into Service . . . . . . . . . . 54 111 3.19.4. Removing Existing Links from Service . . . . . . . . 54 112 3.19.5. Moving to a New Server . . . . . . . . . . . . . . . 54 113 3.20. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 55 114 3.20.1. Source-Specific Multicast (SSM) . . . . . . . . . . 56 115 3.20.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 57 116 3.20.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 58 117 3.21. Operation over Multiple AERO Links (VLANs) . . . . . . . 58 118 3.22. DNS Considerations . . . . . . . . . . . . . . . . . . . 59 119 3.23. Transition Considerations . . . . . . . . . . . . . . . . 60 120 3.24. Detecting and Reacting to Server and Relay Failures . . . 60 121 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 61 122 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 61 123 6. Security Considerations . . . . . . . . . . . . . . . . . . . 61 124 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 63 125 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 64 126 8.1. Normative References . . . . . . . . . . . . . . . . . . 64 127 8.2. Informative References . . . . . . . . . . . . . . . . . 66 128 Appendix A. AERO Alternate Encapsulations . . . . . . . . . . . 73 129 Appendix B. Non-Normative Considerations . . . . . . . . . . . . 75 130 B.1. Implementation Strategies for Route Optimization . . . . 75 131 B.2. Implicit Mobility Management . . . . . . . . . . . . . . 76 132 B.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 76 133 B.4. AERO Clients on the Open Internetwork . . . . . . . . . . 76 134 B.5. Operation on AERO Links with /64 ASPs . . . . . . . . . . 77 135 B.6. AERO Adaptations for SEcure Neighbor Discovery (SEND) . . 77 136 B.7. AERO Critical Infrastructure Considerations . . . . . . . 78 137 B.8. AERO Server Failure Implications . . . . . . . . . . . . 79 138 B.9. AERO Client / Server Architecture . . . . . . . . . . . . 79 139 Appendix C. Change Log . . . . . . . . . . . . . . . . . . . . . 81 140 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 89 142 1. Introduction 144 Asymmetric Extended Route Optimization (AERO) fulfills the 145 requirements of Distributed Mobility Management (DMM) [RFC7333] and 146 route optimization [RFC5522] for aeronautical networking and other 147 network mobility use cases. AERO is based on a Non-Broadcast, 148 Multiple Access (NBMA) virtual link model known as the AERO link. 149 The AERO link is a virtual overlay configured over one or more 150 underlying Internetworks, and nodes on the link can exchange IP 151 packets via tunneling. Multilink operation allows for increased 152 reliability, bandwidth optimization and traffic path diversity. 154 The AERO service comprises Clients, Proxys, Servers and Gateways that 155 are seen as AERO link neighbors. Each node's AERO interface uses an 156 IPv6 link-local address format (known as the AERO address) that 157 supports operation of the IPv6 Neighbor Discovery (ND) protocol 158 [RFC4861] and links ND to IP forwarding. A node's AERO interface can 159 be configured over multiple underlying interfaces, and may therefore 160 appear as a single interface with multiple link-layer addresses. 161 Each link-layer address is subject to change due to mobility and/or 162 QoS fluctuations, and link-layer address changes are signaled by ND 163 messaging the same as for any IPv6 link. 165 AERO links provide a cloud-based service where mobile nodes may use 166 any Server acting as a Mobility Anchor Point (MAP) and fixed nodes 167 may use any Gateway on the link for efficient communications. Fixed 168 nodes forward packets destined to other AERO nodes to the nearest 169 Gateway, which forwards them through the cloud. A mobile node's 170 initial packets are forwarded through the Server, while direct 171 routing is supported through asymmetric extended route optimization 172 while data packets are flowing. Both unicast and multicast 173 communications are supported, and mobile nodes may efficiently move 174 between locations while maintaining continuous communications with 175 correspondents and without changing their IP Address. 177 AERO Relays are interconnected in a secured private BGP overlay 178 routing instance known as the "SPAN". The SPAN provides a hybrid 179 routing/bridging service to join the underlying Internetworks of 180 multiple disjoint administrative domains into a single unified AERO 181 link. Each AERO link instance is characterized by the set of 182 Mobility Service Prefixes (MSPs) common to all mobile nodes. The 183 link extends to the point where a Gateway/Server is on the optimal 184 route from any correspondent node on the link, and provides a gateway 185 between the underlying Internetwork and the SPAN. To the underlying 186 Internetwork, the Gateway/Server is the source of a route to its MSP, 187 and hence uplink traffic to the mobile node is naturally routed to 188 the nearest Gateway/Server. 190 AERO assumes the use of PIM Sparse Mode in support of multicast 191 communication. In support of Source Specific Multicast (SSM) when a 192 Mobile Node is the source, AERO route optimization ensures that a 193 shortest-path multicast tree is established with provisions for 194 mobility and multilink operation. In all other multicast scenarios 195 there are no AERO dependencies. 197 AERO was designed for aeronautical networking for both manned and 198 unmanned aircraft, where the aircraft is treated as a mobile node 199 that can connect an Internet of Things (IoT). AERO is also 200 applicable to a wide variety of other use cases. For example, it can 201 be used to coordinate the Virtual Private Network (VPN) links of 202 mobile nodes (e.g., cellphones, tablets, laptop computers, etc.) that 203 connect into a home enterprise network via public access networks 204 using services such as OpenVPN [OVPN]. Other applicable use cases 205 are also in scope. 207 The following numbered sections present the AERO specification. The 208 appendices at the end of the document are non-normative. 210 2. Terminology 212 The terminology in the normative references applies; the following 213 terms are defined within the scope of this document: 215 IPv6 Neighbor Discovery (ND) 216 an IPv6 control message service for coordinating neighbor 217 relationships between nodes connected to a common link. AERO 218 interfaces use the ND service specified in [RFC4861]. 220 IPv6 Prefix Delegation (PD) 221 a networking service for delegating IPv6 prefixes to nodes on the 222 link. The nominal PD service is DHCPv6 [RFC8415], however 223 alternate services (e.g., based on ND messaging) are also in scope 224 [I-D.templin-v6ops-pdhost][I-D.templin-6man-dhcpv6-ndopt]. Most 225 notably, a minimal form of PD known as "prefix registration" can 226 be used if the Client knows its prefix in advance and can 227 represent it in the IPv6 source address of an ND message. 229 Access Network (ANET) 230 a node's first-hop data link service network, e.g., a radio access 231 network, cellular service provider network, corporate enterprise 232 network, or the public Internet itself. For secured ANETs, link- 233 layer security services such as IEEE 802.1X and physical-layer 234 security prevent unauthorized access internally while border 235 network-layer security services such as firewalls and proxies 236 prevent unauthorized outside access. 238 ANET interface 239 a node's attachment to a link in an ANET. 241 ANET address 242 an IP address assigned to a node's interface connection to an 243 ANET. 245 Internetwork (INET) 246 a connected IP network topology with a coherent routing and 247 addressing plan and that provides a transit backbone service for 248 ANET end systems. INETs also provide an underlay service over 249 which the AERO virtual link is configured. Example INETs include 250 corporate enterprise networks, aviation networks, and the public 251 Internet itself. When there is no administrative boundary between 252 an ANET and the INET, the ANET and INET are one and the same. 254 INET Partition 255 frequently, INETs such as large corporate enterprise networks are 256 sub-divided internally into separate isolated partitions. Each 257 partition is fully connected internally but disconnected from 258 other partitions, and there is no requirement that separate 259 partitions maintain consistent Internet Protocol and/or addressing 260 plans. (An INET partition is the same as a SPAN segment discussed 261 below.) 263 INET interface 264 a node's attachment to a link in an INET. 266 INET address 267 an IP address assigned to a node's interface connection to an 268 INET. 270 AERO link 271 a Non-Broadcast, Multiple Access (NBMA) tunnel virtual overlay 272 configured over one or more underlying INETs. Nodes on the AERO 273 link appear as single-hop neighbors from the perspective of the 274 virtual overlay even though they may be separated by many 275 underlying INET hops. AERO links may be configured over multiple 276 underlying SPAN segments (see below). 278 AERO interface 279 a node's attachment to an AERO link. Since the addresses assigned 280 to an AERO interface are managed for uniqueness, AERO interfaces 281 do not require Duplicate Address Detection (DAD) and therefore set 282 the administrative variable 'DupAddrDetectTransmits' to zero 283 [RFC4862]. 285 underlying interface 286 an ANET or INET interface over which an AERO interface is 287 configured. 289 AERO address 290 an IPv6 link-local address assigned to an AERO interface and 291 constructed as specified in Section 3.4. 293 base AERO address 294 the lowest-numbered AERO address aggregated by the MNP (see 295 Section 3.4). 297 Mobility Service Prefix (MSP) 298 an IP prefix assigned to the AERO link and from which more- 299 specific Mobile Network Prefixes (MNPs) are derived. 301 Mobile Network Prefix (MNP) 302 an IP prefix allocated from an MSP and delegated to an AERO Client 303 or Gateway. 305 AERO node 306 a node that is connected to an AERO link, or that provides 307 services to other nodes on an AERO link. 309 AERO Client ("Client") 310 an AERO node that connects to one or more ANETs and requests MNP 311 PDs from AERO Servers. The Client assigns a Client AERO address 312 to the AERO interface for use in ND exchanges with other AERO 313 nodes and forwards packets to correspondents according to AERO 314 interface neighbor cache state. 316 AERO Server ("Server") 317 an INET node that configures an AERO interface to provide default 318 forwarding and mobility/multilink services for AERO Clients. The 319 Server assigns an administratively-provisioned AERO address to its 320 AERO interface to support the operation of the ND/PD services, and 321 advertises all of its associated MNPs via BGP peerings with 322 Relays. 324 AERO Gateway ("Gateway") 325 an AERO Server that also provides forwarding services between 326 nodes reached via the AERO link and correspondents on other links. 327 AERO Gateways are provisioned with MNPs (i.e., the same as for an 328 AERO Client) and run a dynamic routing protocol to discover any 329 non-MNP IP routes. In both cases, the Gateway advertises the 330 MSP(s) over INET interfaces, and distributes all of its associated 331 MNPs and non-MNP IP routes via BGP peerings with Relays (i.e., the 332 same as for an AERO Server). 334 AERO Relay ("Relay") 335 a node that provides hybrid routing/bridging services (as well as 336 a security trust anchor) for nodes on an AERO link. As a router, 337 the Relay forwards packets using standard IP forwarding. As a 338 bridge, the Relay forwards packets over the AERO link without 339 decrementing the IPv6 Hop Limit. AERO Relays peer with Servers 340 and other Relays to discover the full set of MNPs for the link as 341 well as any non-MNPs that are reachable via Gateways. 343 AERO Proxy ("Proxy") 344 a node that provides proxying services between Clients in an ANET 345 and Servers in external INETs. The AERO Proxy is a conduit 346 between the ANET and external INETs in the same manner as for 347 common web proxies, and behaves in a similar fashion as for ND 348 proxies [RFC4389]. 350 Spanning Partitioned AERO Networks (SPAN) 351 a means for bridging disjoint INET partitions as segments of a 352 unified AERO link the same as for a bridged campus LAN. The SPAN 353 is a mid-layer IPv6 encapsulation service in the AERO routing 354 system that supports a unified AERO link view for all segments. 355 Each segment in the SPAN is a distinct INET partition. 357 SPAN Service Prefix (SSP) 358 a global or unique local /96 IPv6 prefix assigned to the AERO link 359 to support SPAN services. 361 SPAN Partition Prefix (SPP) 362 a sub-prefix of the SPAN Service Prefix uniquely assigned to a 363 single SPAN segment. 365 SPAN Address 366 a global or unique local IPv6 address taken from a SPAN Partition 367 Prefix and constructed as specified in Section 3.5. SPAN 368 addresses are statelessly derived from AERO addresses, and vice- 369 versa. 371 ingress tunnel endpoint (ITE) 372 an AERO interface endpoint that injects encapsulated packets into 373 an AERO link. 375 egress tunnel endpoint (ETE) 376 an AERO interface endpoint that receives encapsulated packets from 377 an AERO link. 379 link-layer address 380 an IP address used as an encapsulation header source or 381 destination address from the perspective of the AERO interface. 383 When an upper layer protocol (e.g., UDP) is used as part of the 384 encapsulation, the port number is also considered as part of the 385 link-layer address. From the perspective of the AERO interface, 386 the link-layer address is either an INET address for intra-segment 387 encapsulation or a SPAN address for inter-segment encapsulation. 389 network layer address 390 the source or destination address of an encapsulated IP packet 391 presented to the AERO interface. 393 end user network (EUN) 394 an internal virtual or external edge IP network that an AERO 395 Client or Gateway connects to the rest of the network via the AERO 396 interface. The Client/Gateway sees each EUN as a "downstream" 397 network, and sees the AERO interface as the point of attachment to 398 the "upstream" network. 400 Mobile Node (MN) 401 an AERO Client and all of its downstream-attached networks that 402 move together as a single unit, i.e., an end system that connects 403 an Internet of Things. 405 Mobile Router (MR) 406 a MN's on-board router that forwards packets between any 407 downstream-attached networks and the AERO link. 409 Route Optimization Source (ROS) 410 the AERO node nearest the source that initiates route 411 optimization. The ROS may be a Server or Proxy acting on behalf 412 of the source Client. 414 Route Optimization responder (ROR) 415 the AERO node nearest the target destination that responds to 416 route optimization requests. The ROR may be a Server acting on 417 behalf of a target MNP Client, or a Gateway for a non-MNP 418 destination. 420 MAP List 421 a geographically and/or topologically referenced list of AERO 422 addresses of all Servers within the same AERO link. There is a 423 single MAP list for the entire AERO link. 425 ROS List 426 a list of AERO/SPAN-to-INET address mappings of all ROSes within 427 the same SPAN segment. There is a distinct ROS list for each 428 segment. 430 Distributed Mobility Management (DMM) 431 a BGP-based overlay routing service coordinated by Servers and 432 Relays that tracks all Server-to-Client associations. 434 Mobility Service (MS) 435 the collective set of all Servers, Proxys, Relays and Gateways 436 that provide the AERO Service to Clients. 438 Mobility Service Endpoint MSE) 439 an individual Server, Proxy, Relay or Gateway in the Mobility 440 Service. 442 Throughout the document, the simple terms "Client", "Server", 443 "Relay", "Proxy" and "Gateway" refer to "AERO Client", "AERO Server", 444 "AERO Relay", "AERO Proxy" and "AERO Gateway", respectively. 445 Capitalization is used to distinguish these terms from other common 446 Internetworking uses in which they appear without capitalization. 448 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 449 the names of node variables, messages and protocol constants) is used 450 throughout this document. The terms "All-Routers multicast", "All- 451 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 452 anycast" are defined in [RFC4291] (with Link-Local scope assumed). 453 Also, the term "IP" is used to generically refer to either Internet 454 Protocol version, i.e., IPv4 [RFC0791] or IPv6 [RFC8200]. 456 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 457 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 458 "OPTIONAL" in this document are to be interpreted as described in BCP 459 14 [RFC2119][RFC8174] when, and only when, they appear in all 460 capitals, as shown here. 462 3. Asymmetric Extended Route Optimization (AERO) 464 The following sections specify the operation of IP over Asymmetric 465 Extended Route Optimization (AERO) links: 467 3.1. AERO Link Reference Model 468 +----------------+ 469 | AERO Relay R1 | 470 | Nbr: S1, S2, P1| 471 |(X1->S1; X2->S2)| 472 | MSP M1 | 473 +-+---------+--+-+ 474 +--------------+ | Secured | | +--------------+ 475 |AERO Server S1| | tunnels | | |AERO Server S2| 476 | Nbr: C1, R1 +-----+ | +-----+ Nbr: C2, R1 | 477 | default->R1 | | | default->R1 | 478 | X1->C1 | | | X2->C2 | 479 +-------+------+ | +------+-------+ 480 | AERO Link | | 481 X===+===+===================+==)===============+===+===X 482 | | | | 483 +-----+--------+ +--------+--+-----+ +--------+-----+ 484 |AERO Client C1| | AERO Proxy P1 | |AERO Client C2| 485 | Nbr: S1 | |(Proxy Nbr Cache)| | Nbr: S2 | 486 | default->S1 | +--------+--------+ | default->S2 | 487 | MNP X1 | | | MNP X2 | 488 +------+-------+ .--------+------. +-----+--------+ 489 | (- Proxyed Clients -) | 490 .-. `---------------' .-. 491 ,-( _)-. ,-( _)-. 492 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 493 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 494 `-(______)-' +-------+ +-------+ `-(______)-' 496 Figure 1: AERO Link Reference Model 498 Figure 1 presents the AERO link reference model. In this model: 500 o the AERO link is an overlay network service configured over one or 501 more underlying INET partitions which may be managed by different 502 administrative authorities and have incompatible protocols and/or 503 addressing plans. 505 o AERO Relay R1 aggregates Mobility Service Prefix (MSP) M1, 506 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 507 via BGP peerings over secured tunnels to Servers (S1, S2). Relays 508 use the SPAN service to bridge disjoint segments of a partitioned 509 AERO link. 511 o AERO Servers S1 and S2 configure secured tunnels with Relay R1 and 512 also provide mobility, multilink and default router services for 513 their associated Clients C1 and C2. 515 o AERO Clients C1 and C2 associate with Servers S1 and S2, 516 respectively. They receive Mobile Network Prefix (MNP) 517 delegations X1 and X2, and also act as default routers for their 518 associated physical or internal virtual EUNs. Simple hosts H1 and 519 H2 attach to the EUNs served by Clients C1 and C2, respectively. 521 o AERO Proxy P1 configures a secured tunnel with Relay R1 and 522 provides proxy services for AERO Clients in secured enclaves that 523 cannot associate directly with other AERO link neighbors. 525 Each node on the AERO link maintains an AERO interface neighbor cache 526 and an IP forwarding table the same as for any link. Although the 527 figure shows a limited deployment, in common operational practice 528 there will normally be many additional Relays, Servers, Clients and 529 Proxys. 531 3.2. AERO Node Types 533 AERO Relays provide hybrid routing/bridging services (as well as a 534 security trust anchor) for nodes on an AERO link. Relays use 535 standard IPv6 routing to forward packets both within the same INET 536 partitions and between disjoint INET partitions based on a mid-layer 537 IPv6 encapsulation known as the SPAN header. The inner IP layer 538 experiences a virtual bridging service since the inner IP TTL/Hop 539 Limit is not decremented during forwarding. Each Relay also peers 540 with Servers and other Relays in a dynamic routing protocol instance 541 to provide a Distributed Mobility Management (DMM) service for the 542 list of active MNPs (see Section 3.3). Relays present the AERO link 543 as a set of one or more Mobility Service Prefixes (MSPs) but as link- 544 layer devices need not connect directly to the AERO link themselves 545 unless an administrative interface is desired. Relays configure 546 secured tunnels with Servers, Proxys and other Relays; they further 547 maintain IP forwarding table entries for each Mobile Network Prefix 548 (MNP) and any other reachable non-MNP prefixes. 550 AERO Servers provide default forwarding and mobility/multilink 551 services for AERO Client Mobile Nodes (MNs). Each Server also peers 552 with Relays in a dynamic routing protocol instance to advertise its 553 list of associated MNPs (see Section 3.3). Servers facilitate PD 554 exchanges with Clients, where each delegated prefix becomes an MNP 555 taken from an MSP. Servers forward packets between AERO interface 556 neighbors and track each Client's mobility profiles. 558 AERO Clients register their MNPs through PD exchanges with AERO 559 Servers over the AERO link, and distribute the MNPs to nodes on EUNs. 560 A Client may also be co-resident on the same physical or virtual 561 platform as a Server; in that case, the Client and Server behave as a 562 single functional unit. 564 AERO Proxys provide a conduit for ANET AERO Clients to associate with 565 AERO Servers in external INETs. Client and Servers exchange control 566 plane messages via the Proxy acting as a bridge between the ANET/INET 567 boundary. The Proxy forwards data packets between Clients and the 568 AERO link according to forwarding information in the neighbor cache. 569 The Proxy function is specified in Section 3.16. 571 AERO Gateways are Servers that provide forwarding services between 572 the AERO interface and INET/EUN interfaces. Gateways are provisioned 573 with MNPs the same as for an AERO Client, and also run a dynamic 574 routing protocol to discover any non-MNP IP routes. The Gateway 575 advertises the MSP(s) to INETs, and distributes all of its associated 576 MNPs and non-MNP IP routes via BGP peerings with Relays. 578 AERO Relays, Servers, Proxys and Gateways are critical infrastructure 579 elements in fixed (i.e., non-mobile) INET deployments and hence have 580 permanent and unchanging INET addresses. AERO Clients are MNs that 581 connect via ANET interfaces, i.e., their ANET addresses may change 582 when the Client moves to a new ANET connection. 584 3.3. AERO Routing System 586 The AERO routing system comprises a private instance of the Border 587 Gateway Protocol (BGP) [RFC4271] that is coordinated between Relays 588 and Servers and does not interact with either the public Internet BGP 589 routing system or any underlying INET routing systems. 591 In a reference deployment, each Server is configured as an Autonomous 592 System Border Router (ASBR) for a stub Autonomous System (AS) using 593 an AS Number (ASN) that is unique within the BGP instance, and each 594 Server further uses eBGP to peer with one or more Relays but does not 595 peer with other Servers. Each INET of a multi-segment AERO link must 596 include one or more Relays, which peer with the Servers and Proxys 597 within that INET. All Relays within the same INET are members of the 598 same hub AS using a common ASN, and use iBGP to maintain a consistent 599 view of all active MNPs currently in service. The Relays of 600 different INETs peer with one another using eBGP. 602 Relays advertise the AERO link's MSPs and any non-MNP routes to each 603 of their Servers. This means that any aggregated non-MNPs (including 604 "default") are advertised to all Servers. Each Relay configures a 605 black-hole route for each of its MSPs. By black-holing the MSPs, the 606 Relay will maintain forwarding table entries only for the MNPs that 607 are currently active, and packets destined to all other MNPs will 608 correctly incur Destination Unreachable messages due to the black- 609 hole route. In this way, Servers have only partial topology 610 knowledge (i.e., they know only about the MNPs of their directly 611 associated Clients) and they forward all other packets to Relays 612 which have full topology knowledge. 614 Servers maintain a working set of associated MNPs, and dynamically 615 announce new MNPs and withdraw departed MNPs in eBGP updates to 616 Relays. Servers that are configured as Gateways also redistribute 617 non-MNP routes learned from non-AERO interfaces via their eBGP Relay 618 peerings. 620 Clients are expected to remain associated with their current Servers 621 for extended timeframes, however Servers SHOULD selectively suppress 622 updates for impatient Clients that repeatedly associate and 623 disassociate with them in order to dampen routing churn. Servers 624 that are configured as Gateways advertise the MSPs via INET/EUN 625 interfaces, and forward packets between INET/EUN interfaces and the 626 AERO interface using standard IP forwarding. 628 Scaling properties of the AERO routing system are limited by the 629 number of BGP routes that can be carried by Relays. As of 2015, the 630 global public Internet BGP routing system manages more than 500K 631 routes with linear growth and no signs of router resource exhaustion 632 [BGP]. More recent network emulation studies have also shown that a 633 single Relay can accommodate at least 1M dynamically changing BGP 634 routes even on a lightweight virtual machine, i.e., and without 635 requiring high-end dedicated router hardware. 637 Therefore, assuming each Relay can carry 1M or more routes, this 638 means that at least 1M Clients can be serviced by a single set of 639 Relays. A means of increasing scaling would be to assign a different 640 set of Relays for each set of MSPs. In that case, each Server still 641 peers with one or more Relays, but institutes route filters so that 642 BGP updates are only sent to the specific set of Relays that 643 aggregate the MSP. For example, if the MSP for the AERO link is 644 2001:db8::/32, a first set of Relays could service the MSP 645 2001:db8::/40, a second set of Relays could service 646 2001:db8:0100::/40, a third set could service 2001:db8:0200::/40, 647 etc. 649 Assuming up to 1K sets of Relays, the AERO routing system can then 650 accommodate 1B or more MNPs with no additional overhead (for example, 651 it should be possible to service 1B /64 MNPs taken from a /34 MSP and 652 even more for shorter prefixes). In this way, each set of Relays 653 services a specific set of MSPs that they advertise to the native 654 Internetwork routing system, and each Server configures MSP-specific 655 routes that list the correct set of Relays as next hops. This 656 arrangement also allows for natural incremental deployment, and can 657 support small scale initial deployments followed by dynamic 658 deployment of additional Clients, Servers and Relays without 659 disturbing the already-deployed base. 661 Server and Relays can use the Bidirectional Forwarding Detection 662 (BFD) protocol [RFC5880] to quickly detect link failures that don't 663 result in interface state changes, BGP peer failures, and 664 administrative state changes. BFD is important in environments where 665 rapid response to failures is required for routing reconvergence and, 666 hence, communications continuity. 668 A full discussion of the BGP-based routing system used by AERO is 669 found in [I-D.ietf-rtgwg-atn-bgp]. The system provides for 670 Distributed Mobility Management (DMM) per the distributed mobility 671 anchoring architecture [I-D.ietf-dmm-distributed-mobility-anchoring]. 673 3.3.1. IPv4 Compatibility Routing 675 For IPv6 MNPs, the AERO routing system includes ordinary IPv6 routes. 676 For IPv4 MNPs, the AERO routing system includes IPv6 routes based on 677 an IPv4-embedded IPv6 address format discussed in Section 3.5.1. 679 3.4. AERO Addresses 681 A Client's AERO address is an IPv6 link-local address with an 682 interface identifier based on the Client's delegated MNP. Relay, 683 Server and Proxy AERO addresses are assigned from the range fe80::/96 684 and include an administratively-provisioned value in the lower 32 685 bits. 687 For IPv6, Client AERO addresses begin with the prefix fe80::/64 and 688 include in the interface identifier (i.e., the lower 64 bits) the 689 most-significant 64 bits of the Client's IPv6 MNPs. For example, if 690 the AERO Client receives the IPv6 MNP: 2001:db8:1000:2000::/56 it 691 constructs its corresponding AERO address as: 692 fe80::2001:db8:1000:2000. 694 For IPv4, Client AERO addresses are based on an IPv4-mapped IPv6 695 address [RFC4291] formed from an IPv4 MNP and with a prefix length of 696 96 plus the MNP prefix length. For example, for the IPv4 MNP 697 192.0.2.32/28 the IPv4-mapped IPv6 MNP is: 699 0:0:0:0:0:FFFF:192.0.2.16/124 (also written as 700 0:0:0:0:0:FFFF:c000:0210/124) 702 The Client then constructs its AERO address with the prefix fe80::/64 703 and with the lower 64 bits of the IPv4-mapped IPv6 address in the 704 interface identifier as: fe80::FFFF:192.0.2.16. 706 Mobility Service (MS) AERO addresses (used by Relays, Servers, 707 Gateways and Proxys) are allocated from the range fe80::/96, and MUST 708 be managed for uniqueness. The lower 32 bits of the AERO address 709 includes a unique integer value between 1 and 0xfeffffff (e.g., 710 fe80::1, fe80::2, fe80::3, etc., fe80::feff:ffff) as assigned by the 711 administrative authority for the link. If the link spans multiple 712 SPAN segments, the AERO addresses are assigned to each segment in 1x1 713 correspondence with SPAN addresses (see: Section 3.5). The address 714 fe80:: is the IPv6 link-local Subnet-Router anycast address, and the 715 address fe80::ffff:ffff is the "All-AERO-Servers" address. The 716 address range fe80::ff00:0000/104 is reserved for future use. 718 The Client's Subnet-Router anycast address can be statelessly 719 determined from its AERO address by simply transposing the AERO 720 address into the upper N bits of the Anycast address followed by 721 128-N bits of zeroes. For example, for the AERO address 722 fe80::2001:db8:1:2 the Subnet-Router anycast address is 723 2001:db8:1:2::. 725 AERO addresses for mobile node Clients embed a MNP as discussed 726 above, while AERO addresses for non-MNP destinations are constructed 727 in exactly the same way. A Client AERO address therefore encodes 728 either an MNP if the prefix is reached via the SPAN or a non-MNP if 729 the prefix is reached via a Gateway. 731 3.5. Spanning Partitioned AERO Networks (SPAN) 733 An AERO link configured over a single INET appears as a single 734 unified link with a consistent underlying network addressing plan. 735 In that case, all nodes on the link can exchange packets via simple 736 INET encapsulation, since the underlying INET is connected. In 737 common practice, however, an AERO link may be partitioned into 738 multiple "segments", where each segment is a distinct INET 739 potentially managed under a different administrative authority (e.g., 740 as for worldwide aviation service providers such as ARINC, SITA, 741 Inmarsat, etc.). Individual INETs may also themselves be partitioned 742 internally, in which case each internal partition is seen as a 743 separate segment. 745 The addressing plan of each segment is consistent internally but will 746 often bear no relation to the addressing plans of other segments. 747 Each segment is also likely to be separated from others by network 748 security devices (e.g., firewalls, proxies, packet filtering 749 gateways, etc.), and in many cases disjoint segments may not even 750 have any common physical link connections. Therefore, nodes can only 751 be assured of exchanging packets directly with correspondents in the 752 same segment, and not with those in other segments. The only means 753 for joining the segments therefore is through inter-domain peerings 754 between AERO Relays. 756 The same as for traditional campus LANs, multiple AERO link segments 757 can be joined into a single unified link via a virtual bridging 758 service termed the "SPAN". The SPAN performs link-layer packet 759 forwarding between segments (i.e., bridging) without decrementing the 760 network-layer TTL/Hop Limit. The SPAN model is depicted in Figure 2: 762 . . . . . . . . . . . . . . . . . . . . . . . 763 . . 764 . .-(::::::::) . 765 . .-(::::::::::::)-. +-+ . 766 . (:::: Segment A :::)--|R|---+ . 767 . `-(::::::::::::)-' +-+ | . 768 . `-(::::::)-' | . 769 . | . 770 . .-(::::::::) | . 771 . .-(::::::::::::)-. +-+ | . 772 . (:::: Segment B :::)--|R|---+ . 773 . `-(::::::::::::)-' +-+ | . 774 . `-(::::::)-' | . 775 . | . 776 . .-(::::::::) | . 777 . .-(::::::::::::)-. +-+ | . 778 . (:::: Segment C :::)--|R|---+ . 779 . `-(::::::::::::)-' +-+ | . 780 . `-(::::::)-' | . 781 . | . 782 . ..(etc).. x . 783 . . 784 . . 785 . <- AERO Link Bridged by the SPAN -> . 786 . . . . . . . . . . . . . .. . . . . . . . . 788 Figure 2: The SPAN 790 To support the SPAN, AERO links use the Unique Local Address (ULA) 791 prefix fd80::/10 [RFC4193] as the SPAN Service Prefix (SSP). The 792 prefix length intentionally matches the IPv6 link-local prefix 793 (fe80::/10), and enables a simple 1-bit stateless translation between 794 link-local and SPAN prefixes (i.e., bit 7 is '1' for link-local or 795 '0' for SPAN). 797 Each segment in the SPAN assigns a unique sub-prefix of SSP::/96 798 termed a "SPAN Partition Prefix (SPP)". For example, a first segment 799 could assign fd80::1000/116, a second could assign fd80::2000/116, a 800 third could assign fd80::3000/116, etc. The administrative 801 authorities for each segment must therefore coordinate to assure 802 mutually-exclusive SPP assignments, but internal provisioning of the 803 SPP is an independent local consideration for each administrative 804 authority. 806 An administratively-assigned "SPAN address" is an address taken from 807 a SPP and assigned to a Relay, Server, Gateway or Proxy interface. 808 SPAN addresses are formed by simply clearing bit 7 of an 809 administratively-assigned AERO address. For example, if the SPP is 810 fd80::1000/116, the SPAN address formed from the AERO address 811 fe80::1001 is simply fd80::1001. 813 An "INET address" is an address of a node's interface connection to 814 an INET. AERO/SPAN/INET address mappings are maintained as permanent 815 neighbor cache entires as discussed in Section 3.8. 817 AERO Relays serve as bridges to join multiple segments into a unified 818 AERO link over multiple diverse administrative domains. They support 819 the bridging function by first establishing forwarding table entries 820 for their SPPs either via standard BGP routing or static routes. For 821 example, if three Relays ('A', 'B' and 'C') from different segments 822 serviced the SPPs fd80::1000/116, fd80::2000/116 and fd80::3000/116 823 respectively, then the forwarding tables in each Relay are as 824 follows: 826 A: fd80::1000/116->local, fd80::2000/116->B, fd80::3000/116->C 828 B: fd80::1000/116->A, fd80::2000/116->local, fd80::3000/116->C 830 C: fd80::1000/116->A, fd80::2000/116->B, fd80::3000/116->local 832 These forwarding table entries are permanent and never change, since 833 they correspond to fixed infrastructure elements in their respective 834 segments. This provides the basis for a link-layer forwarding 835 service that cannot be disrupted by routing updates due to node 836 mobility. 838 With the SPPs in place in each Relay's forwarding table, control and 839 data packets sent between AERO nodes in different segments can 840 therefore be carried over the SPAN via encapsulation. For example, 841 when a source AERO node in segment A forwards a packet with IPv6 842 address 2001:db8:1:2::1 to a target AERO node in segment C with IPv6 843 address 2001:db8:1000:2000::1, it first encapsulates the packet in a 844 SPAN header with source SPAN address taken from fd80::1000/116 (e.g., 845 fd80::1001) and destination SPAN address taken from fd80::3000/116 846 (e.g., fd80::3001). Next, it encapsulates the SPAN message in an 847 INET header with source address set to its own INET address (e.g., 848 192.0.2.100) and destination set to the INET address of a Relay 849 (e.g., 192.0.2.1). 851 SPAN encapsulation is based on Generic Packet Tunneling in IPv6 852 [RFC2473]; the encapsulation format in the above example is shown in 853 Figure 3: 855 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 856 | INET Header | 857 | src = 192.0.2.100 | 858 | dst = 192.0.2.1 | 859 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 860 | SPAN Header | 861 | src = fd80::1001 | 862 | dst = fd80::3001 | 863 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 864 | Inner IP Header | 865 | src = 2001:db8:1:2::1 | 866 | dst = 2001:db8:1000:2000::1 | 867 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 868 | | 869 ~ ~ 870 ~ Inner Packet Body ~ 871 ~ ~ 872 | | 873 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 875 Figure 3: SPAN Encapsulation 877 In this format, the inner IP header and packet body are the original 878 IP packet, the SPAN header is an IPv6 header prepared according to 879 [RFC2473], and the INET header is prepared according to Section 3.9. 880 A packet is said to be "forwarded/sent into the SPAN" when it is 881 encapsulated as described above then forwarded via a secured tunnel 882 to a neighboring Relay. 884 This gives rise to a routing system that contains both MNP routes 885 that may change dynamically due to regional node mobility and SPAN 886 routes that never change. The Relays can therefore provide link- 887 layer bridging by sending packets into the SPAN instead of network- 888 layer routing according to MNP routes. As a result, opportunities 889 for packet loss due to node mobility between different segments are 890 mitigated. 892 With reference to Figure 3, for a Client's AERO address the SPAN 893 destination address is simply set to the Subnet-Router anycast 894 address. For non-link-local addresses, the destination SPAN address 895 may not be known in advance for the first few packets of a flow sent 896 via the SPAN. In that case, the SPAN destination address is set to 897 the original packet's destination, and the SPAN routing system will 898 direct the packet to the correct SPAN egress node. (In the above 899 example, the SPAN destination address is simply 900 2001:db8:1000:2000::1.) 902 3.5.1. SPAN Compatibility Addressing 904 For IPv4 MNPs, Servers inject a "SPAN Compatibility Prefix (SCP)" 905 that embeds the MNP into the BGP routing system. The SCP begins with 906 the upper 64 bits of the SSP, followed by the constant string 907 "0000:FFFF" followed by the IPv4 MNP. For example, if the MNP is 908 192.0.2.0/24 then the SCP is fd80::FFFF:192.0.2.0/120. 910 This allows for encapsulation of IPv4 packets in IPv6 headers with 911 "SPAN Compatibility Addresses (SCAs)". In this example, the SCA 912 corresponding to the SCP is simply fd80::FFFF:192.0.2.0, and can be 913 used as the SPAN destination address for packets forwarded via the 914 SPAN. This allows for forwarding of initial IPv4 packets over IPv6 915 SPAN routes, followed by route optimization for direct 916 communications. 918 3.5.2. Client SPAN Addresses 920 When an AERO Client or Proxy encapsulates and fragments a packet 921 (see: Section 3.12), it inserts its "Client SPAN Address" as the IPv6 922 source address of the encapsulation header. This is necessary to 923 provide reassemblers with a source address corresponding to the node 924 that actually inserted the fragment header so that the correct 925 Identification value context is provided. 927 The Client SPAN address is formed by simply clearing bit 7 of the 928 Client's AERO address. For example, for the Client AERO address 929 fe80::2001:db8:1:2 the corresponding Client SPAN address is 930 fd80::2001:db8:1:2. 932 Note that the Client's MNP itself (and not the Client SPAN address) 933 is injected into the routing system due to the /64 assumption in the 934 AERO address construction [RFC7421]. Because of the /64 assumption, 935 the most-significant 64 bits of the Client's MNP are written into the 936 least-significant 64 bits of the AERO address. If MNPs longer than 937 /64 are used in the future (i.e., /65 up to /118) the least- 938 significant bits of the MNP would need to be written into bits 10 939 through 63 of the SPAN address, which would render the address format 940 useless for longest-prefix-match. For more details, see Appendix B 941 of [I-D.templin-6man-omni-interface]. 943 3.6. AERO Interface Characteristics 945 AERO interfaces are virtual interfaces configured over one or more 946 underlying interfaces classified as follows: 948 o Native interfaces have global IP addresses that are reachable from 949 any INET correspondent. All Server, Gateway and Relay interfaces 950 are native interfaces, as are INET-facing interfaces of Proxys. 952 o NATed interfaces connect to a private network behind a Network 953 Address Translator (NAT). The NAT does not participate in any 954 AERO control message signaling, but the Server can issue control 955 messages on behalf of the Client. Clients that are behind a NAT 956 are required to send periodic keepalive messages to keep NAT state 957 alive when there are no data packets flowing. If no other 958 periodic messaging service is available, the Client can send RS 959 messages to receive RA replies from its Server(s). 961 o VPNed interfaces use security encapsulation to a Virtual Private 962 Network (VPN) server that also acts as an AERO Server. As with 963 NATed links, the Server can issue control messages on behalf of 964 the Client, but the Client need not send periodic keepalives in 965 addition to those already used to maintain the VPN connection. 967 o Proxyed interfaces connect to an ANET that is separated from the 968 open INET by an AERO Proxy. Unlike NATed and VPNed interfaces, 969 the Proxy can actively issue control messages on behalf of the 970 Client. 972 o Direct interfaces connect a Client directly to a neighbor without 973 crossing any ANET/INET paths. An example is a line-of-sight link 974 between a remote pilot and an unmanned aircraft. 976 AERO interfaces use encapsulation (see: Section 3.9) to exchange 977 packets with AERO link neighbors over Native, NATed or VPNed 978 interfaces. AERO interfaces do not use encapsulation over Proxyed 979 and Direct underlying interfaces. 981 AERO interfaces maintain a neighbor cache for tracking per-neighbor 982 state the same as for any interface. AERO interfaces use ND messages 983 including Router Solicitation (RS), Router Advertisement (RA), 984 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 985 neighbor cache management. 987 AERO interfaces send ND messages with an Overlay Multilink Network 988 Interface (OMNI) option formatted as specified in 989 [I-D.templin-6man-omni-interface]. The OMNI option includes prefix 990 registration information and "ifIndex-tuples" containing link quality 991 information for the AERO interface's underlying interfaces. 993 When encapsulation is used, AERO interface ND messages MAY also 994 include an AERO Source/Target Link-Layer Address Option (S/TLLAO) 995 formatted as shown in Figure 4: 997 0 1 2 3 998 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 1 999 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1000 | Type | Length | ifIndex[1] |V| Reserved[1] | 1001 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1002 ~ Link Layer Address [1] ~ 1003 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1004 | Port Number [1] | ifIndex[2] |V| Reserved[2] | 1005 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1006 ~ Link Layer Address [2] ~ 1007 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1008 | Port Number [2] | ~ 1009 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 1010 ~ ~ 1011 ~ ... ~ 1012 ~ ~ 1013 ~ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1014 ~ | ifIndex[N] |V| Reserved[N] | 1015 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1016 ~ Link Layer Address [N] ~ 1017 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1018 | Port Number [N] | Trailing zero padding | 1019 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1020 | Trailing zero padding (if necessary) | 1021 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1023 Figure 4: AERO Source/Target Link-Layer Address Option (S/TLLAO) 1024 Format 1026 In this format, Type and Length are set the same as specified for S/ 1027 TLLAOs in [RFC4861], with trailing zero padding octets added as 1028 necessary to produce an integral number of 8 octet blocks. The S/ 1029 TLLAO includes N ifIndex-tuples in correspondence to ifIndex-tuples 1030 that appear in the OMNI option. Each ifIndex-tuple includes the 1031 folllowing information: 1033 o ifIndex[i] - the same value as in the corresponding ifIndex-tuple 1034 included in the OMNI option. 1036 o V[i] - a bit that identifies the IP protocol version of the 1037 address found in the Link Layer Address [i] field. The bit is set 1038 to 0 for IPv4 or 1 for IPv6. 1040 o Reserved[i] - MUST encode the value 0 on transmission, and ignored 1041 on reception. 1043 o Link Layer Address [i] - the IPv4 or IPv6 address used as the 1044 encapsulation source address. The field is 4 bytes in length for 1045 IPv4 or 16 bytes in length for IPv6. 1047 o Port Number [i] - the upper layer protocol port number used as the 1048 encapsulation source port, or 0 when no upper layer protocol 1049 encapsulation is used. The field is 2 bytes in length. 1051 If an S/TLLAO is included, any ifIndex-tuples correspond to a proper 1052 subset of the OMNI option ifIndex-tuples. Any S/TLLAO ifIndex-tuple 1053 having an ifIndex value that does not appear in an OMNI option 1054 ifindex-tuple is ignored. If the same ifIndex value appears in 1055 multiple ifIndex-tuples, the first tuple is processed and the 1056 remaining tuples are ignored. Any S/TLLAO ifIndex-tuples can 1057 therefore be viewed as inter-dependent extensions of their 1058 corresponidng OMNI option ifIndex-tuples, i.e., the OMNI option and 1059 S/TLLAO are companions that are interpreted in conjunction with each 1060 other. 1062 A Client's AERO interface may be configured over multiple underlying 1063 interface connections. For example, common mobile handheld devices 1064 have both wireless local area network ("WLAN") and cellular wireless 1065 links. These links are typically used "one at a time" with low-cost 1066 WLAN preferred and highly-available cellular wireless as a standby. 1067 In a more complex example, aircraft frequently have many wireless 1068 data link types (e.g. satellite-based, cellular, terrestrial, air-to- 1069 air directional, etc.) with diverse performance and cost properties. 1071 If a Client's multiple underlying interfaces are used "one at a time" 1072 (i.e., all other interfaces are in standby mode while one interface 1073 is active), then ND message OMNI options include only a single 1074 ifIndex-tuple and set to a constant value. In that case, the Client 1075 would appear to have a single interface but with a dynamically 1076 changing link-layer address. 1078 If the Client has multiple active underlying interfaces, then from 1079 the perspective of ND it would appear to have multiple link-layer 1080 addresses. In that case, ND message OMNI options MAY include 1081 multiple ifIndex-tuples - each with a value that corresponds to a 1082 specific interface. Every ND message need not include all OMNI and/ 1083 or S/TLLAO ifIndex-tuples; for any ifIndex-tuple not included, the 1084 neighbor considers the status as unchanged. 1086 Relay, Server and Proxy AERO interfaces may be configured over one or 1087 more secured tunnel interfaces. The AERO interface configures both 1088 an AERO address and its corresponding SPAN address, while the 1089 underlying secured tunnel interfaces are either unnumbered or 1090 configure the same SPAN address. The AERO interface encapsulates 1091 each IP packet in a SPAN header and presents the packet to the 1092 underlying secured tunnel interface. For Relays that do not 1093 configure an AERO interface, the secured tunnel interfaces themselves 1094 are exposed to the IP layer with each interface configuring the 1095 Relay's SPAN address. Routing protocols such as BGP therefore run 1096 directly over the Relay's secured tunnel interfaces. For nodes that 1097 configure an AERO interface, routing protocols such as BGP run over 1098 the AERO interface but do not employ SPAN encapsulation. Instead, 1099 the AERO interface presents the routing protocol messages directly to 1100 the underlying secured tunnels without applying encapsulation and 1101 while using the SPAN address as the source address. This distinction 1102 must be honored consistently according to each node's configuration 1103 so that the IP forwarding table will associate discovered IP routes 1104 with the correct interface. 1106 3.7. AERO Interface Initialization 1108 AERO Servers, Proxys and Clients configure AERO interfaces as their 1109 point of attachment to the AERO link. AERO nodes assign the MSPs for 1110 the link to their AERO interfaces (i.e., as a "route-to-interface") 1111 to ensure that packets with destination addresses covered by an MNP 1112 not explicitly assigned to a non-AERO interface are directed to the 1113 AERO interface. 1115 AERO interface initialization procedures for Servers, Proxys, Clients 1116 and Relays are discussed in the following sections. 1118 3.7.1. AERO Server/Gateway Behavior 1120 When a Server enables an AERO interface, it assigns AERO/SPAN 1121 addresses and configures permanent neighbor cache entries for 1122 neighbors in the same SPAN segment by consulting the ROS list for the 1123 segment. The Server also configures secured tunnels with one or more 1124 neighboring Relays and engages in a BGP routing protocol session with 1125 each Relay. 1127 The AERO interface provides a single interface abstraction to the IP 1128 layer, but internally comprises multiple secured tunnels as well as 1129 an NBMA nexus for sending encapsulated data packets to AERO interface 1130 neighbors. The Server further configures a service to facilitate ND/ 1131 PD exchanges with AERO Clients and manages per-Client neighbor cache 1132 entries and IP forwarding table entries based on control message 1133 exchanges. 1135 Gateways are simply Servers that run a dynamic routing protocol 1136 between the AERO interface and INET/EUN interfaces (see: 1137 Section 3.3). The Gateway provisions MNPs to networks on the INET/ 1138 EUN interfaces (i.e., the same as a Client would do) and advertises 1139 the MSP(s) for the AERO link over the INET/EUN interfaces. The 1140 Gateway further provides an attachment point of the AERO link to the 1141 non-MNP-based global topology. 1143 3.7.2. AERO Proxy Behavior 1145 When a Proxy enables an AERO interface, it assigns AERO/SPAN 1146 addresses and configures permanent neighbor cache entries the same as 1147 for Servers. The Proxy also configures secured tunnels with one or 1148 more neighboring Relays and maintains per-Client neighbor cache 1149 entries based on control message exchanges. 1151 3.7.3. AERO Client Behavior 1153 When a Client enables an AERO interface, it sends an RS message with 1154 ND/PD parameters over an ANET interface to a Server in the MAP list, 1155 which returns an RA message with corresponding parameters. (The RS/ 1156 RA messages may pass through a Proxy in the case of a Client's 1157 Proxyed interface.) 1159 After the initial ND/PD message exchange, the Client assigns AERO 1160 addresses to the AERO interface based on the delegated prefix(es). 1161 The Client can then register additional ANET interfaces with the 1162 Server by sending an RS message over each ANET interface. 1164 3.7.4. AERO Relay Behavior 1166 AERO Relays need not connect directly to the AERO link, since they 1167 operate as link-layer forwarding devices instead of network layer 1168 routers. Configuration of AERO interfaces on Relays is therefore 1169 OPTIONAL, e.g., if an administrative interface is needed. Relays 1170 configure secured tunnels with Servers, Proxys and other Relays; they 1171 also configure AERO/SPAN addresses and permanent neighbor cache 1172 entries the same as Servers. Relays engage in a BGP routing protocol 1173 session with a subset of the Servers on the local SPAN segment, and 1174 with other Relays on the SPAN (see: Section 3.3). 1176 3.8. AERO Interface Neighbor Cache Maintenance 1178 Each AERO interface maintains a conceptual neighbor cache that 1179 includes an entry for each neighbor it communicates with on the AERO 1180 link per [RFC4861]. AERO interface neighbor cache entries are said 1181 to be one of "permanent", "symmetric", "asymmetric" or "proxy". 1183 Permanent neighbor cache entries are created through explicit 1184 administrative action; they have no timeout values and remain in 1185 place until explicitly deleted. AERO Servers and Proxys maintain 1186 permanent neighbor cache entries for all other Servers and Proxys 1187 within the same SPAN segment. Each entry maintains the mapping 1188 between the neighbor's network-layer AERO address and corresponding 1189 INET address. The list of all permanent neighbor cache entries for 1190 the SPAN segment is maintained in the segment's ROS list. 1192 Symmetric neighbor cache entries are created and maintained through 1193 RS/RA exchanges as specified in Section 3.15, and remain in place for 1194 durations bounded by ND/PD lifetimes. AERO Servers maintain 1195 symmetric neighbor cache entries for each of their associated 1196 Clients, and AERO Clients maintain symmetric neighbor cache entries 1197 for each of their associated Servers. The list of all Servers on the 1198 AERO link is maintained in the link's MAP list. 1200 Asymmetric neighbor cache entries are created or updated based on 1201 route optimization messaging as specified in Section 3.17, and are 1202 garbage-collected when keepalive timers expire. AERO route 1203 optimization sources (ROSs) maintain asymmetric neighbor cache 1204 entries for active targets with lifetimes based on ND messaging 1205 constants. Asymmetric neighbor cache entries are unidirectional 1206 since only the ROS (and not the target) creates an entry. 1208 Proxy neighbor cache entries are created and maintained by AERO 1209 Proxys when they process Client/Server ND/PD exchanges, and remain in 1210 place for durations bounded by ND/PD lifetimes. AERO Proxys maintain 1211 proxy neighbor cache entries for each of their associated Clients. 1212 Proxy neighbor cache entries track the Client state and the address 1213 of the Client's associated Server. 1215 To the list of neighbor cache entry states in Section 7.3.2 of 1216 [RFC4861], Proxy and Server AERO interfaces add an additional state 1217 DEPARTED that applies to symmetric and proxy neighbor cache entries 1218 for Clients that have recently departed. The interface sets a 1219 "DepartTime" variable for the neighbor cache entry to "DEPARTTIME" 1220 seconds. DepartTime is decremented unless a new ND message causes 1221 the state to return to REACHABLE. While a neighbor cache entry is in 1222 the DEPARTED state, packets destined to the target Client are 1223 forwarded to the Client's new location instead of being dropped. 1225 When DepartTime decrements to 0, the neighbor cache entry is deleted. 1226 It is RECOMMENDED that DEPARTTIME be set to the default constant 1227 value REACHABLETIME plus 10 seconds (40 seconds by default) to allow 1228 a window for packets in flight to be delivered while stale route 1229 optimization state may be present. 1231 When a target Server receives an authentic NS message used for route 1232 optimization, it searches for a symmetric neighbor cache entry for 1233 the target Client. The Server then returns a solicited NA message 1234 without creating a neighbor cache entry for the ROS, but creates or 1235 updates a target Client "Report List" entry for the ROS and sets a 1236 "ReportTime" variable for the entry to REPORTTIME seconds. The 1237 Server resets ReportTime when it receives a new authentic NS message, 1238 and otherwise decrements ReportTime while no authentic NS messages 1239 have been received. It is RECOMMENDED that REPORTTIME be set to the 1240 default constant value REACHABLETIME plus 10 seconds (40 seconds by 1241 default) to allow a window for route optimization to converge before 1242 ReportTime decrements below REACHABLETIME. 1244 When the ROS receives a solicited NA message response to its NS 1245 message used for route optimization, it creates or updates an 1246 asymmetric neighbor cache entry for the target network-layer and 1247 link-layer addresses. The ROS then (re)sets ReachableTime for the 1248 neighbor cache entry to REACHABLETIME seconds and uses this value to 1249 determine whether packets can be forwarded directly to the target, 1250 i.e., instead of via a default route. The ROS otherwise decrements 1251 ReachableTime while no further solicited NA messages arrive. It is 1252 RECOMMENDED that REACHABLETIME be set to the default constant value 1253 30 seconds as specified in [RFC4861]. 1255 The ROS also uses the value MAX_UNICAST_SOLICIT to limit the number 1256 of NS keepalives sent when a correspondent may have gone unreachable, 1257 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1258 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1259 to limit the number of unsolicited NAs that can be sent based on a 1260 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1261 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1262 same as specified in [RFC4861]. 1264 Different values for DEPARTTIME, REPORTTIME, REACHABLETIME, 1265 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1266 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1267 different values are chosen, all nodes on the link MUST consistently 1268 configure the same values. Most importantly, DEPARTTIME and 1269 REPORTTIME SHOULD be set to a value that is sufficiently longer than 1270 REACHABLETIME to avoid packet loss due to stale route optimization 1271 state. 1273 3.9. AERO Interface Encapsulation and Re-encapsulation 1275 Client AERO interfaces avoid encapsulation over Direct underlying 1276 interfaces and Proxyed underlying interfaces for which the first-hop 1277 access router is AERO-aware. Other AERO interfaces encapsulate 1278 packets according to whether they are entering the AERO interface 1279 from the network layer or if they are being re-admitted into the same 1280 AERO link they arrived on. This latter form of encapsulation is 1281 known as "re-encapsulation". 1283 For packets entering the AERO interface from the network layer, the 1284 AERO interface copies the "TTL/Hop Limit", "Type of Service/Traffic 1285 Class" [RFC2983], "Flow Label"[RFC6438] (for IPv6) and "Congestion 1286 Experienced" [RFC3168] values in the packet's IP header into the 1287 corresponding fields in the encapsulation header(s). 1289 For packets undergoing re-encapsulation, the AERO interface instead 1290 copies these values from the original encapsulation header into the 1291 new encapsulation header, i.e., the values are transferred between 1292 encapsulation headers and *not* copied from the encapsulated packet's 1293 network-layer header. (Note especially that by copying the TTL/Hop 1294 Limit between encapsulation headers the value will eventually 1295 decrement to 0 if there is a (temporary) routing loop.) For IPv4 1296 encapsulation/re-encapsulation, the AERO interface sets the DF bit as 1297 discussed in Section 3.12. 1299 AERO interfaces configured over INET underlying interfaces 1300 encapsulate each packet in a SPAN header, then encapsulate the 1301 resulting SPAN packet in an INET header according to the next hop 1302 determined in the forwarding algorithm in Section 3.13. If the next 1303 hop is reached via a secured tunnel, the AERO interface uses an INET 1304 encapsulation format specific to the secured tunnel type (see: 1305 Section 6). If the next hop is reached via an unsecured underlying 1306 interface, the AERO interface instead uses Generic UDP Encapsulation 1307 (GUE) [I-D.ietf-intarea-gue] or an alternate minimal encapsulation 1308 format Appendix A. 1310 When GUE encapsulation is used, the AERO interface next sets the UDP 1311 source port to a constant value that it will use in each successive 1312 packet it sends, and sets the UDP length field to the length of the 1313 SPAN packet plus 8 bytes for the UDP header itself plus the length of 1314 the GUE header (or 0 if GUE direct IP encapsulation is used). For 1315 packets sent to a Server or Relay, the AERO interface sets the UDP 1316 destination port to 8060, i.e., the IANA-registered port number for 1317 AERO. For packets sent to a Client, the AERO interface sets the UDP 1318 destination port to the port value stored in the neighbor cache entry 1319 for this Client. The AERO interface then either includes or omits 1320 the UDP checksum according to the GUE specification. 1322 AERO interfaces observes the packet sizing and fragmentation 1323 considerations found in Section 3.12. 1325 3.10. AERO Interface Decapsulation 1327 AERO interfaces decapsulate packets destined either to the AERO node 1328 itself or to a destination reached via an interface other than the 1329 AERO interface the packet was received on. When the encapsulated 1330 packet arrives in multiple fragments, the AERO interface reassembles 1331 as discussed in Section 3.12. Further decapsulation steps are 1332 performed according to the appropriate encapsulation format 1333 specification. 1335 3.11. AERO Interface Data Origin Authentication 1337 AERO nodes employ simple data origin authentication procedures. In 1338 particular: 1340 o AERO Relays, Servers and Proxys accept encapsulated data packets 1341 and control messages received from secured tunnels via the SPAN. 1343 o AERO Servers and Proxys accept encapsulated data packets and NS 1344 messages used for Neighbor Unreachability Detection (NUD) received 1345 from a member of the ROS list. 1347 o AERO Proxys and Clients accept packets that originate from within 1348 the same secured ANET. 1350 o AERO Clients and Gateways accept packets from downstream network 1351 correspondents based on ingress filtering. 1353 AERO nodes silently drop any packets that do not satisfy the above 1354 data origin authentication procedures. Further security 1355 considerations are discussed Section 6. 1357 3.12. AERO Interface MTU and Fragmentation 1359 All IPv6 interfaces are REQUIRED to configure a minimum Maximum 1360 Transmission Unit (MTU) of 1280 bytes [RFC8200]. (IPv4 interfaces 1361 have a smaller minimum MTU [RFC1122], but SHOULD also observe the 1362 IPv6 minimum MTU if possible.) The AERO link therefore MUST forward 1363 IPv6 packets of at least 1280 bytes without generating an IPv6 Path 1364 MTU Discovery (PMTUD) Packet Too Big (PTB) message [RFC8201]. 1366 The AERO interface configures an MTU of 9180 bytes [RFC2492]; the 1367 size is therefore not a reflection of the underlying interface MTUs, 1368 but rather determines the largest packet the AERO interface can 1369 forward or reassemble. 1371 The AERO interface can employ link-layer IPv6 encapsulation and 1372 fragmentation/reassembly per [RFC2473], but its use is OPTIONAL since 1373 correct operation will result in either case. Implementations that 1374 omit link-layer IPv6 fragmentation/reassembly may be more prone to 1375 dropping large packets and returning a PTB, while those that include 1376 it may see improved performance at the expense of including 1377 additional code. In both cases, AERO interface neighbors are 1378 responsible for advertising their willingness to reassemble. 1380 The AERO interface returns internally-generated PTB messages for 1381 packets admitted into the interface that it deems too large for the 1382 outbound underlying interface (e.g., according to underlying 1383 interface performance characteristics, MTU, etc). For all other 1384 packets, the AERO interface performs PMTUD even if the destination 1385 appears to be on the same link since a proxy on the path could return 1386 a PTB message. This ensures that the path MTU is adaptive and 1387 reflects the current path used for a given data flow. 1389 When a Client's AERO interface sends a packet that is no larger than 1390 the MTU of the selected underlying interface, it sends according to 1391 the underlying interface L2 frame format. When the AERO interface 1392 sends a packet that is larger than the underlying interface MTU, it 1393 drops the packet and returns a PTB if the neighbor is not willing to 1394 reassemble. 1396 Otherwise, the AERO interface encapsulates the packet in an IPv6 1397 header per [RFC2473] with source address set to the Client's link- 1398 local address and destination address set to the link-local address 1399 of the next hop. The AERO interface then uses IPv6 fragmentation to 1400 break the encapsulated packet into fragments that are no larger than 1401 the underlying interface MTU and sends the fragments over the 1402 underlying interface. The next hop then reassembles and conveys the 1403 packet toward the final destination. 1405 When a Proxy or Server receives a fragmented or whole packet from the 1406 INET destined to a Client, it must determine whether to forward or 1407 drop and return a PTB (e.g., according to the underlying interface 1408 performance characteristics, MTU, etc). If the Proxy/Server deems 1409 the packet to be of acceptable size, it first reassembles locally (if 1410 necessary) then forwards the packet to the Client. If the 1411 (reassembled) packet is no larger than the underlying interface MTU, 1412 the Proxy/Server forwards according to the underlying interface L2 1413 frame format. If the packet is larger than the MTU, the Proxy/Server 1414 instead uses link-layer encapsulation and IPv6 fragmentation as above 1415 if the Client accepts fragments or drops and returns a PTB otherwise. 1416 The Client then reassembles and discards the encapsulation header, 1417 then forwards the whole packet to the final destination. 1419 When a Proxy, Server or Gateway forwards a Client's packet over the 1420 SPAN, it uses IPv6 encapsulation with its own SPAN address as the 1421 source address and the SPAN address of the next hop as the 1422 destination, then uses fragmentation to break the SPAN-encapsulated 1423 packet into pieces no larger than 1280 bytes. When a Server or 1424 Gateway forwards a SPAN-encapsulated packet to a destination outside 1425 of the AERO link, it first reassembles if necessary. This implies 1426 that Proxys, Servers and Gateways MUST support fragmentation and 1427 reassembly for packet exchanges over the SPAN even though 1428 fragmentation and reassembly is OPTIONAL for Clients. 1430 Applications that cannot tolerate loss due to MTU restrictions SHOULD 1431 avoid sending packets larger than 1280 bytes, since dynamic path 1432 changes can reduce the path MTU at any time. Applications that may 1433 benefit from sending larger packets even though the path MTU may 1434 change dynamically MAY use larger sizes (i.e., up to the AERO 1435 interface MTU). 1437 Note that when a Proxy/Server forwards a fragmented packet received 1438 from the INET to a Client, it reassembles locally first instead of 1439 blindly forwarding fragments directly to the Client to avoid attacks 1440 such as tiny fragments, overlapping fragments, etc. 1442 Note also that the AERO interface can forward large packets via 1443 encapsulation and fragmentation while at the same time returning 1444 advisory PTB messages, e.g., subject to rate limiting. The interface 1445 can therefore continuously forward large packets without loss while 1446 sending advisory messages recommending a smaller size. Even more 1447 appropriately, the receiving OMNI node that performs reassembly can 1448 send advisory PTB messages if reassembly conditions are currently 1449 unfavorable. 1451 3.13. AERO Interface Forwarding Algorithm 1453 IP packets enter a node's AERO interface either from the network 1454 layer (i.e., from a local application or the IP forwarding system) or 1455 from the link layer (i.e., from an AERO interface neighbor). All 1456 packets entering a node's AERO interface first undergo data origin 1457 authentication as discussed in Section 3.11. Packets that satisfy 1458 data origin authentication are processed further, while all others 1459 are dropped silently. 1461 Packets that enter the AERO interface from the network layer are 1462 forwarded to an AERO interface neighbor. Packets that enter the AERO 1463 interface from the link layer are either re-admitted into the AERO 1464 link or forwarded to the network layer where they are subject to 1465 either local delivery or IP forwarding. In all cases, the AERO 1466 interface itself MUST NOT decrement the network layer TTL/Hop-count 1467 since its forwarding actions occur below the network layer. 1469 AERO interfaces may have multiple underlying interfaces and/or 1470 neighbor cache entries for neighbors with multiple ifIndex-tuple 1471 registrations (see Section 3.6). The AERO interface uses each 1472 packet's DSCP value (and/or other traffic discriminators such as port 1473 number) to select an outgoing underlying interface based on the 1474 node's own QoS preferences, and also to select a destination link- 1475 layer address based on the neighbor's underlying interface with the 1476 highest preference. AERO implementations SHOULD allow for QoS 1477 preference values to be modified at runtime through network 1478 management. 1480 If multiple outgoing interfaces and/or neighbor interfaces have a 1481 preference of "high", the AERO node replicates the packet and sends 1482 one copy via each of the (outgoing / neighbor) interface pairs; 1483 otherwise, the node sends a single copy of the packet via an 1484 interface with the highest preference. AERO nodes keep track of 1485 which underlying interfaces are currently "reachable" or 1486 "unreachable", and only use "reachable" interfaces for forwarding 1487 purposes. 1489 The following sections discuss the AERO interface forwarding 1490 algorithms for Clients, Proxys, Servers and Relays. In the following 1491 discussion, a packet's destination address is said to "match" if it 1492 is the same as a cached address, or if it is covered by a cached 1493 prefix (which may be encoded in an AERO address). 1495 3.13.1. Client Forwarding Algorithm 1497 When an IP packet enters a Client's AERO interface from the network 1498 layer the Client searches for an asymmetric neighbor cache entry that 1499 matches the destination. If there is a match, the Client uses one or 1500 more "reachable" neighbor interfaces in the entry for packet 1501 forwarding. If there is no asymmetric neighbor cache entry, the 1502 Client instead forwards the packet toward a Server (the packet is 1503 intercepted by a Proxy if there is a Proxy on the path). The Client 1504 encapuslates the packet in an IPv6 header and fragments if necessary 1505 according to MTU requirements (see: Section 3.12). 1507 When an IP packet enters a Client's AERO interface from the link- 1508 layer, if the destination matches one of the Client's MNPs or link- 1509 local addresses the Client reassembles and decapsulates as necessary 1510 and delivers the (now-unencapsulated) packet to the network layer. 1511 Otherwise, the Client drops the packet and MAY return a network-layer 1512 ICMP Destination Unreachable message subject to rate limiting (see: 1513 Section 3.14). 1515 3.13.2. Proxy Forwarding Algorithm 1517 For control messages originating from or destined to a Client, the 1518 Proxy intercepts the message and updates its proxy neighbor cache 1519 entry for the Client. The Proxy then forwards a (proxyed) copy of 1520 the control message. (For example, the Proxy forwards a proxied 1521 version of a Client's NS/RS message to the target neighbor, and 1522 forwards a proxied version of the NA/RA reply to the Client.) 1524 When the Proxy receives a data packet from a Client within the ANET, 1525 it first reassembles and decapsulates if the packet was a link-local 1526 fragment. The Proxy next inserts a SPAN header with source address 1527 set to the Proxy's SPAN address and destination address set to the 1528 SPAN address of the next hop. The Proxy then fragments the SPAN 1529 packet if necessary into fragments no larger than 1280 bytes, then 1530 searches for an asymmetric neighbor cache entry that matches the 1531 destination and forwards the fragments as follows: 1533 o if the destination matches an asymmetric neighbor cache entry, the 1534 Proxy uses one or more "reachable" neighbor interfaces in the 1535 entry for packet forwarding via encapsulation. If the neighbor 1536 interface is in the same SPAN segment, the Proxy forwards the 1537 packet directly to the neighbor; otherwise, it forwards the packet 1538 to a Relay. 1540 o else, the Proxy encapsulates and forwards the packet to a Relay 1541 while using the packet's destination address as the SPAN 1542 destination address. (If the destination is an AERO address, the 1543 Proxy instead uses the corresponding Subnet-Router anycast address 1544 for Client AERO addresses and the SPAN address for 1545 administratively-provisioned AERO addresses.). 1547 When the Proxy receives an encapsulated data packet from an INET 1548 neighbor or from a secured tunnel from a Relay, it accepts the packet 1549 only if data origin authentication succeeds and if there is a proxy 1550 neighbor cache entry that matches the inner destination. Next, if 1551 the packet is a SPAN fragment the Proxy adds the fragment to the 1552 reassembly buffer. The Proxy then reassembles the packet (if 1553 necessary) and continues processing. 1555 Next if reassembly is complete and the neighbor cache state is 1556 REACHABLE, the Proxy either drops and returns a PTB (see: 1557 Section 3.12) or forwards the packet to the Client while performing 1558 link-local encapsulation and re-fragmentation to the ANET MTU size if 1559 necessary. If the neighbor cache entry state is DEPARTED, the Proxy 1560 instead changes the SPAN destination address to the address of the 1561 new Server and forwards it to a Relay while performing re- 1562 fragmentation to 1280 bytes if necessary. 1564 When the Proxy forwards a SPAN packet to a REACHABLE Client for which 1565 the packet is no larger than the ANET MTU, it decapsulates the SPAN 1566 header first and forwards the (unencapsulated) packet to the Client 1567 to avoid the unnecessary overhead for carrying the SPAN header. 1569 3.13.3. Server/Gateway Forwarding Algorithm 1571 For control messages destined to a target Client's AERO address that 1572 are received from a secured tunnel, the Server intercepts the message 1573 and sends an appropriate response on behalf of the Client. (For 1574 example, the Server sends an NA message reply in response to an NS 1575 message directed to one of its associated Clients.) If the Client's 1576 neighbor cache entry is in the DEPARTED state, however, the Server 1577 instead forwards the packet to the Client's new Server as discussed 1578 in Section 3.19. 1580 When the Server receives an encapsulated data packet from an INET 1581 neighbor or from a secured tunnel, it accepts the packet only if data 1582 origin authentication succeeds. If the SPAN destination address is 1583 its own address, the Server continues processing as follows: 1585 o if the destination matches a symmetric neighbor cache entry in the 1586 REACHABLE state the Server prepares the packet for forwarding to 1587 the destination Client. For the Client's Proxyed interfaces, the 1588 Server changes the SPAN destination address to the address of the 1589 Proxy and forwards the packet to the Proxy. For the Client's 1590 other interfaces, the Server reassembles then either drops and 1591 returns a PTB (see: Section 3.12) or forwards the packet (while 1592 re-fragmenting if necessary) using SPAN encapsulation for the 1593 Client's Native, NATed or VPNed interfaces, or no encapsulation 1594 for Direct interfaces. 1596 o else, if the destination matches a symmetric neighbor cache entry 1597 in the DEPARETED state the Server re-encapsulates the packet and 1598 forwards it using the SPAN address of the Client's new Server as 1599 the destination. 1601 o else, if the destination matches an asymmetric neighbor cache 1602 entry, the Server uses one or more "reachable" neighbor interfaces 1603 in the entry for packet forwarding via the local INET if the 1604 neighbor is in the same SPAN segment or via a Relay otherwise. 1606 o else, if the destination is an AERO address that is not assigned 1607 on the AERO interface the Server drops the packet. 1609 o else, the Server (acting as a Gateway) reassembles if necessary, 1610 decapsulates the packet and releases it to the network layer for 1611 local delivery or IP forwarding. Based on the information in the 1612 forwarding table, the network layer may return the packet to the 1613 same AERO interface in which case further processing occurs as 1614 below. (Note that this arrangement accommodates common 1615 implementations in which the IP forwarding table is not accessible 1616 from within the AERO interface. If the AERO interface can 1617 directly access the IP forwarding table (such as for in-kernel 1618 implementations) the forwarding table lookup can instead be 1619 performed internally from within the AERO interface itself.) 1621 When the Server's AERO interface receives a data packet from the 1622 network layer or from a NATed/VPNed/Direct Client, it performs SPAN 1623 encapsualtion and fragmentation if necessary, then processes the 1624 packet according to the network-layer destination address as follows: 1626 o if the destination matches a symmetric or asymmetric neighbor 1627 cache entry the Server processes the packet as above. 1629 o else, the Server encapsulates the packet and forwards it to a 1630 Relay. For administratively-assigned AERO address destinations, 1631 the Server uses the SPAN address corresponding to the destination 1632 as the SPAN destination address. For Client AERO address 1633 destinations, the Server uses the Subnet-Router anycast address 1634 corresponding to the destination as the SPAN destination address. 1635 For all others, the Server uses the packet's destination IP 1636 address as the SPAN destination address. 1638 3.13.4. Relay Forwarding Algorithm 1640 Relays forward packets over secured tunnels the same as any IP 1641 router. When the Relay receives an encapsulated packet via a secured 1642 tunnel, it removes the INET header and searches for a forwarding 1643 table entry that matches the destination address in the next header. 1644 The Relay then processes the packet as follows: 1646 o if the destination matches one of the Relay's own addresses, the 1647 Relay submits the packet for local delivery. 1649 o else, if the destination matches a forwarding table entry the 1650 Relay forwards the packet via a secured tunnel to the next hop. 1651 If the destination matches an MSP without matching an MNP, 1652 however, the Relay instead drops the packet and returns an ICMP 1653 Destination Unreachable message subject to rate limiting (see: 1654 Section 3.14). 1656 o else, the Relay drops the packet and returns an ICMP Destination 1657 Unreachable as above. 1659 As for any IP router, the Relay decrements the TTL/Hop Limit when it 1660 forwards the packet. Therefore, only the Hop Limit in the SPAN 1661 header is decremented, and not the TTL/Hop Limit in the inner packet 1662 header. 1664 3.14. AERO Interface Error Handling 1666 When an AERO node admits a packet into the AERO interface, it may 1667 receive link-layer or network-layer error indications. 1669 A link-layer error indication is an ICMP error message generated by a 1670 router in the INET on the path to the neighbor or by the neighbor 1671 itself. The message includes an IP header with the address of the 1672 node that generated the error as the source address and with the 1673 link-layer address of the AERO node as the destination address. 1675 The IP header is followed by an ICMP header that includes an error 1676 Type, Code and Checksum. Valid type values include "Destination 1677 Unreachable", "Time Exceeded" and "Parameter Problem" 1678 [RFC0792][RFC4443]. (AERO interfaces ignore all link-layer IPv4 1679 "Fragmentation Needed" and IPv6 "Packet Too Big" messages since they 1680 only emit packets that are guaranteed to be no larger than the IP 1681 minimum link MTU as discussed in Section 3.12.) 1683 The ICMP header is followed by the leading portion of the packet that 1684 generated the error, also known as the "packet-in-error". For 1685 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1686 much of invoking packet as possible without the ICMPv6 packet 1687 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1688 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1689 "Internet Header + 64 bits of Original Data Datagram", however 1690 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1691 ICMP datagram SHOULD contain as much of the original datagram as 1692 possible without the length of the ICMP datagram exceeding 576 1693 bytes". 1695 The link-layer error message format is shown in Figure 5 (where, "L2" 1696 and "L3" refer to link-layer and network-layer, respectively): 1698 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1699 ~ ~ 1700 | L2 IP Header of | 1701 | error message | 1702 ~ ~ 1703 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1704 | L2 ICMP Header | 1705 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1706 ~ ~ P 1707 | IP and other encapsulation | a 1708 | headers of original L3 packet | c 1709 ~ ~ k 1710 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1711 ~ ~ t 1712 | IP header of | 1713 | original L3 packet | i 1714 ~ ~ n 1715 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1716 ~ ~ e 1717 | Upper layer headers and | r 1718 | leading portion of body | r 1719 | of the original L3 packet | o 1720 ~ ~ r 1721 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1723 Figure 5: AERO Interface Link-Layer Error Message Format 1725 The AERO node rules for processing these link-layer error messages 1726 are as follows: 1728 o When an AERO node receives a link-layer Parameter Problem message, 1729 it processes the message the same as described as for ordinary 1730 ICMP errors in the normative references [RFC0792][RFC4443]. 1732 o When an AERO node receives persistent link-layer Time Exceeded 1733 messages, the IP ID field may be wrapping before earlier fragments 1734 awaiting reassembly have been processed. In that case, the node 1735 should begin including integrity checks and/or institute rate 1736 limits for subsequent packets. 1738 o When an AERO node receives persistent link-layer Destination 1739 Unreachable messages in response to encapsulated packets that it 1740 sends to one of its asymmetric neighbor correspondents, the node 1741 should process the message as an indication that a path may be 1742 failing, and optionally initiate NUD over that path. If it 1743 receives Destination Unreachable messages over multiple paths, the 1744 node should allow future packets destined to the correspondent to 1745 flow through a default route and re-initiate route optimization. 1747 o When an AERO Client receives persistent link-layer Destination 1748 Unreachable messages in response to encapsulated packets that it 1749 sends to one of its symmetric neighbor Servers, the Client should 1750 mark the path as unusable and use another path. If it receives 1751 Destination Unreachable messages on many or all paths, the Client 1752 should associate with a new Server and release its association 1753 with the old Server as specified in Section 3.19.5. 1755 o When an AERO Server receives persistent link-layer Destination 1756 Unreachable messages in response to encapsulated packets that it 1757 sends to one of its symmetric neighbor Clients, the Server should 1758 mark the underlying path as unusable and use another underlying 1759 path. 1761 o When an AERO Server or Proxy receives link-layer Destination 1762 Unreachable messages in response to an encapsulated packet that it 1763 sends to one of its permanent neighbors, it treats the messages as 1764 an indication that the path to the neighbor may be failing. 1765 However, the dynamic routing protocol should soon reconverge and 1766 correct the temporary outage. 1768 When an AERO Relay receives a packet for which the network-layer 1769 destination address is covered by an MSP, if there is no more- 1770 specific routing information for the destination the Relay drops the 1771 packet and returns a network-layer Destination Unreachable message 1772 subject to rate limiting. The Relay writes the network-layer source 1773 address of the original packet as the destination address and uses 1774 one of its non link-local addresses as the source address of the 1775 message. 1777 When an AERO node receives an encapsulated packet for which the 1778 reassembly buffer it too small, it drops the packet and returns a 1779 network-layer Packet Too Big (PTB) message. The node first writes 1780 the MRU value into the PTB message MTU field, writes the network- 1781 layer source address of the original packet as the destination 1782 address and writes one of its non link-local addresses as the source 1783 address. 1785 3.15. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1787 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1788 coordinated as discussed in the following Sections. 1790 3.15.1. AERO ND/PD Service Model 1792 Each AERO Server on the link configures a PD service to facilitate 1793 Client requests. Each Server is provisioned with a database of MNP- 1794 to-Client ID mappings for all Clients enrolled in the AERO service, 1795 as well as any information necessary to authenticate each Client. 1796 The Client database is maintained by a central administrative 1797 authority for the AERO link and securely distributed to all Servers, 1798 e.g., via the Lightweight Directory Access Protocol (LDAP) [RFC4511], 1799 via static configuration, etc. Clients receive the same service 1800 regardless of the Servers they select. 1802 AERO Clients and Servers use ND messages to maintain neighbor cache 1803 entries. AERO Servers configure their AERO interfaces as advertising 1804 NBMA interfaces, and therefore send unicast RA messages with a short 1805 Router Lifetime value (e.g., REACHABLETIME seconds) in response to a 1806 Client's RS message. Thereafter, Clients send additional RS messages 1807 to keep Server state alive. 1809 AERO Clients and Servers include PD parameters in RS/RA messages (see 1810 [I-D.templin-6man-dhcpv6-ndopt] for ND/PD alternatives). The unified 1811 ND/PD messages are exchanged between Client and Server according to 1812 the prefix management schedule required by the PD service. If the 1813 Client knows its MNP in advance, it can instead employ prefix 1814 registration by including its AERO address as the source address of 1815 an RS message and with an OMNI option with valid prefix registration 1816 information for the MNP. If the Server (and Proxy) accept the 1817 Client's MNP assertion, they inject the prefix into the routing 1818 system and establish the necessary neighbor cache state. 1820 The following sections specify the Client and Server behavior. 1822 3.15.2. AERO Client Behavior 1824 AERO Clients discover the addresses of Servers in a similar manner as 1825 described in [RFC5214]. Discovery methods include static 1826 configuration (e.g., from a flat-file map of Server addresses and 1827 locations), or through an automated means such as Domain Name System 1828 (DNS) name resolution [RFC1035]. Alternatively, the Client can 1829 discover Server addresses through a layer 2 data link login exchange, 1830 or through a unicast RA response to a multicast/anycast RS as 1831 described below. In the absence of other information, the Client can 1832 resolve the DNS Fully-Qualified Domain Name (FQDN) 1833 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 1834 text string and "[domainname]" is a DNS suffix for the AERO link 1835 (e.g., "example.com"). 1837 To associate with a Server, the Client acts as a requesting router to 1838 request MNPs. The Client prepares an RS message with PD parameters 1839 and includes a Nonce and Timestamp option if the Client needs to 1840 correlate RA replies. If the Client already knows the Server's AERO 1841 address, it includes the AERO address as the network-layer 1842 destination address; otherwise, it includes the link-scoped All- 1843 Routers multicast (ff02::2) or Subnet-Router anycast (fe80::) address 1844 as the network-layer destination. If the Client already knows its 1845 own AERO address, it uses the AERO address as the network-layer 1846 source address; otherwise, it uses the unspecified IPv6 address 1847 (::/128) as the network-layer source address. 1849 The Client next includes an OMNI option in the RS message to register 1850 its link-layer information with the Server. The Client sets the OMNI 1851 option prefix registration information according to the MNP, and 1852 includes an ifIndex-tuple with S set to '1' corresponding to the 1853 underlying interface over which the Client will send the RS message. 1854 The Client MAY include additional ifIndex-tuples specific to other 1855 underlying interfaces. The Client MAY also include an SLLAO with a 1856 link-layer address corresponding to the OMNI option ifIndex-tuple 1857 with S set to '1'. 1859 The Client then sends the RS message (either directly via Direct 1860 interfaces, via INET encapsulation for NATed interfaces, via a VPN 1861 for VPNed interfaces, via a Proxy for proxyed interfaces or via a 1862 Relay for native interfaces) and waits for an RA message reply (see 1863 Section 3.15.3). The Client retries up to MAX_RTR_SOLICITATIONS 1864 times until an RA is received. If the Client receives no RAs, or if 1865 it receives an RA with Router Lifetime set to 0, the Client SHOULD 1866 abandon this Server and try another Server. Otherwise, the Client 1867 processes the PD information found in the RA message. 1869 Next, the Client creates a symmetric neighbor cache entry with the 1870 Server's AERO address as the network-layer address and the Server's 1871 encapsulation and/or link-layer addresses as the link-layer address. 1872 The Client records the RA Router Lifetime field value in the neighbor 1873 cache entry as the time for which the Server has committed to 1874 maintaining the MNP in the routing system via this underlying 1875 interface, and caches the other RA configuration information 1876 including Cur Hop Limit, M and O flags, Reachable Time and Retrans 1877 Timer. The Client then autoconfigures AERO addresses for each of the 1878 delegated MNPs and assigns them to the AERO interface. The Client 1879 also caches any MSPs included in Route Information Options (RIOs) 1880 [RFC4191] as MSPs to associate with the AERO link, and assigns the 1881 MTU value in the MTU option to the underlying interface. 1883 The Client then registers additional underlying interfaces with the 1884 Server by sending RS messages via each additional interface. The RS 1885 messages include the same parameters as for the initial RS/RA 1886 exchange, but with destination address set to the Server's AERO 1887 address. 1889 Following autoconfiguration, the Client sub-delegates the MNPs to its 1890 attached EUNs and/or the Client's own internal virtual interfaces as 1891 described in [I-D.templin-v6ops-pdhost] to support the Client's 1892 downstream attached "Internet of Things (IoT)". The Client 1893 subsequently sends additional RS messages over each underlying 1894 interface before the Router Lifetime received for that interface 1895 expires. 1897 After the Client registers its underlying interfaces, it may wish to 1898 change one or more registrations, e.g., if an interface changes 1899 address or becomes unavailable, if QoS preferences change, etc. To 1900 do so, the Client prepares an RS message to send over any available 1901 underlying interface. The RS includes an OMNI option with prefix 1902 registration information specific to its MNP, with an ifIndex-tuple 1903 specific to the selected underlying interface with S set to '1', and 1904 with any additional ifIndex-tuples specific to other underlying 1905 interfaces. The Client includes fresh ifIndex-tuple values to update 1906 the Server's neighbor cache entry. When the Client receives the 1907 Server's RA response, it has assurance that the Server has been 1908 updated with the new information. 1910 If the Client wishes to discontinue use of a Server it issues an RS 1911 message over any underlying interface with an OMNI option with a 1912 prefix release indication. When the Server processes the message, it 1913 releases the MNP, sets the symmetric neighbor cache entry state for 1914 the Client to DEPARTED and returns an RA reply with Router Lifetime 1915 set to 0. After a short delay (e.g., 2 seconds), the Server 1916 withdraws the MNP from the routing system. 1918 3.15.3. AERO Server Behavior 1920 AERO Servers act as IP routers and support a PD service for Clients. 1921 Servers arrange to add their AERO addresses to a static map of Server 1922 addresses for the link and/or the DNS resource records for the FQDN 1923 "linkupnetworks.[domainname]" before entering service. Server 1924 addresses should be geographically and/or topologically referenced, 1925 and made available for discovery by Clients on the AERO link. 1927 When a Server receives a prospective Client's RS message on its AERO 1928 interface, it SHOULD return an immediate RA reply with Router 1929 Lifetime set to 0 if it is currently too busy or otherwise unable to 1930 service the Client. Otherwise, the Server authenticates the RS 1931 message and processes the PD parameters. The Server first determines 1932 the correct MNPs to delegate to the Client by searching the Client 1933 database. When the Server delegates the MNPs, it also creates a 1934 forwarding table entry for each MNP so that the MNPs are propagated 1935 into the routing system (see: Section 3.3). For IPv6, the Server 1936 creates an IPv6 forwarding table entry for each MNP. For IPv4, the 1937 Server creates an IPv6 forwarding table entry with the SPAN 1938 Compatibility Prefix (SCP) corresponding to the IPv4 address. 1940 The Server next creates a symmetric neighbor cache entry for the 1941 Client using the base AERO address as the network-layer address and 1942 with lifetime set to no more than the smallest PD lifetime. Next, 1943 the Server updates the neighbor cache entry by recording the 1944 information in each ifIndex-tuple in the RS OMNI option. The Server 1945 also records the actual SPAN/INET addresses in the neighbor cache 1946 entry. If an SLLAO was present, the Server also compares the SLLAO 1947 address information for the first ifIndex-tuple with the SPAN/INET 1948 information to determine if there is a NAT on the path. 1950 Next, the Server prepares an RA message using its AERO address as the 1951 network-layer source address and the network-layer source address of 1952 the RS message as the network-layer destination address. The Server 1953 sets the Router Lifetime to the time for which it will maintain both 1954 this underlying interface individually and the symmetric neighbor 1955 cache entry as a whole. The Server also sets Cur Hop Limit, M and O 1956 flags, Reachable Time and Retrans Timer to values appropriate for the 1957 AERO link. The Server includes the delegated MNPs, any other PD 1958 parameters and an OMNI option with no ifIndex-tuples. The Server 1959 then includes one or more RIOs that encode the MSPs for the AERO 1960 link, plus an MTU option (see Section 3.12). The Server finally 1961 forwards the message to the Client using SPAN/INET, INET, or NULL 1962 encapsulation as necessary. 1964 After the initial RS/RA exchange, the Server maintains a 1965 ReachableTime timer for each of the Client's underlying interfaces 1966 individually (and for the Client's symmetric neighbor cache entry 1967 collectively) set to expire after Router Lifetime seconds. If the 1968 Client (or Proxy) issues additional RS messages, the Server sends an 1969 RA response and resets ReachableTime. If the Server receives an ND 1970 message with PD release indication it sets the Client's symmetric 1971 neighbor cache entry to the DEPARTED state and withdraws the MNP from 1972 the routing system after a short delay (e.g., 2 seconds). If 1973 ReachableTime expires before a new RS is received on an individual 1974 underlying interface, the Server marks the interface as DOWN. If 1975 ReachableTime expires before any new RS is received on any individual 1976 underlying interface, the Server deletes the neighbor cache entry and 1977 withdraws the MNP without delay. 1979 The Server processes any ND/PD messages pertaining to the Client and 1980 returns an NA/RA reply in response to solicitations. The Server may 1981 also issue unsolicited RA messages, e.g., with PD reconfigure 1982 parameters to cause the Client to renegotiate its PDs, with Router 1983 Lifetime set to 0 if it can no longer service this Client, etc. 1984 Finally, If the symmetric neighbor cache entry is in the DEPARTED 1985 state, the Server deletes the entry after DepartTime expires. 1987 Note: Clients SHOULD notify former Servers of their departures, but 1988 Servers are responsible for expiring neighbor cache entries and 1989 withdrawing routes even if no departure notification is received 1990 (e.g., if the Client leaves the network unexpectedly). Servers 1991 SHOULD therefore set Router Lifetime to REACHABLETIME seconds in 1992 solicited RA messages to minimize persistent stale cache information 1993 in the absence of Client departure notifications. A short Router 1994 Lifetime also ensures that proactive Client/Server RS/RA messaging 1995 will keep any NAT state alive (see above). 1997 Note: All Servers on an AERO link MUST advertise consistent values in 1998 the RA Cur Hop Limit, M and O flags, Reachable Time and Retrans Timer 1999 fields the same as for any link, since unpredictable behavior could 2000 result if different Servers on the same link advertised different 2001 values. 2003 3.15.3.1. Lightweight DHCPv6 Relay Agent (LDRA) 2005 When DHCPv6 is used as the ND/PD service back end, AERO Clients and 2006 Servers are always on the same link (i.e., the AERO link) from the 2007 perspective of DHCPv6. However, in some implementations the DHCPv6 2008 server and ND function may be located in separate modules. In that 2009 case, the Server's AERO interface module can act as a Lightweight 2010 DHCPv6 Relay Agent (LDRA)[RFC6221] to relay PD messages to and from 2011 the DHCPv6 server module. 2013 When the LDRA receives an authentic RS message, it extracts the PD 2014 message parameters and uses them to construct an IPv6/UDP/DHCPv6 2015 message. It sets the IPv6 source address to the source address of 2016 the RS message, sets the IPv6 destination address to 2017 'All_DHCP_Relay_Agents_and_Servers' and sets the UDP fields to values 2018 that will be understood by the DHCPv6 server. 2020 The LDRA then wraps the message in a DHCPv6 'Relay-Forward' message 2021 header and includes an 'Interface-Id' option that includes enough 2022 information to allow the LDRA to forward the resulting Reply message 2023 back to the Client (e.g., the Client's link-layer addresses, a 2024 security association identifier, etc.). The LDRA also wraps the OMNI 2025 option and SLLAO into the Interface-Id option, then forwards the 2026 message to the DHCPv6 server. 2028 When the DHCPv6 server prepares a Reply message, it wraps the message 2029 in a 'Relay-Reply' message and echoes the Interface-Id option. The 2030 DHCPv6 server then delivers the Relay-Reply message to the LDRA, 2031 which discards the Relay-Reply wrapper and IPv6/UDP headers, then 2032 uses the DHCPv6 message to construct an RA response to the Client. 2033 The Server uses the information in the Interface-Id option to prepare 2034 the RA message and to cache the link-layer addresses taken from the 2035 OMNI option and SLLAO echoed in the Interface-Id option. 2037 3.16. The AERO Proxy 2039 Clients may connect to ANETs that deploy perimeter security services 2040 to facilitate communications to Servers in outside INETs. In that 2041 case, the ANET can employ an AERO Proxy. The Proxy is located at the 2042 ANET/INET border and listens for RS messages originating from or RA 2043 messages destined to ANET Clients. The Proxy acts on these control 2044 messages as follows: 2046 o when the Proxy receives an RS message from a new ANET Client, it 2047 first authenticates the message then examines the network-layer 2048 destination address. If the destination address is a Server's 2049 AERO address, the Proxy proceeds to the next step. Otherwise, if 2050 the destination is All-Routers multicast or Subnet-Router anycast, 2051 the Proxy selects a "nearby" Server that is likely to be a good 2052 candidate to serve the Client and replaces the destination address 2053 with the Server's AERO address. Next, the Proxy creates a proxy 2054 neighbor cache entry and caches the Client and Server link-layer 2055 addresses along with the OMNI option information and any other 2056 identifying information including Transaction IDs, Client 2057 Identifiers, Nonce values, etc. The Proxy finally encapsulates 2058 the (proxyed) RS message in a SPAN header with source set to the 2059 Proxy's SPAN address and destination set to the Server's SPAN 2060 address then forwards the message into the SPAN. 2062 o when the Server receives the RS, it authenticates the message then 2063 creates or updates a symmetric neighbor cache entry for the Client 2064 with the Proxy's SPAN address as the link-layer address. The 2065 Server then sends an RA message back to the Proxy via the SPAN. 2067 o when the Proxy receives the RA, it authenticates the message and 2068 matches it with the proxy neighbor cache entry created by the RS. 2069 The Proxy then caches the PD route information as a mapping from 2070 the Client's MNPs to the Client's ANET address, caches the 2071 Server's advertised Router Lifetime and sets the neighbor cache 2072 entry state to REACHABLE. The Proxy then sets the P bit in the RA 2073 flags field, optionally rewrites the Router Lifetime and forwards 2074 the (proxyed) message to the Client. The Proxy finally includes 2075 an MTU option (if necessary) with an MTU to use for the underlying 2076 ANET interface. 2078 After the initial RS/RA exchange, the Proxy forwards any Client data 2079 packets for which there is no matching asymmetric neighbor cache 2080 entry to a Relay via the SPAN. The Proxy instead forwards any Client 2081 data destined to an asymmetric neighbor cache target directly to the 2082 target according to the link-layer information - the process of 2083 establishing asymmetric neighbor cache entries is specified in 2084 Section 3.17. 2086 While the Client is still attached to the ANET, the Proxy sends NS, 2087 RS and/or unsolicited NA messages to update the Server's symmetric 2088 neighbor cache entries on behalf of the Client and/or to convey QoS 2089 updates. This allows for higher-frequency Proxy-initiated RS/RA 2090 messaging over well-connected INET infrastructure supplemented by 2091 lower-frequency Client-initiated RS/RA messaging over constrained 2092 ANET data links. 2094 If the Server ceases to send solicited advertisements, the Proxy 2095 sends unsolicited RAs on the ANET interface with destination set to 2096 All-Nodes multicast (ff02::1) and with Router Lifetime set to zero to 2097 inform Clients that the Server has failed. Although the Proxy 2098 engages in ND exchanges on behalf of the Client, the Client can also 2099 send ND messages on its own behalf, e.g., if it is in a better 2100 position than the Proxy to convey QoS changes, etc. For this reason, 2101 the Proxy marks any Client-originated solicitation messages (e.g. by 2102 inserting a Nonce option) so that it can return the solicited 2103 advertisement to the Client instead of processsing it locally. 2105 If the Client becomes unreachable, the Proxy sets the neighbor cache 2106 entry state to DEPARTED and retains the entry for DEPARTTIME seconds. 2107 While the state is DEPARTED, the Proxy forwards any packets destined 2108 to the Client to a Relay. The Relay in turn forwards the packets to 2109 the Client's current Server. When DepartTime expires, the Proxy 2110 deletes the neighbor cache entry and discards any further packets 2111 destined to this (now forgotten) Client. 2113 In some ANETs that employ a Proxy, the Client's MNP can be injected 2114 into the ANET routing system. In that case, the Client can send data 2115 messages without encapsulation so that the ANET native routing system 2116 transports the unencapsulated packets to the Proxy. This can be very 2117 beneficial, e.g., if the Client connects to the ANET via low-end data 2118 links such as some aviation wireless links. 2120 If the first-hop ANET access router is AERO-aware, the Client can 2121 avoid encapsulation for both its control and data messages. When the 2122 Client connects to the link, it can send an unencapsulated RS message 2123 with source address set to its AERO address and with destination 2124 address set to the AERO address of the Client's selected Server or to 2125 All-Routers multicast or Subnet-Router anycast. The Client includes 2126 an OMNI option formatted as specified in 2127 [I-D.templin-6man-omni-interface]. 2129 The Client then sends the unencapsulated RS message, which will be 2130 intercepted by the AERO-Aware access router. The access router then 2131 encapsulates the RS message in an ANET header with its own address as 2132 the source address and the address of a Proxy as the destination 2133 address. The access router further remembers the address of the 2134 Proxy so that it can encapsulate future data packets from the Client 2135 via the same Proxy. If the access router needs to change to a new 2136 Proxy, it simply sends another RS message toward the Server via the 2137 new Proxy on behalf of the Client. 2139 In some cases, the access router and Proxy may be one and the same 2140 node. In that case, the node would be located on the same physical 2141 link as the Client, but its message exchanges with the Server would 2142 need to pass through a security gateway at the ANET/INET border. The 2143 method for deploying access routers and Proxys (i.e. as a single node 2144 or multiple nodes) is an ANET-local administrative consideration. 2146 3.16.1. Detecting and Responding to Server Failures 2148 In environments where fast recovery from Server failure is required, 2149 Proxys SHOULD use proactive Neighbor Unreachability Detection (NUD) 2150 to track Server reachability in a similar fashion as for 2151 Bidirectional Forwarding Detection (BFD) [RFC5880]. Proxys can then 2152 quickly detect and react to failures so that cached information is 2153 re-established through alternate paths. The NUD control messaging is 2154 carried only over well-connected ground domain networks (i.e., and 2155 not low-end aeronautical radio links) and can therefore be tuned for 2156 rapid response. 2158 Proxys perform proactive NUD with Servers for which there are 2159 currently active ANET Clients by sending continuous NS messages in 2160 rapid succession, e.g., one message per second. The Proxy sends the 2161 NS message via the SPAN with the Proxy's AERO address as the source 2162 and the AERO address of the Server as the destination. When the 2163 Proxy is also sending RS messages to the Server on behalf of ANET 2164 Clients, the resulting RA responses can be considered as equivalent 2165 hints of forward progress. This means that the Proxy need not also 2166 send a periodic NS if it has already sent an RS within the same 2167 period. If the Server fails (i.e., if the Proxy ceases to receive 2168 advertisements), the Proxy can quickly inform Clients by sending 2169 multicast RA messages on the ANET interface. 2171 The Proxy sends RA messages on the ANET interface with source address 2172 set to the Server's address, destination address set to All-Nodes 2173 multicast, and Router Lifetime set to 0. The Proxy SHOULD send 2174 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2175 [RFC4861]. Any Clients on the ANET that had been using the failed 2176 Server will receive the RA messages and associate with a new Server. 2178 3.16.2. Point-to-Multipoint Server Coordindation 2180 In environments where Client messaging over ANETs is bandwidth- 2181 limited and/or expensive, Clients can enlist the services of the 2182 Proxy to coordinate with multiple Servers in a single RS/RA message 2183 exchange. The Client can send a single RS message to All-Routers 2184 multicast that includes the ID's of multiple Servers in MS-Register 2185 sub-options of the OMNI option,. 2187 When the Proxy receives the RS and processes the OMNI option, it 2188 performs a separate RS/RA exchange with each MS-Register Server. 2189 When it has received the RA messages, it creates an "aggregate" RA 2190 message to return to the Client with an OMNI option with each 2191 responding Server's ID recorded in an MS-Register sub-option. 2193 Client's can thereafter employ efficient point-to-multipoint Server 2194 coordination under the assistance of the Proxy to dramatically reduce 2195 the number of messages sent over the ANET while enlisting the support 2196 of multiple Servers for fault tolerance. Clients can further include 2197 MS-Release suboptions in RS messages to request the Proxy to release 2198 from former Servers via the procedures discussed in Section 3.19.5. 2200 The OMNI interface specification [I-D.templin-6man-omni-interface] 2201 provides further discussion of the Client/Proxy RS/RA messaging 2202 involved in point-to-multipoint coordination. 2204 3.17. AERO Route Optimization 2206 While data packets are flowing between a source and target node, 2207 route optimization SHOULD be used. Route optimization is initiated 2208 by the first eligible Route Optimization Source (ROS) closest to the 2209 source as follows: 2211 o For Clients on VPNed, NATed and Direct interfaces, the Server is 2212 the ROS. 2214 o For Clients on Proxyed interfaces, the Proxy is the ROS. 2216 o For Clients on native interfaces, the Client itself is the ROS. 2218 o For correspondent nodes on INET/EUN interfaces serviced by a 2219 Gateway, the Gateway is the ROS. 2221 The route optimization procedure is conducted between the ROS and the 2222 target Server/Gateway acting as a Route Optimization Responder (ROR) 2223 in the same manner as for IPv6 ND Address Resolution and using the 2224 same NS/NA messaging. The target may either be a MNP Client serviced 2225 by a Server, or a non-MNP correspondent reachable via a Gateway. 2227 The procedures are specified in the following sections. 2229 3.17.1. Route Optimization Initiation 2231 While data packets are flowing from the source node toward a target 2232 node, the ROS performs address resolution by sending an NS message 2233 for Address Resolution (NS(AR)) to receive a solicited NA message 2234 from the ROR. When the ROS sends an NS(AR), it includes: 2236 o the AERO address of the ROS as the source address. 2238 o the data packet's destination as the Target Address. 2240 o the Solicited-Node multicast address [RFC4291] formed from the 2241 lower 24 bits of the data packet's destination as the destination 2242 address, e.g., for 2001:db8:1:2::10:2000 the NS destination 2243 address is ff02:0:0:0:0:1:ff10:2000. 2245 The NS(AR) message includes an OMNI option with no ifIndex-tuples and 2246 no SLLAO, such that the target will not create a neighbor cache 2247 entry. 2249 The ROS then encapsulates the NS(AR) message in a SPAN header with 2250 source set to its own SPAN address and destination set to the data 2251 packet's destination address, then sends the message into the SPAN 2252 without decrementing the network-layer TTL/Hop Limit field. 2254 3.17.2. Relaying the NS 2256 When the Relay receives the NS(AR) message from the ROS, it discards 2257 the INET header and determines that the ROR is the next hop by 2258 consulting its standard IPv6 forwarding table for the SPAN header 2259 destination address. The Relay then forwards the message toward the 2260 ROR via the SPAN the same as for any IPv6 router. The final-hop 2261 Relay in the SPAN will deliver the message via a secured tunnel to 2262 the ROR. 2264 3.17.3. Processing the NS and Sending the NA 2266 When the ROR receives the NS(AR) message, it examines the Target 2267 Address to determine whether it has a neighbor cache entry and/or 2268 route that matches the target. If there is no match, the ROR drops 2269 the NS(AR) message. Otherwise, the ROR continues processing as 2270 follows: 2272 o if the target belongs to an MNP Client neighbor in the DEPARTED 2273 state the ROR changes the NS(AR) message SPAN destination address 2274 to the SPAN address of the Client's new Server, forwards the 2275 message into the SPAN and returns from processing. 2277 o If the target belongs to an MNP Client neighbor in the REACHABLE 2278 state, the ROR instead adds the AERO source address to the target 2279 Client's Report List with time set to ReportTime. 2281 o If the target belongs to a non-MNP route, the ROR continues 2282 processing without adding an entry to the Report List. 2284 The ROR then prepares a solicited NA message to send back to the ROS 2285 but does not create a neighbor cache entry. The ROR sets the NA 2286 source address to the AERO address corresponding to the target, sets 2287 the Target Addresss to the target of the solicitation, and sets the 2288 destination address to the source of the solicitation. 2290 The ROR then includes an OMNI option with prefix registration length 2291 set to the length of the MNP if the target is an MNP Client; 2292 otherwise, set to the maximum of the non-MNP prefix length and 64. 2293 (Note that a /64 limit is imposed to avoid causing the ROS to set 2294 short prefixes (e.g., "default") that would match destinations for 2295 which the routing system includes more-specific prefixes.) 2297 If the target is an MNP Client, the ROR next includes ifIndex-tuples 2298 in the OMNI option for each of the target Client's underlying 2299 interfaces with current information for each interface and with the S 2300 flag set to 0. The ROR then includes a TLLAO with ifIndex-tuples in 2301 one-to-one correspondence with the tuples that appear in the OMNI 2302 option. For NATed, VPNed and Direct interfaces, the link layer 2303 addresses are the SPAN address of the ROR. For Proxyed interfaces, 2304 the link-layer addresses are the SPAN addresses of the Proxy's INET 2305 interfaces. For native interfaces, the link-layer addresses are the 2306 SPAN addesses of the Client's native interfaces. 2308 The ROR then sets the NA message R flag to 1 (as a router), S flag to 2309 1 (as a response to a solicitation), and O flag to 0 (as a proxy). 2310 The ROR finally encapsulates the NA message in a SPAN header with 2311 source set to its own SPAN address and destination set to the source 2312 SPAN address of the NS(AR) message, then forwards the message into 2313 the SPAN without decrementing the network-layer TTL/Hop Limit field. 2315 3.17.4. Relaying the NA 2317 When the Relay receives the NA message from the ROR, it discards the 2318 INET header and determines that the ROS is the next hop by consulting 2319 its standard IPv6 forwarding table for the SPAN header destination 2320 address. The Relay then forwards the SPAN-encapsulated NA message 2321 toward the ROS the same as for any IPv6 router. The final-hop Relay 2322 in the SPAN will deliver the message via a secured tunnel to the ROS. 2324 3.17.5. Processing the NA 2326 When the ROS receives the solicited NA message, it processes the 2327 message the same as for standard IPv6 Address Resolution [RFC4861]. 2328 In the process, it caches the source SPAN address then creates an 2329 asymmetric neighbor cache entry for the ROR and caches all 2330 information found in the OMNI and TLLAO options. The ROS finally 2331 sets the asymmetric neighbor cache entry lifetime to REACHABLETIME 2332 seconds. 2334 3.17.6. Route Optimization Maintenance 2336 Following route optimization, the ROS forwards future data packets 2337 destined to the target via the addresses found in the cached link- 2338 layer information. The route optimization is shared by all sources 2339 that send packets to the target via the ROS, i.e., and not just the 2340 source on behalf of which the route optimization was initiated. 2342 While new data packets destined to the target are flowing through the 2343 ROS, it sends additional NS(AR) messages to the ROR before 2344 ReachableTime expires to receive a fresh solicited NA message the 2345 same as described in the previous sections (route optimization 2346 refreshment strategies are an implementation matter, with a non- 2347 normative example given in Appendix B.1). The ROS uses the cached 2348 SPAN address of the ROR as the NS(AR) SPAN destination address, and 2349 sends up to MAX_MULTICAST_SOLICIT NS(AR) messages separated by 1 2350 second until an NA is received. If no NA is received, the ROS 2351 assumes that the current ROR has become unreachable and deletes the 2352 neighbor cache entry. Subsequent data packets will trigger a new 2353 route optimization per Section 3.17.1 to discover a new ROR while 2354 initial data packets travel over a suboptimal route. 2356 If an NA is received, the ROS then updates the asymmetric neighbor 2357 cache entry to refresh ReachableTime, while (for MNP destinations) 2358 the ROR adds or updates the ROS address to the target Client's Report 2359 List and with time set to ReportTime. While no data packets are 2360 flowing, the ROS instead allows ReachableTime for the asymmetric 2361 neighbor cache entry to expire. When ReachableTime expires, the ROS 2362 deletes the asymmetric neighbor cache entry. Any future data packets 2363 flowing through the ROS will again trigger a new route optimization. 2365 The ROS may also receive unsolicited NA messages from the ROR at any 2366 time (see: Section 3.19). If there is an asymmetric neighbor cache 2367 entry for the target, the ROS updates the link-layer information but 2368 does not update ReachableTime since the receipt of an unsolicited NA 2369 does not confirm that any forward paths are working. If there is no 2370 asymmetric neighbor cache entry, the ROS simply discards the 2371 unsolicited NA. 2373 In this arrangement, the ROS holds an asymmetric neighbor cache entry 2374 for the ROR, but the ROR does not hold an asymmetric neighbor cache 2375 entry for the ROS. The route optimization neighbor relationship is 2376 therefore asymmetric and unidirectional. If the target node also has 2377 packets to send back to the source node, then a separate route 2378 optimization procedure is performed in the reverse direction. But, 2379 there is no requirement that the forward and reverse paths be 2380 symmetric. 2382 3.18. Neighbor Unreachability Detection (NUD) 2384 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2385 [RFC4861] either reactively in response to persistent link-layer 2386 errors (see Section 3.14) or proactively to confirm reachability. 2387 The NUD algorithm is based on periodic control message exchanges. 2388 The algorithm may further be seeded by ND hints of forward progress, 2389 but care must be taken to avoid inferring reachability based on 2390 spoofed information. For example, authentic IPv6 ND message 2391 exchanges may be considered as acceptable hints of forward progress, 2392 while spurious data packets should not be. 2394 AERO Servers, Proxys and Gateways can use standard NS/NA NUD 2395 exchanges sent over the SPAN to securely test reachability without 2396 risk of DoS attacks from nodes pretending to be a neighbor; Proxys 2397 can further perform NUD to securely verify Server reachability on 2398 behalf of their proxyed Clients. However, a means for a ROS to test 2399 the unsecured forward directions of target route optimized paths is 2400 also necessary. The following paragraphs present the suggested 2401 method. 2403 When an ROR directs an ROS to a neighbor with one or more target 2404 link-layer addresses, the ROS can proactively test each such 2405 unsecured route optimized path by sending "loopback" NS(NUD) 2406 messages. While testing the paths, the ROS can optionally continue 2407 to send packets via the SPAN, maintain a small queue of packets until 2408 target reachability is confirmed, or (optimistically) allow packets 2409 to flow via the route optimized paths. 2411 When the ROS sends a loopback NS(NUD) message, it uses its AERO 2412 address as both the IPv6 source and destination address, and any IPv6 2413 address as the Target Address. The ROS includes a Nonce and 2414 Timestamp option, then encapsulates the message in SPAN/INET headers 2415 with its own SPAN address as the source and the SPAN address of the 2416 route optimization target as the destination. The ROS then forwards 2417 the message to the target (either directly to the link layer address 2418 of the target if the target is in the same SPAN segment, or via a 2419 Relay if the target is in a different SPAN segment). 2421 When the route optimization target receives the NS(NUD) message, it 2422 notices that the IPv6 destination address is the same as the source 2423 address. It then reverses the SPAN source and destination addresses 2424 and returns the message to the ROS (either directly or via the SPAN). 2425 The route optimization target does not decrement the NS(NUD) message 2426 IPv6 Hop-Limit in the process, since the message has not exited the 2427 SPAN. 2429 When the ROS receives the NS(NUD) message, it can determine from the 2430 Nonce, Timestamp and Target Address that the message originated from 2431 itself and that it transited the forward path. The ROS need not 2432 prepare a NA response, since the destination of the response would be 2433 itself and testing the route optimization path again would be 2434 redundant. 2436 The ROS marks route optimization target paths that pass these NUD 2437 tests as "reachable", and those that do not as "unreachable". These 2438 markings inform the AERO interface forwarding algorithm specified in 2439 Section 3.13. 2441 Note that to avoid a DoS vector nodes MUST NOT return loopback 2442 NS(NUD) messages received from an unsecured link-layer source via a 2443 secured SPAN path. 2445 3.19. Mobility Management and Quality of Service (QoS) 2447 AERO is a Distributed Mobility Management (DMM) service. Each Server 2448 is responsible for only a subset of the Clients on the AERO link, as 2449 opposed to a Centralized Mobility Management (CMM) service where 2450 there is a single network mobility collective entity for all Clients. 2451 Clients coordinate with their associated Servers via RS/RA exchanges 2452 to maintain the DMM profile, and the AERO routing system tracks all 2453 current Client/Server peering relationships. 2455 Servers provide default routing and mobility/multilink services for 2456 their dependent Clients. Clients are responsible for maintaining 2457 neighbor relationships with their Servers through periodic RS/RA 2458 exchanges, which also serves to confirm neighbor reachability. When 2459 a Client's underlying interface address and/or QoS information 2460 changes, the Client is responsible for updating the Server with this 2461 new information. Note that for Proxyed interfaces, however, the 2462 Proxy can also perform some RS/RA exchanges on the Client's behalf. 2464 Mobility management considerations are specified in the following 2465 sections. 2467 3.19.1. Mobility Update Messaging 2469 Servers accommodate Client mobility/multilink and/or QoS change 2470 events by sending unsolicited NA (uNA) messages to each ROS in the 2471 target Client's Report List. When a Server sends a uNA message, it 2472 sets the IPv6 source address to the Client's AERO address, sets the 2473 destination address to All-Nodes multicast and sets the Target 2474 Address to the Client's Subnet-Router anycast address. The Server 2475 also includes an OMNI option with prefix registration information and 2476 with ifIndex-tuples for the target Client's remaining interfaces with 2477 S set to 0. The Server then includes a TLLAO with corresponding 2478 ifIndex-tuples with link layer addresses set to the corresponding 2479 target SPAN addresses. The Server sets the NA R flag to 1, the S 2480 flag to 0 and the O flag to 0, then encapsulates the message in a 2481 SPAN header with source set to its own SPAN address and destination 2482 set to the SPAN address of the ROS and sends the message into the 2483 SPAN. 2485 As discussed in Section 7.2.6 of [RFC4861], the transmission and 2486 reception of uNA messages is unreliable but provides a useful 2487 optimization. In well-connected Internetworks with robust data links 2488 uNA messages will be delivered with high probability, but in any case 2489 the Server can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT uNAs 2490 to each ROS to increase the likelihood that at least one will be 2491 received. 2493 When the ROS receives an uNA message, it ignores the message if there 2494 is no existing neighbor cache entry for the Client. Otherwise, it 2495 uses the included OMNI option and TLLAO information to update the 2496 neighbor cache entry, but does not reset ReachableTime since the 2497 receipt of an unsolicited NA message from the target Server does not 2498 provide confirmation that any forward paths to the target Client are 2499 working. 2501 If uNA messages are lost, the ROS may be left with stale address and/ 2502 or QoS information for the Client for up to REACHABLETIME seconds. 2503 During this time, the ROS can continue sending packets according to 2504 its stale neighbor cache information. When ReachableTime is close to 2505 expiring, the ROS will re-initiate route optimization and receive 2506 fresh link-layer address information. 2508 In addition to sending uNA messages to the current set of ROSs for 2509 the Client, the Server also sends uNAs to the former link-layer 2510 address for any ifIndex-tuple for which the link-layer address has 2511 changed. The uNA messages update Proxys that cannot easily detect 2512 (e.g., without active probing) when a formerly-active Client has 2513 departed. 2515 3.19.2. Announcing Link-Layer Address and/or QoS Preference Changes 2517 When a Client needs to change its ANET addresses and/or QoS 2518 preferences (e.g., due to a mobility event), either the Client or its 2519 Proxys send RS messages to the Server via the SPAN with an OMNI 2520 option that includes an ifIndex-tuple with S set to 1 and with the 2521 new link quality and address information. 2523 Up to MAX_RTR_SOLICITATION RS messages MAY be sent in parallel with 2524 sending actual data packets in case one or more RAs are lost. If all 2525 RAs are lost, the Client SHOULD re-associate with a new Server. 2527 When the Server receives the Client's changes, it sends uNA messages 2528 to all nodes in the Report List the same as described in the previous 2529 section. 2531 3.19.3. Bringing New Links Into Service 2533 When a Client needs to bring new underlying interfaces into service 2534 (e.g., when it activates a new data link), it sends an RS message to 2535 the Server via the underlying interface with an OMNI option that 2536 includes an ifIndex-tuple with S set to 1 and appropriate link 2537 quality values and with link-layer address information for the new 2538 link. 2540 3.19.4. Removing Existing Links from Service 2542 When a Client needs to remove existing underlying interfaces from 2543 service (e.g., when it de-activates an existing data link), it sends 2544 an RS or uNA message to its Server with an OMNI option with 2545 appropriate link quality values. 2547 If the Client needs to send RS/uNA messages over an underlying 2548 interface other than the one being removed from service, it MUST 2549 include ifIndex-tuples with appropriate link quality values for any 2550 underlying interfaces being removed from service. 2552 3.19.5. Moving to a New Server 2554 When a Client associates with a new Server, it performs the Client 2555 procedures specified in Section 3.15.2. The Client also includes MS- 2556 Release identifiers in the RS message OMNI option per 2557 [I-D.templin-6man-omni-interface] if it wants the new Server to 2558 notify any old Servers from which the Client is departing. 2560 When the new Server receives the Client's RS message, it returns an 2561 RA as specified in Section 3.15.3 and sends up to 2562 MAX_NEIGHBOR_ADVERTIISEMENT uNA messages to any old Servers listed in 2563 OMNI option MS-Release identifiers. Each uNA message includes the 2564 Client's AERO address as the source address, the old Server's AERO 2565 address as the destination address, and an OMNI option with the 2566 Register/Release bit set to 0. The new Server wraps the uNA in a 2567 SPAN header with its own SPAN address as the source and the old 2568 Server's SPAN address as the destination, then sends the message into 2569 the SPAN. 2571 When an old Server receives the uNA, it changes the Client's neighbor 2572 cache entry state to DEPARTED, sets the link-layer address of the 2573 Client to the new Server's SPAN address, and sets DepartTime to 2574 DEPARTTIME seconds. After a short delay (e.g., 2 seconds) the old 2575 Server withdraws the Client's MNP from the routing system. After 2576 DepartTime expires, the old Server deletes the Client's neighbor 2577 cache entry. 2579 The old Server also sends unsolicited NA messages to all ROSs in the 2580 Client's Report List with an OMNI option with a single ifIndex-tuple 2581 with ifIndex set to 0 and S set to '1', and with the SPAN address of 2582 the new Server in a companion TLLAO. When the ROS receives the NA, 2583 it caches the address of the new Server in the existing asymmetric 2584 neighbor cache entry and marks the entry as STALE. Subsequent data 2585 packets will then flow according to any existing cached link-layer 2586 information and trigger a new NS(AR)/NA exchange via the new Server. 2588 Clients SHOULD NOT move rapidly between Servers in order to avoid 2589 causing excessive oscillations in the AERO routing system. Examples 2590 of when a Client might wish to change to a different Server include a 2591 Server that has gone unreachable, topological movements of 2592 significant distance, movement to a new geographic region, movement 2593 to a new SPAN segment, etc. 2595 When a Client moves to a new Server, some of the fragments of a 2596 multiple fragment packet may have already arrived at the old Server 2597 while others are en route to the new Server, however no special 2598 attention in the reassembly algorithm is necessary when re-routed 2599 fragments are simply treated as loss. 2601 3.20. Multicast 2603 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 2604 [RFC3810] proxy service for its EUNs and/or hosted applications 2605 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 2606 underlying interfaces for which group membership is required. The 2607 IGMP/MLD messages may be further forwarded by a first-hop ANET access 2608 router acting as an IGMP/MLD-snooping switch [RFC4541], then 2609 ultimately delivered to an AERO Proxy/Server acting as a Protocol 2610 Independent Multicast - Sparse-Mode (PIM-SM, or simply "PIM") 2611 Designated Router (DR) [RFC7761]. AERO Gateways also act as PIM 2612 routers (i.e., the same as AERO Proxys/Servers) on behalf of nodes on 2613 INET/EUN networks. The behaviors identified in the following 2614 sections correspond to Source-Specific Multicast (SSM) and Any-Source 2615 Multicast (ASM) operational modes. 2617 3.20.1. Source-Specific Multicast (SSM) 2619 When an ROS (i.e., an AERO Proxy/Server/Gateway) "X" acting as PIM 2620 router receives a Join/Prune message from a node on its downstream 2621 interfaces containing one or more ((S)ource, (G)roup) pairs, it 2622 updates its Multicast Routing Information Base (MRIB) accordingly. 2623 For each S belonging to a prefix reachable via X's non-AERO 2624 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 2625 on those interfaces per [RFC7761]. 2627 For each S belonging to a prefix reachable via X's AERO interface, X 2628 originates a separate copy of the Join/Prune for each (S,G) in the 2629 message using its own AERO address as the source address and ALL-PIM- 2630 ROUTERS as the destination address. X then encapsulates each message 2631 in a SPAN header with source address set to the SPAN address of X and 2632 destination address set to S then forwards the message into the SPAN. 2633 The SPAN in turn forwards the message to AERO Server/Gateway "Y" that 2634 services S. At the same time, if the message was a Join, X sends a 2635 route-optimization NS message toward each S the same as discussed in 2636 Section 3.17. The resulting NAs will return the AERO address for the 2637 prefix that matches S as the network-layer source address and TLLAOs 2638 with the SPAN addresses corresponding to any ifIndex-tuples that are 2639 currently servicing S. 2641 When Y processes the Join/Prune message, if S located behind any 2642 Native, Direct, VPNed or NATed interfaces Y acts as a PIM router and 2643 updates its MRIB to list X as the next hop in the reverse path. If S 2644 is located behind any Proxys "Z"*, Y also forwards the message to 2645 each Z* over the SPAN while continuing to use the AERO address of X 2646 as the source address. Each Z* then updates its MRIB accordingly and 2647 maintains the AERO address of X as the next hop in the reverse path. 2648 Since the Relays in the SPAN do not examine network layer control 2649 messages, this means that the (reverse) multicast tree path is simply 2650 from each Z* (and/or Y) to X with no other multicast-aware routers in 2651 the path. If any Z* (and/or Y) is located on the same SPAN segment 2652 as X, the multicast data traffic sent to X directly using SPAN/INET 2653 encapsulation instead of via a Relay. 2655 Following the initial Join/Prune and NS/NA messaging, X maintains an 2656 asymmetric neighbor cache entry for each S the same as if X was 2657 sending unicast data traffic to S. In particular, X performs 2658 additional NS/NA exchanges to keep the neighbor cache entry alive for 2659 up to t_periodic seconds [RFC7761]. If no new Joins are received 2660 within t_periodic seconds, X allows the neighbor cache entry to 2661 expire. Finally, if X receives any additional Join/Prune messages 2662 for (S,G) it forwards the messages to each Y and Z* in the neighbor 2663 cache entry over the SPAN. 2665 At some later time, Client C that holds an MNP for source S may 2666 depart from a first Proxy Z1 and/or connect via a new Proxy Z2. In 2667 that case, Y sends an unsolicited NA message to X the same as 2668 specified for unicast mobility in Section 3.19. When X receives the 2669 unsolicited NA message, it updates its asymmetric neighbor cache 2670 entry for the AERO address for source S and sends new Join messages 2671 to any new Proxys Z2. There is no requirement to send any Prune 2672 messages to old Proxys Z1 since source S will no longer source any 2673 multicast data traffic via Z1. Instead, the multicast state for 2674 (S,G) in Proxy Z1 will soon time out since no new Joins will arrive. 2676 After some later time, C may move to a new Server Y2 and depart from 2677 old Sever Y1. In that case, Y1 sends Join messages for any of C's 2678 active (S,G) groups to Y2 while including its own AERO address as the 2679 source address. This causes Y2 to include Y1 in the multicast 2680 forwarding tree during the interim time that Y1's symmetric neighbor 2681 cache entry for C is in the DEPARTED state. At the same time, Y1 2682 sends an unsolicited NA message to X with an OMNI option and TLLAO 2683 with ifIndex-tuple set to 0 and a release indication to cause X to 2684 release its asymmetric neighbor cache entry. X then sends a new Join 2685 message to S via the SPAN and re-initiates route optimization the 2686 same as if it were receiving a fresh Join message from a node on a 2687 downstream link. 2689 3.20.2. Any-Source Multicast (ASM) 2691 When an ROS X acting as a PIM router receives a Join/Prune from a 2692 node on its downstream interfaces containing one or more (*,G) pairs, 2693 it updates its Multicast Routing Information Base (MRIB) accordingly. 2694 X then forwards a copy of the message to the Rendezvous Point (RP) R 2695 for each G over the SPAN. X uses its own AERO address as the source 2696 address and ALL-PIM-ROUTERS as the destination address, then 2697 encapsulates each message in a SPAN header with source address set to 2698 the SPAN address of X and destination address set to R, then sends 2699 the message into the SPAN. At the same time, if the message was a 2700 Join X initiates NS/NA route optimization the same as for the SSM 2701 case discussed in Section 3.20.1. 2703 For each source S that sends multicast traffic to group G via R, the 2704 Proxy/Server Z* for the Client that aggregates S encapsulates the 2705 packets in PIM Register messages and forwards them to R via the SPAN. 2706 R may then elect to send a PIM Join to Z* over the SPAN. This will 2707 result in an (S,G) tree rooted at Z* with R as the next hop so that R 2708 will begin to receive two copies of the packet; one native copy from 2709 the (S, G) tree and a second copy from the pre-existing (*, G) tree 2710 that still uses PIM Register encapsulation. R can then issue a PIM 2711 Register-stop message to suppress the Register-encapsulated stream. 2712 At some later time, if C moves to a new Proxy/Server Z*, it resumes 2713 sending packets via PIM Register encapsulation via the new Z*. 2715 At the same time, as multicast listeners discover individual S's for 2716 a given G, they can initiate an (S,G) Join for each S under the same 2717 procedures discussed in Section 3.20.1. Once the (S,G) tree is 2718 established, the listeners can send (S, G) Prune messages to R so 2719 that multicast packets for group G sourced by S will only be 2720 delivered via the (S, G) tree and not from the (*, G) tree rooted at 2721 R. All mobility considerations discussed for SSM apply. 2723 3.20.3. Bi-Directional PIM (BIDIR-PIM) 2725 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 2726 approach to ASM that treats the Rendezvous Point (RP) as a Designated 2727 Forwarder (DF). Further considerations for BIDIR-PIM are out of 2728 scope. 2730 3.21. Operation over Multiple AERO Links (VLANs) 2732 An AERO Client can connect to multiple AERO links the same as for any 2733 data link service. In that case, the Client maintains a distinct 2734 AERO interface for each link, e.g., 'aero0' for the first link, 2735 'aero1' for the second, 'aero2' for the third, etc. Each AERO link 2736 would include its own distinct set of Relays, Servers and Proxys, 2737 thereby providing redundancy in case of failures. 2739 The Relays, Servers and Proxys on each AERO link can assign AERO and 2740 SPAN addresses that use the same or different numberings from those 2741 on other links. Since the links are mutually independent there is no 2742 requirement for avoiding inter-link address duplication, e.g., the 2743 same AERO address such as fe80::1000 could be used to number distinct 2744 nodes that connect to different AERO links. 2746 Each AERO link could utilize the same or different ANET connections. 2747 The links can be distinguished at the link-layer via Virtual Local 2748 Area Network (VLAN) tagging (e.g., IEEE 802.1Q) and/or through 2749 assignment of distinct sets of MSPs on each link. This gives rise to 2750 the opportunity for supporting multiple redundant networked paths, 2751 where each VLAN is distinguished by a different label (e.g., colors 2752 such as Red, Green, Blue, etc.). In particular, the Client can tag 2753 its RS messages with the appropriate label to cause the network to 2754 select the desired VLAN. 2756 Clients that connect to multiple AERO interfaces can select the 2757 outgoing interface appropriate for a given Red/Blue/Green/etc. 2758 traffic profile while (in the reverse direction) correspondent nodes 2759 must have some way of steering their packets destined to a target via 2760 the correct AERO link. 2762 In a first alternative, if each AERO link services different MSPs, 2763 then the Client can receive a distinct MNP from each of the links. 2764 IP routing will therefore assure that the correct Red/Green/Blue/etc. 2765 network is used for both outbound and inbound traffic. This can be 2766 accomplished using existing technologies and approaches, and without 2767 requiring any special supporting code in correspondent nodes or 2768 Relays. 2770 In a second alternative, if each AERO link services the same MSP(s) 2771 then each link could assign a distinct "AERO Link Anycast" address 2772 that is configured by all Relays on the link. Correspondent nodes 2773 then include a "type 4" routing header with the Anycast address for 2774 the AERO link as the IPv6 destination and with the address of the 2775 target encoded as the "next segment" in the routing header 2776 [RFC8402][I-D.ietf-6man-segment-routing-header]. Standard IP routing 2777 will then direct the packet to the nearest Relay for the correct AERO 2778 link, which will replace the destination address with the target 2779 address then forward the packet to the target. 2781 3.22. DNS Considerations 2783 AERO Client MNs and INET correspondent nodes consult the Domain Name 2784 System (DNS) the same as for any Internetworking node. When 2785 correspondent nodes and Client MNs use different IP protocol versions 2786 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 2787 A records for IPv4 address mappings to MNs which must then be 2788 populated in Gateway NAT64 mapping caches. In that way, an IPv4 2789 correspondent node can send packets to the IPv4 address mapping of 2790 the target MN, and the Gateway will translate the IPv4 header and 2791 destination address into an IPv6 header and IPv6 destination address 2792 of the MN. 2794 When an AERO Client registers with an AERO Server, the Server can 2795 return the address(es) of DNS servers in RDNSS options [RFC6106]. 2796 The DNS server provides the IP addresses of other MNs and 2797 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 2799 3.23. Transition Considerations 2801 The SPAN ensures that dissimilar INET partitions can be joined into a 2802 single unified AERO link, even though the partitions themselves may 2803 have differing protocol versions and/or incompatible addressing 2804 plans. However, a commonality can be achieved by incrementally 2805 distributing globally routable (i.e., native) IP prefixes to 2806 eventually reach all nodes (both mobile and fixed) in all SPAN 2807 segments. This can be accomplished by incrementally deploying AERO 2808 Gateways on each INET partition, with each Gateway distributing its 2809 MNPs and/or discovering non-MNP prefixes on its INET links. 2811 This gives rise to the opportunity to eventually distribute native IP 2812 addresses to all nodes, and to present a unified AERO link view 2813 (bridged by the SPAN) even if the INET partitions remain in their 2814 current protocol and addressing plans. In that way, the AERO link 2815 can serve the dual purpose of providing a mobility/multilink service 2816 and a transition service. Or, if an INET partition is transitioned 2817 to a native IP protocol version and addressing scheme that is 2818 compatible with the AERO link MNP-based addressing scheme, the 2819 partition and AERO link can be joined by Gateways. 2821 Gateways that connect INETs/EUNs with dissimilar IP protocol versions 2822 must employ a network address and protocol translation function such 2823 as NAT64[RFC6146]. 2825 3.24. Detecting and Reacting to Server and Relay Failures 2827 In environments where rapid failure recovery is required, Servers and 2828 Relays SHOULD use Bidirectional Forwarding Detection (BFD) [RFC5880]. 2829 Nodes that use BFD can quickly detect and react to failures so that 2830 cached information is re-established through alternate nodes. BFD 2831 control messaging is carried only over well-connected ground domain 2832 networks (i.e., and not low-end radio links) and can therefore be 2833 tuned for rapid response. 2835 Servers and Relays maintain BFD sessions in parallel with their BGP 2836 peerings. If a Server or Relay fails, BGP peers will quickly re- 2837 establish routes through alternate paths the same as for common BGP 2838 deployments. Similarly, Proxys maintain BFD sessions with their 2839 associated Relays even though they do not establish BGP peerings with 2840 them. 2842 Proxys SHOULD use proactive NUD for Servers for which there are 2843 currently active ANET Clients in a manner that parallels BFD, i.e., 2844 by sending unicast NS messages in rapid succession to receive 2845 solicited NA messages. When the Proxy is also sending RS messages on 2846 behalf of ANET Clients, the RS/RA messaging can be considered as 2847 equivalent hints of forward progress. This means that the Proxy need 2848 not also send a periodic NS if it has already sent an RS within the 2849 same period. If a Server fails, the Proxy will cease to receive 2850 advertisements and can quickly inform Clients of the outage by 2851 sending multicast RA messages on the ANET interface. 2853 The Proxy sends multicast RA messages with source address set to the 2854 Server's address, destination address set to All-Nodes multicast, and 2855 Router Lifetime set to 0. The Proxy SHOULD send 2856 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2857 [RFC4861]. Any Clients on the ANET interface that have been using 2858 the (now defunct) Server will receive the RA messages and associate 2859 with a new Server. 2861 4. Implementation Status 2863 An AERO implementation based on OpenVPN (https://openvpn.net/) was 2864 announced on the v6ops mailing list on January 10, 2018 and an 2865 initial public release of the AERO proof-of-concept source code was 2866 announced on the intarea mailing list on August 21, 2015. 2868 5. IANA Considerations 2870 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 2871 AERO in the "enterprise-numbers" registry. 2873 The IANA has assigned the UDP port number "8060" for an earlier 2874 experimental version of AERO [RFC6706]. This document obsoletes 2875 [RFC6706] and claims the UDP port number "8060" for all future use. 2877 No further IANA actions are required. 2879 6. Security Considerations 2881 AERO Relays configure secured tunnels with AERO Servers and Proxys 2882 within their local SPAN segments. Applicable secured tunnel 2883 alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 2884 [RFC6347], WireGuard, etc. The AERO Relays of all SPAN segments in 2885 turn configure secured tunnels for their neighboring AERO Relays 2886 across the SPAN. Therefore, control messages that traverse the SPAN 2887 between any pair of AERO link neighbors are already secured. 2889 AERO Servers, Gateways and Proxys targeted by a route optimization 2890 may also receive packets directly from the INET partitions instead of 2891 via the SPAN. For INET partitions that apply effective ingress 2892 filtering to defeat source address spoofing, the simple data origin 2893 authentication procedures in Section 3.11 can be applied. 2895 For INET partitions that cannot apply effective ingress filtering, 2896 the two options for securing communications include 1) disable route 2897 optimization so that all traffic is conveyed over secured tunnels via 2898 the SPAN, or 2) enable on-demand secure tunnel creation between INET 2899 partition neighbors. Option 1) would result in longer routes than 2900 necessary and traffic concentration on critical infrastructure 2901 elements. Option 2) could be coordinated by establishing a secured 2902 tunnel on-demand instead of performing an NS/NA exchange in the route 2903 optimization procedures. Procedures for establishing on-demand 2904 secured tunnels are out of scope. 2906 AERO Clients that connect to secured enclaves need not apply security 2907 to their ND messages, since the messages will be intercepted by a 2908 perimeter Proxy that applies security on its outward-facing 2909 interface. AERO Clients located outside of secured enclaves SHOULD 2910 use symmetric network and/or transport layer security services, but 2911 when there are many prospective neighbors with dynamically changing 2912 connectivity an asymmetric security service such as SEND may be 2913 needed (see: Appendix B.6). 2915 Application endpoints SHOULD use application-layer security services 2916 such as TLS/SSL, DTLS or SSH [RFC4251] to assure the same level of 2917 protection as for critical secured Internet services. AERO Clients 2918 that require host-based VPN services SHOULD use symmetric network 2919 and/or transport layer security services such as IPsec, TLS/SSL, 2920 DTLS, etc. AERO Proxys and Servers can also provide a network-based 2921 VPN service on behalf of the Client, e.g., if the Client is located 2922 within a secured enclave and cannot establish a VPN on its own 2923 behalf. 2925 AERO Servers and Relays present targets for traffic amplification 2926 Denial of Service (DoS) attacks. This concern is no different than 2927 for widely-deployed VPN security gateways in the Internet, where 2928 attackers could send spoofed packets to the gateways at high data 2929 rates. This can be mitigated by connecting Servers and Relays over 2930 dedicated links with no connections to the Internet and/or when 2931 connections to the Internet are only permitted through well-managed 2932 firewalls. Traffic amplification DoS attacks can also target an AERO 2933 Client's low data rate links. This is a concern not only for Clients 2934 located on the open Internet but also for Clients in secured 2935 enclaves. AERO Servers and Proxys can institute rate limits that 2936 protect Clients from receiving packet floods that could DoS low data 2937 rate links. 2939 AERO Gateways must implement ingress filtering to avoid a spoofing 2940 attack in which spurious SPAN messages are injected into an AERO link 2941 from an outside attacker. AERO Clients MUST ensure that their 2942 connectivity is not used by unauthorized nodes on their EUNs to gain 2943 access to a protected network, i.e., AERO Clients that act as routers 2944 MUST NOT provide routing services for unauthorized nodes. (This 2945 concern is no different than for ordinary hosts that receive an IP 2946 address delegation but then "share" the address with other nodes via 2947 some form of Internet connection sharing such as tethering.) 2949 The MAP list MUST be well-managed and secured from unauthorized 2950 tampering, even though the list contains only public information. 2951 The MAP list can be conveyed to the Client in a similar fashion as in 2952 [RFC5214] (e.g., through layer 2 data link login messaging, secure 2953 upload of a static file, DNS lookups, etc.). 2955 Although public domain and commercial SEND implementations exist, 2956 concerns regarding the strength of the cryptographic hash algorithm 2957 have been documented [RFC6273] [RFC4982]. 2959 Security considerations for accepting link-layer ICMP messages and 2960 reflected packets are discussed throughout the document. 2962 7. Acknowledgements 2964 Discussions in the IETF, aviation standards communities and private 2965 exchanges helped shape some of the concepts in this work. 2966 Individuals who contributed insights include Mikael Abrahamsson, Mark 2967 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 2968 Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian Farrel, Nick Green, 2969 Sri Gundavelli, Brian Haberman, Bernhard Haindl, Joel Halpern, Tom 2970 Herbert, Sascha Hlusiak, Lee Howard, Zdenek Jaron, Andre Kostur, 2971 Hubert Kuenig, Ted Lemon, Andy Malis, Satoru Matsushima, Tomek 2972 Mrugalski, Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal 2973 Skorepa, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, Lloyd 2974 Wood and James Woodyatt. Members of the IESG also provided valuable 2975 input during their review process that greatly improved the document. 2976 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 2977 for their shepherding guidance during the publication of the AERO 2978 first edition. 2980 This work has further been encouraged and supported by Boeing 2981 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 2982 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 2983 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 2984 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 2985 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 2986 Gene MacLean III, Rob Muszkiewicz, Sean O'Sullivan, Vijay 2987 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 2988 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 2989 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 2990 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 2991 Benson, Katie Tran and Eric Yeh are especially acknowledged for 2992 implementing the AERO functions as extensions to the public domain 2993 OpenVPN distribution. 2995 Earlier works on NBMA tunneling approaches are found in 2996 [RFC2529][RFC5214][RFC5569]. 2998 Many of the constructs presented in this second edition of AERO are 2999 based on the author's earlier works, including: 3001 o The Internet Routing Overlay Network (IRON) 3002 [RFC6179][I-D.templin-ironbis] 3004 o Virtual Enterprise Traversal (VET) 3005 [RFC5558][I-D.templin-intarea-vet] 3007 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 3008 [RFC5320][I-D.templin-intarea-seal] 3010 o AERO, First Edition [RFC6706] 3012 Note that these works cite numerous earlier efforts that are not also 3013 cited here due to space limitations. The authors of those earlier 3014 works are acknowledged for their insights. 3016 This work is aligned with the NASA Safe Autonomous Systems Operation 3017 (SASO) program under NASA contract number NNA16BD84C. 3019 This work is aligned with the FAA as per the SE2025 contract number 3020 DTFAWA-15-D-00030. 3022 This work is aligned with the Boeing Commercial Airplanes (BCA) 3023 Internet of Things (IoT) and autonomy programs. 3025 This work is aligned with the Boeing Information Technology (BIT) 3026 MobileNet program. 3028 8. References 3030 8.1. Normative References 3032 [I-D.templin-6man-omni-interface] 3033 Templin, F. and T. Whyman, "Transmission of IPv6 Packets 3034 over Overlay Multilink Network (OMNI) Interfaces", draft- 3035 templin-6man-omni-interface-07 (work in progress), March 3036 2020. 3038 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3039 DOI 10.17487/RFC0791, September 1981, 3040 . 3042 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 3043 RFC 792, DOI 10.17487/RFC0792, September 1981, 3044 . 3046 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3047 Requirement Levels", BCP 14, RFC 2119, 3048 DOI 10.17487/RFC2119, March 1997, 3049 . 3051 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3052 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3053 December 1998, . 3055 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 3056 "Definition of the Differentiated Services Field (DS 3057 Field) in the IPv4 and IPv6 Headers", RFC 2474, 3058 DOI 10.17487/RFC2474, December 1998, 3059 . 3061 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3062 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3063 DOI 10.17487/RFC3971, March 2005, 3064 . 3066 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3067 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3068 . 3070 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3071 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3072 November 2005, . 3074 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3075 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3076 . 3078 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3079 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3080 DOI 10.17487/RFC4861, September 2007, 3081 . 3083 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3084 Address Autoconfiguration", RFC 4862, 3085 DOI 10.17487/RFC4862, September 2007, 3086 . 3088 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3089 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3090 May 2017, . 3092 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3093 (IPv6) Specification", STD 86, RFC 8200, 3094 DOI 10.17487/RFC8200, July 2017, 3095 . 3097 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3098 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3099 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3100 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3101 . 3103 8.2. Informative References 3105 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3106 2016. 3108 [I-D.ietf-6man-segment-routing-header] 3109 Filsfils, C., Dukes, D., Previdi, S., Leddy, J., 3110 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3111 (SRH)", draft-ietf-6man-segment-routing-header-26 (work in 3112 progress), October 2019. 3114 [I-D.ietf-dmm-distributed-mobility-anchoring] 3115 Chan, A., Wei, X., Lee, J., Jeon, S., and C. Bernardos, 3116 "Distributed Mobility Anchoring", draft-ietf-dmm- 3117 distributed-mobility-anchoring-15 (work in progress), 3118 March 2020. 3120 [I-D.ietf-intarea-gue] 3121 Herbert, T., Yong, L., and O. Zia, "Generic UDP 3122 Encapsulation", draft-ietf-intarea-gue-09 (work in 3123 progress), October 2019. 3125 [I-D.ietf-intarea-gue-extensions] 3126 Herbert, T., Yong, L., and F. Templin, "Extensions for 3127 Generic UDP Encapsulation", draft-ietf-intarea-gue- 3128 extensions-06 (work in progress), March 2019. 3130 [I-D.ietf-intarea-tunnels] 3131 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3132 Architecture", draft-ietf-intarea-tunnels-10 (work in 3133 progress), September 2019. 3135 [I-D.ietf-rtgwg-atn-bgp] 3136 Templin, F., Saccone, G., Dawra, G., Lindem, A., and V. 3137 Moreno, "A Simple BGP-based Mobile Routing System for the 3138 Aeronautical Telecommunications Network", draft-ietf- 3139 rtgwg-atn-bgp-05 (work in progress), January 2020. 3141 [I-D.templin-6man-dhcpv6-ndopt] 3142 Templin, F., "A Unified Stateful/Stateless Configuration 3143 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-09 3144 (work in progress), January 2020. 3146 [I-D.templin-intarea-grefrag] 3147 Templin, F., "GRE Tunnel Level Fragmentation", draft- 3148 templin-intarea-grefrag-04 (work in progress), July 2016. 3150 [I-D.templin-intarea-seal] 3151 Templin, F., "The Subnetwork Encapsulation and Adaptation 3152 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 3153 progress), January 2014. 3155 [I-D.templin-intarea-vet] 3156 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 3157 templin-intarea-vet-40 (work in progress), May 2013. 3159 [I-D.templin-ironbis] 3160 Templin, F., "The Interior Routing Overlay Network 3161 (IRON)", draft-templin-ironbis-16 (work in progress), 3162 March 2014. 3164 [I-D.templin-v6ops-pdhost] 3165 Templin, F., "IPv6 Prefix Delegation and Multi-Addressing 3166 Models", draft-templin-v6ops-pdhost-25 (work in progress), 3167 January 2020. 3169 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3171 [RFC1035] Mockapetris, P., "Domain names - implementation and 3172 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3173 November 1987, . 3175 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 3176 Communication Layers", STD 3, RFC 1122, 3177 DOI 10.17487/RFC1122, October 1989, 3178 . 3180 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 3181 DOI 10.17487/RFC1191, November 1990, 3182 . 3184 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3185 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3186 . 3188 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3189 DOI 10.17487/RFC2003, October 1996, 3190 . 3192 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3193 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3194 . 3196 [RFC2492] Armitage, G., Schulter, P., and M. Jork, "IPv6 over ATM 3197 Networks", RFC 2492, DOI 10.17487/RFC2492, January 1999, 3198 . 3200 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3201 Domains without Explicit Tunnels", RFC 2529, 3202 DOI 10.17487/RFC2529, March 1999, 3203 . 3205 [RFC2764] Gleeson, B., Lin, A., Heinanen, J., Armitage, G., and A. 3206 Malis, "A Framework for IP Based Virtual Private 3207 Networks", RFC 2764, DOI 10.17487/RFC2764, February 2000, 3208 . 3210 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 3211 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 3212 DOI 10.17487/RFC2784, March 2000, 3213 . 3215 [RFC2890] Dommety, G., "Key and Sequence Number Extensions to GRE", 3216 RFC 2890, DOI 10.17487/RFC2890, September 2000, 3217 . 3219 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", 3220 RFC 2923, DOI 10.17487/RFC2923, September 2000, 3221 . 3223 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3224 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3225 . 3227 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3228 of Explicit Congestion Notification (ECN) to IP", 3229 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3230 . 3232 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3233 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3234 DOI 10.17487/RFC3810, June 2004, 3235 . 3237 [RFC3819] Karn, P., Ed., Bormann, C., Fairhurst, G., Grossman, D., 3238 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 3239 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 3240 RFC 3819, DOI 10.17487/RFC3819, July 2004, 3241 . 3243 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 3244 for IPv6 Hosts and Routers", RFC 4213, 3245 DOI 10.17487/RFC4213, October 2005, 3246 . 3248 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3249 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3250 January 2006, . 3252 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3253 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3254 DOI 10.17487/RFC4271, January 2006, 3255 . 3257 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3258 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3259 2006, . 3261 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3262 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3263 December 2005, . 3265 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3266 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3267 2006, . 3269 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3270 Control Message Protocol (ICMPv6) for the Internet 3271 Protocol Version 6 (IPv6) Specification", STD 89, 3272 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3273 . 3275 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3276 Protocol (LDAP): The Protocol", RFC 4511, 3277 DOI 10.17487/RFC4511, June 2006, 3278 . 3280 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3281 "Considerations for Internet Group Management Protocol 3282 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3283 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3284 . 3286 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3287 "Internet Group Management Protocol (IGMP) / Multicast 3288 Listener Discovery (MLD)-Based Multicast Forwarding 3289 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3290 August 2006, . 3292 [RFC4607] Holbrook, H. and B. Cain, "Source-Specific Multicast for 3293 IP", RFC 4607, DOI 10.17487/RFC4607, August 2006, 3294 . 3296 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 3297 Errors at High Data Rates", RFC 4963, 3298 DOI 10.17487/RFC4963, July 2007, 3299 . 3301 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3302 Algorithms in Cryptographically Generated Addresses 3303 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3304 . 3306 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3307 "Bidirectional Protocol Independent Multicast (BIDIR- 3308 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3309 . 3311 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3312 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3313 DOI 10.17487/RFC5214, March 2008, 3314 . 3316 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3317 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3318 February 2010, . 3320 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3321 Route Optimization Requirements for Operational Use in 3322 Aeronautics and Space Exploration Mobile Networks", 3323 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3324 . 3326 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3327 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3328 . 3330 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3331 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3332 January 2010, . 3334 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 3335 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 3336 . 3338 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3339 "IPv6 Router Advertisement Options for DNS Configuration", 3340 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3341 . 3343 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3344 NAT64: Network Address and Protocol Translation from IPv6 3345 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3346 April 2011, . 3348 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3349 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3350 . 3352 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3353 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3354 DOI 10.17487/RFC6221, May 2011, 3355 . 3357 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3358 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3359 DOI 10.17487/RFC6273, June 2011, 3360 . 3362 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3363 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3364 January 2012, . 3366 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3367 for Equal Cost Multipath Routing and Link Aggregation in 3368 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3369 . 3371 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3372 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3373 . 3375 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID Field", 3376 RFC 6864, DOI 10.17487/RFC6864, February 2013, 3377 . 3379 [RFC7269] Chen, G., Cao, Z., Xie, C., and D. Binet, "NAT64 3380 Deployment Options and Experience", RFC 7269, 3381 DOI 10.17487/RFC7269, June 2014, 3382 . 3384 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3385 Korhonen, "Requirements for Distributed Mobility 3386 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3387 . 3389 [RFC7421] Carpenter, B., Ed., Chown, T., Gont, F., Jiang, S., 3390 Petrescu, A., and A. Yourtchenko, "Analysis of the 64-bit 3391 Boundary in IPv6 Addressing", RFC 7421, 3392 DOI 10.17487/RFC7421, January 2015, 3393 . 3395 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3396 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3397 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3398 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3399 2016, . 3401 [RFC8086] Yong, L., Ed., Crabbe, E., Xu, X., and T. Herbert, "GRE- 3402 in-UDP Encapsulation", RFC 8086, DOI 10.17487/RFC8086, 3403 March 2017, . 3405 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 3406 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 3407 DOI 10.17487/RFC8201, July 2017, 3408 . 3410 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3411 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3412 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3413 July 2018, . 3415 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3416 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3417 . 3419 Appendix A. AERO Alternate Encapsulations 3421 When GUE encapsulation is not needed, AERO can use common 3422 encapsulations such as IP-in-IP [RFC2003][RFC2473][RFC4213], Generic 3423 Routing Encapsulation (GRE) [RFC2784][RFC2890] and others. The 3424 encapsulation is therefore only differentiated from non-AERO tunnels 3425 through the application of AERO control messaging and not through, 3426 e.g., a well-known UDP port number. 3428 As for GUE encapsulation, alternate AERO encapsulation formats may 3429 require encapsulation layer fragmentation. For simple IP-in-IP 3430 encapsulation, an IPv6 fragment header is inserted directly between 3431 the inner and outer IP headers when needed, i.e., even if the outer 3432 header is IPv4. The IPv6 Fragment Header is identified to the outer 3433 IP layer by its IP protocol number, and the Next Header field in the 3434 IPv6 Fragment Header identifies the inner IP header version. For GRE 3435 encapsulation, a GRE fragment header is inserted within the GRE 3436 header [I-D.templin-intarea-grefrag]. 3438 Figure 6 shows the AERO IP-in-IP encapsulation format before any 3439 fragmentation is applied: 3441 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3442 | Outer IPv4 Header | | Outer IPv6 Header | 3443 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3444 |IPv6 Frag Header (optional)| |IPv6 Frag Header (optional)| 3445 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3446 | Inner IP Header | | Inner IP Header | 3447 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3448 | | | | 3449 ~ ~ ~ ~ 3450 ~ Inner Packet Body ~ ~ Inner Packet Body ~ 3451 ~ ~ ~ ~ 3452 | | | | 3453 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3455 Minimal Encapsulation in IPv4 Minimal Encapsulation in IPv6 3457 Figure 6: Minimal Encapsulation Format using IP-in-IP 3459 Figure 7 shows the AERO GRE encapsulation format before any 3460 fragmentation is applied: 3462 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3463 | Outer IP Header | 3464 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3465 | GRE Header | 3466 | (with checksum, key, etc..) | 3467 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3468 | GRE Fragment Header (optional)| 3469 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3470 | Inner IP Header | 3471 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3472 | | 3473 ~ ~ 3474 ~ Inner Packet Body ~ 3475 ~ ~ 3476 | | 3477 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3479 Figure 7: Minimal Encapsulation Using GRE 3481 Alternate encapsulation may be preferred in environments where GUE 3482 encapsulation would add unnecessary overhead. For example, certain 3483 low-bandwidth wireless data links may benefit from a reduced 3484 encapsulation overhead. 3486 GUE encapsulation can traverse network paths that are inaccessible to 3487 non-UDP encapsulations, e.g., for crossing Network Address 3488 Translators (NATs). More and more, network middleboxes are also 3489 being configured to discard packets that include anything other than 3490 a well-known IP protocol such as UDP and TCP. It may therefore be 3491 necessary to determine the potential for middlebox filtering before 3492 enabling alternate encapsulation in a given environment. 3494 In addition to IP-in-IP, GRE and GUE, AERO can also use security 3495 encapsulations such as IPsec, TLS/SSL, DTLS, etc. In that case, AERO 3496 control messaging and route determination occur before security 3497 encapsulation is applied for outgoing packets and after security 3498 decapsulation is applied for incoming packets. 3500 AERO is especially well suited for use with VPN system encapsulations 3501 such as OpenVPN [OVPN]. 3503 Appendix B. Non-Normative Considerations 3505 AERO can be applied to a multitude of Internetworking scenarios, with 3506 each having its own adaptations. The following considerations are 3507 provided as non-normative guidance: 3509 B.1. Implementation Strategies for Route Optimization 3511 Route optimization as discussed in Section 3.17 results in the route 3512 optimization source (ROS) creating an asymmetric neighbor cache entry 3513 for the target neighbor. The neighbor cache entry is maintained for 3514 at most REACHABLETIME seconds and then deleted unless updated. In 3515 order to refresh the neighbor cache entry lifetime before the 3516 ReachableTime timer expires, the specification requires 3517 implementations to issue a new NS/NA exchange to reset ReachableTime 3518 to REACHABLETIME seconds while data packets are still flowing. 3519 However, the decision of when to initiate a new NS/NA exchange and to 3520 perpetuate the process is left as an implementation detail. 3522 One possible strategy may be to monitor the neighbor cache entry 3523 watching for data packets for (REACHABLETIME - 5) seconds. If any 3524 data packets have been sent to the neighbor within this timeframe, 3525 then send an NS to receive a new NA. If no data packets have been 3526 sent, wait for 5 additional seconds and send an immediate NS if any 3527 data packets are sent within this "expiration pending" 5 second 3528 window. If no additional data packets are sent within the 5 second 3529 window, delete the neighbor cache entry. 3531 The monitoring of the neighbor data packet traffic therefore becomes 3532 an asymmetric ongoing process during the neighbor cache entry 3533 lifetime. If the neighbor cache entry expires, future data packets 3534 will trigger a new NS/NA exchange while the packets themselves are 3535 delivered over a longer path until route optimization state is re- 3536 established. 3538 B.2. Implicit Mobility Management 3540 AERO interface neighbors MAY provide a configuration option that 3541 allows them to perform implicit mobility management in which no ND 3542 messaging is used. In that case, the Client only transmits packets 3543 over a single interface at a time, and the neighbor always observes 3544 packets arriving from the Client from the same link-layer source 3545 address. 3547 If the Client's underlying interface address changes (either due to a 3548 readdressing of the original interface or switching to a new 3549 interface) the neighbor immediately updates the neighbor cache entry 3550 for the Client and begins accepting and sending packets according to 3551 the Client's new address. This implicit mobility method applies to 3552 use cases such as cellphones with both WiFi and Cellular interfaces 3553 where only one of the interfaces is active at a given time, and the 3554 Client automatically switches over to the backup interface if the 3555 primary interface fails. 3557 B.3. Direct Underlying Interfaces 3559 When a Client's AERO interface is configured over a Direct interface, 3560 the neighbor at the other end of the Direct link can receive packets 3561 without any encapsulation. In that case, the Client sends packets 3562 over the Direct link according to QoS preferences. If the Direct 3563 interface has the highest QoS preference, then the Client's IP 3564 packets are transmitted directly to the peer without going through an 3565 ANET/INET. If other interfaces have higher QoS preferences, then the 3566 Client's IP packets are transmitted via a different interface, which 3567 may result in the inclusion of Proxys, Servers and Relays in the 3568 communications path. Direct interfaces must be tested periodically 3569 for reachability, e.g., via NUD. 3571 B.4. AERO Clients on the Open Internetwork 3573 AERO Clients that connect to the open Internetwork via either a 3574 native or NATed interface can establish a VPN to securely connect to 3575 a Server. Alternatively, the Client can exchange ND messages 3576 directly with other AERO nodes on the same SPAN segment using INET 3577 encapsulation only and without joining the SPAN. In that case, 3578 however, the Client must apply asymmetric security for ND messages to 3579 ensure routing and neighbor cache integrity (see: Section 6). 3581 B.5. Operation on AERO Links with /64 ASPs 3583 IPv6 AERO links typically have MSPs that aggregate many candidate 3584 MNPs of length /64 or shorter. However, in some cases it may be 3585 desirable to use AERO over links that have only a /64 MSP. This can 3586 be accommodated by treating all Clients on the AERO link as simple 3587 hosts that receive /128 prefix delegations. 3589 In that case, the Client sends an RS message to the Server the same 3590 as for ordinary AERO links. The Server responds with an RA message 3591 that includes one or more /128 prefixes (i.e., singleton addresses) 3592 that include the /64 MSP prefix along with an interface identifier 3593 portion to be assigned to the Client. The Client and Server then 3594 configure their AERO addresses based on the interface identifier 3595 portions of the /128s (i.e., the lower 64 bits) and not based on the 3596 /64 prefix (i.e., the upper 64 bits). 3598 For example, if the MSP for the host-only IPv6 AERO link is 3599 2001:db8:1000:2000::/64, each Client will receive one or more /128 3600 IPv6 prefix delegations such as 2001:db8:1000:2000::1/128, 3601 2001:db8:1000:2000::2/128, etc. When the Client receives the prefix 3602 delegations, it assigns the AERO addresses fe80::1, fe80::2, etc. to 3603 the AERO interface, and assigns the global IPv6 addresses (i.e., the 3604 /128s) to either the AERO interface or an internal virtual interface 3605 such as a loopback. In this arrangement, the Client conducts route 3606 optimization in the same sense as discussed in Section 3.17. 3608 This specification has applicability for nodes that act as a Client 3609 on an "upstream" AERO link, but also act as a Server on "downstream" 3610 AERO links. More specifically, if the node acts as a Client to 3611 receive a /64 prefix from the upstream AERO link it can then act as a 3612 Server to provision /128s to Clients on downstream AERO links. 3614 B.6. AERO Adaptations for SEcure Neighbor Discovery (SEND) 3616 SEcure Neighbor Discovery (SEND) [RFC3971] and Cryptographically 3617 Generated Addresses (CGAs) [RFC3972] were designed to secure IPv6 ND 3618 messaging in environments where symmetric network and/or transport- 3619 layer security services are impractical (see: Section 6). AERO nodes 3620 that use SEND/CGA employ the following adaptations. 3622 When a source AERO node prepares a SEND-protected ND message, it uses 3623 a link-local CGA as the IPv6 source address and writes the prefix 3624 embedded in its AERO address (i.e., instead of fe80::/64) in the CGA 3625 parameters Subnet Prefix field. When the neighbor receives the ND 3626 message, it first verifies the message checksum and SEND/CGA 3627 parameters while using the link-local prefix fe80::/64 (i.e., instead 3628 of the value in the Subnet Prefix field) to match against the IPv6 3629 source address of the ND message. 3631 The neighbor then derives the AERO address of the source by using the 3632 value in the Subnet Prefix field as the interface identifier of an 3633 AERO address. For example, if the Subnet Prefix field contains 3634 2001:db8:1:2, the neighbor constructs the AERO address as 3635 fe80::2001:db8:1:2. The neighbor then caches the AERO address in the 3636 neighbor cache entry it creates for the source, and uses the AERO 3637 address as the IPv6 destination address of any ND message replies. 3639 B.7. AERO Critical Infrastructure Considerations 3641 AERO Relays can be either Commercial off-the Shelf (COTS) standard IP 3642 routers or virtual machines in the cloud. Relays must be 3643 provisioned, supported and managed by the INET administrative 3644 authority, and connected to the Relays of other INETs via inter- 3645 domain peerings. Cost for purchasing, configuring and managing 3646 Relays is nominal even for very large AERO links. 3648 AERO Servers can be standard dedicated server platforms, but most 3649 often will be deployed as virtual machines in the cloud. The only 3650 requirements for Servers are that they can run the AERO user-level 3651 code and have at least one network interface connection to the INET. 3652 As with Relays, Servers must be provisioned, supported and managed by 3653 the INET administrative authority. Cost for purchasing, configuring 3654 and managing Servers is nominal especially for virtual Servers hosted 3655 in the cloud. 3657 AERO Proxys are most often standard dedicated server platforms with 3658 one network interface connected to the ANET and a second interface 3659 connected to an INET. As with Servers, the only requirements are 3660 that they can run the AERO user-level code and have at least one 3661 interface connection to the INET. Proxys must be provisioned, 3662 supported and managed by the ANET administrative authority. Cost for 3663 purchasing, configuring and managing Proxys is nominal, and borne by 3664 the ANET administrative authority. 3666 AERO Gateways can be any dedicated server or COTS router platform 3667 connected to INETs and/or EUNs. The Gateway joins the SPAN and 3668 engages in eBGP peering with one or more Relays as a stub AS. The 3669 Gateway then injects its MNPs and/or non-MNP prefixes into the BGP 3670 routing system, and provisions the prefixes to its downstream- 3671 attached networks. The Gateway can perform ROS/ROR services the same 3672 as for any Server, and can route between the MNP and non-MNP address 3673 spaces. 3675 B.8. AERO Server Failure Implications 3677 AERO Servers may appear as a single point of failure in the 3678 architecture, but such is not the case since all Servers on the link 3679 provide identical services and loss of a Server does not imply 3680 immediate and/or comprehensive communication failures. Although 3681 Clients typically associate with a single Server at a time, Server 3682 failure is quickly detected and conveyed by Bidirectional Forward 3683 Detection (BFD) and/or proactive NUD allowing Clients to migrate to 3684 new Servers. 3686 If a Server fails, ongoing packet forwarding to Clients will continue 3687 by virtue of the asymmetric neighbor cache entries that have already 3688 been established in route optimization sources (ROSs). If a Client 3689 also experiences mobility events at roughly the same time the Server 3690 fails, unsolicited NA messages may be lost but proxy neighbor cache 3691 entries in the DEPARTED state will ensure that packet forwarding to 3692 the Client's new locations will continue for up to DEPARTTIME 3693 seconds. 3695 If a Client is left without a Server for an extended timeframe (e.g., 3696 greater than REACHABLETIIME seconds) then existing asymmetric 3697 neighbor cache entries will eventually expire and both ongoing and 3698 new communications will fail. The original source will continue to 3699 retransmit until the Client has established a new Server 3700 relationship, after which time continuous communications will resume. 3702 Therefore, providing many Servers on the link with high availability 3703 profiles provides resilience against loss of individual Servers and 3704 assurance that Clients can establish new Server relationships quickly 3705 in event of a Server failure. 3707 B.9. AERO Client / Server Architecture 3709 The AERO architectural model is client / server in the control plane, 3710 with route optimization in the data plane. The same as for common 3711 Internet services, the AERO Client discovers the addresses of AERO 3712 Servers and selects one Server to connect to. The AERO service is 3713 analogous to common Internet services such as google.com, yahoo.com, 3714 cnn.com, etc. However, there is only one AERO service for the link 3715 and all Servers provide identical services. 3717 Common Internet services provide differing strategies for advertising 3718 server addresses to clients. The strategy is conveyed through the 3719 DNS resource records returned in response to name resolution queries. 3720 As of January 2020 Internet-based 'nslookup' services were used to 3721 determine the following: 3723 o When a client resolves the domainname "google.com", the DNS always 3724 returns one A record (i.e., an IPv4 address) and one AAAA record 3725 (i.e., an IPv6 address). The client receives the same addresses 3726 each time it resolves the domainname via the same DNS resolver, 3727 but may receive different addresses when it resolves the 3728 domainname via different DNS resolvers. But, in each case, 3729 exactly one A and one AAAA record are returned. 3731 o When a client resolves the domainname "ietf.org", the DNS always 3732 returns one A record and one AAAA record with the same addresses 3733 regardless of which DNS resolver is used. 3735 o When a client resolves the domainname "yahoo.com", the DNS always 3736 returns a list of 4 A records and 4 AAAA records. Each time the 3737 client resolves the domainname via the same DNS resolver, the same 3738 list of addresses are returned but in randomized order (i.e., 3739 consistent with a DNS round-robin strategy). But, interestingly, 3740 the same addresses are returned (albeit in randomized order) when 3741 the domainname is resolved via different DNS resolvers. 3743 o When a client resolves the domainname "amazon.com", the DNS always 3744 returns a list of 3 A records and no AAAA records. As with 3745 "yahoo.com", the same three A records are returned from any 3746 worldwide Internet connection point in randomized order. 3748 The above example strategies show differing approaches to Internet 3749 resilience and service distribution offered by major Internet 3750 services. The Google approach exposes only a single IPv4 and a 3751 single IPv6 address to clients. Clients can then select whichever IP 3752 protocol version offers the best response, but will always use the 3753 same IP address according to the current Internet connection point. 3754 This means that the IP address offered by the network must lead to a 3755 highly-available server and/or service distribution point. In other 3756 words, resilience is predicated on high availability within the 3757 network and with no client-initiated failovers expected (i.e., it is 3758 all-or-nothing from the client's perspective). However, Google does 3759 provide for worldwide distributed service distribution by virtue of 3760 the fact that each Internet connection point responds with a 3761 different IPv6 and IPv4 address. The IETF approach is like google 3762 (all-or-nothing from the client's perspective), but provides only a 3763 single IPv4 or IPv6 address on a worldwide basis. This means that 3764 the addresses must be made highly-available at the network level with 3765 no client failover possibility, and if there is any worldwide service 3766 distribution it would need to be conducted by a network element that 3767 is reached via the IP address acting as a service distribution point. 3769 In contrast to the Google and IETF philosophies, Yahoo and Amazon 3770 both provide clients with a (short) list of IP addresses with Yahoo 3771 providing both IP protocol versions and Amazon as IPv4-only. The 3772 order of the list is randomized with each name service query 3773 response, with the effect of round-robin load balancing for service 3774 distribution. With a short list of addresses, there is still 3775 expectation that the network will implement high availability for 3776 each address but in case any single address fails the client can 3777 switch over to using a different address. The balance then becomes 3778 one of function in the network vs function in the end system. 3780 The same implications observed for common highly-available services 3781 in the Internet apply also to the AERO client/server architecture. 3782 When an AERO Client connects to one or more ANETs, it discovers one 3783 or more AERO Server addresses through the mechanisms discussed in 3784 earlier sections. Each Server address presumably leads to a fault- 3785 tolerant clustering arrangement such as supported by Linux-HA, 3786 Extended Virtual Synchrony or Paxos. Such an arrangement has 3787 precedence in common Internet service deployments in lightweight 3788 virtual machines without requiring expensive hardware deployment. 3789 Similarly, common Internet service deployments set service IP 3790 addresses on service distribution points that may relay requests to 3791 many different servers. 3793 For AERO, the expectation is that a combination of the Google/IETF 3794 and Yahoo/Amazon philosophies would be employed. The AERO Client 3795 connects to different ANET access points and can receive 1-2 Server 3796 AERO addresses at each point. It then selects one AERO Server 3797 address, and engages in RS/RA exchanges with the same Server from all 3798 ANET connections. The Client remains with this Server unless or 3799 until the Server fails, in which case it can switch over to an 3800 alternate Server. The Client can likewise switch over to a different 3801 Server at any time if there is some reason for it to do so. So, the 3802 AERO expectation is for a balance of function in the network and end 3803 system, with fault tolerance and resilience at both levels. 3805 Appendix C. Change Log 3807 << RFC Editor - remove prior to publication >> 3809 Changes from draft-templin-intarea-6706bis-32 to draft-templin- 3810 intrea-6706bis-33: 3812 o Updated Proxy discussion with "point-to-multipoint" server 3813 coordination 3815 o Significant updates to Address Resolution and NUD to include 3816 correct addresses in messages 3818 o Differentiate between NS(AR) and NS(NUD) as their addresses and 3819 use cases differ. 3821 Changes from draft-templin-intarea-6706bis-30 to draft-templin- 3822 intrea-6706bis-31: 3824 o Added "advisory PTB messages" under FAA SE2025 contract number 3825 DTFAWA-15-D-00030. 3827 Changes from draft-templin-intarea-6706bis-29 to draft-templin- 3828 intrea-6706bis-30: 3830 o Deprecate "primary" concept. Now, RS/RA keepalives are 3831 mainatained over *all* underlying interfaces (i.e., and not just 3832 one primary). 3834 Changes from draft-templin-intarea-6706bis-28 to draft-templin- 3835 intrea-6706bis-29: 3837 o Changed OMNI interface citation to "draft-templin-6man-omni- 3838 interface" 3840 o Changed SPAN Service Prefix to fd80::/10. 3842 o Changed S/TLLAO format to include 'S' bit for ifIndex 3843 corresponding to the underlying interface that is Source of ND 3844 message. 3846 o Updated Path MTU 3848 Changes from draft-templin-intarea-6706bis-27 to draft-templin- 3849 intrea-6706bis-28: 3851 o MTU and fragmentation. 3853 Changes from draft-templin-intarea-6706bis-26 to draft-templin- 3854 intrea-6706bis-27: 3856 o MTU and fragmentation. 3858 o SPAN Service Prefix set to fd00::/10 3860 o Client SPAN addresses defined. 3862 Changes from draft-templin-intarea-6706bis-25 to draft-templin- 3863 intrea-6706bis-26: 3865 o MTU and RA configuration information updated. 3867 Changes from draft-templin-intarea-6706bis-24 to draft-templin- 3868 intrea-6706bis-25: 3870 o Added concept of "primary" to allow for proxyed RS/RA over only 3871 selected underlying interfaces. 3873 o General Cleanup. 3875 Changes from draft-templin-intarea-6706bis-23 to draft-templin- 3876 intrea-6706bis-24: 3878 o OMNI interface spec now a normative reference. 3880 o Use REACHABLETIME as the nominal Router Lifetime to return in RAs. 3882 o General cleanup. 3884 Changes from draft-templin-intarea-6706bis-22 to draft-templin- 3885 intrea-6706bis-23: 3887 o Choice of using either RS/RA or unsolicited NA for old Server 3888 notification. 3890 o General cleanup. 3892 Changes from draft-templin-intarea-6706bis-21 to draft-templin- 3893 intrea-6706bis-22: 3895 o Tightened up text on Proxy. 3897 o Removed unnecessarily restrictive texts. 3899 o General cleanup. 3901 Changes from draft-templin-intarea-6706bis-20 to draft-templin- 3902 intrea-6706bis-21: 3904 o Clarified relationship between OMNI and S/TLLAO ifIndex-tuples. 3906 o Important text in Section 13.15.3 on Servers timing out Clients 3907 that have gone silent without sending a departure notification. 3909 o New text on RS/RA as "hints of forward progress" for proactive 3910 NUD. 3912 Changes from draft-templin-intarea-6706bis-19 to draft-templin- 3913 intrea-6706bis-20: 3915 o Included new route optimization source and destination addressing 3916 strategy. Now, route optimization maintenance uses the address of 3917 the existing Server instead of the data packet destination address 3918 so that less pressure is placed on the BGP routing system 3919 convergence time and Server constancy is supported. 3921 o Included new method for releasing from old MSE without requiring 3922 Client messaging. 3924 o Included references to new OMNI interface spec (including the OMNI 3925 option). 3927 o New appendix on AERO Client/Server architecture. 3929 Changes from draft-templin-intarea-6706bis-18 to draft-templin- 3930 intrea-6706bis-19: 3932 o Changed Proxy/Server keepalives to use "proactive NUD" in a manner 3933 tha paralles BFD 3935 Changes from draft-templin-intarea-6706bis-17 to draft-templin- 3936 intrea-6706bis-18: 3938 o Discuss how AERO option is used in relation to S/TLLAOs 3940 o New text on Bidirectional Forwarding Detection (BFD) 3942 o Cleaned up usage (and non-usage) of unsolicited NAs 3944 o New appendix on Server failures 3946 Changes from draft-templin-intarea-6706bis-15 to draft-templin- 3947 intrea-6706bis-17: 3949 o S/TLLAO now includes multiple link-layer addresses within a single 3950 option instead of requiring multiple options 3952 o New unsolicited NA message to inform the old link that a Client 3953 has moved to a new link 3955 Changes from draft-templin-intarea-6706bis-14 to draft-templin- 3956 intrea-6706bis-15: 3958 o MTU and fragmentation 3960 o New details in movement to new Server 3961 Changes from draft-templin-intarea-6706bis-13 to draft-templin- 3962 intrea-6706bis-14: 3964 o Security based on secured tunnels, ingress filtering, MAP list and 3965 ROS list 3967 Changes from draft-templin-intarea-6706bis-12 to draft-templin- 3968 intrea-6706bis-13: 3970 o New paragraph in Section 3.6 on AERO interface layering over 3971 secured tunnels 3973 o Removed extraneous text in Section 3.7 3975 o Added new detail to the forwarding algorithm in Section 3.9 3977 o Clarified use of fragmentation 3979 o Route optimization now supported for both MNP and non-MNP-based 3980 prefixes 3982 o Relays are now seen as link-layer elements in the architecture. 3984 o Built out multicast section in detail. 3986 o New Appendix on implementation considerations for route 3987 optimization. 3989 Changes from draft-templin-intarea-6706bis-11 to draft-templin- 3990 intrea-6706bis-12: 3992 o Introduced Gateways as a new AERO element for connecting 3993 Correspondent Nodes on INET links 3995 o Introduced terms "Access Network (ANET)" and "Internetwork (INET)" 3997 o Changed "ASP" to "MSP", and "ACP" to "MNP" 3999 o New figure on the relation of Segments to the SPAN and AERO link 4001 o New "S" bit in S/TLLAO to indicate the "Source" S/TLLAO as opposed 4002 to additional S/TLLAOs 4004 o Changed Interface ID for Servers from 255 to 0xffff 4006 o Significant updates to Route Optimization, NUD, and Mobility 4007 Management 4009 o New Section on Multicast 4011 o New Section on AERO Clients in the open Internetwork 4013 o New Section on Operation over multiple AERO links (VLANs over the 4014 SPAN) 4016 o New Sections on DNS considerations and Transition considerations 4018 o 4020 Changes from draft-templin-intarea-6706bis-10 to draft-templin- 4021 intrea-6706bis-11: 4023 o Added The SPAN 4025 Changes from draft-templin-intarea-6706bis-09 to draft-templin- 4026 intrea-6706bis-10: 4028 o Orphaned packets in flight (e.g., when a neighbor cache entry is 4029 in the DEPARTED state) are now forwarded at the link layer instead 4030 of at the network layer. Forwarding at the network layer can 4031 result in routing loops and/or excessive delays of forwarded 4032 packets while the routing system is still reconverging. 4034 o Update route optimization to clarify the unsecured nature of the 4035 first NS used for route discovery 4037 o Many cleanups and clarifications on ND messaging parameters 4039 Changes from draft-templin-intarea-6706bis-08 to draft-templin- 4040 intrea-6706bis-09: 4042 o Changed PRL to "MAP list" 4044 o For neighbor cache entries, changed "static" to "symmetric", and 4045 "dynamic" to "asymmetric" 4047 o Specified Proxy RS/RA exchanges with Servers on behalf of Clients 4049 o Added discussion of unsolicited NAs in Section 3.16, and included 4050 forward reference to Section 3.18 4052 o Added discussion of AERO Clients used as critical infrastructure 4053 elements to connect fixed networks. 4055 o Added network-based VPN under security considerations 4056 Changes from draft-templin-intarea-6706bis-07 to draft-templin- 4057 intrea-6706bis-08: 4059 o New section on AERO-Aware Access Router 4061 Changes from draft-templin-intarea-6706bis-06 to draft-templin- 4062 intrea-6706bis-07: 4064 o Added "R" bit for release of PDs. Now have a full RS/RA service 4065 that can do PD without requiring DHCPv6 messaging over-the-air 4067 o Clarifications on solicited vs unsolicited NAs 4069 o Clarified use of MAX_NEIGHBOR_ADVERTISEMENTS for the purpose of 4070 increase reliability 4072 Changes from draft-templin-intarea-6706bis-05 to draft-templin- 4073 intrea-6706bis-06: 4075 o Major re-work and simplification of Route Optimization function 4077 o Added Distributed Mobility Management (DMM) and Mobility Anchor 4078 Point (MAP) terminology 4080 o New section on "AERO Critical Infrastructure Element 4081 Considerations" demonstrating low overall cost for the service 4083 o minor text revisions and deletions 4085 o removed extraneous appendices 4087 Changes from draft-templin-intarea-6706bis-04 to draft-templin- 4088 intrea-6706bis-05: 4090 o New Appendix E on S/TLLAO Extensions for special-purpose links. 4091 Discussed ATN/IPS as example. 4093 o New sentence in introduction to declare appendices as non- 4094 normative. 4096 Changes from draft-templin-intarea-6706bis-03 to draft-templin- 4097 intrea-6706bis-04: 4099 o Added definitions for Potential Router List (PRL) and secure 4100 enclave 4102 o Included text on mapping transport layer port numbers to network 4103 layer DSCP values 4105 o Added reference to DTLS and DMM Distributed Mobility Anchoring 4106 working group document 4108 o Reworked Security Considerations 4110 o Updated references. 4112 Changes from draft-templin-intarea-6706bis-02 to draft-templin- 4113 intrea-6706bis-03: 4115 o Added new section on SEND. 4117 o Clarifications on "AERO Address" section. 4119 o Updated references and added new reference for RFC8086. 4121 o Security considerations updates. 4123 o General text clarifications and cleanup. 4125 Changes from draft-templin-intarea-6706bis-01 to draft-templin- 4126 intrea-6706bis-02: 4128 o Note on encapsulation avoidance in Section 4. 4130 Changes from draft-templin-intarea-6706bis-00 to draft-templin- 4131 intrea-6706bis-01: 4133 o Remove DHCPv6 Server Release procedures that leveraged the old way 4134 Relays used to "route" between Server link-local addresses 4136 o Remove all text relating to Relays needing to do any AERO-specific 4137 operations 4139 o Proxy sends RS and receives RA from Server using SEND. Use CGAs 4140 as source addresses, and destination address of RA reply is to the 4141 AERO address corresponding to the Client's ACP. 4143 o Proxy uses SEND to protect RS and authenticate RA (Client does not 4144 use SEND, but rather relies on subnetwork security. When the 4145 Proxy receives an RS from the Client, it creates a new RS using 4146 its own addresses as the source and uses SEND with CGAs to send a 4147 new RS to the Server. 4149 o Emphasize distributed mobility management 4151 o AERO address-based RS injection of ACP into underlying routing 4152 system. 4154 Changes from draft-templin-aerolink-82 to draft-templin-intarea- 4155 6706bis-00: 4157 o Document use of NUD (NS/NA) for reliable link-layer address 4158 updates as an alternative to unreliable unsolicited NA. 4159 Consistent with Section 7.2.6 of RFC4861. 4161 o Server adds additional layer of encapsulation between outer and 4162 inner headers of NS/NA messages for transmission through Relays 4163 that act as vanilla IPv6 routers. The messages include the AERO 4164 Server Subnet Router Anycast address as the source and the Subnet 4165 Router Anycast address corresponding to the Client's ACP as the 4166 destination. 4168 o Clients use Subnet Router Anycast address as the encapsulation 4169 source address when the access network does not provide a 4170 topologically-fixed address. 4172 Author's Address 4174 Fred L. Templin (editor) 4175 Boeing Research & Technology 4176 P.O. Box 3707 4177 Seattle, WA 98124 4178 USA 4180 Email: fltemplin@acm.org