idnits 2.17.1 draft-templin-intarea-6706bis-63.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 (September 26, 2020) is 1307 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 914 -- Looks like a reference, but probably isn't: '2' on line 914 -- Looks like a reference, but probably isn't: '3' on line 887 == Unused Reference: 'BGP' is defined on line 3237, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-frag-fragile' is defined on line 3250, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-dhcpv6-ndopt' is defined on line 3267, but no explicit reference was found in the text == Outdated reference: A later version (-99) exists of draft-templin-6man-omni-interface-38 == Outdated reference: A later version (-31) exists of draft-bonica-6man-comp-rtg-hdr-22 == Outdated reference: A later version (-04) exists of draft-bonica-6man-crh-helper-opt-01 == Outdated reference: A later version (-13) exists of draft-ietf-intarea-tunnels-10 == Outdated reference: A later version (-26) exists of draft-ietf-rtgwg-atn-bgp-06 == Outdated reference: A later version (-11) exists of draft-templin-6man-dhcpv6-ndopt-10 == Outdated reference: A later version (-27) exists of draft-templin-v6ops-pdhost-26 -- Obsolete informational reference (is this intentional?): RFC 3330 (Obsoleted by RFC 5735) -- 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 (~~), 12 warnings (==), 7 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, September 26, 2020 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: March 30, 2021 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-intarea-6706bis-63 13 Abstract 15 This document specifies the operation of IP over Overlay Multilink 16 Network (OMNI) interfaces using the Asymmetric Extended Route 17 Optimization (AERO) internetworking and mobility management service. 18 AERO/OMNI use an IPv6 link-local address format that supports 19 operation of the IPv6 Neighbor Discovery (ND) protocol and links ND 20 to IP forwarding. Prefix delegation/registration services are 21 employed for network admission and to manage the routing system. 22 Multilink operation, mobility management, quality of service (QoS) 23 signaling and route optimization are naturally supported through 24 dynamic neighbor cache updates. Standard IP multicasting services 25 are also supported. AERO is a widely-applicable mobile 26 internetworking service especially well-suited to aviation services, 27 intelligent transportation systems, mobile Virtual Private Networks 28 (VPNs) and many other applications. 30 Status of This Memo 32 This Internet-Draft is submitted in full conformance with the 33 provisions of BCP 78 and BCP 79. 35 Internet-Drafts are working documents of the Internet Engineering 36 Task Force (IETF). Note that other groups may also distribute 37 working documents as Internet-Drafts. The list of current Internet- 38 Drafts is at https://datatracker.ietf.org/drafts/current/. 40 Internet-Drafts are draft documents valid for a maximum of six months 41 and may be updated, replaced, or obsoleted by other documents at any 42 time. It is inappropriate to use Internet-Drafts as reference 43 material or to cite them other than as "work in progress." 45 This Internet-Draft will expire on March 30, 2021. 47 Copyright Notice 49 Copyright (c) 2020 IETF Trust and the persons identified as the 50 document authors. All rights reserved. 52 This document is subject to BCP 78 and the IETF Trust's Legal 53 Provisions Relating to IETF Documents 54 (https://trustee.ietf.org/license-info) in effect on the date of 55 publication of this document. Please review these documents 56 carefully, as they describe your rights and restrictions with respect 57 to this document. Code Components extracted from this document must 58 include Simplified BSD License text as described in Section 4.e of 59 the Trust Legal Provisions and are provided without warranty as 60 described in the Simplified BSD License. 62 Table of Contents 64 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 65 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 66 3. Asymmetric Extended Route Optimization (AERO) . . . . . . . . 10 67 3.1. AERO Node Types . . . . . . . . . . . . . . . . . . . . . 10 68 3.2. The AERO Service over OMNI Links . . . . . . . . . . . . 12 69 3.2.1. AERO/OMNI Reference Model . . . . . . . . . . . . . . 12 70 3.2.2. Link-Local Addresses (LLAs) and Unique Local 71 Addresses (ULAs) . . . . . . . . . . . . . . . . . . 14 72 3.2.3. AERO Routing System . . . . . . . . . . . . . . . . . 15 73 3.2.4. OMNI Link Encapsulation . . . . . . . . . . . . . . . 16 74 3.2.5. Segment Routing Topologies (SRTs) . . . . . . . . . . 18 75 3.2.6. Segment Routing To the OMNI Link . . . . . . . . . . 18 76 3.2.7. Segment Routing Within the OMNI Link . . . . . . . . 19 77 3.2.8. Segment Routing Header Compression . . . . . . . . . 21 78 3.3. OMNI Interface Characteristics . . . . . . . . . . . . . 21 79 3.4. OMNI Interface Initialization . . . . . . . . . . . . . . 23 80 3.4.1. AERO Server/Relay Behavior . . . . . . . . . . . . . 23 81 3.4.2. AERO Proxy Behavior . . . . . . . . . . . . . . . . . 24 82 3.4.3. AERO Client Behavior . . . . . . . . . . . . . . . . 24 83 3.4.4. AERO Bridge Behavior . . . . . . . . . . . . . . . . 24 84 3.5. OMNI Interface Neighbor Cache Maintenance . . . . . . . . 24 85 3.5.1. OMNI Neighbor Interface Attributes . . . . . . . . . 26 86 3.6. OMNI Interface Encapsulation and Re-encapsulation . . . . 27 87 3.7. OMNI Interface Decapsulation . . . . . . . . . . . . . . 28 88 3.8. OMNI Interface Data Origin Authentication . . . . . . . . 28 89 3.9. OMNI Adaptation Layer and OMNI Interface MTU . . . . . . 29 90 3.10. OMNI Interface Forwarding Algorithm . . . . . . . . . . . 29 91 3.10.1. Client Forwarding Algorithm . . . . . . . . . . . . 30 92 3.10.2. Proxy Forwarding Algorithm . . . . . . . . . . . . . 30 93 3.10.3. Server/Relay Forwarding Algorithm . . . . . . . . . 31 94 3.10.4. Bridge Forwarding Algorithm . . . . . . . . . . . . 32 96 3.11. OMNI Interface Error Handling . . . . . . . . . . . . . . 33 97 3.12. AERO Router Discovery, Prefix Delegation and 98 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 36 99 3.12.1. AERO Service Model . . . . . . . . . . . . . . . . . 36 100 3.12.2. AERO Client Behavior . . . . . . . . . . . . . . . . 37 101 3.12.3. AERO Server Behavior . . . . . . . . . . . . . . . . 39 102 3.13. The AERO Proxy . . . . . . . . . . . . . . . . . . . . . 41 103 3.13.1. Combined Proxy/Servers . . . . . . . . . . . . . . . 43 104 3.13.2. Detecting and Responding to Server Failures . . . . 44 105 3.13.3. Point-to-Multipoint Server Coordination . . . . . . 44 106 3.14. AERO Route Optimization / Address Resolution . . . . . . 45 107 3.14.1. Route Optimization Initiation . . . . . . . . . . . 46 108 3.14.2. Relaying the NS . . . . . . . . . . . . . . . . . . 46 109 3.14.3. Processing the NS and Sending the NA . . . . . . . . 46 110 3.14.4. Relaying the NA . . . . . . . . . . . . . . . . . . 47 111 3.14.5. Processing the NA . . . . . . . . . . . . . . . . . 48 112 3.14.6. Route Optimization Maintenance . . . . . . . . . . . 48 113 3.15. Neighbor Unreachability Detection (NUD) . . . . . . . . . 49 114 3.16. Mobility Management and Quality of Service (QoS) . . . . 50 115 3.16.1. Mobility Update Messaging . . . . . . . . . . . . . 51 116 3.16.2. Announcing Link-Layer Address and/or QoS Preference 117 Changes . . . . . . . . . . . . . . . . . . . . . . 52 118 3.16.3. Bringing New Links Into Service . . . . . . . . . . 52 119 3.16.4. Removing Existing Links from Service . . . . . . . . 52 120 3.16.5. Moving to a New Server . . . . . . . . . . . . . . . 52 121 3.17. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 54 122 3.17.1. Source-Specific Multicast (SSM) . . . . . . . . . . 54 123 3.17.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 55 124 3.17.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 56 125 3.18. Operation over Multiple OMNI Links . . . . . . . . . . . 56 126 3.19. DNS Considerations . . . . . . . . . . . . . . . . . . . 57 127 3.20. Transition Considerations . . . . . . . . . . . . . . . . 57 128 3.21. Detecting and Reacting to Server and Bridge Failures . . 58 129 3.22. AERO Clients on the Open Internet . . . . . . . . . . . . 59 130 3.22.1. Use of SEND and CGA . . . . . . . . . . . . . . . . 61 131 3.23. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . 63 132 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 63 133 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 63 134 6. Security Considerations . . . . . . . . . . . . . . . . . . . 64 135 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 66 136 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 67 137 8.1. Normative References . . . . . . . . . . . . . . . . . . 67 138 8.2. Informative References . . . . . . . . . . . . . . . . . 69 139 Appendix A. Non-Normative Considerations . . . . . . . . . . . . 74 140 A.1. Implementation Strategies for Route Optimization . . . . 74 141 A.2. Implicit Mobility Management . . . . . . . . . . . . . . 75 142 A.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 75 143 A.4. AERO Critical Infrastructure Considerations . . . . . . . 76 144 A.5. AERO Server Failure Implications . . . . . . . . . . . . 76 145 A.6. AERO Client / Server Architecture . . . . . . . . . . . . 77 146 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 79 147 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 80 149 1. Introduction 151 Asymmetric Extended Route Optimization (AERO) fulfills the 152 requirements of Distributed Mobility Management (DMM) [RFC7333] and 153 route optimization [RFC5522] for aeronautical networking and other 154 network mobility use cases such as intelligent transportation 155 systems. AERO is an internetworking and mobility management service 156 based on the Overlay Multilink Network Interface (OMNI) 157 [I-D.templin-6man-omni-interface] Non-Broadcast, Multiple Access 158 (NBMA) virtual link model. The OMNI link is a virtual overlay 159 configured over one or more underlying Internetworks, and nodes on 160 the link can exchange IP packets via tunneling. The OMNI Adaptation 161 Layer (OAL) supports multilink operation for increased reliability, 162 bandwidth optimization and traffic path selection while accommodating 163 Maximum Transmission Unit (MTU) diversity. 165 The AERO service comprises Clients, Proxys, Servers and Relays that 166 are seen as OMNI link neighbors as well as Bridges that interconnect 167 OMNI link segments. Each node's OMNI interface uses an IPv6 link- 168 local address format that supports operation of the IPv6 Neighbor 169 Discovery (ND) protocol [RFC4861] and links ND to IP forwarding. A 170 node's OMNI interface can be configured over multiple underlying 171 interfaces, and may therefore appear as a single interface with 172 multiple link-layer addresses. Each link-layer address is subject to 173 change due to mobility and/or QoS fluctuations, and link-layer 174 address changes are signaled by ND messaging the same as for any IPv6 175 link. 177 AERO provides a cloud-based service where mobile nodes may use any 178 Server acting as a Mobility Anchor Point (MAP) and fixed nodes may 179 use any Relay on the link for efficient communications. Fixed nodes 180 forward packets destined to other AERO nodes to the nearest Relay, 181 which forwards them through the cloud. A mobile node's initial 182 packets are forwarded through the Server, while direct routing is 183 supported through asymmetric extended route optimization while data 184 packets are flowing. Both unicast and multicast communications are 185 supported, and mobile nodes may efficiently move between locations 186 while maintaining continuous communications with correspondents and 187 without changing their IP Address. 189 AERO Bridges are interconnected in a secured private BGP overlay 190 routing instance using encapsulation to provide a hybrid routing/ 191 bridging service that joins the underlying Internetworks of multiple 192 disjoint administrative domains into a single unified OMNI link. 193 Each OMNI link instance is characterized by the set of Mobility 194 Service Prefixes (MSPs) common to all mobile nodes. The link extends 195 to the point where a Relay/Server is on the optimal route from any 196 correspondent node on the link, and provides a conduit between the 197 underlying Internetwork and the OMNI link. To the underlying 198 Internetwork, the Relay/Server is the source of a route to the MSP, 199 and hence uplink traffic to the mobile node is naturally routed to 200 the nearest Relay/Server. 202 AERO assumes the use of PIM Sparse Mode in support of multicast 203 communication. In support of Source Specific Multicast (SSM) when a 204 Mobile Node is the source, AERO route optimization ensures that a 205 shortest-path multicast tree is established with provisions for 206 mobility and multilink operation. In all other multicast scenarios 207 there are no AERO dependencies. 209 AERO was designed for aeronautical networking for both manned and 210 unmanned aircraft, where the aircraft is treated as a mobile node 211 that can connect an Internet of Things (IoT). AERO is also 212 applicable to a wide variety of other use cases. For example, it can 213 be used to coordinate the Virtual Private Network (VPN) links of 214 mobile nodes (e.g., cellphones, tablets, laptop computers, etc.) that 215 connect into a home enterprise network via public access networks 216 using services such as OpenVPN [OVPN]. It can also be used to 217 facilitate vehicular and pedestrian communications services for 218 intelligent transportation systems. Other applicable use cases are 219 also in scope. 221 The following numbered sections present the AERO specification. The 222 appendices at the end of the document are non-normative. 224 2. Terminology 226 The terminology in the normative references applies; especially, the 227 terminology in the OMNI specification 228 [I-D.templin-6man-omni-interface] is used extensively throughout. 229 The following terms are defined within the scope of this document: 231 IPv6 Neighbor Discovery (ND) 232 an IPv6 control message service for coordinating neighbor 233 relationships between nodes connected to a common link. AERO uses 234 the ND service specified in [RFC4861]. 236 IPv6 Prefix Delegation 237 a networking service for delegating IPv6 prefixes to nodes on the 238 link. The nominal service is DHCPv6 [RFC8415], however alternate 239 services (e.g., based on ND messaging) are also in scope. Most 240 notably, a minimal form of prefix delegation known as "prefix 241 registration" can be used if the Client knows its prefix in 242 advance and can represent it in the IPv6 source address of an ND 243 message. 245 Access Network (ANET) 246 a node's first-hop data link service network (e.g., a radio access 247 network, cellular service provider network, corporate enterprise 248 network, etc.) that often provides link-layer security services 249 such as IEEE 802.1X and physical-layer security prevent 250 unauthorized access internally and with border network-layer 251 security services such as firewalls and proxies that prevent 252 unauthorized outside access. 254 ANET interface 255 a node's attachment to a link in an ANET. 257 Internetwork (INET) 258 a connected IP network topology with a coherent routing and 259 addressing plan and that provides a transit backbone service for 260 ANET end systems. INETs also provide an underlay service over 261 which the AERO virtual link is configured. Example INETs include 262 corporate enterprise networks, aviation networks, and the public 263 Internet itself. When there is no administrative boundary between 264 an ANET and the INET, the ANET and INET are one and the same. 266 INET Partition 267 frequently, INETs such as large corporate enterprise networks are 268 sub-divided internally into separate isolated partitions. Each 269 partition is fully connected internally but disconnected from 270 other partitions, and there is no requirement that separate 271 partitions maintain consistent Internet Protocol and/or addressing 272 plans. (Each INET partition is seen as a separate OMNI link 273 segment as discussed below.) 275 INET interface 276 a node's attachment to a link in an INET. 278 INET address 279 an IP address assigned to a node's interface connection to an 280 INET. 282 INET encapsulation 283 the encapsulation of a packet in an outer header or headers that 284 can be routed within the scope of the local INET partition. 286 OMNI link 287 the same as defined in [I-D.templin-6man-omni-interface], and 288 manifested by IPv6 encapsulation [RFC2473]. The OMNI link spans 289 underlying INET segments joined by virtual bridges in a spanning 290 tree the same as a bridged campus LAN. AERO nodes on the OMNI 291 link appear as single-hop neighbors even though they may be 292 separated by multiple underlying INET hops, and can use Segment 293 Routing [RFC8402] to cause packets to visit selected waypoints on 294 the link. 296 OMNI Interface 297 a node's attachment to an OMNI link. Since the addresses assigned 298 to an OMNI interface are managed for uniqueness, OMNI interfaces 299 do not require Duplicate Address Detection (DAD) and therefore set 300 the administrative variable 'DupAddrDetectTransmits' to zero 301 [RFC4862]. 303 OMNI Adaptation Layer (OAL) 304 an OMNI interface process whereby packets admitted into the 305 interface are wrapped in a mid-layer IPv6 header and fragmented/ 306 reassembled if necessary to support the OMNI link Maximum 307 Transmission Unit (MTU). The OAL is also responsible for 308 generating MTU-related control messages as necessary, and for 309 providing addressing context for spanning multiple segments of a 310 bridged OMNI link. 312 OMNI Link-Local Address (LLA) 313 a link local IPv6 address per [RFC4291] constructed as specified 314 in Section 3.2.2. 316 OMNI Unique-Local Address (ULA) 317 a unique local IPv6 address per [RFC4193] constructed as specified 318 in Section 3.2.2. OMNI ULAs are statelessly derived from OMNI 319 LLAs, and vice-versa. 321 underlying interface 322 an ANET or INET interface over which an OMNI interface is 323 configured. 325 Mobility Service Prefix (MSP) 326 an IP prefix assigned to the OMNI link and from which more- 327 specific Mobile Network Prefixes (MNPs) are derived. 329 Mobile Network Prefix (MNP) 330 an IP prefix allocated from an MSP and delegated to an AERO Client 331 or Relay. 333 AERO node 334 a node that is connected to an OMNI link and participates in the 335 AERO internetworking and mobility service. 337 AERO Client ("Client") 338 an AERO node that connects over one or more underlying interfaces 339 and requests MNP delegation/registration service from AERO 340 Servers. The Client assigns a Client LLA to the OMNI interface 341 for use in ND exchanges with other AERO nodes and forwards packets 342 to correspondents according to OMNI interface neighbor cache 343 state. 345 AERO Server ("Server") 346 an INET node that configures an OMNI interface to provide default 347 forwarding and mobility/multilink services for AERO Clients. The 348 Server assigns an administratively-provisioned LLA to its OMNI 349 interface to support the operation of the ND services, and 350 advertises all of its associated MNPs via BGP peerings with 351 Bridges. 353 AERO Relay ("Relay") 354 an AERO Server that also provides forwarding services between 355 nodes reached via the OMNI link and correspondents on other links. 356 AERO Relays are provisioned with MNPs (i.e., the same as for an 357 AERO Client) and run a dynamic routing protocol to discover any 358 non-MNP IP routes. In both cases, the Relay advertises the MSP(s) 359 to its downstream networks, and distributes all of its associated 360 MNPs and non-MNP IP routes via BGP peerings with Bridges (i.e., 361 the same as for an AERO Server). 363 AERO Bridge ("Bridge") 364 a node that provides hybrid routing/bridging services (as well as 365 a security trust anchor) for nodes on an OMNI link. As a router, 366 the Bridge forwards packets using standard IP forwarding. As a 367 bridge, the Bridge forwards packets over the OMNI link without 368 decrementing the IPv6 Hop Limit. AERO Bridges peer with Servers 369 and other Bridges to discover the full set of MNPs for the link as 370 well as any non-MNPs that are reachable via Relays. 372 AERO Proxy ("Proxy") 373 a node that provides proxying services between Clients in an ANET 374 and Servers in external INETs. The AERO Proxy is a conduit 375 between the ANET and external INETs in the same manner as for 376 common web proxies, and behaves in a similar fashion as for ND 377 proxies [RFC4389]. A node may be configured to act as either a 378 Proxy and/or a Server, depending on Client Server selection 379 criteria. 381 ingress tunnel endpoint (ITE) 382 an OMNI interface endpoint that injects encapsulated packets into 383 an OMNI link. 385 egress tunnel endpoint (ETE) 386 an OMNI interface endpoint that receives encapsulated packets from 387 an OMNI link. 389 link-layer address 390 an IP address used as an encapsulation header source or 391 destination address from the perspective of the OMNI interface. 392 When an upper layer protocol (e.g., UDP) is used as part of the 393 encapsulation, the port number is also considered as part of the 394 link-layer address. 396 network layer address 397 the source or destination address of an encapsulated IP packet 398 presented to the OMNI interface. 400 end user network (EUN) 401 an internal virtual or external edge IP network that an AERO 402 Client or Relay connects to the rest of the network via the OMNI 403 interface. The Client/Relay sees each EUN as a "downstream" 404 network, and sees the OMNI interface as the point of attachment to 405 the "upstream" network. 407 Mobile Node (MN) 408 an AERO Client and all of its downstream-attached networks that 409 move together as a single unit, i.e., an end system that connects 410 an Internet of Things. 412 Mobile Router (MR) 413 a MN's on-board router that forwards packets between any 414 downstream-attached networks and the OMNI link. 416 Route Optimization Source (ROS) 417 the AERO node nearest the source that initiates route 418 optimization. The ROS may be a Server or Proxy acting on behalf 419 of the source Client. 421 Route Optimization responder (ROR) 422 the AERO node nearest the target destination that responds to 423 route optimization requests. The ROR may be a Server acting on 424 behalf of a target MNP Client, or a Relay for a non-MNP 425 destination. 427 MAP List 428 a geographically and/or topologically referenced list of addresses 429 of all Servers within the same OMNI link. There is a single MAP 430 list for the entire OMNI link. 432 Distributed Mobility Management (DMM) 433 a BGP-based overlay routing service coordinated by Servers and 434 Bridges that tracks all Server-to-Client associations. 436 Mobility Service (MS) 437 the collective set of all Servers, Proxys, Bridges and Relays that 438 provide the AERO Service to Clients. 440 Mobility Service Endpoint MSE) 441 an individual Server, Proxy, Bridge or Relay in the Mobility 442 Service. 444 Throughout the document, the simple terms "Client", "Server", 445 "Bridge", "Proxy" and "Relay" refer to "AERO Client", "AERO Server", 446 "AERO Bridge", "AERO Proxy" and "AERO Relay", respectively. 447 Capitalization is used to distinguish these terms from other common 448 Internetworking uses in which they appear without capitalization. 450 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 451 the names of node variables, messages and protocol constants) is used 452 throughout this document. The terms "All-Routers multicast", "All- 453 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 454 anycast" are defined in [RFC4291]. Also, the term "IP" is used to 455 generically refer to either Internet Protocol version, i.e., IPv4 456 [RFC0791] or IPv6 [RFC8200]. 458 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 459 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 460 "OPTIONAL" in this document are to be interpreted as described in BCP 461 14 [RFC2119][RFC8174] when, and only when, they appear in all 462 capitals, as shown here. 464 3. Asymmetric Extended Route Optimization (AERO) 466 The following sections specify the operation of IP over OMNI links 467 using the AERO service: 469 3.1. AERO Node Types 471 AERO Clients are Mobile Nodes (MNs) that connect via underlying 472 interfaces with addresses that may change when the Client moves to a 473 new network connection point. AERO Clients register their Mobile 474 Network Prefixes (MNPs) with the AERO service, and distribute the 475 MNPs to nodes on EUNs. AERO Bridges, Servers, Proxys and Relays are 476 critical infrastructure elements in fixed (i.e., non-mobile) INET 477 deployments and hence have permanent and unchanging INET addresses. 478 Together, they constitute the AERO service which provides an OMNI 479 link virtual overlay for connecting AERO Clients. 481 AERO Bridges provide hybrid routing/bridging services (as well as a 482 security trust anchor) for nodes on an OMNI link. Bridges use 483 standard IPv6 routing to forward packets both within the same INET 484 partitions and between disjoint INET partitions based on a mid-layer 485 IPv6 encapsulation per [RFC2473]. The inner IP layer experiences a 486 virtual bridging service since the inner IP TTL/Hop Limit is not 487 decremented during forwarding. Each Bridge also peers with Servers 488 and other Bridges in a dynamic routing protocol instance to provide a 489 Distributed Mobility Management (DMM) service for the list of active 490 MNPs (see Section 3.2.3). Bridges present the OMNI link as a set of 491 one or more Mobility Service Prefixes (MSPs) and configure secured 492 tunnels with Servers, Relays, Proxys and other Bridges; they further 493 maintain IP forwarding table entries for each MNP and any other 494 reachable non-MNP prefixes. 496 AERO Servers provide default forwarding and mobility/multilink 497 services for AERO Client Mobile Nodes (MNs). Each Server also peers 498 with Bridges in a dynamic routing protocol instance to advertise its 499 list of associated MNPs (see Section 3.2.3). Servers facilitate 500 prefix delegation/registration exchanges with Clients, where each 501 delegated prefix becomes an MNP taken from an MSP. Servers forward 502 packets between OMNI interface neighbors and track each Client's 503 mobility profiles. Servers may further act as Servers for some sets 504 of Clients and as Proxies for others. 506 AERO Proxys provide a conduit for ANET Clients to associate with 507 Servers in external INETs. Client and Servers exchange control plane 508 messages via the Proxy acting as a bridge between the ANET/INET 509 boundary. The Proxy forwards data packets between Clients and the 510 OMNI link according to forwarding information in the neighbor cache. 511 The Proxy function is specified in Section 3.13. Proxys may further 512 act as Proxys for some sets of Clients and as Servers for others. 514 AERO Relays are Servers that provide forwarding services between the 515 OMNI interface and INET/EUN interfaces. Relays are provisioned with 516 MNPs the same as for an AERO Client, and also run a dynamic routing 517 protocol to discover any non-MNP IP routes. The Relay advertises the 518 MSP(s) to its connected networks, and distributes all of its 519 associated MNPs and non-MNP IP routes via BGP peerings with Bridges 521 3.2. The AERO Service over OMNI Links 523 3.2.1. AERO/OMNI Reference Model 525 Figure 1 presents the basic OMNI link reference model: 527 +----------------+ 528 | AERO Bridge B1 | 529 | Nbr: S1, S2, P1| 530 |(X1->S1; X2->S2)| 531 | MSP M1 | 532 +-+---------+--+-+ 533 +--------------+ | Secured | | +--------------+ 534 |AERO Server S1| | tunnels | | |AERO Server S2| 535 | Nbr: C1, B1 +-----+ | +-----+ Nbr: C2, B1 | 536 | default->B1 | | | default->B1 | 537 | X1->C1 | | | X2->C2 | 538 +-------+------+ | +------+-------+ 539 | OMNI link | | 540 X===+===+===================+==)===============+===+===X 541 | | | | 542 +-----+--------+ +--------+--+-----+ +--------+-----+ 543 |AERO Client C1| | AERO Proxy P1 | |AERO Client C2| 544 | Nbr: S1 | |(Proxy Nbr Cache)| | Nbr: S2 | 545 | default->S1 | +--------+--------+ | default->S2 | 546 | MNP X1 | | | MNP X2 | 547 +------+-------+ .--------+------. +-----+--------+ 548 | (- Proxyed Clients -) | 549 .-. `---------------' .-. 550 ,-( _)-. ,-( _)-. 551 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 552 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 553 `-(______)-' +-------+ +-------+ `-(______)-' 555 Figure 1: AERO/OMNI Reference Model 557 In this model: 559 o the OMNI link is an overlay network service configured over one or 560 more underlying INET partitions which may be managed by different 561 administrative authorities and have incompatible protocols and/or 562 addressing plans. 564 o AERO Bridge B1 aggregates Mobility Service Prefix (MSP) M1, 565 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 566 via BGP peerings over secured tunnels to Servers (S1, S2). 567 Bridges connect the disjoint segments of a partitioned OMNI link. 569 o AERO Servers/Relays S1 and S2 configure secured tunnels with 570 Bridge B1 and also provide mobility, multilink and default router 571 services for their associated Clients C1 and C2. 573 o AERO Clients C1 and C2 associate with Servers S1 and S2, 574 respectively. They receive Mobile Network Prefix (MNP) 575 delegations X1 and X2, and also act as default routers for their 576 associated physical or internal virtual EUNs. Simple hosts H1 and 577 H2 attach to the EUNs served by Clients C1 and C2, respectively. 579 o AERO Proxy P1 configures a secured tunnel with Bridge B1 and 580 provides proxy services for AERO Clients in secured enclaves that 581 cannot associate directly with other OMNI link neighbors. 583 An OMNI link configured over a single INET appears as a single 584 unified link with a consistent underlying network addressing plan. 585 In that case, all nodes on the link can exchange packets via simple 586 INET encapsulation, since the underlying INET is connected. In 587 common practice, however, an OMNI link may be partitioned into 588 multiple "segments", where each segment is a distinct INET 589 potentially managed under a different administrative authority (e.g., 590 as for worldwide aviation service providers such as ARINC, SITA, 591 Inmarsat, etc.). Individual INETs may also themselves be partitioned 592 internally, in which case each internal partition is seen as a 593 separate segment. 595 The addressing plan of each segment is consistent internally but will 596 often bear no relation to the addressing plans of other segments. 597 Each segment is also likely to be separated from others by network 598 security devices (e.g., firewalls, proxies, packet filtering 599 gateways, etc.), and in many cases disjoint segments may not even 600 have any common physical link connections. Therefore, nodes can only 601 be assured of exchanging packets directly with correspondents in the 602 same segment, and not with those in other segments. The only means 603 for joining the segments therefore is through inter-domain peerings 604 between AERO Bridges. 606 The same as for traditional campus LANs, multiple OMNI link segments 607 can be joined into a single unified link via a virtual bridging 608 service using the OMNI Adaptation Layer (OAL) which inserts a mid- 609 layer IPv6 encpasulation per [RFC2473] that supports inter-segment 610 forwarding (i.e., bridging) without decrementing the network-layer 611 TTL/Hop Limit. This bridging of OMNI link segments is shown in 612 Figure 2: 614 . . . . . . . . . . . . . . . . . . . . . . . 615 . . 616 . .-(::::::::) . 617 . .-(::::::::::::)-. +-+ . 618 . (:::: Segment A :::)--|B|---+ . 619 . `-(::::::::::::)-' +-+ | . 620 . `-(::::::)-' | . 621 . | . 622 . .-(::::::::) | . 623 . .-(::::::::::::)-. +-+ | . 624 . (:::: Segment B :::)--|B|---+ . 625 . `-(::::::::::::)-' +-+ | . 626 . `-(::::::)-' | . 627 . | . 628 . .-(::::::::) | . 629 . .-(::::::::::::)-. +-+ | . 630 . (:::: Segment C :::)--|B|---+ . 631 . `-(::::::::::::)-' +-+ | . 632 . `-(::::::)-' | . 633 . | . 634 . ..(etc).. x . 635 . . 636 . . 637 . <- OMNI link Bridged by encapsulation -> . 638 . . . . . . . . . . . . . .. . . . . . . . . 640 Figure 2: Bridging OMNI Link Segments 642 Bridges, Servers, Relays and Proxys connect via secured INET tunnels 643 over their respecitve segments in a spanning tree topology rooted at 644 the Bridges. The secured spanning tree supports strong 645 authentication for IPv6 ND control messages and may also be used to 646 convey the initial data packets in a flow. Route optimization can 647 then be employed to cause data packets to take more direct paths 648 between OMNI link neighbors without having to strictly follow the 649 spanning tree. 651 3.2.2. Link-Local Addresses (LLAs) and Unique Local Addresses (ULAs) 653 AERO nodes on OMNI links use the Link-Local Address (LLA) prefix 654 fe80::/10 [RFC4193] to assign LLAs used for network-layer addresses 655 in IPv6 ND and data messages. They also use the Unique Local Address 656 (ULA) prefix fc80::/10 [RFC4193] to form ULAs used for OAL header 657 source and desitnation addresses. See 658 [I-D.templin-6man-omni-interface] for a full specification of the 659 LLAs and ULAs used by AERO nodes on OMNI links. 661 For routing system organization (see Section 3.2.3), ULAs are 662 organized in partition prefixes, e.g., fc80::1000/116. For each such 663 partition prefix, the Bridge(s) that connect that segment assign the 664 all-zero's address of the prefix as a Subnet Router Anycast address. 665 For example, the Subnet Router Anycast address for fc80::1000/116 is 666 simply fc80::1000. 668 3.2.3. AERO Routing System 670 The AERO routing system comprises a private instance of the Border 671 Gateway Protocol (BGP) [RFC4271] that is coordinated between Bridges 672 and Servers and does not interact with either the public Internet BGP 673 routing system or any underlying INET routing systems. 675 In a reference deployment, each Server is configured as an Autonomous 676 System Border Router (ASBR) for a stub Autonomous System (AS) using 677 an AS Number (ASN) that is unique within the BGP instance, and each 678 Server further uses eBGP to peer with one or more Bridges but does 679 not peer with other Servers. Each INET of a multi-segment OMNI link 680 must include one or more Bridges, which peer with the Servers and 681 Proxys within that INET. All Bridges within the same INET are 682 members of the same hub AS using a common ASN, and use iBGP to 683 maintain a consistent view of all active MNPs currently in service. 684 The Bridges of different INETs peer with one another using eBGP. 686 Bridges advertise the OMNI link's MSPs and any non-MNP routes to each 687 of their Servers. This means that any aggregated non-MNPs (including 688 "default") are advertised to all Servers. Each Bridge configures a 689 black-hole route for each of its MSPs. By black-holing the MSPs, the 690 Bridge will maintain forwarding table entries only for the MNPs that 691 are currently active, and packets destined to all other MNPs will 692 correctly incur Destination Unreachable messages due to the black- 693 hole route. In this way, Servers have only partial topology 694 knowledge (i.e., they know only about the MNPs of their directly 695 associated Clients) and they forward all other packets to Bridges 696 which have full topology knowledge. 698 Each OMNI link segment assigns a unique sub-prefix of fc80::/96 known 699 as the ULA partition prefix. For example, a first segment could 700 assign fc80::1000/116, a second could assign fc80::2000/116, a third 701 could assign fc80::3000/116, etc. The administrative authorities for 702 each segment must therefore coordinate to assure mutually-exclusive 703 partiton prefix assignments, but internal provisioning of each prefix 704 is an independent local consideration for each administrative 705 authority. 707 ULA partition prefixes are statitcally represented in Bridge 708 forwarding tables. Bridges join multiple segments into a unified 709 OMNI link over multiple diverse administrative domains. They support 710 a bridging function by first establishing forwarding table entries 711 for their partiion prefixes either via standard BGP routing or static 712 routes. For example, if three Bridges ('A', 'B' and 'C') from 713 different segments serviced fc80::1000/116, fc80::2000/116 and 714 fc80::3000/116 respectively, then the forwarding tables in each 715 Bridge are as follows: 717 A: fc80::1000/116->local, fc80::2000/116->B, fc80::3000/116->C 719 B: fc80::1000/116->A, fc80::2000/116->local, fc80::3000/116->C 721 C: fc80::1000/116->A, fc80::2000/116->B, fc80::3000/116->local 723 These forwarding table entries are permanent and never change, since 724 they correspond to fixed infrastructure elements in their respective 725 segments. 727 ULA Client prefixes are instead dynamically advertised in the AERO 728 routing system by Servers and Relays that provide service for their 729 corresponding MNPs. For example, if three Servers ('D', 'E' and 'F') 730 service the MNPs 2001:db8:1000:2000::/56, 2001:db8:3000:4000::/56 and 731 2001:db8:5000:6000::/56 then the routing system would include: 733 D: fc80:2001:db8:1000:2000::/72 735 E: fc80:2001:db8:3000:4000::/72 737 F: fc80:2001:db8:5000:6000::/72 739 A full discussion of the BGP-based routing system used by AERO is 740 found in [I-D.ietf-rtgwg-atn-bgp]. 742 3.2.4. OMNI Link Encapsulation 744 With the Client and partition prefixes in place in each Bridge's 745 forwarding table, control and data packets sent between AERO nodes in 746 different segments can therefore be carried over the spanning treee 747 via mid-layer encapsulation using the OMNI Adaptation Layer (OAL) 748 header. For example, when an AERO service node forwards a packet 749 with IPv6 address 2001:db8:1:2::1 to a target AERO node with IPv6 750 address 2001:db8:1000:2000::1, it first encapsulates the packet in an 751 OAL header with source address set to its own ULA address (e.g., 752 fc80::1000:2000) and destination address set to 753 fc80:2001:db8:1000:2000::. Next, it encapsulates the resulting OAL 754 packet in an INET header with source address set to its own INET 755 address (e.g., 192.0.2.100) and destination set to the INET address 756 of a Bridge (e.g., 192.0.2.1). 758 OAL encapsulation is based on Generic Packet Tunneling in IPv6 759 [RFC2473]; the encapsulation format in the above example is shown in 760 Figure 3: 762 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 763 | INET Header | 764 | src = 192.0.2.100 | 765 | dst = 192.0.2.1 | 766 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 767 | OAL Header | 768 | src = fc80::1000:2000 | 769 | dst=fc80:2001:db8:1000:2000:: | 770 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 771 | Inner IP Header | 772 | src = 2001:db8:1:2::1 | 773 | dst = 2001:db8:1000:2000::1 | 774 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 775 | | 776 ~ ~ 777 ~ Inner Packet Body ~ 778 ~ ~ 779 | | 780 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 782 Figure 3: SPAN Encapsulation 784 In this format, the inner IP header and packet body are the original 785 IP packet, the OAL header is an IPv6 header prepared according to 786 [RFC2473], and the INET header is prepared as discussed in 787 Section 3.6. 789 This gives rise to a routing system that contains both Client prefix 790 routes that may change dynamically due to regional node mobility and 791 partion prefix routes that rarely if ever change. The Bridges can 792 therefore provide link-layer bridging by sending packets over the 793 spanning tree instead of network-layer routing according to MNP 794 routes. As a result, opportunities for packet loss due to node 795 mobility between different segments are mitigated. 797 In normal operations, IPv6 ND messages are conveyed over secured 798 paths between OMNI link neighbors so that specific Proxys, Servers or 799 Relays can be addressed without being subject to mobility events. 800 Conversely, only the first few packets destined to Clients need to 801 traverse secured paths until route optimization can determine a more 802 direct path. 804 3.2.5. Segment Routing Topologies (SRTs) 806 The 16-bit sub-prefixes of fc80::/10 identify up to 64 distinct 807 Segment Routing Topologies (SRTs). Each SRT is a mutually-exclusive 808 OMNI link overlay instance using a mutually-exclusive set of ULAs, 809 and emulates a Virtual LAN (VLAN) service for the OMNI link. In some 810 cases (e.g., when redundant topologies are needed for fault tolerance 811 and reliability) it may be beneficial to deploy multiple SRTs that 812 act as independent overlay instances. A communication failure in one 813 instance therefore will not affect communications in other instances. 815 Each SRT is identified by a distinct value in bits 10-15 of fc80::10, 816 i.e., as fc80::/16, fc81::/16, fc82::/16, etc. This document asserts 817 that up to four SRTs provide a level of safety sufficient for 818 critical communications such as civil aviation. Each SRT is 819 designated with a color that identifies a different OMNI link 820 instance as follows: 822 o Red - corresponds to fc80::/16 824 o Green - corresponds to fc81::/16 826 o Blue-1 - corresponds to fc82::/16 828 o Blue-2 - corresponds to fc83::/16 830 o fc84::/16 through fcbf::/16 are available for additional SRTs. 832 Each OMNI interface assigns an anycast ULA corresponding to its SRT 833 prefix. For example, the anycast ULA for the Green SRT is simply 834 fc81::. The anycast ULA is used for OMNI interface determination in 835 Safety-Based Multilink (SBM) as discussed in 836 [I-D.templin-6man-omni-interface]. Each OMNI interface further 837 applies Performance-Based Multilink (PBM) internally. 839 3.2.6. Segment Routing To the OMNI Link 841 An original IPv6 source can direct a packet to an OMNI link Client by 842 including a Segment Routing Header (SRH) with the anycast ULA for the 843 selected SRT as either the IPv6 destination or as an intermediate hop 844 within the SRH. This allows the original source to determine the 845 specific topology a packet will traverse when there may be multiple 846 alternatives to choose from. Since the SRH contains no useful 847 information for the destination, the Client may elect to delete the 848 SRH before forwarding in order to reduce overhead. This form of 849 Segment Routing supports Safety-Based Multilink (SBM), and can be 850 exercised through general-purpose SRH types such as [RFC8754]. 852 3.2.7. Segment Routing Within the OMNI Link 854 AERO nodes that insert an OAL header can use Segment Routing within 855 the OMNI link when necessary to influence the path of packets 856 destined to targets in remote segments without requiring all packets 857 to traverse strict spanning tree paths. 859 When a Client, Proxy or Server has a packet to send to a target 860 discovered through route optimization located in the same OMNI link 861 segment, it encapsulates the packet in an OAL header with the ULA of 862 the target as the destination address if fragmentation is necessary; 863 otherwise, it may omit the OAL header. The node then uses the 864 target's Link Layer Address (L2ADDR) information for INET 865 encapsulation without including an SRH. 867 When a Client, Proxy or Server has a packet to send to a route 868 optimization target located in a remote OMNI link segment, it 869 encapsulates the packet in an OAL header with its own ULA as the 870 source address. The node then SHOULD include an SRH [RFC8754] while 871 forwarding the packet to a Bridge. 873 When the SRH is omitted, the node sets the OAL header destination 874 address to the ULA of the target Client/Proxy/Server and packet 875 forwarding is via spanning tree paths. When the SRH is included, the 876 node first sets the destination address to the ULA Subnet Router 877 Anycast address of the remote segment and sets the lower 32 bits of 878 the ULA of the target's Proxy/Server as the Last Hop Segment (LHS). 879 The node also includes an AERO Route Optimization specification in 880 the SRH TLV section as shown in Figure 4: 882 0 1 2 3 883 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 1 884 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 885 | Type=TBD | Length | MNPlen|V| FMT | MNP[1] | 886 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 887 | MNP[2] | MNP[3] | ... | MNP[i] | 888 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 889 ~ Link Layer Address (L2ADDR) ~ 890 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 892 Figure 4: AERO Route Optimization SRH TLV 894 In this format: 896 o Type is TBD to be assigned according to the Segment Routing Header 897 TLV registry [RFC8754]. 899 o Length is the length of the body of the TLV in bytes, excluding 900 the Type and Length fields. 902 o MNPlen encodes a value between 0 and 15 that, when added to 1, 903 indicates the number of octets of the IPv4/IPv6 MNP prefix that 904 follows. (For example, when MNPlen encodes the value 2 the number 905 of octets 'i' is set to the value 3.) 907 o V indicates the IP protocol version of the MNP that follows. V is 908 set to 0 for IPv4 or 1 for IPv6. 910 o FMT is a three bit code that determines the context and format of 911 the L2ADDR exactly as specified in the OMNI Interface Attributes 912 sub-option [I-D.templin-6man-omni-interface]. 914 o MNP[1], MNP[2], etc. up to MNP[i] encode the leading 'i' octets of 915 the MNP, beginning with the most significant octet followed by the 916 next most significant octet, etc. The number of MNP octets to be 917 included is determined by the number of trailing zero octets in 918 the prefix. For example, for the IPv6 MNP 2001:db8:1:2::/64, 'i' 919 is set to 8 and only the leftmost 8 octets of the MNP are 920 included. In the same way, for the IPv4 MNP 192.0.2/24, 'i' is 921 set to 3 and only the leftmost 3 octets of the MNP are included. 923 o Link Layer Address (L2ADDR) is a UDP Port Number and IP address 924 encoded according to FMT exactly as specified in 925 [I-D.templin-6man-omni-interface]. 927 The node then forwards the packet via a local Bridge, which will 928 eventually direct it to a Bridge on the same segment as the target. 930 When a Bridge receives a packet with Segments Left=1 and with LHS on 931 a local segment, it checks to see if there is an AERO Route 932 Optimization TLV. If so, the Bridge creates a ULA destination 933 according to FMT. If FMT indicates that L2ADDR corresponds to a 934 target Proxy/Server, the Bridge concatenates the SRT fc*::/96 prefix 935 with the 32 bit LHS value to form the ULA destination. Otherwise, 936 the Bridge concatenates the SRT fc*::/16 prefix with the leading 937 MNPlen octets of the MNP and sets the remaining rightmost bits to 0 938 to form a Subnet Router Anycast ULA destination. The Bridge then 939 writes the ULA into the OAL header destination address and 940 encapsulates the packet in an INET header with the target's L2ADDR as 941 the destination then forwards the packet. Since the SRH contains no 942 useful information for the destination, the Bridge may elect to 943 delete the SRH before forwarding in order to reduce overhead. 945 In this way, the Bridge participates in route optimization to reduce 946 traffic load and suboptimal routing through strict spanning tree 947 paths. Note that if the Bridge does not recognize the AERO Route 948 Optimization TLV, it instead places the SRT fc*::/96 prefix 949 concatenated with the 32 bit LHS in the IPv6 destination address and 950 forwards according to the spanning tree. (Note that this is the same 951 behavior that would occur if the AERO Route Optimization TLV were not 952 present). 954 3.2.8. Segment Routing Header Compression 956 In the Segment Routing use cases discussed above, the segment routing 957 headers must be kept to a minimum size since source and target 958 Clients may be located behind low-end wireless links (e.g., 1Mbps or 959 less). The Compressed Routing Header (CRH) 960 [I-D.bonica-6man-comp-rtg-hdr] provides a compact form that reduces 961 the header size by omitting invariant information. The CRH Helper 962 option [I-D.bonica-6man-crh-helper-opt] can be used to encode the 963 AERO Route Optimization TLV, and the final hop Bridge that performs 964 route optimization may remove the CRH and its helper before 965 encapsulating and forwarding to the target. 967 The CRH and its companion helper option are therefore seen as 968 critical architectural elements that should be quickly progressed 969 through the standards process. Implementations SHOULD use the CRH 970 and its companion helper option instead of other Routing Header types 971 whenever possible to conserve bandwidth. 973 3.3. OMNI Interface Characteristics 975 OMNI interfaces are virtual interfaces configured over one or more 976 underlying interfaces classified as follows: 978 o INET interfaces connect to an INET either natively or through one 979 or several IPv4 Network Address Translators (NATs). Native INET 980 interfaces have global IP addresses that are reachable from any 981 INET correspondent. All Server, Relay and Bridge interfaces are 982 native interfaces, as are INET-facing interfaces of Proxys. NATed 983 INET interfaces connect to a private network behind one or more 984 NATs that provide INET access. Clients that are behind a NAT are 985 required to send periodic keepalive messages to keep NAT state 986 alive when there are no data packets flowing. 988 o ANET interfaces connect to an ANET that is separated from the open 989 INET by a Proxy. Proxys can actively issue control messages over 990 the INET on behalf of the Client to reduce ANET congestion. 992 o VPNed interfaces use security encapsulation over the INET to a 993 Virtual Private Network (VPN) server that also acts as a Server or 994 Proxy. Other than the link-layer encapsulation format, VPNed 995 interfaces behave the same as Direct interfaces. 997 o Direct interfaces connect a Client directly to a Server or Proxy 998 without crossing any ANET/INET paths. An example is a line-of- 999 sight link between a remote pilot and an unmanned aircraft. The 1000 same Client considerations apply as for VPNed interfaces. 1002 OMNI interfaces use OAL encapsulation as necessary as discussed in 1003 Section 3.2.4. OMNI interfaces use link-layer encapsulation (see: 1004 Section 3.6) to exchange packets with OMNI link neighbors over INET 1005 or VPNed interfaces. OMNI interfaces do not use link-layer 1006 encapsulation over ANET and Direct underlying interfaces. 1008 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 1009 state the same as for any interface. OMNI interfaces use ND messages 1010 including Router Solicitation (RS), Router Advertisement (RA), 1011 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 1012 neighbor cache management. 1014 OMNI interfaces send ND messages with an OMNI option formatted as 1015 specified in [I-D.templin-6man-omni-interface]. The OMNI option 1016 includes prefix registration information and Interface Attributes 1017 containing link information parameters for the OMNI interface's 1018 underlying interfaces. Each OMNI option may include multiple 1019 Interface Attributes sub-options, each idenfiied by an ifIndex value. 1021 A Client's OMNI interface may be configured over multiple underlying 1022 interface connections. For example, common mobile handheld devices 1023 have both wireless local area network ("WLAN") and cellular wireless 1024 links. These links are often used "one at a time" with low-cost WLAN 1025 preferred and highly-available cellular wireless as a standby, but a 1026 simultaneous-use capability could provide benefits. In a more 1027 complex example, aircraft frequently have many wireless data link 1028 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1029 directional, etc.) with diverse performance and cost properties. 1031 If a Client's multiple underlying interfaces are used "one at a time" 1032 (i.e., all other interfaces are in standby mode while one interface 1033 is active), then ND message OMNI options include only a single 1034 Interface Attribues sub-option set to constant values. In that case, 1035 the Client would appear to have a single interface but with a 1036 dynamically changing link-layer address. 1038 If the Client has multiple active underlying interfaces, then from 1039 the perspective of ND it would appear to have multiple link-layer 1040 addresses. In that case, ND message OMNI options MAY include 1041 multiple Interface Attributes sub-options - each with values that 1042 correspond to a specific interface. Every ND message need not 1043 include Interface Attributes for all underlying interfaces; for any 1044 attributes not included, the neighbor considers the status as 1045 unchanged. 1047 Bridge, Server and Proxy OMNI interfaces may be configured over one 1048 or more secured tunnel interfaces. The OMNI interface configures 1049 both an LLA and its corresponding ULA, while the underlying secured 1050 tunnel interfaces are either unnumbered or configure the same ULA. 1051 The OMNI interface encapsulates each IP packet in an OAL header and 1052 presents the packet to the underlying secured tunnel interface. 1053 Routing protocols such as BGP that run over the OMNI interface do not 1054 employ OAL encapsulation, but rather present the routing protocol 1055 messages directly to the underlying secured tunnels while using the 1056 ULA as the source address. This distinction must be honored 1057 consistently according to each node's configuration so that the IP 1058 forwarding table will associate discovered IP routes with the correct 1059 interface. 1061 3.4. OMNI Interface Initialization 1063 AERO Servers, Proxys and Clients configure OMNI interfaces as their 1064 point of attachment to the OMNI link. AERO nodes assign the MSPs for 1065 the link to their OMNI interfaces (i.e., as a "route-to-interface") 1066 to ensure that packets with destination addresses covered by an MNP 1067 not explicitly assigned to a non-OMNI interface are directed to the 1068 OMNI interface. 1070 OMNI interface initialization procedures for Servers, Proxys, Clients 1071 and Bridges are discussed in the following sections. 1073 3.4.1. AERO Server/Relay Behavior 1075 When a Server enables an OMNI interface, it assigns an LLA/ULA 1076 appropriate for the given OMNI link segment. The Server also 1077 configures secured tunnels with one or more neighboring Bridges and 1078 engages in a BGP routing protocol session with each Bridge. 1080 The OMNI interface provides a single interface abstraction to the IP 1081 layer, but internally comprises multiple secured tunnels as well as 1082 an NBMA nexus for sending encapsulated data packets to OMNI interface 1083 neighbors. The Server further configures a service to facilitate ND 1084 exchanges with AERO Clients and manages per-Client neighbor cache 1085 entries and IP forwarding table entries based on control message 1086 exchanges. 1088 Relays are simply Servers that run a dynamic routing protocol to 1089 redistribute routes between the OMNI interface and INET/EUN 1090 interfaces (see: Section 3.2.3). The Relay provisions MNPs to 1091 networks on the INET/EUN interfaces (i.e., the same as a Client would 1092 do) and advertises the MSP(s) for the OMNI link over the INET/EUN 1093 interfaces. The Relay further provides an attachment point of the 1094 OMNI link to a non-MNP-based global topology. 1096 3.4.2. AERO Proxy Behavior 1098 When a Proxy enables an OMNI interface, it assigns an LLA/ULA and 1099 configures permanent neighbor cache entries the same as for Servers. 1100 The Proxy also configures secured tunnels with one or more 1101 neighboring Bridges and maintains per-Client neighbor cache entries 1102 based on control message exchanges. Importantly Proxys are often 1103 configured to act as Servers, and vice-versa. 1105 3.4.3. AERO Client Behavior 1107 When a Client enables an OMNI interface, it sends RS messages with ND 1108 parameters over its underlying interfaces to a Server, which returns 1109 an RA message with corresponding parameters. (The RS/RA messages may 1110 pass through a Proxy in the case of a Client's ANET interface, or 1111 through one or more NATs in the case of a Client's INET interface.) 1113 3.4.4. AERO Bridge Behavior 1115 AERO Bridges configure an OMNI interface and assign the ULA Subnet 1116 Router Anycast address for each OMNI link segment they connect to. 1117 Bridges configure secured tunnels with Servers, Proxys and other 1118 Bridges; they also configure LLAs/ULAs and permanent neighbor cache 1119 entries the same as Servers. Bridges engage in a BGP routing 1120 protocol session with a subset of the Servers and other Bridges on 1121 the spanning tree (see: Section 3.2.3). 1123 3.5. OMNI Interface Neighbor Cache Maintenance 1125 Each OMNI interface maintains a conceptual neighbor cache that 1126 includes an entry for each neighbor it communicates with on the OMNI 1127 link per [RFC4861]. OMNI interface neighbor cache entries are said 1128 to be one of "permanent", "symmetric", "asymmetric" or "proxy". 1130 Permanent neighbor cache entries are created through explicit 1131 administrative action; they have no timeout values and remain in 1132 place until explicitly deleted. AERO Bridges maintain permanent 1133 neighbor cache entries for their associated Proxys/Servers (and vice- 1134 versa). Each entry maintains the mapping between the neighbor's 1135 network-layer LLA and corresponding INET address. 1137 Symmetric neighbor cache entries are created and maintained through 1138 RS/RA exchanges as specified in Section 3.12, and remain in place for 1139 durations bounded by prefix lifetimes. AERO Servers maintain 1140 symmetric neighbor cache entries for each of their associated 1141 Clients, and AERO Clients maintain symmetric neighbor cache entries 1142 for each of their associated Servers. 1144 Asymmetric neighbor cache entries are created or updated based on 1145 route optimization messaging as specified in Section 3.14, and are 1146 garbage-collected when keepalive timers expire. AERO ROSs maintain 1147 asymmetric neighbor cache entries for active targets with lifetimes 1148 based on ND messaging constants. Asymmetric neighbor cache entries 1149 are unidirectional since only the ROS (and not the ROR) creates an 1150 entry. 1152 Proxy neighbor cache entries are created and maintained by AERO 1153 Proxys when they process Client/Server ND exchanges, and remain in 1154 place for durations bounded by ND and prefix lifetimes. AERO Proxys 1155 maintain proxy neighbor cache entries for each of their associated 1156 Clients. Proxy neighbor cache entries track the Client state and the 1157 address of the Client's associated Server(s). 1159 To the list of neighbor cache entry states in Section 7.3.2 of 1160 [RFC4861], Proxy and Server OMNI interfaces add an additional state 1161 DEPARTED that applies to symmetric and proxy neighbor cache entries 1162 for Clients that have recently departed. The interface sets a 1163 "DepartTime" variable for the neighbor cache entry to "DEPART_TIME" 1164 seconds. DepartTime is decremented unless a new ND message causes 1165 the state to return to REACHABLE. While a neighbor cache entry is in 1166 the DEPARTED state, packets destined to the target Client are 1167 forwarded to the Client's new location instead of being dropped. 1168 When DepartTime decrements to 0, the neighbor cache entry is deleted. 1169 It is RECOMMENDED that DEPART_TIME be set to the default constant 1170 value REACHABLE_TIME plus 10 seconds (40 seconds by default) to allow 1171 a window for packets in flight to be delivered while stale route 1172 optimization state may be present. 1174 When an ROR receives an authentic NS message used for route 1175 optimization, it searches for a symmetric neighbor cache entry for 1176 the target Client. The ROR then returns a solicited NA message 1177 without creating a neighbor cache entry for the ROS, but creates or 1178 updates a target Client "Report List" entry for the ROS and sets a 1179 "ReportTime" variable for the entry to REPORT_TIME seconds. The ROR 1180 resets ReportTime when it receives a new authentic NS message, and 1181 otherwise decrements ReportTime while no authentic NS messages have 1182 been received. It is RECOMMENDED that REPORT_TIME be set to the 1183 default constant value REACHABLE_TIME plus 10 seconds (40 seconds by 1184 default) to allow a window for route optimization to converge before 1185 ReportTime decrements below REACHABLE_TIME. 1187 When the ROS receives a solicited NA message response to its NS 1188 message used for route optimization, it creates or updates an 1189 asymmetric neighbor cache entry for the target network-layer and 1190 link-layer addresses. The ROS then (re)sets ReachableTime for the 1191 neighbor cache entry to REACHABLE_TIME seconds and uses this value to 1192 determine whether packets can be forwarded directly to the target, 1193 i.e., instead of via a default route. The ROS otherwise decrements 1194 ReachableTime while no further solicited NA messages arrive. It is 1195 RECOMMENDED that REACHABLE_TIME be set to the default constant value 1196 30 seconds as specified in [RFC4861]. 1198 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1199 of NS keepalives sent when a correspondent may have gone unreachable, 1200 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1201 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1202 to limit the number of unsolicited NAs that can be sent based on a 1203 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1204 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1205 same as specified in [RFC4861]. 1207 Different values for DEPART_TIME, REPORT_TIME, REACHABLE_TIME, 1208 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1209 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1210 different values are chosen, all nodes on the link MUST consistently 1211 configure the same values. Most importantly, DEPART_TIME and 1212 REPORT_TIME SHOULD be set to a value that is sufficiently longer than 1213 REACHABLE_TIME to avoid packet loss due to stale route optimization 1214 state. 1216 3.5.1. OMNI Neighbor Interface Attributes 1218 OMNI interface IPv6 ND messages include OMNI options 1219 [I-D.templin-6man-omni-interface] with Interface Attributes that 1220 provide Link-Layer Address and QoS Preference information for the 1221 neighbor's underlying interfaces. This information is stored in the 1222 neighbor cache and provides the basis for the forwarding algorithm 1223 specified in Section 3.10. The information is cumulative and 1224 reflects the union of the OMNI information from the most recent ND 1225 messages received from the neighbor; it is therefore not required 1226 that each ND message contain all neighbor information. 1228 The OMNI option Interface Attributes for each underlying interface 1229 includes a two-part "Link-Layer Address" consisting of a simple IP 1230 encapsulation address determined by the FMT and L2ADDR fields and an 1231 OAL ULA determined by the SRT and LHS fields. If the neighbor is 1232 located in the local OMNI link segment (and, if any necessary NAT 1233 state has been established) forwarding via simple IP encapsulation 1234 can be used; otherwise, OAL encapsulation must be used. Underlying 1235 interfaces are further selected based on their associated preference 1236 values "high", "medium", "low" or "disabled". 1238 Note: the OMNI option is distinct from any Source/Target Link-Layer 1239 Address Options (S/TLLAOs) that may appear in an ND message according 1240 to the appropriate IPv6 over specific link layer specification (e.g., 1241 [RFC2464]). If both an OMNI option and S/TLLAO appear, the former 1242 pertains to encapsulation addresses while the latter pertains to the 1243 native L2 address format of the underlying media. 1245 3.6. OMNI Interface Encapsulation and Re-encapsulation 1247 The OMNI Adaptation Layer (OAL) inserts a mid-layer IPv6 header known 1248 as the OAL header when necessary as discussed in the following 1249 sections. After either inserting or omitting the OAL header, the 1250 OMNI interface also inserts or omits an outer encapsulation header as 1251 discussed below. 1253 OMNI interfaces avoid outer encapsulation over Direct underlying 1254 interfaces and ANET underlying interfaces for which the first-hop 1255 access router is connected to the same underlying link. Otherwise, 1256 OMNI interfaces encapsulate packets according to whether they are 1257 entering the OMNI interface from the network layer or if they are 1258 being re-admitted into the same OMNI link they arrived on. This 1259 latter form of encapsulation is known as "re-encapsulation". 1261 For packets entering the OMNI interface from the network layer, the 1262 OMNI interface copies the "TTL/Hop Limit", "Type of Service/Traffic 1263 Class" [RFC2983], "Flow Label"[RFC6438] (for IPv6) and "Congestion 1264 Experienced" [RFC3168] values in the inner packet's IP header into 1265 the corresponding fields in the OAL and outer encapsulation 1266 header(s). 1268 For packets undergoing re-encapsulation, the OMNI interface instead 1269 copies these values from the original encapsulation header into the 1270 new encapsulation header, i.e., the values are transferred between 1271 encapsulation headers and *not* copied from the encapsulated packet's 1272 network-layer header. (Note especially that by copying the TTL/Hop 1273 Limit between encapsulation headers the value will eventually 1274 decrement to 0 if there is a (temporary) routing loop.) 1276 OMNI interfaces configured over ANET underlying interfaces which 1277 employ a different IP protocol version and/or may be located multiple 1278 IP hops from the nearest Proxy/Server use IP-in-IP encapsulation so 1279 that the inner packet can traverse the ANET. IPv6 underlying ANET 1280 interfaces use [RFC2473] encapsulation, while IPv4 interfaces use the 1281 appropriate encapsulation per one of [RFC2529][RFC5214][RFC2003]. 1283 OMNI interfaces configured over INET underlying interfaces 1284 encapsulate packets in INET headers according to the next hop 1285 determined in the forwarding algorithm in Section 3.10. If the next 1286 hop is reached via a secured tunnel, the OMNI interface uses an 1287 encapsulation format specific to the secured tunnel type (see: 1288 Section 6). If the next hop is reached via an unsecured INET 1289 interface, the OMNI interface instead uses UDP/IP encapsulation per 1290 [RFC4380] and as extended in [RFC6081]. 1292 When UDP/IP encapsulation is used, the OMNI interface next sets the 1293 UDP source port to a constant value that it will use in each 1294 successive packet it sends, and sets the UDP length field to the 1295 length of the encapsulated packet plus 8 bytes for the UDP header 1296 itself plus the length of any included extension headers or trailers. 1297 The encapsulated packet may be either IPv6 or IPv4, as distinguished 1298 by the version number found in the first four bits. 1300 For UDP/IP-encapsulated packets sent to a Server, Relay or Bridge, 1301 the OMNI interface sets the UDP destination port to 8060, i.e., the 1302 IANA-registered port number for AERO. For packets sent to a Client, 1303 the OMNI interface sets the UDP destination port to the port value 1304 stored in the neighbor cache entry for this Client. The OMNI 1305 interface finally includes/omits the UDP checksum according to 1306 [RFC6935][RFC6936]. 1308 3.7. OMNI Interface Decapsulation 1310 OMNI interfaces decapsulate packets destined either to the AERO node 1311 itself or to a destination reached via an interface other than the 1312 OMNI interface the packet was received on. When the encapsulated 1313 packet arrives in multiple OAL fragments, the OMNI interface 1314 reassembles as discussed in Section 3.9. Further decapsulation steps 1315 are performed according to the appropriate encapsulation format 1316 specification. 1318 3.8. OMNI Interface Data Origin Authentication 1320 AERO nodes employ simple data origin authentication procedures. In 1321 particular: 1323 o AERO Bridges, Servers and Proxys accept encapsulated data packets 1324 and control messages received from the (secured) spanning tree. 1326 o AERO Proxys and Clients accept packets that originate from within 1327 the same secured ANET. 1329 o AERO Clients and Relays accept packets from downstream network 1330 correspondents based on ingress filtering. 1332 o AERO Clients, Relays and Servers verify the outer UDP/IP 1333 encapsulation addresses according to [RFC4380]. 1335 AERO nodes silently drop any packets that do not satisfy the above 1336 data origin authentication procedures. Further security 1337 considerations are discussed in Section 6. 1339 3.9. OMNI Adaptation Layer and OMNI Interface MTU 1341 The OMNI interface observes the link nature of tunnels, including the 1342 Maximum Transmission Unit (MTU), Maximum Reassembly Unit (MRU) and 1343 the role of fragmentation and reassembly[I-D.ietf-intarea-tunnels]. 1344 The OMNI interface employs an OMNI Adaptation Layer (OAL) for 1345 accommodating multiple underlying links with diverse MTUs. The 1346 functions of the OAL and the OMNI interface MTU/MRU are specified in 1347 Section 5 of [I-D.templin-6man-omni-interface], with MTU/MRU both set 1348 to the constant value 9180 bytes. 1350 3.10. OMNI Interface Forwarding Algorithm 1352 IP packets enter a node's OMNI interface either from the network 1353 layer (i.e., from a local application or the IP forwarding system) or 1354 from the link layer (i.e., from an OMNI interface neighbor). All 1355 packets entering a node's OMNI interface first undergo data origin 1356 authentication as discussed in Section 3.8. Packets that satisfy 1357 data origin authentication are processed further, while all others 1358 are dropped silently. The OMNI interface OAL wraps accepted packets 1359 in an OAL header and SRH if necessary as discussed above. 1361 Packets that enter the OMNI interface from the network layer are 1362 forwarded to an OMNI interface neighbor. Packets that enter the OMNI 1363 interface from the link layer are either re-admitted into the OMNI 1364 link or forwarded to the network layer where they are subject to 1365 either local delivery or IP forwarding. In all cases, the OMNI 1366 interface itself MUST NOT decrement the network layer TTL/Hop-count 1367 since its forwarding actions occur below the network layer. 1369 OMNI interfaces may have multiple underlying interfaces and/or 1370 neighbor cache entries for neighbors with multiple underlying 1371 interfaces (see Section 3.3). The OMNI interface uses traffic 1372 classifiers (e.g., DSCP value, port number, etc.) to select an 1373 outgoing underlying interface for each packet based on the node's own 1374 QoS preferences, and also to select a destination link-layer address 1375 based on the neighbor's underlying interface with the highest 1376 preference. AERO implementations SHOULD allow for QoS preference 1377 values to be modified at runtime through network management. 1379 If multiple outgoing interfaces and/or neighbor interfaces have a 1380 preference of "high", the AERO node replicates the packet and sends 1381 one copy via each of the (outgoing / neighbor) interface pairs; 1382 otherwise, the node sends a single copy of the packet via an 1383 interface with the highest preference. AERO nodes keep track of 1384 which underlying interfaces are currently "reachable" or 1385 "unreachable", and only use "reachable" interfaces for forwarding 1386 purposes. 1388 The following sections discuss the OMNI interface forwarding 1389 algorithms for Clients, Proxys, Servers and Bridges. In the 1390 following discussion, a packet's destination address is said to 1391 "match" if it is the same as a cached address, or if it is covered by 1392 a cached prefix (which may be encoded in an LLA). 1394 3.10.1. Client Forwarding Algorithm 1396 When an IP packet enters a Client's OMNI interface from the network 1397 layer the Client searches for an asymmetric neighbor cache entry that 1398 matches the destination. If there is a match, the Client uses one or 1399 more "reachable" neighbor interfaces in the entry for packet 1400 forwarding. If there is no asymmetric neighbor cache entry, the 1401 Client instead forwards the packet toward a Server (the packet is 1402 intercepted by a Proxy if there is a Proxy on the path). The Client 1403 encapsulates the packet in an OAL header and SRH if necessary and 1404 fragments according to MTU requirements (see: Section 3.9). 1406 When an IP packet enters a Client's OMNI interface from the link- 1407 layer, if the destination matches one of the Client's MNPs or link- 1408 local addresses the Client reassembles and decapsulates as necessary 1409 and delivers the inner packet to the network layer. Otherwise, the 1410 Client drops the packet and MAY return a network-layer ICMP 1411 Destination Unreachable message subject to rate limiting (see: 1412 Section 3.11). 1414 3.10.2. Proxy Forwarding Algorithm 1416 For control messages originating from or destined to a Client, the 1417 Proxy intercepts the message and updates its proxy neighbor cache 1418 entry for the Client. The Proxy then forwards a (proxyed) copy of 1419 the control message. (For example, the Proxy forwards a proxied 1420 version of a Client's NS/RS message to the target neighbor, and 1421 forwards a proxied version of the NA/RA reply to the Client.) 1422 When the Proxy receives a data packet from a Client within the ANET, 1423 the Proxy reassembles and re-fragments if necessary then searches for 1424 an asymmetric neighbor cache entry that matches the destination and 1425 forwards as follows: 1427 o if the destination matches an asymmetric neighbor cache entry, the 1428 Proxy uses one or more "reachable" neighbor interfaces in the 1429 entry for packet forwarding using OAL encapsulation and including 1430 a SRH if necessary according to the cached link-layer address 1431 information. If the neighbor interface is in the same OMNI link 1432 segment, the Proxy forwards the packet directly to the neighbor; 1433 otherwise, it forwards the packet to a Bridge. 1435 o else, the Proxy uses OAL encapsulation and forwards the packet to 1436 a Bridge while using the ULA corresponding to the packet's 1437 destination as the OAL destination address. 1439 When the Proxy receives an encapsulated data packet from an INET 1440 neighbor or from a secured tunnel from a Bridge, it accepts the 1441 packet only if data origin authentication succeeds and if there is a 1442 proxy neighbor cache entry that matches the inner destination. Next, 1443 the Proxy reassembles the packet (if necessary) and continues 1444 processing. If the reassembly is complete and the neighbor cache 1445 state is REACHABLE, the Proxy then returns a PTB if necessary (see: 1446 Section 3.9) then either drops or forwards the packet to the Client 1447 while performing OAL encapsulation and re-fragmentation if necessary. 1448 If the neighbor cache entry state is DEPARTED, the Proxy instead 1449 changes the OAL destination address to the address of the new Server 1450 and forwards it to a Bridge while performing OAL re-fragmentation if 1451 necessary. 1453 3.10.3. Server/Relay Forwarding Algorithm 1455 For control messages destined to a target Client's LLA that are 1456 received from a secured tunnel, the Server intercepts the message and 1457 sends a Proxyed response on behalf of the Client. (For example, the 1458 Server sends a Proxyed NA message reply in response to an NS message 1459 directed to one of its associated Clients.) If the Client's neighbor 1460 cache entry is in the DEPARTED state, however, the Server instead 1461 forwards the packet to the Client's new Server as discussed in 1462 Section 3.16. 1464 When the Server receives an encapsulated data packet from an INET 1465 neighbor or from a secured tunnel, it accepts the packet only if data 1466 origin authentication succeeds. The Server then continues processing 1467 as follows: 1469 o if the network layer destination matches a symmetric neighbor 1470 cache entry in the REACHABLE state the Server prepares the packet 1471 for forwarding to the destination Client. The Server first 1472 reassembles (if necessary) and forwards the packet (while re- 1473 fragmenting if necessary) as specified in Section 3.9. 1475 o else, if the destination matches a symmetric neighbor cache entry 1476 in the DEPARETED state the Server re-encapsulates the packet and 1477 forwards it using the ULA of the Client's new Server as the OAL 1478 header destination. 1480 o else, if the destination matches an asymmetric neighbor cache 1481 entry, the Server uses one or more "reachable" neighbor interfaces 1482 in the entry for packet forwarding via the local INET if the 1483 neighbor is in the same OMNI link segment or using OAL 1484 encapsulation and Segment Routing if necessary with the final 1485 destination set to the neighbor's ULA otherwise. 1487 o else, if the destination matches a non-MNP route in the IP 1488 forwarding table or an LLA assigned to the Server's OMNI 1489 interface, the Server reassembles if necessary, decapsulates the 1490 packet and releases it to the network layer for local delivery or 1491 IP forwarding. 1493 o else, the Server drops the packet. 1495 When the Server's OMNI interface receives a data packet from the 1496 network layer or from a VPNed or Direct Client, it performs OAL 1497 encapsulation and fragmentation if necessary, then processes the 1498 packet according to the network-layer destination address as follows: 1500 o if the destination matches a symmetric or asymmetric neighbor 1501 cache entry the Server processes the packet as above. 1503 o else, the Server encapsulates the packet in an OAL header and 1504 forwards it to a Bridge using its own ULA as the source and the 1505 ULA corresponding to the destination as the destination. 1507 3.10.4. Bridge Forwarding Algorithm 1509 Bridges forward OAL-encapsulated packets over secured tunnels the 1510 same as any IP router. When the Bridge receives an OAL-encapsulated 1511 packet via a secured tunnel, it removes the outer INET header and 1512 searches for a forwarding table entry that matches the OAL 1513 destination address. The Bridge then processes the packet as 1514 follows: 1516 o if the destination matches its ULA Subnet Router Anycast address, 1517 the Bridge checks for a SRH. If there is a SRH with Segments 1518 Left=1, with the ULA of a Proxy/Server on the local segment as the 1519 LHS, and with an AERO Route Optimization TLV, the Bridge examines 1520 the FMT to determine if the target is behind a NAT. If no NAT is 1521 indicated, the Bridge copies the MNP Subnet Router Anycast address 1522 if an MNP is included (otherwise copies the Proxy/Server ULA) into 1523 the destination address then forwards the packet directly to the 1524 L2ADDR using link-layer (UDP/IP) encapsulation. If a NAT is 1525 indicated, the Bridge MAY perform NAT traversal procedures by 1526 sending bubbles per [RFC4380]. The Bridge then either applies 1527 AERO route optimization if NAT traversal procedures have been 1528 successfully applied, or forwards the packet directly to the 1529 Server. 1531 o if the destination matches one of the Bridge's own addresses, the 1532 Bridge submits the packet for local delivery. 1534 o else, if the destination matches a forwarding table entry the 1535 Bridge forwards the packet via a secured tunnel to the next hop. 1536 If the destination matches an MSP without matching an MNP, 1537 however, the Bridge instead drops the packet and returns an ICMP 1538 Destination Unreachable message subject to rate limiting (see: 1539 Section 3.11). 1541 o else, the Bridge drops the packet and returns an ICMP Destination 1542 Unreachable as above. 1544 As for any IP router, the Bridge decrements the TTL/Hop Limit when it 1545 forwards the packet. Therefore, only the Hop Limit in the OAL header 1546 is decremented, and not the TTL/Hop Limit in the inner packet header. 1547 Bridges do not insert OAL headers themselves; instead, they act as 1548 IPv6 routers and forward packets based on the destination address 1549 found in the OAL headers of packets they receive. 1551 3.11. OMNI Interface Error Handling 1553 When an AERO node admits a packet into the OMNI interface, it may 1554 receive link-layer or network-layer error indications. 1556 A link-layer error indication is an ICMP error message generated by a 1557 router in the INET on the path to the neighbor or by the neighbor 1558 itself. The message includes an IP header with the address of the 1559 node that generated the error as the source address and with the 1560 link-layer address of the AERO node as the destination address. 1562 The IP header is followed by an ICMP header that includes an error 1563 Type, Code and Checksum. Valid type values include "Destination 1564 Unreachable", "Time Exceeded" and "Parameter Problem" 1565 [RFC0792][RFC4443]. (OMNI interfaces ignore all link-layer IPv4 1566 "Fragmentation Needed" and IPv6 "Packet Too Big" messages since they 1567 only emit packets that are guaranteed to be no larger than the IP 1568 minimum link MTU as discussed in Section 3.9.) 1570 The ICMP header is followed by the leading portion of the packet that 1571 generated the error, also known as the "packet-in-error". For 1572 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1573 much of invoking packet as possible without the ICMPv6 packet 1574 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1575 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1576 "Internet Header + 64 bits of Original Data Datagram", however 1577 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1578 ICMP datagram SHOULD contain as much of the original datagram as 1579 possible without the length of the ICMP datagram exceeding 576 1580 bytes". 1582 The link-layer error message format is shown in Figure 5 (where, "L2" 1583 and "L3" refer to link-layer and network-layer, respectively): 1585 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1586 ~ ~ 1587 | L2 IP Header of | 1588 | error message | 1589 ~ ~ 1590 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1591 | L2 ICMP Header | 1592 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1593 ~ ~ P 1594 | IP and other encapsulation | a 1595 | headers of original L3 packet | c 1596 ~ ~ k 1597 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1598 ~ ~ t 1599 | IP header of | 1600 | original L3 packet | i 1601 ~ ~ n 1602 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1603 ~ ~ e 1604 | Upper layer headers and | r 1605 | leading portion of body | r 1606 | of the original L3 packet | o 1607 ~ ~ r 1608 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1610 Figure 5: OMNI Interface Link-Layer Error Message Format 1612 The AERO node rules for processing these link-layer error messages 1613 are as follows: 1615 o When an AERO node receives a link-layer Parameter Problem message, 1616 it processes the message the same as described as for ordinary 1617 ICMP errors in the normative references [RFC0792][RFC4443]. 1619 o When an AERO node receives persistent link-layer Time Exceeded 1620 messages, the IP ID field may be wrapping before earlier fragments 1621 awaiting reassembly have been processed. In that case, the node 1622 should begin including integrity checks and/or institute rate 1623 limits for subsequent packets. 1625 o When an AERO node receives persistent link-layer Destination 1626 Unreachable messages in response to encapsulated packets that it 1627 sends to one of its asymmetric neighbor correspondents, the node 1628 should process the message as an indication that a path may be 1629 failing, and optionally initiate NUD over that path. If it 1630 receives Destination Unreachable messages over multiple paths, the 1631 node should allow future packets destined to the correspondent to 1632 flow through a default route and re-initiate route optimization. 1634 o When an AERO Client receives persistent link-layer Destination 1635 Unreachable messages in response to encapsulated packets that it 1636 sends to one of its symmetric neighbor Servers, the Client should 1637 mark the path as unusable and use another path. If it receives 1638 Destination Unreachable messages on many or all paths, the Client 1639 should associate with a new Server and release its association 1640 with the old Server as specified in Section 3.16.5. 1642 o When an AERO Server receives persistent link-layer Destination 1643 Unreachable messages in response to encapsulated packets that it 1644 sends to one of its symmetric neighbor Clients, the Server should 1645 mark the underlying path as unusable and use another underlying 1646 path. 1648 o When an AERO Server or Proxy receives link-layer Destination 1649 Unreachable messages in response to an encapsulated packet that it 1650 sends to one of its permanent neighbors, it treats the messages as 1651 an indication that the path to the neighbor may be failing. 1652 However, the dynamic routing protocol should soon reconverge and 1653 correct the temporary outage. 1655 When an AERO Bridge receives a packet for which the network-layer 1656 destination address is covered by an MSP, if there is no more- 1657 specific routing information for the destination the Bridge drops the 1658 packet and returns a network-layer Destination Unreachable message 1659 subject to rate limiting. The Bridge writes the network-layer source 1660 address of the original packet as the destination address and uses 1661 one of its non link-local addresses as the source address of the 1662 message. 1664 When an AERO node receives an encapsulated packet for which the 1665 reassembly buffer it too small, it drops the packet and returns a 1666 network-layer Packet Too Big (PTB) message. The node first writes 1667 the MRU value into the PTB message MTU field, writes the network- 1668 layer source address of the original packet as the destination 1669 address and writes one of its non link-local addresses as the source 1670 address. 1672 3.12. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1674 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1675 coordinated as discussed in the following Sections. 1677 3.12.1. AERO Service Model 1679 Each AERO Server on the OMNI link is configured to facilitate Client 1680 prefix delegation/registration requests. Each Server is provisioned 1681 with a database of MNP-to-Client ID mappings for all Clients enrolled 1682 in the AERO service, as well as any information necessary to 1683 authenticate each Client. The Client database is maintained by a 1684 central administrative authority for the OMNI link and securely 1685 distributed to all Servers, e.g., via the Lightweight Directory 1686 Access Protocol (LDAP) [RFC4511], via static configuration, etc. 1687 Clients receive the same service regardless of the Servers they 1688 select. 1690 AERO Clients and Servers use ND messages to maintain neighbor cache 1691 entries. AERO Servers configure their OMNI interfaces as advertising 1692 NBMA interfaces, and therefore send unicast RA messages with a short 1693 Router Lifetime value (e.g., ReachableTime seconds) in response to a 1694 Client's RS message. Thereafter, Clients send additional RS messages 1695 to keep Server state alive. 1697 AERO Clients and Servers include prefix delegation and/or 1698 registration parameters in RS/RA messages (see 1699 [I-D.templin-6man-omni-interface]). The ND messages are exchanged 1700 between Client and Server according to the prefix management schedule 1701 required by the service. If the Client knows its MNP in advance, it 1702 can employ prefix registration by including its LLA as the source 1703 address of an RS message and with an OMNI option with valid prefix 1704 registration information for the MNP. If the Server (and Proxy) 1705 accept the Client's MNP assertion, they inject the prefix into the 1706 routing system and establish the necessary neighbor cache state. 1708 The following sections specify the Client and Server behavior. 1710 3.12.2. AERO Client Behavior 1712 AERO Clients discover the addresses of Servers in a similar manner as 1713 described in [RFC5214]. Discovery methods include static 1714 configuration (e.g., from a flat-file map of Server addresses and 1715 locations), or through an automated means such as Domain Name System 1716 (DNS) name resolution [RFC1035]. Alternatively, the Client can 1717 discover Server addresses through a layer 2 data link login exchange, 1718 or through a unicast RA response to a multicast/anycast RS as 1719 described below. In the absence of other information, the Client can 1720 resolve the DNS Fully-Qualified Domain Name (FQDN) 1721 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 1722 text string and "[domainname]" is a DNS suffix for the OMNI link 1723 (e.g., "example.com"). 1725 To associate with a Server, the Client acts as a requesting router to 1726 request MNPs. The Client prepares an RS message with prefix 1727 management parameters and includes a Nonce and Timestamp option if 1728 the Client needs to correlate RA replies. If the Client already 1729 knows the Server's LLA, it includes the LLA as the network-layer 1730 destination address; otherwise, it includes (link-local) All-Routers 1731 multicast as the network-layer destination. If the Client already 1732 knows its own LLA, it uses the LLA as the network-layer source 1733 address; otherwise, it uses the unspecified IPv6 address (::/128) as 1734 the network-layer source address. 1736 The Client next includes an OMNI option in the RS message to register 1737 its link-layer information with the Server. The Client sets the OMNI 1738 option prefix registration information according to the MNP, and 1739 includes Interface Attributes corresponding to the underlying 1740 interface over which the Client will send the RS message. The Client 1741 MAY include additional Interface Attributes specific to other 1742 underlying interfaces. 1744 The Client then sends the RS message (either directly via Direct 1745 interfaces, via a VPN for VPNed interfaces, via a Proxy for ANET 1746 interfaces or via INET encapsulation for INET interfaces) and waits 1747 for an RA message reply (see Section 3.12.3). The Client retries up 1748 to MAX_RTR_SOLICITATIONS times until an RA is received. If the 1749 Client receives no RAs, or if it receives an RA with Router Lifetime 1750 set to 0, the Client SHOULD abandon this Server and try another 1751 Server. Otherwise, the Client processes the prefix information found 1752 in the RA message. 1754 Next, the Client creates a symmetric neighbor cache entry with the 1755 Server's LLA as the network-layer address and the Server's 1756 encapsulation and/or link-layer addresses as the link-layer address. 1757 The Client records the RA Router Lifetime field value in the neighbor 1758 cache entry as the time for which the Server has committed to 1759 maintaining the MNP in the routing system via this underlying 1760 interface, and caches the other RA configuration information 1761 including Cur Hop Limit, M and O flags, Reachable Time and Retrans 1762 Timer. The Client then autoconfigures LLAs for each of the delegated 1763 MNPs and assigns them to the OMNI interface. The Client also caches 1764 any MSPs included in Route Information Options (RIOs) [RFC4191] as 1765 MSPs to associate with the OMNI link, and assigns the MTU value in 1766 the MTU option to the underlying interface. 1768 The Client then registers additional underlying interfaces with the 1769 Server by sending RS messages via each additional interface. The RS 1770 messages include the same parameters as for the initial RS/RA 1771 exchange, but with destination address set to the Server's LLA. 1773 Following autoconfiguration, the Client sub-delegates the MNPs to its 1774 attached EUNs and/or the Client's own internal virtual interfaces as 1775 described in [I-D.templin-v6ops-pdhost] to support the Client's 1776 downstream attached "Internet of Things (IoT)". The Client 1777 subsequently sends additional RS messages over each underlying 1778 interface before the Router Lifetime received for that interface 1779 expires. 1781 After the Client registers its underlying interfaces, it may wish to 1782 change one or more registrations, e.g., if an interface changes 1783 address or becomes unavailable, if QoS preferences change, etc. To 1784 do so, the Client prepares an RS message to send over any available 1785 underlying interface. The RS includes an OMNI option with prefix 1786 registration information specific to its MNP, with Interface 1787 Attributes specific to the selected underlying interface, and with 1788 any additional Interface Attributes specific to other underlying 1789 interfaces. When the Client receives the Server's RA response, it 1790 has assurance that the Server has been updated with the new 1791 information. 1793 If the Client wishes to discontinue use of a Server it issues an RS 1794 message over any underlying interface with an OMNI option with a 1795 prefix release indication. When the Server processes the message, it 1796 releases the MNP, sets the symmetric neighbor cache entry state for 1797 the Client to DEPARTED and returns an RA reply with Router Lifetime 1798 set to 0. After a short delay (e.g., 2 seconds), the Server 1799 withdraws the MNP from the routing system. 1801 3.12.3. AERO Server Behavior 1803 AERO Servers act as IP routers and support a prefix delegation/ 1804 registration service for Clients. Servers arrange to add their LLAs 1805 to a static map of Server addresses for the link and/or the DNS 1806 resource records for the FQDN "linkupnetworks.[domainname]" before 1807 entering service. Server addresses should be geographically and/or 1808 topologically referenced, and made available for discovery by Clients 1809 on the OMNI link. 1811 When a Server receives a prospective Client's RS message on its OMNI 1812 interface, it SHOULD return an immediate RA reply with Router 1813 Lifetime set to 0 if it is currently too busy or otherwise unable to 1814 service the Client. Otherwise, the Server authenticates the RS 1815 message and processes the prefix delegation/registration parameters. 1816 The Server first determines the correct MNPs to provide to the Client 1817 by searching the Client database. When the Server returns the MNPs, 1818 it also creates a forwarding table entry for the ULA corredponding to 1819 each MNP so that the MNPs are propagated into the routing system 1820 (see: Section 3.2.3). For IPv6, the Server creates an IPv6 1821 forwarding table entry for each MNP. For IPv4, the Server creates an 1822 IPv6 forwarding table entry with the IPv4-compatibility ULA prefix 1823 corresponding to the IPv4 address. 1825 The Server next creates a symmetric neighbor cache entry for the 1826 Client using the base LLA as the network-layer address and with 1827 lifetime set to no more than the smallest prefix lifetime. Next, the 1828 Server updates the neighbor cache entry by recording the information 1829 in each Interface Attributes sub-option in the RS OMNI option. The 1830 Server also records the actual OAL/INET addresses in the neighbor 1831 cache entry. 1833 Next, the Server prepares an RA message using its LLA as the network- 1834 layer source address and the network-layer source address of the RS 1835 message as the network-layer destination address. The Server sets 1836 the Router Lifetime to the time for which it will maintain both this 1837 underlying interface individually and the symmetric neighbor cache 1838 entry as a whole. The Server also sets Cur Hop Limit, M and O flags, 1839 Reachable Time and Retrans Timer to values appropriate for the OMNI 1840 link. The Server includes the MNPs, any other prefix management 1841 parameters and an OMNI option with no Interface Attributes. The 1842 Server then includes one or more RIOs that encode the MSPs for the 1843 OMNI link, plus an MTU option (see Section 3.9). The Server finally 1844 forwards the message to the Client using OAL/INET, INET, or NULL 1845 encapsulation as necessary. 1847 After the initial RS/RA exchange, the Server maintains a 1848 ReachableTime timer for each of the Client's underlying interfaces 1849 individually (and for the Client's symmetric neighbor cache entry 1850 collectively) set to expire after ReachableTime seconds. If the 1851 Client (or Proxy) issues additional RS messages, the Server sends an 1852 RA response and resets ReachableTime. If the Server receives an ND 1853 message with a prefix release indication it sets the Client's 1854 symmetric neighbor cache entry to the DEPARTED state and withdraws 1855 the MNP from the routing system after a short delay (e.g., 2 1856 seconds). If ReachableTime expires before a new RS is received on an 1857 individual underlying interface, the Server marks the interface as 1858 DOWN. If ReachableTime expires before any new RS is received on any 1859 individual underlying interface, the Server sets the symmetric 1860 neighbor cache entry state to STALE and sets a 10 second timer. If 1861 the Server has not received a new RS or ND message with a prefix 1862 release indication before the 10 second timer expires, it deletes the 1863 neighbor cache entry and withdraws the MNP from the routing system. 1865 The Server processes any ND messages pertaining to the Client and 1866 returns an NA/RA reply in response to solicitations. The Server may 1867 also issue unsolicited RA messages, e.g., with reconfigure parameters 1868 to cause the Client to renegotiate its prefix delegation/ 1869 registrations, with Router Lifetime set to 0 if it can no longer 1870 service this Client, etc. Finally, If the symmetric neighbor cache 1871 entry is in the DEPARTED state, the Server deletes the entry after 1872 DepartTime expires. 1874 Note: Clients SHOULD notify former Servers of their departures, but 1875 Servers are responsible for expiring neighbor cache entries and 1876 withdrawing routes even if no departure notification is received 1877 (e.g., if the Client leaves the network unexpectedly). Servers 1878 SHOULD therefore set Router Lifetime to ReachableTime seconds in 1879 solicited RA messages to minimize persistent stale cache information 1880 in the absence of Client departure notifications. A short Router 1881 Lifetime also ensures that proactive Client/Server RS/RA messaging 1882 will keep any NAT state alive (see above). 1884 Note: All Servers on an OMNI link MUST advertise consistent values in 1885 the RA Cur Hop Limit, M and O flags, Reachable Time and Retrans Timer 1886 fields the same as for any link, since unpredictable behavior could 1887 result if different Servers on the same link advertised different 1888 values. 1890 3.12.3.1. Lightweight DHCPv6 Relay Agent (LDRA) 1892 When the OMNI option includes a DHCPv6 Unique Identifier (DUID) sub- 1893 option, the use of DHCPv6 Prefix Delegation is indicated. AERO 1894 Clients and Servers are always on the same link (i.e., the OMNI link) 1895 from the perspective of DHCPv6, however in some implementations the 1896 DHCPv6 server and ND function may be located in separate modules. In 1897 that case, the Server's OMNI interface module can act as a 1898 Lightweight DHCPv6 Relay Agent (LDRA)[RFC6221] to relay prefix 1899 delegation messages to and from the DHCPv6 server module. 1901 When the LDRA receives an authentic RS message, it constructs a 1902 corresponding IPv6/UDP/DHCPv6 message. It sets the IPv6 source 1903 address to the source address of the RS message, sets the IPv6 1904 destination address to 'All_DHCP_Relay_Agents_and_Servers' and sets 1905 the UDP fields to values that will be understood by the DHCPv6 1906 server. 1908 The LDRA then wraps the message in a DHCPv6 'Relay-Forward' message 1909 header and includes an 'Interface-Id' option that includes enough 1910 information to allow the LDRA to forward the resulting Reply message 1911 back to the Client (e.g., the Client's link-layer addresses, a 1912 security association identifier, etc.). The LDRA also wraps the OMNI 1913 option into the Interface-Id option, then forwards the message to the 1914 DHCPv6 server. 1916 When the DHCPv6 server prepares a Reply message, it wraps the message 1917 in a 'Relay-Reply' message and echoes the Interface-Id option. The 1918 DHCPv6 server then delivers the Relay-Reply message to the LDRA, 1919 which discards the Relay-Reply wrapper and IPv6/UDP headers, then 1920 uses the DHCPv6 message to construct an RA response to the Client. 1921 The Server uses the information in the Interface-Id option to prepare 1922 the RA message and to cache the link-layer addresses taken from the 1923 OMNI option echoed in the Interface-Id option. 1925 3.13. The AERO Proxy 1927 Clients may connect to protected-spectrum ANETs that employ physical 1928 and/or link-layer security services to facilitate communications to 1929 Servers in outside INETs. In that case, the ANET can employ an AERO 1930 Proxy. The Proxy is located at the ANET/INET border and listens for 1931 RS messages originating from or RA messages destined to ANET Clients. 1932 The Proxy acts on these control messages as follows: 1934 o when the Proxy receives an RS message from a new ANET Client, it 1935 first authenticates the message then examines the network-layer 1936 destination address. If the destination address is a Server's 1937 LLA, the Proxy proceeds to the next step. Otherwise, if the 1938 destination is (link-local) All-Routers multicast, the Proxy 1939 selects a "nearby" Server that is likely to be a good candidate to 1940 serve the Client and replaces the destination address with the 1941 Server's LLA. Next, the Proxy creates a proxy neighbor cache 1942 entry and caches the Client and Server link-layer addresses along 1943 with the OMNI option information and any other identifying 1944 information including Transaction IDs, Client Identifiers, Nonce 1945 values, etc. The Proxy finally encapsulates the (proxyed) RS 1946 message in an OAL header with source set to the Proxy's ULA and 1947 destination set to the Server's ULA then forwards the message into 1948 the OMNI link. 1950 o when the Server receives the RS, it authenticates the message then 1951 creates or updates a symmetric neighbor cache entry for the Client 1952 with the Proxy's ULA as the link-layer address. The Server then 1953 sends an RA message back to the Proxy via the spanning tree. 1955 o when the Proxy receives the RA, it authenticates the message and 1956 matches it with the proxy neighbor cache entry created by the RS. 1957 The Proxy then caches the prefix information as a mapping from the 1958 Client's MNPs to the Client's link-layer address, caches the 1959 Server's advertised Router Lifetime and sets the neighbor cache 1960 entry state to REACHABLE. The Proxy then optionally rewrites the 1961 Router Lifetime and forwards the (proxyed) message to the Client. 1962 The Proxy finally includes an MTU option (if necessary) with an 1963 MTU to use for the underlying ANET interface. 1965 After the initial RS/RA exchange, the Proxy forwards any Client data 1966 packets for which there is no matching asymmetric neighbor cache 1967 entry to a Bridge using OAL encapsulation with its own ULA as the 1968 source and the ULA corresponding to the Client as the destination. 1969 The Proxy instead forwards any Client data destined to an asymmetric 1970 neighbor cache target directly to the target according to the OAL/ 1971 link-layer information - the process of establishing asymmetric 1972 neighbor cache entries is specified in Section 3.14. 1974 While the Client is still attached to the ANET, the Proxy sends NS, 1975 RS and/or unsolicited NA messages to update the Server's symmetric 1976 neighbor cache entries on behalf of the Client and/or to convey QoS 1977 updates. This allows for higher-frequency Proxy-initiated RS/RA 1978 messaging over well-connected INET infrastructure supplemented by 1979 lower-frequency Client-initiated RS/RA messaging over constrained 1980 ANET data links. 1982 If the Server ceases to send solicited advertisements, the Proxy 1983 sends unsolicited RAs on the ANET interface with destination set to 1984 (link-local) All-Nodes multicast and with Router Lifetime set to zero 1985 to inform Clients that the Server has failed. Although the Proxy 1986 engages in ND exchanges on behalf of the Client, the Client can also 1987 send ND messages on its own behalf, e.g., if it is in a better 1988 position than the Proxy to convey QoS changes, etc. For this reason, 1989 the Proxy marks any Client-originated solicitation messages (e.g. by 1990 inserting a Nonce option) so that it can return the solicited 1991 advertisement to the Client instead of processing it locally. 1993 If the Client becomes unreachable, the Proxy sets the neighbor cache 1994 entry state to DEPARTED and retains the entry for DepartTime seconds. 1995 While the state is DEPARTED, the Proxy forwards any packets destined 1996 to the Client to a Bridge via OAL encapsulation with the Client's 1997 current Server as the destination. The Bridge in turn forwards the 1998 packets to the Client's current Server. When DepartTime expires, the 1999 Proxy deletes the neighbor cache entry and discards any further 2000 packets destined to this (now forgotten) Client. 2002 In some ANETs that employ a Proxy, the Client's MNP can be injected 2003 into the ANET routing system. In that case, the Client can send data 2004 messages without encapsulation so that the ANET routing system 2005 transports the unencapsulated packets to the Proxy. This can be very 2006 beneficial, e.g., if the Client connects to the ANET via low-end data 2007 links such as some aviation wireless links. 2009 If the first-hop ANET access router is on the same underlying link 2010 and recognizes the AERO/OMNI protocol, the Client can avoid 2011 encapsulation for both its control and data messages. When the 2012 Client connects to the link, it can send an unencapsulated RS message 2013 with source address set to its LLA and with destination address set 2014 to the LLA of the Client's selected Server or to (link-local) All- 2015 Routers multicast. The Client includes an OMNI option formatted as 2016 specified in [I-D.templin-6man-omni-interface]. 2018 The Client then sends the unencapsulated RS message, which will be 2019 intercepted by the AERO-Aware access router. The access router then 2020 encapsulates the RS message in an ANET header with its own address as 2021 the source address and the address of a Proxy as the destination 2022 address. The access router further remembers the address of the 2023 Proxy so that it can encapsulate future data packets from the Client 2024 via the same Proxy. If the access router needs to change to a new 2025 Proxy, it simply sends another RS message toward the Server via the 2026 new Proxy on behalf of the Client. 2028 In some cases, the access router and Proxy may be one and the same 2029 node. In that case, the node would be located on the same physical 2030 link as the Client, but its message exchanges with the Server would 2031 need to pass through a security gateway at the ANET/INET border. The 2032 method for deploying access routers and Proxys (i.e. as a single node 2033 or multiple nodes) is an ANET-local administrative consideration. 2035 3.13.1. Combined Proxy/Servers 2037 Clients may need to connect directly to Servers via INET, Direct and 2038 VPNed interfaces (i.e., non-ANET interfaces). If the Client's 2039 underlying interfaces all connect via the same INET partition, then 2040 it can connect to a single controlling Server via all interfaces. 2042 If some Client interfaces connect via different INET partitions, 2043 however, the Client still selects a set of controlling Servers and 2044 sends RS messages via their directly-connected Servers while using 2045 the LLA of the controlling Server as the destination. 2047 When a Server receives an RS with destination set to the LLA of a 2048 controlling Server, it acts as a Proxy to forward the message to the 2049 controlling Server while forwarding the corresponding RA reply to the 2050 Client. 2052 3.13.2. Detecting and Responding to Server Failures 2054 In environments where fast recovery from Server failure is required, 2055 Proxys SHOULD use proactive Neighbor Unreachability Detection (NUD) 2056 to track Server reachability in a similar fashion as for 2057 Bidirectional Forwarding Detection (BFD) [RFC5880]. Proxys can then 2058 quickly detect and react to failures so that cached information is 2059 re-established through alternate paths. The NUD control messaging is 2060 carried only over well-connected ground domain networks (i.e., and 2061 not low-end aeronautical radio links) and can therefore be tuned for 2062 rapid response. 2064 Proxys perform proactive NUD with Servers for which there are 2065 currently active ANET Clients by sending continuous NS messages in 2066 rapid succession, e.g., one message per second. The Proxy sends the 2067 NS message via the spanning tree with the Proxy's LLA as the source 2068 and the LLA of the Server as the destination. When the Proxy is also 2069 sending RS messages to the Server on behalf of ANET Clients, the 2070 resulting RA responses can be considered as equivalent hints of 2071 forward progress. This means that the Proxy need not also send a 2072 periodic NS if it has already sent an RS within the same period. If 2073 the Server fails (i.e., if the Proxy ceases to receive 2074 advertisements), the Proxy can quickly inform Clients by sending 2075 multicast RA messages on the ANET interface. 2077 The Proxy sends RA messages on the ANET interface with source address 2078 set to the Server's address, destination address set to (link-local) 2079 All-Nodes multicast, and Router Lifetime set to 0. The Proxy SHOULD 2080 send MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small 2081 delays [RFC4861]. Any Clients on the ANET that had been using the 2082 failed Server will receive the RA messages and associate with a new 2083 Server. 2085 3.13.3. Point-to-Multipoint Server Coordination 2087 In environments where Client messaging over ANETs is bandwidth- 2088 limited and/or expensive, Clients can enlist the services of the 2089 Proxy to coordinate with multiple Servers in a single RS/RA message 2090 exchange. The Client can send a single RS message to (link-local) 2091 All-Routers multicast that includes the ID's of multiple Servers in 2092 MS-Register sub-options of the OMNI option. 2094 When the Proxy receives the RS and processes the OMNI option, it 2095 sends a separate RS to each MS-Register Server ID. When the Proxy 2096 receives an RA, it can optionally return an immediate "singleton" RA 2097 to the Client or record the Server's ID for inclusion in a pending 2098 "aggregate" RA message. The Proxy can then return aggregate RA 2099 messages to the Client including multiple Server IDs in order to 2100 conserve bandwidth. Each RA includes a proper subset of the Server 2101 IDs from the original RS message, and the Proxy must ensure that the 2102 message contents of each RA are consistent with the information 2103 received from the (aggregated) Servers. 2105 Clients can thereafter employ efficient point-to-multipoint Server 2106 coordination under the assistance of the Proxy to reduce the number 2107 of messages sent over the ANET while enlisting the support of 2108 multiple Servers for fault tolerance. Clients can further include 2109 MS-Release suboptions in IPv6 ND messages to request the Proxy to 2110 release from former Servers via the procedures discussed in 2111 Section 3.16.5. 2113 The OMNI interface specification [I-D.templin-6man-omni-interface] 2114 provides further discussion of the Client/Proxy RS/RA messaging 2115 involved in point-to-multipoint coordination. 2117 3.14. AERO Route Optimization / Address Resolution 2119 While data packets are flowing between a source and target node, 2120 route optimization SHOULD be used. Route optimization is initiated 2121 by the first eligible Route Optimization Source (ROS) closest to the 2122 source as follows: 2124 o For Clients on VPNed and Direct interfaces, the Server is the ROS. 2126 o For Clients on ANET interfaces, the Proxy is the ROS. 2128 o For Clients on INET interfaces, the Client itself is the ROS. 2130 o For correspondent nodes on INET/EUN interfaces serviced by a 2131 Relay, the Relay is the ROS. 2133 The route optimization procedure is conducted between the ROS and the 2134 target Server/Relay acting as a Route Optimization Responder (ROR) in 2135 the same manner as for IPv6 ND Address Resolution and using the same 2136 NS/NA messaging. The target may either be a MNP Client serviced by a 2137 Server, or a non-MNP correspondent reachable via a Relay. 2139 The procedures are specified in the following sections. 2141 3.14.1. Route Optimization Initiation 2143 While data packets are flowing from the source node toward a target 2144 node, the ROS performs address resolution by sending an NS message 2145 for Address Resolution (NS(AR)) to receive a solicited NA message 2146 from the ROR. When the ROS sends an NS(AR), it includes: 2148 o the LLA of the ROS as the source address. 2150 o the data packet's destination as the Target Address. 2152 o the Solicited-Node multicast address [RFC4291] formed from the 2153 lower 24 bits of the data packet's destination as the destination 2154 address, e.g., for 2001:db8:1:2::10:2000 the NS destination 2155 address is ff02:0:0:0:0:1:ff10:2000. 2157 The NS(AR) message includes an OMNI option with no Interface 2158 Attributes, such that the target will not create a neighbor cache 2159 entry. The Prefix Length in the OMNI option is set to the Prefix 2160 Length associated with the ROS's LLA. 2162 The ROS then encapsulates the NS(AR) message in an OAL header with 2163 source set to its own ULA and destination set to the ULA 2164 corresponding to the packet's final destination, then sends the 2165 message into the spanning tree without decrementing the network-layer 2166 TTL/Hop Limit field. 2168 3.14.2. Relaying the NS 2170 When the Bridge receives the NS(AR) message from the ROS, it discards 2171 the INET header and determines that the ROR is the next hop by 2172 consulting its standard IPv6 forwarding table for the OAL header 2173 destination address. The Bridge then forwards the message toward the 2174 ROR via the spanning tree the same as for any IPv6 router. The 2175 final-hop Bridge in the spanning tree will deliver the message via a 2176 secured tunnel to the ROR. 2178 3.14.3. Processing the NS and Sending the NA 2180 When the ROR receives the NS(AR) message, it examines the Target 2181 Address to determine whether it has a neighbor cache entry and/or 2182 route that matches the target. If there is no match, the ROR drops 2183 the message. Otherwise, the ROR continues processing as follows: 2185 o if the target belongs to an MNP Client neighbor in the DEPARTED 2186 state the ROR changes the NS(AR) message OAL destination address 2187 to the ULA of the Client's new Server, forwards the message into 2188 the spanning tree and returns from processing. 2190 o If the target belongs to an MNP Client neighbor in the REACHABLE 2191 state, the ROR instead adds the AERO source address to the target 2192 Client's Report List with time set to ReportTime. 2194 o If the target belongs to a non-MNP route, the ROR continues 2195 processing without adding an entry to the Report List. 2197 The ROR then prepares a solicited NA message to send back to the ROS 2198 but does not create a neighbor cache entry. The ROR sets the NA 2199 source address to the LLA corresponding to the target, sets the 2200 Target Address to the target of the solicitation, and sets the 2201 destination address to the source of the solicitation. The ROR then 2202 includes an OMNI option with Prefix Length set to the length 2203 associated with the LLA. 2205 If the target is an MNP Client, the ROR next includes Interface 2206 Attributes in the OMNI option for each of the target Client's 2207 underlying interfaces with current information for each interface and 2208 with the S/T-ifIndex field in the OMNI header set to 0 to indicate 2209 that the message originated from the ROR and not the Client. 2211 For each Interface Attributes sub-option, the ROR sets the L2ADDR 2212 according to its own INET address for VPNed or Direct interfaces, to 2213 the INET address of the Proxy or to the Client's INET address for 2214 INET interfaces. The ROR then includes the lower 32 bits of its own 2215 ULA (or the ULA of the Proxy) as the LHS, encodes the ULA prefix 2216 length code in the SRT field and sets the FMT code accordingly as 2217 specified in Section 3.3. 2219 The ROR then sets the NA message R flag to 1 (as a router), S flag to 2220 1 (as a response to a solicitation), and O flag to 0 (as a proxy). 2221 The ROR finally encapsulates the NA message in an OAL header with 2222 source set to its own ULA and destination set to the source ULA of 2223 the NS(AR) message, then forwards the message into the spanning tree 2224 without decrementing the network-layer TTL/Hop Limit field. 2226 3.14.4. Relaying the NA 2228 When the Bridge receives the NA message from the ROR, it discards the 2229 INET header and determines that the ROS is the next hop by consulting 2230 its standard IPv6 forwarding table for the OAL header destination 2231 address. The Bridge then forwards the OAL-encapsulated NA message 2232 toward the ROS the same as for any IPv6 router. The final-hop Bridge 2233 in the spanning tree will deliver the message via a secured tunnel to 2234 the ROS. 2236 3.14.5. Processing the NA 2238 When the ROS receives the solicited NA message, it processes the 2239 message the same as for standard IPv6 Address Resolution [RFC4861]. 2240 In the process, it caches the source ULA then creates an asymmetric 2241 neighbor cache entry for the target and caches all information found 2242 in the OMNI option. The ROS finally sets the asymmetric neighbor 2243 cache entry lifetime to ReachableTime seconds. 2245 3.14.6. Route Optimization Maintenance 2247 Following route optimization, the ROS forwards future data packets 2248 destined to the target via the addresses found in the cached link- 2249 layer information. The route optimization is shared by all sources 2250 that send packets to the target via the ROS, i.e., and not just the 2251 source on behalf of which the route optimization was initiated. 2253 While new data packets destined to the target are flowing through the 2254 ROS, it sends additional NS(AR) messages to the ROR before 2255 ReachableTime expires to receive a fresh solicited NA message the 2256 same as described in the previous sections (route optimization 2257 refreshment strategies are an implementation matter, with a non- 2258 normative example given in Appendix A.1). The ROS uses the cached 2259 ULA of the ROR as the NS(AR) OAL destination address, and sends up to 2260 MAX_MULTICAST_SOLICIT NS(AR) messages separated by 1 second until an 2261 NA is received. If no NA is received, the ROS assumes that the 2262 current ROR has become unreachable and deletes the target neighbor 2263 cache entry. Subsequent data packets will trigger a new route 2264 optimization per Section 3.14.1 to discover a new ROR while initial 2265 data packets travel over a suboptimal route. 2267 If an NA is received, the ROS then updates the asymmetric neighbor 2268 cache entry to refresh ReachableTime, while (for MNP destinations) 2269 the ROR adds or updates the ROS address to the target's Report List 2270 and with time set to ReportTime. While no data packets are flowing, 2271 the ROS instead allows ReachableTime for the asymmetric neighbor 2272 cache entry to expire. When ReachableTime expires, the ROS deletes 2273 the asymmetric neighbor cache entry. Any future data packets flowing 2274 through the ROS will again trigger a new route optimization. 2276 The ROS may also receive unsolicited NA messages from the ROR at any 2277 time (see: Section 3.16). If there is an asymmetric neighbor cache 2278 entry for the target, the ROS updates the link-layer information but 2279 does not update ReachableTime since the receipt of an unsolicited NA 2280 does not confirm that any forward paths are working. If there is no 2281 asymmetric neighbor cache entry, the ROS simply discards the 2282 unsolicited NA. 2284 In this arrangement, the ROS holds an asymmetric neighbor cache entry 2285 for the target via the ROR, but the ROR does not hold an asymmetric 2286 neighbor cache entry for the ROS. The route optimization neighbor 2287 relationship is therefore asymmetric and unidirectional. If the 2288 target node also has packets to send back to the source node, then a 2289 separate route optimization procedure is performed in the reverse 2290 direction. But, there is no requirement that the forward and reverse 2291 paths be symmetric. 2293 3.15. Neighbor Unreachability Detection (NUD) 2295 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2296 [RFC4861] either reactively in response to persistent link-layer 2297 errors (see Section 3.11) or proactively to confirm reachability. 2298 The NUD algorithm is based on periodic control message exchanges. 2299 The algorithm may further be seeded by ND hints of forward progress, 2300 but care must be taken to avoid inferring reachability based on 2301 spoofed information. For example, authentic IPv6 ND message 2302 exchanges may be considered as acceptable hints of forward progress, 2303 while spurious data packets should not be. 2305 AERO Servers, Proxys and Relays can use (OAL-encapsulated) standard 2306 NS/NA NUD exchanges sent over the OMNI link spanning tree to securely 2307 test reachability without risk of DoS attacks from nodes pretending 2308 to be a neighbor; Proxys can further perform NUD to securely verify 2309 Server reachability on behalf of their proxyed Clients. However, a 2310 means for an ROS to test the unsecured forward directions of target 2311 route optimized paths is also necessary. 2313 When an ROR directs an ROS to a neighbor with one or more target 2314 link-layer addresses, the ROS can proactively test each such 2315 unsecured route optimized path by sending "loopback" NS(NUD) 2316 messages. While testing the paths, the ROS can optionally continue 2317 to send packets via the spanning tree, maintain a small queue of 2318 packets until target reachability is confirmed, or (optimistically) 2319 allow packets to flow via the route optimized paths. 2321 When the ROS sends a loopback NS(NUD) message, it uses its LLA as 2322 both the IPv6 source and destination address, and the MNP Subnet- 2323 Router anycast address as the Target Address. The ROS includes a 2324 Nonce and Timestamp option, then encapsulates the message in OAL/INET 2325 headers with its own ULA as the source and the ULA of the route 2326 optimization target as the destination. The ROS then forwards the 2327 message to the target (either directly to the L2ADDR of the target if 2328 the target is in the same OMNI link segment, or via a Bridge if the 2329 target is in a different OMNI link segment). 2331 When the route optimization target receives the NS(NUD) message, it 2332 notices that the IPv6 destination address is the same as the source 2333 address. It then reverses the OAL header source and destination 2334 addresses and returns the message to the ROS (either directly or via 2335 the spanning tree). The route optimization target does not decrement 2336 the NS(NUD) message IPv6 Hop-Limit in the process, since the message 2337 has not exited the OMNI link. 2339 When the ROS receives the NS(NUD) message, it can determine from the 2340 Nonce, Timestamp and Target Address that the message originated from 2341 itself and that it transited the forward path. The ROS need not 2342 prepare an NA response, since the destination of the response would 2343 be itself and testing the route optimization path again would be 2344 redundant. 2346 The ROS marks route optimization target paths that pass these NUD 2347 tests as "reachable", and those that do not as "unreachable". These 2348 markings inform the OMNI interface forwarding algorithm specified in 2349 Section 3.10. 2351 Note that to avoid a DoS vector nodes MUST NOT return loopback 2352 NS(NUD) messages received from an unsecured link-layer source via the 2353 spanning tree. 2355 3.16. Mobility Management and Quality of Service (QoS) 2357 AERO is a Distributed Mobility Management (DMM) service. Each Server 2358 is responsible for only a subset of the Clients on the OMNI link, as 2359 opposed to a Centralized Mobility Management (CMM) service where 2360 there is a single network mobility collective entity for all Clients. 2361 Clients coordinate with their associated Servers via RS/RA exchanges 2362 to maintain the DMM profile, and the AERO routing system tracks all 2363 current Client/Server peering relationships. 2365 Servers provide default routing and mobility/multilink services for 2366 their dependent Clients. Clients are responsible for maintaining 2367 neighbor relationships with their Servers through periodic RS/RA 2368 exchanges, which also serves to confirm neighbor reachability. When 2369 a Client's underlying interface address and/or QoS information 2370 changes, the Client is responsible for updating the Server with this 2371 new information. Note that when there is a Proxy in the path, the 2372 Proxy can also perform some RS/RA exchanges on the Client's behalf. 2374 Mobility management messaging is based on the transmission and 2375 reception of unsolicited Neighbor Advertisment (uNA) messages. Each 2376 uNA message sets the IPv6 destination address to (link-local) All- 2377 Nodes multicast to convey a general update of Interface Attributes to 2378 (possibly) multiple recipients, or to a specific unicast LLA to 2379 announce a departure event to a specific recipient. Implementations 2380 must therefore examine the destination address to determine the 2381 nature of the mobility event (i.e., update vs departure). 2383 Mobility management considerations are specified in the following 2384 sections. 2386 3.16.1. Mobility Update Messaging 2388 Servers accommodate Client mobility/multilink and/or QoS change 2389 events by sending unsolicited NA (uNA) messages to each ROS in the 2390 target Client's Report List. When a Server sends a uNA message, it 2391 sets the IPv6 source address to the Client's LLA, sets the 2392 destination address to (link-local) All-Nodes multicast and sets the 2393 Target Address to the Client's Subnet-Router anycast address. The 2394 Server also includes an OMNI option with Prefix Length set to the 2395 length associated with the Client's LLA, with Interface Attributes 2396 for the target Client's underlying interfaces and with the OMNI 2397 header S/T-ifIndex set to 0. The Server then sets the NA R flag to 2398 1, the S flag to 0 and the O flag to 0, then encapsulates the message 2399 in an OAL header with source set to its own ULA and destination set 2400 to the ULA of the ROS and sends the message into the spanning tree. 2402 As discussed in Section 7.2.6 of [RFC4861], the transmission and 2403 reception of uNA messages is unreliable but provides a useful 2404 optimization. In well-connected Internetworks with robust data links 2405 uNA messages will be delivered with high probability, but in any case 2406 the Server can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT uNAs 2407 to each ROS to increase the likelihood that at least one will be 2408 received. 2410 When the ROS receives a uNA message prepared as above, it ignores the 2411 message if there is no existing neighbor cache entry for the Client. 2412 Otherwise, it uses the included OMNI option information to update the 2413 neighbor cache entry, but does not reset ReachableTime since the 2414 receipt of an unsolicited NA message from the target Server does not 2415 provide confirmation that any forward paths to the target Client are 2416 working. 2418 If uNA messages are lost, the ROS may be left with stale address and/ 2419 or QoS information for the Client for up to ReachableTime seconds. 2420 During this time, the ROS can continue sending packets according to 2421 its stale neighbor cache information. When ReachableTime is close to 2422 expiring, the ROS will re-initiate route optimization and receive 2423 fresh link-layer address information. 2425 In addition to sending uNA messages to the current set of ROSs for 2426 the Client, the Server also sends uNAs to the former link-layer 2427 address for any underlying interface for which the link-layer address 2428 has changed. The uNA messages update Proxys that cannot easily 2429 detect (e.g., without active probing) when a formerly-active Client 2430 has departed. Note however that the uNA is formatted the same as for 2431 announcing the Client's departure to an old Server per 2432 Section 3.16.5, i.e., and not per the above instructions. 2434 3.16.2. Announcing Link-Layer Address and/or QoS Preference Changes 2436 When a Client needs to change its underlying interface addresses and/ 2437 or QoS preferences (e.g., due to a mobility event), either the Client 2438 or its Proxys send RS messages to the Server via the spanning tree 2439 with an OMNI option that includes Interface attributes with the new 2440 link quality and address information. 2442 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 2443 sending actual data packets in case one or more RAs are lost. If all 2444 RAs are lost, the Client SHOULD re-associate with a new Server. 2446 When the Server receives the Client's changes, it sends uNA messages 2447 to all nodes in the Report List the same as described in the previous 2448 section. 2450 3.16.3. Bringing New Links Into Service 2452 When a Client needs to bring new underlying interfaces into service 2453 (e.g., when it activates a new data link), it sends an RS message to 2454 the Server via the underlying interface with an OMNI option that 2455 includes Interface Attributes with appropriate link quality values 2456 and with link-layer address information for the new link. 2458 3.16.4. Removing Existing Links from Service 2460 When a Client needs to remove existing underlying interfaces from 2461 service (e.g., when it de-activates an existing data link), it sends 2462 an RS or uNA message to its Server with an OMNI option with 2463 appropriate link quality values. 2465 If the Client needs to send RS/uNA messages over an underlying 2466 interface other than the one being removed from service, it MUST 2467 include Interface Attributes with appropriate link quality values for 2468 any underlying interfaces being removed from service. 2470 3.16.5. Moving to a New Server 2472 When a Client associates with a new Server, it performs the Client 2473 procedures specified in Section 3.12.2. The Client also includes MS- 2474 Release identifiers in the RS message OMNI option per 2476 [I-D.templin-6man-omni-interface] if it wants the new Server to 2477 notify any old Servers from which the Client is departing. 2479 When the new Server receives the Client's RS message, it returns an 2480 RA as specified in Section 3.12.3 and sends up to 2481 MAX_NEIGHBOR_ADVERTIISEMENT uNA messages to any old Servers listed in 2482 OMNI option MS-Release identifiers. When the new Server sends a uNA 2483 message, it sets the IPv6 source address to the Client's LLA, sets 2484 the destination address to the old Server's LLA, and sets the Target 2485 Address to the Client's Subnet-Router anycast address. The new 2486 Server also includes an OMNI option with Prefix Length set to the 2487 length associated with the Client's LLA, with Interface Attributes 2488 for its own underlying interface, and with the OMNI header S/ 2489 T-ifIndex set to 0. The new Server then sets the NA R flag to 1, the 2490 S flag to 0 and the O flag to 0, then encapsulates the message in an 2491 OAL header with source set to its own ULA and destination set to the 2492 ULA of the old Server and sends the message into the spanning tree. 2494 When an old Server receives the uNA, it changes the Client's neighbor 2495 cache entry state to DEPARTED, sets the link-layer address of the 2496 Client to the new Server's ULA, and resets DepartTime. After a short 2497 delay (e.g., 2 seconds) the old Server withdraws the Client's MNP 2498 from the routing system. After DepartTime expires, the old Server 2499 deletes the Client's neighbor cache entry. 2501 The old Server also sends a "proxy" copy of the uNA message to all 2502 ROSs in the Client's Report List. In particular, it changes the IPv6 2503 and OAL destination addresses to those of the ROS. When the ROS 2504 receives the NA, it caches the address of the new Server in the 2505 existing asymmetric neighbor cache entry and marks the entry as STALE 2506 for a period of 10 seconds after which the cache entry is deleted. 2507 While in the STALE state, subsequent data packets flow according to 2508 any existing cached link-layer information and trigger a new 2509 NS(AR)/NA exchange via the new Server. 2511 Clients SHOULD NOT move rapidly between Servers in order to avoid 2512 causing excessive oscillations in the AERO routing system. Examples 2513 of when a Client might wish to change to a different Server include a 2514 Server that has gone unreachable, topological movements of 2515 significant distance, movement to a new geographic region, movement 2516 to a new OMNI link segment, etc. 2518 When a Client moves to a new Server, some of the fragments of a 2519 multiple fragment packet may have already arrived at the old Server 2520 while others are en route to the new Server, however no special 2521 attention in the reassembly algorithm is necessary when re-routed 2522 fragments are simply treated as loss. 2524 3.17. Multicast 2526 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 2527 [RFC3810] proxy service for its EUNs and/or hosted applications 2528 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 2529 underlying interfaces for which group membership is required. The 2530 IGMP/MLD messages may be further forwarded by a first-hop ANET access 2531 router acting as an IGMP/MLD-snooping switch [RFC4541], then 2532 ultimately delivered to an AERO Proxy/Server acting as a Protocol 2533 Independent Multicast - Sparse-Mode (PIM-SM, or simply "PIM") 2534 Designated Router (DR) [RFC7761]. AERO Relays also act as PIM 2535 routers (i.e., the same as AERO Proxys/Servers) on behalf of nodes on 2536 INET/EUN networks. The behaviors identified in the following 2537 sections correspond to Source-Specific Multicast (SSM) and Any-Source 2538 Multicast (ASM) operational modes. 2540 3.17.1. Source-Specific Multicast (SSM) 2542 When an ROS (i.e., an AERO Proxy/Server/Relay) "X" acting as PIM 2543 router receives a Join/Prune message from a node on its downstream 2544 interfaces containing one or more ((S)ource, (G)roup) pairs, it 2545 updates its Multicast Routing Information Base (MRIB) accordingly. 2546 For each S belonging to a prefix reachable via X's non-OMNI 2547 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 2548 on those interfaces per [RFC7761]. 2550 For each S belonging to a prefix reachable via X's OMNI interface, X 2551 originates a separate copy of the Join/Prune for each (S,G) in the 2552 message using its own LLA as the source address and ALL-PIM-ROUTERS 2553 as the destination address. X then encapsulates each message in an 2554 OAL header with source address set to the ULA of X and destination 2555 address set to S then forwards the message into the spanning tree, 2556 which delivers it to AERO Server/Relay "Y" that services S. At the 2557 same time, if the message was a Join, X sends a route-optimization NS 2558 message toward each S the same as discussed in Section 3.14. The 2559 resulting NAs will return the LLA for the prefix that matches S as 2560 the network-layer source address and with an OMNI option with the ULA 2561 corresponding to any underlying interfaces that are currently 2562 servicing S. 2564 When Y processes the Join/Prune message, if S located behind any 2565 INET, Direct, or VPNed interfaces Y acts as a PIM router and updates 2566 its MRIB to list X as the next hop in the reverse path. If S is 2567 located behind any Proxys "Z"*, Y also forwards the message to each 2568 Z* over the spanning tree while continuing to use the LLA of X as the 2569 source address. Each Z* then updates its MRIB accordingly and 2570 maintains the LLA of X as the next hop in the reverse path. Since 2571 the Bridges do not examine network layer control messages, this means 2572 that the (reverse) multicast tree path is simply from each Z* (and/or 2573 Y) to X with no other multicast-aware routers in the path. If any Z* 2574 (and/or Y) is located on the same OMNI link segment as X, the 2575 multicast data traffic sent to X directly using OAL/INET 2576 encapsulation instead of via a Bridge. 2578 Following the initial Join/Prune and NS/NA messaging, X maintains an 2579 asymmetric neighbor cache entry for each S the same as if X was 2580 sending unicast data traffic to S. In particular, X performs 2581 additional NS/NA exchanges to keep the neighbor cache entry alive for 2582 up to t_periodic seconds [RFC7761]. If no new Joins are received 2583 within t_periodic seconds, X allows the neighbor cache entry to 2584 expire. Finally, if X receives any additional Join/Prune messages 2585 for (S,G) it forwards the messages to each Y and Z* in the neighbor 2586 cache entry over the spanning tree. 2588 At some later time, Client C that holds an MNP for source S may 2589 depart from a first Proxy Z1 and/or connect via a new Proxy Z2. In 2590 that case, Y sends an unsolicited NA message to X the same as 2591 specified for unicast mobility in Section 3.16. When X receives the 2592 unsolicited NA message, it updates its asymmetric neighbor cache 2593 entry for the LLA for source S and sends new Join messages to any new 2594 Proxys Z2. There is no requirement to send any Prune messages to old 2595 Proxys Z1 since source S will no longer source any multicast data 2596 traffic via Z1. Instead, the multicast state for (S,G) in Proxy Z1 2597 will soon time out since no new Joins will arrive. 2599 After some later time, C may move to a new Server Y2 and depart from 2600 old Sever Y1. In that case, Y1 sends Join messages for any of C's 2601 active (S,G) groups to Y2 while including its own LLA as the source 2602 address. This causes Y2 to include Y1 in the multicast forwarding 2603 tree during the interim time that Y1's symmetric neighbor cache entry 2604 for C is in the DEPARTED state. At the same time, Y1 sends an 2605 unsolicited NA message to X with an OMNI option with S/T-ifIndex in 2606 the header set to 0 and a release indication to cause X to release 2607 its asymmetric neighbor cache entry. X then sends a new Join message 2608 to S via the spanning tree and re-initiates route optimization the 2609 same as if it were receiving a fresh Join message from a node on a 2610 downstream link. 2612 3.17.2. Any-Source Multicast (ASM) 2614 When an ROS X acting as a PIM router receives a Join/Prune from a 2615 node on its downstream interfaces containing one or more (*,G) pairs, 2616 it updates its Multicast Routing Information Base (MRIB) accordingly. 2617 X then forwards a copy of the message to the Rendezvous Point (RP) R 2618 for each G over the spanning tree. X uses its own LLA as the source 2619 address and ALL-PIM-ROUTERS as the destination address, then 2620 encapsulates each message in an OAL header with source address set to 2621 the ULA of X and destination address set to R, then sends the message 2622 into the spanning tree. At the same time, if the message was a Join 2623 X initiates NS/NA route optimization the same as for the SSM case 2624 discussed in Section 3.17.1. 2626 For each source S that sends multicast traffic to group G via R, the 2627 Proxy/Server Z* for the Client that aggregates S encapsulates the 2628 packets in PIM Register messages and forwards them to R via the 2629 spanning tree, which may then elect to send a PIM Join to Z*. This 2630 will result in an (S,G) tree rooted at Z* with R as the next hop so 2631 that R will begin to receive two copies of the packet; one native 2632 copy from the (S, G) tree and a second copy from the pre-existing (*, 2633 G) tree that still uses PIM Register encapsulation. R can then issue 2634 a PIM Register-stop message to suppress the Register-encapsulated 2635 stream. At some later time, if C moves to a new Proxy/Server Z*, it 2636 resumes sending packets via PIM Register encapsulation via the new 2637 Z*. 2639 At the same time, as multicast listeners discover individual S's for 2640 a given G, they can initiate an (S,G) Join for each S under the same 2641 procedures discussed in Section 3.17.1. Once the (S,G) tree is 2642 established, the listeners can send (S, G) Prune messages to R so 2643 that multicast packets for group G sourced by S will only be 2644 delivered via the (S, G) tree and not from the (*, G) tree rooted at 2645 R. All mobility considerations discussed for SSM apply. 2647 3.17.3. Bi-Directional PIM (BIDIR-PIM) 2649 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 2650 approach to ASM that treats the Rendezvous Point (RP) as a Designated 2651 Forwarder (DF). Further considerations for BIDIR-PIM are out of 2652 scope. 2654 3.18. Operation over Multiple OMNI Links 2656 An AERO Client can connect to multiple OMNI links the same as for any 2657 data link service. In that case, the Client maintains a distinct 2658 OMNI interface for each link, e.g., 'omni0' for the first link, 2659 'omni1' for the second, 'omni2' for the third, etc. Each OMNI link 2660 would include its own distinct set of Bridges, Servers and Proxys, 2661 thereby providing redundancy in case of failures. 2663 Each OMNI link could utilize the same or different ANET connections. 2664 The links can be distinguished at the link-layer via the SRT prefix 2665 in a similar fashion as for Virtual Local Area Network (VLAN) tagging 2666 (e.g., IEEE 802.1Q) and/or through assignment of distinct sets of 2667 MSPs on each link. This gives rise to the opportunity for supporting 2668 multiple redundant networked paths, with each VLAN distinguished by a 2669 different SRT "color" (see: Section 3.2.5). 2671 The Client's IP layer can select the outgoing OMNI interface 2672 appropriate for a given traffic profile while (in the reverse 2673 direction) correspondent nodes must have some way of steering their 2674 packets destined to a target via the correct OMNI link. 2676 In a first alternative, if each OMNI link services different MSPs, 2677 then the Client can receive a distinct MNP from each of the links. 2678 IP routing will therefore assure that the correct Red/Green/Blue/etc. 2679 network is used for both outbound and inbound traffic. This can be 2680 accomplished using existing technologies and approaches, and without 2681 requiring any special supporting code in correspondent nodes or 2682 Bridges. 2684 In a second alternative, if each OMNI link services the same MSP(s) 2685 then each link could assign a distinct "OMNI link Anycast" address 2686 that is configured by all Bridges on the link. Correspondent nodes 2687 can then perform Segment Routing to select the correct SRT, which 2688 will then direct the packet over multiple hops to the target. 2690 3.19. DNS Considerations 2692 AERO Client MNs and INET correspondent nodes consult the Domain Name 2693 System (DNS) the same as for any Internetworking node. When 2694 correspondent nodes and Client MNs use different IP protocol versions 2695 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 2696 A records for IPv4 address mappings to MNs which must then be 2697 populated in Relay NAT64 mapping caches. In that way, an IPv4 2698 correspondent node can send packets to the IPv4 address mapping of 2699 the target MN, and the Relay will translate the IPv4 header and 2700 destination address into an IPv6 header and IPv6 destination address 2701 of the MN. 2703 When an AERO Client registers with an AERO Server, the Server can 2704 return the address(es) of DNS servers in RDNSS options [RFC6106]. 2705 The DNS server provides the IP addresses of other MNs and 2706 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 2708 3.20. Transition Considerations 2710 OAL encapsulation ensures that dissimilar INET partitions can be 2711 joined into a single unified OMNI link, even though the partitions 2712 themselves may have differing protocol versions and/or incompatible 2713 addressing plans. However, a commonality can be achieved by 2714 incrementally distributing globally routable (i.e., native) IP 2715 prefixes to eventually reach all nodes (both mobile and fixed) in all 2716 OMNI link segments. This can be accomplished by incrementally 2717 deploying AERO Relays on each INET partition, with each Relay 2718 distributing its MNPs and/or discovering non-MNP prefixes on its INET 2719 links. 2721 This gives rise to the opportunity to eventually distribute native IP 2722 addresses to all nodes, and to present a unified OMNI link view even 2723 if the INET partitions remain in their current protocol and 2724 addressing plans. In that way, the OMNI link can serve the dual 2725 purpose of providing a mobility/multilink service and a transition 2726 service. Or, if an INET partition is transitioned to a native IP 2727 protocol version and addressing scheme that is compatible with the 2728 OMNI link MNP-based addressing scheme, the partition and OMNI link 2729 can be joined by Relays. 2731 Relays that connect INETs/EUNs with dissimilar IP protocol versions 2732 may need to employ a network address and protocol translation 2733 function such as NAT64[RFC6146]. 2735 3.21. Detecting and Reacting to Server and Bridge Failures 2737 In environments where rapid failure recovery is required, Servers and 2738 Bridges SHOULD use Bidirectional Forwarding Detection (BFD) 2739 [RFC5880]. Nodes that use BFD can quickly detect and react to 2740 failures so that cached information is re-established through 2741 alternate nodes. BFD control messaging is carried only over well- 2742 connected ground domain networks (i.e., and not low-end radio links) 2743 and can therefore be tuned for rapid response. 2745 Servers and Bridges maintain BFD sessions in parallel with their BGP 2746 peerings. If a Server or Bridge fails, BGP peers will quickly re- 2747 establish routes through alternate paths the same as for common BGP 2748 deployments. Similarly, Proxys maintain BFD sessions with their 2749 associated Bridges even though they do not establish BGP peerings 2750 with them. 2752 Proxys SHOULD use proactive NUD for Servers for which there are 2753 currently active ANET Clients in a manner that parallels BFD, i.e., 2754 by sending unicast NS messages in rapid succession to receive 2755 solicited NA messages. When the Proxy is also sending RS messages on 2756 behalf of ANET Clients, the RS/RA messaging can be considered as 2757 equivalent hints of forward progress. This means that the Proxy need 2758 not also send a periodic NS if it has already sent an RS within the 2759 same period. If a Server fails, the Proxy will cease to receive 2760 advertisements and can quickly inform Clients of the outage by 2761 sending multicast RA messages on the ANET interface. 2763 The Proxy sends multicast RA messages with source address set to the 2764 Server's address, destination address set to (link-local) All-Nodes 2765 multicast, and Router Lifetime set to 0. The Proxy SHOULD send 2766 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2767 [RFC4861]. Any Clients on the ANET interface that have been using 2768 the (now defunct) Server will receive the RA messages and associate 2769 with a new Server. 2771 3.22. AERO Clients on the Open Internet 2773 AERO Clients that connect to the open Internet via INET interfaces 2774 can establish a VPN or direct link to securely connect to a Server in 2775 a "tethered" arrangement with all of the Client's traffic transiting 2776 the Server. Alternatively, the Client can associate with an INET 2777 Server using UDP/IP encapsulation and asymmetric securing services as 2778 discussed in the following sections. 2780 When a Client's OMNI interface enables an INET underlying interface, 2781 it first determines whether the interface is likely to be behind a 2782 NAT. For IPv4, the Client assumes it is on the open Internet if the 2783 INET address is not a special-use IPv4 address per [RFC3330]. 2784 Similarly for IPv6, the Client assumes it is on the open Internet if 2785 the INET address is not a link-local [RFC4291] or unique-local 2786 [RFC4193] IPv6 address. 2788 The Client then prepares a UDP/IP-encapsulated RS message with IPv6 2789 source address set to its LLA, with IPv6 destination set to (link- 2790 local) All-Routers multicast and with an OMNI option with underlying 2791 interface parameters. If the Client believes that it is on the open 2792 Internet, it SHOULD include Interface Attributes with the L2ADDR used 2793 for INET encapsulation (otherwise, it MAY omit L2ADDR). If the 2794 underlying address is IPv4, the Client includes the Port Number and 2795 IPv4 address written in obfuscated form [RFC4380] as discussed in 2796 Section 3.3. If the underlying interface address is IPv6, the Client 2797 instead includes the Port Number and IPv6 address in obfuscated form. 2798 The Client finally includes an Authentication option per [RFC4380] to 2799 provide message authentication, sets the UDP/IP source to its INET 2800 address and UDP port, sets the UDP/IP destination to the Server's 2801 INET address and the AERO service port number (8060), then sends the 2802 message to the Server. 2804 When the Server receives the RS, it authenticates the message and 2805 registers the Client's MNP and INET interface information according 2806 to the OMNI option parameters. If the RS message includes an L2ADDR 2807 in the OMNI option, the Server compares the encapsulation IP address 2808 and UDP port number with the (unobfuscated) values. If the values 2809 are the same, the Server caches the Client's information as "INET" 2810 addresses meaning that the Client is likely to accept direct messages 2811 without requiring NAT traversal exchanges. If the values are 2812 different (or, if the OMNI option did not include an L2ADDR) the 2813 Server instead caches the Client's information as "NAT" addresses 2814 meaning that NAT traversal exchanges may be necessary. 2816 The Server then returns an RA message with IPv6 source and 2817 destination set corresponding to the addresses in the RS, and with an 2818 Authentication option per [RFC4380]. For IPv4, the Server also 2819 includes an Origin option per [RFC4380] with the mapped and 2820 obfuscated Port Number and IPv4 address observed in the encapsulation 2821 headers. For IPv6, the Server instead includes an IPv6 Origin option 2822 per Figure 6 with the mapped and obfuscated observed Port Number and 2823 IPv6 address (note that the value 0x02 in the second octet 2824 differentiates from other [RFC4380] option types). 2826 +--------+--------+-----------------+ 2827 | 0x00 | 0x02 | Origin port # | 2828 +--------+--------+-----------------+ 2829 ~ Origin IPv6 address ~ 2830 +-----------------------------------+ 2832 Figure 6: IPv6 Origin Option 2834 When the Client receives the RA message, it compares the mapped Port 2835 Number and IP address from the Origin option with its own address. 2836 If the addresses are the same, the Client assumes the open Internet / 2837 Cone NAT principle; if the addresses are different, the Client 2838 instead assumes that further qualification procedures are necessary 2839 to detect the type of NAT and proceeds according to standard 2840 [RFC4380] procedures. 2842 After the Client has registered its INET interfaces in such RS/RA 2843 exchanges it sends periodic RS messages to receive fresh RA messages 2844 before the Router Lifetime received on each INET interface expires. 2845 The Client also maintains default routes via its Servers, i.e., the 2846 same as described in earlier sections. 2848 When the Client sends messages to target IP addresses, it also 2849 invokes route optimization per Section 3.14 using IPv6 ND address 2850 resolution messaging. The Client sends the NS(AR) message to the 2851 Server wrapped in a UDP/IP header with an Authentication option with 2852 the NS source address set to the Client's LLA and destination address 2853 set to the target solicited node multicast address. The Server 2854 authenticates the message and sends a corresponding NS(AR) message 2855 over the spanning tree the same as if it were the ROS, but with the 2856 OAL source address set to the Server's ULA and destination set to the 2857 ULA of the target. When the ROR receives the NS(AR), it adds the 2858 Server's ULA and Client's LLA to the target's Report List, and 2859 returns an NA with OMNI option information for the target. The 2860 Server then returns a UDP/IP encapsulated NA message with an 2861 Authentication option to the Client. 2863 Following route optimization for targets in the same OMNI link 2864 segment, if the target's L2ADDR is on the open INET, the Client 2865 forwards data packets directly to the target INET address. If the 2866 target is behind a NAT, the Client first establishes NAT state for 2867 the L2ADDR using the "bubble" mechanisms specified in 2868 [RFC6081][RFC4380]. The Client continues to send data packets via 2869 its Server until NAT state is populated, then begins forwarding 2870 packets via the direct path through the NAT to the target. For 2871 targets in different OMNI link segments, the Client inserts an SRH 2872 and forwards data packets to the Bridge that returned the NA message. 2874 The ROR may return uNAs via the Server if the target moves, and the 2875 Server will send corresponding Authentication-protected uNAs to the 2876 Client. The Client can also send "loopback" NS(NUD) messages to test 2877 forward path reachability even though there is no security 2878 association between the Client and the target. 2880 The Client sends UDP/IP encapsulated IPv6 packets no larger than 1280 2881 bytes in one piece. In order to accommodate larger IPv6 packets (up 2882 to the OMNI interface MTU), the Client inserts an OAL header with 2883 source set to its own ULA and destination set to the ULA of the 2884 target and uses IPv6 fragmentation according to Section 3.9. The 2885 Client then encapsulates each fragment in a UDP/IP header and sends 2886 the fragments to the next hop. 2888 3.22.1. Use of SEND and CGA 2890 In some environments, use of the [RFC4380] Authentication option 2891 alone may be sufficient for assuring IPv6 ND message authentication 2892 between Clients and Servers. When additional protection is 2893 necessary, nodes should employ SEcure Neighbor Discovery (SEND) 2894 [RFC3971] with Cryptographically-Generated Addresses (CGA) [RFC3972]. 2896 When SEND/CGA are used, the Client prepares RS messages with its 2897 link-local CGA as the IPv6 source and (link-local) All-Routers 2898 multicast as the IPv6 Destination, includes any SEND options and 2899 wraps the message in an OAL header. The Client sets the OAL source 2900 address to its own ULA and sets the OAL destination address to (site- 2901 local) All-Routers multicast. The Client then wraps the RS message 2902 in UDP/IP headers according to [RFC4380] and sends the message to the 2903 Server. 2905 When the Server receives the message, it first verifies the 2906 Authentication option (if present) then uses the OAL source address 2907 to determine the MNP of the Client. The Server then processes the 2908 SEND options to authenticate the RS message and prepares an RA 2909 message response. The Server prepares the RA with its own link-local 2910 CGA as the IPv6 source and the CGA of the Client as the IPv6 2911 destination, includes any SEND options and wraps the message in an 2912 OAL header. The Server sets the OAL source address to its own ULA 2913 and sets the OAL destination address to the Client's ULA. The Server 2914 then wraps the RA message in UDP/IP headers according to [RFC4380] 2915 and sends the message to the Client. Thereafter, the Client/Server 2916 send additional RS/RA messages to maintain their association and any 2917 NAT state. 2919 The Client and Server also may exchange NS/NA messages using their 2920 own CGA as the source and with OAL encapsulation as above. When a 2921 Client sends an NS(AR), it sets the IPv6 source to its CGA and sets 2922 the IPv6 destination to the Solicited-Node Multicast address of the 2923 target. The Client then wraps the message in an OAL header with its 2924 own ULA as the source and the ULA of the target as the destination 2925 and sends it to the Server. The Server authenticates the message, 2926 then changes the IPv6 source address to the Client's LLA, removes the 2927 SEND options, and sends a corresponding NS(AR) into the spanning 2928 tree. When the Server receives the corresponding OAL-encapsulated 2929 NA, it changes the IPv6 destination address to the Client's CGA, 2930 inserts SEND options, then wraps the message in UDP/IP headers and 2931 sends it to the Client. 2933 When a Client sends a uNA, it sets the IPv6 source address to its own 2934 CGA and sets the IPv6 destination address to (link-local) All-Nodes 2935 multicast, includes SEND options, wraps the message in OAL and UDP/IP 2936 headers and sends the message to the Server. The Server 2937 authenticates the message, then changes the IPv6 address to the 2938 Client's LLA, removes the SEND options and forwards the message the 2939 same as discussed in Section 3.16.1. In the reverse direction, when 2940 the Server forwards a uNA to the Client, it changes the IPv6 address 2941 to its own CGA and inserts SEND options then forwards the message to 2942 the Client. 2944 When a Client sends an NS(NUD), it sets both the IPv6 source and 2945 destination address to its own LLA, wraps the message in an OAL 2946 header and UDP/IP headers, then sends the message directly to the 2947 peer which will loop the message back. In this case alone, the 2948 Client does not use the Server as a trust broker for forwarding the 2949 ND message. 2951 3.23. Time-Varying MNPs 2953 In some use cases, it is desirable, beneficial and efficient for the 2954 Client to receive a constant MNP that travels with the Client 2955 wherever it moves. For example, this would allow air traffic 2956 controllers to easily track aircraft, etc. In other cases, however 2957 (e.g., intelligent transportation systems), the MN may be willing to 2958 sacrifice a modicum of efficiency in order to have time-varying MNPs 2959 that can be changed every so often to defeat adversarial tracking. 2961 The DHCPv6 service offers a way for Clients that desire time-varying 2962 MNPs to obtain short-lived prefixes (e.g., on the order of a small 2963 number of minutes). In that case, the identity of the Client would 2964 not be bound to the MNP but rather the Client's identity would be 2965 bound to the DHCPv6 Device Unique Identifier (DUID) and used as the 2966 seed for Prefix Delegation. The Client would then be obligated to 2967 renumber its internal networks whenever its MNP (and therefore also 2968 its LLA) changes. This should not present a challenge for Clients 2969 with automated network renumbering services, however presents limits 2970 for the durations of ongoing sessions that would prefer to use a 2971 constant address. 2973 4. Implementation Status 2975 An early AERO implementation based on OpenVPN (https://openvpn.net/) 2976 was announced on the v6ops mailing list on January 10, 2018 and an 2977 initial public release of the AERO proof-of-concept source code was 2978 announced on the intarea mailing list on August 21, 2015. 2980 AERO Release-3.0.1 was tagged on September 11, 2020, and is 2981 undergoing internal testing. Additional releases expected Q42020, 2982 with first public release expected before year-end. 2984 5. IANA Considerations 2986 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 2987 AERO in the "enterprise-numbers" registry. 2989 The IANA has assigned the UDP port number "8060" for an earlier 2990 experimental version of AERO [RFC6706]. This document obsoletes 2991 [RFC6706] and claims the UDP port number "8060" for all future use. 2993 The IANA is instructed to assign a new type value TBD in the Segment 2994 Routing Header TLV registry [RFC8754]. 2996 No further IANA actions are required. 2998 6. Security Considerations 3000 AERO Bridges configure secured tunnels with AERO Servers, Realys and 3001 Proxys within their local OMNI link segments. Applicable secured 3002 tunnel alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 3003 [RFC6347], WireGuard [WG], etc. The AERO Bridges of all OMNI link 3004 segments in turn configure secured tunnels for their neighboring AERO 3005 Bridges in a spanning tree topology. Therefore, control messages 3006 exchanged between any pair of OMNI link neighbors on the spanning 3007 tree are already secured. 3009 AERO Servers, Relays and Proxys targeted by a route optimization may 3010 also receive data packets directly from arbitrary nodes in INET 3011 partitions instead of via the spanning tree. For INET partitions 3012 that apply effective ingress filtering to defeat source address 3013 spoofing, the simple data origin authentication procedures in 3014 Section 3.8 can be applied. 3016 For INET partitions that require strong security in the data plane, 3017 two options for securing communications include 1) disable route 3018 optimization so that all traffic is conveyed over secured tunnels, or 3019 2) enable on-demand secure tunnel creation between INET partition 3020 neighbors. Option 1) would result in longer routes than necessary 3021 and traffic concentration on critical infrastructure elements. 3022 Option 2) could be coordinated by establishing a secured tunnel on- 3023 demand instead of performing an NS/NA exchange in the route 3024 optimization procedures. Procedures for establishing on-demand 3025 secured tunnels are out of scope. 3027 AERO Clients that connect to secured ANETs need not apply security to 3028 their ND messages, since the messages will be intercepted by a 3029 perimeter Proxy that applies security on its INET-facing interface as 3030 part of the spanning tree (see above). AERO Clients connected to the 3031 open INET can use symmetric network and/or transport layer security 3032 services such as VPNs or can by some other means establish a direct 3033 link. When a VPN or direct link may be impractical, however, an 3034 asymmetric security service such as SEcure Neighbor Discovery (SEND) 3035 [RFC3971] with Cryptographically Generated Addresses (CGAs) [RFC3972] 3036 and/or the Authentication option [RFC4380] can be applied. 3038 Application endpoints SHOULD use application-layer security services 3039 such as TLS/SSL, DTLS or SSH [RFC4251] to assure the same level of 3040 protection as for critical secured Internet services. AERO Clients 3041 that require host-based VPN services SHOULD use symmetric network 3042 and/or transport layer security services such as IPsec, TLS/SSL, 3043 DTLS, etc. AERO Proxys and Servers can also provide a network-based 3044 VPN service on behalf of the Client, e.g., if the Client is located 3045 within a secured enclave and cannot establish a VPN on its own 3046 behalf. 3048 AERO Servers and Bridges present targets for traffic amplification 3049 Denial of Service (DoS) attacks. This concern is no different than 3050 for widely-deployed VPN security gateways in the Internet, where 3051 attackers could send spoofed packets to the gateways at high data 3052 rates. This can be mitigated by connecting Servers and Bridges over 3053 dedicated links with no connections to the Internet and/or when 3054 connections to the Internet are only permitted through well-managed 3055 firewalls. Traffic amplification DoS attacks can also target an AERO 3056 Client's low data rate links. This is a concern not only for Clients 3057 located on the open Internet but also for Clients in secured 3058 enclaves. AERO Servers and Proxys can institute rate limits that 3059 protect Clients from receiving packet floods that could DoS low data 3060 rate links. 3062 AERO Relays must implement ingress filtering to avoid a spoofing 3063 attack in which spurious messages with ULA addresses are injected 3064 into an OMNI link from an outside attacker. AERO Clients MUST ensure 3065 that their connectivity is not used by unauthorized nodes on their 3066 EUNs to gain access to a protected network, i.e., AERO Clients that 3067 act as routers MUST NOT provide routing services for unauthorized 3068 nodes. (This concern is no different than for ordinary hosts that 3069 receive an IP address delegation but then "share" the address with 3070 other nodes via some form of Internet connection sharing such as 3071 tethering.) 3073 The MAP list MUST be well-managed and secured from unauthorized 3074 tampering, even though the list contains only public information. 3075 The MAP list can be conveyed to the Client in a similar fashion as in 3076 [RFC5214] (e.g., through layer 2 data link login messaging, secure 3077 upload of a static file, DNS lookups, etc.). 3079 Although public domain and commercial SEND implementations exist, 3080 concerns regarding the strength of the cryptographic hash algorithm 3081 have been documented [RFC6273] [RFC4982]. 3083 SRH authentication facilities are specified in [RFC8754]. 3085 Security considerations for accepting link-layer ICMP messages and 3086 reflected packets are discussed throughout the document. 3088 Security considerations for IPv6 fragmentation and reassembly are 3089 discussed in [I-D.templin-6man-omni-interface]. 3091 7. Acknowledgements 3093 Discussions in the IETF, aviation standards communities and private 3094 exchanges helped shape some of the concepts in this work. 3095 Individuals who contributed insights include Mikael Abrahamsson, Mark 3096 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 3097 Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian Farrel, Nick Green, 3098 Sri Gundavelli, Brian Haberman, Bernhard Haindl, Joel Halpern, Tom 3099 Herbert, Sascha Hlusiak, Lee Howard, Zdenek Jaron, Andre Kostur, 3100 Hubert Kuenig, Ted Lemon, Andy Malis, Satoru Matsushima, Tomek 3101 Mrugalski, Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal 3102 Skorepa, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, Lloyd 3103 Wood and James Woodyatt. Members of the IESG also provided valuable 3104 input during their review process that greatly improved the document. 3105 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 3106 for their shepherding guidance during the publication of the AERO 3107 first edition. 3109 This work has further been encouraged and supported by Boeing 3110 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 3111 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 3112 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 3113 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 3114 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 3115 Gene MacLean III, Rob Muszkiewicz, Sean O'Sullivan, Vijay 3116 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 3117 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 3118 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 3119 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 3120 Benson, Katie Tran and Eric Yeh are especially acknowledged for 3121 implementing the AERO functions as extensions to the public domain 3122 OpenVPN distribution. 3124 Earlier works on NBMA tunneling approaches are found in 3125 [RFC2529][RFC5214][RFC5569]. 3127 Many of the constructs presented in this second edition of AERO are 3128 based on the author's earlier works, including: 3130 o The Internet Routing Overlay Network (IRON) 3131 [RFC6179][I-D.templin-ironbis] 3133 o Virtual Enterprise Traversal (VET) 3134 [RFC5558][I-D.templin-intarea-vet] 3136 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 3137 [RFC5320][I-D.templin-intarea-seal] 3139 o AERO, First Edition [RFC6706] 3141 Note that these works cite numerous earlier efforts that are not also 3142 cited here due to space limitations. The authors of those earlier 3143 works are acknowledged for their insights. 3145 This work is aligned with the NASA Safe Autonomous Systems Operation 3146 (SASO) program under NASA contract number NNA16BD84C. 3148 This work is aligned with the FAA as per the SE2025 contract number 3149 DTFAWA-15-D-00030. 3151 This work is aligned with the Boeing Commercial Airplanes (BCA) 3152 Internet of Things (IoT) and autonomy programs. 3154 This work is aligned with the Boeing Information Technology (BIT) 3155 MobileNet program. 3157 8. References 3159 8.1. Normative References 3161 [I-D.templin-6man-omni-interface] 3162 Templin, F. and T. Whyman, "Transmission of IP Packets 3163 over Overlay Multilink Network (OMNI) Interfaces", draft- 3164 templin-6man-omni-interface-38 (work in progress), 3165 September 2020. 3167 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3168 DOI 10.17487/RFC0791, September 1981, 3169 . 3171 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 3172 RFC 792, DOI 10.17487/RFC0792, September 1981, 3173 . 3175 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3176 Requirement Levels", BCP 14, RFC 2119, 3177 DOI 10.17487/RFC2119, March 1997, 3178 . 3180 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3181 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3182 December 1998, . 3184 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3185 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3186 DOI 10.17487/RFC3971, March 2005, 3187 . 3189 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3190 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3191 . 3193 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3194 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3195 November 2005, . 3197 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3198 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3199 . 3201 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 3202 Network Address Translations (NATs)", RFC 4380, 3203 DOI 10.17487/RFC4380, February 2006, 3204 . 3206 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3207 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3208 DOI 10.17487/RFC4861, September 2007, 3209 . 3211 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3212 Address Autoconfiguration", RFC 4862, 3213 DOI 10.17487/RFC4862, September 2007, 3214 . 3216 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 3217 DOI 10.17487/RFC6081, January 2011, 3218 . 3220 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3221 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3222 May 2017, . 3224 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3225 (IPv6) Specification", STD 86, RFC 8200, 3226 DOI 10.17487/RFC8200, July 2017, 3227 . 3229 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3230 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3231 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3232 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3233 . 3235 8.2. Informative References 3237 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3238 2016. 3240 [I-D.bonica-6man-comp-rtg-hdr] 3241 Bonica, R., Kamite, Y., Niwa, T., Alston, A., and L. 3242 Jalil, "The IPv6 Compact Routing Header (CRH)", draft- 3243 bonica-6man-comp-rtg-hdr-22 (work in progress), May 2020. 3245 [I-D.bonica-6man-crh-helper-opt] 3246 Li, X., Bao, C., Ruan, E., and R. Bonica, "Compressed 3247 Routing Header (CRH) Helper Option", draft-bonica-6man- 3248 crh-helper-opt-01 (work in progress), May 2020. 3250 [I-D.ietf-intarea-frag-fragile] 3251 Bonica, R., Baker, F., Huston, G., Hinden, R., Troan, O., 3252 and F. Gont, "IP Fragmentation Considered Fragile", draft- 3253 ietf-intarea-frag-fragile-17 (work in progress), September 3254 2019. 3256 [I-D.ietf-intarea-tunnels] 3257 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3258 Architecture", draft-ietf-intarea-tunnels-10 (work in 3259 progress), September 2019. 3261 [I-D.ietf-rtgwg-atn-bgp] 3262 Templin, F., Saccone, G., Dawra, G., Lindem, A., and V. 3263 Moreno, "A Simple BGP-based Mobile Routing System for the 3264 Aeronautical Telecommunications Network", draft-ietf- 3265 rtgwg-atn-bgp-06 (work in progress), June 2020. 3267 [I-D.templin-6man-dhcpv6-ndopt] 3268 Templin, F., "A Unified Stateful/Stateless Configuration 3269 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-10 3270 (work in progress), June 2020. 3272 [I-D.templin-intarea-seal] 3273 Templin, F., "The Subnetwork Encapsulation and Adaptation 3274 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 3275 progress), January 2014. 3277 [I-D.templin-intarea-vet] 3278 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 3279 templin-intarea-vet-40 (work in progress), May 2013. 3281 [I-D.templin-ironbis] 3282 Templin, F., "The Interior Routing Overlay Network 3283 (IRON)", draft-templin-ironbis-16 (work in progress), 3284 March 2014. 3286 [I-D.templin-v6ops-pdhost] 3287 Templin, F., "IPv6 Prefix Delegation and Multi-Addressing 3288 Models", draft-templin-v6ops-pdhost-26 (work in progress), 3289 June 2020. 3291 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3293 [RFC1035] Mockapetris, P., "Domain names - implementation and 3294 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3295 November 1987, . 3297 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3298 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3299 . 3301 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3302 DOI 10.17487/RFC2003, October 1996, 3303 . 3305 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3306 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3307 . 3309 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 3310 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 3311 . 3313 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3314 Domains without Explicit Tunnels", RFC 2529, 3315 DOI 10.17487/RFC2529, March 1999, 3316 . 3318 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3319 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3320 . 3322 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3323 of Explicit Congestion Notification (ECN) to IP", 3324 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3325 . 3327 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 3328 DOI 10.17487/RFC3330, September 2002, 3329 . 3331 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3332 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3333 DOI 10.17487/RFC3810, June 2004, 3334 . 3336 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3337 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3338 January 2006, . 3340 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3341 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3342 DOI 10.17487/RFC4271, January 2006, 3343 . 3345 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3346 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3347 2006, . 3349 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3350 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3351 December 2005, . 3353 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3354 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3355 2006, . 3357 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3358 Control Message Protocol (ICMPv6) for the Internet 3359 Protocol Version 6 (IPv6) Specification", STD 89, 3360 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3361 . 3363 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3364 Protocol (LDAP): The Protocol", RFC 4511, 3365 DOI 10.17487/RFC4511, June 2006, 3366 . 3368 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3369 "Considerations for Internet Group Management Protocol 3370 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3371 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3372 . 3374 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3375 "Internet Group Management Protocol (IGMP) / Multicast 3376 Listener Discovery (MLD)-Based Multicast Forwarding 3377 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3378 August 2006, . 3380 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3381 Algorithms in Cryptographically Generated Addresses 3382 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3383 . 3385 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3386 "Bidirectional Protocol Independent Multicast (BIDIR- 3387 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3388 . 3390 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3391 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3392 DOI 10.17487/RFC5214, March 2008, 3393 . 3395 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3396 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3397 February 2010, . 3399 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3400 Route Optimization Requirements for Operational Use in 3401 Aeronautics and Space Exploration Mobile Networks", 3402 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3403 . 3405 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3406 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3407 . 3409 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3410 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3411 January 2010, . 3413 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 3414 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 3415 . 3417 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3418 "IPv6 Router Advertisement Options for DNS Configuration", 3419 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3420 . 3422 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3423 NAT64: Network Address and Protocol Translation from IPv6 3424 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3425 April 2011, . 3427 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3428 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3429 . 3431 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3432 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3433 DOI 10.17487/RFC6221, May 2011, 3434 . 3436 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3437 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3438 DOI 10.17487/RFC6273, June 2011, 3439 . 3441 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3442 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3443 January 2012, . 3445 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3446 for Equal Cost Multipath Routing and Link Aggregation in 3447 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3448 . 3450 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3451 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3452 . 3454 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 3455 UDP Checksums for Tunneled Packets", RFC 6935, 3456 DOI 10.17487/RFC6935, April 2013, 3457 . 3459 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 3460 for the Use of IPv6 UDP Datagrams with Zero Checksums", 3461 RFC 6936, DOI 10.17487/RFC6936, April 2013, 3462 . 3464 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3465 Korhonen, "Requirements for Distributed Mobility 3466 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3467 . 3469 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3470 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3471 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3472 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3473 2016, . 3475 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3476 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3477 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3478 July 2018, . 3480 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3481 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3482 . 3484 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 3485 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3486 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 3487 . 3489 [WG] Wireguard, "Wireguard, https://www.wireguard.com", August 3490 2020. 3492 Appendix A. Non-Normative Considerations 3494 AERO can be applied to a multitude of Internetworking scenarios, with 3495 each having its own adaptations. The following considerations are 3496 provided as non-normative guidance: 3498 A.1. Implementation Strategies for Route Optimization 3500 Route optimization as discussed in Section 3.14 results in the route 3501 optimization source (ROS) creating an asymmetric neighbor cache entry 3502 for the target neighbor. The neighbor cache entry is maintained for 3503 at most ReachableTime seconds and then deleted unless updated. In 3504 order to refresh the neighbor cache entry lifetime before the 3505 ReachableTime timer expires, the specification requires 3506 implementations to issue a new NS/NA exchange to reset ReachableTime 3507 while data packets are still flowing. However, the decision of when 3508 to initiate a new NS/NA exchange and to perpetuate the process is 3509 left as an implementation detail. 3511 One possible strategy may be to monitor the neighbor cache entry 3512 watching for data packets for (ReachableTime - 5) seconds. If any 3513 data packets have been sent to the neighbor within this timeframe, 3514 then send an NS to receive a new NA. If no data packets have been 3515 sent, wait for 5 additional seconds and send an immediate NS if any 3516 data packets are sent within this "expiration pending" 5 second 3517 window. If no additional data packets are sent within the 5 second 3518 window, delete the neighbor cache entry. 3520 The monitoring of the neighbor data packet traffic therefore becomes 3521 an asymmetric ongoing process during the neighbor cache entry 3522 lifetime. If the neighbor cache entry expires, future data packets 3523 will trigger a new NS/NA exchange while the packets themselves are 3524 delivered over a longer path until route optimization state is re- 3525 established. 3527 A.2. Implicit Mobility Management 3529 OMNI interface neighbors MAY provide a configuration option that 3530 allows them to perform implicit mobility management in which no ND 3531 messaging is used. In that case, the Client only transmits packets 3532 over a single interface at a time, and the neighbor always observes 3533 packets arriving from the Client from the same link-layer source 3534 address. 3536 If the Client's underlying interface address changes (either due to a 3537 readdressing of the original interface or switching to a new 3538 interface) the neighbor immediately updates the neighbor cache entry 3539 for the Client and begins accepting and sending packets according to 3540 the Client's new address. This implicit mobility method applies to 3541 use cases such as cellphones with both WiFi and Cellular interfaces 3542 where only one of the interfaces is active at a given time, and the 3543 Client automatically switches over to the backup interface if the 3544 primary interface fails. 3546 A.3. Direct Underlying Interfaces 3548 When a Client's OMNI interface is configured over a Direct interface, 3549 the neighbor at the other end of the Direct link can receive packets 3550 without any encapsulation. In that case, the Client sends packets 3551 over the Direct link according to QoS preferences. If the Direct 3552 interface has the highest QoS preference, then the Client's IP 3553 packets are transmitted directly to the peer without going through an 3554 ANET/INET. If other interfaces have higher QoS preferences, then the 3555 Client's IP packets are transmitted via a different interface, which 3556 may result in the inclusion of Proxys, Servers and Bridges in the 3557 communications path. Direct interfaces must be tested periodically 3558 for reachability, e.g., via NUD. 3560 A.4. AERO Critical Infrastructure Considerations 3562 AERO Bridges can be either Commercial off-the Shelf (COTS) standard 3563 IP routers or virtual machines in the cloud. Bridges must be 3564 provisioned, supported and managed by the INET administrative 3565 authority, and connected to the Bridges of other INETs via inter- 3566 domain peerings. Cost for purchasing, configuring and managing 3567 Bridges is nominal even for very large OMNI links. 3569 AERO Servers can be standard dedicated server platforms, but most 3570 often will be deployed as virtual machines in the cloud. The only 3571 requirements for Servers are that they can run the AERO user-level 3572 code and have at least one network interface connection to the INET. 3573 As with Bridges, Servers must be provisioned, supported and managed 3574 by the INET administrative authority. Cost for purchasing, 3575 configuring and managing Servers is nominal especially for virtual 3576 Servers hosted in the cloud. 3578 AERO Proxys are most often standard dedicated server platforms with 3579 one network interface connected to the ANET and a second interface 3580 connected to an INET. As with Servers, the only requirements are 3581 that they can run the AERO user-level code and have at least one 3582 interface connection to the INET. Proxys must be provisioned, 3583 supported and managed by the ANET administrative authority. Cost for 3584 purchasing, configuring and managing Proxys is nominal, and borne by 3585 the ANET administrative authority. 3587 AERO Relays can be any dedicated server or COTS router platform 3588 connected to INETs and/or EUNs. The Relay connects to the OMNI link 3589 and engages in eBGP peering with one or more Bridges as a stub AS. 3590 The Relay then injects its MNPs and/or non-MNP prefixes into the BGP 3591 routing system, and provisions the prefixes to its downstream- 3592 attached networks. The Relay can perform ROS/ROR services the same 3593 as for any Server, and can route between the MNP and non-MNP address 3594 spaces. 3596 A.5. AERO Server Failure Implications 3598 AERO Servers may appear as a single point of failure in the 3599 architecture, but such is not the case since all Servers on the link 3600 provide identical services and loss of a Server does not imply 3601 immediate and/or comprehensive communication failures. Although 3602 Clients typically associate with a single Server at a time, Server 3603 failure is quickly detected and conveyed by Bidirectional Forward 3604 Detection (BFD) and/or proactive NUD allowing Clients to migrate to 3605 new Servers. 3607 If a Server fails, ongoing packet forwarding to Clients will continue 3608 by virtue of the asymmetric neighbor cache entries that have already 3609 been established in route optimization sources (ROSs). If a Client 3610 also experiences mobility events at roughly the same time the Server 3611 fails, unsolicited NA messages may be lost but proxy neighbor cache 3612 entries in the DEPARTED state will ensure that packet forwarding to 3613 the Client's new locations will continue for up to DepartTime 3614 seconds. 3616 If a Client is left without a Server for an extended timeframe (e.g., 3617 greater than ReachableTime seconds) then existing asymmetric neighbor 3618 cache entries will eventually expire and both ongoing and new 3619 communications will fail. The original source will continue to 3620 retransmit until the Client has established a new Server 3621 relationship, after which time continuous communications will resume. 3623 Therefore, providing many Servers on the link with high availability 3624 profiles provides resilience against loss of individual Servers and 3625 assurance that Clients can establish new Server relationships quickly 3626 in event of a Server failure. 3628 A.6. AERO Client / Server Architecture 3630 The AERO architectural model is client / server in the control plane, 3631 with route optimization in the data plane. The same as for common 3632 Internet services, the AERO Client discovers the addresses of AERO 3633 Servers and selects one Server to connect to. The AERO service is 3634 analogous to common Internet services such as google.com, yahoo.com, 3635 cnn.com, etc. However, there is only one AERO service for the link 3636 and all Servers provide identical services. 3638 Common Internet services provide differing strategies for advertising 3639 server addresses to clients. The strategy is conveyed through the 3640 DNS resource records returned in response to name resolution queries. 3641 As of January 2020 Internet-based 'nslookup' services were used to 3642 determine the following: 3644 o When a client resolves the domainname "google.com", the DNS always 3645 returns one A record (i.e., an IPv4 address) and one AAAA record 3646 (i.e., an IPv6 address). The client receives the same addresses 3647 each time it resolves the domainname via the same DNS resolver, 3648 but may receive different addresses when it resolves the 3649 domainname via different DNS resolvers. But, in each case, 3650 exactly one A and one AAAA record are returned. 3652 o When a client resolves the domainname "ietf.org", the DNS always 3653 returns one A record and one AAAA record with the same addresses 3654 regardless of which DNS resolver is used. 3656 o When a client resolves the domainname "yahoo.com", the DNS always 3657 returns a list of 4 A records and 4 AAAA records. Each time the 3658 client resolves the domainname via the same DNS resolver, the same 3659 list of addresses are returned but in randomized order (i.e., 3660 consistent with a DNS round-robin strategy). But, interestingly, 3661 the same addresses are returned (albeit in randomized order) when 3662 the domainname is resolved via different DNS resolvers. 3664 o When a client resolves the domainname "amazon.com", the DNS always 3665 returns a list of 3 A records and no AAAA records. As with 3666 "yahoo.com", the same three A records are returned from any 3667 worldwide Internet connection point in randomized order. 3669 The above example strategies show differing approaches to Internet 3670 resilience and service distribution offered by major Internet 3671 services. The Google approach exposes only a single IPv4 and a 3672 single IPv6 address to clients. Clients can then select whichever IP 3673 protocol version offers the best response, but will always use the 3674 same IP address according to the current Internet connection point. 3675 This means that the IP address offered by the network must lead to a 3676 highly-available server and/or service distribution point. In other 3677 words, resilience is predicated on high availability within the 3678 network and with no client-initiated failovers expected (i.e., it is 3679 all-or-nothing from the client's perspective). However, Google does 3680 provide for worldwide distributed service distribution by virtue of 3681 the fact that each Internet connection point responds with a 3682 different IPv6 and IPv4 address. The IETF approach is like google 3683 (all-or-nothing from the client's perspective), but provides only a 3684 single IPv4 or IPv6 address on a worldwide basis. This means that 3685 the addresses must be made highly-available at the network level with 3686 no client failover possibility, and if there is any worldwide service 3687 distribution it would need to be conducted by a network element that 3688 is reached via the IP address acting as a service distribution point. 3690 In contrast to the Google and IETF philosophies, Yahoo and Amazon 3691 both provide clients with a (short) list of IP addresses with Yahoo 3692 providing both IP protocol versions and Amazon as IPv4-only. The 3693 order of the list is randomized with each name service query 3694 response, with the effect of round-robin load balancing for service 3695 distribution. With a short list of addresses, there is still 3696 expectation that the network will implement high availability for 3697 each address but in case any single address fails the client can 3698 switch over to using a different address. The balance then becomes 3699 one of function in the network vs function in the end system. 3701 The same implications observed for common highly-available services 3702 in the Internet apply also to the AERO client/server architecture. 3703 When an AERO Client connects to one or more ANETs, it discovers one 3704 or more AERO Server addresses through the mechanisms discussed in 3705 earlier sections. Each Server address presumably leads to a fault- 3706 tolerant clustering arrangement such as supported by Linux-HA, 3707 Extended Virtual Synchrony or Paxos. Such an arrangement has 3708 precedence in common Internet service deployments in lightweight 3709 virtual machines without requiring expensive hardware deployment. 3710 Similarly, common Internet service deployments set service IP 3711 addresses on service distribution points that may relay requests to 3712 many different servers. 3714 For AERO, the expectation is that a combination of the Google/IETF 3715 and Yahoo/Amazon philosophies would be employed. The AERO Client 3716 connects to different ANET access points and can receive 1-2 Server 3717 LLAs at each point. It then selects one AERO Server address, and 3718 engages in RS/RA exchanges with the same Server from all ANET 3719 connections. The Client remains with this Server unless or until the 3720 Server fails, in which case it can switch over to an alternate 3721 Server. The Client can likewise switch over to a different Server at 3722 any time if there is some reason for it to do so. So, the AERO 3723 expectation is for a balance of function in the network and end 3724 system, with fault tolerance and resilience at both levels. 3726 Appendix B. Change Log 3728 << RFC Editor - remove prior to publication >> 3730 Changes from draft-templin-intarea-6706bis-61 to draft-templin- 3731 intrea-6706bis-62: 3733 o New sub-section on OMNI Neighbor Interface Attributes 3735 Changes from draft-templin-intarea-6706bis-59 to draft-templin- 3736 intrea-6706bis-60: 3738 o Removed all references to S/TLLAO - all Interface Attributes are 3739 now maintained completely in the OMNI option. 3741 Changes from draft-templin-intarea-6706bis-58 to draft-templin- 3742 intrea-6706bis-59: 3744 o The term "Relay"used in older draft versions is now "Bridge". 3745 "Relay" now refers to what was formally called: "Gateway". 3747 o Fine-grained cleanup of Forwarding Algorithm; IPv6 ND message 3748 addressing; OMNI Prefix Lengths, etc. 3750 Changes from draft-templin-intarea-6706bis-54 to draft-templin- 3751 intrea-6706bis-55: 3753 o Updates on Segment Routing and S/TLLAO contents. 3755 o Various editorials and addressing cleanups. 3757 Changes from draft-templin-intarea-6706bis-52 to draft-templin- 3758 intrea-6706bis-53: 3760 o Normative reference to the OMNI spec, and remove portions that are 3761 already specified in OMNI. 3763 o Renamed "AERO interface/link" to "OMIN interface/link" throughout 3764 the document. 3766 o Truncated obsolete back section matter. 3768 Author's Address 3770 Fred L. Templin (editor) 3771 Boeing Research & Technology 3772 P.O. Box 3707 3773 Seattle, WA 98124 3774 USA 3776 Email: fltemplin@acm.org