idnits 2.17.1 draft-templin-intarea-6706bis-76.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 (December 11, 2020) is 1222 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) == Missing Reference: 'ULA' is mentioned on line 959, but not defined == Missing Reference: 'ULA0' is mentioned on line 965, but not defined == Missing Reference: 'ULA1' is mentioned on line 979, but not defined == Missing Reference: 'ULA2' is mentioned on line 969, but not defined == Missing Reference: 'ULA3' is mentioned on line 971, but not defined == Unused Reference: 'RFC3971' is defined on line 3266, but no explicit reference was found in the text == Unused Reference: 'RFC3972' is defined on line 3271, but no explicit reference was found in the text == Unused Reference: 'BGP' is defined on line 3319, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-comp-rtg-hdr' is defined on line 3322, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-crh-helper-opt' is defined on line 3328, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-frag-fragile' is defined on line 3333, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-dhcpv6-ndopt' is defined on line 3350, but no explicit reference was found in the text == Unused Reference: 'RFC4982' is defined on line 3472, but no explicit reference was found in the text == Unused Reference: 'RFC6273' is defined on line 3528, but no explicit reference was found in the text == Outdated reference: A later version (-99) exists of draft-templin-6man-omni-interface-56 == Outdated reference: A later version (-31) exists of draft-bonica-6man-comp-rtg-hdr-23 == Outdated reference: A later version (-04) exists of draft-bonica-6man-crh-helper-opt-02 == 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 (~~), 23 warnings (==), 4 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, December 11, 2020 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: June 14, 2021 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-intarea-6706bis-76 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 June 14, 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) . . . . . . . . . . 20 75 3.2.6. Segment Routing For OMNI Link Selection . . . . . . . 21 76 3.2.7. Segment Routing Within the OMNI Link . . . . . . . . 21 77 3.3. OMNI Interface Characteristics . . . . . . . . . . . . . 22 78 3.4. OMNI Interface Initialization . . . . . . . . . . . . . . 24 79 3.4.1. AERO Server/Relay Behavior . . . . . . . . . . . . . 24 80 3.4.2. AERO Proxy Behavior . . . . . . . . . . . . . . . . . 25 81 3.4.3. AERO Client Behavior . . . . . . . . . . . . . . . . 25 82 3.4.4. AERO Bridge Behavior . . . . . . . . . . . . . . . . 25 83 3.5. OMNI Interface Neighbor Cache Maintenance . . . . . . . . 25 84 3.5.1. OMNI Neighbor Interface Attributes . . . . . . . . . 27 85 3.5.2. OMNI Neighbor Advertisement Message Flags . . . . . . 28 86 3.6. OMNI Interface Encapsulation and Re-encapsulation . . . . 28 87 3.7. OMNI Interface Decapsulation . . . . . . . . . . . . . . 30 88 3.8. OMNI Interface Data Origin Authentication . . . . . . . . 30 89 3.9. OMNI Adaptation Layer and OMNI Interface MTU . . . . . . 30 90 3.10. OMNI Interface Forwarding Algorithm . . . . . . . . . . . 31 91 3.10.1. Client Forwarding Algorithm . . . . . . . . . . . . 32 92 3.10.2. Proxy Forwarding Algorithm . . . . . . . . . . . . . 32 93 3.10.3. Server/Relay Forwarding Algorithm . . . . . . . . . 33 94 3.10.4. Bridge Forwarding Algorithm . . . . . . . . . . . . 34 96 3.11. OMNI Interface Error Handling . . . . . . . . . . . . . . 35 97 3.12. AERO Router Discovery, Prefix Delegation and 98 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 38 99 3.12.1. AERO Service Model . . . . . . . . . . . . . . . . . 38 100 3.12.2. AERO Client Behavior . . . . . . . . . . . . . . . . 38 101 3.12.3. AERO Server Behavior . . . . . . . . . . . . . . . . 40 102 3.13. The AERO Proxy . . . . . . . . . . . . . . . . . . . . . 43 103 3.13.1. Combined Proxy/Servers . . . . . . . . . . . . . . . 45 104 3.13.2. Detecting and Responding to Server Failures . . . . 46 105 3.13.3. Point-to-Multipoint Server Coordination . . . . . . 46 106 3.14. AERO Route Optimization / Address Resolution . . . . . . 47 107 3.14.1. Route Optimization Initiation . . . . . . . . . . . 48 108 3.14.2. Relaying the NS . . . . . . . . . . . . . . . . . . 48 109 3.14.3. Processing the NS and Sending the NA . . . . . . . . 48 110 3.14.4. Relaying the NA . . . . . . . . . . . . . . . . . . 49 111 3.14.5. Processing the NA . . . . . . . . . . . . . . . . . 50 112 3.14.6. Route Optimization Maintenance . . . . . . . . . . . 50 113 3.15. Neighbor Unreachability Detection (NUD) . . . . . . . . . 51 114 3.16. Mobility Management and Quality of Service (QoS) . . . . 52 115 3.16.1. Mobility Update Messaging . . . . . . . . . . . . . 53 116 3.16.2. Announcing Link-Layer Address and/or QoS Preference 117 Changes . . . . . . . . . . . . . . . . . . . . . . 54 118 3.16.3. Bringing New Links Into Service . . . . . . . . . . 55 119 3.16.4. Deactivating Existing Links . . . . . . . . . . . . 55 120 3.16.5. Moving Between Servers . . . . . . . . . . . . . . . 55 121 3.17. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 56 122 3.17.1. Source-Specific Multicast (SSM) . . . . . . . . . . 57 123 3.17.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 58 124 3.17.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 59 125 3.18. Operation over Multiple OMNI Links . . . . . . . . . . . 59 126 3.19. DNS Considerations . . . . . . . . . . . . . . . . . . . 60 127 3.20. Transition Considerations . . . . . . . . . . . . . . . . 60 128 3.21. Detecting and Reacting to Server and Bridge Failures . . 61 129 3.22. AERO Clients on the Open Internet . . . . . . . . . . . . 62 130 3.23. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . 64 131 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 65 132 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 65 133 6. Security Considerations . . . . . . . . . . . . . . . . . . . 65 134 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 67 135 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 68 136 8.1. Normative References . . . . . . . . . . . . . . . . . . 68 137 8.2. Informative References . . . . . . . . . . . . . . . . . 70 138 Appendix A. Non-Normative Considerations . . . . . . . . . . . . 76 139 A.1. Implementation Strategies for Route Optimization . . . . 76 140 A.2. Implicit Mobility Management . . . . . . . . . . . . . . 76 141 A.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 77 142 A.4. AERO Critical Infrastructure Considerations . . . . . . . 77 143 A.5. AERO Server Failure Implications . . . . . . . . . . . . 78 144 A.6. AERO Client / Server Architecture . . . . . . . . . . . . 79 145 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 81 146 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 82 148 1. Introduction 150 Asymmetric Extended Route Optimization (AERO) fulfills the 151 requirements of Distributed Mobility Management (DMM) [RFC7333] and 152 route optimization [RFC5522] for aeronautical networking and other 153 network mobility use cases such as intelligent transportation 154 systems. AERO is an internetworking and mobility management service 155 based on the Overlay Multilink Network Interface (OMNI) 156 [I-D.templin-6man-omni-interface] Non-Broadcast, Multiple Access 157 (NBMA) virtual link model. The OMNI link is a virtual overlay 158 configured over one or more underlying Internetworks, and nodes on 159 the link can exchange IP packets via tunneling. The OMNI Adaptation 160 Layer (OAL) supports multilink operation for increased reliability, 161 bandwidth optimization and traffic path selection while accommodating 162 Maximum Transmission Unit (MTU) diversity. 164 The AERO service comprises Clients, Proxys, Servers and Relays that 165 are seen as OMNI link neighbors as well as Bridges that interconnect 166 OMNI link segments. Each node's OMNI interface uses an IPv6 link- 167 local address format that supports operation of the IPv6 Neighbor 168 Discovery (ND) protocol [RFC4861] and links ND to IP forwarding. A 169 node's OMNI interface can be configured over multiple underlying 170 interfaces, and may therefore appear as a single interface with 171 multiple link-layer addresses. Each link-layer address is subject to 172 change due to mobility and/or QoS fluctuations, and link-layer 173 address changes are signaled by ND messaging the same as for any IPv6 174 link. 176 AERO provides a cloud-based service where mobile nodes may use any 177 Server acting as a Mobility Anchor Point (MAP) and fixed nodes may 178 use any Relay on the link for efficient communications. Fixed nodes 179 forward packets destined to other AERO nodes to the nearest Relay, 180 which forwards them through the cloud. A mobile node's initial 181 packets are forwarded through the Server, while direct routing is 182 supported through asymmetric extended route optimization while data 183 packets are flowing. Both unicast and multicast communications are 184 supported, and mobile nodes may efficiently move between locations 185 while maintaining continuous communications with correspondents and 186 without changing their IP Address. 188 AERO Bridges are interconnected in a secured private BGP overlay 189 routing instance using encapsulation to provide a hybrid routing/ 190 bridging service that joins the underlying Internetworks of multiple 191 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 domain 297 a set of affiliated OMNI links that collectively provide services 298 under a common (set of) Mobility Service Prefixes (MSPs). 300 OMNI Interface 301 a node's attachment to an OMNI link. Since the addresses assigned 302 to an OMNI interface are managed for uniqueness, OMNI interfaces 303 do not require Duplicate Address Detection (DAD) and therefore set 304 the administrative variable 'DupAddrDetectTransmits' to zero 305 [RFC4862]. 307 OMNI Adaptation Layer (OAL) 308 an OMNI interface process whereby packets admitted into the 309 interface are wrapped in a mid-layer IPv6 header and fragmented/ 310 reassembled if necessary to support the OMNI link Maximum 311 Transmission Unit (MTU). The OAL is also responsible for 312 generating MTU-related control messages as necessary, and for 313 providing addressing context for spanning multiple segments of a 314 bridged OMNI link. 316 OMNI Link-Local Address (LLA) 317 a link local IPv6 address per [RFC4291] constructed as specified 318 in Section 3.2.2. 320 OMNI Unique-Local Address (ULA) 321 an IPv6 address from the IPv6 ULA prefix fd00::/8 [RFC4193], and 322 constructed as specified in [I-D.templin-6man-omni-interface]. 323 OMNI ULAs are statelessly derived from OMNI LLAs, and vice-versa. 325 underlying interface 326 an ANET or INET interface over which an OMNI interface is 327 configured. 329 Mobility Service Prefix (MSP) 330 an IP prefix assigned to the OMNI link and from which more- 331 specific Mobile Network Prefixes (MNPs) are derived. For IPv6, 332 the MSP is typically taken from an IPv6 Globally Unqiue Address 333 (GUA) prefix, e.g., obtained from an assigned numbers registry. 335 Mobile Network Prefix (MNP) 336 an IP prefix allocated from an MSP and delegated to an AERO Client 337 or Relay. 339 AERO node 340 a node that is connected to an OMNI link and participates in the 341 AERO internetworking and mobility service. 343 AERO Client ("Client") 344 an AERO node that connects over one or more underlying interfaces 345 and requests MNP delegation/registration service from AERO 346 Servers. The Client assigns a Client LLA to the OMNI interface 347 for use in ND exchanges with other AERO nodes and forwards packets 348 to correspondents according to OMNI interface neighbor cache 349 state. 351 AERO Server ("Server") 352 an INET node that configures an OMNI interface to provide default 353 forwarding and mobility/multilink services for AERO Clients. The 354 Server assigns an administratively-provisioned LLA to its OMNI 355 interface to support the operation of the ND services, and 356 advertises all of its associated MNPs via BGP peerings with 357 Bridges. 359 AERO Relay ("Relay") 360 an AERO Server that also provides forwarding services between 361 nodes reached via the OMNI link and correspondents on other links. 362 AERO Relays are provisioned with MNPs (i.e., the same as for an 363 AERO Client) and run a dynamic routing protocol to discover any 364 non-MNP IP routes. In both cases, the Relay advertises the MSP(s) 365 to its downstream networks, and distributes all of its associated 366 MNPs and non-MNP IP routes via BGP peerings with Bridges (i.e., 367 the same as for an AERO Server). 369 AERO Bridge ("Bridge") 370 a node that provides hybrid routing/bridging services (as well as 371 a security trust anchor) for nodes on an OMNI link. As a router, 372 the Bridge forwards packets using standard IP forwarding. As a 373 bridge, the Bridge forwards packets over the OMNI link without 374 decrementing the IPv6 Hop Limit. AERO Bridges peer with Servers 375 and other Bridges to discover the full set of MNPs for the link as 376 well as any non-MNPs that are reachable via Relays. 378 AERO Proxy ("Proxy") 379 a node that provides proxying services between Clients in an ANET 380 and Servers in external INETs. The AERO Proxy is a conduit 381 between the ANET and external INETs in the same manner as for 382 common web proxies, and behaves in a similar fashion as for ND 383 proxies [RFC4389]. A node may be configured to act as either a 384 Proxy and/or a Server, depending on Client Server selection 385 criteria. 387 ingress tunnel endpoint (ITE) 388 an OMNI interface endpoint that injects encapsulated packets into 389 an OMNI link. 391 egress tunnel endpoint (ETE) 392 an OMNI interface endpoint that receives encapsulated packets from 393 an OMNI link. 395 link-layer address 396 an IP address used as an encapsulation header source or 397 destination address from the perspective of the OMNI interface. 398 When an upper layer protocol (e.g., UDP) is used as part of the 399 encapsulation, the port number is also considered as part of the 400 link-layer address. 402 network layer address 403 the source or destination address of an encapsulated IP packet 404 presented to the OMNI interface. 406 end user network (EUN) 407 an internal virtual or external edge IP network that an AERO 408 Client or Relay connects to the rest of the network via the OMNI 409 interface. The Client/Relay sees each EUN as a "downstream" 410 network, and sees the OMNI interface as the point of attachment to 411 the "upstream" network. 413 Mobile Node (MN) 414 an AERO Client and all of its downstream-attached networks that 415 move together as a single unit, i.e., an end system that connects 416 an Internet of Things. 418 Mobile Router (MR) 419 a MN's on-board router that forwards packets between any 420 downstream-attached networks and the OMNI link. 422 Route Optimization Source (ROS) 423 the AERO node nearest the source that initiates route 424 optimization. The ROS may be a Server or Proxy acting on behalf 425 of the source Client. 427 Route Optimization responder (ROR) 428 the AERO node nearest the target destination that responds to 429 route optimization requests. The ROR may be a Server acting on 430 behalf of a target MNP Client, or a Relay for a non-MNP 431 destination. 433 MAP List 434 a geographically and/or topologically referenced list of addresses 435 of all Servers within the same OMNI link. There is a single MAP 436 list for the entire OMNI link. 438 Distributed Mobility Management (DMM) 439 a BGP-based overlay routing service coordinated by Servers and 440 Bridges that tracks all Server-to-Client associations. 442 Mobility Service (MS) 443 the collective set of all Servers, Proxys, Bridges and Relays that 444 provide the AERO Service to Clients. 446 Mobility Service Endpoint MSE) 447 an individual Server, Proxy, Bridge or Relay in the Mobility 448 Service. 450 Throughout the document, the simple terms "Client", "Server", 451 "Bridge", "Proxy" and "Relay" refer to "AERO Client", "AERO Server", 452 "AERO Bridge", "AERO Proxy" and "AERO Relay", respectively. 453 Capitalization is used to distinguish these terms from other common 454 Internetworking uses in which they appear without capitalization. 456 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 457 the names of node variables, messages and protocol constants) is used 458 throughout this document. The terms "All-Routers multicast", "All- 459 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 460 anycast" are defined in [RFC4291]. Also, the term "IP" is used to 461 generically refer to either Internet Protocol version, i.e., IPv4 462 [RFC0791] or IPv6 [RFC8200]. 464 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 465 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 466 "OPTIONAL" in this document are to be interpreted as described in BCP 467 14 [RFC2119][RFC8174] when, and only when, they appear in all 468 capitals, as shown here. 470 3. Asymmetric Extended Route Optimization (AERO) 472 The following sections specify the operation of IP over OMNI links 473 using the AERO service: 475 3.1. AERO Node Types 477 AERO Clients are Mobile Nodes (MNs) that connect via underlying 478 interfaces with addresses that may change when the Client moves to a 479 new network connection point. AERO Clients register their Mobile 480 Network Prefixes (MNPs) with the AERO service, and distribute the 481 MNPs to nodes on EUNs. AERO Bridges, Servers, Proxys and Relays are 482 critical infrastructure elements in fixed (i.e., non-mobile) INET 483 deployments and hence have permanent and unchanging INET addresses. 484 Together, they constitute the AERO service which provides an OMNI 485 link virtual overlay for connecting AERO Clients. 487 AERO Bridges provide hybrid routing/bridging services (as well as a 488 security trust anchor) for nodes on an OMNI link. Bridges use 489 standard IPv6 routing to forward packets both within the same INET 490 partitions and between disjoint INET partitions based on a mid-layer 491 IPv6 encapsulation per [RFC2473]. The inner IP layer experiences a 492 virtual bridging service since the inner IP TTL/Hop Limit is not 493 decremented during forwarding. Each Bridge also peers with Servers 494 and other Bridges in a dynamic routing protocol instance to provide a 495 Distributed Mobility Management (DMM) service for the list of active 496 MNPs (see Section 3.2.3). Bridges present the OMNI link as a set of 497 one or more Mobility Service Prefixes (MSPs) and configure secured 498 tunnels with Servers, Relays, Proxys and other Bridges; they further 499 maintain IP forwarding table entries for each MNP and any other 500 reachable non-MNP prefixes. 502 AERO Servers provide default forwarding and mobility/multilink 503 services for AERO Client Mobile Nodes (MNs). Each Server also peers 504 with Bridges in a dynamic routing protocol instance to advertise its 505 list of associated MNPs (see Section 3.2.3). Servers facilitate 506 prefix delegation/registration exchanges with Clients, where each 507 delegated prefix becomes an MNP taken from an MSP. Servers forward 508 packets between OMNI interface neighbors and track each Client's 509 mobility profiles. Servers may further act as Servers for some sets 510 of Clients and as Proxies for others. 512 AERO Proxys provide a conduit for ANET Clients to associate with 513 Servers in external INETs. Client and Servers exchange control plane 514 messages via the Proxy acting as a bridge between the ANET/INET 515 boundary. The Proxy forwards data packets between Clients and the 516 OMNI link according to forwarding information in the neighbor cache. 517 The Proxy function is specified in Section 3.13. Proxys may further 518 act as Proxys for some sets of Clients and as Servers for others. 520 AERO Relays are Servers that provide forwarding services between the 521 OMNI interface and INET/EUN interfaces. Relays are provisioned with 522 MNPs the same as for an AERO Client, and also run a dynamic routing 523 protocol to discover any non-MNP IP routes. The Relay advertises the 524 MSP(s) to its connected networks, and distributes all of its 525 associated MNPs and non-MNP IP routes via BGP peerings with Bridges 527 3.2. The AERO Service over OMNI Links 529 3.2.1. AERO/OMNI Reference Model 531 Figure 1 presents the basic OMNI link reference model: 533 +----------------+ 534 | AERO Bridge B1 | 535 | Nbr: S1, S2, P1| 536 |(X1->S1; X2->S2)| 537 | MSP M1 | 538 +-+---------+--+-+ 539 +--------------+ | Secured | | +--------------+ 540 |AERO Server S1| | tunnels | | |AERO Server S2| 541 | Nbr: C1, B1 +-----+ | +-----+ Nbr: C2, B1 | 542 | default->B1 | | | default->B1 | 543 | X1->C1 | | | X2->C2 | 544 +-------+------+ | +------+-------+ 545 | OMNI link | | 546 X===+===+===================+==)===============+===+===X 547 | | | | 548 +-----+--------+ +--------+--+-----+ +--------+-----+ 549 |AERO Client C1| | AERO Proxy P1 | |AERO Client C2| 550 | Nbr: S1 | |(Proxy Nbr Cache)| | Nbr: S2 | 551 | default->S1 | +--------+--------+ | default->S2 | 552 | MNP X1 | | | MNP X2 | 553 +------+-------+ .--------+------. +-----+--------+ 554 | (- Proxyed Clients -) | 555 .-. `---------------' .-. 556 ,-( _)-. ,-( _)-. 557 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 558 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 559 `-(______)-' +-------+ +-------+ `-(______)-' 561 Figure 1: AERO/OMNI Reference Model 563 In this model: 565 o the OMNI link is an overlay network service configured over one or 566 more underlying INET partitions which may be managed by different 567 administrative authorities and have incompatible protocols and/or 568 addressing plans. 570 o AERO Bridge B1 aggregates Mobility Service Prefix (MSP) M1, 571 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 572 via BGP peerings over secured tunnels to Servers (S1, S2). 573 Bridges connect the disjoint segments of a partitioned OMNI link. 575 o AERO Servers/Relays S1 and S2 configure secured tunnels with 576 Bridge B1 and also provide mobility, multilink and default router 577 services for their associated Clients C1 and C2. 579 o AERO Clients C1 and C2 associate with Servers S1 and S2, 580 respectively. They receive Mobile Network Prefix (MNP) 581 delegations X1 and X2, and also act as default routers for their 582 associated physical or internal virtual EUNs. Simple hosts H1 and 583 H2 attach to the EUNs served by Clients C1 and C2, respectively. 585 o AERO Proxy P1 configures a secured tunnel with Bridge B1 and 586 provides proxy services for AERO Clients in secured enclaves that 587 cannot associate directly with other OMNI link neighbors. 589 An OMNI link configured over a single INET appears as a single 590 unified link with a consistent underlying network addressing plan. 591 In that case, all nodes on the link can exchange packets via simple 592 INET encapsulation, since the underlying INET is connected. In 593 common practice, however, an OMNI link may be partitioned into 594 multiple "segments", where each segment is a distinct INET 595 potentially managed under a different administrative authority (e.g., 596 as for worldwide aviation service providers such as ARINC, SITA, 597 Inmarsat, etc.). Individual INETs may also themselves be partitioned 598 internally, in which case each internal partition is seen as a 599 separate segment. 601 The addressing plan of each segment is consistent internally but will 602 often bear no relation to the addressing plans of other segments. 603 Each segment is also likely to be separated from others by network 604 security devices (e.g., firewalls, proxies, packet filtering 605 gateways, etc.), and in many cases disjoint segments may not even 606 have any common physical link connections. Therefore, nodes can only 607 be assured of exchanging packets directly with correspondents in the 608 same segment, and not with those in other segments. The only means 609 for joining the segments therefore is through inter-domain peerings 610 between AERO Bridges. 612 The same as for traditional campus LANs, multiple OMNI link segments 613 can be joined into a single unified link via a virtual bridging 614 service using the OMNI Adaptation Layer (OAL) which inserts a mid- 615 layer IPv6 encapsulation per [RFC2473] that supports inter-segment 616 forwarding (i.e., bridging) without decrementing the network-layer 617 TTL/Hop Limit. This bridging of OMNI link segments is shown in 618 Figure 2: 620 . . . . . . . . . . . . . . . . . . . . . . . 621 . . 622 . .-(::::::::) . 623 . .-(::::::::::::)-. +-+ . 624 . (:::: Segment A :::)--|B|---+ . 625 . `-(::::::::::::)-' +-+ | . 626 . `-(::::::)-' | . 627 . | . 628 . .-(::::::::) | . 629 . .-(::::::::::::)-. +-+ | . 630 . (:::: Segment B :::)--|B|---+ . 631 . `-(::::::::::::)-' +-+ | . 632 . `-(::::::)-' | . 633 . | . 634 . .-(::::::::) | . 635 . .-(::::::::::::)-. +-+ | . 636 . (:::: Segment C :::)--|B|---+ . 637 . `-(::::::::::::)-' +-+ | . 638 . `-(::::::)-' | . 639 . | . 640 . ..(etc).. x . 641 . . 642 . . 643 . <- OMNI link Bridged by encapsulation -> . 644 . . . . . . . . . . . . . .. . . . . . . . . 646 Figure 2: Bridging OMNI Link Segments 648 Bridges, Servers, Relays and Proxys connect via secured INET tunnels 649 over their respective segments in a spanning tree topology rooted at 650 the Bridges. The secured spanning tree supports strong 651 authentication for IPv6 ND control messages and may also be used to 652 convey the initial data packets in a flow. Route optimization can 653 then be employed to cause data packets to take more direct paths 654 between OMNI link neighbors without having to strictly follow the 655 spanning tree. 657 3.2.2. Link-Local Addresses (LLAs) and Unique Local Addresses (ULAs) 659 AERO nodes on OMNI links use the Link-Local Address (LLA) prefix 660 fe80::/64 [RFC4291] to assign LLAs used for network-layer addresses 661 in IPv6 ND and data messages. They also use the Unique Local Address 662 (ULA) prefix [ULA]::/48 to form per-link ULA sub-prefixes of the form 663 [ULA*]::/64, where '*' is a 16-bit OMNI link identifier. Each per- 664 link ULA sub-prefix is in turn used to form ULAs used for OAL header 665 source and destination addresses. See 666 [I-D.templin-6man-omni-interface] for a full specification of the 667 LLAs and ULAs used by AERO nodes on OMNI links. 669 IPv6 routers are not permitted to forward packets with LLA addresses. 670 Therefore, IPv6 packets with LLA addresses must be encapsulated in 671 IPv6 headers with ULA addresses in order to traverse the OMNI link. 672 ULA addresses therefore must be routable within the OMNI link domain- 673 local area. 675 For routing system organization (see Section 3.2.3), ULAs are 676 organized in partition prefixes, e.g., [ULA*]::1000/116. For each 677 such partition prefix, the Bridge(s) that connect that segment assign 678 the all-zero's address of the prefix as a Subnet Router Anycast 679 address. For example, the Subnet Router Anycast address for 680 [ULA*]::1000/116 is simply [ULA*]::1000. 682 3.2.3. AERO Routing System 684 The AERO routing system comprises a private instance of the Border 685 Gateway Protocol (BGP) [RFC4271] that is coordinated between Bridges 686 and Servers and does not interact with either the public Internet BGP 687 routing system or any underlying INET routing systems. 689 In a reference deployment, each Server is configured as an Autonomous 690 System Border Router (ASBR) for a stub Autonomous System (AS) using 691 an AS Number (ASN) that is unique within the BGP instance, and each 692 Server further uses eBGP to peer with one or more Bridges but does 693 not peer with other Servers. Each INET of a multi-segment OMNI link 694 must include one or more Bridges, which peer with the Servers and 695 Proxys within that INET. All Bridges within the same INET are 696 members of the same hub AS using a common ASN, and use iBGP to 697 maintain a consistent view of all active MNPs currently in service. 698 The Bridges of different INETs peer with one another using eBGP. 700 Bridges advertise the OMNI link's MSPs and any non-MNP routes to each 701 of their Servers. This means that any aggregated non-MNPs (including 702 "default") are advertised to all Servers. Each Bridge configures a 703 black-hole route for each of its MSPs. By black-holing the MSPs, the 704 Bridge will maintain forwarding table entries only for the MNPs that 705 are currently active, and packets destined to all other MNPs will 706 correctly incur Destination Unreachable messages due to the black- 707 hole route. In this way, Servers have only partial topology 708 knowledge (i.e., they know only about the MNPs of their directly 709 associated Clients) and they forward all other packets to Bridges 710 which have full topology knowledge. 712 Each OMNI link segment assigns a unique sub-prefix of [ULA*]::/96 713 known as the ULA partition prefix. For example, a first segment 714 could assign [ULA*]::1000/116, a second could assign 715 [ULA*]::2000/116, a third could assign [ULA*]::3000/116, etc. The 716 administrative authorities for each segment must therefore coordinate 717 to assure mutually-exclusive partition prefix assignments, but 718 internal provisioning of each prefix is an independent local 719 consideration for each administrative authority. 721 ULA partition prefixes are statically represented in Bridge 722 forwarding tables. Bridges join multiple segments into a unified 723 OMNI link over multiple diverse administrative domains. They support 724 a bridging function by first establishing forwarding table entries 725 for their partition prefixes either via standard BGP routing or 726 static routes. For example, if three Bridges ('A', 'B' and 'C') from 727 different segments serviced [ULA*]::1000/116, [ULA*]::2000/116 and 728 [ULA*]::3000/116 respectively, then the forwarding tables in each 729 Bridge are as follows: 731 A: [ULA*]::1000/116->local, [ULA*]::2000/116->B, [ULA*]::3000/116->C 733 B: [ULA*]::1000/116->A, [ULA*]::2000/116->local, [ULA*]::3000/116->C 735 C: [ULA*]::1000/116->A, [ULA*]::2000/116->B, [ULA*]::3000/116->local 737 These forwarding table entries are permanent and never change, since 738 they correspond to fixed infrastructure elements in their respective 739 segments. 741 ULA Client prefixes are instead dynamically advertised in the AERO 742 routing system by Servers and Relays that provide service for their 743 corresponding MNPs. For example, if three Servers ('D', 'E' and 'F') 744 service the MNPs 2001:db8:1000:2000::/56, 2001:db8:3000:4000::/56 and 745 2001:db8:5000:6000::/56 then the routing system would include: 747 D: [ULA*]:2001:db8:1000:2000/120 749 E: [ULA*]:2001:db8:3000:4000/120 751 F: [ULA*]:2001:db8:5000:6000/120 753 A full discussion of the BGP-based routing system used by AERO is 754 found in [I-D.ietf-rtgwg-atn-bgp]. 756 3.2.4. OMNI Link Encapsulation 758 With the Client and partition prefixes in place in Bridge forwarding 759 tables, the OMNI interface sends control and data packets toward AERO 760 destination nodes located in different OMNI link segments over the 761 spanning tree via mid-layer encapsulation using the OMNI Adaptation 762 Layer (OAL) header based on Generic Packet Tunneling in IPv6 763 [RFC2473]. When necessary, the OMNI interface also includes an OMNI 764 Routing Header (ORH) as an extension to the OAL header if final 765 segment forwarding information is available, e.g., in the neighbor 766 cache. The ORH is formatted as shown in Figure 3: 768 0 1 2 3 769 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 1 770 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 771 | Next Header | Hdr Ext Len | Routing Type | SRT | FMT | 772 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 773 | Last Hop Segment-id (LHS) | 774 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 775 ~ Link Layer Address (L2ADDR) ~ 776 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 777 ~ Destination Suffix (if necessary) ~ 778 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 779 ~ Null Padding (if necessary) ~ 780 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 782 Figure 3: OMNI Routing Header (ORH) Format 784 In this format: 786 o Next Header identifies the type of header immediately following 787 the ORH. 789 o Hdr Ext Len is the length of the Routing header in 8-octet units 790 (not including the first 8 octets), with trailing padding added if 791 necessary to produce an integral number of 8-octet units. 793 o Routing Type is set to TBD (see IANA Considerations). 795 o Segments Left is omitted, and replaced by a 5-bit SRT and 3-bit 796 FMT field. 798 o SRT - a 5-bit Segment Routing Topology prefix length value that 799 (when added to 96) determines the prefix length to apply to the 800 ULA formed from concatenating [ULA*]::/96 with the 32 bit LHS MSID 801 value that follows. For example, the value 16 corresponds to the 802 prefix length 112. 804 o FMT - a 3-bit "Framework/Mode/Type" code corresponding to the 805 included Link Layer Address as follows: 807 * When the most significant bit (i.e., "Framework") is set to 0, 808 L2ADDR is the INET encapsulation address of the Server/Proxy 809 named in the LHS; otherwise, it is the address for the target 810 Client, and the Destination Suffix is included. 812 * When the next most significant bit (i.e., "Mode") is set to 0, 813 the Source/Target L2ADDR is on the open INET; otherwise, it is 814 (likely) located behind a Network Address Translator (NAT). 816 * When the least significant bit (i.e., "Type") is set to 0, 817 L2ADDR includes a UDP Port Number followed by an IPv4 address; 818 otherwise, it includes a UDP Port Number followed by an IPv6 819 address. 821 o LHS - the 32 bit MSID of a service node in the Last Hop Segment on 822 the path to the target. When SRT and LHS are both set to 0, the 823 LHS is considered unspecified. When SRT is set to 0 and LHS is 824 non-zero, the prefix length is set to 128. SRT and LHS provide 825 guidance to the OMNI interface forwarding algorithm. 826 Specifically, if SRT/LHS is located in the local OMNI link segment 827 then the OMNI interface can omit OAL/ORH encapsualtion and send 828 directly to the target using INET encapsulation according to FMT/ 829 L2ADDR; else, it must perform INET/OAL/ORH encapsulation and 830 forward according to the OMNI link spanning tree. 832 o Link Layer Address (L2ADDR) - Formatted according to FMT, and 833 identifies the link-layer address (i.e., the encapsulation 834 address) of the source/target. The UDP Port Number appears in the 835 first two octets and the IP address appears in the next 4 octets 836 for IPv4 or 16 octets for IPv6. The Port Number and IP address 837 are recorded in ones-compliment "obfuscated" form per [RFC4380]. 838 The OMNI interface forwarding algorithm uses FMT/L2ADDR to 839 determine the INET encapsulation address for local forwarding when 840 SRT/LHS is located in the same OMNI link segment. 842 o Destination Suffix is a 64-bit field included only for OAL- 843 encapsulated packets that are destined directly to the ULA of the 844 Client (i.e., according to the FMT code). When present, 845 Destination Suffix encodes the 64-bit ULA suffix for the Client 846 that will receive packet. For example, if the Client ULA is 847 [ULA*]:2001:db8:1:2 then the Destination suffix encodes the value 848 2001:db8:1:2. 850 o Null Padding contains zero-valued octets as necessary to pad the 851 ORH to an integral number of 8-octet units. 853 When an AERO node uses OAL encapsulation for a packet with addresses 854 such as 2001:db8:1:2::1 and 2001:db8:1234:5678::1, it sets the OAL 855 header source address to its own ULA address (e.g., 856 [ULA*]::1000:2000). The node also sets the destination address to 857 the ULA of the Client (e.g., [ULA*]::2001:db8:1234:5678) when the 858 Client is addressed directly; otherwise, it sets the destination 859 address to the ULA of the Client's Proxy/Server (e.g., 861 [ULA*]::4321:9876). If the neighbor cache includes Last Hop Segment 862 information for the target destination, the node next inserts an ORH 863 immediately following the OAL header while including the correct SRT, 864 FMT, LHS, L2ADDR and (if necessary) Destination Suffix information. 865 Next, the node overwrites the OAL header destination address with the 866 LHS Subnet Router Anycast address (for example, for LHS 1000:2000 867 with SRT 16, the Subnet Router Anycast address is [ULA*]::1000:0000). 869 The node then fragments the OAL/ORH packet if necessary, with each 870 resulting fragment including the OAL/ORH headers while only the first 871 fragment includes the original IPv6 header. The node finally 872 encapsulates each resulting OAL/ORH packet/fragment in an INET header 873 with source address set to its own INET address (e.g., 192.0.2.100) 874 and destination set to the INET address of a Bridge (e.g., 875 192.0.2.1). 877 The encapsulation format in the above example is shown in Figure 4: 879 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 880 | INET Header | 881 | src = 192.0.2.100 | 882 | dst = 192.0.2.1 | 883 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 884 | OAL Header | 885 | src = [ULA*]::1000:2000 | 886 | dst= ULA for inner IP dst | 887 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 888 | ORH Header | 889 | (if necessary) | 890 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 891 | Inner IP Header | 892 | src = 2001:db8:1:2::1 | 893 | dst = 2001:db8:1234:5678::1 | 894 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 895 | | 896 ~ ~ 897 ~ Inner Packet Body ~ 898 ~ ~ 899 | | 900 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 902 Figure 4: OAL/ORH Encapsulation 904 In this format, the inner IP header and packet body are the original 905 IP packet, the OAL header is an IPv6 header prepared according to 906 [RFC2473], the ORH is a Routing Header extension of the OAL header, 907 and the INET header is prepared as discussed in Section 3.6. 909 This gives rise to a routing system that contains both Client prefix 910 routes that may change dynamically due to regional node mobility and 911 partition prefix routes that rarely if ever change. The Bridges can 912 therefore provide link-layer bridging by sending packets over the 913 spanning tree instead of network-layer routing according to MNP 914 routes. As a result, opportunities for packet loss due to node 915 mobility between different segments are mitigated. 917 In normal operations, IPv6 ND messages are conveyed over secured 918 paths between OMNI link neighbors so that specific Proxys, Servers or 919 Relays can be addressed without being subject to mobility events. 920 Conversely, only the first few packets destined to Clients need to 921 traverse secured paths until route optimization can determine a more 922 direct path. 924 Note: An IPv6 "minimal encapsulation" format (i.e., an IPv6 variant 925 of [RFC2004]) based on extensions to the ORH was considered, analyzed 926 and rejected. In the approach, the ORH would be inserted as an 927 extension header to the original IPv6 packet header. The IPv6 928 destination address would then be written into the ORH, and the ULA 929 corresponding to the destination would be overwritten in the IPv6 930 destination address. This would seemingly convey enough forwarding 931 information so that OAL encapsulation could be avoided. However, 932 this "minimal encapsulation" IPv6 packet would then have a non-ULA 933 source address and ULA destination address, an incorrect value in 934 upper layer protocol checksums, and a Hop Limit that is decremented 935 within the spanning tree when it should not be. The insertion and 936 removal of the ORH would also entail rewriting the Payload Length and 937 Next Header fields - again, invalidating upper layer checksums. 938 These irregularities would result in implementation challenges and 939 the potential for operational issues, e.g., since actionable ICMPv6 940 error reports could not be delivered to the original source. In 941 order to address the issues, still more information such as the 942 original IPv6 source address could be written into the ORH. However, 943 with the additional information the benefit of the "minimal 944 encapsulation" savings quickly diminishes, and becomes overshadowed 945 by the implementation and operational irregularities. 947 3.2.5. Segment Routing Topologies (SRTs) 949 The 64-bit sub-prefixes of [ULA]::/48 identify up to 2^16 distinct 950 Segment Routing Topologies (SRTs). Each SRT is a mutually-exclusive 951 OMNI link overlay instance using a distinct set of ULAs, and emulates 952 a Virtual LAN (VLAN) service for the OMNI link. In some cases (e.g., 953 when redundant topologies are needed for fault tolerance and 954 reliability) it may be beneficial to deploy multiple SRTs that act as 955 independent overlay instances. A communication failure in one 956 instance therefore will not affect communications in other instances. 958 Each SRT is identified by a distinct value in bits 48-63 of 959 [ULA]::/48, i.e., as [ULA0]::/64, [ULA1]::/64, [ULA2]::/64, etc. 960 This document asserts that up to four SRTs provide a level of safety 961 sufficient for critical communications such as civil aviation. Each 962 SRT is designated with a color that identifies a different OMNI link 963 instance as follows: 965 o Red - corresponds to [ULA0]::/64 967 o Green - corresponds to [ULA1]::/64 969 o Blue-1 - corresponds to [ULA2]::/64 971 o Blue-2 - corresponds to [ULA3]::/64 973 o the remaining [ULA*]::/64 sub prefixes are available for 974 additional SRTs. 976 Each OMNI interface is identified by a unique interface name (e.g., 977 omni0, omni1, omni2, etc.) and assigns an anycast ULA corresponding 978 to its SRT prefix. For example, the anycast ULA for the Green SRT is 979 simply [ULA1]::. The anycast ULA is used for OMNI interface 980 determination in Safety-Based Multilink (SBM) as discussed in 981 [I-D.templin-6man-omni-interface]. Each OMNI interface further 982 applies Performance-Based Multilink (PBM) internally. 984 3.2.6. Segment Routing For OMNI Link Selection 986 An original IPv6 source can direct an IPv6 packet to an AERO node by 987 including a standard IPv6 Segment Routing Header (SRH) [RFC8754] with 988 the anycast ULA for the selected SRT as either the IPv6 destination 989 or as an intermediate hop within the SRH. This allows the original 990 source to determine the specific OMNI link topology a packet will 991 traverse when there may be multiple alternatives. 993 When the AERO node processes the SRH and forwards the packet to the 994 correct OMNI interface, the OMNI interface writes the next IPv6 995 address from the SRH into the IPv6 destination address and decrements 996 Segments Left. If decrementing would cause Segments Left to become 997 0, the OMNI interface deletes the SRH before forwarding. This form 998 of Segment Routing supports Safety-Based Multilink (SBM). 1000 3.2.7. Segment Routing Within the OMNI Link 1002 AERO node OMNI interfaces can insert OAL/ORH headers for Segment 1003 Routing within the OMNI link to influence the paths of packets 1004 destined to targets in remote segments without requiring all packets 1005 to traverse strict spanning tree paths. 1007 When an AERO node's OMNI interface has a packet to send to a target 1008 discovered through route optimization located in the same OMNI link 1009 segment, it encapsulates the packet in OAL/ORH headers if necessary 1010 as discussed above. The node then uses the target's Link Layer 1011 Address (L2ADDR) information for INET encapsulation. 1013 When an AERO node's OMNI interface has a packet to send to a route 1014 optimization target located in a remote OMNI link segment, it 1015 encapsulates the packet in OAL/ORH headers as discussed above while 1016 forwarding the packet to a Bridge with destination set to the Subnet 1017 Router Anycast address for the final OMNI link segment. 1019 When a Bridge receives a packet destined to its Subnet Router Anycast 1020 address with an OAL/ORH with SRT/LHS values corresponding to the 1021 local segment, it examines the L2ADDR according to FMT and removes 1022 the ORH from the packet. If the ORH includes a saved Destination 1023 Suffix, the Bridge then writes the corresponding ULA into the OAL 1024 destination address; otherwise, it writes the ULA corresponding to 1025 the SRT/LHS fields into the destination. The Bridge then 1026 encapsulates the packet in an INET header according to L2ADDR and 1027 forwards the packet within the INET either to the LHS Server/Proxy or 1028 directly to the destination itself. In this way, the Bridge 1029 participates in route optimization to reduce traffic load and 1030 suboptimal routing through strict spanning tree paths. 1032 3.3. OMNI Interface Characteristics 1034 OMNI interfaces are virtual interfaces configured over one or more 1035 underlying interfaces classified as follows: 1037 o INET interfaces connect to an INET either natively or through one 1038 or several IPv4 Network Address Translators (NATs). Native INET 1039 interfaces have global IP addresses that are reachable from any 1040 INET correspondent. All Server, Relay and Bridge interfaces are 1041 native interfaces, as are INET-facing interfaces of Proxys. NATed 1042 INET interfaces connect to a private network behind one or more 1043 NATs that provide INET access. Clients that are behind a NAT are 1044 required to send periodic keepalive messages to keep NAT state 1045 alive when there are no data packets flowing. 1047 o ANET interfaces connect to an ANET that is separated from the open 1048 INET by a Proxy. Proxys can actively issue control messages over 1049 the INET on behalf of the Client to reduce ANET congestion. 1051 o VPNed interfaces use security encapsulation over the INET to a 1052 Virtual Private Network (VPN) server that also acts as a Server or 1053 Proxy. Other than the link-layer encapsulation format, VPNed 1054 interfaces behave the same as Direct interfaces. 1056 o Direct interfaces connect a Client directly to a Server or Proxy 1057 without crossing any ANET/INET paths. An example is a line-of- 1058 sight link between a remote pilot and an unmanned aircraft. The 1059 same Client considerations apply as for VPNed interfaces. 1061 OMNI interfaces use OAL/ORH encapsulation as necessary as discussed 1062 in Section 3.2.4. OMNI interfaces use link-layer encapsulation (see: 1063 Section 3.6) to exchange packets with OMNI link neighbors over INET 1064 or VPNed interfaces as well as over ANET interfaces for which the 1065 Client and Proxy may be multiple IP hops away. OMNI interfaces do 1066 not use link-layer encapsulation over Direct underlying interfaces or 1067 ANET interfaces when the Client and Proxy are known to be on the same 1068 underlying link. 1070 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 1071 state the same as for any interface. OMNI interfaces use ND messages 1072 including Router Solicitation (RS), Router Advertisement (RA), 1073 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 1074 neighbor cache management. 1076 OMNI interfaces send ND messages with an OMNI option formatted as 1077 specified in [I-D.templin-6man-omni-interface]. The OMNI option 1078 includes prefix registration information and Interface Attributes 1079 containing link information parameters for the OMNI interface's 1080 underlying interfaces. Each OMNI option may include multiple 1081 Interface Attributes sub-options, each identified by an ifIndex 1082 value. 1084 A Client's OMNI interface may be configured over multiple underlying 1085 interface connections. For example, common mobile handheld devices 1086 have both wireless local area network ("WLAN") and cellular wireless 1087 links. These links are often used "one at a time" with low-cost WLAN 1088 preferred and highly-available cellular wireless as a standby, but a 1089 simultaneous-use capability could provide benefits. In a more 1090 complex example, aircraft frequently have many wireless data link 1091 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1092 directional, etc.) with diverse performance and cost properties. 1094 If a Client's multiple underlying interfaces are used "one at a time" 1095 (i.e., all other interfaces are in standby mode while one interface 1096 is active), then ND message OMNI options include only a single 1097 Interface Attributes sub-option set to constant values. In that 1098 case, the Client would appear to have a single interface but with a 1099 dynamically changing link-layer address. 1101 If the Client has multiple active underlying interfaces, then from 1102 the perspective of ND it would appear to have multiple link-layer 1103 addresses. In that case, ND message OMNI options MAY include 1104 multiple Interface Attributes sub-options - each with values that 1105 correspond to a specific interface. Every ND message need not 1106 include Interface Attributes for all underlying interfaces; for any 1107 attributes not included, the neighbor considers the status as 1108 unchanged. 1110 Bridge, Server and Proxy OMNI interfaces may be configured over one 1111 or more secured tunnel interfaces. The OMNI interface configures 1112 both an LLA and its corresponding ULA, while the underlying secured 1113 tunnel interfaces are either unnumbered or configure the same ULA. 1114 The OMNI interface encapsulates each IP packet in OAL/ORH headers and 1115 presents the packet to the underlying secured tunnel interface. 1116 Routing protocols such as BGP that run over the OMNI interface do not 1117 employ OAL/ORH encapsulation, but rather present the routing protocol 1118 messages directly to the underlying secured tunnels while using the 1119 ULA as the source address. This distinction must be honored 1120 consistently according to each node's configuration so that the IP 1121 forwarding table will associate discovered IP routes with the correct 1122 interface. 1124 3.4. OMNI Interface Initialization 1126 AERO Servers, Proxys and Clients configure OMNI interfaces as their 1127 point of attachment to the OMNI link. AERO nodes assign the MSPs for 1128 the link to their OMNI interfaces (i.e., as a "route-to-interface") 1129 to ensure that packets with destination addresses covered by an MNP 1130 not explicitly assigned to a non-OMNI interface are directed to the 1131 OMNI interface. 1133 OMNI interface initialization procedures for Servers, Proxys, Clients 1134 and Bridges are discussed in the following sections. 1136 3.4.1. AERO Server/Relay Behavior 1138 When a Server enables an OMNI interface, it assigns an LLA/ULA 1139 appropriate for the given OMNI link segment. The Server also 1140 configures secured tunnels with one or more neighboring Bridges and 1141 engages in a BGP routing protocol session with each Bridge. 1143 The OMNI interface provides a single interface abstraction to the IP 1144 layer, but internally comprises multiple secured tunnels as well as 1145 an NBMA nexus for sending encapsulated data packets to OMNI interface 1146 neighbors. The Server further configures a service to facilitate ND 1147 exchanges with AERO Clients and manages per-Client neighbor cache 1148 entries and IP forwarding table entries based on control message 1149 exchanges. 1151 Relays are simply Servers that run a dynamic routing protocol to 1152 redistribute routes between the OMNI interface and INET/EUN 1153 interfaces (see: Section 3.2.3). The Relay provisions MNPs to 1154 networks on the INET/EUN interfaces (i.e., the same as a Client would 1155 do) and advertises the MSP(s) for the OMNI link over the INET/EUN 1156 interfaces. The Relay further provides an attachment point of the 1157 OMNI link to a non-MNP-based global topology. 1159 3.4.2. AERO Proxy Behavior 1161 When a Proxy enables an OMNI interface, it assigns an LLA/ULA and 1162 configures permanent neighbor cache entries the same as for Servers. 1163 The Proxy also configures secured tunnels with one or more 1164 neighboring Bridges and maintains per-Client neighbor cache entries 1165 based on control message exchanges. Importantly Proxys are often 1166 configured to act as Servers, and vice-versa. 1168 3.4.3. AERO Client Behavior 1170 When a Client enables an OMNI interface, it sends RS messages with ND 1171 parameters over its underlying interfaces to a Server, which returns 1172 an RA message with corresponding parameters. (The RS/RA messages may 1173 pass through a Proxy in the case of a Client's ANET interface, or 1174 through one or more NATs in the case of a Client's INET interface.) 1176 3.4.4. AERO Bridge Behavior 1178 AERO Bridges configure an OMNI interface and assign the ULA Subnet 1179 Router Anycast address for each OMNI link segment they connect to. 1180 Bridges configure secured tunnels with Servers, Proxys and other 1181 Bridges; they also configure LLAs/ULAs and permanent neighbor cache 1182 entries the same as Servers. Bridges engage in a BGP routing 1183 protocol session with a subset of the Servers and other Bridges on 1184 the spanning tree (see: Section 3.2.3). 1186 3.5. OMNI Interface Neighbor Cache Maintenance 1188 Each OMNI interface maintains a conceptual neighbor cache that 1189 includes an entry for each neighbor it communicates with on the OMNI 1190 link per [RFC4861]. OMNI interface neighbor cache entries are said 1191 to be one of "permanent", "symmetric", "asymmetric" or "proxy". 1193 Permanent neighbor cache entries are created through explicit 1194 administrative action; they have no timeout values and remain in 1195 place until explicitly deleted. AERO Bridges maintain permanent 1196 neighbor cache entries for their associated Proxys/Servers (and vice- 1197 versa). Each entry maintains the mapping between the neighbor's 1198 network-layer LLA and corresponding INET address. 1200 Symmetric neighbor cache entries are created and maintained through 1201 RS/RA exchanges as specified in Section 3.12, and remain in place for 1202 durations bounded by prefix lifetimes. AERO Servers maintain 1203 symmetric neighbor cache entries for each of their associated 1204 Clients, and AERO Clients maintain symmetric neighbor cache entries 1205 for each of their associated Servers. 1207 Asymmetric neighbor cache entries are created or updated based on 1208 route optimization messaging as specified in Section 3.14, and are 1209 garbage-collected when keepalive timers expire. AERO ROSs maintain 1210 asymmetric neighbor cache entries for active targets with lifetimes 1211 based on ND messaging constants. Asymmetric neighbor cache entries 1212 are unidirectional since only the ROS (and not the ROR) creates an 1213 entry. 1215 Proxy neighbor cache entries are created and maintained by AERO 1216 Proxys when they process Client/Server ND exchanges, and remain in 1217 place for durations bounded by ND and prefix lifetimes. AERO Proxys 1218 maintain proxy neighbor cache entries for each of their associated 1219 Clients. Proxy neighbor cache entries track the Client state and the 1220 address of the Client's associated Server(s). 1222 To the list of neighbor cache entry states in Section 7.3.2 of 1223 [RFC4861], Proxy and Server OMNI interfaces add an additional state 1224 DEPARTED that applies to symmetric and proxy neighbor cache entries 1225 for Clients that have recently departed. The interface sets a 1226 "DepartTime" variable for the neighbor cache entry to "DEPART_TIME" 1227 seconds. DepartTime is decremented unless a new ND message causes 1228 the state to return to REACHABLE. While a neighbor cache entry is in 1229 the DEPARTED state, packets destined to the target Client are 1230 forwarded to the Client's new location instead of being dropped. 1231 When DepartTime decrements to 0, the neighbor cache entry is deleted. 1232 It is RECOMMENDED that DEPART_TIME be set to the default constant 1233 value REACHABLE_TIME plus 10 seconds (40 seconds by default) to allow 1234 a window for packets in flight to be delivered while stale route 1235 optimization state may be present. 1237 When an ROR receives an authentic NS message used for route 1238 optimization, it searches for a symmetric neighbor cache entry for 1239 the target Client. The ROR then returns a solicited NA message 1240 without creating a neighbor cache entry for the ROS, but creates or 1241 updates a target Client "Report List" entry for the ROS and sets a 1242 "ReportTime" variable for the entry to REPORT_TIME seconds. The ROR 1243 resets ReportTime when it receives a new authentic NS message, and 1244 otherwise decrements ReportTime while no authentic NS messages have 1245 been received. It is RECOMMENDED that REPORT_TIME be set to the 1246 default constant value REACHABLE_TIME plus 10 seconds (40 seconds by 1247 default) to allow a window for route optimization to converge before 1248 ReportTime decrements below REACHABLE_TIME. 1250 When the ROS receives a solicited NA message response to its NS 1251 message used for route optimization, it creates or updates an 1252 asymmetric neighbor cache entry for the target network-layer and 1253 link-layer addresses. The ROS then (re)sets ReachableTime for the 1254 neighbor cache entry to REACHABLE_TIME seconds and uses this value to 1255 determine whether packets can be forwarded directly to the target, 1256 i.e., instead of via a default route. The ROS otherwise decrements 1257 ReachableTime while no further solicited NA messages arrive. It is 1258 RECOMMENDED that REACHABLE_TIME be set to the default constant value 1259 30 seconds as specified in [RFC4861]. 1261 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1262 of NS keepalives sent when a correspondent may have gone unreachable, 1263 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1264 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1265 to limit the number of unsolicited NAs that can be sent based on a 1266 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1267 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1268 same as specified in [RFC4861]. 1270 Different values for DEPART_TIME, REPORT_TIME, REACHABLE_TIME, 1271 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1272 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1273 different values are chosen, all nodes on the link MUST consistently 1274 configure the same values. Most importantly, DEPART_TIME and 1275 REPORT_TIME SHOULD be set to a value that is sufficiently longer than 1276 REACHABLE_TIME to avoid packet loss due to stale route optimization 1277 state. 1279 3.5.1. OMNI Neighbor Interface Attributes 1281 OMNI interface IPv6 ND messages include OMNI options 1282 [I-D.templin-6man-omni-interface] with Interface Attributes that 1283 provide Link-Layer Address and QoS Preference information for the 1284 neighbor's underlying interfaces. This information is stored in the 1285 neighbor cache and provides the basis for the forwarding algorithm 1286 specified in Section 3.10. The information is cumulative and 1287 reflects the union of the OMNI information from the most recent ND 1288 messages received from the neighbor; it is therefore not required 1289 that each ND message contain all neighbor information. 1291 The OMNI option Interface Attributes for each underlying interface 1292 includes a two-part "Link-Layer Address" consisting of a simple IP 1293 encapsulation address determined by the FMT and L2ADDR fields and an 1294 OAL ULA determined by the SRT and LHS fields. If the neighbor is 1295 located in the local OMNI link segment (and, if any necessary NAT 1296 state has been established) forwarding via simple IP encapsulation 1297 can be used; otherwise, OAL encapsulation must be used. Underlying 1298 interfaces are further selected based on their associated preference 1299 values "high", "medium", "low" or "disabled". 1301 Note: the OMNI option is distinct from any Source/Target Link-Layer 1302 Address Options (S/TLLAOs) that may appear in an ND message according 1303 to the appropriate IPv6 over specific link layer specification (e.g., 1304 [RFC2464]). If both an OMNI option and S/TLLAO appear, the former 1305 pertains to encapsulation addresses while the latter pertains to the 1306 native L2 address format of the underlying media. 1308 3.5.2. OMNI Neighbor Advertisement Message Flags 1310 As discussed in Section 4.4 of [RFC4861] NA messages include three 1311 flag bits R, S and O. OMNI interface NA messages treat the flags as 1312 follows: 1314 o R: The R ("Router") flag is set to 1 in the NA messages sent by 1315 all AERO/OMNI node types. Simple hosts that would set R to 0 do 1316 not occur on the OMNI link itself, but may occur on the downstream 1317 links of Clients and Relays. 1319 o S: The S ("Solicited") flag is set exactly as specified in 1320 Section 4.4. of [RFC4861], i.e., it is set to 1 for Solicited NAs 1321 and set to 0 for Unsolicited NAs (both unicast and multicast). 1323 o O: The O ("Override") flag is set to 0 for solicited proxy NAs and 1324 set to 1 for all other solicited and unsolicited NAs. For further 1325 study is whether solicited NAs for anycast targets apply for OMNI 1326 links. Since OMNI LLAs must be uniquely assigned to Clients to 1327 support correct ND protocol operation, however, no role is 1328 currently seen for assigning the same OMNI LLA to multiple 1329 Clients. 1331 3.6. OMNI Interface Encapsulation and Re-encapsulation 1333 The OMNI Adaptation Layer (OAL) inserts mid-layer IPv6 headers known 1334 as the OAL/ORH headers when necessary as discussed in the following 1335 sections. After either inserting or omitting the OAL/ORH headers, 1336 the OMNI interface also inserts or omits an outer encapsulation 1337 header as discussed below. 1339 OMNI interfaces avoid outer encapsulation over Direct underlying 1340 interfaces and ANET underlying interfaces for which the Client and 1341 Proxy are connected to the same underlying link. Otherwise, OMNI 1342 interfaces encapsulate packets according to whether they are entering 1343 the OMNI interface from the network layer or if they are being re- 1344 admitted into the same OMNI link they arrived on. This latter form 1345 of encapsulation is known as "re-encapsulation". 1347 For packets entering the OMNI interface from the network layer, the 1348 OMNI interface copies the "TTL/Hop Limit", "Type of Service/Traffic 1349 Class" [RFC2983], "Flow Label"[RFC6438] (for IPv6) and "Congestion 1350 Experienced" [RFC3168] values in the inner packet's IP header into 1351 the corresponding fields in the OAL and outer encapsulation 1352 header(s). 1354 For packets undergoing re-encapsulation, the OMNI interface instead 1355 copies these values from the original encapsulation header into the 1356 new encapsulation header, i.e., the values are transferred between 1357 encapsulation headers and *not* copied from the encapsulated packet's 1358 network-layer header. (Note especially that by copying the TTL/Hop 1359 Limit between encapsulation headers the value will eventually 1360 decrement to 0 if there is a (temporary) routing loop.) 1362 OMNI interfaces configured over ANET underlying interfaces which 1363 employ a different IP protocol version (and/or when the Client and 1364 Proxy may be separated by multiple ANET IP hops) use IP-in-IP 1365 encapsulation so that the inner packet can traverse the ANET without 1366 decrementing the TTL/Hop-Limit. IPv6 underlying ANET interfaces use 1367 [RFC2473] encapsulation, while IPv4 interfaces use the appropriate 1368 encapsulation per one of [RFC5214][RFC2003]. 1370 OMNI interfaces configured over INET underlying interfaces 1371 encapsulate packets in INET headers according to the next hop 1372 determined in the forwarding algorithm in Section 3.10. If the next 1373 hop is reached via a secured tunnel, the OMNI interface uses an 1374 encapsulation format specific to the secured tunnel type (see: 1375 Section 6). If the next hop is reached via an unsecured INET 1376 interface, the OMNI interface instead uses UDP/IP encapsulation per 1377 [RFC4380] and as extended in [RFC6081]. 1379 When UDP/IP encapsulation is used, the OMNI interface next sets the 1380 UDP source port to a constant value that it will use in each 1381 successive packet it sends, and sets the UDP length field to the 1382 length of the encapsulated packet plus 8 bytes for the UDP header 1383 itself plus the length of any included extension headers or trailers. 1384 The encapsulated packet may be either IPv6 or IPv4, as distinguished 1385 by the version number found in the first four bits. 1387 For UDP/IP-encapsulated packets sent to a Server, Relay or Bridge, 1388 the OMNI interface sets the UDP destination port to 8060, i.e., the 1389 IANA-registered port number for AERO. For packets sent to a Client, 1390 the OMNI interface sets the UDP destination port to the port value 1391 stored in the neighbor cache entry for this Client. The OMNI 1392 interface finally includes/omits the UDP checksum according to 1393 [RFC6935][RFC6936]. 1395 3.7. OMNI Interface Decapsulation 1397 OMNI interfaces decapsulate packets destined either to the AERO node 1398 itself or to a destination reached via an interface other than the 1399 OMNI interface the packet was received on. When the encapsulated 1400 packet arrives in multiple OAL fragments, the OMNI interface 1401 reassembles as discussed in Section 3.9. Further decapsulation steps 1402 are performed according to the appropriate encapsulation format 1403 specification. 1405 3.8. OMNI Interface Data Origin Authentication 1407 AERO nodes employ simple data origin authentication procedures. In 1408 particular: 1410 o AERO Bridges, Servers and Proxys accept encapsulated data packets 1411 and control messages received from the (secured) spanning tree. 1413 o AERO Proxys and Clients accept packets that originate from within 1414 the same secured ANET. 1416 o AERO Clients and Relays accept packets from downstream network 1417 correspondents based on ingress filtering. 1419 o AERO Clients, Relays and Servers verify the outer UDP/IP 1420 encapsulation addresses according to [RFC4380]. 1422 AERO nodes silently drop any packets that do not satisfy the above 1423 data origin authentication procedures. Further security 1424 considerations are discussed in Section 6. 1426 3.9. OMNI Adaptation Layer and OMNI Interface MTU 1428 The OMNI interface observes the link nature of tunnels, including the 1429 Maximum Transmission Unit (MTU), Maximum Reassembly Unit (MRU) and 1430 the role of fragmentation and reassembly [I-D.ietf-intarea-tunnels]. 1431 The OMNI interface employs an OMNI Adaptation Layer (OAL) for 1432 accommodating multiple underlying links with diverse MTUs. The 1433 functions of the OAL and the OMNI interface MTU/MRU are specified in 1434 Section 5 of [I-D.templin-6man-omni-interface], with MTU/MRU both set 1435 to the constant value 9180 bytes. 1437 When the network layer presents an IP packet to the OMNI interface, 1438 the OAL encapsulates the packet in OAL/ORH headers. The OAL then 1439 fragments the encapsulated packet if necessary such that the OAL/ORH 1440 headers appear in each fragment while the original IP packet header 1441 appears only in the first fragment. The OAL then transmits each OAL/ 1442 ORH packet/fragment over an underlying linterface connected to either 1443 a physical link such as Ethernet, WiFi and the like or a virtual link 1444 such as an Internet or higher-layer tunnel (see the definition of 1445 link in [RFC8200]). 1447 Note: Although the ORH may be removed by a Bridge on the path (see: 1448 Section 3.10.4), this does not interfere with the destination's 1449 ability to reassemble in the event that the packet was fragmented. 1450 This is due to the fact that the ORH is not included in the 1451 fragmentabe part; therefore, its removal does not invalidate the 1452 offset values in any fragment headers. 1454 3.10. OMNI Interface Forwarding Algorithm 1456 IP packets enter a node's OMNI interface either from the network 1457 layer (i.e., from a local application or the IP forwarding system) or 1458 from the link layer (i.e., from an OMNI interface neighbor). All 1459 packets entering a node's OMNI interface first undergo data origin 1460 authentication as discussed in Section 3.8. Packets that satisfy 1461 data origin authentication are processed further, while all others 1462 are dropped silently. The OMNI interface OAL wraps accepted packets 1463 in OAL/ORH headers if necessary as discussed above. 1465 Packets that enter the OMNI interface from the network layer are 1466 forwarded to an OMNI interface neighbor. Packets that enter the OMNI 1467 interface from the link layer are either re-admitted into the OMNI 1468 link or forwarded to the network layer where they are subject to 1469 either local delivery or IP forwarding. In all cases, the OMNI 1470 interface itself MUST NOT decrement the network layer TTL/Hop-count 1471 since its forwarding actions occur below the network layer. 1473 OMNI interfaces may have multiple underlying interfaces and/or 1474 neighbor cache entries for neighbors with multiple underlying 1475 interfaces (see Section 3.3). The OMNI interface uses interface 1476 attributes and/or traffic classifiers (e.g., DSCP value, port number, 1477 etc.) to select an outgoing underlying interface for each packet 1478 based on the node's own QoS preferences, and also to select a 1479 destination link-layer address based on the neighbor's underlying 1480 interface with the highest preference. AERO implementations SHOULD 1481 allow for QoS preference values to be modified at runtime through 1482 network management. 1484 If multiple outgoing interfaces and/or neighbor interfaces have a 1485 preference of "high", the AERO node replicates the packet and sends 1486 one copy via each of the (outgoing / neighbor) interface pairs; 1487 otherwise, the node sends a single copy of the packet via an 1488 interface with the highest preference. AERO nodes keep track of 1489 which underlying interfaces are currently "reachable" or 1490 "unreachable", and only use "reachable" interfaces for forwarding 1491 purposes. 1493 The following sections discuss the OMNI interface forwarding 1494 algorithms for Clients, Proxys, Servers and Bridges. In the 1495 following discussion, a packet's destination address is said to 1496 "match" if it is the same as a cached address, or if it is covered by 1497 a cached prefix (which may be encoded in an LLA). 1499 3.10.1. Client Forwarding Algorithm 1501 When an IP packet enters a Client's OMNI interface from the network 1502 layer the Client searches for an asymmetric neighbor cache entry that 1503 matches the destination. If there is a match, the Client uses one or 1504 more "reachable" neighbor interfaces in the entry for packet 1505 forwarding. If there is no asymmetric neighbor cache entry, the 1506 Client instead forwards the packet toward a Server (the packet is 1507 intercepted by a Proxy if there is a Proxy on the path). The Client 1508 encapsulates the packet in OAL/ORH headers if necessary and fragments 1509 according to MTU requirements (see: Section 3.9). 1511 When an IP packet enters a Client's OMNI interface from the link- 1512 layer, if the destination matches one of the Client's MNPs or link- 1513 local addresses the Client reassembles and decapsulates as necessary 1514 and delivers the inner packet to the network layer. Otherwise, the 1515 Client drops the packet and MAY return a network-layer ICMP 1516 Destination Unreachable message subject to rate limiting (see: 1517 Section 3.11). 1519 3.10.2. Proxy Forwarding Algorithm 1521 For control messages originating from or destined to a Client, the 1522 Proxy intercepts the message and updates its proxy neighbor cache 1523 entry for the Client. The Proxy then forwards a (proxyed) copy of 1524 the control message. (For example, the Proxy forwards a proxied 1525 version of a Client's NS/RS message to the target neighbor, and 1526 forwards a proxied version of the NA/RA reply to the Client.) 1528 When the Proxy receives a data packet from a Client within the ANET, 1529 the Proxy reassembles and re-fragments if necessary then searches for 1530 an asymmetric neighbor cache entry that matches the destination and 1531 forwards as follows: 1533 o if the destination matches an asymmetric neighbor cache entry, the 1534 Proxy uses one or more "reachable" neighbor interfaces in the 1535 entry for packet forwarding using OAL/ORH encapsulation if 1536 necessary according to the cached link-layer address information. 1537 If the neighbor interface is in the same OMNI link segment, the 1538 Proxy forwards the packet directly to the neighbor; otherwise, it 1539 forwards the packet to a Bridge. 1541 o else, the Proxy uses OAL/ORH encapsulation and forwards the packet 1542 to a Bridge while using the ULA corresponding to the packet's 1543 destination as the destination address. 1545 When the Proxy receives an encapsulated data packet from an INET 1546 neighbor or from a secured tunnel from a Bridge, it accepts the 1547 packet only if data origin authentication succeeds and if there is a 1548 proxy neighbor cache entry that matches the inner destination. Next, 1549 the Proxy reassembles the packet (if necessary) and continues 1550 processing. If the reassembly is complete and the neighbor cache 1551 state is REACHABLE, the Proxy then returns a PTB if necessary (see: 1552 Section 3.9) then either drops or forwards the packet to the Client 1553 while performing OAL/ORH encapsulation and re-fragmentation if 1554 necessary. If the neighbor cache entry state is DEPARTED, the Proxy 1555 instead changes the destination address to the address of the new 1556 Server and forwards it to a Bridge while performing OAL/ORH re- 1557 fragmentation if necessary. 1559 3.10.3. Server/Relay Forwarding Algorithm 1561 For control messages destined to a target Client's LLA that are 1562 received from a secured tunnel, the Server intercepts the message and 1563 sends a Proxyed response on behalf of the Client. (For example, the 1564 Server sends a Proxyed NA message reply in response to an NS message 1565 directed to one of its associated Clients.) If the Client's neighbor 1566 cache entry is in the DEPARTED state, however, the Server instead 1567 forwards the packet to the Client's new Server as discussed in 1568 Section 3.16. 1570 When the Server receives an encapsulated data packet from an INET 1571 neighbor or from a secured tunnel, it accepts the packet only if data 1572 origin authentication succeeds. The Server then continues processing 1573 as follows: 1575 o if the network layer destination matches a symmetric neighbor 1576 cache entry in the REACHABLE state the Server prepares the packet 1577 for forwarding to the destination Client. The Server first 1578 reassembles (if necessary) and forwards the packet (while re- 1579 fragmenting if necessary) as specified in Section 3.9. 1581 o else, if the destination matches a symmetric neighbor cache entry 1582 in the DEPARETED state the Server re-encapsulates the packet and 1583 forwards it using the ULA of the Client's new Server as the 1584 destination. 1586 o else, if the destination matches an asymmetric neighbor cache 1587 entry, the Server uses one or more "reachable" neighbor interfaces 1588 in the entry for packet forwarding via the local INET if the 1589 neighbor is in the same OMNI link segment or using OAL/ORH 1590 encapsulation if necessary with the final destination set to the 1591 neighbor's ULA otherwise. 1593 o else, if the destination matches a non-MNP route in the IP 1594 forwarding table or an LLA assigned to the Server's OMNI 1595 interface, the Server reassembles if necessary, decapsulates the 1596 packet and releases it to the network layer for local delivery or 1597 IP forwarding. 1599 o else, the Server drops the packet. 1601 When the Server's OMNI interface receives a data packet from the 1602 network layer or from a VPNed or Direct Client, it performs OAL/ORH 1603 encapsulation and fragmentation if necessary, then processes the 1604 packet according to the network-layer destination address as follows: 1606 o if the destination matches a symmetric or asymmetric neighbor 1607 cache entry the Server processes the packet as above. 1609 o else, the Server encapsulates the packet in OLA/ORH headers and 1610 forwards it to a Bridge using its own ULA as the source and the 1611 ULA corresponding to the destination as the destination. 1613 3.10.4. Bridge Forwarding Algorithm 1615 Bridges forward OAL/ORH-encapsulated packets over secured tunnels the 1616 same as any IP router. When the Bridge receives an OAL/ORH- 1617 encapsulated packet via a secured tunnel, it removes the outer INET 1618 header and searches for a forwarding table entry that matches the 1619 destination address. The Bridge then processes the packet as 1620 follows: 1622 o if the destination matches its ULA Subnet Router Anycast address, 1623 the Bridge determines if the next header is an ORH. If so, the 1624 Bridge removes the ORH from the packet while decrementing the OAL 1625 header Payload Length field. If the ORH includes a Destination 1626 Suffix the Bridge also writes the ULA formed from the Destination 1627 Suffix into the OAL header destination address; otherwise, it 1628 writes the ULA formed from the SRT/LHS values. Next, the Bridge 1629 examines the FMT to determine if the target is behind a NAT. If 1630 no NAT is indicated, the Bridge forwards the packet directly to 1631 the L2ADDR using link-layer (UDP/IP) encapsulation. If a NAT is 1632 indicated, the Bridge MAY perform NAT traversal procedures by 1633 sending bubbles per [RFC4380]. The Bridge then either applies 1634 AERO route optimization after NAT traversal procedures have 1635 converged, or simply forwards the packet directly to the Server 1636 indicated by SRT/LHS. 1638 o if the destination matches one of the Bridge's own addresses, the 1639 Bridge submits the packet for local delivery. 1641 o else, if the destination matches a forwarding table entry the 1642 Bridge forwards the packet via a secured tunnel to the next hop. 1643 If the destination matches an MSP without matching an MNP, 1644 however, the Bridge instead drops the packet and returns an ICMP 1645 Destination Unreachable message subject to rate limiting (see: 1646 Section 3.11). 1648 o else, the Bridge drops the packet and returns an ICMP Destination 1649 Unreachable as above. 1651 As for any IP router, the Bridge decrements the TTL/Hop Limit when it 1652 forwards the packet. Therefore, when an OAL header is present only 1653 the Hop Limit in the OAL header is decremented and not the TTL/Hop 1654 Limit in the inner packet header. Bridges do not insert OAL/ORH 1655 headers themselves; instead, they act as IPv6 routers and forward 1656 packets based on the destination address found in the headers of 1657 packets they receive. 1659 3.11. OMNI Interface Error Handling 1661 When an AERO node admits a packet into the OMNI interface, it may 1662 receive link-layer or network-layer error indications. 1664 A link-layer error indication is an ICMP error message generated by a 1665 router in the INET on the path to the neighbor or by the neighbor 1666 itself. The message includes an IP header with the address of the 1667 node that generated the error as the source address and with the 1668 link-layer address of the AERO node as the destination address. 1670 The IP header is followed by an ICMP header that includes an error 1671 Type, Code and Checksum. Valid type values include "Destination 1672 Unreachable", "Time Exceeded" and "Parameter Problem" 1673 [RFC0792][RFC4443]. (OMNI interfaces ignore all link-layer IPv4 1674 "Fragmentation Needed" and IPv6 "Packet Too Big" messages since they 1675 only emit packets that are guaranteed to be no larger than the IP 1676 minimum link MTU as discussed in Section 3.9.) 1677 The ICMP header is followed by the leading portion of the packet that 1678 generated the error, also known as the "packet-in-error". For 1679 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1680 much of invoking packet as possible without the ICMPv6 packet 1681 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1682 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1683 "Internet Header + 64 bits of Original Data Datagram", however 1684 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1685 ICMP datagram SHOULD contain as much of the original datagram as 1686 possible without the length of the ICMP datagram exceeding 576 1687 bytes". 1689 The link-layer error message format is shown in Figure 5 (where, "L2" 1690 and "L3" refer to link-layer and network-layer, respectively): 1692 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1693 ~ ~ 1694 | L2 IP Header of | 1695 | error message | 1696 ~ ~ 1697 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1698 | L2 ICMP Header | 1699 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1700 ~ ~ P 1701 | IP and other encapsulation | a 1702 | headers of original L3 packet | c 1703 ~ ~ k 1704 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1705 ~ ~ t 1706 | IP header of | 1707 | original L3 packet | i 1708 ~ ~ n 1709 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1710 ~ ~ e 1711 | Upper layer headers and | r 1712 | leading portion of body | r 1713 | of the original L3 packet | o 1714 ~ ~ r 1715 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1717 Figure 5: OMNI Interface Link-Layer Error Message Format 1719 The AERO node rules for processing these link-layer error messages 1720 are as follows: 1722 o When an AERO node receives a link-layer Parameter Problem message, 1723 it processes the message the same as described as for ordinary 1724 ICMP errors in the normative references [RFC0792][RFC4443]. 1726 o When an AERO node receives persistent link-layer Time Exceeded 1727 messages, the IP ID field may be wrapping before earlier fragments 1728 awaiting reassembly have been processed. In that case, the node 1729 should begin including integrity checks and/or institute rate 1730 limits for subsequent packets. 1732 o When an AERO node receives persistent link-layer Destination 1733 Unreachable messages in response to encapsulated packets that it 1734 sends to one of its asymmetric neighbor correspondents, the node 1735 should process the message as an indication that a path may be 1736 failing, and optionally initiate NUD over that path. If it 1737 receives Destination Unreachable messages over multiple paths, the 1738 node should allow future packets destined to the correspondent to 1739 flow through a default route and re-initiate route optimization. 1741 o When an AERO Client receives persistent link-layer Destination 1742 Unreachable messages in response to encapsulated packets that it 1743 sends to one of its symmetric neighbor Servers, the Client should 1744 mark the path as unusable and use another path. If it receives 1745 Destination Unreachable messages on many or all paths, the Client 1746 should associate with a new Server and release its association 1747 with the old Server as specified in Section 3.16.5. 1749 o When an AERO Server receives persistent link-layer Destination 1750 Unreachable messages in response to encapsulated packets that it 1751 sends to one of its symmetric neighbor Clients, the Server should 1752 mark the underlying path as unusable and use another underlying 1753 path. 1755 o When an AERO Server or Proxy receives link-layer Destination 1756 Unreachable messages in response to an encapsulated packet that it 1757 sends to one of its permanent neighbors, it treats the messages as 1758 an indication that the path to the neighbor may be failing. 1759 However, the dynamic routing protocol should soon reconverge and 1760 correct the temporary outage. 1762 When an AERO Bridge receives a packet for which the network-layer 1763 destination address is covered by an MSP, if there is no more- 1764 specific routing information for the destination the Bridge drops the 1765 packet and returns a network-layer Destination Unreachable message 1766 subject to rate limiting. The Bridge writes the network-layer source 1767 address of the original packet as the destination address and uses 1768 one of its non link-local addresses as the source address of the 1769 message. 1771 When an AERO node receives an encapsulated packet for which the 1772 reassembly buffer it too small, it drops the packet and returns a 1773 network-layer Packet Too Big (PTB) message. The node first writes 1774 the MRU value into the PTB message MTU field, writes the network- 1775 layer source address of the original packet as the destination 1776 address and writes one of its non link-local addresses as the source 1777 address. 1779 3.12. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1781 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1782 coordinated as discussed in the following Sections. 1784 3.12.1. AERO Service Model 1786 Each AERO Server on the OMNI link is configured to facilitate Client 1787 prefix delegation/registration requests. Each Server is provisioned 1788 with a database of MNP-to-Client ID mappings for all Clients enrolled 1789 in the AERO service, as well as any information necessary to 1790 authenticate each Client. The Client database is maintained by a 1791 central administrative authority for the OMNI link and securely 1792 distributed to all Servers, e.g., via the Lightweight Directory 1793 Access Protocol (LDAP) [RFC4511], via static configuration, etc. 1794 Clients receive the same service regardless of the Servers they 1795 select. 1797 AERO Clients and Servers use ND messages to maintain neighbor cache 1798 entries. AERO Servers configure their OMNI interfaces as advertising 1799 NBMA interfaces, and therefore send unicast RA messages with a short 1800 Router Lifetime value (e.g., ReachableTime seconds) in response to a 1801 Client's RS message. Thereafter, Clients send additional RS messages 1802 to keep Server state alive. 1804 AERO Clients and Servers include prefix delegation and/or 1805 registration parameters in RS/RA messages (see 1806 [I-D.templin-6man-omni-interface]). The ND messages are exchanged 1807 between Client and Server according to the prefix management schedule 1808 required by the service. If the Client knows its MNP in advance, it 1809 can employ prefix registration by including its LLA as the source 1810 address of an RS message and with an OMNI option with valid prefix 1811 registration information for the MNP. If the Server (and Proxy) 1812 accept the Client's MNP assertion, they inject the prefix into the 1813 routing system and establish the necessary neighbor cache state. 1815 The following sections specify the Client and Server behavior. 1817 3.12.2. AERO Client Behavior 1819 AERO Clients discover the addresses of Servers in a similar manner as 1820 described in [RFC5214]. Discovery methods include static 1821 configuration (e.g., from a flat-file map of Server addresses and 1822 locations), or through an automated means such as Domain Name System 1823 (DNS) name resolution [RFC1035]. Alternatively, the Client can 1824 discover Server addresses through a layer 2 data link login exchange, 1825 or through a unicast RA response to a multicast/anycast RS as 1826 described below. In the absence of other information, the Client can 1827 resolve the DNS Fully-Qualified Domain Name (FQDN) 1828 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 1829 text string and "[domainname]" is a DNS suffix for the OMNI link 1830 (e.g., "example.com"). 1832 To associate with a Server, the Client acts as a requesting router to 1833 request MNPs. The Client prepares an RS message with prefix 1834 management parameters and includes a Nonce and Timestamp option if 1835 the Client needs to correlate RA replies. If the Client already 1836 knows the Server's LLA, it includes the LLA as the network-layer 1837 destination address; otherwise, it includes (link-local) All-Routers 1838 multicast as the network-layer destination. If the Client already 1839 knows its own LLA, it uses the LLA as the network-layer source 1840 address; otherwise, it uses an OMNI Temporary LLA as the network- 1841 layer source address and includes a DHCP Unique Identifier (DUID) 1842 sub-option in the OMNI option (see: 1843 [I-D.templin-6man-omni-interface]). 1845 The Client next includes an OMNI option in the RS message to register 1846 its link-layer information with the Server. The Client sets the OMNI 1847 option prefix registration information according to the MNP, and 1848 includes Interface Attributes corresponding to the underlying 1849 interface over which the Client will send the RS message. The Client 1850 MAY include additional Interface Attributes specific to other 1851 underlying interfaces. 1853 The Client then sends the RS message (either directly via Direct 1854 interfaces, via a VPN for VPNed interfaces, via a Proxy for ANET 1855 interfaces or via INET encapsulation for INET interfaces) and waits 1856 for an RA message reply (see Section 3.12.3). The Client retries up 1857 to MAX_RTR_SOLICITATIONS times until an RA is received. If the 1858 Client receives no RAs, or if it receives an RA with Router Lifetime 1859 set to 0, the Client SHOULD abandon this Server and try another 1860 Server. Otherwise, the Client processes the prefix information found 1861 in the RA message. 1863 Next, the Client creates a symmetric neighbor cache entry with the 1864 Server's LLA as the network-layer address and the Server's 1865 encapsulation and/or link-layer addresses as the link-layer address. 1866 The Client records the RA Router Lifetime field value in the neighbor 1867 cache entry as the time for which the Server has committed to 1868 maintaining the MNP in the routing system via this underlying 1869 interface, and caches the other RA configuration information 1870 including Cur Hop Limit, M and O flags, Reachable Time and Retrans 1871 Timer. The Client then autoconfigures LLAs for each of the delegated 1872 MNPs and assigns them to the OMNI interface. The Client also caches 1873 any MSPs included in Route Information Options (RIOs) [RFC4191] as 1874 MSPs to associate with the OMNI link, and assigns the MTU value in 1875 the MTU option to the underlying interface. 1877 The Client then registers additional underlying interfaces with the 1878 Server by sending RS messages via each additional interface. The RS 1879 messages include the same parameters as for the initial RS/RA 1880 exchange, but with destination address set to the Server's LLA. 1882 Following autoconfiguration, the Client sub-delegates the MNPs to its 1883 attached EUNs and/or the Client's own internal virtual interfaces as 1884 described in [I-D.templin-v6ops-pdhost] to support the Client's 1885 downstream attached "Internet of Things (IoT)". The Client 1886 subsequently sends additional RS messages over each underlying 1887 interface before the Router Lifetime received for that interface 1888 expires. 1890 After the Client registers its underlying interfaces, it may wish to 1891 change one or more registrations, e.g., if an interface changes 1892 address or becomes unavailable, if QoS preferences change, etc. To 1893 do so, the Client prepares an RS message to send over any available 1894 underlying interface. The RS includes an OMNI option with prefix 1895 registration information specific to its MNP, with Interface 1896 Attributes specific to the selected underlying interface, and with 1897 any additional Interface Attributes specific to other underlying 1898 interfaces. When the Client receives the Server's RA response, it 1899 has assurance that the Server has been updated with the new 1900 information. 1902 If the Client wishes to discontinue use of a Server it issues an RS 1903 message over any underlying interface with an OMNI option with a 1904 prefix release indication. When the Server processes the message, it 1905 releases the MNP, sets the symmetric neighbor cache entry state for 1906 the Client to DEPARTED and returns an RA reply with Router Lifetime 1907 set to 0. After a short delay (e.g., 2 seconds), the Server 1908 withdraws the MNP from the routing system. 1910 3.12.3. AERO Server Behavior 1912 AERO Servers act as IP routers and support a prefix delegation/ 1913 registration service for Clients. Servers arrange to add their LLAs 1914 to a static map of Server addresses for the link and/or the DNS 1915 resource records for the FQDN "linkupnetworks.[domainname]" before 1916 entering service. Server addresses should be geographically and/or 1917 topologically referenced, and made available for discovery by Clients 1918 on the OMNI link. 1920 When a Server receives a prospective Client's RS message on its OMNI 1921 interface, it SHOULD return an immediate RA reply with Router 1922 Lifetime set to 0 if it is currently too busy or otherwise unable to 1923 service the Client. Otherwise, the Server authenticates the RS 1924 message and processes the prefix delegation/registration parameters. 1925 The Server first determines the correct MNPs to provide to the Client 1926 by searching the Client database. When the Server returns the MNPs, 1927 it also creates a forwarding table entry for the ULA corresponding to 1928 each MNP so that the MNPs are propagated into the routing system 1929 (see: Section 3.2.3). For IPv6, the Server creates an IPv6 1930 forwarding table entry for each MNP. For IPv4, the Server creates an 1931 IPv6 forwarding table entry with the IPv4-compatibility ULA prefix 1932 corresponding to the IPv4 address. 1934 The Server next creates a symmetric neighbor cache entry for the 1935 Client using the base LLA as the network-layer address and with 1936 lifetime set to no more than the smallest prefix lifetime. Next, the 1937 Server updates the neighbor cache entry by recording the information 1938 in each Interface Attributes sub-option in the RS OMNI option. The 1939 Server also records the actual OAL/INET addresses in the neighbor 1940 cache entry. 1942 Next, the Server prepares an RA message using its LLA as the network- 1943 layer source address and the network-layer source address of the RS 1944 message as the network-layer destination address. The Server sets 1945 the Router Lifetime to the time for which it will maintain both this 1946 underlying interface individually and the symmetric neighbor cache 1947 entry as a whole. The Server also sets Cur Hop Limit, M and O flags, 1948 Reachable Time and Retrans Timer to values appropriate for the OMNI 1949 link. The Server includes the MNPs, any other prefix management 1950 parameters and an OMNI option with no Interface Attributes. The 1951 Server then includes one or more RIOs that encode the MSPs for the 1952 OMNI link, plus an MTU option (see Section 3.9). The Server finally 1953 forwards the message to the Client using OAL/INET, INET, or NULL 1954 encapsulation as necessary. 1956 After the initial RS/RA exchange, the Server maintains a 1957 ReachableTime timer for each of the Client's underlying interfaces 1958 individually (and for the Client's symmetric neighbor cache entry 1959 collectively) set to expire after ReachableTime seconds. If the 1960 Client (or Proxy) issues additional RS messages, the Server sends an 1961 RA response and resets ReachableTime. If the Server receives an ND 1962 message with a prefix release indication it sets the Client's 1963 symmetric neighbor cache entry to the DEPARTED state and withdraws 1964 the MNP from the routing system after a short delay (e.g., 2 1965 seconds). If ReachableTime expires before a new RS is received on an 1966 individual underlying interface, the Server marks the interface as 1967 DOWN. If ReachableTime expires before any new RS is received on any 1968 individual underlying interface, the Server sets the symmetric 1969 neighbor cache entry state to STALE and sets a 10 second timer. If 1970 the Server has not received a new RS or ND message with a prefix 1971 release indication before the 10 second timer expires, it deletes the 1972 neighbor cache entry and withdraws the MNP from the routing system. 1974 The Server processes any ND messages pertaining to the Client and 1975 returns an NA/RA reply in response to solicitations. The Server may 1976 also issue unsolicited RA messages, e.g., with reconfigure parameters 1977 to cause the Client to renegotiate its prefix delegation/ 1978 registrations, with Router Lifetime set to 0 if it can no longer 1979 service this Client, etc. Finally, If the symmetric neighbor cache 1980 entry is in the DEPARTED state, the Server deletes the entry after 1981 DepartTime expires. 1983 Note: Clients SHOULD notify former Servers of their departures, but 1984 Servers are responsible for expiring neighbor cache entries and 1985 withdrawing routes even if no departure notification is received 1986 (e.g., if the Client leaves the network unexpectedly). Servers 1987 SHOULD therefore set Router Lifetime to ReachableTime seconds in 1988 solicited RA messages to minimize persistent stale cache information 1989 in the absence of Client departure notifications. A short Router 1990 Lifetime also ensures that proactive Client/Server RS/RA messaging 1991 will keep any NAT state alive (see above). 1993 Note: All Servers on an OMNI link MUST advertise consistent values in 1994 the RA Cur Hop Limit, M and O flags, Reachable Time and Retrans Timer 1995 fields the same as for any link, since unpredictable behavior could 1996 result if different Servers on the same link advertised different 1997 values. 1999 3.12.3.1. DHCPv6-Based Prefix Registration 2001 When a Client is not pre-provisioned with an OMNI LLA containing a 2002 MNP, it will need for the Server to select one or more MNPs on its 2003 behalf and set up the correct state in the AERO routing service. (A 2004 Client with a pre-provisioned MNP may also request the Server to 2005 select additional MNPs.) The DHCPv6 service [RFC8415] is used to 2006 support this requirement. 2008 When a Client needs to have the Server select MNPs, it sends an RS 2009 message with an OMNI option that includes a DHCPv6 message suboption 2010 with DHCPv6 Prefix Delegation (DHCPv6-PD) parameters. When the 2011 Server receives the RS message, it extracts the DHCPv6-PD message 2012 from the OMNI option. 2014 The Server then acts as a "Proxy DHCPv6 Client" in a message exchange 2015 with the locally-resident DHCPv6 server, which delegates MNPs and 2016 returns a DHCPv6-PD Reply message. (If the Server wishes to defer 2017 creation of MN state until the DHCPv6-PD Reply is received, it can 2018 instead act as a Lightweight DHCPv6 Relay Agent per [RFC6221] by 2019 encapsulating the DHCPv6-PD message in a Relay-forward/reply exchange 2020 with Relay Message and Interface ID options.) 2022 When the Server receives the DHCPv6-PD Reply, it adds a route to the 2023 routing system and creates an OMNI MN LLA based on the delegated MNP. 2024 The Server then sends an RA back to the Client with the (newly- 2025 created) OMNI MN LLA as the destination address and with the 2026 DHCPv6-PD Reply message coded in the OMNI option. When the Client 2027 receives the RA, it creates a default route, assigns the Subnet 2028 Router Anycast address and sets its OMNI LLA based on the delegated 2029 MNP. 2031 3.13. The AERO Proxy 2033 Clients may connect to protected-spectrum ANETs that employ physical 2034 and/or link-layer security services to facilitate communications to 2035 Servers in outside INETs. In that case, the ANET can employ an AERO 2036 Proxy. The Proxy is located at the ANET/INET border and listens for 2037 RS messages originating from or RA messages destined to ANET Clients. 2038 The Proxy acts on these control messages as follows: 2040 o when the Proxy receives an RS message from a new ANET Client, it 2041 first authenticates the message then examines the network-layer 2042 destination address. If the destination address is a Server's 2043 LLA, the Proxy proceeds to the next step. Otherwise, if the 2044 destination is (link-local) All-Routers multicast, the Proxy 2045 selects a "nearby" Server that is likely to be a good candidate to 2046 serve the Client and replaces the destination address with the 2047 Server's LLA. Next, the Proxy creates a proxy neighbor cache 2048 entry and caches the Client and Server link-layer addresses along 2049 with the OMNI option information and any other identifying 2050 information including Transaction IDs, Client Identifiers, Nonce 2051 values, etc. The Proxy finally encapsulates the (proxyed) RS 2052 message in an OAL header with source set to the Proxy's ULA and 2053 destination set to the Server's ULA. The Proxy also includes an 2054 OMNI header with an Interface Attributes option that includes its 2055 own INET address plus a unique Port Number for this Client, then 2056 forwards the message into the OMNI link spanning tree. 2058 o when the Server receives the RS, it authenticates the message then 2059 creates or updates a symmetric neighbor cache entry for the Client 2060 with the Proxy's ULA, INET address and Port Number as the link- 2061 layer address information. The Server then sends an RA message 2062 back to the Proxy via the spanning tree. 2064 o when the Proxy receives the RA, it authenticates the message and 2065 matches it with the proxy neighbor cache entry created by the RS. 2066 The Proxy then caches the prefix information as a mapping from the 2067 Client's MNPs to the Client's link-layer address, caches the 2068 Server's advertised Router Lifetime and sets the neighbor cache 2069 entry state to REACHABLE. The Proxy then optionally rewrites the 2070 Router Lifetime and forwards the (proxyed) message to the Client. 2071 The Proxy finally includes an MTU option (if necessary) with an 2072 MTU to use for the underlying ANET interface. 2074 After the initial RS/RA exchange, the Proxy forwards any Client data 2075 packets for which there is no matching asymmetric neighbor cache 2076 entry to a Bridge using OAL encapsulation with its own ULA as the 2077 source and the ULA corresponding to the Client as the destination. 2078 The Proxy instead forwards any Client data destined to an asymmetric 2079 neighbor cache target directly to the target according to the OAL/ 2080 link-layer information - the process of establishing asymmetric 2081 neighbor cache entries is specified in Section 3.14. 2083 While the Client is still attached to the ANET, the Proxy sends NS, 2084 RS and/or unsolicited NA messages to update the Server's symmetric 2085 neighbor cache entries on behalf of the Client and/or to convey QoS 2086 updates. This allows for higher-frequency Proxy-initiated RS/RA 2087 messaging over well-connected INET infrastructure supplemented by 2088 lower-frequency Client-initiated RS/RA messaging over constrained 2089 ANET data links. 2091 If the Server ceases to send solicited advertisements, the Proxy 2092 sends unsolicited RAs on the ANET interface with destination set to 2093 (link-local) All-Nodes multicast and with Router Lifetime set to zero 2094 to inform Clients that the Server has failed. Although the Proxy 2095 engages in ND exchanges on behalf of the Client, the Client can also 2096 send ND messages on its own behalf, e.g., if it is in a better 2097 position than the Proxy to convey QoS changes, etc. For this reason, 2098 the Proxy marks any Client-originated solicitation messages (e.g. by 2099 inserting a Nonce option) so that it can return the solicited 2100 advertisement to the Client instead of processing it locally. 2102 If the Client becomes unreachable, the Proxy sets the neighbor cache 2103 entry state to DEPARTED and retains the entry for DepartTime seconds. 2104 While the state is DEPARTED, the Proxy forwards any packets destined 2105 to the Client to a Bridge via OAL encapsulation with the Client's 2106 current Server as the destination. The Bridge in turn forwards the 2107 packets to the Client's current Server. When DepartTime expires, the 2108 Proxy deletes the neighbor cache entry and discards any further 2109 packets destined to this (now forgotten) Client. 2111 In some ANETs that employ a Proxy, the Client's MNP can be injected 2112 into the ANET routing system. In that case, the Client can send data 2113 messages without encapsulation so that the ANET routing system 2114 transports the unencapsulated packets to the Proxy. This can be very 2115 beneficial, e.g., if the Client connects to the ANET via low-end data 2116 links such as some aviation wireless links. 2118 If the first-hop ANET access router is on the same underlying link 2119 and recognizes the AERO/OMNI protocol, the Client can avoid 2120 encapsulation for both its control and data messages. When the 2121 Client connects to the link, it can send an unencapsulated RS message 2122 with source address set to its LLA and with destination address set 2123 to the LLA of the Client's selected Server or to (link-local) All- 2124 Routers multicast. The Client includes an OMNI option formatted as 2125 specified in [I-D.templin-6man-omni-interface]. 2127 The Client then sends the unencapsulated RS message, which will be 2128 intercepted by the AERO-Aware access router. The access router then 2129 encapsulates the RS message in an ANET header with its own address as 2130 the source address and the address of a Proxy as the destination 2131 address. The access router further remembers the address of the 2132 Proxy so that it can encapsulate future data packets from the Client 2133 via the same Proxy. If the access router needs to change to a new 2134 Proxy, it simply sends another RS message toward the Server via the 2135 new Proxy on behalf of the Client. 2137 In some cases, the access router and Proxy may be one and the same 2138 node. In that case, the node would be located on the same physical 2139 link as the Client, but its message exchanges with the Server would 2140 need to pass through a security gateway at the ANET/INET border. The 2141 method for deploying access routers and Proxys (i.e. as a single node 2142 or multiple nodes) is an ANET-local administrative consideration. 2144 3.13.1. Combined Proxy/Servers 2146 Clients may need to connect directly to Servers via INET, Direct and 2147 VPNed interfaces (i.e., non-ANET interfaces). If the Client's 2148 underlying interfaces all connect via the same INET partition, then 2149 it can connect to a single controlling Server via all interfaces. 2151 If some Client interfaces connect via different INET partitions, 2152 however, the Client still selects a set of controlling Servers and 2153 sends RS messages via their directly-connected Servers while using 2154 the LLA of the controlling Server as the destination. 2156 When a Server receives an RS with destination set to the LLA of a 2157 controlling Server, it acts as a Proxy to forward the message to the 2158 controlling Server while forwarding the corresponding RA reply to the 2159 Client. 2161 3.13.2. Detecting and Responding to Server Failures 2163 In environments where fast recovery from Server failure is required, 2164 Proxys SHOULD use proactive Neighbor Unreachability Detection (NUD) 2165 to track Server reachability in a similar fashion as for 2166 Bidirectional Forwarding Detection (BFD) [RFC5880]. Proxys can then 2167 quickly detect and react to failures so that cached information is 2168 re-established through alternate paths. The NUD control messaging is 2169 carried only over well-connected ground domain networks (i.e., and 2170 not low-end aeronautical radio links) and can therefore be tuned for 2171 rapid response. 2173 Proxys perform proactive NUD with Servers for which there are 2174 currently active ANET Clients by sending continuous NS messages in 2175 rapid succession, e.g., one message per second. The Proxy sends the 2176 NS message via the spanning tree with the Proxy's LLA as the source 2177 and the LLA of the Server as the destination. When the Proxy is also 2178 sending RS messages to the Server on behalf of ANET Clients, the 2179 resulting RA responses can be considered as equivalent hints of 2180 forward progress. This means that the Proxy need not also send a 2181 periodic NS if it has already sent an RS within the same period. If 2182 the Server fails (i.e., if the Proxy ceases to receive 2183 advertisements), the Proxy can quickly inform Clients by sending 2184 multicast RA messages on the ANET interface. 2186 The Proxy sends RA messages on the ANET interface with source address 2187 set to the Server's address, destination address set to (link-local) 2188 All-Nodes multicast, and Router Lifetime set to 0. The Proxy SHOULD 2189 send MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small 2190 delays [RFC4861]. Any Clients on the ANET that had been using the 2191 failed Server will receive the RA messages and associate with a new 2192 Server. 2194 3.13.3. Point-to-Multipoint Server Coordination 2196 In environments where Client messaging over ANETs is bandwidth- 2197 limited and/or expensive, Clients can enlist the services of the 2198 Proxy to coordinate with multiple Servers in a single RS/RA message 2199 exchange. The Client can send a single RS message to (link-local) 2200 All-Routers multicast that includes the ID's of multiple Servers in 2201 MS-Register sub-options of the OMNI option. 2203 When the Proxy receives the RS and processes the OMNI option, it 2204 sends a separate RS to each MS-Register Server ID. When the Proxy 2205 receives an RA, it can optionally return an immediate "singleton" RA 2206 to the Client or record the Server's ID for inclusion in a pending 2207 "aggregate" RA message. The Proxy can then return aggregate RA 2208 messages to the Client including multiple Server IDs in order to 2209 conserve bandwidth. Each RA includes a proper subset of the Server 2210 IDs from the original RS message, and the Proxy must ensure that the 2211 message contents of each RA are consistent with the information 2212 received from the (aggregated) Servers. 2214 Clients can thereafter employ efficient point-to-multipoint Server 2215 coordination under the assistance of the Proxy to reduce the number 2216 of messages sent over the ANET while enlisting the support of 2217 multiple Servers for fault tolerance. Clients can further include 2218 MS-Release sub-options in IPv6 ND messages to request the Proxy to 2219 release from former Servers via the procedures discussed in 2220 Section 3.16.5. 2222 The OMNI interface specification [I-D.templin-6man-omni-interface] 2223 provides further discussion of the Client/Proxy RS/RA messaging 2224 involved in point-to-multipoint coordination. 2226 3.14. AERO Route Optimization / Address Resolution 2228 While data packets are flowing between a source and target node, 2229 route optimization SHOULD be used. Route optimization is initiated 2230 by the first eligible Route Optimization Source (ROS) closest to the 2231 source as follows: 2233 o For Clients on VPNed and Direct interfaces, the Server is the ROS. 2235 o For Clients on ANET interfaces, the Proxy is the ROS. 2237 o For Clients on INET interfaces, the Client itself is the ROS. 2239 o For correspondent nodes on INET/EUN interfaces serviced by a 2240 Relay, the Relay is the ROS. 2242 The route optimization procedure is conducted between the ROS and the 2243 target Server/Relay acting as a Route Optimization Responder (ROR) in 2244 the same manner as for IPv6 ND Address Resolution and using the same 2245 NS/NA messaging. The target may either be a MNP Client serviced by a 2246 Server, or a non-MNP correspondent reachable via a Relay. 2248 The procedures are specified in the following sections. 2250 3.14.1. Route Optimization Initiation 2252 While data packets are flowing from the source node toward a target 2253 node, the ROS performs address resolution by sending an NS message 2254 for Address Resolution (NS(AR)) to receive a solicited NA message 2255 from the ROR. When the ROS sends an NS(AR), it includes: 2257 o the LLA of the ROS as the source address. 2259 o the data packet's destination as the Target Address. 2261 o the Solicited-Node multicast address [RFC4291] formed from the 2262 lower 24 bits of the data packet's destination as the destination 2263 address, e.g., for 2001:db8:1:2::10:2000 the NS destination 2264 address is ff02:0:0:0:0:1:ff10:2000. 2266 The NS(AR) message includes an OMNI option with no Interface 2267 Attributes, such that the target will not create a neighbor cache 2268 entry. The Prefix Length in the OMNI option is set to the Prefix 2269 Length associated with the ROS's LLA. 2271 The ROS then encapsulates the NS(AR) message in an OAL header with 2272 source set to its own ULA and destination set to the ULA 2273 corresponding to the target, then sends the message into the spanning 2274 tree without decrementing the network-layer TTL/Hop Limit field. 2275 (When the ROS is a Client, it instead securely sends the NS(AR) to 2276 one of its current Servers by including an Authentication option per 2277 [RFC4380]. The Server then forwards the message into the spanning 2278 tree on behalf of the Client, while setting the IPv6 source address 2279 and the OAL source address to the LLA and ULA of the Client, 2280 respectively.) 2282 3.14.2. Relaying the NS 2284 When the Bridge receives the NS(AR) message from the ROS, it discards 2285 the INET header and determines that the ROR is the next hop by 2286 consulting its standard IPv6 forwarding table for the OAL header 2287 destination address. The Bridge then forwards the message toward the 2288 ROR via the spanning tree the same as for any IPv6 router. The 2289 final-hop Bridge in the spanning tree will deliver the message via a 2290 secured tunnel to the ROR. 2292 3.14.3. Processing the NS and Sending the NA 2294 When the ROR receives the NS(AR) message, it examines the Target 2295 Address to determine whether it has a neighbor cache entry and/or 2296 route that matches the target. If there is no match, the ROR drops 2297 the message. Otherwise, the ROR continues processing as follows: 2299 o if the target belongs to an MNP Client neighbor in the DEPARTED 2300 state the ROR changes the NS(AR) message OAL destination address 2301 to the ULA of the Client's new Server, forwards the message into 2302 the spanning tree and returns from processing. 2304 o If the target belongs to an MNP Client neighbor in the REACHABLE 2305 state, the ROR instead adds the AERO source address to the target 2306 Client's Report List with time set to ReportTime. 2308 o If the target belongs to a non-MNP route, the ROR continues 2309 processing without adding an entry to the Report List. 2311 The ROR then prepares a solicited NA message to send back to the ROS 2312 but does not create a neighbor cache entry. The ROR sets the NA 2313 source address to the LLA corresponding to the target, sets the 2314 Target Address to the target of the solicitation, and sets the 2315 destination address to the source of the solicitation. The ROR then 2316 includes an OMNI option with Prefix Length set to the length 2317 associated with the LLA. 2319 If the target is an MNP Client, the ROR next includes Interface 2320 Attributes in the OMNI option for each of the target Client's 2321 underlying interfaces with current information for each interface and 2322 with the S/T-ifIndex field in the OMNI header set to 0 to indicate 2323 that the message originated from the ROR and not the Client. 2325 For each Interface Attributes sub-option, the ROR sets the L2ADDR 2326 according to its own INET address for VPNed or Direct interfaces, to 2327 the INET address of the Proxy or to the Client's INET address for 2328 INET interfaces. The ROR then includes the lower 32 bits of its own 2329 ULA (or the ULA of the Proxy) as the LHS, encodes the ULA prefix 2330 length code in the SRT field and sets the FMT code accordingly as 2331 specified in Section 3.3. 2333 The ROR then sets the NA message R flag to 1 (as a router), S flag to 2334 1 (as a response to a solicitation), and O flag to 0 (as a proxy). 2335 The ROR finally encapsulates the NA message in an OAL header with 2336 source set to its own ULA and destination set to the source ULA of 2337 the NS(AR) message, then forwards the message into the spanning tree 2338 without decrementing the network-layer TTL/Hop Limit field. 2340 3.14.4. Relaying the NA 2342 When the Bridge receives the NA message from the ROR, it discards the 2343 INET header and determines that the ROS is the next hop by consulting 2344 its standard IPv6 forwarding table for the OAL header destination 2345 address. The Bridge then forwards the OAL-encapsulated NA message 2346 toward the ROS the same as for any IPv6 router. The final-hop Bridge 2347 in the spanning tree will deliver the message via a secured tunnel to 2348 the ROS. 2350 3.14.5. Processing the NA 2352 When the ROS receives the solicited NA message, it processes the 2353 message the same as for standard IPv6 Address Resolution [RFC4861]. 2354 In the process, it caches the source ULA then creates an asymmetric 2355 neighbor cache entry for the target and caches all information found 2356 in the OMNI option. The ROS finally sets the asymmetric neighbor 2357 cache entry lifetime to ReachableTime seconds. (When the ROS is a 2358 Client, the solicited NA message will first be delivered via the 2359 spanning tree to one of its current Servers, which then securely 2360 forwards the message to the Client by including an Authentication 2361 option per [RFC4380]. 2363 3.14.6. Route Optimization Maintenance 2365 Following route optimization, the ROS forwards future data packets 2366 destined to the target via the addresses found in the cached link- 2367 layer information. The route optimization is shared by all sources 2368 that send packets to the target via the ROS, i.e., and not just the 2369 source on behalf of which the route optimization was initiated. 2371 While new data packets destined to the target are flowing through the 2372 ROS, it sends additional NS(AR) messages to the ROR before 2373 ReachableTime expires to receive a fresh solicited NA message the 2374 same as described in the previous sections (route optimization 2375 refreshment strategies are an implementation matter, with a non- 2376 normative example given in Appendix A.1). The ROS uses the cached 2377 ULA of the ROR as the NS(AR) OAL destination address (i.e., instead 2378 of using the ULA corresponding to the target as was the case for the 2379 initial NS(AR)), and sends up to MAX_MULTICAST_SOLICIT NS(AR) 2380 messages separated by 1 second until an NA is received. If no NA is 2381 received, the ROS assumes that the current ROR has become unreachable 2382 and deletes the target neighbor cache entry. Subsequent data packets 2383 will trigger a new route optimization with an NS with OAL destination 2384 address set to the ULA corresponding to the target per Section 3.14.1 2385 to discover a new ROR while initial data packets travel over a 2386 suboptimal route. 2388 If an NA is received, the ROS then updates the asymmetric neighbor 2389 cache entry to refresh ReachableTime, while (for MNP destinations) 2390 the ROR adds or updates the ROS address to the target's Report List 2391 and with time set to ReportTime. While no data packets are flowing, 2392 the ROS instead allows ReachableTime for the asymmetric neighbor 2393 cache entry to expire. When ReachableTime expires, the ROS deletes 2394 the asymmetric neighbor cache entry. Any future data packets flowing 2395 through the ROS will again trigger a new route optimization. 2397 The ROS may also receive unsolicited NA messages from the ROR at any 2398 time (see: Section 3.16). If there is an asymmetric neighbor cache 2399 entry for the target, the ROS updates the link-layer information but 2400 does not update ReachableTime since the receipt of an unsolicited NA 2401 does not confirm that any forward paths are working. If there is no 2402 asymmetric neighbor cache entry, the ROS simply discards the 2403 unsolicited NA. 2405 In this arrangement, the ROS holds an asymmetric neighbor cache entry 2406 for the target via the ROR, but the ROR does not hold an asymmetric 2407 neighbor cache entry for the ROS. The route optimization neighbor 2408 relationship is therefore asymmetric and unidirectional. If the 2409 target node also has packets to send back to the source node, then a 2410 separate route optimization procedure is performed in the reverse 2411 direction. But, there is no requirement that the forward and reverse 2412 paths be symmetric. 2414 3.15. Neighbor Unreachability Detection (NUD) 2416 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2417 [RFC4861] either reactively in response to persistent link-layer 2418 errors (see Section 3.11) or proactively to confirm reachability. 2419 The NUD algorithm is based on periodic control message exchanges. 2420 The algorithm may further be seeded by ND hints of forward progress, 2421 but care must be taken to avoid inferring reachability based on 2422 spoofed information. For example, authentic IPv6 ND message 2423 exchanges may be considered as acceptable hints of forward progress, 2424 while spurious data packets should not be. 2426 AERO Servers, Proxys and Relays can use (OAL-encapsulated) standard 2427 NS/NA NUD exchanges sent over the OMNI link spanning tree to securely 2428 test reachability without risk of DoS attacks from nodes pretending 2429 to be a neighbor (these NS/NA(NUD) messages use the unicast LLAs and 2430 ULAs of the two parties involved in the NUD test the same as for 2431 standard IPv6 ND). Proxys can further perform NUD to securely verify 2432 Server reachability on behalf of their proxyed Clients. However, a 2433 means for an ROS to test the unsecured forward directions of target 2434 route optimized paths is also necessary. 2436 When an ROR directs an ROS to a neighbor with one or more target 2437 link-layer addresses, the ROS can proactively test each such 2438 unsecured route optimized path by sending "loopback" NS(NUD) 2439 messages. While testing the paths, the ROS can optionally continue 2440 to send packets via the spanning tree, maintain a small queue of 2441 packets until target reachability is confirmed, or (optimistically) 2442 allow packets to flow via the route optimized paths. 2444 When the ROS sends a loopback NS(NUD) message, it uses its own LLA as 2445 both the IPv6 source and destination address, and the MNP Subnet- 2446 Router anycast address as the Target Address. The ROS includes a 2447 Nonce and Timestamp option, then encapsulates the message in OAL/INET 2448 headers with its own ULA as the source and the ULA of the route 2449 optimization target as the destination. The ROS then forwards the 2450 message to the target (either directly to the L2ADDR of the target if 2451 the target is in the same OMNI link segment, or via a Bridge if the 2452 target is in a different OMNI link segment). 2454 When the route optimization target receives the NS(NUD) message, it 2455 notices that the IPv6 destination address is the same as the source 2456 address. It then reverses the OAL header source and destination 2457 addresses and returns the message to the ROS (either directly or via 2458 the spanning tree). The route optimization target does not decrement 2459 the NS(NUD) message IPv6 Hop-Limit in the process, since the message 2460 has not exited the OMNI link. 2462 When the ROS receives the NS(NUD) message, it can determine from the 2463 Nonce, Timestamp and Target Address that the message originated from 2464 itself and that it transited the forward path. The ROS need not 2465 prepare an NA response, since the destination of the response would 2466 be itself and testing the route optimization path again would be 2467 redundant. 2469 The ROS marks route optimization target paths that pass these NUD 2470 tests as "reachable", and those that do not as "unreachable". These 2471 markings inform the OMNI interface forwarding algorithm specified in 2472 Section 3.10. 2474 Note that to avoid a DoS vector nodes MUST NOT return loopback 2475 NS(NUD) messages received from an unsecured link-layer source via the 2476 (secured) spanning tree. 2478 3.16. Mobility Management and Quality of Service (QoS) 2480 AERO is a Distributed Mobility Management (DMM) service. Each Server 2481 is responsible for only a subset of the Clients on the OMNI link, as 2482 opposed to a Centralized Mobility Management (CMM) service where 2483 there is a single network mobility collective entity for all Clients. 2484 Clients coordinate with their associated Servers via RS/RA exchanges 2485 to maintain the DMM profile, and the AERO routing system tracks all 2486 current Client/Server peering relationships. 2488 Servers provide default routing and mobility/multilink services for 2489 their dependent Clients. Clients are responsible for maintaining 2490 neighbor relationships with their Servers through periodic RS/RA 2491 exchanges, which also serves to confirm neighbor reachability. When 2492 a Client's underlying interface address and/or QoS information 2493 changes, the Client is responsible for updating the Server with this 2494 new information. Note that when there is a Proxy in the path, the 2495 Proxy can also perform some RS/RA exchanges on the Client's behalf. 2497 Mobility management messaging is based on the transmission and 2498 reception of unsolicited Neighbor Advertisement (uNA) messages. Each 2499 uNA message sets the IPv6 destination address to (link-local) All- 2500 Nodes multicast to convey a general update of Interface Attributes to 2501 (possibly) multiple recipients, or to a specific unicast LLA to 2502 announce a departure event to a specific recipient. Implementations 2503 must therefore examine the destination address to determine the 2504 nature of the mobility event (i.e., update vs departure). 2506 Mobility management considerations are specified in the following 2507 sections. 2509 3.16.1. Mobility Update Messaging 2511 Servers accommodate Client mobility, multilink and/or QoS change 2512 events by sending unsolicited NA (uNA) messages to each ROS in the 2513 target Client's Report List. When a Server sends a uNA message, it 2514 sets the IPv6 source address to the Client's LLA, sets the 2515 destination address to (link-local) All-Nodes multicast and sets the 2516 Target Address to the Client's Subnet-Router anycast address. The 2517 Server also includes an OMNI option with Prefix Length set to the 2518 length associated with the Client's LLA, with Interface Attributes 2519 for the target Client's underlying interfaces and with the OMNI 2520 header S/T-ifIndex set to 0. The Server then sets the NA R flag to 2521 1, the S flag to 0 and the O flag to 1, then encapsulates the message 2522 in an OAL header with source set to its own ULA and destination set 2523 to the ULA of the ROS and sends the message into the spanning tree. 2525 As discussed in Section 7.2.6 of [RFC4861], the transmission and 2526 reception of uNA messages is unreliable but provides a useful 2527 optimization. In well-connected Internetworks with robust data links 2528 uNA messages will be delivered with high probability, but in any case 2529 the Server can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT uNAs 2530 to each ROS to increase the likelihood that at least one will be 2531 received. 2533 When the ROS receives a uNA message prepared as above, it ignores the 2534 message if there is no existing neighbor cache entry for the Client. 2535 Otherwise, it uses the included OMNI option information to update the 2536 neighbor cache entry, but does not reset ReachableTime since the 2537 receipt of an unsolicited NA message from the target Server does not 2538 provide confirmation that any forward paths to the target Client are 2539 working. 2541 If uNA messages are lost, the ROS may be left with stale address and/ 2542 or QoS information for the Client for up to ReachableTime seconds. 2543 During this time, the ROS can continue sending packets according to 2544 its stale neighbor cache information. When ReachableTime is close to 2545 expiring, the ROS will re-initiate route optimization and receive 2546 fresh link-layer address information. 2548 In addition to sending uNA messages to the current set of ROSs for 2549 the Client, the Server also sends uNAs to the ULA associated with the 2550 link-layer address for any underlying interface for which the link- 2551 layer address has changed. These uNA messages update an old Proxy/ 2552 Server that cannot easily detect (e.g., without active probing) when 2553 a formerly-active Client has departed. When the Server sends the 2554 uNA, it sets the IPv6 source address to the Client's LLA, sets the 2555 destination address to the old Proxy/Server's LLA, and sets the 2556 Target Address to the Client's Subnet-Router anycast address. The 2557 Server also includes an OMNI option with Prefix Length set to the 2558 length associated with the Client's LLA, with Interface Attributes 2559 for the changed underlying interface, and with the OMNI header S/ 2560 T-ifIndex set to 0. The Server then sets the NA R flag to 1, the S 2561 flag to 0 and the O flag to 1, then encapsulates the message in an 2562 OAL header with source set to its own ULA and destination set to the 2563 ULA of the old Proxy/Server and sends the message into the spanning 2564 tree. 2566 3.16.2. Announcing Link-Layer Address and/or QoS Preference Changes 2568 When a Client needs to change its underlying interface addresses and/ 2569 or QoS preferences (e.g., due to a mobility event), either the Client 2570 or its Proxys send RS messages to the Server via the spanning tree 2571 with an OMNI option that includes Interface attributes with the new 2572 link quality and address information. 2574 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 2575 sending actual data packets in case one or more RAs are lost. If all 2576 RAs are lost, the Client SHOULD re-associate with a new Server. 2578 When the Server receives the Client's changes, it sends uNA messages 2579 to all nodes in the Report List the same as described in the previous 2580 section. 2582 3.16.3. Bringing New Links Into Service 2584 When a Client needs to bring new underlying interfaces into service 2585 (e.g., when it activates a new data link), it sends an RS message to 2586 the Server via the underlying interface with an OMNI option that 2587 includes Interface Attributes with appropriate link quality values 2588 and with link-layer address information for the new link. 2590 3.16.4. Deactivating Existing Links 2592 When a Client needs to deactivate an existing underlying interface, 2593 it sends an RS or uNA message to its Server with an OMNI option with 2594 appropriate Interface Attribute values - in particular, the link 2595 quality value 0 assures that neighbors will cease to use the link. 2597 If the Client needs to send RS/uNA messages over an underlying 2598 interface other than the one being deactivated, it MUST include 2599 Interface Attributes with appropriate link quality values for any 2600 underlying interfaces being deactivated. 2602 Note that when a Client deactivates an underlying interface, 2603 neighbors that have received the RS/uNA messages need not purge all 2604 references for the underlying interface from their neighbor cache 2605 entries. The Client may reactivate or reuse the underlying interface 2606 and/or its ifIndex at a later point in time, when it will send RS/uNA 2607 messages with fresh Interface Attributes to update any neighbors. 2609 3.16.5. Moving Between Servers 2611 The Client performs the procedures specified in Section 3.12.2 when 2612 it first associates with a new Server or renews its association with 2613 an existing Server. The Client also includes MS-Release identifiers 2614 in the RS message OMNI option per [I-D.templin-6man-omni-interface] 2615 if it wants the new Server to notify any old Servers from which the 2616 Client is departing. 2618 When the new Server receives the Client's RS message, it returns an 2619 RA as specified in Section 3.12.3 and sends up to 2620 MAX_NEIGHBOR_ADVERTIISEMENT uNA messages to any old Servers listed in 2621 OMNI option MS-Release identifiers. When the new Server sends a uNA 2622 message, it sets the IPv6 source address to the Client's LLA, sets 2623 the destination address to the old Server's LLA, and sets the Target 2624 Address to the Client's Subnet-Router anycast address. The new 2625 Server also includes an OMNI option with Prefix Length set to the 2626 length associated with the Client's LLA, with Interface Attributes 2627 for its own underlying interface, and with the OMNI header S/ 2628 T-ifIndex set to 0. The new Server then sets the NA R flag to 1, the 2629 S flag to 0 and the O flag to 1, then encapsulates the message in an 2630 OAL header with source set to its own ULA and destination set to the 2631 ULA of the old Server and sends the message into the spanning tree. 2633 When an old Server receives the uNA, it changes the Client's neighbor 2634 cache entry state to DEPARTED, sets the link-layer address of the 2635 Client to the new Server's ULA, and resets DepartTime. After a short 2636 delay (e.g., 2 seconds) the old Server withdraws the Client's MNP 2637 from the routing system. After DepartTime expires, the old Server 2638 deletes the Client's neighbor cache entry. 2640 The old Server also iteratively forwards a copy of the uNA message to 2641 each ROS in the Client's Report List by changing the OAL destination 2642 address to the ULA of the ROS while leaving all other fields of the 2643 message unmodified. When the ROS receives the uNA, it examines the 2644 Target address to determine the correct asymmetric neighbor cache 2645 entry and verifies that the IPv6 destination address matches the old 2646 Server. The ROS then caches the IPv6 source address as the new 2647 Server for the existing asymmetric neighbor cache entry and marks the 2648 entry as STALE. While in the STALE state, the ROS allows new data 2649 packets to flow according to any existing cached link-layer 2650 information and sends new NS(AR) messages using its own ULA as the 2651 OAL source and the ULA of the new Server as the OAL destination 2652 address to elicit NA messages that reset the asymmetric neighbor 2653 cache entry state to REACHABLE. If no new NA message is received for 2654 10 seconds while in the STALE state, the ROS deletes the neighbor 2655 cache entry. 2657 Clients SHOULD NOT move rapidly between Servers in order to avoid 2658 causing excessive oscillations in the AERO routing system. Examples 2659 of when a Client might wish to change to a different Server include a 2660 Server that has gone unreachable, topological movements of 2661 significant distance, movement to a new geographic region, movement 2662 to a new OMNI link segment, etc. 2664 When a Client moves to a new Server, some of the fragments of a 2665 multiple fragment packet may have already arrived at the old Server 2666 while others are en route to the new Server, however no special 2667 attention in the reassembly algorithm is necessary when re-routed 2668 fragments are simply treated as loss. 2670 3.17. Multicast 2672 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 2673 [RFC3810] proxy service for its EUNs and/or hosted applications 2674 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 2675 underlying interfaces for which group membership is required. The 2676 IGMP/MLD messages may be further forwarded by a first-hop ANET access 2677 router acting as an IGMP/MLD-snooping switch [RFC4541], then 2678 ultimately delivered to an AERO Proxy/Server acting as a Protocol 2679 Independent Multicast - Sparse-Mode (PIM-SM, or simply "PIM") 2680 Designated Router (DR) [RFC7761]. AERO Relays also act as PIM 2681 routers (i.e., the same as AERO Proxys/Servers) on behalf of nodes on 2682 INET/EUN networks. The behaviors identified in the following 2683 sections correspond to Source-Specific Multicast (SSM) and Any-Source 2684 Multicast (ASM) operational modes. 2686 3.17.1. Source-Specific Multicast (SSM) 2688 When an ROS (i.e., an AERO Proxy/Server/Relay) "X" acting as PIM 2689 router receives a Join/Prune message from a node on its downstream 2690 interfaces containing one or more ((S)ource, (G)roup) pairs, it 2691 updates its Multicast Routing Information Base (MRIB) accordingly. 2692 For each S belonging to a prefix reachable via X's non-OMNI 2693 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 2694 on those interfaces per [RFC7761]. 2696 For each S belonging to a prefix reachable via X's OMNI interface, X 2697 originates a separate copy of the Join/Prune for each (S,G) in the 2698 message using its own LLA as the source address and ALL-PIM-ROUTERS 2699 as the destination address. X then encapsulates each message in an 2700 OAL header with source address set to the ULA of X and destination 2701 address set to S then forwards the message into the spanning tree, 2702 which delivers it to AERO Server/Relay "Y" that services S. At the 2703 same time, if the message was a Join, X sends a route-optimization NS 2704 message toward each S the same as discussed in Section 3.14. The 2705 resulting NAs will return the LLA for the prefix that matches S as 2706 the network-layer source address and with an OMNI option with the ULA 2707 corresponding to any underlying interfaces that are currently 2708 servicing S. 2710 When Y processes the Join/Prune message, if S located behind any 2711 INET, Direct, or VPNed interfaces Y acts as a PIM router and updates 2712 its MRIB to list X as the next hop in the reverse path. If S is 2713 located behind any Proxys "Z"*, Y also forwards the message to each 2714 Z* over the spanning tree while continuing to use the LLA of X as the 2715 source address. Each Z* then updates its MRIB accordingly and 2716 maintains the LLA of X as the next hop in the reverse path. Since 2717 the Bridges do not examine network layer control messages, this means 2718 that the (reverse) multicast tree path is simply from each Z* (and/or 2719 Y) to X with no other multicast-aware routers in the path. If any Z* 2720 (and/or Y) is located on the same OMNI link segment as X, the 2721 multicast data traffic sent to X directly using OAL/INET 2722 encapsulation instead of via a Bridge. 2724 Following the initial Join/Prune and NS/NA messaging, X maintains an 2725 asymmetric neighbor cache entry for each S the same as if X was 2726 sending unicast data traffic to S. In particular, X performs 2727 additional NS/NA exchanges to keep the neighbor cache entry alive for 2728 up to t_periodic seconds [RFC7761]. If no new Joins are received 2729 within t_periodic seconds, X allows the neighbor cache entry to 2730 expire. Finally, if X receives any additional Join/Prune messages 2731 for (S,G) it forwards the messages to each Y and Z* in the neighbor 2732 cache entry over the spanning tree. 2734 At some later time, Client C that holds an MNP for source S may 2735 depart from a first Proxy Z1 and/or connect via a new Proxy Z2. In 2736 that case, Y sends an unsolicited NA message to X the same as 2737 specified for unicast mobility in Section 3.16. When X receives the 2738 unsolicited NA message, it updates its asymmetric neighbor cache 2739 entry for the LLA for source S and sends new Join messages to any new 2740 Proxys Z2. There is no requirement to send any Prune messages to old 2741 Proxys Z1 since source S will no longer source any multicast data 2742 traffic via Z1. Instead, the multicast state for (S,G) in Proxy Z1 2743 will soon time out since no new Joins will arrive. 2745 After some later time, C may move to a new Server Y2 and depart from 2746 old Sever Y1. In that case, Y1 sends Join messages for any of C's 2747 active (S,G) groups to Y2 while including its own LLA as the source 2748 address. This causes Y2 to include Y1 in the multicast forwarding 2749 tree during the interim time that Y1's symmetric neighbor cache entry 2750 for C is in the DEPARTED state. At the same time, Y1 sends an 2751 unsolicited NA message to X with an OMNI option with S/T-ifIndex in 2752 the header set to 0 and a release indication to cause X to release 2753 its asymmetric neighbor cache entry. X then sends a new Join message 2754 to S via the spanning tree and re-initiates route optimization the 2755 same as if it were receiving a fresh Join message from a node on a 2756 downstream link. 2758 3.17.2. Any-Source Multicast (ASM) 2760 When an ROS X acting as a PIM router receives a Join/Prune from a 2761 node on its downstream interfaces containing one or more (*,G) pairs, 2762 it updates its Multicast Routing Information Base (MRIB) accordingly. 2763 X then forwards a copy of the message to the Rendezvous Point (RP) R 2764 for each G over the spanning tree. X uses its own LLA as the source 2765 address and ALL-PIM-ROUTERS as the destination address, then 2766 encapsulates each message in an OAL header with source address set to 2767 the ULA of X and destination address set to R, then sends the message 2768 into the spanning tree. At the same time, if the message was a Join 2769 X initiates NS/NA route optimization the same as for the SSM case 2770 discussed in Section 3.17.1. 2772 For each source S that sends multicast traffic to group G via R, the 2773 Proxy/Server Z* for the Client that aggregates S encapsulates the 2774 packets in PIM Register messages and forwards them to R via the 2775 spanning tree, which may then elect to send a PIM Join to Z*. This 2776 will result in an (S,G) tree rooted at Z* with R as the next hop so 2777 that R will begin to receive two copies of the packet; one native 2778 copy from the (S, G) tree and a second copy from the pre-existing (*, 2779 G) tree that still uses PIM Register encapsulation. R can then issue 2780 a PIM Register-stop message to suppress the Register-encapsulated 2781 stream. At some later time, if C moves to a new Proxy/Server Z*, it 2782 resumes sending packets via PIM Register encapsulation via the new 2783 Z*. 2785 At the same time, as multicast listeners discover individual S's for 2786 a given G, they can initiate an (S,G) Join for each S under the same 2787 procedures discussed in Section 3.17.1. Once the (S,G) tree is 2788 established, the listeners can send (S, G) Prune messages to R so 2789 that multicast packets for group G sourced by S will only be 2790 delivered via the (S, G) tree and not from the (*, G) tree rooted at 2791 R. All mobility considerations discussed for SSM apply. 2793 3.17.3. Bi-Directional PIM (BIDIR-PIM) 2795 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 2796 approach to ASM that treats the Rendezvous Point (RP) as a Designated 2797 Forwarder (DF). Further considerations for BIDIR-PIM are out of 2798 scope. 2800 3.18. Operation over Multiple OMNI Links 2802 An AERO Client can connect to multiple OMNI links the same as for any 2803 data link service. In that case, the Client maintains a distinct 2804 OMNI interface for each link, e.g., 'omni0' for the first link, 2805 'omni1' for the second, 'omni2' for the third, etc. Each OMNI link 2806 would include its own distinct set of Bridges, Servers and Proxys, 2807 thereby providing redundancy in case of failures. 2809 Each OMNI link could utilize the same or different ANET connections. 2810 The links can be distinguished at the link-layer via the SRT prefix 2811 in a similar fashion as for Virtual Local Area Network (VLAN) tagging 2812 (e.g., IEEE 802.1Q) and/or through assignment of distinct sets of 2813 MSPs on each link. This gives rise to the opportunity for supporting 2814 multiple redundant networked paths, with each VLAN distinguished by a 2815 different SRT "color" (see: Section 3.2.5). 2817 The Client's IP layer can select the outgoing OMNI interface 2818 appropriate for a given traffic profile while (in the reverse 2819 direction) correspondent nodes must have some way of steering their 2820 packets destined to a target via the correct OMNI link. 2822 In a first alternative, if each OMNI link services different MSPs, 2823 then the Client can receive a distinct MNP from each of the links. 2824 IP routing will therefore assure that the correct Red/Green/Blue/etc. 2825 network is used for both outbound and inbound traffic. This can be 2826 accomplished using existing technologies and approaches, and without 2827 requiring any special supporting code in correspondent nodes or 2828 Bridges. 2830 In a second alternative, if each OMNI link services the same MSP(s) 2831 then each link could assign a distinct "OMNI link Anycast" address 2832 that is configured by all Bridges on the link. Correspondent nodes 2833 can then perform Segment Routing to select the correct SRT, which 2834 will then direct the packet over multiple hops to the target. 2836 3.19. DNS Considerations 2838 AERO Client MNs and INET correspondent nodes consult the Domain Name 2839 System (DNS) the same as for any Internetworking node. When 2840 correspondent nodes and Client MNs use different IP protocol versions 2841 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 2842 A records for IPv4 address mappings to MNs which must then be 2843 populated in Relay NAT64 mapping caches. In that way, an IPv4 2844 correspondent node can send packets to the IPv4 address mapping of 2845 the target MN, and the Relay will translate the IPv4 header and 2846 destination address into an IPv6 header and IPv6 destination address 2847 of the MN. 2849 When an AERO Client registers with an AERO Server, the Server can 2850 return the address(es) of DNS servers in RDNSS options [RFC6106]. 2851 The DNS server provides the IP addresses of other MNs and 2852 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 2854 3.20. Transition Considerations 2856 OAL encapsulation ensures that dissimilar INET partitions can be 2857 joined into a single unified OMNI link, even though the partitions 2858 themselves may have differing protocol versions and/or incompatible 2859 addressing plans. However, a commonality can be achieved by 2860 incrementally distributing globally routable (i.e., native) IP 2861 prefixes to eventually reach all nodes (both mobile and fixed) in all 2862 OMNI link segments. This can be accomplished by incrementally 2863 deploying AERO Relays on each INET partition, with each Relay 2864 distributing its MNPs and/or discovering non-MNP prefixes on its INET 2865 links. 2867 This gives rise to the opportunity to eventually distribute native IP 2868 addresses to all nodes, and to present a unified OMNI link view even 2869 if the INET partitions remain in their current protocol and 2870 addressing plans. In that way, the OMNI link can serve the dual 2871 purpose of providing a mobility/multilink service and a transition 2872 service. Or, if an INET partition is transitioned to a native IP 2873 protocol version and addressing scheme that is compatible with the 2874 OMNI link MNP-based addressing scheme, the partition and OMNI link 2875 can be joined by Relays. 2877 Relays that connect INETs/EUNs with dissimilar IP protocol versions 2878 may need to employ a network address and protocol translation 2879 function such as NAT64 [RFC6146]. 2881 3.21. Detecting and Reacting to Server and Bridge Failures 2883 In environments where rapid failure recovery is required, Servers and 2884 Bridges SHOULD use Bidirectional Forwarding Detection (BFD) 2885 [RFC5880]. Nodes that use BFD can quickly detect and react to 2886 failures so that cached information is re-established through 2887 alternate nodes. BFD control messaging is carried only over well- 2888 connected ground domain networks (i.e., and not low-end radio links) 2889 and can therefore be tuned for rapid response. 2891 Servers and Bridges maintain BFD sessions in parallel with their BGP 2892 peerings. If a Server or Bridge fails, BGP peers will quickly re- 2893 establish routes through alternate paths the same as for common BGP 2894 deployments. Similarly, Proxys maintain BFD sessions with their 2895 associated Bridges even though they do not establish BGP peerings 2896 with them. 2898 Proxys SHOULD use proactive NUD for Servers for which there are 2899 currently active ANET Clients in a manner that parallels BFD, i.e., 2900 by sending unicast NS messages in rapid succession to receive 2901 solicited NA messages. When the Proxy is also sending RS messages on 2902 behalf of ANET Clients, the RS/RA messaging can be considered as 2903 equivalent hints of forward progress. This means that the Proxy need 2904 not also send a periodic NS if it has already sent an RS within the 2905 same period. If a Server fails, the Proxy will cease to receive 2906 advertisements and can quickly inform Clients of the outage by 2907 sending multicast RA messages on the ANET interface. 2909 The Proxy sends multicast RA messages with source address set to the 2910 Server's address, destination address set to (link-local) All-Nodes 2911 multicast, and Router Lifetime set to 0. The Proxy SHOULD send 2912 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2913 [RFC4861]. Any Clients on the ANET interface that have been using 2914 the (now defunct) Server will receive the RA messages and associate 2915 with a new Server. 2917 3.22. AERO Clients on the Open Internet 2919 AERO Clients that connect to the open Internet via INET interfaces 2920 can establish a VPN or direct link to securely connect to a Server in 2921 a "tethered" arrangement with all of the Client's traffic transiting 2922 the Server. Alternatively, the Client can associate with an INET 2923 Server using UDP/IP encapsulation and asymmetric securing services as 2924 discussed in the following sections. 2926 When a Client's OMNI interface enables an INET underlying interface, 2927 it first determines whether the interface is likely to be behind a 2928 NAT. For IPv4, the Client assumes it is on the open Internet if the 2929 INET address is not a special-use IPv4 address per [RFC3330]. 2930 Similarly for IPv6, the Client assumes it is on the open Internet if 2931 the INET address is not a link-local [RFC4291] or unique-local 2932 [RFC4193] IPv6 address. 2934 The Client then prepares a UDP/IP-encapsulated RS message with IPv6 2935 source address set to its LLA, with IPv6 destination set to (link- 2936 local) All-Routers multicast and with an OMNI option with underlying 2937 interface attributes. If the Client believes that it is on the open 2938 Internet, it SHOULD include Interface Attributes with the L2ADDR used 2939 for INET encapsulation (otherwise, it MAY omit L2ADDR). If the 2940 underlying address is IPv4, the Client includes the Port Number and 2941 IPv4 address written in obfuscated form [RFC4380] as discussed in 2942 Section 3.3. If the underlying interface address is IPv6, the Client 2943 instead includes the Port Number and IPv6 address in obfuscated form. 2944 The Client finally includes an Authentication option per [RFC4380] to 2945 provide message authentication, sets the UDP/IP source to its INET 2946 address and UDP port, sets the UDP/IP destination to the Server's 2947 INET address and the AERO service port number (8060), then sends the 2948 message to the Server. 2950 When the Server receives the RS, it authenticates the message and 2951 registers the Client's MNP and INET interface information according 2952 to the OMNI option parameters. If the RS message includes an L2ADDR 2953 in the OMNI option, the Server compares the encapsulation IP address 2954 and UDP port number with the (unobfuscated) values. If the values 2955 are the same, the Server caches the Client's information as "INET" 2956 addresses meaning that the Client is likely to accept direct messages 2957 without requiring NAT traversal exchanges. If the values are 2958 different (or, if the OMNI option did not include an L2ADDR) the 2959 Server instead caches the Client's information as "NAT" addresses 2960 meaning that NAT traversal exchanges may be necessary. 2962 The Server then returns an RA message with IPv6 source and 2963 destination set corresponding to the addresses in the RS, and with an 2964 Authentication option per [RFC4380]. For IPv4, the Server also 2965 includes an Origin option per [RFC4380] with the mapped and 2966 obfuscated Port Number and IPv4 address observed in the encapsulation 2967 headers. For IPv6, the Server instead includes an IPv6 Origin option 2968 per Figure 6 with the mapped and obfuscated observed Port Number and 2969 IPv6 address (note that the value 0x02 in the second octet 2970 differentiates from other [RFC4380] option types). 2972 +--------+--------+-----------------+ 2973 | 0x00 | 0x02 | Origin port # | 2974 +--------+--------+-----------------+ 2975 ~ Origin IPv6 address ~ 2976 +-----------------------------------+ 2978 Figure 6: IPv6 Origin Option 2980 When the Client receives the RA message, it compares the mapped Port 2981 Number and IP address from the Origin option with its own address. 2982 If the addresses are the same, the Client assumes the open Internet / 2983 Cone NAT principle; if the addresses are different, the Client 2984 instead assumes that further qualification procedures are necessary 2985 to detect the type of NAT and proceeds according to standard 2986 [RFC4380] procedures. 2988 After the Client has registered its INET interfaces in such RS/RA 2989 exchanges it sends periodic RS messages to receive fresh RA messages 2990 before the Router Lifetime received on each INET interface expires. 2991 The Client also maintains default routes via its Servers, i.e., the 2992 same as described in earlier sections. 2994 When the Client sends messages to target IP addresses, it also 2995 invokes route optimization per Section 3.14 using IPv6 ND address 2996 resolution messaging. The Client sends the NS(AR) message to the 2997 Server wrapped in a UDP/IP header with an Authentication option with 2998 the NS source address set to the Client's LLA and destination address 2999 set to the target solicited node multicast address. The Server 3000 authenticates the message and sends a corresponding NS(AR) message 3001 over the spanning tree the same as if it were the ROS, but with the 3002 OAL source address set to the Server's ULA and destination set to the 3003 ULA of the target. When the ROR receives the NS(AR), it adds the 3004 Server's ULA and Client's LLA to the target's Report List, and 3005 returns an NA with OMNI option information for the target. The 3006 Server then returns a UDP/IP encapsulated NA message with an 3007 Authentication option to the Client. 3009 Following route optimization for targets in the same OMNI link 3010 segment, if the target's L2ADDR is on the open INET, the Client 3011 forwards data packets directly to the target INET address. If the 3012 target is behind a NAT, the Client first establishes NAT state for 3013 the L2ADDR using the "bubble" mechanisms specified in 3014 [RFC6081][RFC4380]. The Client continues to send data packets via 3015 its Server until NAT state is populated, then begins forwarding 3016 packets via the direct path through the NAT to the target. For 3017 targets in different OMNI link segments, the Client uses OAL/ORH 3018 encapsulation and forwards data packets to the Bridge that returned 3019 the NA message. 3021 The ROR may return uNAs via the Server if the target moves, and the 3022 Server will send corresponding Authentication-protected uNAs to the 3023 Client. The Client can also send "loopback" NS(NUD) messages to test 3024 forward path reachability even though there is no security 3025 association between the Client and the target. 3027 The Client sends UDP/IP encapsulated IPv6 packets no larger than 1280 3028 bytes in one piece. In order to accommodate larger IPv6 packets (up 3029 to the OMNI interface MTU), the Client inserts an OAL header with 3030 source set to its own ULA and destination set to the ULA of the 3031 target and uses IPv6 fragmentation according to Section 3.9. The 3032 Client then encapsulates each fragment in a UDP/IP header and sends 3033 the fragments to the next hop. 3035 3.23. Time-Varying MNPs 3037 In some use cases, it is desirable, beneficial and efficient for the 3038 Client to receive a constant MNP that travels with the Client 3039 wherever it moves. For example, this would allow air traffic 3040 controllers to easily track aircraft, etc. In other cases, however 3041 (e.g., intelligent transportation systems), the MN may be willing to 3042 sacrifice a modicum of efficiency in order to have time-varying MNPs 3043 that can be changed every so often to defeat adversarial tracking. 3045 The DHCPv6 service offers a way for Clients that desire time-varying 3046 MNPs to obtain short-lived prefixes (e.g., on the order of a small 3047 number of minutes). In that case, the identity of the Client would 3048 not be bound to the MNP but rather the Client's identity would be 3049 bound to the DHCPv6 Device Unique Identifier (DUID) and used as the 3050 seed for Prefix Delegation. The Client would then be obligated to 3051 renumber its internal networks whenever its MNP (and therefore also 3052 its LLA) changes. This should not present a challenge for Clients 3053 with automated network renumbering services, however presents limits 3054 for the durations of ongoing sessions that would prefer to use a 3055 constant address. 3057 4. Implementation Status 3059 An early AERO implementation based on OpenVPN (https://openvpn.net/) 3060 was announced on the v6ops mailing list on January 10, 2018 and an 3061 initial public release of the AERO proof-of-concept source code was 3062 announced on the intarea mailing list on August 21, 2015. 3064 AERO Release-3.0.2 was tagged on October 15, 2020, and is undergoing 3065 internal testing. Additional releases expected Q42020, with first 3066 public release expected before year-end. 3068 5. IANA Considerations 3070 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 3071 AERO in the "enterprise-numbers" registry. 3073 The IANA has assigned the UDP port number "8060" for an earlier 3074 experimental version of AERO [RFC6706]. This document obsoletes 3075 [RFC6706] and claims the UDP port number "8060" for all future use. 3077 The IANA is instructed to assign a new type value TBD in the IPv6 3078 Routing Types registry. 3080 No further IANA actions are required. 3082 6. Security Considerations 3084 AERO Bridges configure secured tunnels with AERO Servers, Relays and 3085 Proxys within their local OMNI link segments. Applicable secured 3086 tunnel alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 3087 [RFC6347], WireGuard [WG], etc. The AERO Bridges of all OMNI link 3088 segments in turn configure secured tunnels for their neighboring AERO 3089 Bridges in a spanning tree topology. Therefore, control messages 3090 exchanged between any pair of OMNI link neighbors on the spanning 3091 tree are already secured. 3093 AERO Servers, Relays and Proxys targeted by a route optimization may 3094 also receive data packets directly from arbitrary nodes in INET 3095 partitions instead of via the spanning tree. For INET partitions 3096 that apply effective ingress filtering to defeat source address 3097 spoofing, the simple data origin authentication procedures in 3098 Section 3.8 can be applied. 3100 For INET partitions that require strong security in the data plane, 3101 two options for securing communications include 1) disable route 3102 optimization so that all traffic is conveyed over secured tunnels, or 3103 2) enable on-demand secure tunnel creation between INET partition 3104 neighbors. Option 1) would result in longer routes than necessary 3105 and traffic concentration on critical infrastructure elements. 3106 Option 2) could be coordinated by establishing a secured tunnel on- 3107 demand instead of performing an NS/NA exchange in the route 3108 optimization procedures. Procedures for establishing on-demand 3109 secured tunnels are out of scope. 3111 AERO Clients that connect to secured ANETs need not apply security to 3112 their ND messages, since the messages will be intercepted by a 3113 perimeter Proxy that applies security on its INET-facing interface as 3114 part of the spanning tree (see above). AERO Clients connected to the 3115 open INET can use symmetric network and/or transport layer security 3116 services such as VPNs or can by some other means establish a direct 3117 link. When a VPN or direct link may be impractical, however, an 3118 asymmetric security service such as the Authentication option 3119 specified in [RFC4380] should be applied. The Authentication option 3120 requires a unique Client identifier, which can be obtained per the 3121 Universally Unique IDentifier (UUID) [RFC4122] service and also used 3122 as a DHCP Unique Identifier (DUID) per [RFC6355]. 3124 Application endpoints SHOULD use application-layer security services 3125 such as TLS/SSL, DTLS or SSH [RFC4251] to assure the same level of 3126 protection as for critical secured Internet services. AERO Clients 3127 that require host-based VPN services SHOULD use symmetric network 3128 and/or transport layer security services such as IPsec, TLS/SSL, 3129 DTLS, etc. AERO Proxys and Servers can also provide a network-based 3130 VPN service on behalf of the Client, e.g., if the Client is located 3131 within a secured enclave and cannot establish a VPN on its own 3132 behalf. 3134 AERO Servers and Bridges present targets for traffic amplification 3135 Denial of Service (DoS) attacks. This concern is no different than 3136 for widely-deployed VPN security gateways in the Internet, where 3137 attackers could send spoofed packets to the gateways at high data 3138 rates. This can be mitigated by connecting Servers and Bridges over 3139 dedicated links with no connections to the Internet and/or when 3140 connections to the Internet are only permitted through well-managed 3141 firewalls. Traffic amplification DoS attacks can also target an AERO 3142 Client's low data rate links. This is a concern not only for Clients 3143 located on the open Internet but also for Clients in secured 3144 enclaves. AERO Servers and Proxys can institute rate limits that 3145 protect Clients from receiving packet floods that could DoS low data 3146 rate links. 3148 AERO Relays must implement ingress filtering to avoid a spoofing 3149 attack in which spurious messages with ULA addresses are injected 3150 into an OMNI link from an outside attacker. AERO Clients MUST ensure 3151 that their connectivity is not used by unauthorized nodes on their 3152 EUNs to gain access to a protected network, i.e., AERO Clients that 3153 act as routers MUST NOT provide routing services for unauthorized 3154 nodes. (This concern is no different than for ordinary hosts that 3155 receive an IP address delegation but then "share" the address with 3156 other nodes via some form of Internet connection sharing such as 3157 tethering.) 3159 The MAP list MUST be well-managed and secured from unauthorized 3160 tampering, even though the list contains only public information. 3161 The MAP list can be conveyed to the Client in a similar fashion as in 3162 [RFC5214] (e.g., through layer 2 data link login messaging, secure 3163 upload of a static file, DNS lookups, etc.). 3165 SRH authentication facilities are specified in [RFC8754]. 3167 Security considerations for accepting link-layer ICMP messages and 3168 reflected packets are discussed throughout the document. 3170 Security considerations for IPv6 fragmentation and reassembly are 3171 discussed in [I-D.templin-6man-omni-interface]. 3173 7. Acknowledgements 3175 Discussions in the IETF, aviation standards communities and private 3176 exchanges helped shape some of the concepts in this work. 3177 Individuals who contributed insights include Mikael Abrahamsson, Mark 3178 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 3179 Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian Farrel, Nick Green, 3180 Sri Gundavelli, Brian Haberman, Bernhard Haindl, Joel Halpern, Tom 3181 Herbert, Sascha Hlusiak, Lee Howard, Zdenek Jaron, Andre Kostur, 3182 Hubert Kuenig, Ted Lemon, Andy Malis, Satoru Matsushima, Tomek 3183 Mrugalski, Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal 3184 Skorepa, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, Lloyd 3185 Wood and James Woodyatt. Members of the IESG also provided valuable 3186 input during their review process that greatly improved the document. 3187 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 3188 for their shepherding guidance during the publication of the AERO 3189 first edition. 3191 This work has further been encouraged and supported by Boeing 3192 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 3193 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 3194 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 3195 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 3196 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 3197 Gene MacLean III, Rob Muszkiewicz, Sean O'Sullivan, Vijay 3198 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 3199 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 3200 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 3201 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 3202 Benson, Katie Tran and Eric Yeh are especially acknowledged for 3203 implementing the AERO functions as extensions to the public domain 3204 OpenVPN distribution. 3206 Earlier works on NBMA tunneling approaches are found in 3207 [RFC2529][RFC5214][RFC5569]. 3209 Many of the constructs presented in this second edition of AERO are 3210 based on the author's earlier works, including: 3212 o The Internet Routing Overlay Network (IRON) 3213 [RFC6179][I-D.templin-ironbis] 3215 o Virtual Enterprise Traversal (VET) 3216 [RFC5558][I-D.templin-intarea-vet] 3218 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 3219 [RFC5320][I-D.templin-intarea-seal] 3221 o AERO, First Edition [RFC6706] 3223 Note that these works cite numerous earlier efforts that are not also 3224 cited here due to space limitations. The authors of those earlier 3225 works are acknowledged for their insights. 3227 This work is aligned with the NASA Safe Autonomous Systems Operation 3228 (SASO) program under NASA contract number NNA16BD84C. 3230 This work is aligned with the FAA as per the SE2025 contract number 3231 DTFAWA-15-D-00030. 3233 This work is aligned with the Boeing Commercial Airplanes (BCA) 3234 Internet of Things (IoT) and autonomy programs. 3236 This work is aligned with the Boeing Information Technology (BIT) 3237 MobileNet program. 3239 8. References 3241 8.1. Normative References 3243 [I-D.templin-6man-omni-interface] 3244 Templin, F. and T. Whyman, "Transmission of IP Packets 3245 over Overlay Multilink Network (OMNI) Interfaces", draft- 3246 templin-6man-omni-interface-56 (work in progress), 3247 December 2020. 3249 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3250 DOI 10.17487/RFC0791, September 1981, 3251 . 3253 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 3254 RFC 792, DOI 10.17487/RFC0792, September 1981, 3255 . 3257 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3258 Requirement Levels", BCP 14, RFC 2119, 3259 DOI 10.17487/RFC2119, March 1997, 3260 . 3262 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3263 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3264 December 1998, . 3266 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3267 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3268 DOI 10.17487/RFC3971, March 2005, 3269 . 3271 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3272 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3273 . 3275 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3276 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3277 November 2005, . 3279 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3280 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3281 . 3283 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 3284 Network Address Translations (NATs)", RFC 4380, 3285 DOI 10.17487/RFC4380, February 2006, 3286 . 3288 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3289 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3290 DOI 10.17487/RFC4861, September 2007, 3291 . 3293 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3294 Address Autoconfiguration", RFC 4862, 3295 DOI 10.17487/RFC4862, September 2007, 3296 . 3298 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 3299 DOI 10.17487/RFC6081, January 2011, 3300 . 3302 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3303 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3304 May 2017, . 3306 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3307 (IPv6) Specification", STD 86, RFC 8200, 3308 DOI 10.17487/RFC8200, July 2017, 3309 . 3311 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3312 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3313 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3314 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3315 . 3317 8.2. Informative References 3319 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3320 2016. 3322 [I-D.bonica-6man-comp-rtg-hdr] 3323 Bonica, R., Kamite, Y., Alston, A., Henriques, D., and L. 3324 Jalil, "The IPv6 Compact Routing Header (CRH)", draft- 3325 bonica-6man-comp-rtg-hdr-23 (work in progress), October 3326 2020. 3328 [I-D.bonica-6man-crh-helper-opt] 3329 Li, X., Bao, C., Ruan, E., and R. Bonica, "Compressed 3330 Routing Header (CRH) Helper Option", draft-bonica-6man- 3331 crh-helper-opt-02 (work in progress), October 2020. 3333 [I-D.ietf-intarea-frag-fragile] 3334 Bonica, R., Baker, F., Huston, G., Hinden, R., Troan, O., 3335 and F. Gont, "IP Fragmentation Considered Fragile", draft- 3336 ietf-intarea-frag-fragile-17 (work in progress), September 3337 2019. 3339 [I-D.ietf-intarea-tunnels] 3340 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3341 Architecture", draft-ietf-intarea-tunnels-10 (work in 3342 progress), September 2019. 3344 [I-D.ietf-rtgwg-atn-bgp] 3345 Templin, F., Saccone, G., Dawra, G., Lindem, A., and V. 3346 Moreno, "A Simple BGP-based Mobile Routing System for the 3347 Aeronautical Telecommunications Network", draft-ietf- 3348 rtgwg-atn-bgp-06 (work in progress), June 2020. 3350 [I-D.templin-6man-dhcpv6-ndopt] 3351 Templin, F., "A Unified Stateful/Stateless Configuration 3352 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-10 3353 (work in progress), June 2020. 3355 [I-D.templin-intarea-seal] 3356 Templin, F., "The Subnetwork Encapsulation and Adaptation 3357 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 3358 progress), January 2014. 3360 [I-D.templin-intarea-vet] 3361 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 3362 templin-intarea-vet-40 (work in progress), May 2013. 3364 [I-D.templin-ironbis] 3365 Templin, F., "The Interior Routing Overlay Network 3366 (IRON)", draft-templin-ironbis-16 (work in progress), 3367 March 2014. 3369 [I-D.templin-v6ops-pdhost] 3370 Templin, F., "IPv6 Prefix Delegation and Multi-Addressing 3371 Models", draft-templin-v6ops-pdhost-26 (work in progress), 3372 June 2020. 3374 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3376 [RFC1035] Mockapetris, P., "Domain names - implementation and 3377 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3378 November 1987, . 3380 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3381 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3382 . 3384 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3385 DOI 10.17487/RFC2003, October 1996, 3386 . 3388 [RFC2004] Perkins, C., "Minimal Encapsulation within IP", RFC 2004, 3389 DOI 10.17487/RFC2004, October 1996, 3390 . 3392 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3393 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3394 . 3396 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 3397 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 3398 . 3400 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3401 Domains without Explicit Tunnels", RFC 2529, 3402 DOI 10.17487/RFC2529, March 1999, 3403 . 3405 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3406 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3407 . 3409 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3410 of Explicit Congestion Notification (ECN) to IP", 3411 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3412 . 3414 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 3415 DOI 10.17487/RFC3330, September 2002, 3416 . 3418 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3419 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3420 DOI 10.17487/RFC3810, June 2004, 3421 . 3423 [RFC4122] Leach, P., Mealling, M., and R. Salz, "A Universally 3424 Unique IDentifier (UUID) URN Namespace", RFC 4122, 3425 DOI 10.17487/RFC4122, July 2005, 3426 . 3428 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3429 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3430 January 2006, . 3432 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3433 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3434 DOI 10.17487/RFC4271, January 2006, 3435 . 3437 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3438 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3439 2006, . 3441 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3442 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3443 December 2005, . 3445 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3446 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3447 2006, . 3449 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3450 Control Message Protocol (ICMPv6) for the Internet 3451 Protocol Version 6 (IPv6) Specification", STD 89, 3452 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3453 . 3455 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3456 Protocol (LDAP): The Protocol", RFC 4511, 3457 DOI 10.17487/RFC4511, June 2006, 3458 . 3460 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3461 "Considerations for Internet Group Management Protocol 3462 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3463 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3464 . 3466 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3467 "Internet Group Management Protocol (IGMP) / Multicast 3468 Listener Discovery (MLD)-Based Multicast Forwarding 3469 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3470 August 2006, . 3472 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3473 Algorithms in Cryptographically Generated Addresses 3474 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3475 . 3477 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3478 "Bidirectional Protocol Independent Multicast (BIDIR- 3479 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3480 . 3482 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3483 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3484 DOI 10.17487/RFC5214, March 2008, 3485 . 3487 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3488 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3489 February 2010, . 3491 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3492 Route Optimization Requirements for Operational Use in 3493 Aeronautics and Space Exploration Mobile Networks", 3494 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3495 . 3497 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3498 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3499 . 3501 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3502 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3503 January 2010, . 3505 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 3506 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 3507 . 3509 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3510 "IPv6 Router Advertisement Options for DNS Configuration", 3511 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3512 . 3514 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3515 NAT64: Network Address and Protocol Translation from IPv6 3516 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3517 April 2011, . 3519 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3520 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3521 . 3523 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3524 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3525 DOI 10.17487/RFC6221, May 2011, 3526 . 3528 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3529 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3530 DOI 10.17487/RFC6273, June 2011, 3531 . 3533 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3534 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3535 January 2012, . 3537 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 3538 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, 3539 DOI 10.17487/RFC6355, August 2011, 3540 . 3542 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3543 for Equal Cost Multipath Routing and Link Aggregation in 3544 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3545 . 3547 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3548 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3549 . 3551 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 3552 UDP Checksums for Tunneled Packets", RFC 6935, 3553 DOI 10.17487/RFC6935, April 2013, 3554 . 3556 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 3557 for the Use of IPv6 UDP Datagrams with Zero Checksums", 3558 RFC 6936, DOI 10.17487/RFC6936, April 2013, 3559 . 3561 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3562 Korhonen, "Requirements for Distributed Mobility 3563 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3564 . 3566 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3567 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3568 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3569 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3570 2016, . 3572 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3573 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3574 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3575 July 2018, . 3577 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3578 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3579 . 3581 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 3582 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3583 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 3584 . 3586 [WG] Wireguard, "Wireguard, https://www.wireguard.com", August 3587 2020. 3589 Appendix A. Non-Normative Considerations 3591 AERO can be applied to a multitude of Internetworking scenarios, with 3592 each having its own adaptations. The following considerations are 3593 provided as non-normative guidance: 3595 A.1. Implementation Strategies for Route Optimization 3597 Route optimization as discussed in Section 3.14 results in the route 3598 optimization source (ROS) creating an asymmetric neighbor cache entry 3599 for the target neighbor. The neighbor cache entry is maintained for 3600 at most ReachableTime seconds and then deleted unless updated. In 3601 order to refresh the neighbor cache entry lifetime before the 3602 ReachableTime timer expires, the specification requires 3603 implementations to issue a new NS/NA exchange to reset ReachableTime 3604 while data packets are still flowing. However, the decision of when 3605 to initiate a new NS/NA exchange and to perpetuate the process is 3606 left as an implementation detail. 3608 One possible strategy may be to monitor the neighbor cache entry 3609 watching for data packets for (ReachableTime - 5) seconds. If any 3610 data packets have been sent to the neighbor within this timeframe, 3611 then send an NS to receive a new NA. If no data packets have been 3612 sent, wait for 5 additional seconds and send an immediate NS if any 3613 data packets are sent within this "expiration pending" 5 second 3614 window. If no additional data packets are sent within the 5 second 3615 window, delete the neighbor cache entry. 3617 The monitoring of the neighbor data packet traffic therefore becomes 3618 an asymmetric ongoing process during the neighbor cache entry 3619 lifetime. If the neighbor cache entry expires, future data packets 3620 will trigger a new NS/NA exchange while the packets themselves are 3621 delivered over a longer path until route optimization state is re- 3622 established. 3624 A.2. Implicit Mobility Management 3626 OMNI interface neighbors MAY provide a configuration option that 3627 allows them to perform implicit mobility management in which no ND 3628 messaging is used. In that case, the Client only transmits packets 3629 over a single interface at a time, and the neighbor always observes 3630 packets arriving from the Client from the same link-layer source 3631 address. 3633 If the Client's underlying interface address changes (either due to a 3634 readdressing of the original interface or switching to a new 3635 interface) the neighbor immediately updates the neighbor cache entry 3636 for the Client and begins accepting and sending packets according to 3637 the Client's new address. This implicit mobility method applies to 3638 use cases such as cellphones with both WiFi and Cellular interfaces 3639 where only one of the interfaces is active at a given time, and the 3640 Client automatically switches over to the backup interface if the 3641 primary interface fails. 3643 A.3. Direct Underlying Interfaces 3645 When a Client's OMNI interface is configured over a Direct interface, 3646 the neighbor at the other end of the Direct link can receive packets 3647 without any encapsulation. In that case, the Client sends packets 3648 over the Direct link according to QoS preferences. If the Direct 3649 interface has the highest QoS preference, then the Client's IP 3650 packets are transmitted directly to the peer without going through an 3651 ANET/INET. If other interfaces have higher QoS preferences, then the 3652 Client's IP packets are transmitted via a different interface, which 3653 may result in the inclusion of Proxys, Servers and Bridges in the 3654 communications path. Direct interfaces must be tested periodically 3655 for reachability, e.g., via NUD. 3657 A.4. AERO Critical Infrastructure Considerations 3659 AERO Bridges can be either Commercial off-the Shelf (COTS) standard 3660 IP routers or virtual machines in the cloud. Bridges must be 3661 provisioned, supported and managed by the INET administrative 3662 authority, and connected to the Bridges of other INETs via inter- 3663 domain peerings. Cost for purchasing, configuring and managing 3664 Bridges is nominal even for very large OMNI links. 3666 AERO Servers can be standard dedicated server platforms, but most 3667 often will be deployed as virtual machines in the cloud. The only 3668 requirements for Servers are that they can run the AERO user-level 3669 code and have at least one network interface connection to the INET. 3670 As with Bridges, Servers must be provisioned, supported and managed 3671 by the INET administrative authority. Cost for purchasing, 3672 configuring and managing Servers is nominal especially for virtual 3673 Servers hosted in the cloud. 3675 AERO Proxys are most often standard dedicated server platforms with 3676 one network interface connected to the ANET and a second interface 3677 connected to an INET. As with Servers, the only requirements are 3678 that they can run the AERO user-level code and have at least one 3679 interface connection to the INET. Proxys must be provisioned, 3680 supported and managed by the ANET administrative authority. Cost for 3681 purchasing, configuring and managing Proxys is nominal, and borne by 3682 the ANET administrative authority. 3684 AERO Relays can be any dedicated server or COTS router platform 3685 connected to INETs and/or EUNs. The Relay connects to the OMNI link 3686 and engages in eBGP peering with one or more Bridges as a stub AS. 3687 The Relay then injects its MNPs and/or non-MNP prefixes into the BGP 3688 routing system, and provisions the prefixes to its downstream- 3689 attached networks. The Relay can perform ROS/ROR services the same 3690 as for any Server, and can route between the MNP and non-MNP address 3691 spaces. 3693 A.5. AERO Server Failure Implications 3695 AERO Servers may appear as a single point of failure in the 3696 architecture, but such is not the case since all Servers on the link 3697 provide identical services and loss of a Server does not imply 3698 immediate and/or comprehensive communication failures. Although 3699 Clients typically associate with a single Server at a time, Server 3700 failure is quickly detected and conveyed by Bidirectional Forward 3701 Detection (BFD) and/or proactive NUD allowing Clients to migrate to 3702 new Servers. 3704 If a Server fails, ongoing packet forwarding to Clients will continue 3705 by virtue of the asymmetric neighbor cache entries that have already 3706 been established in route optimization sources (ROSs). If a Client 3707 also experiences mobility events at roughly the same time the Server 3708 fails, unsolicited NA messages may be lost but proxy neighbor cache 3709 entries in the DEPARTED state will ensure that packet forwarding to 3710 the Client's new locations will continue for up to DepartTime 3711 seconds. 3713 If a Client is left without a Server for an extended timeframe (e.g., 3714 greater than ReachableTime seconds) then existing asymmetric neighbor 3715 cache entries will eventually expire and both ongoing and new 3716 communications will fail. The original source will continue to 3717 retransmit until the Client has established a new Server 3718 relationship, after which time continuous communications will resume. 3720 Therefore, providing many Servers on the link with high availability 3721 profiles provides resilience against loss of individual Servers and 3722 assurance that Clients can establish new Server relationships quickly 3723 in event of a Server failure. 3725 A.6. AERO Client / Server Architecture 3727 The AERO architectural model is client / server in the control plane, 3728 with route optimization in the data plane. The same as for common 3729 Internet services, the AERO Client discovers the addresses of AERO 3730 Servers and selects one Server to connect to. The AERO service is 3731 analogous to common Internet services such as google.com, yahoo.com, 3732 cnn.com, etc. However, there is only one AERO service for the link 3733 and all Servers provide identical services. 3735 Common Internet services provide differing strategies for advertising 3736 server addresses to clients. The strategy is conveyed through the 3737 DNS resource records returned in response to name resolution queries. 3738 As of January 2020 Internet-based 'nslookup' services were used to 3739 determine the following: 3741 o When a client resolves the domainname "google.com", the DNS always 3742 returns one A record (i.e., an IPv4 address) and one AAAA record 3743 (i.e., an IPv6 address). The client receives the same addresses 3744 each time it resolves the domainname via the same DNS resolver, 3745 but may receive different addresses when it resolves the 3746 domainname via different DNS resolvers. But, in each case, 3747 exactly one A and one AAAA record are returned. 3749 o When a client resolves the domainname "ietf.org", the DNS always 3750 returns one A record and one AAAA record with the same addresses 3751 regardless of which DNS resolver is used. 3753 o When a client resolves the domainname "yahoo.com", the DNS always 3754 returns a list of 4 A records and 4 AAAA records. Each time the 3755 client resolves the domainname via the same DNS resolver, the same 3756 list of addresses are returned but in randomized order (i.e., 3757 consistent with a DNS round-robin strategy). But, interestingly, 3758 the same addresses are returned (albeit in randomized order) when 3759 the domainname is resolved via different DNS resolvers. 3761 o When a client resolves the domainname "amazon.com", the DNS always 3762 returns a list of 3 A records and no AAAA records. As with 3763 "yahoo.com", the same three A records are returned from any 3764 worldwide Internet connection point in randomized order. 3766 The above example strategies show differing approaches to Internet 3767 resilience and service distribution offered by major Internet 3768 services. The Google approach exposes only a single IPv4 and a 3769 single IPv6 address to clients. Clients can then select whichever IP 3770 protocol version offers the best response, but will always use the 3771 same IP address according to the current Internet connection point. 3772 This means that the IP address offered by the network must lead to a 3773 highly-available server and/or service distribution point. In other 3774 words, resilience is predicated on high availability within the 3775 network and with no client-initiated failovers expected (i.e., it is 3776 all-or-nothing from the client's perspective). However, Google does 3777 provide for worldwide distributed service distribution by virtue of 3778 the fact that each Internet connection point responds with a 3779 different IPv6 and IPv4 address. The IETF approach is like google 3780 (all-or-nothing from the client's perspective), but provides only a 3781 single IPv4 or IPv6 address on a worldwide basis. This means that 3782 the addresses must be made highly-available at the network level with 3783 no client failover possibility, and if there is any worldwide service 3784 distribution it would need to be conducted by a network element that 3785 is reached via the IP address acting as a service distribution point. 3787 In contrast to the Google and IETF philosophies, Yahoo and Amazon 3788 both provide clients with a (short) list of IP addresses with Yahoo 3789 providing both IP protocol versions and Amazon as IPv4-only. The 3790 order of the list is randomized with each name service query 3791 response, with the effect of round-robin load balancing for service 3792 distribution. With a short list of addresses, there is still 3793 expectation that the network will implement high availability for 3794 each address but in case any single address fails the client can 3795 switch over to using a different address. The balance then becomes 3796 one of function in the network vs function in the end system. 3798 The same implications observed for common highly-available services 3799 in the Internet apply also to the AERO client/server architecture. 3800 When an AERO Client connects to one or more ANETs, it discovers one 3801 or more AERO Server addresses through the mechanisms discussed in 3802 earlier sections. Each Server address presumably leads to a fault- 3803 tolerant clustering arrangement such as supported by Linux-HA, 3804 Extended Virtual Synchrony or Paxos. Such an arrangement has 3805 precedence in common Internet service deployments in lightweight 3806 virtual machines without requiring expensive hardware deployment. 3807 Similarly, common Internet service deployments set service IP 3808 addresses on service distribution points that may relay requests to 3809 many different servers. 3811 For AERO, the expectation is that a combination of the Google/IETF 3812 and Yahoo/Amazon philosophies would be employed. The AERO Client 3813 connects to different ANET access points and can receive 1-2 Server 3814 LLAs at each point. It then selects one AERO Server address, and 3815 engages in RS/RA exchanges with the same Server from all ANET 3816 connections. The Client remains with this Server unless or until the 3817 Server fails, in which case it can switch over to an alternate 3818 Server. The Client can likewise switch over to a different Server at 3819 any time if there is some reason for it to do so. So, the AERO 3820 expectation is for a balance of function in the network and end 3821 system, with fault tolerance and resilience at both levels. 3823 Appendix B. Change Log 3825 << RFC Editor - remove prior to publication >> 3827 Changes from draft-templin-intarea-6706bis-61 to draft-templin- 3828 intrea-6706bis-62: 3830 o New sub-section on OMNI Neighbor Interface Attributes 3832 Changes from draft-templin-intarea-6706bis-59 to draft-templin- 3833 intrea-6706bis-60: 3835 o Removed all references to S/TLLAO - all Interface Attributes are 3836 now maintained completely in the OMNI option. 3838 Changes from draft-templin-intarea-6706bis-58 to draft-templin- 3839 intrea-6706bis-59: 3841 o The term "Relay" used in older draft versions is now "Bridge". 3842 "Relay" now refers to what was formally called: "Gateway". 3844 o Fine-grained cleanup of Forwarding Algorithm; IPv6 ND message 3845 addressing; OMNI Prefix Lengths, etc. 3847 Changes from draft-templin-intarea-6706bis-54 to draft-templin- 3848 intrea-6706bis-55: 3850 o Updates on Segment Routing and S/TLLAO contents. 3852 o Various editorials and addressing cleanups. 3854 Changes from draft-templin-intarea-6706bis-52 to draft-templin- 3855 intrea-6706bis-53: 3857 o Normative reference to the OMNI spec, and remove portions that are 3858 already specified in OMNI. 3860 o Renamed "AERO interface/link" to "OMIN interface/link" throughout 3861 the document. 3863 o Truncated obsolete back section matter. 3865 Author's Address 3867 Fred L. Templin (editor) 3868 Boeing Research & Technology 3869 P.O. Box 3707 3870 Seattle, WA 98124 3871 USA 3873 Email: fltemplin@acm.org