idnits 2.17.1 draft-templin-intarea-6706bis-73.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 8, 2020) is 1234 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: 'DLA' is mentioned on line 953, but not defined == Missing Reference: 'DLA0' is mentioned on line 960, but not defined == Missing Reference: 'DLA1' is mentioned on line 974, but not defined == Missing Reference: 'DLA2' is mentioned on line 964, but not defined == Missing Reference: 'DLA3' is mentioned on line 966, but not defined == Unused Reference: 'RFC3971' is defined on line 3244, but no explicit reference was found in the text == Unused Reference: 'RFC3972' is defined on line 3249, but no explicit reference was found in the text == Unused Reference: 'BGP' is defined on line 3297, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-comp-rtg-hdr' is defined on line 3300, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-crh-helper-opt' is defined on line 3306, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-frag-fragile' is defined on line 3311, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-dhcpv6-ndopt' is defined on line 3328, but no explicit reference was found in the text == Unused Reference: 'RFC4982' is defined on line 3450, but no explicit reference was found in the text == Unused Reference: 'RFC6273' is defined on line 3506, but no explicit reference was found in the text == Outdated reference: A later version (-99) exists of draft-templin-6man-omni-interface-54 == 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 8, 2020 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: June 11, 2021 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-intarea-6706bis-73 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 11, 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 Domain Local 71 Addresses (DLAs) . . . . . . . . . . . . . . . . . . 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 . . . . . . . . . . . . . . . . . . . . 37 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 . . . . 45 105 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 . . . . . . . . . . . 47 108 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 . . . . . . . . . . . . . . . . . 49 112 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 . . . . . . . . . . 54 119 3.16.4. Deactivating Existing Links . . . . . . . . . . . . 54 120 3.16.5. Moving Between Servers . . . . . . . . . . . . . . . 55 121 3.17. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 56 122 3.17.1. Source-Specific Multicast (SSM) . . . . . . . . . . 56 123 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 . . . . . . . . . . . . 61 130 3.23. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . 64 131 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 64 132 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 Domain-Local Address (DLA) 321 an IPv6 address from the prefix [DLA]::/10 constructed as 322 specified in [I-D.templin-6man-omni-interface]. OMNI DLAs are 323 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. 333 Mobile Network Prefix (MNP) 334 an IP prefix allocated from an MSP and delegated to an AERO Client 335 or Relay. 337 AERO node 338 a node that is connected to an OMNI link and participates in the 339 AERO internetworking and mobility service. 341 AERO Client ("Client") 342 an AERO node that connects over one or more underlying interfaces 343 and requests MNP delegation/registration service from AERO 344 Servers. The Client assigns a Client LLA to the OMNI interface 345 for use in ND exchanges with other AERO nodes and forwards packets 346 to correspondents according to OMNI interface neighbor cache 347 state. 349 AERO Server ("Server") 350 an INET node that configures an OMNI interface to provide default 351 forwarding and mobility/multilink services for AERO Clients. The 352 Server assigns an administratively-provisioned LLA to its OMNI 353 interface to support the operation of the ND services, and 354 advertises all of its associated MNPs via BGP peerings with 355 Bridges. 357 AERO Relay ("Relay") 358 an AERO Server that also provides forwarding services between 359 nodes reached via the OMNI link and correspondents on other links. 360 AERO Relays are provisioned with MNPs (i.e., the same as for an 361 AERO Client) and run a dynamic routing protocol to discover any 362 non-MNP IP routes. In both cases, the Relay advertises the MSP(s) 363 to its downstream networks, and distributes all of its associated 364 MNPs and non-MNP IP routes via BGP peerings with Bridges (i.e., 365 the same as for an AERO Server). 367 AERO Bridge ("Bridge") 368 a node that provides hybrid routing/bridging services (as well as 369 a security trust anchor) for nodes on an OMNI link. As a router, 370 the Bridge forwards packets using standard IP forwarding. As a 371 bridge, the Bridge forwards packets over the OMNI link without 372 decrementing the IPv6 Hop Limit. AERO Bridges peer with Servers 373 and other Bridges to discover the full set of MNPs for the link as 374 well as any non-MNPs that are reachable via Relays. 376 AERO Proxy ("Proxy") 377 a node that provides proxying services between Clients in an ANET 378 and Servers in external INETs. The AERO Proxy is a conduit 379 between the ANET and external INETs in the same manner as for 380 common web proxies, and behaves in a similar fashion as for ND 381 proxies [RFC4389]. A node may be configured to act as either a 382 Proxy and/or a Server, depending on Client Server selection 383 criteria. 385 ingress tunnel endpoint (ITE) 386 an OMNI interface endpoint that injects encapsulated packets into 387 an OMNI link. 389 egress tunnel endpoint (ETE) 390 an OMNI interface endpoint that receives encapsulated packets from 391 an OMNI link. 393 link-layer address 394 an IP address used as an encapsulation header source or 395 destination address from the perspective of the OMNI interface. 396 When an upper layer protocol (e.g., UDP) is used as part of the 397 encapsulation, the port number is also considered as part of the 398 link-layer address. 400 network layer address 401 the source or destination address of an encapsulated IP packet 402 presented to the OMNI interface. 404 end user network (EUN) 405 an internal virtual or external edge IP network that an AERO 406 Client or Relay connects to the rest of the network via the OMNI 407 interface. The Client/Relay sees each EUN as a "downstream" 408 network, and sees the OMNI interface as the point of attachment to 409 the "upstream" network. 411 Mobile Node (MN) 412 an AERO Client and all of its downstream-attached networks that 413 move together as a single unit, i.e., an end system that connects 414 an Internet of Things. 416 Mobile Router (MR) 417 a MN's on-board router that forwards packets between any 418 downstream-attached networks and the OMNI link. 420 Route Optimization Source (ROS) 421 the AERO node nearest the source that initiates route 422 optimization. The ROS may be a Server or Proxy acting on behalf 423 of the source Client. 425 Route Optimization responder (ROR) 426 the AERO node nearest the target destination that responds to 427 route optimization requests. The ROR may be a Server acting on 428 behalf of a target MNP Client, or a Relay for a non-MNP 429 destination. 431 MAP List 432 a geographically and/or topologically referenced list of addresses 433 of all Servers within the same OMNI link. There is a single MAP 434 list for the entire OMNI link. 436 Distributed Mobility Management (DMM) 437 a BGP-based overlay routing service coordinated by Servers and 438 Bridges that tracks all Server-to-Client associations. 440 Mobility Service (MS) 441 the collective set of all Servers, Proxys, Bridges and Relays that 442 provide the AERO Service to Clients. 444 Mobility Service Endpoint MSE) 445 an individual Server, Proxy, Bridge or Relay in the Mobility 446 Service. 448 Throughout the document, the simple terms "Client", "Server", 449 "Bridge", "Proxy" and "Relay" refer to "AERO Client", "AERO Server", 450 "AERO Bridge", "AERO Proxy" and "AERO Relay", respectively. 451 Capitalization is used to distinguish these terms from other common 452 Internetworking uses in which they appear without capitalization. 454 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 455 the names of node variables, messages and protocol constants) is used 456 throughout this document. The terms "All-Routers multicast", "All- 457 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 458 anycast" are defined in [RFC4291]. Also, the term "IP" is used to 459 generically refer to either Internet Protocol version, i.e., IPv4 460 [RFC0791] or IPv6 [RFC8200]. 462 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 463 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 464 "OPTIONAL" in this document are to be interpreted as described in BCP 465 14 [RFC2119][RFC8174] when, and only when, they appear in all 466 capitals, as shown here. 468 3. Asymmetric Extended Route Optimization (AERO) 470 The following sections specify the operation of IP over OMNI links 471 using the AERO service: 473 3.1. AERO Node Types 475 AERO Clients are Mobile Nodes (MNs) that connect via underlying 476 interfaces with addresses that may change when the Client moves to a 477 new network connection point. AERO Clients register their Mobile 478 Network Prefixes (MNPs) with the AERO service, and distribute the 479 MNPs to nodes on EUNs. AERO Bridges, Servers, Proxys and Relays are 480 critical infrastructure elements in fixed (i.e., non-mobile) INET 481 deployments and hence have permanent and unchanging INET addresses. 482 Together, they constitute the AERO service which provides an OMNI 483 link virtual overlay for connecting AERO Clients. 485 AERO Bridges provide hybrid routing/bridging services (as well as a 486 security trust anchor) for nodes on an OMNI link. Bridges use 487 standard IPv6 routing to forward packets both within the same INET 488 partitions and between disjoint INET partitions based on a mid-layer 489 IPv6 encapsulation per [RFC2473]. The inner IP layer experiences a 490 virtual bridging service since the inner IP TTL/Hop Limit is not 491 decremented during forwarding. Each Bridge also peers with Servers 492 and other Bridges in a dynamic routing protocol instance to provide a 493 Distributed Mobility Management (DMM) service for the list of active 494 MNPs (see Section 3.2.3). Bridges present the OMNI link as a set of 495 one or more Mobility Service Prefixes (MSPs) and configure secured 496 tunnels with Servers, Relays, Proxys and other Bridges; they further 497 maintain IP forwarding table entries for each MNP and any other 498 reachable non-MNP prefixes. 500 AERO Servers provide default forwarding and mobility/multilink 501 services for AERO Client Mobile Nodes (MNs). Each Server also peers 502 with Bridges in a dynamic routing protocol instance to advertise its 503 list of associated MNPs (see Section 3.2.3). Servers facilitate 504 prefix delegation/registration exchanges with Clients, where each 505 delegated prefix becomes an MNP taken from an MSP. Servers forward 506 packets between OMNI interface neighbors and track each Client's 507 mobility profiles. Servers may further act as Servers for some sets 508 of Clients and as Proxies for others. 510 AERO Proxys provide a conduit for ANET Clients to associate with 511 Servers in external INETs. Client and Servers exchange control plane 512 messages via the Proxy acting as a bridge between the ANET/INET 513 boundary. The Proxy forwards data packets between Clients and the 514 OMNI link according to forwarding information in the neighbor cache. 515 The Proxy function is specified in Section 3.13. Proxys may further 516 act as Proxys for some sets of Clients and as Servers for others. 518 AERO Relays are Servers that provide forwarding services between the 519 OMNI interface and INET/EUN interfaces. Relays are provisioned with 520 MNPs the same as for an AERO Client, and also run a dynamic routing 521 protocol to discover any non-MNP IP routes. The Relay advertises the 522 MSP(s) to its connected networks, and distributes all of its 523 associated MNPs and non-MNP IP routes via BGP peerings with Bridges 525 3.2. The AERO Service over OMNI Links 527 3.2.1. AERO/OMNI Reference Model 529 Figure 1 presents the basic OMNI link reference model: 531 +----------------+ 532 | AERO Bridge B1 | 533 | Nbr: S1, S2, P1| 534 |(X1->S1; X2->S2)| 535 | MSP M1 | 536 +-+---------+--+-+ 537 +--------------+ | Secured | | +--------------+ 538 |AERO Server S1| | tunnels | | |AERO Server S2| 539 | Nbr: C1, B1 +-----+ | +-----+ Nbr: C2, B1 | 540 | default->B1 | | | default->B1 | 541 | X1->C1 | | | X2->C2 | 542 +-------+------+ | +------+-------+ 543 | OMNI link | | 544 X===+===+===================+==)===============+===+===X 545 | | | | 546 +-----+--------+ +--------+--+-----+ +--------+-----+ 547 |AERO Client C1| | AERO Proxy P1 | |AERO Client C2| 548 | Nbr: S1 | |(Proxy Nbr Cache)| | Nbr: S2 | 549 | default->S1 | +--------+--------+ | default->S2 | 550 | MNP X1 | | | MNP X2 | 551 +------+-------+ .--------+------. +-----+--------+ 552 | (- Proxyed Clients -) | 553 .-. `---------------' .-. 554 ,-( _)-. ,-( _)-. 555 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 556 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 557 `-(______)-' +-------+ +-------+ `-(______)-' 559 Figure 1: AERO/OMNI Reference Model 561 In this model: 563 o the OMNI link is an overlay network service configured over one or 564 more underlying INET partitions which may be managed by different 565 administrative authorities and have incompatible protocols and/or 566 addressing plans. 568 o AERO Bridge B1 aggregates Mobility Service Prefix (MSP) M1, 569 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 570 via BGP peerings over secured tunnels to Servers (S1, S2). 571 Bridges connect the disjoint segments of a partitioned OMNI link. 573 o AERO Servers/Relays S1 and S2 configure secured tunnels with 574 Bridge B1 and also provide mobility, multilink and default router 575 services for their associated Clients C1 and C2. 577 o AERO Clients C1 and C2 associate with Servers S1 and S2, 578 respectively. They receive Mobile Network Prefix (MNP) 579 delegations X1 and X2, and also act as default routers for their 580 associated physical or internal virtual EUNs. Simple hosts H1 and 581 H2 attach to the EUNs served by Clients C1 and C2, respectively. 583 o AERO Proxy P1 configures a secured tunnel with Bridge B1 and 584 provides proxy services for AERO Clients in secured enclaves that 585 cannot associate directly with other OMNI link neighbors. 587 An OMNI link configured over a single INET appears as a single 588 unified link with a consistent underlying network addressing plan. 589 In that case, all nodes on the link can exchange packets via simple 590 INET encapsulation, since the underlying INET is connected. In 591 common practice, however, an OMNI link may be partitioned into 592 multiple "segments", where each segment is a distinct INET 593 potentially managed under a different administrative authority (e.g., 594 as for worldwide aviation service providers such as ARINC, SITA, 595 Inmarsat, etc.). Individual INETs may also themselves be partitioned 596 internally, in which case each internal partition is seen as a 597 separate segment. 599 The addressing plan of each segment is consistent internally but will 600 often bear no relation to the addressing plans of other segments. 601 Each segment is also likely to be separated from others by network 602 security devices (e.g., firewalls, proxies, packet filtering 603 gateways, etc.), and in many cases disjoint segments may not even 604 have any common physical link connections. Therefore, nodes can only 605 be assured of exchanging packets directly with correspondents in the 606 same segment, and not with those in other segments. The only means 607 for joining the segments therefore is through inter-domain peerings 608 between AERO Bridges. 610 The same as for traditional campus LANs, multiple OMNI link segments 611 can be joined into a single unified link via a virtual bridging 612 service using the OMNI Adaptation Layer (OAL) which inserts a mid- 613 layer IPv6 encapsulation per [RFC2473] that supports inter-segment 614 forwarding (i.e., bridging) without decrementing the network-layer 615 TTL/Hop Limit. This bridging of OMNI link segments is shown in 616 Figure 2: 618 . . . . . . . . . . . . . . . . . . . . . . . 619 . . 620 . .-(::::::::) . 621 . .-(::::::::::::)-. +-+ . 622 . (:::: Segment A :::)--|B|---+ . 623 . `-(::::::::::::)-' +-+ | . 624 . `-(::::::)-' | . 625 . | . 626 . .-(::::::::) | . 627 . .-(::::::::::::)-. +-+ | . 628 . (:::: Segment B :::)--|B|---+ . 629 . `-(::::::::::::)-' +-+ | . 630 . `-(::::::)-' | . 631 . | . 632 . .-(::::::::) | . 633 . .-(::::::::::::)-. +-+ | . 634 . (:::: Segment C :::)--|B|---+ . 635 . `-(::::::::::::)-' +-+ | . 636 . `-(::::::)-' | . 637 . | . 638 . ..(etc).. x . 639 . . 640 . . 641 . <- OMNI link Bridged by encapsulation -> . 642 . . . . . . . . . . . . . .. . . . . . . . . 644 Figure 2: Bridging OMNI Link Segments 646 Bridges, Servers, Relays and Proxys connect via secured INET tunnels 647 over their respective segments in a spanning tree topology rooted at 648 the Bridges. The secured spanning tree supports strong 649 authentication for IPv6 ND control messages and may also be used to 650 convey the initial data packets in a flow. Route optimization can 651 then be employed to cause data packets to take more direct paths 652 between OMNI link neighbors without having to strictly follow the 653 spanning tree. 655 3.2.2. Link-Local Addresses (LLAs) and Domain Local Addresses (DLAs) 657 AERO nodes on OMNI links use the Link-Local Address (LLA) prefix 658 fe80::/64 [RFC4291] to assign LLAs used for network-layer addresses 659 in IPv6 ND and data messages. They also use the Domain Local Address 660 (DLA) prefix [DLA]::/48 to form per-link DLA sub-prefixes of the form 661 [DLA*]::/64, where '*' is a 16-bit OMNI link identifier. Each per- 662 link DLA sub-prefix is in turn used to form DLAs used for OAL header 663 source and destination addresses. See 664 [I-D.templin-6man-omni-interface] for a full specification of the 665 LLAs and DLAs used by AERO nodes on OMNI links. 667 IPv6 routers are not permitted to forward packets with LLA addresses. 668 Therefore, IPv6 packets with LLA addresses must be encapsulated in 669 IPv6 headers with DLA addresses in order to traverse the OMNI link. 670 DLA addresses therefore must be routable within the OMNI link domain- 671 local area. 673 For routing system organization (see Section 3.2.3), DLAs are 674 organized in partition prefixes, e.g., [DLA]::1000/116. For each 675 such partition prefix, the Bridge(s) that connect that segment assign 676 the all-zero's address of the prefix as a Subnet Router Anycast 677 address. For example, the Subnet Router Anycast address for 678 [DLA*]::1000/116 is simply [DLA*]::1000. 680 3.2.3. AERO Routing System 682 The AERO routing system comprises a private instance of the Border 683 Gateway Protocol (BGP) [RFC4271] that is coordinated between Bridges 684 and Servers and does not interact with either the public Internet BGP 685 routing system or any underlying INET routing systems. 687 In a reference deployment, each Server is configured as an Autonomous 688 System Border Router (ASBR) for a stub Autonomous System (AS) using 689 an AS Number (ASN) that is unique within the BGP instance, and each 690 Server further uses eBGP to peer with one or more Bridges but does 691 not peer with other Servers. Each INET of a multi-segment OMNI link 692 must include one or more Bridges, which peer with the Servers and 693 Proxys within that INET. All Bridges within the same INET are 694 members of the same hub AS using a common ASN, and use iBGP to 695 maintain a consistent view of all active MNPs currently in service. 696 The Bridges of different INETs peer with one another using eBGP. 698 Bridges advertise the OMNI link's MSPs and any non-MNP routes to each 699 of their Servers. This means that any aggregated non-MNPs (including 700 "default") are advertised to all Servers. Each Bridge configures a 701 black-hole route for each of its MSPs. By black-holing the MSPs, the 702 Bridge will maintain forwarding table entries only for the MNPs that 703 are currently active, and packets destined to all other MNPs will 704 correctly incur Destination Unreachable messages due to the black- 705 hole route. In this way, Servers have only partial topology 706 knowledge (i.e., they know only about the MNPs of their directly 707 associated Clients) and they forward all other packets to Bridges 708 which have full topology knowledge. 710 Each OMNI link segment assigns a unique sub-prefix of [DLA*]::/96 711 known as the DLA partition prefix. For example, a first segment 712 could assign [DLA*]::1000/116, a second could assign 713 [DLA*]::2000/116, a third could assign [DLA*]::3000/116, etc. The 714 administrative authorities for each segment must therefore coordinate 715 to assure mutually-exclusive partition prefix assignments, but 716 internal provisioning of each prefix is an independent local 717 consideration for each administrative authority. 719 DLA partition prefixes are statically represented in Bridge 720 forwarding tables. Bridges join multiple segments into a unified 721 OMNI link over multiple diverse administrative domains. They support 722 a bridging function by first establishing forwarding table entries 723 for their partition prefixes either via standard BGP routing or 724 static routes. For example, if three Bridges ('A', 'B' and 'C') from 725 different segments serviced [DLA*]::1000/116, [DLA*]::2000/116 and 726 [DLA*]::3000/116 respectively, then the forwarding tables in each 727 Bridge are as follows: 729 A: [DLA*]::1000/116->local, [DLA*]::2000/116->B, [DLA*]::3000/116->C 731 B: [DLA*]::1000/116->A, [DLA*]::2000/116->local, [DLA*]::3000/116->C 733 C: [DLA*]::1000/116->A, [DLA*]::2000/116->B, [DLA*]::3000/116->local 735 These forwarding table entries are permanent and never change, since 736 they correspond to fixed infrastructure elements in their respective 737 segments. 739 DLA Client prefixes are instead dynamically advertised in the AERO 740 routing system by Servers and Relays that provide service for their 741 corresponding MNPs. For example, if three Servers ('D', 'E' and 'F') 742 service the MNPs 2001:db8:1000:2000::/56, 2001:db8:3000:4000::/56 and 743 2001:db8:5000:6000::/56 then the routing system would include: 745 D: [DLA*]:2001:db8:1000:2000/120 747 E: [DLA*]:2001:db8:3000:4000/120 749 F: [DLA*]:2001:db8:5000:6000/120 751 A full discussion of the BGP-based routing system used by AERO is 752 found in [I-D.ietf-rtgwg-atn-bgp]. 754 3.2.4. OMNI Link Encapsulation 756 With the Client and partition prefixes in place in Bridge forwarding 757 tables, the OMNI interface sends control and data packets toward AERO 758 destination nodes located in different OMNI link segments over the 759 spanning tree via mid-layer encapsulation using the OMNI Adaptation 760 Layer (OAL) header based on Generic Packet Tunneling in IPv6 761 [RFC2473]. When necessary, the OMNI interface also includes an OMNI 762 Routing Header (ORH) as an extension to the OAL header if final 763 segment forwarding information is available, e.g., in the neighbor 764 cache. The ORH is formatted as shown in Figure 3: 766 0 1 2 3 767 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 1 768 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 769 | Next Header | Hdr Ext Len | Routing Type | SRT | FMT | 770 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 771 | Last Hop Segment-id (LHS) | 772 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 773 ~ Link Layer Address (L2ADDR) ~ 774 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 775 ~ Destination Suffix (if necessary) ~ 776 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 777 ~ Null Padding (if necessary) ~ 778 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 780 Figure 3: OMNI Routing Header (ORH) Format 782 In this format: 784 o Next Header identifies the type of header immediately following 785 the ORH. 787 o Hdr Ext Len is the length of the Routing header in 8-octet units 788 (not including the first 8 octets), with trailing padding added if 789 necessary to produce an integral number of 8-octet units. 791 o Routing Type is set to TBD (see IANA Considerations). 793 o Segments Left is omitted, and replaced by a 5-bit SRT and 3-bit 794 FMT field. 796 o SRT - a 5-bit Segment Routing Topology prefix length value that 797 (when added to 96) determines the prefix length to apply to the 798 DLA formed from concatenating [DLA*]::/96 with the 32 bit LHS MSID 799 value that follows. For example, the value 16 corresponds to the 800 prefix length 112. 802 o FMT - a 3-bit "Framework/Mode/Type" code corresponding to the 803 included Link Layer Address as follows: 805 * When the most significant bit (i.e., "Framework") is set to 0, 806 L2ADDR is the INET encapsulation address of the Server/Proxy 807 named in the LHS; otherwise, it is the address for the target 808 Client, and the Destination Suffix is included. 810 * When the next most significant bit (i.e., "Mode") is set to 0, 811 the Source/Target L2ADDR is on the open INET; otherwise, it is 812 (likely) located behind a Network Address Translator (NAT). 814 * When the least significant bit (i.e., "Type") is set to 0, 815 L2ADDR includes a UDP Port Number followed by an IPv4 address; 816 otherwise, it includes a UDP Port Number followed by an IPv6 817 address. 819 o LHS - the 32 bit MSID of the Last Hop Server/Proxy on the path to 820 the target. When SRT and LHS are both set to 0, the LHS is 821 considered unspecified. When SRT is set to 0 and LHS is non-zero, 822 the prefix length is set to 128. SRT and LHS provide guidance to 823 the OMNI interface forwarding algorithm. Specifically, if SRT/LHS 824 is located in the local OMNI link segment then the OMNI interface 825 can omit OAL encapsualtion and encapsulate according to FMT/ 826 L2ADDR; else, it must perform OAL encapsulation and forward 827 according to the OMNI link spanning tree. 829 o Link Layer Address (L2ADDR) - Formatted according to FMT, and 830 identifies the link-layer address (i.e., the encapsulation 831 address) of the source/target. The UDP Port Number appears in the 832 first two octets and the IP address appears in the next 4 octets 833 for IPv4 or 16 octets for IPv6. The Port Number and IP address 834 are recorded in ones-compliment "obfuscated" form per [RFC4380]. 835 The OMNI interface forwarding algorithm uses FMT/L2ADDR to 836 determine the encapsulation address for forwarding when SRT/LHS is 837 located in the local OMNI link segment. 839 o Destination Suffix is a 64-bit field included only for OAL- 840 encapsulated packets that are destined directly to the DLA of the 841 Client (i.e., according to the FMT code). When present, 842 Destination Suffix encodes the 64-bit DLA suffix for the Client 843 that will receive packet. For example, if the Client DLA is 844 [DLA*]:2001:db8:1:2 then the Destination suffix encodes the value 845 2001:db8:1:2. 847 o Null Padding contains zero-valued octets as necessary to pad the 848 ORH to an integral number of 8-octet units. 850 When an AERO node uses OAL encapsulation for a packet with addresses 851 such as 2001:db8:1:2::1 and 2001:db8:1234:5678::1, it sets the OAL 852 header source address to its own DLA address (e.g., 853 [DLA*]::1000:2000). The node also sets the destination address to 854 the DLA of the Client (e.g., [DLA*]::2001:db8:1234:5678) when the 855 Client can be directly addressed; otherwise, it sets the destination 856 address to the DLA of the Client's Proxy/Server (e.g., 857 [DLA*]::4321:9876). The node then fragments the mid-layer packet if 858 necessary then encapsulates each resulting OAL packet in an INET 859 header with source address set to its own INET address (e.g., 860 192.0.2.100) and destination set to the INET address of a Bridge 861 (e.g., 192.0.2.1). 863 If the neighbor cache includes final segment addressing information 864 for the target destination, the node inserts an ORH immediately 865 following the OAL header of each packet/fragment while including the 866 correct SRT, FMT, LHS, L2ADDR and Destination Suffix information. 867 Next, the node overwrites the IPv6 destination address with the 868 Subnet Router Anycast address corresponding to the LHS (for example, 869 for LHS [DLA]::1000:2000 the Subnet Router Anycast address is 870 [DLA*]::1000:0000). The encapsulation format in the above example is 871 shown in Figure 4: 873 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 874 | INET Header | 875 | src = 192.0.2.100 | 876 | dst = 192.0.2.1 | 877 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 878 | OAL Header | 879 | src = [DLA*]::1000:2000 | 880 | dst= DLA for inner IP dst | 881 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 882 | ORH Header | 883 | (if necessary) | 884 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 885 | Inner IP Header | 886 | src = 2001:db8:1:2::1 | 887 | dst = 2001:db8:1234:5678::1 | 888 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 889 | | 890 ~ ~ 891 ~ Inner Packet Body ~ 892 ~ ~ 893 | | 894 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 896 Figure 4: OAL/ORH Encapsulation 898 In this format, the inner IP header and packet body are the original 899 IP packet, the OAL header is an IPv6 header prepared according to 900 [RFC2473], the ORH is a Routing Header extension of the OAL header, 901 and the INET header is prepared as discussed in Section 3.6. 903 This gives rise to a routing system that contains both Client prefix 904 routes that may change dynamically due to regional node mobility and 905 partition prefix routes that rarely if ever change. The Bridges can 906 therefore provide link-layer bridging by sending packets over the 907 spanning tree instead of network-layer routing according to MNP 908 routes. As a result, opportunities for packet loss due to node 909 mobility between different segments are mitigated. 911 In normal operations, IPv6 ND messages are conveyed over secured 912 paths between OMNI link neighbors so that specific Proxys, Servers or 913 Relays can be addressed without being subject to mobility events. 914 Conversely, only the first few packets destined to Clients need to 915 traverse secured paths until route optimization can determine a more 916 direct path. 918 Note: An IPv6 "minimal encapsulation" format (i.e., an IPv6 variant 919 of [RFC2004]) based on extensions to the ORH was considered, analyzed 920 and rejected. In the approach, the ORH would be inserted as an 921 extension header to the original IPv6 packet header. The IPv6 922 destination address would then be written into the ORH, and the DLA 923 corresponding to the destination would be overwritten in the IPv6 924 destination address. This would seemingly convey enough forwarding 925 information so that OAL encapsulation could be avoided. However, 926 this "minimal encapsulation" IPv6 packet would then have a non-DLA 927 source address and DLA destination address, an incorrect value in 928 upper layer protocol checksums, and a Hop Limit that is decremented 929 within the spanning tree when it should not be. The insertion and 930 removal of the ORH would also entail rewriting the Payload Length and 931 Next Header fields - again, invalidating upper layer checksums. 932 These irregularities would result in implementation challenges and 933 the potential for operational issues, e.g., since actionable ICMPv6 934 error reports could not be delivered to the original source. In 935 order to address the issues, still more information such as the 936 original IPv6 source address could be written into the ORH. However, 937 with the additional information the benefit of the "minimal 938 encapsulation" savings quickly diminishes, and becomes overshadowed 939 by the implementation and operational irregularities. 941 3.2.5. Segment Routing Topologies (SRTs) 943 The 64-bit sub-prefixes of [DLA]::/48 identify up to 2^16 distinct 944 Segment Routing Topologies (SRTs). Each SRT is a mutually-exclusive 945 OMNI link overlay instance using a distinct set of DLAs, and emulates 946 a Virtual LAN (VLAN) service for the OMNI link. In some cases (e.g., 947 when redundant topologies are needed for fault tolerance and 948 reliability) it may be beneficial to deploy multiple SRTs that act as 949 independent overlay instances. A communication failure in one 950 instance therefore will not affect communications in other instances. 952 Each SRT is identified by a distinct value in bits 48-63 of 953 [DLA]::/48, i.e., as [DLA0]::/64, [DLA1]::/64, [DLA2]::/64, etc. 955 This document asserts that up to four SRTs provide a level of safety 956 sufficient for critical communications such as civil aviation. Each 957 SRT is designated with a color that identifies a different OMNI link 958 instance as follows: 960 o Red - corresponds to [DLA0]::/64 962 o Green - corresponds to [DLA1]::/64 964 o Blue-1 - corresponds to [DLA2]::/64 966 o Blue-2 - corresponds to [DLA3]::/64 968 o the remaining [DLA*]::/64 sub prefixes are available for 969 additional SRTs. 971 Each OMNI interface is identified by a unique interface name (e.g., 972 omni0, omni1, omni2, etc.) and assigns an anycast DLA corresponding 973 to its SRT prefix. For example, the anycast DLA for the Green SRT is 974 simply [DLA1]::. The anycast DLA is used for OMNI interface 975 determination in Safety-Based Multilink (SBM) as discussed in 976 [I-D.templin-6man-omni-interface]. Each OMNI interface further 977 applies Performance-Based Multilink (PBM) internally. 979 3.2.6. Segment Routing For OMNI Link Selection 981 An original IPv6 source can direct an IPv6 packet to an AERO node by 982 including a standard IPv6 Segment Routing Header (SRH) [RFC8754] with 983 the anycast DLA for the selected SRT as either the IPv6 destination 984 or as an intermediate hop within the SRH. This allows the original 985 source to determine the specific OMNI link topology a packet will 986 traverse when there may be multiple alternatives. 988 When the AERO node processes the SRH and forwards the packet to the 989 correct OMNI interface, the OMNI interface writes the next IPv6 990 address from the SRH into the IPv6 destination address and decrements 991 Segments Left. If decrementing would cause Segments Left to become 992 0, the OMNI interface deletes the SRH before forwarding. This form 993 of Segment Routing supports Safety-Based Multilink (SBM). 995 3.2.7. Segment Routing Within the OMNI Link 997 AERO node OMNI interfaces can insert OAL/ORH headers for Segment 998 Routing within the OMNI link to influence the paths of packets 999 destined to targets in remote segments without requiring all packets 1000 to traverse strict spanning tree paths. 1002 When an AERO node's OMNI interface has a packet to send to a target 1003 discovered through route optimization located in the same OMNI link 1004 segment, it encapsulates the packet in OAL/ORH headers if necessary 1005 as discussed above. The node then uses the target's Link Layer 1006 Address (L2ADDR) information for INET encapsulation. 1008 When an AERO node's OMNI interface has a packet to send to a route 1009 optimization target located in a remote OMNI link segment, it 1010 encapsulates the packet in OAL/ORH headers as discussed above while 1011 forwarding the packet to a Bridge with destination set to the Subnet 1012 Router Anycast address for the final OMNI link segment. 1014 When a Bridge receives a packet destined to its Subnet Router Anycast 1015 address with an OAL/ORH with SRT/LHS values corresponding to the 1016 local segment, it examines the L2ADDR according to FMT and removes 1017 the ORH from the packet. If the ORH includes a saved Destination 1018 Suffix, the Bridge then writes the corresponding DLA into the OAL 1019 destination address; otherwise, it writes the DLA corresponding to 1020 the SRT/LHS fields into the destination. The Bridge then 1021 encapsulates the packet in an INET header according to L2ADDR and 1022 forwards the packet within the INET either to the LHS Server/Proxy or 1023 directly to the destination itself. In this way, the Bridge 1024 participates in route optimization to reduce traffic load and 1025 suboptimal routing through strict spanning tree paths. 1027 3.3. OMNI Interface Characteristics 1029 OMNI interfaces are virtual interfaces configured over one or more 1030 underlying interfaces classified as follows: 1032 o INET interfaces connect to an INET either natively or through one 1033 or several IPv4 Network Address Translators (NATs). Native INET 1034 interfaces have global IP addresses that are reachable from any 1035 INET correspondent. All Server, Relay and Bridge interfaces are 1036 native interfaces, as are INET-facing interfaces of Proxys. NATed 1037 INET interfaces connect to a private network behind one or more 1038 NATs that provide INET access. Clients that are behind a NAT are 1039 required to send periodic keepalive messages to keep NAT state 1040 alive when there are no data packets flowing. 1042 o ANET interfaces connect to an ANET that is separated from the open 1043 INET by a Proxy. Proxys can actively issue control messages over 1044 the INET on behalf of the Client to reduce ANET congestion. 1046 o VPNed interfaces use security encapsulation over the INET to a 1047 Virtual Private Network (VPN) server that also acts as a Server or 1048 Proxy. Other than the link-layer encapsulation format, VPNed 1049 interfaces behave the same as Direct interfaces. 1051 o Direct interfaces connect a Client directly to a Server or Proxy 1052 without crossing any ANET/INET paths. An example is a line-of- 1053 sight link between a remote pilot and an unmanned aircraft. The 1054 same Client considerations apply as for VPNed interfaces. 1056 OMNI interfaces use OAL/ORH encapsulation as necessary as discussed 1057 in Section 3.2.4. OMNI interfaces use link-layer encapsulation (see: 1058 Section 3.6) to exchange packets with OMNI link neighbors over INET 1059 or VPNed interfaces as well as over ANET interfaces for which the 1060 Client and Proxy may be multiple IP hops away. OMNI interfaces do 1061 not use link-layer encapsulation over Direct underlying interfaces or 1062 ANET interfaces when the Client and Proxy are known to be on the same 1063 underlying link. 1065 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 1066 state the same as for any interface. OMNI interfaces use ND messages 1067 including Router Solicitation (RS), Router Advertisement (RA), 1068 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 1069 neighbor cache management. 1071 OMNI interfaces send ND messages with an OMNI option formatted as 1072 specified in [I-D.templin-6man-omni-interface]. The OMNI option 1073 includes prefix registration information and Interface Attributes 1074 containing link information parameters for the OMNI interface's 1075 underlying interfaces. Each OMNI option may include multiple 1076 Interface Attributes sub-options, each identified by an ifIndex 1077 value. 1079 A Client's OMNI interface may be configured over multiple underlying 1080 interface connections. For example, common mobile handheld devices 1081 have both wireless local area network ("WLAN") and cellular wireless 1082 links. These links are often used "one at a time" with low-cost WLAN 1083 preferred and highly-available cellular wireless as a standby, but a 1084 simultaneous-use capability could provide benefits. In a more 1085 complex example, aircraft frequently have many wireless data link 1086 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1087 directional, etc.) with diverse performance and cost properties. 1089 If a Client's multiple underlying interfaces are used "one at a time" 1090 (i.e., all other interfaces are in standby mode while one interface 1091 is active), then ND message OMNI options include only a single 1092 Interface Attributes sub-option set to constant values. In that 1093 case, the Client would appear to have a single interface but with a 1094 dynamically changing link-layer address. 1096 If the Client has multiple active underlying interfaces, then from 1097 the perspective of ND it would appear to have multiple link-layer 1098 addresses. In that case, ND message OMNI options MAY include 1099 multiple Interface Attributes sub-options - each with values that 1100 correspond to a specific interface. Every ND message need not 1101 include Interface Attributes for all underlying interfaces; for any 1102 attributes not included, the neighbor considers the status as 1103 unchanged. 1105 Bridge, Server and Proxy OMNI interfaces may be configured over one 1106 or more secured tunnel interfaces. The OMNI interface configures 1107 both an LLA and its corresponding DLA, while the underlying secured 1108 tunnel interfaces are either unnumbered or configure the same DLA. 1109 The OMNI interface encapsulates each IP packet in OAL/ORH headers and 1110 presents the packet to the underlying secured tunnel interface. 1111 Routing protocols such as BGP that run over the OMNI interface do not 1112 employ OAL/ORH encapsulation, but rather present the routing protocol 1113 messages directly to the underlying secured tunnels while using the 1114 DLA as the source address. This distinction must be honored 1115 consistently according to each node's configuration so that the IP 1116 forwarding table will associate discovered IP routes with the correct 1117 interface. 1119 3.4. OMNI Interface Initialization 1121 AERO Servers, Proxys and Clients configure OMNI interfaces as their 1122 point of attachment to the OMNI link. AERO nodes assign the MSPs for 1123 the link to their OMNI interfaces (i.e., as a "route-to-interface") 1124 to ensure that packets with destination addresses covered by an MNP 1125 not explicitly assigned to a non-OMNI interface are directed to the 1126 OMNI interface. 1128 OMNI interface initialization procedures for Servers, Proxys, Clients 1129 and Bridges are discussed in the following sections. 1131 3.4.1. AERO Server/Relay Behavior 1133 When a Server enables an OMNI interface, it assigns an LLA/DLA 1134 appropriate for the given OMNI link segment. The Server also 1135 configures secured tunnels with one or more neighboring Bridges and 1136 engages in a BGP routing protocol session with each Bridge. 1138 The OMNI interface provides a single interface abstraction to the IP 1139 layer, but internally comprises multiple secured tunnels as well as 1140 an NBMA nexus for sending encapsulated data packets to OMNI interface 1141 neighbors. The Server further configures a service to facilitate ND 1142 exchanges with AERO Clients and manages per-Client neighbor cache 1143 entries and IP forwarding table entries based on control message 1144 exchanges. 1146 Relays are simply Servers that run a dynamic routing protocol to 1147 redistribute routes between the OMNI interface and INET/EUN 1148 interfaces (see: Section 3.2.3). The Relay provisions MNPs to 1149 networks on the INET/EUN interfaces (i.e., the same as a Client would 1150 do) and advertises the MSP(s) for the OMNI link over the INET/EUN 1151 interfaces. The Relay further provides an attachment point of the 1152 OMNI link to a non-MNP-based global topology. 1154 3.4.2. AERO Proxy Behavior 1156 When a Proxy enables an OMNI interface, it assigns an LLA/DLA and 1157 configures permanent neighbor cache entries the same as for Servers. 1158 The Proxy also configures secured tunnels with one or more 1159 neighboring Bridges and maintains per-Client neighbor cache entries 1160 based on control message exchanges. Importantly Proxys are often 1161 configured to act as Servers, and vice-versa. 1163 3.4.3. AERO Client Behavior 1165 When a Client enables an OMNI interface, it sends RS messages with ND 1166 parameters over its underlying interfaces to a Server, which returns 1167 an RA message with corresponding parameters. (The RS/RA messages may 1168 pass through a Proxy in the case of a Client's ANET interface, or 1169 through one or more NATs in the case of a Client's INET interface.) 1171 3.4.4. AERO Bridge Behavior 1173 AERO Bridges configure an OMNI interface and assign the DLA Subnet 1174 Router Anycast address for each OMNI link segment they connect to. 1175 Bridges configure secured tunnels with Servers, Proxys and other 1176 Bridges; they also configure LLAs/DLAs and permanent neighbor cache 1177 entries the same as Servers. Bridges engage in a BGP routing 1178 protocol session with a subset of the Servers and other Bridges on 1179 the spanning tree (see: Section 3.2.3). 1181 3.5. OMNI Interface Neighbor Cache Maintenance 1183 Each OMNI interface maintains a conceptual neighbor cache that 1184 includes an entry for each neighbor it communicates with on the OMNI 1185 link per [RFC4861]. OMNI interface neighbor cache entries are said 1186 to be one of "permanent", "symmetric", "asymmetric" or "proxy". 1188 Permanent neighbor cache entries are created through explicit 1189 administrative action; they have no timeout values and remain in 1190 place until explicitly deleted. AERO Bridges maintain permanent 1191 neighbor cache entries for their associated Proxys/Servers (and vice- 1192 versa). Each entry maintains the mapping between the neighbor's 1193 network-layer LLA and corresponding INET address. 1195 Symmetric neighbor cache entries are created and maintained through 1196 RS/RA exchanges as specified in Section 3.12, and remain in place for 1197 durations bounded by prefix lifetimes. AERO Servers maintain 1198 symmetric neighbor cache entries for each of their associated 1199 Clients, and AERO Clients maintain symmetric neighbor cache entries 1200 for each of their associated Servers. 1202 Asymmetric neighbor cache entries are created or updated based on 1203 route optimization messaging as specified in Section 3.14, and are 1204 garbage-collected when keepalive timers expire. AERO ROSs maintain 1205 asymmetric neighbor cache entries for active targets with lifetimes 1206 based on ND messaging constants. Asymmetric neighbor cache entries 1207 are unidirectional since only the ROS (and not the ROR) creates an 1208 entry. 1210 Proxy neighbor cache entries are created and maintained by AERO 1211 Proxys when they process Client/Server ND exchanges, and remain in 1212 place for durations bounded by ND and prefix lifetimes. AERO Proxys 1213 maintain proxy neighbor cache entries for each of their associated 1214 Clients. Proxy neighbor cache entries track the Client state and the 1215 address of the Client's associated Server(s). 1217 To the list of neighbor cache entry states in Section 7.3.2 of 1218 [RFC4861], Proxy and Server OMNI interfaces add an additional state 1219 DEPARTED that applies to symmetric and proxy neighbor cache entries 1220 for Clients that have recently departed. The interface sets a 1221 "DepartTime" variable for the neighbor cache entry to "DEPART_TIME" 1222 seconds. DepartTime is decremented unless a new ND message causes 1223 the state to return to REACHABLE. While a neighbor cache entry is in 1224 the DEPARTED state, packets destined to the target Client are 1225 forwarded to the Client's new location instead of being dropped. 1226 When DepartTime decrements to 0, the neighbor cache entry is deleted. 1227 It is RECOMMENDED that DEPART_TIME be set to the default constant 1228 value REACHABLE_TIME plus 10 seconds (40 seconds by default) to allow 1229 a window for packets in flight to be delivered while stale route 1230 optimization state may be present. 1232 When an ROR receives an authentic NS message used for route 1233 optimization, it searches for a symmetric neighbor cache entry for 1234 the target Client. The ROR then returns a solicited NA message 1235 without creating a neighbor cache entry for the ROS, but creates or 1236 updates a target Client "Report List" entry for the ROS and sets a 1237 "ReportTime" variable for the entry to REPORT_TIME seconds. The ROR 1238 resets ReportTime when it receives a new authentic NS message, and 1239 otherwise decrements ReportTime while no authentic NS messages have 1240 been received. It is RECOMMENDED that REPORT_TIME be set to the 1241 default constant value REACHABLE_TIME plus 10 seconds (40 seconds by 1242 default) to allow a window for route optimization to converge before 1243 ReportTime decrements below REACHABLE_TIME. 1245 When the ROS receives a solicited NA message response to its NS 1246 message used for route optimization, it creates or updates an 1247 asymmetric neighbor cache entry for the target network-layer and 1248 link-layer addresses. The ROS then (re)sets ReachableTime for the 1249 neighbor cache entry to REACHABLE_TIME seconds and uses this value to 1250 determine whether packets can be forwarded directly to the target, 1251 i.e., instead of via a default route. The ROS otherwise decrements 1252 ReachableTime while no further solicited NA messages arrive. It is 1253 RECOMMENDED that REACHABLE_TIME be set to the default constant value 1254 30 seconds as specified in [RFC4861]. 1256 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1257 of NS keepalives sent when a correspondent may have gone unreachable, 1258 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1259 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1260 to limit the number of unsolicited NAs that can be sent based on a 1261 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1262 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1263 same as specified in [RFC4861]. 1265 Different values for DEPART_TIME, REPORT_TIME, REACHABLE_TIME, 1266 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1267 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1268 different values are chosen, all nodes on the link MUST consistently 1269 configure the same values. Most importantly, DEPART_TIME and 1270 REPORT_TIME SHOULD be set to a value that is sufficiently longer than 1271 REACHABLE_TIME to avoid packet loss due to stale route optimization 1272 state. 1274 3.5.1. OMNI Neighbor Interface Attributes 1276 OMNI interface IPv6 ND messages include OMNI options 1277 [I-D.templin-6man-omni-interface] with Interface Attributes that 1278 provide Link-Layer Address and QoS Preference information for the 1279 neighbor's underlying interfaces. This information is stored in the 1280 neighbor cache and provides the basis for the forwarding algorithm 1281 specified in Section 3.10. The information is cumulative and 1282 reflects the union of the OMNI information from the most recent ND 1283 messages received from the neighbor; it is therefore not required 1284 that each ND message contain all neighbor information. 1286 The OMNI option Interface Attributes for each underlying interface 1287 includes a two-part "Link-Layer Address" consisting of a simple IP 1288 encapsulation address determined by the FMT and L2ADDR fields and an 1289 OAL DLA determined by the SRT and LHS fields. If the neighbor is 1290 located in the local OMNI link segment (and, if any necessary NAT 1291 state has been established) forwarding via simple IP encapsulation 1292 can be used; otherwise, OAL encapsulation must be used. Underlying 1293 interfaces are further selected based on their associated preference 1294 values "high", "medium", "low" or "disabled". 1296 Note: the OMNI option is distinct from any Source/Target Link-Layer 1297 Address Options (S/TLLAOs) that may appear in an ND message according 1298 to the appropriate IPv6 over specific link layer specification (e.g., 1299 [RFC2464]). If both an OMNI option and S/TLLAO appear, the former 1300 pertains to encapsulation addresses while the latter pertains to the 1301 native L2 address format of the underlying media. 1303 3.5.2. OMNI Neighbor Advertisement Message Flags 1305 As discussed in Section 4.4 of [RFC4861] NA messages include three 1306 flag bits R, S and O. OMNI interface NA messages treat the flags as 1307 follows: 1309 o R: The R ("Router") flag is set to 1 in the NA messages sent by 1310 all AERO/OMNI node types. Simple hosts that would set R to 0 do 1311 not occur on the OMNI link itself, but may occur on the downstream 1312 links of Clients and Relays. 1314 o S: The S ("Solicited") flag is set exactly as specified in 1315 Section 4.4. of [RFC4861], i.e., it is set to 1 for Solicited NAs 1316 and set to 0 for Unsolicited NAs (both unicast and multicast). 1318 o O: The O ("Override") flag is set to 0 for solicited proxy NAs and 1319 set to 1 for all other solicited and unsolicited NAs. For further 1320 study is whether solicited NAs for anycast targets apply for OMNI 1321 links. Since OMNI LLAs must be uniquely assigned to Clients to 1322 support correct ND protocol operation, however, no role is 1323 currently seen for assigning the same OMNI LLA to multiple 1324 Clients. 1326 3.6. OMNI Interface Encapsulation and Re-encapsulation 1328 The OMNI Adaptation Layer (OAL) inserts mid-layer IPv6 headers known 1329 as the OAL/ORH headers when necessary as discussed in the following 1330 sections. After either inserting or omitting the OAL/ORH headers, 1331 the OMNI interface also inserts or omits an outer encapsulation 1332 header as discussed below. 1334 OMNI interfaces avoid outer encapsulation over Direct underlying 1335 interfaces and ANET underlying interfaces for which the Client and 1336 Proxy are connected to the same underlying link. Otherwise, OMNI 1337 interfaces encapsulate packets according to whether they are entering 1338 the OMNI interface from the network layer or if they are being re- 1339 admitted into the same OMNI link they arrived on. This latter form 1340 of encapsulation is known as "re-encapsulation". 1342 For packets entering the OMNI interface from the network layer, the 1343 OMNI interface copies the "TTL/Hop Limit", "Type of Service/Traffic 1344 Class" [RFC2983], "Flow Label"[RFC6438] (for IPv6) and "Congestion 1345 Experienced" [RFC3168] values in the inner packet's IP header into 1346 the corresponding fields in the OAL and outer encapsulation 1347 header(s). 1349 For packets undergoing re-encapsulation, the OMNI interface instead 1350 copies these values from the original encapsulation header into the 1351 new encapsulation header, i.e., the values are transferred between 1352 encapsulation headers and *not* copied from the encapsulated packet's 1353 network-layer header. (Note especially that by copying the TTL/Hop 1354 Limit between encapsulation headers the value will eventually 1355 decrement to 0 if there is a (temporary) routing loop.) 1357 OMNI interfaces configured over ANET underlying interfaces which 1358 employ a different IP protocol version (and/or when the Client and 1359 Proxy may be separated by multiple ANET IP hops) use IP-in-IP 1360 encapsulation so that the inner packet can traverse the ANET without 1361 decrementing the TTL/Hop-Limit. IPv6 underlying ANET interfaces use 1362 [RFC2473] encapsulation, while IPv4 interfaces use the appropriate 1363 encapsulation per one of [RFC5214][RFC2003]. 1365 OMNI interfaces configured over INET underlying interfaces 1366 encapsulate packets in INET headers according to the next hop 1367 determined in the forwarding algorithm in Section 3.10. If the next 1368 hop is reached via a secured tunnel, the OMNI interface uses an 1369 encapsulation format specific to the secured tunnel type (see: 1370 Section 6). If the next hop is reached via an unsecured INET 1371 interface, the OMNI interface instead uses UDP/IP encapsulation per 1372 [RFC4380] and as extended in [RFC6081]. 1374 When UDP/IP encapsulation is used, the OMNI interface next sets the 1375 UDP source port to a constant value that it will use in each 1376 successive packet it sends, and sets the UDP length field to the 1377 length of the encapsulated packet plus 8 bytes for the UDP header 1378 itself plus the length of any included extension headers or trailers. 1379 The encapsulated packet may be either IPv6 or IPv4, as distinguished 1380 by the version number found in the first four bits. 1382 For UDP/IP-encapsulated packets sent to a Server, Relay or Bridge, 1383 the OMNI interface sets the UDP destination port to 8060, i.e., the 1384 IANA-registered port number for AERO. For packets sent to a Client, 1385 the OMNI interface sets the UDP destination port to the port value 1386 stored in the neighbor cache entry for this Client. The OMNI 1387 interface finally includes/omits the UDP checksum according to 1388 [RFC6935][RFC6936]. 1390 3.7. OMNI Interface Decapsulation 1392 OMNI interfaces decapsulate packets destined either to the AERO node 1393 itself or to a destination reached via an interface other than the 1394 OMNI interface the packet was received on. When the encapsulated 1395 packet arrives in multiple OAL fragments, the OMNI interface 1396 reassembles as discussed in Section 3.9. Further decapsulation steps 1397 are performed according to the appropriate encapsulation format 1398 specification. 1400 3.8. OMNI Interface Data Origin Authentication 1402 AERO nodes employ simple data origin authentication procedures. In 1403 particular: 1405 o AERO Bridges, Servers and Proxys accept encapsulated data packets 1406 and control messages received from the (secured) spanning tree. 1408 o AERO Proxys and Clients accept packets that originate from within 1409 the same secured ANET. 1411 o AERO Clients and Relays accept packets from downstream network 1412 correspondents based on ingress filtering. 1414 o AERO Clients, Relays and Servers verify the outer UDP/IP 1415 encapsulation addresses according to [RFC4380]. 1417 AERO nodes silently drop any packets that do not satisfy the above 1418 data origin authentication procedures. Further security 1419 considerations are discussed in Section 6. 1421 3.9. OMNI Adaptation Layer and OMNI Interface MTU 1423 The OMNI interface observes the link nature of tunnels, including the 1424 Maximum Transmission Unit (MTU), Maximum Reassembly Unit (MRU) and 1425 the role of fragmentation and reassembly [I-D.ietf-intarea-tunnels]. 1426 The OMNI interface employs an OMNI Adaptation Layer (OAL) for 1427 accommodating multiple underlying links with diverse MTUs. The 1428 functions of the OAL and the OMNI interface MTU/MRU are specified in 1429 Section 5 of [I-D.templin-6man-omni-interface], with MTU/MRU both set 1430 to the constant value 9180 bytes. 1432 3.10. OMNI Interface Forwarding Algorithm 1434 IP packets enter a node's OMNI interface either from the network 1435 layer (i.e., from a local application or the IP forwarding system) or 1436 from the link layer (i.e., from an OMNI interface neighbor). All 1437 packets entering a node's OMNI interface first undergo data origin 1438 authentication as discussed in Section 3.8. Packets that satisfy 1439 data origin authentication are processed further, while all others 1440 are dropped silently. The OMNI interface OAL wraps accepted packets 1441 in OAL/ORH headers if necessary as discussed above. 1443 Packets that enter the OMNI interface from the network layer are 1444 forwarded to an OMNI interface neighbor. Packets that enter the OMNI 1445 interface from the link layer are either re-admitted into the OMNI 1446 link or forwarded to the network layer where they are subject to 1447 either local delivery or IP forwarding. In all cases, the OMNI 1448 interface itself MUST NOT decrement the network layer TTL/Hop-count 1449 since its forwarding actions occur below the network layer. 1451 OMNI interfaces may have multiple underlying interfaces and/or 1452 neighbor cache entries for neighbors with multiple underlying 1453 interfaces (see Section 3.3). The OMNI interface uses interface 1454 attributes and/or traffic classifiers (e.g., DSCP value, port number, 1455 etc.) to select an outgoing underlying interface for each packet 1456 based on the node's own QoS preferences, and also to select a 1457 destination link-layer address based on the neighbor's underlying 1458 interface with the highest preference. AERO implementations SHOULD 1459 allow for QoS preference values to be modified at runtime through 1460 network management. 1462 If multiple outgoing interfaces and/or neighbor interfaces have a 1463 preference of "high", the AERO node replicates the packet and sends 1464 one copy via each of the (outgoing / neighbor) interface pairs; 1465 otherwise, the node sends a single copy of the packet via an 1466 interface with the highest preference. AERO nodes keep track of 1467 which underlying interfaces are currently "reachable" or 1468 "unreachable", and only use "reachable" interfaces for forwarding 1469 purposes. 1471 The following sections discuss the OMNI interface forwarding 1472 algorithms for Clients, Proxys, Servers and Bridges. In the 1473 following discussion, a packet's destination address is said to 1474 "match" if it is the same as a cached address, or if it is covered by 1475 a cached prefix (which may be encoded in an LLA). 1477 3.10.1. Client Forwarding Algorithm 1479 When an IP packet enters a Client's OMNI interface from the network 1480 layer the Client searches for an asymmetric neighbor cache entry that 1481 matches the destination. If there is a match, the Client uses one or 1482 more "reachable" neighbor interfaces in the entry for packet 1483 forwarding. If there is no asymmetric neighbor cache entry, the 1484 Client instead forwards the packet toward a Server (the packet is 1485 intercepted by a Proxy if there is a Proxy on the path). The Client 1486 encapsulates the packet in OAL/ORH headers if necessary and fragments 1487 according to MTU requirements (see: Section 3.9). 1489 When an IP packet enters a Client's OMNI interface from the link- 1490 layer, if the destination matches one of the Client's MNPs or link- 1491 local addresses the Client reassembles and decapsulates as necessary 1492 and delivers the inner packet to the network layer. Otherwise, the 1493 Client drops the packet and MAY return a network-layer ICMP 1494 Destination Unreachable message subject to rate limiting (see: 1495 Section 3.11). 1497 3.10.2. Proxy Forwarding Algorithm 1499 For control messages originating from or destined to a Client, the 1500 Proxy intercepts the message and updates its proxy neighbor cache 1501 entry for the Client. The Proxy then forwards a (proxyed) copy of 1502 the control message. (For example, the Proxy forwards a proxied 1503 version of a Client's NS/RS message to the target neighbor, and 1504 forwards a proxied version of the NA/RA reply to the Client.) 1506 When the Proxy receives a data packet from a Client within the ANET, 1507 the Proxy reassembles and re-fragments if necessary then searches for 1508 an asymmetric neighbor cache entry that matches the destination and 1509 forwards as follows: 1511 o if the destination matches an asymmetric neighbor cache entry, the 1512 Proxy uses one or more "reachable" neighbor interfaces in the 1513 entry for packet forwarding using OAL/ORH encapsulation if 1514 necessary according to the cached link-layer address information. 1515 If the neighbor interface is in the same OMNI link segment, the 1516 Proxy forwards the packet directly to the neighbor; otherwise, it 1517 forwards the packet to a Bridge. 1519 o else, the Proxy uses OAL/ORH encapsulation and forwards the packet 1520 to a Bridge while using the DLA corresponding to the packet's 1521 destination as the destination address. 1523 When the Proxy receives an encapsulated data packet from an INET 1524 neighbor or from a secured tunnel from a Bridge, it accepts the 1525 packet only if data origin authentication succeeds and if there is a 1526 proxy neighbor cache entry that matches the inner destination. Next, 1527 the Proxy reassembles the packet (if necessary) and continues 1528 processing. If the reassembly is complete and the neighbor cache 1529 state is REACHABLE, the Proxy then returns a PTB if necessary (see: 1530 Section 3.9) then either drops or forwards the packet to the Client 1531 while performing OAL/ORH encapsulation and re-fragmentation if 1532 necessary. If the neighbor cache entry state is DEPARTED, the Proxy 1533 instead changes the destination address to the address of the new 1534 Server and forwards it to a Bridge while performing OAL/ORH re- 1535 fragmentation if necessary. 1537 3.10.3. Server/Relay Forwarding Algorithm 1539 For control messages destined to a target Client's LLA that are 1540 received from a secured tunnel, the Server intercepts the message and 1541 sends a Proxyed response on behalf of the Client. (For example, the 1542 Server sends a Proxyed NA message reply in response to an NS message 1543 directed to one of its associated Clients.) If the Client's neighbor 1544 cache entry is in the DEPARTED state, however, the Server instead 1545 forwards the packet to the Client's new Server as discussed in 1546 Section 3.16. 1548 When the Server receives an encapsulated data packet from an INET 1549 neighbor or from a secured tunnel, it accepts the packet only if data 1550 origin authentication succeeds. The Server then continues processing 1551 as follows: 1553 o if the network layer destination matches a symmetric neighbor 1554 cache entry in the REACHABLE state the Server prepares the packet 1555 for forwarding to the destination Client. The Server first 1556 reassembles (if necessary) and forwards the packet (while re- 1557 fragmenting if necessary) as specified in Section 3.9. 1559 o else, if the destination matches a symmetric neighbor cache entry 1560 in the DEPARETED state the Server re-encapsulates the packet and 1561 forwards it using the DLA of the Client's new Server as the 1562 destination. 1564 o else, if the destination matches an asymmetric neighbor cache 1565 entry, the Server uses one or more "reachable" neighbor interfaces 1566 in the entry for packet forwarding via the local INET if the 1567 neighbor is in the same OMNI link segment or using OAL/ORH 1568 encapsulation if necessary with the final destination set to the 1569 neighbor's DLA otherwise. 1571 o else, if the destination matches a non-MNP route in the IP 1572 forwarding table or an LLA assigned to the Server's OMNI 1573 interface, the Server reassembles if necessary, decapsulates the 1574 packet and releases it to the network layer for local delivery or 1575 IP forwarding. 1577 o else, the Server drops the packet. 1579 When the Server's OMNI interface receives a data packet from the 1580 network layer or from a VPNed or Direct Client, it performs OAL/ORH 1581 encapsulation and fragmentation if necessary, then processes the 1582 packet according to the network-layer destination address as follows: 1584 o if the destination matches a symmetric or asymmetric neighbor 1585 cache entry the Server processes the packet as above. 1587 o else, the Server encapsulates the packet in OLA/ORH headers and 1588 forwards it to a Bridge using its own DLA as the source and the 1589 DLA corresponding to the destination as the destination. 1591 3.10.4. Bridge Forwarding Algorithm 1593 Bridges forward OAL/ORH-encapsulated packets over secured tunnels the 1594 same as any IP router. When the Bridge receives an OAL/ORH- 1595 encapsulated packet via a secured tunnel, it removes the outer INET 1596 header and searches for a forwarding table entry that matches the 1597 destination address. The Bridge then processes the packet as 1598 follows: 1600 o if the destination matches its DLA Subnet Router Anycast address, 1601 the Bridge checks for an ORH. If there is an ORH with SRT/LHS 1602 located on the local segment, the Bridge removes the ORH from the 1603 packet. If the ORH includes a Destination Suffix the Bridge also 1604 writes the DLA formed from the Destination Suffix into the OAL 1605 header destination address; otherwise, it writes the DLA formed 1606 from the SRT/LHS values into the OAL header destination. Next, 1607 the Bridge examines the FMT to determine if the target is behind a 1608 NAT. If no NAT is indicated, the Bridge forwards the packet 1609 directly to the L2ADDR using link-layer (UDP/IP) encapsulation. 1610 If a NAT is indicated, the Bridge MAY perform NAT traversal 1611 procedures by sending bubbles per [RFC4380]. The Bridge then 1612 either applies AERO route optimization if NAT traversal procedures 1613 have been successfully applied, or forwards the packet directly to 1614 the Server. 1616 o if the destination matches one of the Bridge's own addresses, the 1617 Bridge submits the packet for local delivery. 1619 o else, if the destination matches a forwarding table entry the 1620 Bridge forwards the packet via a secured tunnel to the next hop. 1622 If the destination matches an MSP without matching an MNP, 1623 however, the Bridge instead drops the packet and returns an ICMP 1624 Destination Unreachable message subject to rate limiting (see: 1625 Section 3.11). 1627 o else, the Bridge drops the packet and returns an ICMP Destination 1628 Unreachable as above. 1630 As for any IP router, the Bridge decrements the TTL/Hop Limit when it 1631 forwards the packet. Therefore, when an OAL header is present only 1632 the Hop Limit in the OAL header is decremented and not the TTL/Hop 1633 Limit in the inner packet header. Bridges do not insert OAL/ORH 1634 headers themselves; instead, they act as IPv6 routers and forward 1635 packets based on the destination address found in the headers of 1636 packets they receive. 1638 3.11. OMNI Interface Error Handling 1640 When an AERO node admits a packet into the OMNI interface, it may 1641 receive link-layer or network-layer error indications. 1643 A link-layer error indication is an ICMP error message generated by a 1644 router in the INET on the path to the neighbor or by the neighbor 1645 itself. The message includes an IP header with the address of the 1646 node that generated the error as the source address and with the 1647 link-layer address of the AERO node as the destination address. 1649 The IP header is followed by an ICMP header that includes an error 1650 Type, Code and Checksum. Valid type values include "Destination 1651 Unreachable", "Time Exceeded" and "Parameter Problem" 1652 [RFC0792][RFC4443]. (OMNI interfaces ignore all link-layer IPv4 1653 "Fragmentation Needed" and IPv6 "Packet Too Big" messages since they 1654 only emit packets that are guaranteed to be no larger than the IP 1655 minimum link MTU as discussed in Section 3.9.) 1657 The ICMP header is followed by the leading portion of the packet that 1658 generated the error, also known as the "packet-in-error". For 1659 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1660 much of invoking packet as possible without the ICMPv6 packet 1661 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1662 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1663 "Internet Header + 64 bits of Original Data Datagram", however 1664 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1665 ICMP datagram SHOULD contain as much of the original datagram as 1666 possible without the length of the ICMP datagram exceeding 576 1667 bytes". 1669 The link-layer error message format is shown in Figure 5 (where, "L2" 1670 and "L3" refer to link-layer and network-layer, respectively): 1672 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1673 ~ ~ 1674 | L2 IP Header of | 1675 | error message | 1676 ~ ~ 1677 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1678 | L2 ICMP Header | 1679 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1680 ~ ~ P 1681 | IP and other encapsulation | a 1682 | headers of original L3 packet | c 1683 ~ ~ k 1684 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1685 ~ ~ t 1686 | IP header of | 1687 | original L3 packet | i 1688 ~ ~ n 1689 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1690 ~ ~ e 1691 | Upper layer headers and | r 1692 | leading portion of body | r 1693 | of the original L3 packet | o 1694 ~ ~ r 1695 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1697 Figure 5: OMNI Interface Link-Layer Error Message Format 1699 The AERO node rules for processing these link-layer error messages 1700 are as follows: 1702 o When an AERO node receives a link-layer Parameter Problem message, 1703 it processes the message the same as described as for ordinary 1704 ICMP errors in the normative references [RFC0792][RFC4443]. 1706 o When an AERO node receives persistent link-layer Time Exceeded 1707 messages, the IP ID field may be wrapping before earlier fragments 1708 awaiting reassembly have been processed. In that case, the node 1709 should begin including integrity checks and/or institute rate 1710 limits for subsequent packets. 1712 o When an AERO node receives persistent link-layer Destination 1713 Unreachable messages in response to encapsulated packets that it 1714 sends to one of its asymmetric neighbor correspondents, the node 1715 should process the message as an indication that a path may be 1716 failing, and optionally initiate NUD over that path. If it 1717 receives Destination Unreachable messages over multiple paths, the 1718 node should allow future packets destined to the correspondent to 1719 flow through a default route and re-initiate route optimization. 1721 o When an AERO Client receives persistent link-layer Destination 1722 Unreachable messages in response to encapsulated packets that it 1723 sends to one of its symmetric neighbor Servers, the Client should 1724 mark the path as unusable and use another path. If it receives 1725 Destination Unreachable messages on many or all paths, the Client 1726 should associate with a new Server and release its association 1727 with the old Server as specified in Section 3.16.5. 1729 o When an AERO Server receives persistent link-layer Destination 1730 Unreachable messages in response to encapsulated packets that it 1731 sends to one of its symmetric neighbor Clients, the Server should 1732 mark the underlying path as unusable and use another underlying 1733 path. 1735 o When an AERO Server or Proxy receives link-layer Destination 1736 Unreachable messages in response to an encapsulated packet that it 1737 sends to one of its permanent neighbors, it treats the messages as 1738 an indication that the path to the neighbor may be failing. 1739 However, the dynamic routing protocol should soon reconverge and 1740 correct the temporary outage. 1742 When an AERO Bridge receives a packet for which the network-layer 1743 destination address is covered by an MSP, if there is no more- 1744 specific routing information for the destination the Bridge drops the 1745 packet and returns a network-layer Destination Unreachable message 1746 subject to rate limiting. The Bridge writes the network-layer source 1747 address of the original packet as the destination address and uses 1748 one of its non link-local addresses as the source address of the 1749 message. 1751 When an AERO node receives an encapsulated packet for which the 1752 reassembly buffer it too small, it drops the packet and returns a 1753 network-layer Packet Too Big (PTB) message. The node first writes 1754 the MRU value into the PTB message MTU field, writes the network- 1755 layer source address of the original packet as the destination 1756 address and writes one of its non link-local addresses as the source 1757 address. 1759 3.12. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1761 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1762 coordinated as discussed in the following Sections. 1764 3.12.1. AERO Service Model 1766 Each AERO Server on the OMNI link is configured to facilitate Client 1767 prefix delegation/registration requests. Each Server is provisioned 1768 with a database of MNP-to-Client ID mappings for all Clients enrolled 1769 in the AERO service, as well as any information necessary to 1770 authenticate each Client. The Client database is maintained by a 1771 central administrative authority for the OMNI link and securely 1772 distributed to all Servers, e.g., via the Lightweight Directory 1773 Access Protocol (LDAP) [RFC4511], via static configuration, etc. 1774 Clients receive the same service regardless of the Servers they 1775 select. 1777 AERO Clients and Servers use ND messages to maintain neighbor cache 1778 entries. AERO Servers configure their OMNI interfaces as advertising 1779 NBMA interfaces, and therefore send unicast RA messages with a short 1780 Router Lifetime value (e.g., ReachableTime seconds) in response to a 1781 Client's RS message. Thereafter, Clients send additional RS messages 1782 to keep Server state alive. 1784 AERO Clients and Servers include prefix delegation and/or 1785 registration parameters in RS/RA messages (see 1786 [I-D.templin-6man-omni-interface]). The ND messages are exchanged 1787 between Client and Server according to the prefix management schedule 1788 required by the service. If the Client knows its MNP in advance, it 1789 can employ prefix registration by including its LLA as the source 1790 address of an RS message and with an OMNI option with valid prefix 1791 registration information for the MNP. If the Server (and Proxy) 1792 accept the Client's MNP assertion, they inject the prefix into the 1793 routing system and establish the necessary neighbor cache state. 1795 The following sections specify the Client and Server behavior. 1797 3.12.2. AERO Client Behavior 1799 AERO Clients discover the addresses of Servers in a similar manner as 1800 described in [RFC5214]. Discovery methods include static 1801 configuration (e.g., from a flat-file map of Server addresses and 1802 locations), or through an automated means such as Domain Name System 1803 (DNS) name resolution [RFC1035]. Alternatively, the Client can 1804 discover Server addresses through a layer 2 data link login exchange, 1805 or through a unicast RA response to a multicast/anycast RS as 1806 described below. In the absence of other information, the Client can 1807 resolve the DNS Fully-Qualified Domain Name (FQDN) 1808 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 1809 text string and "[domainname]" is a DNS suffix for the OMNI link 1810 (e.g., "example.com"). 1812 To associate with a Server, the Client acts as a requesting router to 1813 request MNPs. The Client prepares an RS message with prefix 1814 management parameters and includes a Nonce and Timestamp option if 1815 the Client needs to correlate RA replies. If the Client already 1816 knows the Server's LLA, it includes the LLA as the network-layer 1817 destination address; otherwise, it includes (link-local) All-Routers 1818 multicast as the network-layer destination. If the Client already 1819 knows its own LLA, it uses the LLA as the network-layer source 1820 address; otherwise, it uses an OMNI Temporary LLA as the network- 1821 layer source address and includes a DHCP Unique Identifier (DUID) 1822 sub-option in the OMNI option (see: 1823 [I-D.templin-6man-omni-interface]). 1825 The Client next includes an OMNI option in the RS message to register 1826 its link-layer information with the Server. The Client sets the OMNI 1827 option prefix registration information according to the MNP, and 1828 includes Interface Attributes corresponding to the underlying 1829 interface over which the Client will send the RS message. The Client 1830 MAY include additional Interface Attributes specific to other 1831 underlying interfaces. 1833 The Client then sends the RS message (either directly via Direct 1834 interfaces, via a VPN for VPNed interfaces, via a Proxy for ANET 1835 interfaces or via INET encapsulation for INET interfaces) and waits 1836 for an RA message reply (see Section 3.12.3). The Client retries up 1837 to MAX_RTR_SOLICITATIONS times until an RA is received. If the 1838 Client receives no RAs, or if it receives an RA with Router Lifetime 1839 set to 0, the Client SHOULD abandon this Server and try another 1840 Server. Otherwise, the Client processes the prefix information found 1841 in the RA message. 1843 Next, the Client creates a symmetric neighbor cache entry with the 1844 Server's LLA as the network-layer address and the Server's 1845 encapsulation and/or link-layer addresses as the link-layer address. 1846 The Client records the RA Router Lifetime field value in the neighbor 1847 cache entry as the time for which the Server has committed to 1848 maintaining the MNP in the routing system via this underlying 1849 interface, and caches the other RA configuration information 1850 including Cur Hop Limit, M and O flags, Reachable Time and Retrans 1851 Timer. The Client then autoconfigures LLAs for each of the delegated 1852 MNPs and assigns them to the OMNI interface. The Client also caches 1853 any MSPs included in Route Information Options (RIOs) [RFC4191] as 1854 MSPs to associate with the OMNI link, and assigns the MTU value in 1855 the MTU option to the underlying interface. 1857 The Client then registers additional underlying interfaces with the 1858 Server by sending RS messages via each additional interface. The RS 1859 messages include the same parameters as for the initial RS/RA 1860 exchange, but with destination address set to the Server's LLA. 1862 Following autoconfiguration, the Client sub-delegates the MNPs to its 1863 attached EUNs and/or the Client's own internal virtual interfaces as 1864 described in [I-D.templin-v6ops-pdhost] to support the Client's 1865 downstream attached "Internet of Things (IoT)". The Client 1866 subsequently sends additional RS messages over each underlying 1867 interface before the Router Lifetime received for that interface 1868 expires. 1870 After the Client registers its underlying interfaces, it may wish to 1871 change one or more registrations, e.g., if an interface changes 1872 address or becomes unavailable, if QoS preferences change, etc. To 1873 do so, the Client prepares an RS message to send over any available 1874 underlying interface. The RS includes an OMNI option with prefix 1875 registration information specific to its MNP, with Interface 1876 Attributes specific to the selected underlying interface, and with 1877 any additional Interface Attributes specific to other underlying 1878 interfaces. When the Client receives the Server's RA response, it 1879 has assurance that the Server has been updated with the new 1880 information. 1882 If the Client wishes to discontinue use of a Server it issues an RS 1883 message over any underlying interface with an OMNI option with a 1884 prefix release indication. When the Server processes the message, it 1885 releases the MNP, sets the symmetric neighbor cache entry state for 1886 the Client to DEPARTED and returns an RA reply with Router Lifetime 1887 set to 0. After a short delay (e.g., 2 seconds), the Server 1888 withdraws the MNP from the routing system. 1890 3.12.3. AERO Server Behavior 1892 AERO Servers act as IP routers and support a prefix delegation/ 1893 registration service for Clients. Servers arrange to add their LLAs 1894 to a static map of Server addresses for the link and/or the DNS 1895 resource records for the FQDN "linkupnetworks.[domainname]" before 1896 entering service. Server addresses should be geographically and/or 1897 topologically referenced, and made available for discovery by Clients 1898 on the OMNI link. 1900 When a Server receives a prospective Client's RS message on its OMNI 1901 interface, it SHOULD return an immediate RA reply with Router 1902 Lifetime set to 0 if it is currently too busy or otherwise unable to 1903 service the Client. Otherwise, the Server authenticates the RS 1904 message and processes the prefix delegation/registration parameters. 1905 The Server first determines the correct MNPs to provide to the Client 1906 by searching the Client database. When the Server returns the MNPs, 1907 it also creates a forwarding table entry for the DLA corresponding to 1908 each MNP so that the MNPs are propagated into the routing system 1909 (see: Section 3.2.3). For IPv6, the Server creates an IPv6 1910 forwarding table entry for each MNP. For IPv4, the Server creates an 1911 IPv6 forwarding table entry with the IPv4-compatibility DLA prefix 1912 corresponding to the IPv4 address. 1914 The Server next creates a symmetric neighbor cache entry for the 1915 Client using the base LLA as the network-layer address and with 1916 lifetime set to no more than the smallest prefix lifetime. Next, the 1917 Server updates the neighbor cache entry by recording the information 1918 in each Interface Attributes sub-option in the RS OMNI option. The 1919 Server also records the actual OAL/INET addresses in the neighbor 1920 cache entry. 1922 Next, the Server prepares an RA message using its LLA as the network- 1923 layer source address and the network-layer source address of the RS 1924 message as the network-layer destination address. The Server sets 1925 the Router Lifetime to the time for which it will maintain both this 1926 underlying interface individually and the symmetric neighbor cache 1927 entry as a whole. The Server also sets Cur Hop Limit, M and O flags, 1928 Reachable Time and Retrans Timer to values appropriate for the OMNI 1929 link. The Server includes the MNPs, any other prefix management 1930 parameters and an OMNI option with no Interface Attributes. The 1931 Server then includes one or more RIOs that encode the MSPs for the 1932 OMNI link, plus an MTU option (see Section 3.9). The Server finally 1933 forwards the message to the Client using OAL/INET, INET, or NULL 1934 encapsulation as necessary. 1936 After the initial RS/RA exchange, the Server maintains a 1937 ReachableTime timer for each of the Client's underlying interfaces 1938 individually (and for the Client's symmetric neighbor cache entry 1939 collectively) set to expire after ReachableTime seconds. If the 1940 Client (or Proxy) issues additional RS messages, the Server sends an 1941 RA response and resets ReachableTime. If the Server receives an ND 1942 message with a prefix release indication it sets the Client's 1943 symmetric neighbor cache entry to the DEPARTED state and withdraws 1944 the MNP from the routing system after a short delay (e.g., 2 1945 seconds). If ReachableTime expires before a new RS is received on an 1946 individual underlying interface, the Server marks the interface as 1947 DOWN. If ReachableTime expires before any new RS is received on any 1948 individual underlying interface, the Server sets the symmetric 1949 neighbor cache entry state to STALE and sets a 10 second timer. If 1950 the Server has not received a new RS or ND message with a prefix 1951 release indication before the 10 second timer expires, it deletes the 1952 neighbor cache entry and withdraws the MNP from the routing system. 1954 The Server processes any ND messages pertaining to the Client and 1955 returns an NA/RA reply in response to solicitations. The Server may 1956 also issue unsolicited RA messages, e.g., with reconfigure parameters 1957 to cause the Client to renegotiate its prefix delegation/ 1958 registrations, with Router Lifetime set to 0 if it can no longer 1959 service this Client, etc. Finally, If the symmetric neighbor cache 1960 entry is in the DEPARTED state, the Server deletes the entry after 1961 DepartTime expires. 1963 Note: Clients SHOULD notify former Servers of their departures, but 1964 Servers are responsible for expiring neighbor cache entries and 1965 withdrawing routes even if no departure notification is received 1966 (e.g., if the Client leaves the network unexpectedly). Servers 1967 SHOULD therefore set Router Lifetime to ReachableTime seconds in 1968 solicited RA messages to minimize persistent stale cache information 1969 in the absence of Client departure notifications. A short Router 1970 Lifetime also ensures that proactive Client/Server RS/RA messaging 1971 will keep any NAT state alive (see above). 1973 Note: All Servers on an OMNI link MUST advertise consistent values in 1974 the RA Cur Hop Limit, M and O flags, Reachable Time and Retrans Timer 1975 fields the same as for any link, since unpredictable behavior could 1976 result if different Servers on the same link advertised different 1977 values. 1979 3.12.3.1. DHCPv6-Based Prefix Registration 1981 When a Client is not pre-provisioned with an OMNI LLA containing a 1982 MNP, it will need for the Server to select one or more MNPs on its 1983 behalf and set up the correct state in the AERO routing service. (A 1984 Client with a pre-provisioned MNP may also request the Server to 1985 select additional MNPs.) The DHCPv6 service [RFC8415] is used to 1986 support this requirement. 1988 When a Client needs to have the Server select MNPs, it sends an RS 1989 message with an OMNI option that includes a DHCPv6 message suboption 1990 with DHCPv6 Prefix Delegation (DHCPv6-PD) parameters. When the 1991 Server receives the RS message, it extracts the DHCPv6-PD message 1992 from the OMNI option. 1994 The Server then acts as a "Proxy DHCPv6 Client" in a message exchange 1995 with the locally-resident DHCPv6 server, which delegates MNPs and 1996 returns a DHCPv6-PD Reply message. (If the Server wishes to defer 1997 creation of MN state until the DHCPv6-PD Reply is received, it can 1998 instead act as a Lightweight DHCPv6 Relay Agent per [RFC6221] by 1999 encapsulating the DHCPv6-PD message in a Relay-forward/reply exchange 2000 with Relay Message and Interface ID options.) 2001 When the Server receives the DHCPv6-PD Reply, it adds a route to the 2002 routing system and creates an OMNI MN LLA based on the delegated MNP. 2003 The Server then sends an RA back to the Client with the (newly- 2004 created) OMNI MN LLA as the destination address and with the 2005 DHCPv6-PD Reply message coded in the OMNI option. When the Client 2006 receives the RA, it creates a default route, assigns the Subnet 2007 Router Anycast address and sets its OMNI LLA based on the delegated 2008 MNP. 2010 3.13. The AERO Proxy 2012 Clients may connect to protected-spectrum ANETs that employ physical 2013 and/or link-layer security services to facilitate communications to 2014 Servers in outside INETs. In that case, the ANET can employ an AERO 2015 Proxy. The Proxy is located at the ANET/INET border and listens for 2016 RS messages originating from or RA messages destined to ANET Clients. 2017 The Proxy acts on these control messages as follows: 2019 o when the Proxy receives an RS message from a new ANET Client, it 2020 first authenticates the message then examines the network-layer 2021 destination address. If the destination address is a Server's 2022 LLA, the Proxy proceeds to the next step. Otherwise, if the 2023 destination is (link-local) All-Routers multicast, the Proxy 2024 selects a "nearby" Server that is likely to be a good candidate to 2025 serve the Client and replaces the destination address with the 2026 Server's LLA. Next, the Proxy creates a proxy neighbor cache 2027 entry and caches the Client and Server link-layer addresses along 2028 with the OMNI option information and any other identifying 2029 information including Transaction IDs, Client Identifiers, Nonce 2030 values, etc. The Proxy finally encapsulates the (proxyed) RS 2031 message in an OAL header with source set to the Proxy's DLA and 2032 destination set to the Server's DLA. The Proxy also includes an 2033 OMNI header with an Interface Attributes option that includes its 2034 own INET address plus a unique Port Number for this Client, then 2035 forwards the message into the OMNI link spanning tree. 2037 o when the Server receives the RS, it authenticates the message then 2038 creates or updates a symmetric neighbor cache entry for the Client 2039 with the Proxy's DLA, INET address and Port Number as the link- 2040 layer address information. The Server then sends an RA message 2041 back to the Proxy via the spanning tree. 2043 o when the Proxy receives the RA, it authenticates the message and 2044 matches it with the proxy neighbor cache entry created by the RS. 2045 The Proxy then caches the prefix information as a mapping from the 2046 Client's MNPs to the Client's link-layer address, caches the 2047 Server's advertised Router Lifetime and sets the neighbor cache 2048 entry state to REACHABLE. The Proxy then optionally rewrites the 2049 Router Lifetime and forwards the (proxyed) message to the Client. 2050 The Proxy finally includes an MTU option (if necessary) with an 2051 MTU to use for the underlying ANET interface. 2053 After the initial RS/RA exchange, the Proxy forwards any Client data 2054 packets for which there is no matching asymmetric neighbor cache 2055 entry to a Bridge using OAL encapsulation with its own DLA as the 2056 source and the DLA corresponding to the Client as the destination. 2057 The Proxy instead forwards any Client data destined to an asymmetric 2058 neighbor cache target directly to the target according to the OAL/ 2059 link-layer information - the process of establishing asymmetric 2060 neighbor cache entries is specified in Section 3.14. 2062 While the Client is still attached to the ANET, the Proxy sends NS, 2063 RS and/or unsolicited NA messages to update the Server's symmetric 2064 neighbor cache entries on behalf of the Client and/or to convey QoS 2065 updates. This allows for higher-frequency Proxy-initiated RS/RA 2066 messaging over well-connected INET infrastructure supplemented by 2067 lower-frequency Client-initiated RS/RA messaging over constrained 2068 ANET data links. 2070 If the Server ceases to send solicited advertisements, the Proxy 2071 sends unsolicited RAs on the ANET interface with destination set to 2072 (link-local) All-Nodes multicast and with Router Lifetime set to zero 2073 to inform Clients that the Server has failed. Although the Proxy 2074 engages in ND exchanges on behalf of the Client, the Client can also 2075 send ND messages on its own behalf, e.g., if it is in a better 2076 position than the Proxy to convey QoS changes, etc. For this reason, 2077 the Proxy marks any Client-originated solicitation messages (e.g. by 2078 inserting a Nonce option) so that it can return the solicited 2079 advertisement to the Client instead of processing it locally. 2081 If the Client becomes unreachable, the Proxy sets the neighbor cache 2082 entry state to DEPARTED and retains the entry for DepartTime seconds. 2083 While the state is DEPARTED, the Proxy forwards any packets destined 2084 to the Client to a Bridge via OAL encapsulation with the Client's 2085 current Server as the destination. The Bridge in turn forwards the 2086 packets to the Client's current Server. When DepartTime expires, the 2087 Proxy deletes the neighbor cache entry and discards any further 2088 packets destined to this (now forgotten) Client. 2090 In some ANETs that employ a Proxy, the Client's MNP can be injected 2091 into the ANET routing system. In that case, the Client can send data 2092 messages without encapsulation so that the ANET routing system 2093 transports the unencapsulated packets to the Proxy. This can be very 2094 beneficial, e.g., if the Client connects to the ANET via low-end data 2095 links such as some aviation wireless links. 2097 If the first-hop ANET access router is on the same underlying link 2098 and recognizes the AERO/OMNI protocol, the Client can avoid 2099 encapsulation for both its control and data messages. When the 2100 Client connects to the link, it can send an unencapsulated RS message 2101 with source address set to its LLA and with destination address set 2102 to the LLA of the Client's selected Server or to (link-local) All- 2103 Routers multicast. The Client includes an OMNI option formatted as 2104 specified in [I-D.templin-6man-omni-interface]. 2106 The Client then sends the unencapsulated RS message, which will be 2107 intercepted by the AERO-Aware access router. The access router then 2108 encapsulates the RS message in an ANET header with its own address as 2109 the source address and the address of a Proxy as the destination 2110 address. The access router further remembers the address of the 2111 Proxy so that it can encapsulate future data packets from the Client 2112 via the same Proxy. If the access router needs to change to a new 2113 Proxy, it simply sends another RS message toward the Server via the 2114 new Proxy on behalf of the Client. 2116 In some cases, the access router and Proxy may be one and the same 2117 node. In that case, the node would be located on the same physical 2118 link as the Client, but its message exchanges with the Server would 2119 need to pass through a security gateway at the ANET/INET border. The 2120 method for deploying access routers and Proxys (i.e. as a single node 2121 or multiple nodes) is an ANET-local administrative consideration. 2123 3.13.1. Combined Proxy/Servers 2125 Clients may need to connect directly to Servers via INET, Direct and 2126 VPNed interfaces (i.e., non-ANET interfaces). If the Client's 2127 underlying interfaces all connect via the same INET partition, then 2128 it can connect to a single controlling Server via all interfaces. 2130 If some Client interfaces connect via different INET partitions, 2131 however, the Client still selects a set of controlling Servers and 2132 sends RS messages via their directly-connected Servers while using 2133 the LLA of the controlling Server as the destination. 2135 When a Server receives an RS with destination set to the LLA of a 2136 controlling Server, it acts as a Proxy to forward the message to the 2137 controlling Server while forwarding the corresponding RA reply to the 2138 Client. 2140 3.13.2. Detecting and Responding to Server Failures 2142 In environments where fast recovery from Server failure is required, 2143 Proxys SHOULD use proactive Neighbor Unreachability Detection (NUD) 2144 to track Server reachability in a similar fashion as for 2145 Bidirectional Forwarding Detection (BFD) [RFC5880]. Proxys can then 2146 quickly detect and react to failures so that cached information is 2147 re-established through alternate paths. The NUD control messaging is 2148 carried only over well-connected ground domain networks (i.e., and 2149 not low-end aeronautical radio links) and can therefore be tuned for 2150 rapid response. 2152 Proxys perform proactive NUD with Servers for which there are 2153 currently active ANET Clients by sending continuous NS messages in 2154 rapid succession, e.g., one message per second. The Proxy sends the 2155 NS message via the spanning tree with the Proxy's LLA as the source 2156 and the LLA of the Server as the destination. When the Proxy is also 2157 sending RS messages to the Server on behalf of ANET Clients, the 2158 resulting RA responses can be considered as equivalent hints of 2159 forward progress. This means that the Proxy need not also send a 2160 periodic NS if it has already sent an RS within the same period. If 2161 the Server fails (i.e., if the Proxy ceases to receive 2162 advertisements), the Proxy can quickly inform Clients by sending 2163 multicast RA messages on the ANET interface. 2165 The Proxy sends RA messages on the ANET interface with source address 2166 set to the Server's address, destination address set to (link-local) 2167 All-Nodes multicast, and Router Lifetime set to 0. The Proxy SHOULD 2168 send MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small 2169 delays [RFC4861]. Any Clients on the ANET that had been using the 2170 failed Server will receive the RA messages and associate with a new 2171 Server. 2173 3.13.3. Point-to-Multipoint Server Coordination 2175 In environments where Client messaging over ANETs is bandwidth- 2176 limited and/or expensive, Clients can enlist the services of the 2177 Proxy to coordinate with multiple Servers in a single RS/RA message 2178 exchange. The Client can send a single RS message to (link-local) 2179 All-Routers multicast that includes the ID's of multiple Servers in 2180 MS-Register sub-options of the OMNI option. 2182 When the Proxy receives the RS and processes the OMNI option, it 2183 sends a separate RS to each MS-Register Server ID. When the Proxy 2184 receives an RA, it can optionally return an immediate "singleton" RA 2185 to the Client or record the Server's ID for inclusion in a pending 2186 "aggregate" RA message. The Proxy can then return aggregate RA 2187 messages to the Client including multiple Server IDs in order to 2188 conserve bandwidth. Each RA includes a proper subset of the Server 2189 IDs from the original RS message, and the Proxy must ensure that the 2190 message contents of each RA are consistent with the information 2191 received from the (aggregated) Servers. 2193 Clients can thereafter employ efficient point-to-multipoint Server 2194 coordination under the assistance of the Proxy to reduce the number 2195 of messages sent over the ANET while enlisting the support of 2196 multiple Servers for fault tolerance. Clients can further include 2197 MS-Release sub-options in IPv6 ND messages to request the Proxy to 2198 release from former Servers via the procedures discussed in 2199 Section 3.16.5. 2201 The OMNI interface specification [I-D.templin-6man-omni-interface] 2202 provides further discussion of the Client/Proxy RS/RA messaging 2203 involved in point-to-multipoint coordination. 2205 3.14. AERO Route Optimization / Address Resolution 2207 While data packets are flowing between a source and target node, 2208 route optimization SHOULD be used. Route optimization is initiated 2209 by the first eligible Route Optimization Source (ROS) closest to the 2210 source as follows: 2212 o For Clients on VPNed and Direct interfaces, the Server is the ROS. 2214 o For Clients on ANET interfaces, the Proxy is the ROS. 2216 o For Clients on INET interfaces, the Client itself is the ROS. 2218 o For correspondent nodes on INET/EUN interfaces serviced by a 2219 Relay, the Relay is the ROS. 2221 The route optimization procedure is conducted between the ROS and the 2222 target Server/Relay acting as a Route Optimization Responder (ROR) in 2223 the same manner as for IPv6 ND Address Resolution and using the same 2224 NS/NA messaging. The target may either be a MNP Client serviced by a 2225 Server, or a non-MNP correspondent reachable via a Relay. 2227 The procedures are specified in the following sections. 2229 3.14.1. Route Optimization Initiation 2231 While data packets are flowing from the source node toward a target 2232 node, the ROS performs address resolution by sending an NS message 2233 for Address Resolution (NS(AR)) to receive a solicited NA message 2234 from the ROR. When the ROS sends an NS(AR), it includes: 2236 o the LLA of the ROS as the source address. 2238 o the data packet's destination as the Target Address. 2240 o the Solicited-Node multicast address [RFC4291] formed from the 2241 lower 24 bits of the data packet's destination as the destination 2242 address, e.g., for 2001:db8:1:2::10:2000 the NS destination 2243 address is ff02:0:0:0:0:1:ff10:2000. 2245 The NS(AR) message includes an OMNI option with no Interface 2246 Attributes, such that the target will not create a neighbor cache 2247 entry. The Prefix Length in the OMNI option is set to the Prefix 2248 Length associated with the ROS's LLA. 2250 The ROS then encapsulates the NS(AR) message in an OAL header with 2251 source set to its own DLA and destination set to the DLA 2252 corresponding to the target, then sends the message into the spanning 2253 tree without decrementing the network-layer TTL/Hop Limit field. 2254 (When the ROS is a Client, it instead securely sends the NS(AR) to 2255 one of its current Servers by including an Authentication option per 2256 [RFC4380]. The Server then forwards the message into the spanning 2257 tree on behalf of the Client, while setting the IPv6 source address 2258 and the OAL source address to the LLA and DLA of the Client, 2259 respectively.) 2261 3.14.2. Relaying the NS 2263 When the Bridge receives the NS(AR) message from the ROS, it discards 2264 the INET header and determines that the ROR is the next hop by 2265 consulting its standard IPv6 forwarding table for the OAL header 2266 destination address. The Bridge then forwards the message toward the 2267 ROR via the spanning tree the same as for any IPv6 router. The 2268 final-hop Bridge in the spanning tree will deliver the message via a 2269 secured tunnel to the ROR. 2271 3.14.3. Processing the NS and Sending the NA 2273 When the ROR receives the NS(AR) message, it examines the Target 2274 Address to determine whether it has a neighbor cache entry and/or 2275 route that matches the target. If there is no match, the ROR drops 2276 the message. Otherwise, the ROR continues processing as follows: 2278 o if the target belongs to an MNP Client neighbor in the DEPARTED 2279 state the ROR changes the NS(AR) message OAL destination address 2280 to the DLA of the Client's new Server, forwards the message into 2281 the spanning tree and returns from processing. 2283 o If the target belongs to an MNP Client neighbor in the REACHABLE 2284 state, the ROR instead adds the AERO source address to the target 2285 Client's Report List with time set to ReportTime. 2287 o If the target belongs to a non-MNP route, the ROR continues 2288 processing without adding an entry to the Report List. 2290 The ROR then prepares a solicited NA message to send back to the ROS 2291 but does not create a neighbor cache entry. The ROR sets the NA 2292 source address to the LLA corresponding to the target, sets the 2293 Target Address to the target of the solicitation, and sets the 2294 destination address to the source of the solicitation. The ROR then 2295 includes an OMNI option with Prefix Length set to the length 2296 associated with the LLA. 2298 If the target is an MNP Client, the ROR next includes Interface 2299 Attributes in the OMNI option for each of the target Client's 2300 underlying interfaces with current information for each interface and 2301 with the S/T-ifIndex field in the OMNI header set to 0 to indicate 2302 that the message originated from the ROR and not the Client. 2304 For each Interface Attributes sub-option, the ROR sets the L2ADDR 2305 according to its own INET address for VPNed or Direct interfaces, to 2306 the INET address of the Proxy or to the Client's INET address for 2307 INET interfaces. The ROR then includes the lower 32 bits of its own 2308 DLA (or the DLA of the Proxy) as the LHS, encodes the DLA prefix 2309 length code in the SRT field and sets the FMT code accordingly as 2310 specified in Section 3.3. 2312 The ROR then sets the NA message R flag to 1 (as a router), S flag to 2313 1 (as a response to a solicitation), and O flag to 0 (as a proxy). 2314 The ROR finally encapsulates the NA message in an OAL header with 2315 source set to its own DLA and destination set to the source DLA of 2316 the NS(AR) message, then forwards the message into the spanning tree 2317 without decrementing the network-layer TTL/Hop Limit field. 2319 3.14.4. Relaying the NA 2321 When the Bridge receives the NA message from the ROR, it discards the 2322 INET header and determines that the ROS is the next hop by consulting 2323 its standard IPv6 forwarding table for the OAL header destination 2324 address. The Bridge then forwards the OAL-encapsulated NA message 2325 toward the ROS the same as for any IPv6 router. The final-hop Bridge 2326 in the spanning tree will deliver the message via a secured tunnel to 2327 the ROS. 2329 3.14.5. Processing the NA 2331 When the ROS receives the solicited NA message, it processes the 2332 message the same as for standard IPv6 Address Resolution [RFC4861]. 2333 In the process, it caches the source DLA then creates an asymmetric 2334 neighbor cache entry for the target and caches all information found 2335 in the OMNI option. The ROS finally sets the asymmetric neighbor 2336 cache entry lifetime to ReachableTime seconds. (When the ROS is a 2337 Client, the solicited NA message will first be delivered via the 2338 spanning tree to one of its current Servers, which then securely 2339 forwards the message to the Client by including an Authentication 2340 option per [RFC4380]. 2342 3.14.6. Route Optimization Maintenance 2344 Following route optimization, the ROS forwards future data packets 2345 destined to the target via the addresses found in the cached link- 2346 layer information. The route optimization is shared by all sources 2347 that send packets to the target via the ROS, i.e., and not just the 2348 source on behalf of which the route optimization was initiated. 2350 While new data packets destined to the target are flowing through the 2351 ROS, it sends additional NS(AR) messages to the ROR before 2352 ReachableTime expires to receive a fresh solicited NA message the 2353 same as described in the previous sections (route optimization 2354 refreshment strategies are an implementation matter, with a non- 2355 normative example given in Appendix A.1). The ROS uses the cached 2356 DLA of the ROR as the NS(AR) OAL destination address (i.e., instead 2357 of using the DLA corresponding to the target as was the case for the 2358 initial NS(AR)), and sends up to MAX_MULTICAST_SOLICIT NS(AR) 2359 messages separated by 1 second until an NA is received. If no NA is 2360 received, the ROS assumes that the current ROR has become unreachable 2361 and deletes the target neighbor cache entry. Subsequent data packets 2362 will trigger a new route optimization with an NS with OAL destination 2363 address set to the DLA corresponding to the target per Section 3.14.1 2364 to discover a new ROR while initial data packets travel over a 2365 suboptimal route. 2367 If an NA is received, the ROS then updates the asymmetric neighbor 2368 cache entry to refresh ReachableTime, while (for MNP destinations) 2369 the ROR adds or updates the ROS address to the target's Report List 2370 and with time set to ReportTime. While no data packets are flowing, 2371 the ROS instead allows ReachableTime for the asymmetric neighbor 2372 cache entry to expire. When ReachableTime expires, the ROS deletes 2373 the asymmetric neighbor cache entry. Any future data packets flowing 2374 through the ROS will again trigger a new route optimization. 2376 The ROS may also receive unsolicited NA messages from the ROR at any 2377 time (see: Section 3.16). If there is an asymmetric neighbor cache 2378 entry for the target, the ROS updates the link-layer information but 2379 does not update ReachableTime since the receipt of an unsolicited NA 2380 does not confirm that any forward paths are working. If there is no 2381 asymmetric neighbor cache entry, the ROS simply discards the 2382 unsolicited NA. 2384 In this arrangement, the ROS holds an asymmetric neighbor cache entry 2385 for the target via the ROR, but the ROR does not hold an asymmetric 2386 neighbor cache entry for the ROS. The route optimization neighbor 2387 relationship is therefore asymmetric and unidirectional. If the 2388 target node also has packets to send back to the source node, then a 2389 separate route optimization procedure is performed in the reverse 2390 direction. But, there is no requirement that the forward and reverse 2391 paths be symmetric. 2393 3.15. Neighbor Unreachability Detection (NUD) 2395 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2396 [RFC4861] either reactively in response to persistent link-layer 2397 errors (see Section 3.11) or proactively to confirm reachability. 2398 The NUD algorithm is based on periodic control message exchanges. 2399 The algorithm may further be seeded by ND hints of forward progress, 2400 but care must be taken to avoid inferring reachability based on 2401 spoofed information. For example, authentic IPv6 ND message 2402 exchanges may be considered as acceptable hints of forward progress, 2403 while spurious data packets should not be. 2405 AERO Servers, Proxys and Relays can use (OAL-encapsulated) standard 2406 NS/NA NUD exchanges sent over the OMNI link spanning tree to securely 2407 test reachability without risk of DoS attacks from nodes pretending 2408 to be a neighbor (these NS/NA(NUD) messages use the unicast LLAs and 2409 DLAs of the two parties involved in the NUD test the same as for 2410 standard IPv6 ND). Proxys can further perform NUD to securely verify 2411 Server reachability on behalf of their proxyed Clients. However, a 2412 means for an ROS to test the unsecured forward directions of target 2413 route optimized paths is also necessary. 2415 When an ROR directs an ROS to a neighbor with one or more target 2416 link-layer addresses, the ROS can proactively test each such 2417 unsecured route optimized path by sending "loopback" NS(NUD) 2418 messages. While testing the paths, the ROS can optionally continue 2419 to send packets via the spanning tree, maintain a small queue of 2420 packets until target reachability is confirmed, or (optimistically) 2421 allow packets to flow via the route optimized paths. 2423 When the ROS sends a loopback NS(NUD) message, it uses its own LLA as 2424 both the IPv6 source and destination address, and the MNP Subnet- 2425 Router anycast address as the Target Address. The ROS includes a 2426 Nonce and Timestamp option, then encapsulates the message in OAL/INET 2427 headers with its own DLA as the source and the DLA of the route 2428 optimization target as the destination. The ROS then forwards the 2429 message to the target (either directly to the L2ADDR of the target if 2430 the target is in the same OMNI link segment, or via a Bridge if the 2431 target is in a different OMNI link segment). 2433 When the route optimization target receives the NS(NUD) message, it 2434 notices that the IPv6 destination address is the same as the source 2435 address. It then reverses the OAL header source and destination 2436 addresses and returns the message to the ROS (either directly or via 2437 the spanning tree). The route optimization target does not decrement 2438 the NS(NUD) message IPv6 Hop-Limit in the process, since the message 2439 has not exited the OMNI link. 2441 When the ROS receives the NS(NUD) message, it can determine from the 2442 Nonce, Timestamp and Target Address that the message originated from 2443 itself and that it transited the forward path. The ROS need not 2444 prepare an NA response, since the destination of the response would 2445 be itself and testing the route optimization path again would be 2446 redundant. 2448 The ROS marks route optimization target paths that pass these NUD 2449 tests as "reachable", and those that do not as "unreachable". These 2450 markings inform the OMNI interface forwarding algorithm specified in 2451 Section 3.10. 2453 Note that to avoid a DoS vector nodes MUST NOT return loopback 2454 NS(NUD) messages received from an unsecured link-layer source via the 2455 (secured) spanning tree. 2457 3.16. Mobility Management and Quality of Service (QoS) 2459 AERO is a Distributed Mobility Management (DMM) service. Each Server 2460 is responsible for only a subset of the Clients on the OMNI link, as 2461 opposed to a Centralized Mobility Management (CMM) service where 2462 there is a single network mobility collective entity for all Clients. 2463 Clients coordinate with their associated Servers via RS/RA exchanges 2464 to maintain the DMM profile, and the AERO routing system tracks all 2465 current Client/Server peering relationships. 2467 Servers provide default routing and mobility/multilink services for 2468 their dependent Clients. Clients are responsible for maintaining 2469 neighbor relationships with their Servers through periodic RS/RA 2470 exchanges, which also serves to confirm neighbor reachability. When 2471 a Client's underlying interface address and/or QoS information 2472 changes, the Client is responsible for updating the Server with this 2473 new information. Note that when there is a Proxy in the path, the 2474 Proxy can also perform some RS/RA exchanges on the Client's behalf. 2476 Mobility management messaging is based on the transmission and 2477 reception of unsolicited Neighbor Advertisement (uNA) messages. Each 2478 uNA message sets the IPv6 destination address to (link-local) All- 2479 Nodes multicast to convey a general update of Interface Attributes to 2480 (possibly) multiple recipients, or to a specific unicast LLA to 2481 announce a departure event to a specific recipient. Implementations 2482 must therefore examine the destination address to determine the 2483 nature of the mobility event (i.e., update vs departure). 2485 Mobility management considerations are specified in the following 2486 sections. 2488 3.16.1. Mobility Update Messaging 2490 Servers accommodate Client mobility, multilink and/or QoS change 2491 events by sending unsolicited NA (uNA) messages to each ROS in the 2492 target Client's Report List. When a Server sends a uNA message, it 2493 sets the IPv6 source address to the Client's LLA, sets the 2494 destination address to (link-local) All-Nodes multicast and sets the 2495 Target Address to the Client's Subnet-Router anycast address. The 2496 Server also includes an OMNI option with Prefix Length set to the 2497 length associated with the Client's LLA, with Interface Attributes 2498 for the target Client's underlying interfaces and with the OMNI 2499 header S/T-ifIndex set to 0. The Server then sets the NA R flag to 2500 1, the S flag to 0 and the O flag to 1, then encapsulates the message 2501 in an OAL header with source set to its own DLA and destination set 2502 to the DLA of the ROS and sends the message into the spanning tree. 2504 As discussed in Section 7.2.6 of [RFC4861], the transmission and 2505 reception of uNA messages is unreliable but provides a useful 2506 optimization. In well-connected Internetworks with robust data links 2507 uNA messages will be delivered with high probability, but in any case 2508 the Server can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT uNAs 2509 to each ROS to increase the likelihood that at least one will be 2510 received. 2512 When the ROS receives a uNA message prepared as above, it ignores the 2513 message if there is no existing neighbor cache entry for the Client. 2514 Otherwise, it uses the included OMNI option information to update the 2515 neighbor cache entry, but does not reset ReachableTime since the 2516 receipt of an unsolicited NA message from the target Server does not 2517 provide confirmation that any forward paths to the target Client are 2518 working. 2520 If uNA messages are lost, the ROS may be left with stale address and/ 2521 or QoS information for the Client for up to ReachableTime seconds. 2522 During this time, the ROS can continue sending packets according to 2523 its stale neighbor cache information. When ReachableTime is close to 2524 expiring, the ROS will re-initiate route optimization and receive 2525 fresh link-layer address information. 2527 In addition to sending uNA messages to the current set of ROSs for 2528 the Client, the Server also sends uNAs to the DLA associated with the 2529 link-layer address for any underlying interface for which the link- 2530 layer address has changed. These uNA messages update an old Proxy/ 2531 Server that cannot easily detect (e.g., without active probing) when 2532 a formerly-active Client has departed. When the Server sends the 2533 uNA, it sets the IPv6 source address to the Client's LLA, sets the 2534 destination address to the old Proxy/Server's LLA, and sets the 2535 Target Address to the Client's Subnet-Router anycast address. The 2536 Server also includes an OMNI option with Prefix Length set to the 2537 length associated with the Client's LLA, with Interface Attributes 2538 for the changed underlying interface, and with the OMNI header S/ 2539 T-ifIndex set to 0. The Server then sets the NA R flag to 1, the S 2540 flag to 0 and the O flag to 1, then encapsulates the message in an 2541 OAL header with source set to its own DLA and destination set to the 2542 DLA of the old Proxy/Server and sends the message into the spanning 2543 tree. 2545 3.16.2. Announcing Link-Layer Address and/or QoS Preference Changes 2547 When a Client needs to change its underlying interface addresses and/ 2548 or QoS preferences (e.g., due to a mobility event), either the Client 2549 or its Proxys send RS messages to the Server via the spanning tree 2550 with an OMNI option that includes Interface attributes with the new 2551 link quality and address information. 2553 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 2554 sending actual data packets in case one or more RAs are lost. If all 2555 RAs are lost, the Client SHOULD re-associate with a new Server. 2557 When the Server receives the Client's changes, it sends uNA messages 2558 to all nodes in the Report List the same as described in the previous 2559 section. 2561 3.16.3. Bringing New Links Into Service 2563 When a Client needs to bring new underlying interfaces into service 2564 (e.g., when it activates a new data link), it sends an RS message to 2565 the Server via the underlying interface with an OMNI option that 2566 includes Interface Attributes with appropriate link quality values 2567 and with link-layer address information for the new link. 2569 3.16.4. Deactivating Existing Links 2571 When a Client needs to deactivate an existing underlying interface, 2572 it sends an RS or uNA message to its Server with an OMNI option with 2573 appropriate Interface Attribute values - in particular, the link 2574 quality value 0 assures that neighbors will cease to use the link. 2576 If the Client needs to send RS/uNA messages over an underlying 2577 interface other than the one being deactivated, it MUST include 2578 Interface Attributes with appropriate link quality values for any 2579 underlying interfaces being deactivated. 2581 Note that when a Client deactivates an underlying interface, 2582 neighbors that have received the RS/uNA messages need not purge all 2583 references for the underlying interface from their neighbor cache 2584 entries. The Client may reactivate or reuse the underlying interface 2585 and/or its ifIndex at a later point in time, when it will send RS/uNA 2586 messages with fresh Interface Attributes to update any neighbors. 2588 3.16.5. Moving Between Servers 2590 The Client performs the procedures specified in Section 3.12.2 when 2591 it first associates with a new Server or renews its association with 2592 an existing Server. The Client also includes MS-Release identifiers 2593 in the RS message OMNI option per [I-D.templin-6man-omni-interface] 2594 if it wants the new Server to notify any old Servers from which the 2595 Client is departing. 2597 When the new Server receives the Client's RS message, it returns an 2598 RA as specified in Section 3.12.3 and sends up to 2599 MAX_NEIGHBOR_ADVERTIISEMENT uNA messages to any old Servers listed in 2600 OMNI option MS-Release identifiers. When the new Server sends a uNA 2601 message, it sets the IPv6 source address to the Client's LLA, sets 2602 the destination address to the old Server's LLA, and sets the Target 2603 Address to the Client's Subnet-Router anycast address. The new 2604 Server also includes an OMNI option with Prefix Length set to the 2605 length associated with the Client's LLA, with Interface Attributes 2606 for its own underlying interface, and with the OMNI header S/ 2607 T-ifIndex set to 0. The new Server then sets the NA R flag to 1, the 2608 S flag to 0 and the O flag to 1, then encapsulates the message in an 2609 OAL header with source set to its own DLA and destination set to the 2610 DLA of the old Server and sends the message into the spanning tree. 2612 When an old Server receives the uNA, it changes the Client's neighbor 2613 cache entry state to DEPARTED, sets the link-layer address of the 2614 Client to the new Server's DLA, and resets DepartTime. After a short 2615 delay (e.g., 2 seconds) the old Server withdraws the Client's MNP 2616 from the routing system. After DepartTime expires, the old Server 2617 deletes the Client's neighbor cache entry. 2619 The old Server also iteratively forwards a copy of the uNA message to 2620 each ROS in the Client's Report List by changing the OAL destination 2621 address to the DLA of the ROS while leaving all other fields of the 2622 message unmodified. When the ROS receives the uNA, it examines the 2623 Target address to determine the correct asymmetric neighbor cache 2624 entry and verifies that the IPv6 destination address matches the old 2625 Server. The ROS then caches the IPv6 source address as the new 2626 Server for the existing asymmetric neighbor cache entry and marks the 2627 entry as STALE. While in the STALE state, the ROS allows new data 2628 packets to flow according to any existing cached link-layer 2629 information and sends new NS(AR) messages using its own DLA as the 2630 OAL source and the DLA of the new Server as the OAL destination 2631 address to elicit NA messages that reset the asymmetric neighbor 2632 cache entry state to REACHABLE. If no new NA message is received for 2633 10 seconds while in the STALE state, the ROS deletes the neighbor 2634 cache entry. 2636 Clients SHOULD NOT move rapidly between Servers in order to avoid 2637 causing excessive oscillations in the AERO routing system. Examples 2638 of when a Client might wish to change to a different Server include a 2639 Server that has gone unreachable, topological movements of 2640 significant distance, movement to a new geographic region, movement 2641 to a new OMNI link segment, etc. 2643 When a Client moves to a new Server, some of the fragments of a 2644 multiple fragment packet may have already arrived at the old Server 2645 while others are en route to the new Server, however no special 2646 attention in the reassembly algorithm is necessary when re-routed 2647 fragments are simply treated as loss. 2649 3.17. Multicast 2651 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 2652 [RFC3810] proxy service for its EUNs and/or hosted applications 2653 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 2654 underlying interfaces for which group membership is required. The 2655 IGMP/MLD messages may be further forwarded by a first-hop ANET access 2656 router acting as an IGMP/MLD-snooping switch [RFC4541], then 2657 ultimately delivered to an AERO Proxy/Server acting as a Protocol 2658 Independent Multicast - Sparse-Mode (PIM-SM, or simply "PIM") 2659 Designated Router (DR) [RFC7761]. AERO Relays also act as PIM 2660 routers (i.e., the same as AERO Proxys/Servers) on behalf of nodes on 2661 INET/EUN networks. The behaviors identified in the following 2662 sections correspond to Source-Specific Multicast (SSM) and Any-Source 2663 Multicast (ASM) operational modes. 2665 3.17.1. Source-Specific Multicast (SSM) 2667 When an ROS (i.e., an AERO Proxy/Server/Relay) "X" acting as PIM 2668 router receives a Join/Prune message from a node on its downstream 2669 interfaces containing one or more ((S)ource, (G)roup) pairs, it 2670 updates its Multicast Routing Information Base (MRIB) accordingly. 2671 For each S belonging to a prefix reachable via X's non-OMNI 2672 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 2673 on those interfaces per [RFC7761]. 2675 For each S belonging to a prefix reachable via X's OMNI interface, X 2676 originates a separate copy of the Join/Prune for each (S,G) in the 2677 message using its own LLA as the source address and ALL-PIM-ROUTERS 2678 as the destination address. X then encapsulates each message in an 2679 OAL header with source address set to the DLA of X and destination 2680 address set to S then forwards the message into the spanning tree, 2681 which delivers it to AERO Server/Relay "Y" that services S. At the 2682 same time, if the message was a Join, X sends a route-optimization NS 2683 message toward each S the same as discussed in Section 3.14. The 2684 resulting NAs will return the LLA for the prefix that matches S as 2685 the network-layer source address and with an OMNI option with the DLA 2686 corresponding to any underlying interfaces that are currently 2687 servicing S. 2689 When Y processes the Join/Prune message, if S located behind any 2690 INET, Direct, or VPNed interfaces Y acts as a PIM router and updates 2691 its MRIB to list X as the next hop in the reverse path. If S is 2692 located behind any Proxys "Z"*, Y also forwards the message to each 2693 Z* over the spanning tree while continuing to use the LLA of X as the 2694 source address. Each Z* then updates its MRIB accordingly and 2695 maintains the LLA of X as the next hop in the reverse path. Since 2696 the Bridges do not examine network layer control messages, this means 2697 that the (reverse) multicast tree path is simply from each Z* (and/or 2698 Y) to X with no other multicast-aware routers in the path. If any Z* 2699 (and/or Y) is located on the same OMNI link segment as X, the 2700 multicast data traffic sent to X directly using OAL/INET 2701 encapsulation instead of via a Bridge. 2703 Following the initial Join/Prune and NS/NA messaging, X maintains an 2704 asymmetric neighbor cache entry for each S the same as if X was 2705 sending unicast data traffic to S. In particular, X performs 2706 additional NS/NA exchanges to keep the neighbor cache entry alive for 2707 up to t_periodic seconds [RFC7761]. If no new Joins are received 2708 within t_periodic seconds, X allows the neighbor cache entry to 2709 expire. Finally, if X receives any additional Join/Prune messages 2710 for (S,G) it forwards the messages to each Y and Z* in the neighbor 2711 cache entry over the spanning tree. 2713 At some later time, Client C that holds an MNP for source S may 2714 depart from a first Proxy Z1 and/or connect via a new Proxy Z2. In 2715 that case, Y sends an unsolicited NA message to X the same as 2716 specified for unicast mobility in Section 3.16. When X receives the 2717 unsolicited NA message, it updates its asymmetric neighbor cache 2718 entry for the LLA for source S and sends new Join messages to any new 2719 Proxys Z2. There is no requirement to send any Prune messages to old 2720 Proxys Z1 since source S will no longer source any multicast data 2721 traffic via Z1. Instead, the multicast state for (S,G) in Proxy Z1 2722 will soon time out since no new Joins will arrive. 2724 After some later time, C may move to a new Server Y2 and depart from 2725 old Sever Y1. In that case, Y1 sends Join messages for any of C's 2726 active (S,G) groups to Y2 while including its own LLA as the source 2727 address. This causes Y2 to include Y1 in the multicast forwarding 2728 tree during the interim time that Y1's symmetric neighbor cache entry 2729 for C is in the DEPARTED state. At the same time, Y1 sends an 2730 unsolicited NA message to X with an OMNI option with S/T-ifIndex in 2731 the header set to 0 and a release indication to cause X to release 2732 its asymmetric neighbor cache entry. X then sends a new Join message 2733 to S via the spanning tree and re-initiates route optimization the 2734 same as if it were receiving a fresh Join message from a node on a 2735 downstream link. 2737 3.17.2. Any-Source Multicast (ASM) 2739 When an ROS X acting as a PIM router receives a Join/Prune from a 2740 node on its downstream interfaces containing one or more (*,G) pairs, 2741 it updates its Multicast Routing Information Base (MRIB) accordingly. 2742 X then forwards a copy of the message to the Rendezvous Point (RP) R 2743 for each G over the spanning tree. X uses its own LLA as the source 2744 address and ALL-PIM-ROUTERS as the destination address, then 2745 encapsulates each message in an OAL header with source address set to 2746 the DLA of X and destination address set to R, then sends the message 2747 into the spanning tree. At the same time, if the message was a Join 2748 X initiates NS/NA route optimization the same as for the SSM case 2749 discussed in Section 3.17.1. 2751 For each source S that sends multicast traffic to group G via R, the 2752 Proxy/Server Z* for the Client that aggregates S encapsulates the 2753 packets in PIM Register messages and forwards them to R via the 2754 spanning tree, which may then elect to send a PIM Join to Z*. This 2755 will result in an (S,G) tree rooted at Z* with R as the next hop so 2756 that R will begin to receive two copies of the packet; one native 2757 copy from the (S, G) tree and a second copy from the pre-existing (*, 2758 G) tree that still uses PIM Register encapsulation. R can then issue 2759 a PIM Register-stop message to suppress the Register-encapsulated 2760 stream. At some later time, if C moves to a new Proxy/Server Z*, it 2761 resumes sending packets via PIM Register encapsulation via the new 2762 Z*. 2764 At the same time, as multicast listeners discover individual S's for 2765 a given G, they can initiate an (S,G) Join for each S under the same 2766 procedures discussed in Section 3.17.1. Once the (S,G) tree is 2767 established, the listeners can send (S, G) Prune messages to R so 2768 that multicast packets for group G sourced by S will only be 2769 delivered via the (S, G) tree and not from the (*, G) tree rooted at 2770 R. All mobility considerations discussed for SSM apply. 2772 3.17.3. Bi-Directional PIM (BIDIR-PIM) 2774 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 2775 approach to ASM that treats the Rendezvous Point (RP) as a Designated 2776 Forwarder (DF). Further considerations for BIDIR-PIM are out of 2777 scope. 2779 3.18. Operation over Multiple OMNI Links 2781 An AERO Client can connect to multiple OMNI links the same as for any 2782 data link service. In that case, the Client maintains a distinct 2783 OMNI interface for each link, e.g., 'omni0' for the first link, 2784 'omni1' for the second, 'omni2' for the third, etc. Each OMNI link 2785 would include its own distinct set of Bridges, Servers and Proxys, 2786 thereby providing redundancy in case of failures. 2788 Each OMNI link could utilize the same or different ANET connections. 2789 The links can be distinguished at the link-layer via the SRT prefix 2790 in a similar fashion as for Virtual Local Area Network (VLAN) tagging 2791 (e.g., IEEE 802.1Q) and/or through assignment of distinct sets of 2792 MSPs on each link. This gives rise to the opportunity for supporting 2793 multiple redundant networked paths, with each VLAN distinguished by a 2794 different SRT "color" (see: Section 3.2.5). 2796 The Client's IP layer can select the outgoing OMNI interface 2797 appropriate for a given traffic profile while (in the reverse 2798 direction) correspondent nodes must have some way of steering their 2799 packets destined to a target via the correct OMNI link. 2801 In a first alternative, if each OMNI link services different MSPs, 2802 then the Client can receive a distinct MNP from each of the links. 2803 IP routing will therefore assure that the correct Red/Green/Blue/etc. 2804 network is used for both outbound and inbound traffic. This can be 2805 accomplished using existing technologies and approaches, and without 2806 requiring any special supporting code in correspondent nodes or 2807 Bridges. 2809 In a second alternative, if each OMNI link services the same MSP(s) 2810 then each link could assign a distinct "OMNI link Anycast" address 2811 that is configured by all Bridges on the link. Correspondent nodes 2812 can then perform Segment Routing to select the correct SRT, which 2813 will then direct the packet over multiple hops to the target. 2815 3.19. DNS Considerations 2817 AERO Client MNs and INET correspondent nodes consult the Domain Name 2818 System (DNS) the same as for any Internetworking node. When 2819 correspondent nodes and Client MNs use different IP protocol versions 2820 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 2821 A records for IPv4 address mappings to MNs which must then be 2822 populated in Relay NAT64 mapping caches. In that way, an IPv4 2823 correspondent node can send packets to the IPv4 address mapping of 2824 the target MN, and the Relay will translate the IPv4 header and 2825 destination address into an IPv6 header and IPv6 destination address 2826 of the MN. 2828 When an AERO Client registers with an AERO Server, the Server can 2829 return the address(es) of DNS servers in RDNSS options [RFC6106]. 2830 The DNS server provides the IP addresses of other MNs and 2831 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 2833 3.20. Transition Considerations 2835 OAL encapsulation ensures that dissimilar INET partitions can be 2836 joined into a single unified OMNI link, even though the partitions 2837 themselves may have differing protocol versions and/or incompatible 2838 addressing plans. However, a commonality can be achieved by 2839 incrementally distributing globally routable (i.e., native) IP 2840 prefixes to eventually reach all nodes (both mobile and fixed) in all 2841 OMNI link segments. This can be accomplished by incrementally 2842 deploying AERO Relays on each INET partition, with each Relay 2843 distributing its MNPs and/or discovering non-MNP prefixes on its INET 2844 links. 2846 This gives rise to the opportunity to eventually distribute native IP 2847 addresses to all nodes, and to present a unified OMNI link view even 2848 if the INET partitions remain in their current protocol and 2849 addressing plans. In that way, the OMNI link can serve the dual 2850 purpose of providing a mobility/multilink service and a transition 2851 service. Or, if an INET partition is transitioned to a native IP 2852 protocol version and addressing scheme that is compatible with the 2853 OMNI link MNP-based addressing scheme, the partition and OMNI link 2854 can be joined by Relays. 2856 Relays that connect INETs/EUNs with dissimilar IP protocol versions 2857 may need to employ a network address and protocol translation 2858 function such as NAT64 [RFC6146]. 2860 3.21. Detecting and Reacting to Server and Bridge Failures 2862 In environments where rapid failure recovery is required, Servers and 2863 Bridges SHOULD use Bidirectional Forwarding Detection (BFD) 2864 [RFC5880]. Nodes that use BFD can quickly detect and react to 2865 failures so that cached information is re-established through 2866 alternate nodes. BFD control messaging is carried only over well- 2867 connected ground domain networks (i.e., and not low-end radio links) 2868 and can therefore be tuned for rapid response. 2870 Servers and Bridges maintain BFD sessions in parallel with their BGP 2871 peerings. If a Server or Bridge fails, BGP peers will quickly re- 2872 establish routes through alternate paths the same as for common BGP 2873 deployments. Similarly, Proxys maintain BFD sessions with their 2874 associated Bridges even though they do not establish BGP peerings 2875 with them. 2877 Proxys SHOULD use proactive NUD for Servers for which there are 2878 currently active ANET Clients in a manner that parallels BFD, i.e., 2879 by sending unicast NS messages in rapid succession to receive 2880 solicited NA messages. When the Proxy is also sending RS messages on 2881 behalf of ANET Clients, the RS/RA messaging can be considered as 2882 equivalent hints of forward progress. This means that the Proxy need 2883 not also send a periodic NS if it has already sent an RS within the 2884 same period. If a Server fails, the Proxy will cease to receive 2885 advertisements and can quickly inform Clients of the outage by 2886 sending multicast RA messages on the ANET interface. 2888 The Proxy sends multicast RA messages with source address set to the 2889 Server's address, destination address set to (link-local) All-Nodes 2890 multicast, and Router Lifetime set to 0. The Proxy SHOULD send 2891 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2892 [RFC4861]. Any Clients on the ANET interface that have been using 2893 the (now defunct) Server will receive the RA messages and associate 2894 with a new Server. 2896 3.22. AERO Clients on the Open Internet 2898 AERO Clients that connect to the open Internet via INET interfaces 2899 can establish a VPN or direct link to securely connect to a Server in 2900 a "tethered" arrangement with all of the Client's traffic transiting 2901 the Server. Alternatively, the Client can associate with an INET 2902 Server using UDP/IP encapsulation and asymmetric securing services as 2903 discussed in the following sections. 2905 When a Client's OMNI interface enables an INET underlying interface, 2906 it first determines whether the interface is likely to be behind a 2907 NAT. For IPv4, the Client assumes it is on the open Internet if the 2908 INET address is not a special-use IPv4 address per [RFC3330]. 2909 Similarly for IPv6, the Client assumes it is on the open Internet if 2910 the INET address is not a link-local [RFC4291] or unique-local 2911 [RFC4193] IPv6 address. 2913 The Client then prepares a UDP/IP-encapsulated RS message with IPv6 2914 source address set to its LLA, with IPv6 destination set to (link- 2915 local) All-Routers multicast and with an OMNI option with underlying 2916 interface attributes. If the Client believes that it is on the open 2917 Internet, it SHOULD include Interface Attributes with the L2ADDR used 2918 for INET encapsulation (otherwise, it MAY omit L2ADDR). If the 2919 underlying address is IPv4, the Client includes the Port Number and 2920 IPv4 address written in obfuscated form [RFC4380] as discussed in 2921 Section 3.3. If the underlying interface address is IPv6, the Client 2922 instead includes the Port Number and IPv6 address in obfuscated form. 2923 The Client finally includes an Authentication option per [RFC4380] to 2924 provide message authentication, sets the UDP/IP source to its INET 2925 address and UDP port, sets the UDP/IP destination to the Server's 2926 INET address and the AERO service port number (8060), then sends the 2927 message to the Server. 2929 When the Server receives the RS, it authenticates the message and 2930 registers the Client's MNP and INET interface information according 2931 to the OMNI option parameters. If the RS message includes an L2ADDR 2932 in the OMNI option, the Server compares the encapsulation IP address 2933 and UDP port number with the (unobfuscated) values. If the values 2934 are the same, the Server caches the Client's information as "INET" 2935 addresses meaning that the Client is likely to accept direct messages 2936 without requiring NAT traversal exchanges. If the values are 2937 different (or, if the OMNI option did not include an L2ADDR) the 2938 Server instead caches the Client's information as "NAT" addresses 2939 meaning that NAT traversal exchanges may be necessary. 2941 The Server then returns an RA message with IPv6 source and 2942 destination set corresponding to the addresses in the RS, and with an 2943 Authentication option per [RFC4380]. For IPv4, the Server also 2944 includes an Origin option per [RFC4380] with the mapped and 2945 obfuscated Port Number and IPv4 address observed in the encapsulation 2946 headers. For IPv6, the Server instead includes an IPv6 Origin option 2947 per Figure 6 with the mapped and obfuscated observed Port Number and 2948 IPv6 address (note that the value 0x02 in the second octet 2949 differentiates from other [RFC4380] option types). 2951 +--------+--------+-----------------+ 2952 | 0x00 | 0x02 | Origin port # | 2953 +--------+--------+-----------------+ 2954 ~ Origin IPv6 address ~ 2955 +-----------------------------------+ 2957 Figure 6: IPv6 Origin Option 2959 When the Client receives the RA message, it compares the mapped Port 2960 Number and IP address from the Origin option with its own address. 2961 If the addresses are the same, the Client assumes the open Internet / 2962 Cone NAT principle; if the addresses are different, the Client 2963 instead assumes that further qualification procedures are necessary 2964 to detect the type of NAT and proceeds according to standard 2965 [RFC4380] procedures. 2967 After the Client has registered its INET interfaces in such RS/RA 2968 exchanges it sends periodic RS messages to receive fresh RA messages 2969 before the Router Lifetime received on each INET interface expires. 2970 The Client also maintains default routes via its Servers, i.e., the 2971 same as described in earlier sections. 2973 When the Client sends messages to target IP addresses, it also 2974 invokes route optimization per Section 3.14 using IPv6 ND address 2975 resolution messaging. The Client sends the NS(AR) message to the 2976 Server wrapped in a UDP/IP header with an Authentication option with 2977 the NS source address set to the Client's LLA and destination address 2978 set to the target solicited node multicast address. The Server 2979 authenticates the message and sends a corresponding NS(AR) message 2980 over the spanning tree the same as if it were the ROS, but with the 2981 OAL source address set to the Server's DLA and destination set to the 2982 DLA of the target. When the ROR receives the NS(AR), it adds the 2983 Server's DLA and Client's LLA to the target's Report List, and 2984 returns an NA with OMNI option information for the target. The 2985 Server then returns a UDP/IP encapsulated NA message with an 2986 Authentication option to the Client. 2988 Following route optimization for targets in the same OMNI link 2989 segment, if the target's L2ADDR is on the open INET, the Client 2990 forwards data packets directly to the target INET address. If the 2991 target is behind a NAT, the Client first establishes NAT state for 2992 the L2ADDR using the "bubble" mechanisms specified in 2993 [RFC6081][RFC4380]. The Client continues to send data packets via 2994 its Server until NAT state is populated, then begins forwarding 2995 packets via the direct path through the NAT to the target. For 2996 targets in different OMNI link segments, the Client uses OAL/ORH 2997 encapsulation and forwards data packets to the Bridge that returned 2998 the NA message. 3000 The ROR may return uNAs via the Server if the target moves, and the 3001 Server will send corresponding Authentication-protected uNAs to the 3002 Client. The Client can also send "loopback" NS(NUD) messages to test 3003 forward path reachability even though there is no security 3004 association between the Client and the target. 3006 The Client sends UDP/IP encapsulated IPv6 packets no larger than 1280 3007 bytes in one piece. In order to accommodate larger IPv6 packets (up 3008 to the OMNI interface MTU), the Client inserts an OAL header with 3009 source set to its own DLA and destination set to the DLA of the 3010 target and uses IPv6 fragmentation according to Section 3.9. The 3011 Client then encapsulates each fragment in a UDP/IP header and sends 3012 the fragments to the next hop. 3014 3.23. Time-Varying MNPs 3016 In some use cases, it is desirable, beneficial and efficient for the 3017 Client to receive a constant MNP that travels with the Client 3018 wherever it moves. For example, this would allow air traffic 3019 controllers to easily track aircraft, etc. In other cases, however 3020 (e.g., intelligent transportation systems), the MN may be willing to 3021 sacrifice a modicum of efficiency in order to have time-varying MNPs 3022 that can be changed every so often to defeat adversarial tracking. 3024 The DHCPv6 service offers a way for Clients that desire time-varying 3025 MNPs to obtain short-lived prefixes (e.g., on the order of a small 3026 number of minutes). In that case, the identity of the Client would 3027 not be bound to the MNP but rather the Client's identity would be 3028 bound to the DHCPv6 Device Unique Identifier (DUID) and used as the 3029 seed for Prefix Delegation. The Client would then be obligated to 3030 renumber its internal networks whenever its MNP (and therefore also 3031 its LLA) changes. This should not present a challenge for Clients 3032 with automated network renumbering services, however presents limits 3033 for the durations of ongoing sessions that would prefer to use a 3034 constant address. 3036 4. Implementation Status 3038 An early AERO implementation based on OpenVPN (https://openvpn.net/) 3039 was announced on the v6ops mailing list on January 10, 2018 and an 3040 initial public release of the AERO proof-of-concept source code was 3041 announced on the intarea mailing list on August 21, 2015. 3043 AERO Release-3.0.2 was tagged on October 15, 2020, and is undergoing 3044 internal testing. Additional releases expected Q42020, with first 3045 public release expected before year-end. 3047 5. IANA Considerations 3049 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 3050 AERO in the "enterprise-numbers" registry. 3052 The IANA has assigned the UDP port number "8060" for an earlier 3053 experimental version of AERO [RFC6706]. This document obsoletes 3054 [RFC6706] and claims the UDP port number "8060" for all future use. 3056 The IANA is instructed to assign a new type value TBD in the IPv6 3057 Routing Types registry. 3059 No further IANA actions are required. 3061 6. Security Considerations 3063 AERO Bridges configure secured tunnels with AERO Servers, Relays and 3064 Proxys within their local OMNI link segments. Applicable secured 3065 tunnel alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 3066 [RFC6347], WireGuard [WG], etc. The AERO Bridges of all OMNI link 3067 segments in turn configure secured tunnels for their neighboring AERO 3068 Bridges in a spanning tree topology. Therefore, control messages 3069 exchanged between any pair of OMNI link neighbors on the spanning 3070 tree are already secured. 3072 AERO Servers, Relays and Proxys targeted by a route optimization may 3073 also receive data packets directly from arbitrary nodes in INET 3074 partitions instead of via the spanning tree. For INET partitions 3075 that apply effective ingress filtering to defeat source address 3076 spoofing, the simple data origin authentication procedures in 3077 Section 3.8 can be applied. 3079 For INET partitions that require strong security in the data plane, 3080 two options for securing communications include 1) disable route 3081 optimization so that all traffic is conveyed over secured tunnels, or 3082 2) enable on-demand secure tunnel creation between INET partition 3083 neighbors. Option 1) would result in longer routes than necessary 3084 and traffic concentration on critical infrastructure elements. 3085 Option 2) could be coordinated by establishing a secured tunnel on- 3086 demand instead of performing an NS/NA exchange in the route 3087 optimization procedures. Procedures for establishing on-demand 3088 secured tunnels are out of scope. 3090 AERO Clients that connect to secured ANETs need not apply security to 3091 their ND messages, since the messages will be intercepted by a 3092 perimeter Proxy that applies security on its INET-facing interface as 3093 part of the spanning tree (see above). AERO Clients connected to the 3094 open INET can use symmetric network and/or transport layer security 3095 services such as VPNs or can by some other means establish a direct 3096 link. When a VPN or direct link may be impractical, however, an 3097 asymmetric security service such as the Authentication option 3098 specified in [RFC4380] should be applied. The Authentication option 3099 requires a unique Client identifier, which can be obtained per the 3100 Universally Unique IDentifier (UUID) [RFC4122] service and also used 3101 as a DHCP Unique Identifier (DUID) per [RFC6355]. 3103 Application endpoints SHOULD use application-layer security services 3104 such as TLS/SSL, DTLS or SSH [RFC4251] to assure the same level of 3105 protection as for critical secured Internet services. AERO Clients 3106 that require host-based VPN services SHOULD use symmetric network 3107 and/or transport layer security services such as IPsec, TLS/SSL, 3108 DTLS, etc. AERO Proxys and Servers can also provide a network-based 3109 VPN service on behalf of the Client, e.g., if the Client is located 3110 within a secured enclave and cannot establish a VPN on its own 3111 behalf. 3113 AERO Servers and Bridges present targets for traffic amplification 3114 Denial of Service (DoS) attacks. This concern is no different than 3115 for widely-deployed VPN security gateways in the Internet, where 3116 attackers could send spoofed packets to the gateways at high data 3117 rates. This can be mitigated by connecting Servers and Bridges over 3118 dedicated links with no connections to the Internet and/or when 3119 connections to the Internet are only permitted through well-managed 3120 firewalls. Traffic amplification DoS attacks can also target an AERO 3121 Client's low data rate links. This is a concern not only for Clients 3122 located on the open Internet but also for Clients in secured 3123 enclaves. AERO Servers and Proxys can institute rate limits that 3124 protect Clients from receiving packet floods that could DoS low data 3125 rate links. 3127 AERO Relays must implement ingress filtering to avoid a spoofing 3128 attack in which spurious messages with DLA addresses are injected 3129 into an OMNI link from an outside attacker. AERO Clients MUST ensure 3130 that their connectivity is not used by unauthorized nodes on their 3131 EUNs to gain access to a protected network, i.e., AERO Clients that 3132 act as routers MUST NOT provide routing services for unauthorized 3133 nodes. (This concern is no different than for ordinary hosts that 3134 receive an IP address delegation but then "share" the address with 3135 other nodes via some form of Internet connection sharing such as 3136 tethering.) 3137 The MAP list MUST be well-managed and secured from unauthorized 3138 tampering, even though the list contains only public information. 3139 The MAP list can be conveyed to the Client in a similar fashion as in 3140 [RFC5214] (e.g., through layer 2 data link login messaging, secure 3141 upload of a static file, DNS lookups, etc.). 3143 SRH authentication facilities are specified in [RFC8754]. 3145 Security considerations for accepting link-layer ICMP messages and 3146 reflected packets are discussed throughout the document. 3148 Security considerations for IPv6 fragmentation and reassembly are 3149 discussed in [I-D.templin-6man-omni-interface]. 3151 7. Acknowledgements 3153 Discussions in the IETF, aviation standards communities and private 3154 exchanges helped shape some of the concepts in this work. 3155 Individuals who contributed insights include Mikael Abrahamsson, Mark 3156 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 3157 Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian Farrel, Nick Green, 3158 Sri Gundavelli, Brian Haberman, Bernhard Haindl, Joel Halpern, Tom 3159 Herbert, Sascha Hlusiak, Lee Howard, Zdenek Jaron, Andre Kostur, 3160 Hubert Kuenig, Ted Lemon, Andy Malis, Satoru Matsushima, Tomek 3161 Mrugalski, Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal 3162 Skorepa, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, Lloyd 3163 Wood and James Woodyatt. Members of the IESG also provided valuable 3164 input during their review process that greatly improved the document. 3165 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 3166 for their shepherding guidance during the publication of the AERO 3167 first edition. 3169 This work has further been encouraged and supported by Boeing 3170 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 3171 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 3172 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 3173 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 3174 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 3175 Gene MacLean III, Rob Muszkiewicz, Sean O'Sullivan, Vijay 3176 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 3177 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 3178 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 3179 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 3180 Benson, Katie Tran and Eric Yeh are especially acknowledged for 3181 implementing the AERO functions as extensions to the public domain 3182 OpenVPN distribution. 3184 Earlier works on NBMA tunneling approaches are found in 3185 [RFC2529][RFC5214][RFC5569]. 3187 Many of the constructs presented in this second edition of AERO are 3188 based on the author's earlier works, including: 3190 o The Internet Routing Overlay Network (IRON) 3191 [RFC6179][I-D.templin-ironbis] 3193 o Virtual Enterprise Traversal (VET) 3194 [RFC5558][I-D.templin-intarea-vet] 3196 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 3197 [RFC5320][I-D.templin-intarea-seal] 3199 o AERO, First Edition [RFC6706] 3201 Note that these works cite numerous earlier efforts that are not also 3202 cited here due to space limitations. The authors of those earlier 3203 works are acknowledged for their insights. 3205 This work is aligned with the NASA Safe Autonomous Systems Operation 3206 (SASO) program under NASA contract number NNA16BD84C. 3208 This work is aligned with the FAA as per the SE2025 contract number 3209 DTFAWA-15-D-00030. 3211 This work is aligned with the Boeing Commercial Airplanes (BCA) 3212 Internet of Things (IoT) and autonomy programs. 3214 This work is aligned with the Boeing Information Technology (BIT) 3215 MobileNet program. 3217 8. References 3219 8.1. Normative References 3221 [I-D.templin-6man-omni-interface] 3222 Templin, F. and T. Whyman, "Transmission of IP Packets 3223 over Overlay Multilink Network (OMNI) Interfaces", draft- 3224 templin-6man-omni-interface-54 (work in progress), 3225 December 2020. 3227 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3228 DOI 10.17487/RFC0791, September 1981, 3229 . 3231 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 3232 RFC 792, DOI 10.17487/RFC0792, September 1981, 3233 . 3235 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3236 Requirement Levels", BCP 14, RFC 2119, 3237 DOI 10.17487/RFC2119, March 1997, 3238 . 3240 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3241 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3242 December 1998, . 3244 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3245 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3246 DOI 10.17487/RFC3971, March 2005, 3247 . 3249 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3250 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3251 . 3253 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3254 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3255 November 2005, . 3257 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3258 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3259 . 3261 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 3262 Network Address Translations (NATs)", RFC 4380, 3263 DOI 10.17487/RFC4380, February 2006, 3264 . 3266 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3267 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3268 DOI 10.17487/RFC4861, September 2007, 3269 . 3271 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3272 Address Autoconfiguration", RFC 4862, 3273 DOI 10.17487/RFC4862, September 2007, 3274 . 3276 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 3277 DOI 10.17487/RFC6081, January 2011, 3278 . 3280 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3281 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3282 May 2017, . 3284 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3285 (IPv6) Specification", STD 86, RFC 8200, 3286 DOI 10.17487/RFC8200, July 2017, 3287 . 3289 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3290 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3291 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3292 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3293 . 3295 8.2. Informative References 3297 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3298 2016. 3300 [I-D.bonica-6man-comp-rtg-hdr] 3301 Bonica, R., Kamite, Y., Alston, A., Henriques, D., and L. 3302 Jalil, "The IPv6 Compact Routing Header (CRH)", draft- 3303 bonica-6man-comp-rtg-hdr-23 (work in progress), October 3304 2020. 3306 [I-D.bonica-6man-crh-helper-opt] 3307 Li, X., Bao, C., Ruan, E., and R. Bonica, "Compressed 3308 Routing Header (CRH) Helper Option", draft-bonica-6man- 3309 crh-helper-opt-02 (work in progress), October 2020. 3311 [I-D.ietf-intarea-frag-fragile] 3312 Bonica, R., Baker, F., Huston, G., Hinden, R., Troan, O., 3313 and F. Gont, "IP Fragmentation Considered Fragile", draft- 3314 ietf-intarea-frag-fragile-17 (work in progress), September 3315 2019. 3317 [I-D.ietf-intarea-tunnels] 3318 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3319 Architecture", draft-ietf-intarea-tunnels-10 (work in 3320 progress), September 2019. 3322 [I-D.ietf-rtgwg-atn-bgp] 3323 Templin, F., Saccone, G., Dawra, G., Lindem, A., and V. 3324 Moreno, "A Simple BGP-based Mobile Routing System for the 3325 Aeronautical Telecommunications Network", draft-ietf- 3326 rtgwg-atn-bgp-06 (work in progress), June 2020. 3328 [I-D.templin-6man-dhcpv6-ndopt] 3329 Templin, F., "A Unified Stateful/Stateless Configuration 3330 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-10 3331 (work in progress), June 2020. 3333 [I-D.templin-intarea-seal] 3334 Templin, F., "The Subnetwork Encapsulation and Adaptation 3335 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 3336 progress), January 2014. 3338 [I-D.templin-intarea-vet] 3339 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 3340 templin-intarea-vet-40 (work in progress), May 2013. 3342 [I-D.templin-ironbis] 3343 Templin, F., "The Interior Routing Overlay Network 3344 (IRON)", draft-templin-ironbis-16 (work in progress), 3345 March 2014. 3347 [I-D.templin-v6ops-pdhost] 3348 Templin, F., "IPv6 Prefix Delegation and Multi-Addressing 3349 Models", draft-templin-v6ops-pdhost-26 (work in progress), 3350 June 2020. 3352 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3354 [RFC1035] Mockapetris, P., "Domain names - implementation and 3355 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3356 November 1987, . 3358 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3359 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3360 . 3362 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3363 DOI 10.17487/RFC2003, October 1996, 3364 . 3366 [RFC2004] Perkins, C., "Minimal Encapsulation within IP", RFC 2004, 3367 DOI 10.17487/RFC2004, October 1996, 3368 . 3370 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3371 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3372 . 3374 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 3375 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 3376 . 3378 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3379 Domains without Explicit Tunnels", RFC 2529, 3380 DOI 10.17487/RFC2529, March 1999, 3381 . 3383 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3384 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3385 . 3387 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3388 of Explicit Congestion Notification (ECN) to IP", 3389 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3390 . 3392 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 3393 DOI 10.17487/RFC3330, September 2002, 3394 . 3396 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3397 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3398 DOI 10.17487/RFC3810, June 2004, 3399 . 3401 [RFC4122] Leach, P., Mealling, M., and R. Salz, "A Universally 3402 Unique IDentifier (UUID) URN Namespace", RFC 4122, 3403 DOI 10.17487/RFC4122, July 2005, 3404 . 3406 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3407 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3408 January 2006, . 3410 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3411 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3412 DOI 10.17487/RFC4271, January 2006, 3413 . 3415 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3416 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3417 2006, . 3419 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3420 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3421 December 2005, . 3423 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3424 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3425 2006, . 3427 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3428 Control Message Protocol (ICMPv6) for the Internet 3429 Protocol Version 6 (IPv6) Specification", STD 89, 3430 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3431 . 3433 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3434 Protocol (LDAP): The Protocol", RFC 4511, 3435 DOI 10.17487/RFC4511, June 2006, 3436 . 3438 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3439 "Considerations for Internet Group Management Protocol 3440 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3441 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3442 . 3444 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3445 "Internet Group Management Protocol (IGMP) / Multicast 3446 Listener Discovery (MLD)-Based Multicast Forwarding 3447 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3448 August 2006, . 3450 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3451 Algorithms in Cryptographically Generated Addresses 3452 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3453 . 3455 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3456 "Bidirectional Protocol Independent Multicast (BIDIR- 3457 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3458 . 3460 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3461 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3462 DOI 10.17487/RFC5214, March 2008, 3463 . 3465 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3466 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3467 February 2010, . 3469 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3470 Route Optimization Requirements for Operational Use in 3471 Aeronautics and Space Exploration Mobile Networks", 3472 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3473 . 3475 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3476 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3477 . 3479 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3480 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3481 January 2010, . 3483 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 3484 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 3485 . 3487 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3488 "IPv6 Router Advertisement Options for DNS Configuration", 3489 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3490 . 3492 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3493 NAT64: Network Address and Protocol Translation from IPv6 3494 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3495 April 2011, . 3497 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3498 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3499 . 3501 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3502 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3503 DOI 10.17487/RFC6221, May 2011, 3504 . 3506 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3507 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3508 DOI 10.17487/RFC6273, June 2011, 3509 . 3511 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3512 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3513 January 2012, . 3515 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 3516 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, 3517 DOI 10.17487/RFC6355, August 2011, 3518 . 3520 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3521 for Equal Cost Multipath Routing and Link Aggregation in 3522 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3523 . 3525 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3526 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3527 . 3529 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 3530 UDP Checksums for Tunneled Packets", RFC 6935, 3531 DOI 10.17487/RFC6935, April 2013, 3532 . 3534 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 3535 for the Use of IPv6 UDP Datagrams with Zero Checksums", 3536 RFC 6936, DOI 10.17487/RFC6936, April 2013, 3537 . 3539 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3540 Korhonen, "Requirements for Distributed Mobility 3541 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3542 . 3544 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3545 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3546 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3547 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3548 2016, . 3550 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3551 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3552 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3553 July 2018, . 3555 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3556 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3557 . 3559 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 3560 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3561 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 3562 . 3564 [WG] Wireguard, "Wireguard, https://www.wireguard.com", August 3565 2020. 3567 Appendix A. Non-Normative Considerations 3569 AERO can be applied to a multitude of Internetworking scenarios, with 3570 each having its own adaptations. The following considerations are 3571 provided as non-normative guidance: 3573 A.1. Implementation Strategies for Route Optimization 3575 Route optimization as discussed in Section 3.14 results in the route 3576 optimization source (ROS) creating an asymmetric neighbor cache entry 3577 for the target neighbor. The neighbor cache entry is maintained for 3578 at most ReachableTime seconds and then deleted unless updated. In 3579 order to refresh the neighbor cache entry lifetime before the 3580 ReachableTime timer expires, the specification requires 3581 implementations to issue a new NS/NA exchange to reset ReachableTime 3582 while data packets are still flowing. However, the decision of when 3583 to initiate a new NS/NA exchange and to perpetuate the process is 3584 left as an implementation detail. 3586 One possible strategy may be to monitor the neighbor cache entry 3587 watching for data packets for (ReachableTime - 5) seconds. If any 3588 data packets have been sent to the neighbor within this timeframe, 3589 then send an NS to receive a new NA. If no data packets have been 3590 sent, wait for 5 additional seconds and send an immediate NS if any 3591 data packets are sent within this "expiration pending" 5 second 3592 window. If no additional data packets are sent within the 5 second 3593 window, delete the neighbor cache entry. 3595 The monitoring of the neighbor data packet traffic therefore becomes 3596 an asymmetric ongoing process during the neighbor cache entry 3597 lifetime. If the neighbor cache entry expires, future data packets 3598 will trigger a new NS/NA exchange while the packets themselves are 3599 delivered over a longer path until route optimization state is re- 3600 established. 3602 A.2. Implicit Mobility Management 3604 OMNI interface neighbors MAY provide a configuration option that 3605 allows them to perform implicit mobility management in which no ND 3606 messaging is used. In that case, the Client only transmits packets 3607 over a single interface at a time, and the neighbor always observes 3608 packets arriving from the Client from the same link-layer source 3609 address. 3611 If the Client's underlying interface address changes (either due to a 3612 readdressing of the original interface or switching to a new 3613 interface) the neighbor immediately updates the neighbor cache entry 3614 for the Client and begins accepting and sending packets according to 3615 the Client's new address. This implicit mobility method applies to 3616 use cases such as cellphones with both WiFi and Cellular interfaces 3617 where only one of the interfaces is active at a given time, and the 3618 Client automatically switches over to the backup interface if the 3619 primary interface fails. 3621 A.3. Direct Underlying Interfaces 3623 When a Client's OMNI interface is configured over a Direct interface, 3624 the neighbor at the other end of the Direct link can receive packets 3625 without any encapsulation. In that case, the Client sends packets 3626 over the Direct link according to QoS preferences. If the Direct 3627 interface has the highest QoS preference, then the Client's IP 3628 packets are transmitted directly to the peer without going through an 3629 ANET/INET. If other interfaces have higher QoS preferences, then the 3630 Client's IP packets are transmitted via a different interface, which 3631 may result in the inclusion of Proxys, Servers and Bridges in the 3632 communications path. Direct interfaces must be tested periodically 3633 for reachability, e.g., via NUD. 3635 A.4. AERO Critical Infrastructure Considerations 3637 AERO Bridges can be either Commercial off-the Shelf (COTS) standard 3638 IP routers or virtual machines in the cloud. Bridges must be 3639 provisioned, supported and managed by the INET administrative 3640 authority, and connected to the Bridges of other INETs via inter- 3641 domain peerings. Cost for purchasing, configuring and managing 3642 Bridges is nominal even for very large OMNI links. 3644 AERO Servers can be standard dedicated server platforms, but most 3645 often will be deployed as virtual machines in the cloud. The only 3646 requirements for Servers are that they can run the AERO user-level 3647 code and have at least one network interface connection to the INET. 3648 As with Bridges, Servers must be provisioned, supported and managed 3649 by the INET administrative authority. Cost for purchasing, 3650 configuring and managing Servers is nominal especially for virtual 3651 Servers hosted in the cloud. 3653 AERO Proxys are most often standard dedicated server platforms with 3654 one network interface connected to the ANET and a second interface 3655 connected to an INET. As with Servers, the only requirements are 3656 that they can run the AERO user-level code and have at least one 3657 interface connection to the INET. Proxys must be provisioned, 3658 supported and managed by the ANET administrative authority. Cost for 3659 purchasing, configuring and managing Proxys is nominal, and borne by 3660 the ANET administrative authority. 3662 AERO Relays can be any dedicated server or COTS router platform 3663 connected to INETs and/or EUNs. The Relay connects to the OMNI link 3664 and engages in eBGP peering with one or more Bridges as a stub AS. 3665 The Relay then injects its MNPs and/or non-MNP prefixes into the BGP 3666 routing system, and provisions the prefixes to its downstream- 3667 attached networks. The Relay can perform ROS/ROR services the same 3668 as for any Server, and can route between the MNP and non-MNP address 3669 spaces. 3671 A.5. AERO Server Failure Implications 3673 AERO Servers may appear as a single point of failure in the 3674 architecture, but such is not the case since all Servers on the link 3675 provide identical services and loss of a Server does not imply 3676 immediate and/or comprehensive communication failures. Although 3677 Clients typically associate with a single Server at a time, Server 3678 failure is quickly detected and conveyed by Bidirectional Forward 3679 Detection (BFD) and/or proactive NUD allowing Clients to migrate to 3680 new Servers. 3682 If a Server fails, ongoing packet forwarding to Clients will continue 3683 by virtue of the asymmetric neighbor cache entries that have already 3684 been established in route optimization sources (ROSs). If a Client 3685 also experiences mobility events at roughly the same time the Server 3686 fails, unsolicited NA messages may be lost but proxy neighbor cache 3687 entries in the DEPARTED state will ensure that packet forwarding to 3688 the Client's new locations will continue for up to DepartTime 3689 seconds. 3691 If a Client is left without a Server for an extended timeframe (e.g., 3692 greater than ReachableTime seconds) then existing asymmetric neighbor 3693 cache entries will eventually expire and both ongoing and new 3694 communications will fail. The original source will continue to 3695 retransmit until the Client has established a new Server 3696 relationship, after which time continuous communications will resume. 3698 Therefore, providing many Servers on the link with high availability 3699 profiles provides resilience against loss of individual Servers and 3700 assurance that Clients can establish new Server relationships quickly 3701 in event of a Server failure. 3703 A.6. AERO Client / Server Architecture 3705 The AERO architectural model is client / server in the control plane, 3706 with route optimization in the data plane. The same as for common 3707 Internet services, the AERO Client discovers the addresses of AERO 3708 Servers and selects one Server to connect to. The AERO service is 3709 analogous to common Internet services such as google.com, yahoo.com, 3710 cnn.com, etc. However, there is only one AERO service for the link 3711 and all Servers provide identical services. 3713 Common Internet services provide differing strategies for advertising 3714 server addresses to clients. The strategy is conveyed through the 3715 DNS resource records returned in response to name resolution queries. 3716 As of January 2020 Internet-based 'nslookup' services were used to 3717 determine the following: 3719 o When a client resolves the domainname "google.com", the DNS always 3720 returns one A record (i.e., an IPv4 address) and one AAAA record 3721 (i.e., an IPv6 address). The client receives the same addresses 3722 each time it resolves the domainname via the same DNS resolver, 3723 but may receive different addresses when it resolves the 3724 domainname via different DNS resolvers. But, in each case, 3725 exactly one A and one AAAA record are returned. 3727 o When a client resolves the domainname "ietf.org", the DNS always 3728 returns one A record and one AAAA record with the same addresses 3729 regardless of which DNS resolver is used. 3731 o When a client resolves the domainname "yahoo.com", the DNS always 3732 returns a list of 4 A records and 4 AAAA records. Each time the 3733 client resolves the domainname via the same DNS resolver, the same 3734 list of addresses are returned but in randomized order (i.e., 3735 consistent with a DNS round-robin strategy). But, interestingly, 3736 the same addresses are returned (albeit in randomized order) when 3737 the domainname is resolved via different DNS resolvers. 3739 o When a client resolves the domainname "amazon.com", the DNS always 3740 returns a list of 3 A records and no AAAA records. As with 3741 "yahoo.com", the same three A records are returned from any 3742 worldwide Internet connection point in randomized order. 3744 The above example strategies show differing approaches to Internet 3745 resilience and service distribution offered by major Internet 3746 services. The Google approach exposes only a single IPv4 and a 3747 single IPv6 address to clients. Clients can then select whichever IP 3748 protocol version offers the best response, but will always use the 3749 same IP address according to the current Internet connection point. 3750 This means that the IP address offered by the network must lead to a 3751 highly-available server and/or service distribution point. In other 3752 words, resilience is predicated on high availability within the 3753 network and with no client-initiated failovers expected (i.e., it is 3754 all-or-nothing from the client's perspective). However, Google does 3755 provide for worldwide distributed service distribution by virtue of 3756 the fact that each Internet connection point responds with a 3757 different IPv6 and IPv4 address. The IETF approach is like google 3758 (all-or-nothing from the client's perspective), but provides only a 3759 single IPv4 or IPv6 address on a worldwide basis. This means that 3760 the addresses must be made highly-available at the network level with 3761 no client failover possibility, and if there is any worldwide service 3762 distribution it would need to be conducted by a network element that 3763 is reached via the IP address acting as a service distribution point. 3765 In contrast to the Google and IETF philosophies, Yahoo and Amazon 3766 both provide clients with a (short) list of IP addresses with Yahoo 3767 providing both IP protocol versions and Amazon as IPv4-only. The 3768 order of the list is randomized with each name service query 3769 response, with the effect of round-robin load balancing for service 3770 distribution. With a short list of addresses, there is still 3771 expectation that the network will implement high availability for 3772 each address but in case any single address fails the client can 3773 switch over to using a different address. The balance then becomes 3774 one of function in the network vs function in the end system. 3776 The same implications observed for common highly-available services 3777 in the Internet apply also to the AERO client/server architecture. 3778 When an AERO Client connects to one or more ANETs, it discovers one 3779 or more AERO Server addresses through the mechanisms discussed in 3780 earlier sections. Each Server address presumably leads to a fault- 3781 tolerant clustering arrangement such as supported by Linux-HA, 3782 Extended Virtual Synchrony or Paxos. Such an arrangement has 3783 precedence in common Internet service deployments in lightweight 3784 virtual machines without requiring expensive hardware deployment. 3785 Similarly, common Internet service deployments set service IP 3786 addresses on service distribution points that may relay requests to 3787 many different servers. 3789 For AERO, the expectation is that a combination of the Google/IETF 3790 and Yahoo/Amazon philosophies would be employed. The AERO Client 3791 connects to different ANET access points and can receive 1-2 Server 3792 LLAs at each point. It then selects one AERO Server address, and 3793 engages in RS/RA exchanges with the same Server from all ANET 3794 connections. The Client remains with this Server unless or until the 3795 Server fails, in which case it can switch over to an alternate 3796 Server. The Client can likewise switch over to a different Server at 3797 any time if there is some reason for it to do so. So, the AERO 3798 expectation is for a balance of function in the network and end 3799 system, with fault tolerance and resilience at both levels. 3801 Appendix B. Change Log 3803 << RFC Editor - remove prior to publication >> 3805 Changes from draft-templin-intarea-6706bis-61 to draft-templin- 3806 intrea-6706bis-62: 3808 o New sub-section on OMNI Neighbor Interface Attributes 3810 Changes from draft-templin-intarea-6706bis-59 to draft-templin- 3811 intrea-6706bis-60: 3813 o Removed all references to S/TLLAO - all Interface Attributes are 3814 now maintained completely in the OMNI option. 3816 Changes from draft-templin-intarea-6706bis-58 to draft-templin- 3817 intrea-6706bis-59: 3819 o The term "Relay" used in older draft versions is now "Bridge". 3820 "Relay" now refers to what was formally called: "Gateway". 3822 o Fine-grained cleanup of Forwarding Algorithm; IPv6 ND message 3823 addressing; OMNI Prefix Lengths, etc. 3825 Changes from draft-templin-intarea-6706bis-54 to draft-templin- 3826 intrea-6706bis-55: 3828 o Updates on Segment Routing and S/TLLAO contents. 3830 o Various editorials and addressing cleanups. 3832 Changes from draft-templin-intarea-6706bis-52 to draft-templin- 3833 intrea-6706bis-53: 3835 o Normative reference to the OMNI spec, and remove portions that are 3836 already specified in OMNI. 3838 o Renamed "AERO interface/link" to "OMIN interface/link" throughout 3839 the document. 3841 o Truncated obsolete back section matter. 3843 Author's Address 3845 Fred L. Templin (editor) 3846 Boeing Research & Technology 3847 P.O. Box 3707 3848 Seattle, WA 98124 3849 USA 3851 Email: fltemplin@acm.org