idnits 2.17.1 draft-templin-intarea-6706bis-77.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 14, 2020) is 1227 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: 'ULA' is mentioned on line 949, but not defined == Missing Reference: 'ULA0' is mentioned on line 955, but not defined == Missing Reference: 'ULA1' is mentioned on line 969, but not defined == Missing Reference: 'ULA2' is mentioned on line 959, but not defined == Missing Reference: 'ULA3' is mentioned on line 961, but not defined == Unused Reference: 'RFC3971' is defined on line 3256, but no explicit reference was found in the text == Unused Reference: 'RFC3972' is defined on line 3261, but no explicit reference was found in the text == Unused Reference: 'BGP' is defined on line 3309, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-comp-rtg-hdr' is defined on line 3312, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-crh-helper-opt' is defined on line 3318, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-frag-fragile' is defined on line 3323, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-dhcpv6-ndopt' is defined on line 3340, but no explicit reference was found in the text == Unused Reference: 'RFC4982' is defined on line 3462, but no explicit reference was found in the text == Unused Reference: 'RFC6273' is defined on line 3518, but no explicit reference was found in the text == Outdated reference: A later version (-99) exists of draft-templin-6man-omni-interface-57 == 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 14, 2020 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: June 17, 2021 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-intarea-6706bis-77 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 17, 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), Unique Local Addresses 71 (ULAs) and Global Unicast Addresses (GUA) . . . . . . 14 72 3.2.3. AERO Routing System . . . . . . . . . . . . . . . . . 15 73 3.2.4. OMNI Link Encapsulation . . . . . . . . . . . . . . . 16 74 3.2.5. Segment Routing Topologies (SRTs) . . . . . . . . . . 20 75 3.2.6. Segment Routing For OMNI Link Selection . . . . . . . 21 76 3.2.7. Segment Routing Within the OMNI Link . . . . . . . . 21 77 3.3. OMNI Interface Characteristics . . . . . . . . . . . . . 22 78 3.4. OMNI Interface Initialization . . . . . . . . . . . . . . 24 79 3.4.1. AERO Server/Relay Behavior . . . . . . . . . . . . . 24 80 3.4.2. AERO Proxy Behavior . . . . . . . . . . . . . . . . . 25 81 3.4.3. AERO Client Behavior . . . . . . . . . . . . . . . . 25 82 3.4.4. AERO Bridge Behavior . . . . . . . . . . . . . . . . 25 83 3.5. OMNI Interface Neighbor Cache Maintenance . . . . . . . . 25 84 3.5.1. OMNI Neighbor Interface Attributes . . . . . . . . . 27 85 3.5.2. OMNI Neighbor Advertisement Message Flags . . . . . . 28 86 3.6. OMNI Interface Encapsulation and Re-encapsulation . . . . 28 87 3.7. OMNI Interface Decapsulation . . . . . . . . . . . . . . 30 88 3.8. OMNI Interface Data Origin Authentication . . . . . . . . 30 89 3.9. OMNI Adaptation Layer and OMNI Interface MTU . . . . . . 30 90 3.10. OMNI Interface Forwarding Algorithm . . . . . . . . . . . 31 91 3.10.1. Client Forwarding Algorithm . . . . . . . . . . . . 32 92 3.10.2. Proxy Forwarding Algorithm . . . . . . . . . . . . . 32 93 3.10.3. Server/Relay Forwarding Algorithm . . . . . . . . . 33 94 3.10.4. Bridge Forwarding Algorithm . . . . . . . . . . . . 34 96 3.11. OMNI Interface Error Handling . . . . . . . . . . . . . . 35 97 3.12. AERO Router Discovery, Prefix Delegation and 98 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 38 99 3.12.1. AERO Service Model . . . . . . . . . . . . . . . . . 38 100 3.12.2. AERO Client Behavior . . . . . . . . . . . . . . . . 38 101 3.12.3. AERO Server Behavior . . . . . . . . . . . . . . . . 40 102 3.13. The AERO Proxy . . . . . . . . . . . . . . . . . . . . . 43 103 3.13.1. Combined Proxy/Servers . . . . . . . . . . . . . . . 45 104 3.13.2. Detecting and Responding to Server Failures . . . . 46 105 3.13.3. Point-to-Multipoint Server Coordination . . . . . . 46 106 3.14. AERO Route Optimization / Address Resolution . . . . . . 47 107 3.14.1. Route Optimization Initiation . . . . . . . . . . . 48 108 3.14.2. Relaying the NS . . . . . . . . . . . . . . . . . . 48 109 3.14.3. Processing the NS and Sending the NA . . . . . . . . 48 110 3.14.4. Relaying the NA . . . . . . . . . . . . . . . . . . 49 111 3.14.5. Processing the NA . . . . . . . . . . . . . . . . . 50 112 3.14.6. Route Optimization Maintenance . . . . . . . . . . . 50 113 3.15. Neighbor Unreachability Detection (NUD) . . . . . . . . . 51 114 3.16. Mobility Management and Quality of Service (QoS) . . . . 52 115 3.16.1. Mobility Update Messaging . . . . . . . . . . . . . 53 116 3.16.2. Announcing Link-Layer Address and/or QoS Preference 117 Changes . . . . . . . . . . . . . . . . . . . . . . 54 118 3.16.3. Bringing New Links Into Service . . . . . . . . . . 55 119 3.16.4. Deactivating Existing Links . . . . . . . . . . . . 55 120 3.16.5. Moving Between Servers . . . . . . . . . . . . . . . 55 121 3.17. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 56 122 3.17.1. Source-Specific Multicast (SSM) . . . . . . . . . . 57 123 3.17.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 58 124 3.17.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 59 125 3.18. Operation over Multiple OMNI Links . . . . . . . . . . . 59 126 3.19. DNS Considerations . . . . . . . . . . . . . . . . . . . 60 127 3.20. Transition Considerations . . . . . . . . . . . . . . . . 60 128 3.21. Detecting and Reacting to Server and Bridge Failures . . 61 129 3.22. AERO Clients on the Open Internet . . . . . . . . . . . . 62 130 3.23. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . 64 131 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 65 132 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 65 133 6. Security Considerations . . . . . . . . . . . . . . . . . . . 65 134 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 67 135 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 68 136 8.1. Normative References . . . . . . . . . . . . . . . . . . 68 137 8.2. Informative References . . . . . . . . . . . . . . . . . 70 138 Appendix A. Non-Normative Considerations . . . . . . . . . . . . 76 139 A.1. Implementation Strategies for Route Optimization . . . . 76 140 A.2. Implicit Mobility Management . . . . . . . . . . . . . . 76 141 A.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 77 142 A.4. AERO Critical Infrastructure Considerations . . . . . . . 77 143 A.5. AERO Server Failure Implications . . . . . . . . . . . . 78 144 A.6. AERO Client / Server Architecture . . . . . . . . . . . . 79 145 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 81 146 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 82 148 1. Introduction 150 Asymmetric Extended Route Optimization (AERO) fulfills the 151 requirements of Distributed Mobility Management (DMM) [RFC7333] and 152 route optimization [RFC5522] for aeronautical networking and other 153 network mobility use cases such as intelligent transportation 154 systems. AERO is an internetworking and mobility management service 155 based on the Overlay Multilink Network Interface (OMNI) 156 [I-D.templin-6man-omni-interface] Non-Broadcast, Multiple Access 157 (NBMA) virtual link model. The OMNI link is a virtual overlay 158 configured over one or more underlying Internetworks, and nodes on 159 the link can exchange IP packets via tunneling. The OMNI Adaptation 160 Layer (OAL) supports multilink operation for increased reliability, 161 bandwidth optimization and traffic path selection while accommodating 162 Maximum Transmission Unit (MTU) diversity. 164 The AERO service comprises Clients, Proxys, Servers and Relays that 165 are seen as OMNI link neighbors as well as Bridges that interconnect 166 OMNI link segments. Each node's OMNI interface uses an IPv6 link- 167 local address format that supports operation of the IPv6 Neighbor 168 Discovery (ND) protocol [RFC4861] and links ND to IP forwarding. A 169 node's OMNI interface can be configured over multiple underlying 170 interfaces, and may therefore appear as a single interface with 171 multiple link-layer addresses. Each link-layer address is subject to 172 change due to mobility and/or QoS fluctuations, and link-layer 173 address changes are signaled by ND messaging the same as for any IPv6 174 link. 176 AERO provides a cloud-based service where mobile nodes may use any 177 Server acting as a Mobility Anchor Point (MAP) and fixed nodes may 178 use any Relay on the link for efficient communications. Fixed nodes 179 forward packets destined to other AERO nodes to the nearest Relay, 180 which forwards them through the cloud. A mobile node's initial 181 packets are forwarded through the Server, while direct routing is 182 supported through asymmetric extended route optimization while data 183 packets are flowing. Both unicast and multicast communications are 184 supported, and mobile nodes may efficiently move between locations 185 while maintaining continuous communications with correspondents and 186 without changing their IP Address. 188 AERO Bridges are interconnected in a secured private BGP overlay 189 routing instance using encapsulation to provide a hybrid routing/ 190 bridging service that joins the underlying Internetworks of multiple 191 disjoint administrative domains into a single unified OMNI link. 193 Each OMNI link instance is characterized by the set of Mobility 194 Service Prefixes (MSPs) common to all mobile nodes. The link extends 195 to the point where a Relay/Server is on the optimal route from any 196 correspondent node on the link, and provides a conduit between the 197 underlying Internetwork and the OMNI link. To the underlying 198 Internetwork, the Relay/Server is the source of a route to the MSP, 199 and hence uplink traffic to the mobile node is naturally routed to 200 the nearest Relay/Server. 202 AERO assumes the use of PIM Sparse Mode in support of multicast 203 communication. In support of Source Specific Multicast (SSM) when a 204 Mobile Node is the source, AERO route optimization ensures that a 205 shortest-path multicast tree is established with provisions for 206 mobility and multilink operation. In all other multicast scenarios 207 there are no AERO dependencies. 209 AERO was designed for aeronautical networking for both manned and 210 unmanned aircraft, where the aircraft is treated as a mobile node 211 that can connect an Internet of Things (IoT). AERO is also 212 applicable to a wide variety of other use cases. For example, it can 213 be used to coordinate the Virtual Private Network (VPN) links of 214 mobile nodes (e.g., cellphones, tablets, laptop computers, etc.) that 215 connect into a home enterprise network via public access networks 216 using services such as OpenVPN [OVPN]. It can also be used to 217 facilitate vehicular and pedestrian communications services for 218 intelligent transportation systems. Other applicable use cases are 219 also in scope. 221 The following numbered sections present the AERO specification. The 222 appendices at the end of the document are non-normative. 224 2. Terminology 226 The terminology in the normative references applies; especially, the 227 terminology in the OMNI specification 228 [I-D.templin-6man-omni-interface] is used extensively throughout. 229 The following terms are defined within the scope of this document: 231 IPv6 Neighbor Discovery (ND) 232 an IPv6 control message service for coordinating neighbor 233 relationships between nodes connected to a common link. AERO uses 234 the ND service specified in [RFC4861]. 236 IPv6 Prefix Delegation 237 a networking service for delegating IPv6 prefixes to nodes on the 238 link. The nominal service is DHCPv6 [RFC8415], however alternate 239 services (e.g., based on ND messaging) are also in scope. Most 240 notably, a minimal form of prefix delegation known as "prefix 241 registration" can be used if the Client knows its prefix in 242 advance and can represent it in the IPv6 source address of an ND 243 message. 245 Access Network (ANET) 246 a node's first-hop data link service network (e.g., a radio access 247 network, cellular service provider network, corporate enterprise 248 network, etc.) that often provides link-layer security services 249 such as IEEE 802.1X and physical-layer security prevent 250 unauthorized access internally and with border network-layer 251 security services such as firewalls and proxies that prevent 252 unauthorized outside access. 254 ANET interface 255 a node's attachment to a link in an ANET. 257 Internetwork (INET) 258 a connected IP network topology with a coherent routing and 259 addressing plan and that provides a transit backbone service for 260 ANET end systems. INETs also provide an underlay service over 261 which the AERO virtual link is configured. Example INETs include 262 corporate enterprise networks, aviation networks, and the public 263 Internet itself. When there is no administrative boundary between 264 an ANET and the INET, the ANET and INET are one and the same. 266 INET Partition 267 frequently, INETs such as large corporate enterprise networks are 268 sub-divided internally into separate isolated partitions. Each 269 partition is fully connected internally but disconnected from 270 other partitions, and there is no requirement that separate 271 partitions maintain consistent Internet Protocol and/or addressing 272 plans. (Each INET partition is seen as a separate OMNI link 273 segment as discussed below.) 275 INET interface 276 a node's attachment to a link in an INET. 278 INET address 279 an IP address assigned to a node's interface connection to an 280 INET. 282 INET encapsulation 283 the encapsulation of a packet in an outer header or headers that 284 can be routed within the scope of the local INET partition. 286 OMNI link 287 the same as defined in [I-D.templin-6man-omni-interface], and 288 manifested by IPv6 encapsulation [RFC2473]. The OMNI link spans 289 underlying INET segments joined by virtual bridges in a spanning 290 tree the same as a bridged campus LAN. AERO nodes on the OMNI 291 link appear as single-hop neighbors even though they may be 292 separated by multiple underlying INET hops, and can use Segment 293 Routing [RFC8402] to cause packets to visit selected waypoints on 294 the link. 296 OMNI Interface 297 a node's attachment to an OMNI link. Since the addresses assigned 298 to an OMNI interface are managed for uniqueness, OMNI interfaces 299 do not require Duplicate Address Detection (DAD) and therefore set 300 the administrative variable 'DupAddrDetectTransmits' to zero 301 [RFC4862]. 303 OMNI Adaptation Layer (OAL) 304 an OMNI interface process whereby packets admitted into the 305 interface are wrapped in a mid-layer IPv6 header and fragmented/ 306 reassembled if necessary to support the OMNI link Maximum 307 Transmission Unit (MTU). The OAL is also responsible for 308 generating MTU-related control messages as necessary, and for 309 providing addressing context for spanning multiple segments of a 310 bridged OMNI link. 312 OMNI Link-Local Address (LLA) 313 a link local IPv6 address per [RFC4291] constructed as specified 314 in Section 3.2.2. 316 OMNI Unique-Local Address (ULA) 317 an IPv6 address from the IPv6 ULA prefix fd00::/8 [RFC4193], and 318 constructed as specified in [I-D.templin-6man-omni-interface]. 319 OMNI ULAs are statelessly derived from OMNI LLAs, and vice-versa. 321 underlying interface 322 an ANET or INET interface over which an OMNI interface is 323 configured. 325 Mobility Service Prefix (MSP) 326 an IP prefix assigned to the OMNI link and from which more- 327 specific Mobile Network Prefixes (MNPs) are derived. For IPv6, 328 the MSP is typically taken from an IPv6 Globally Unqiue Address 329 (GUA) prefix, e.g., obtained from an assigned numbers registry. 331 Mobile Network Prefix (MNP) 332 an IP prefix allocated from an MSP and delegated to an AERO Client 333 or Relay. 335 AERO node 336 a node that is connected to an OMNI link and participates in the 337 AERO internetworking and mobility service. 339 AERO Client ("Client") 340 an AERO node that connects over one or more underlying interfaces 341 and requests MNP delegation/registration service from AERO 342 Servers. The Client assigns a Client LLA to the OMNI interface 343 for use in ND exchanges with other AERO nodes and forwards packets 344 to correspondents according to OMNI interface neighbor cache 345 state. 347 AERO Server ("Server") 348 an INET node that configures an OMNI interface to provide default 349 forwarding and mobility/multilink services for AERO Clients. The 350 Server assigns an administratively-provisioned LLA to its OMNI 351 interface to support the operation of the ND services, and 352 advertises all of its associated MNPs via BGP peerings with 353 Bridges. 355 AERO Relay ("Relay") 356 an AERO Server that also provides forwarding services between 357 nodes reached via the OMNI link and correspondents on other links. 358 AERO Relays are provisioned with MNPs (i.e., the same as for an 359 AERO Client) and run a dynamic routing protocol to discover any 360 non-MNP IP routes. In both cases, the Relay advertises the MSP(s) 361 to its downstream networks, and distributes all of its associated 362 MNPs and non-MNP IP routes via BGP peerings with Bridges (i.e., 363 the same as for an AERO Server). 365 AERO Bridge ("Bridge") 366 a node that provides hybrid routing/bridging services (as well as 367 a security trust anchor) for nodes on an OMNI link. As a router, 368 the Bridge forwards packets using standard IP forwarding. As a 369 bridge, the Bridge forwards packets over the OMNI link without 370 decrementing the IPv6 Hop Limit. AERO Bridges peer with Servers 371 and other Bridges to discover the full set of MNPs for the link as 372 well as any non-MNPs that are reachable via Relays. 374 AERO Proxy ("Proxy") 375 a node that provides proxying services between Clients in an ANET 376 and Servers in external INETs. The AERO Proxy is a conduit 377 between the ANET and external INETs in the same manner as for 378 common web proxies, and behaves in a similar fashion as for ND 379 proxies [RFC4389]. A node may be configured to act as either a 380 Proxy and/or a Server, depending on Client Server selection 381 criteria. 383 ingress tunnel endpoint (ITE) 384 an OMNI interface endpoint that injects encapsulated packets into 385 an OMNI link. 387 egress tunnel endpoint (ETE) 388 an OMNI interface endpoint that receives encapsulated packets from 389 an OMNI link. 391 link-layer address 392 an IP address used as an encapsulation header source or 393 destination address from the perspective of the OMNI interface. 394 When an upper layer protocol (e.g., UDP) is used as part of the 395 encapsulation, the port number is also considered as part of the 396 link-layer address. 398 network layer address 399 the source or destination address of an encapsulated IP packet 400 presented to the OMNI interface. 402 end user network (EUN) 403 an internal virtual or external edge IP network that an AERO 404 Client or Relay connects to the rest of the network via the OMNI 405 interface. The Client/Relay sees each EUN as a "downstream" 406 network, and sees the OMNI interface as the point of attachment to 407 the "upstream" network. 409 Mobile Node (MN) 410 an AERO Client and all of its downstream-attached networks that 411 move together as a single unit, i.e., an end system that connects 412 an Internet of Things. 414 Mobile Router (MR) 415 a MN's on-board router that forwards packets between any 416 downstream-attached networks and the OMNI link. 418 Route Optimization Source (ROS) 419 the AERO node nearest the source that initiates route 420 optimization. The ROS may be a Server or Proxy acting on behalf 421 of the source Client. 423 Route Optimization responder (ROR) 424 the AERO node nearest the target destination that responds to 425 route optimization requests. The ROR may be a Server acting on 426 behalf of a target MNP Client, or a Relay for a non-MNP 427 destination. 429 MAP List 430 a geographically and/or topologically referenced list of addresses 431 of all Servers within the same OMNI link. There is a single MAP 432 list for the entire OMNI link. 434 Distributed Mobility Management (DMM) 435 a BGP-based overlay routing service coordinated by Servers and 436 Bridges that tracks all Server-to-Client associations. 438 Mobility Service (MS) 439 the collective set of all Servers, Proxys, Bridges and Relays that 440 provide the AERO Service to Clients. 442 Mobility Service Endpoint MSE) 443 an individual Server, Proxy, Bridge or Relay in the Mobility 444 Service. 446 Throughout the document, the simple terms "Client", "Server", 447 "Bridge", "Proxy" and "Relay" refer to "AERO Client", "AERO Server", 448 "AERO Bridge", "AERO Proxy" and "AERO Relay", respectively. 449 Capitalization is used to distinguish these terms from other common 450 Internetworking uses in which they appear without capitalization. 452 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 453 the names of node variables, messages and protocol constants) is used 454 throughout this document. The terms "All-Routers multicast", "All- 455 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 456 anycast" are defined in [RFC4291]. Also, the term "IP" is used to 457 generically refer to either Internet Protocol version, i.e., IPv4 458 [RFC0791] or IPv6 [RFC8200]. 460 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 461 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 462 "OPTIONAL" in this document are to be interpreted as described in BCP 463 14 [RFC2119][RFC8174] when, and only when, they appear in all 464 capitals, as shown here. 466 3. Asymmetric Extended Route Optimization (AERO) 468 The following sections specify the operation of IP over OMNI links 469 using the AERO service: 471 3.1. AERO Node Types 473 AERO Clients are Mobile Nodes (MNs) that connect via underlying 474 interfaces with addresses that may change when the Client moves to a 475 new network connection point. AERO Clients register their Mobile 476 Network Prefixes (MNPs) with the AERO service, and distribute the 477 MNPs to nodes on EUNs. AERO Bridges, Servers, Proxys and Relays are 478 critical infrastructure elements in fixed (i.e., non-mobile) INET 479 deployments and hence have permanent and unchanging INET addresses. 480 Together, they constitute the AERO service which provides an OMNI 481 link virtual overlay for connecting AERO Clients. 483 AERO Bridges provide hybrid routing/bridging services (as well as a 484 security trust anchor) for nodes on an OMNI link. Bridges use 485 standard IPv6 routing to forward packets both within the same INET 486 partitions and between disjoint INET partitions based on a mid-layer 487 IPv6 encapsulation per [RFC2473]. The inner IP layer experiences a 488 virtual bridging service since the inner IP TTL/Hop Limit is not 489 decremented during forwarding. Each Bridge also peers with Servers 490 and other Bridges in a dynamic routing protocol instance to provide a 491 Distributed Mobility Management (DMM) service for the list of active 492 MNPs (see Section 3.2.3). Bridges present the OMNI link as a set of 493 one or more Mobility Service Prefixes (MSPs) and configure secured 494 tunnels with Servers, Relays, Proxys and other Bridges; they further 495 maintain IP forwarding table entries for each MNP and any other 496 reachable non-MNP prefixes. 498 AERO Servers provide default forwarding and mobility/multilink 499 services for AERO Client Mobile Nodes (MNs). Each Server also peers 500 with Bridges in a dynamic routing protocol instance to advertise its 501 list of associated MNPs (see Section 3.2.3). Servers facilitate 502 prefix delegation/registration exchanges with Clients, where each 503 delegated prefix becomes an MNP taken from an MSP. Servers forward 504 packets between OMNI interface neighbors and track each Client's 505 mobility profiles. Servers may further act as Servers for some sets 506 of Clients and as Proxies for others. 508 AERO Proxys provide a conduit for ANET Clients to associate with 509 Servers in external INETs. Client and Servers exchange control plane 510 messages via the Proxy acting as a bridge between the ANET/INET 511 boundary. The Proxy forwards data packets between Clients and the 512 OMNI link according to forwarding information in the neighbor cache. 513 The Proxy function is specified in Section 3.13. Proxys may further 514 act as Proxys for some sets of Clients and as Servers for others. 516 AERO Relays are Servers that provide forwarding services between the 517 OMNI interface and INET/EUN interfaces. Relays are provisioned with 518 MNPs the same as for an AERO Client, and also run a dynamic routing 519 protocol to discover any non-MNP IP routes. The Relay advertises the 520 MSP(s) to its connected networks, and distributes all of its 521 associated MNPs and non-MNP IP routes via BGP peerings with Bridges 523 3.2. The AERO Service over OMNI Links 525 3.2.1. AERO/OMNI Reference Model 527 Figure 1 presents the basic OMNI link reference model: 529 +----------------+ 530 | AERO Bridge B1 | 531 | Nbr: S1, S2, P1| 532 |(X1->S1; X2->S2)| 533 | MSP M1 | 534 +-+---------+--+-+ 535 +--------------+ | Secured | | +--------------+ 536 |AERO Server S1| | tunnels | | |AERO Server S2| 537 | Nbr: C1, B1 +-----+ | +-----+ Nbr: C2, B1 | 538 | default->B1 | | | default->B1 | 539 | X1->C1 | | | X2->C2 | 540 +-------+------+ | +------+-------+ 541 | OMNI link | | 542 X===+===+===================+==)===============+===+===X 543 | | | | 544 +-----+--------+ +--------+--+-----+ +--------+-----+ 545 |AERO Client C1| | AERO Proxy P1 | |AERO Client C2| 546 | Nbr: S1 | |(Proxy Nbr Cache)| | Nbr: S2 | 547 | default->S1 | +--------+--------+ | default->S2 | 548 | MNP X1 | | | MNP X2 | 549 +------+-------+ .--------+------. +-----+--------+ 550 | (- Proxyed Clients -) | 551 .-. `---------------' .-. 552 ,-( _)-. ,-( _)-. 553 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 554 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 555 `-(______)-' +-------+ +-------+ `-(______)-' 557 Figure 1: AERO/OMNI Reference Model 559 In this model: 561 o the OMNI link is an overlay network service configured over one or 562 more underlying INET partitions which may be managed by different 563 administrative authorities and have incompatible protocols and/or 564 addressing plans. 566 o AERO Bridge B1 aggregates Mobility Service Prefix (MSP) M1, 567 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 568 via BGP peerings over secured tunnels to Servers (S1, S2). 569 Bridges connect the disjoint segments of a partitioned OMNI link. 571 o AERO Servers/Relays S1 and S2 configure secured tunnels with 572 Bridge B1 and also provide mobility, multilink and default router 573 services for their associated Clients C1 and C2. 575 o AERO Clients C1 and C2 associate with Servers S1 and S2, 576 respectively. They receive Mobile Network Prefix (MNP) 577 delegations X1 and X2, and also act as default routers for their 578 associated physical or internal virtual EUNs. Simple hosts H1 and 579 H2 attach to the EUNs served by Clients C1 and C2, respectively. 581 o AERO Proxy P1 configures a secured tunnel with Bridge B1 and 582 provides proxy services for AERO Clients in secured enclaves that 583 cannot associate directly with other OMNI link neighbors. 585 An OMNI link configured over a single INET appears as a single 586 unified link with a consistent underlying network addressing plan. 587 In that case, all nodes on the link can exchange packets via simple 588 INET encapsulation, since the underlying INET is connected. In 589 common practice, however, an OMNI link may be partitioned into 590 multiple "segments", where each segment is a distinct INET 591 potentially managed under a different administrative authority (e.g., 592 as for worldwide aviation service providers such as ARINC, SITA, 593 Inmarsat, etc.). Individual INETs may also themselves be partitioned 594 internally, in which case each internal partition is seen as a 595 separate segment. 597 The addressing plan of each segment is consistent internally but will 598 often bear no relation to the addressing plans of other segments. 599 Each segment is also likely to be separated from others by network 600 security devices (e.g., firewalls, proxies, packet filtering 601 gateways, etc.), and in many cases disjoint segments may not even 602 have any common physical link connections. Therefore, nodes can only 603 be assured of exchanging packets directly with correspondents in the 604 same segment, and not with those in other segments. The only means 605 for joining the segments therefore is through inter-domain peerings 606 between AERO Bridges. 608 The same as for traditional campus LANs, multiple OMNI link segments 609 can be joined into a single unified link via a virtual bridging 610 service using the OMNI Adaptation Layer (OAL) which inserts a mid- 611 layer IPv6 encapsulation per [RFC2473] that supports inter-segment 612 forwarding (i.e., bridging) without decrementing the network-layer 613 TTL/Hop Limit. This bridging of OMNI link segments is shown in 614 Figure 2: 616 . . . . . . . . . . . . . . . . . . . . . . . 617 . . 618 . .-(::::::::) . 619 . .-(::::::::::::)-. +-+ . 620 . (:::: Segment A :::)--|B|---+ . 621 . `-(::::::::::::)-' +-+ | . 622 . `-(::::::)-' | . 623 . | . 624 . .-(::::::::) | . 625 . .-(::::::::::::)-. +-+ | . 626 . (:::: Segment B :::)--|B|---+ . 627 . `-(::::::::::::)-' +-+ | . 628 . `-(::::::)-' | . 629 . | . 630 . .-(::::::::) | . 631 . .-(::::::::::::)-. +-+ | . 632 . (:::: Segment C :::)--|B|---+ . 633 . `-(::::::::::::)-' +-+ | . 634 . `-(::::::)-' | . 635 . | . 636 . ..(etc).. x . 637 . . 638 . . 639 . <- OMNI link Bridged by encapsulation -> . 640 . . . . . . . . . . . . . .. . . . . . . . . 642 Figure 2: Bridging OMNI Link Segments 644 Bridges, Servers, Relays and Proxys connect via secured INET tunnels 645 over their respective segments in a spanning tree topology rooted at 646 the Bridges. The secured spanning tree supports strong 647 authentication for IPv6 ND control messages and may also be used to 648 convey the initial data packets in a flow. Route optimization can 649 then be employed to cause data packets to take more direct paths 650 between OMNI link neighbors without having to strictly follow the 651 spanning tree. 653 3.2.2. Link-Local Addresses (LLAs), Unique Local Addresses (ULAs) and 654 Global Unicast Addresses (GUA) 656 AERO nodes on OMNI links use the Link-Local Address (LLA) prefix 657 fe80::/64 [RFC4291] to assign LLAs used for network-layer addresses 658 in link-scoped IPv6 ND and data messages. They also use the Unique 659 Local Address (ULA) prefix fd00::/8 followed by a pseudo-random 660 40-bit global ID to form the prefix [ULA]::/48, then include a 16-bit 661 OMNI link identifier '*' to form the prefix [ULA*]::/64 to assign OAL 662 addresses. Finally, the MSPs and MNPs used by AERO nodes are 663 typically based on Global Unicast Addresses (GUAs), but in some cases 664 may be based on private-use addresses. See 665 [I-D.templin-6man-omni-interface] for a full specification of LLAs, 666 ULAs and GUAs used by AERO nodes on OMNI links. 668 3.2.3. AERO Routing System 670 The AERO routing system comprises a private instance of the Border 671 Gateway Protocol (BGP) [RFC4271] that is coordinated between Bridges 672 and Servers and does not interact with either the public Internet BGP 673 routing system or any underlying INET routing systems. 675 In a reference deployment, each Server is configured as an Autonomous 676 System Border Router (ASBR) for a stub Autonomous System (AS) using 677 an AS Number (ASN) that is unique within the BGP instance, and each 678 Server further uses eBGP to peer with one or more Bridges but does 679 not peer with other Servers. Each INET of a multi-segment OMNI link 680 must include one or more Bridges, which peer with the Servers and 681 Proxys within that INET. All Bridges within the same INET are 682 members of the same hub AS using a common ASN, and use iBGP to 683 maintain a consistent view of all active MNPs currently in service. 684 The Bridges of different INETs peer with one another using eBGP. 686 Bridges advertise the OMNI link's MSPs and any non-MNP routes to each 687 of their Servers. This means that any aggregated non-MNPs (including 688 "default") are advertised to all Servers. Each Bridge configures a 689 black-hole route for each of its MSPs. By black-holing the MSPs, the 690 Bridge will maintain forwarding table entries only for the MNPs that 691 are currently active, and packets destined to all other MNPs will 692 correctly incur Destination Unreachable messages due to the black- 693 hole route. In this way, Servers have only partial topology 694 knowledge (i.e., they know only about the MNPs of their directly 695 associated Clients) and they forward all other packets to Bridges 696 which have full topology knowledge. 698 Each OMNI link segment assigns a unique sub-prefix of [ULA*]::/96 699 known as the ULA partition prefix. For example, a first segment 700 could assign [ULA*]::1000/116, a second could assign 701 [ULA*]::2000/116, a third could assign [ULA*]::3000/116, etc. The 702 administrative authorities for each segment must therefore coordinate 703 to assure mutually-exclusive partition prefix assignments, but 704 internal provisioning of each prefix is an independent local 705 consideration for each administrative authority. For each partition 706 prefix, the Bridge(s) that connect that segment assign the all-zero's 707 address of the prefix as a Subnet Router Anycast address. For 708 example, the Subnet Router Anycast address for [ULA*]::1000/116 is 709 simply [ULA*]::1000. 711 ULA partition prefixes are statically represented in Bridge 712 forwarding tables. Bridges join multiple segments into a unified 713 OMNI link over multiple diverse administrative domains. They support 714 a bridging function by first establishing forwarding table entries 715 for their partition prefixes either via standard BGP routing or 716 static routes. For example, if three Bridges ('A', 'B' and 'C') from 717 different segments serviced [ULA*]::1000/116, [ULA*]::2000/116 and 718 [ULA*]::3000/116 respectively, then the forwarding tables in each 719 Bridge are as follows: 721 A: [ULA*]::1000/116->local, [ULA*]::2000/116->B, [ULA*]::3000/116->C 723 B: [ULA*]::1000/116->A, [ULA*]::2000/116->local, [ULA*]::3000/116->C 725 C: [ULA*]::1000/116->A, [ULA*]::2000/116->B, [ULA*]::3000/116->local 727 These forwarding table entries are permanent and never change, since 728 they correspond to fixed infrastructure elements in their respective 729 segments. 731 ULA Client prefixes are instead dynamically advertised in the AERO 732 routing system by Servers and Relays that provide service for their 733 corresponding MNPs. For example, if three Servers ('D', 'E' and 'F') 734 service the MNPs 2001:db8:1000:2000::/56, 2001:db8:3000:4000::/56 and 735 2001:db8:5000:6000::/56 then the routing system would include: 737 D: [ULA*]:2001:db8:1000:2000/120 739 E: [ULA*]:2001:db8:3000:4000/120 741 F: [ULA*]:2001:db8:5000:6000/120 743 A full discussion of the BGP-based routing system used by AERO is 744 found in [I-D.ietf-rtgwg-atn-bgp]. 746 3.2.4. OMNI Link Encapsulation 748 With the Client and partition prefixes in place in Bridge forwarding 749 tables, the OMNI interface sends control and data packets toward AERO 750 destination nodes located in different OMNI link segments over the 751 spanning tree via mid-layer encapsulation using the OMNI Adaptation 752 Layer (OAL) header based on Generic Packet Tunneling in IPv6 753 [RFC2473]. When necessary, the OMNI interface also includes an OMNI 754 Routing Header (ORH) as an extension to the OAL header if final 755 segment forwarding information is available, e.g., in the neighbor 756 cache. The ORH is formatted as shown in Figure 3: 758 0 1 2 3 759 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 1 760 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 761 | Next Header | Hdr Ext Len | Routing Type | SRT | FMT | 762 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 763 | Last Hop Segment-id (LHS) | 764 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 765 ~ Link Layer Address (L2ADDR) ~ 766 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 767 ~ Destination Suffix (if necessary) ~ 768 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 769 ~ Null Padding (if necessary) ~ 770 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 772 Figure 3: OMNI Routing Header (ORH) Format 774 In this format: 776 o Next Header identifies the type of header immediately following 777 the ORH. 779 o Hdr Ext Len is the length of the Routing header in 8-octet units 780 (not including the first 8 octets), with trailing padding added if 781 necessary to produce an integral number of 8-octet units. 783 o Routing Type is set to TBD (see IANA Considerations). 785 o Segments Left is omitted, and replaced by a 5-bit SRT and 3-bit 786 FMT field. 788 o SRT - a 5-bit Segment Routing Topology prefix length value that 789 (when added to 96) determines the prefix length to apply to the 790 ULA formed from concatenating [ULA*]::/96 with the 32 bit LHS MSID 791 value that follows. For example, the value 16 corresponds to the 792 prefix length 112. 794 o FMT - a 3-bit "Framework/Mode/Type" code corresponding to the 795 included Link Layer Address as follows: 797 * When the most significant bit (i.e., "Framework") is set to 0, 798 L2ADDR is the INET encapsulation address of the Server/Proxy 799 named in the LHS; otherwise, it is the address for the target 800 Client, and the Destination Suffix is included. 802 * When the next most significant bit (i.e., "Mode") is set to 0, 803 the Source/Target L2ADDR is on the open INET; otherwise, it is 804 (likely) located behind a Network Address Translator (NAT). 806 * When the least significant bit (i.e., "Type") is set to 0, 807 L2ADDR includes a UDP Port Number followed by an IPv4 address; 808 otherwise, it includes a UDP Port Number followed by an IPv6 809 address. 811 o LHS - the 32 bit MSID of a service node in the Last Hop Segment on 812 the path to the target. When SRT and LHS are both set to 0, the 813 LHS is considered unspecified. When SRT is set to 0 and LHS is 814 non-zero, the prefix length is set to 128. SRT and LHS provide 815 guidance to the OMNI interface forwarding algorithm. 816 Specifically, if SRT/LHS is located in the local OMNI link segment 817 then the OMNI interface can omit OAL/ORH encapsualtion and send 818 directly to the target using INET encapsulation according to FMT/ 819 L2ADDR; else, it must perform INET/OAL/ORH encapsulation and 820 forward according to the OMNI link spanning tree. 822 o Link Layer Address (L2ADDR) - Formatted according to FMT, and 823 identifies the link-layer address (i.e., the encapsulation 824 address) of the source/target. The UDP Port Number appears in the 825 first two octets and the IP address appears in the next 4 octets 826 for IPv4 or 16 octets for IPv6. The Port Number and IP address 827 are recorded in ones-compliment "obfuscated" form per [RFC4380]. 828 The OMNI interface forwarding algorithm uses FMT/L2ADDR to 829 determine the INET encapsulation address for local forwarding when 830 SRT/LHS is located in the same OMNI link segment. 832 o Destination Suffix is a 64-bit field included only for OAL- 833 encapsulated packets that are destined directly to the ULA of the 834 Client (i.e., according to the FMT code). When present, 835 Destination Suffix encodes the 64-bit ULA suffix for the Client 836 that will receive packet. For example, if the Client ULA is 837 [ULA*]:2001:db8:1:2 then the Destination suffix encodes the value 838 2001:db8:1:2. 840 o Null Padding contains zero-valued octets as necessary to pad the 841 ORH to an integral number of 8-octet units. 843 When an AERO node uses OAL encapsulation for a packet with addresses 844 such as 2001:db8:1:2::1 and 2001:db8:1234:5678::1, it sets the OAL 845 header source address to its own ULA address (e.g., 846 [ULA*]::1000:2000). The node also sets the destination address to 847 the ULA of the Client (e.g., [ULA*]::2001:db8:1234:5678) when the 848 Client is addressed directly; otherwise, it sets the destination 849 address to the ULA of the Client's Proxy/Server (e.g., 850 [ULA*]::4321:9876). If the neighbor cache includes Last Hop Segment 851 information for the target destination, the node next inserts an ORH 852 immediately following the OAL header while including the correct SRT, 853 FMT, LHS, L2ADDR and (if necessary) Destination Suffix information. 855 Next, the node overwrites the OAL header destination address with the 856 LHS Subnet Router Anycast address (for example, for LHS 1000:2000 857 with SRT 16, the Subnet Router Anycast address is [ULA*]::1000:0000). 859 The node then fragments the OAL/ORH packet if necessary, with each 860 resulting fragment including the OAL/ORH headers while only the first 861 fragment includes the original IPv6 header. The node finally 862 encapsulates each resulting OAL/ORH packet/fragment in an INET header 863 with source address set to its own INET address (e.g., 192.0.2.100) 864 and destination set to the INET address of a Bridge (e.g., 865 192.0.2.1). 867 The encapsulation format in the above example is shown in Figure 4: 869 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 870 | INET Header | 871 | src = 192.0.2.100 | 872 | dst = 192.0.2.1 | 873 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 874 | OAL Header | 875 | src = [ULA*]::1000:2000 | 876 | dst= ULA for inner IP dst | 877 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 878 | ORH Header | 879 | (if necessary) | 880 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 881 | Inner IP Header | 882 | src = 2001:db8:1:2::1 | 883 | dst = 2001:db8:1234:5678::1 | 884 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 885 | | 886 ~ ~ 887 ~ Inner Packet Body ~ 888 ~ ~ 889 | | 890 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 892 Figure 4: OAL/ORH Encapsulation 894 In this format, the inner IP header and packet body are the original 895 IP packet, the OAL header is an IPv6 header prepared according to 896 [RFC2473], the ORH is a Routing Header extension of the OAL header, 897 and the INET header is prepared as discussed in Section 3.6. 899 This gives rise to a routing system that contains both Client prefix 900 routes that may change dynamically due to regional node mobility and 901 partition prefix routes that rarely if ever change. The Bridges can 902 therefore provide link-layer bridging by sending packets over the 903 spanning tree instead of network-layer routing according to MNP 904 routes. As a result, opportunities for packet loss due to node 905 mobility between different segments are mitigated. 907 In normal operations, IPv6 ND messages are conveyed over secured 908 paths between OMNI link neighbors so that specific Proxys, Servers or 909 Relays can be addressed without being subject to mobility events. 910 Conversely, only the first few packets destined to Clients need to 911 traverse secured paths until route optimization can determine a more 912 direct path. 914 Note: An IPv6 "minimal encapsulation" format (i.e., an IPv6 variant 915 of [RFC2004]) based on extensions to the ORH was considered, analyzed 916 and rejected. In the approach, the ORH would be inserted as an 917 extension header to the original IPv6 packet header. The IPv6 918 destination address would then be written into the ORH, and the ULA 919 corresponding to the destination would be overwritten in the IPv6 920 destination address. This would seemingly convey enough forwarding 921 information so that OAL encapsulation could be avoided. However, 922 this "minimal encapsulation" IPv6 packet would then have a non-ULA 923 source address and ULA destination address, an incorrect value in 924 upper layer protocol checksums, and a Hop Limit that is decremented 925 within the spanning tree when it should not be. The insertion and 926 removal of the ORH would also entail rewriting the Payload Length and 927 Next Header fields - again, invalidating upper layer checksums. 928 These irregularities would result in implementation challenges and 929 the potential for operational issues, e.g., since actionable ICMPv6 930 error reports could not be delivered to the original source. In 931 order to address the issues, still more information such as the 932 original IPv6 source address could be written into the ORH. However, 933 with the additional information the benefit of the "minimal 934 encapsulation" savings quickly diminishes, and becomes overshadowed 935 by the implementation and operational irregularities. 937 3.2.5. Segment Routing Topologies (SRTs) 939 The 64-bit sub-prefixes of [ULA]::/48 identify up to 2^16 distinct 940 Segment Routing Topologies (SRTs). Each SRT is a mutually-exclusive 941 OMNI link overlay instance using a distinct set of ULAs, and emulates 942 a Virtual LAN (VLAN) service for the OMNI link. In some cases (e.g., 943 when redundant topologies are needed for fault tolerance and 944 reliability) it may be beneficial to deploy multiple SRTs that act as 945 independent overlay instances. A communication failure in one 946 instance therefore will not affect communications in other instances. 948 Each SRT is identified by a distinct value in bits 48-63 of 949 [ULA]::/48, i.e., as [ULA0]::/64, [ULA1]::/64, [ULA2]::/64, etc. 950 This document asserts that up to four SRTs provide a level of safety 951 sufficient for critical communications such as civil aviation. Each 952 SRT is designated with a color that identifies a different OMNI link 953 instance as follows: 955 o Red - corresponds to [ULA0]::/64 957 o Green - corresponds to [ULA1]::/64 959 o Blue-1 - corresponds to [ULA2]::/64 961 o Blue-2 - corresponds to [ULA3]::/64 963 o the remaining [ULA*]::/64 sub prefixes are available for 964 additional SRTs. 966 Each OMNI interface is identified by a unique interface name (e.g., 967 omni0, omni1, omni2, etc.) and assigns an anycast ULA corresponding 968 to its SRT prefix. For example, the anycast ULA for the Green SRT is 969 simply [ULA1]::. The anycast ULA is used for OMNI interface 970 determination in Safety-Based Multilink (SBM) as discussed in 971 [I-D.templin-6man-omni-interface]. Each OMNI interface further 972 applies Performance-Based Multilink (PBM) internally. 974 3.2.6. Segment Routing For OMNI Link Selection 976 An original IPv6 source can direct an IPv6 packet to an AERO node by 977 including a standard IPv6 Segment Routing Header (SRH) [RFC8754] with 978 the anycast ULA for the selected SRT as either the IPv6 destination 979 or as an intermediate hop within the SRH. This allows the original 980 source to determine the specific OMNI link topology a packet will 981 traverse when there may be multiple alternatives. 983 When the AERO node processes the SRH and forwards the packet to the 984 correct OMNI interface, the OMNI interface writes the next IPv6 985 address from the SRH into the IPv6 destination address and decrements 986 Segments Left. If decrementing would cause Segments Left to become 987 0, the OMNI interface deletes the SRH before forwarding. This form 988 of Segment Routing supports Safety-Based Multilink (SBM). 990 3.2.7. Segment Routing Within the OMNI Link 992 AERO node OMNI interfaces can insert OAL/ORH headers for Segment 993 Routing within the OMNI link to influence the paths of packets 994 destined to targets in remote segments without requiring all packets 995 to traverse strict spanning tree paths. 997 When an AERO node's OMNI interface has a packet to send to a target 998 discovered through route optimization located in the same OMNI link 999 segment, it encapsulates the packet in OAL/ORH headers if necessary 1000 as discussed above. The node then uses the target's Link Layer 1001 Address (L2ADDR) information for INET encapsulation. 1003 When an AERO node's OMNI interface has a packet to send to a route 1004 optimization target located in a remote OMNI link segment, it 1005 encapsulates the packet in OAL/ORH headers as discussed above while 1006 forwarding the packet to a Bridge with destination set to the Subnet 1007 Router Anycast address for the final OMNI link segment. 1009 When a Bridge receives a packet destined to its Subnet Router Anycast 1010 address with an OAL/ORH with SRT/LHS values corresponding to the 1011 local segment, it examines the L2ADDR according to FMT and removes 1012 the ORH from the packet. If the ORH includes a saved Destination 1013 Suffix, the Bridge then writes the corresponding ULA into the OAL 1014 destination address; otherwise, it writes the ULA corresponding to 1015 the SRT/LHS fields into the destination. The Bridge then 1016 encapsulates the packet in an INET header according to L2ADDR and 1017 forwards the packet within the INET either to the LHS Server/Proxy or 1018 directly to the destination itself. In this way, the Bridge 1019 participates in route optimization to reduce traffic load and 1020 suboptimal routing through strict spanning tree paths. 1022 3.3. OMNI Interface Characteristics 1024 OMNI interfaces are virtual interfaces configured over one or more 1025 underlying interfaces classified as follows: 1027 o INET interfaces connect to an INET either natively or through one 1028 or several IPv4 Network Address Translators (NATs). Native INET 1029 interfaces have global IP addresses that are reachable from any 1030 INET correspondent. All Server, Relay and Bridge interfaces are 1031 native interfaces, as are INET-facing interfaces of Proxys. NATed 1032 INET interfaces connect to a private network behind one or more 1033 NATs that provide INET access. Clients that are behind a NAT are 1034 required to send periodic keepalive messages to keep NAT state 1035 alive when there are no data packets flowing. 1037 o ANET interfaces connect to an ANET that is separated from the open 1038 INET by a Proxy. Proxys can actively issue control messages over 1039 the INET on behalf of the Client to reduce ANET congestion. 1041 o VPNed interfaces use security encapsulation over the INET to a 1042 Virtual Private Network (VPN) server that also acts as a Server or 1043 Proxy. Other than the link-layer encapsulation format, VPNed 1044 interfaces behave the same as Direct interfaces. 1046 o Direct interfaces connect a Client directly to a Server or Proxy 1047 without crossing any ANET/INET paths. An example is a line-of- 1048 sight link between a remote pilot and an unmanned aircraft. The 1049 same Client considerations apply as for VPNed interfaces. 1051 OMNI interfaces use OAL/ORH encapsulation as necessary as discussed 1052 in Section 3.2.4. OMNI interfaces use link-layer encapsulation (see: 1053 Section 3.6) to exchange packets with OMNI link neighbors over INET 1054 or VPNed interfaces as well as over ANET interfaces for which the 1055 Client and Proxy may be multiple IP hops away. OMNI interfaces do 1056 not use link-layer encapsulation over Direct underlying interfaces or 1057 ANET interfaces when the Client and Proxy are known to be on the same 1058 underlying link. 1060 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 1061 state the same as for any interface. OMNI interfaces use ND messages 1062 including Router Solicitation (RS), Router Advertisement (RA), 1063 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 1064 neighbor cache management. 1066 OMNI interfaces send ND messages with an OMNI option formatted as 1067 specified in [I-D.templin-6man-omni-interface]. The OMNI option 1068 includes prefix registration information and Interface Attributes 1069 containing link information parameters for the OMNI interface's 1070 underlying interfaces. Each OMNI option may include multiple 1071 Interface Attributes sub-options, each identified by an ifIndex 1072 value. 1074 A Client's OMNI interface may be configured over multiple underlying 1075 interface connections. For example, common mobile handheld devices 1076 have both wireless local area network ("WLAN") and cellular wireless 1077 links. These links are often used "one at a time" with low-cost WLAN 1078 preferred and highly-available cellular wireless as a standby, but a 1079 simultaneous-use capability could provide benefits. In a more 1080 complex example, aircraft frequently have many wireless data link 1081 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1082 directional, etc.) with diverse performance and cost properties. 1084 If a Client's multiple underlying interfaces are used "one at a time" 1085 (i.e., all other interfaces are in standby mode while one interface 1086 is active), then ND message OMNI options include only a single 1087 Interface Attributes sub-option set to constant values. In that 1088 case, the Client would appear to have a single interface but with a 1089 dynamically changing link-layer address. 1091 If the Client has multiple active underlying interfaces, then from 1092 the perspective of ND it would appear to have multiple link-layer 1093 addresses. In that case, ND message OMNI options MAY include 1094 multiple Interface Attributes sub-options - each with values that 1095 correspond to a specific interface. Every ND message need not 1096 include Interface Attributes for all underlying interfaces; for any 1097 attributes not included, the neighbor considers the status as 1098 unchanged. 1100 Bridge, Server and Proxy OMNI interfaces may be configured over one 1101 or more secured tunnel interfaces. The OMNI interface configures 1102 both an LLA and its corresponding ULA, while the underlying secured 1103 tunnel interfaces are either unnumbered or configure the same ULA. 1104 The OMNI interface encapsulates each IP packet in OAL/ORH headers and 1105 presents the packet to the underlying secured tunnel interface. 1106 Routing protocols such as BGP that run over the OMNI interface do not 1107 employ OAL/ORH encapsulation, but rather present the routing protocol 1108 messages directly to the underlying secured tunnels while using the 1109 ULA as the source address. This distinction must be honored 1110 consistently according to each node's configuration so that the IP 1111 forwarding table will associate discovered IP routes with the correct 1112 interface. 1114 3.4. OMNI Interface Initialization 1116 AERO Servers, Proxys and Clients configure OMNI interfaces as their 1117 point of attachment to the OMNI link. AERO nodes assign the MSPs for 1118 the link to their OMNI interfaces (i.e., as a "route-to-interface") 1119 to ensure that packets with destination addresses covered by an MNP 1120 not explicitly assigned to a non-OMNI interface are directed to the 1121 OMNI interface. 1123 OMNI interface initialization procedures for Servers, Proxys, Clients 1124 and Bridges are discussed in the following sections. 1126 3.4.1. AERO Server/Relay Behavior 1128 When a Server enables an OMNI interface, it assigns an LLA/ULA 1129 appropriate for the given OMNI link segment. The Server also 1130 configures secured tunnels with one or more neighboring Bridges and 1131 engages in a BGP routing protocol session with each Bridge. 1133 The OMNI interface provides a single interface abstraction to the IP 1134 layer, but internally comprises multiple secured tunnels as well as 1135 an NBMA nexus for sending encapsulated data packets to OMNI interface 1136 neighbors. The Server further configures a service to facilitate ND 1137 exchanges with AERO Clients and manages per-Client neighbor cache 1138 entries and IP forwarding table entries based on control message 1139 exchanges. 1141 Relays are simply Servers that run a dynamic routing protocol to 1142 redistribute routes between the OMNI interface and INET/EUN 1143 interfaces (see: Section 3.2.3). The Relay provisions MNPs to 1144 networks on the INET/EUN interfaces (i.e., the same as a Client would 1145 do) and advertises the MSP(s) for the OMNI link over the INET/EUN 1146 interfaces. The Relay further provides an attachment point of the 1147 OMNI link to a non-MNP-based global topology. 1149 3.4.2. AERO Proxy Behavior 1151 When a Proxy enables an OMNI interface, it assigns an LLA/ULA and 1152 configures permanent neighbor cache entries the same as for Servers. 1153 The Proxy also configures secured tunnels with one or more 1154 neighboring Bridges and maintains per-Client neighbor cache entries 1155 based on control message exchanges. Importantly Proxys are often 1156 configured to act as Servers, and vice-versa. 1158 3.4.3. AERO Client Behavior 1160 When a Client enables an OMNI interface, it sends RS messages with ND 1161 parameters over its underlying interfaces to a Server, which returns 1162 an RA message with corresponding parameters. (The RS/RA messages may 1163 pass through a Proxy in the case of a Client's ANET interface, or 1164 through one or more NATs in the case of a Client's INET interface.) 1166 3.4.4. AERO Bridge Behavior 1168 AERO Bridges configure an OMNI interface and assign the ULA Subnet 1169 Router Anycast address for each OMNI link segment they connect to. 1170 Bridges configure secured tunnels with Servers, Proxys and other 1171 Bridges; they also configure LLAs/ULAs and permanent neighbor cache 1172 entries the same as Servers. Bridges engage in a BGP routing 1173 protocol session with a subset of the Servers and other Bridges on 1174 the spanning tree (see: Section 3.2.3). 1176 3.5. OMNI Interface Neighbor Cache Maintenance 1178 Each OMNI interface maintains a conceptual neighbor cache that 1179 includes an entry for each neighbor it communicates with on the OMNI 1180 link per [RFC4861]. OMNI interface neighbor cache entries are said 1181 to be one of "permanent", "symmetric", "asymmetric" or "proxy". 1183 Permanent neighbor cache entries are created through explicit 1184 administrative action; they have no timeout values and remain in 1185 place until explicitly deleted. AERO Bridges maintain permanent 1186 neighbor cache entries for their associated Proxys/Servers (and vice- 1187 versa). Each entry maintains the mapping between the neighbor's 1188 network-layer LLA and corresponding INET address. 1190 Symmetric neighbor cache entries are created and maintained through 1191 RS/RA exchanges as specified in Section 3.12, and remain in place for 1192 durations bounded by prefix lifetimes. AERO Servers maintain 1193 symmetric neighbor cache entries for each of their associated 1194 Clients, and AERO Clients maintain symmetric neighbor cache entries 1195 for each of their associated Servers. 1197 Asymmetric neighbor cache entries are created or updated based on 1198 route optimization messaging as specified in Section 3.14, and are 1199 garbage-collected when keepalive timers expire. AERO ROSs maintain 1200 asymmetric neighbor cache entries for active targets with lifetimes 1201 based on ND messaging constants. Asymmetric neighbor cache entries 1202 are unidirectional since only the ROS (and not the ROR) creates an 1203 entry. 1205 Proxy neighbor cache entries are created and maintained by AERO 1206 Proxys when they process Client/Server ND exchanges, and remain in 1207 place for durations bounded by ND and prefix lifetimes. AERO Proxys 1208 maintain proxy neighbor cache entries for each of their associated 1209 Clients. Proxy neighbor cache entries track the Client state and the 1210 address of the Client's associated Server(s). 1212 To the list of neighbor cache entry states in Section 7.3.2 of 1213 [RFC4861], Proxy and Server OMNI interfaces add an additional state 1214 DEPARTED that applies to symmetric and proxy neighbor cache entries 1215 for Clients that have recently departed. The interface sets a 1216 "DepartTime" variable for the neighbor cache entry to "DEPART_TIME" 1217 seconds. DepartTime is decremented unless a new ND message causes 1218 the state to return to REACHABLE. While a neighbor cache entry is in 1219 the DEPARTED state, packets destined to the target Client are 1220 forwarded to the Client's new location instead of being dropped. 1221 When DepartTime decrements to 0, the neighbor cache entry is deleted. 1222 It is RECOMMENDED that DEPART_TIME be set to the default constant 1223 value REACHABLE_TIME plus 10 seconds (40 seconds by default) to allow 1224 a window for packets in flight to be delivered while stale route 1225 optimization state may be present. 1227 When an ROR receives an authentic NS message used for route 1228 optimization, it searches for a symmetric neighbor cache entry for 1229 the target Client. The ROR then returns a solicited NA message 1230 without creating a neighbor cache entry for the ROS, but creates or 1231 updates a target Client "Report List" entry for the ROS and sets a 1232 "ReportTime" variable for the entry to REPORT_TIME seconds. The ROR 1233 resets ReportTime when it receives a new authentic NS message, and 1234 otherwise decrements ReportTime while no authentic NS messages have 1235 been received. It is RECOMMENDED that REPORT_TIME be set to the 1236 default constant value REACHABLE_TIME plus 10 seconds (40 seconds by 1237 default) to allow a window for route optimization to converge before 1238 ReportTime decrements below REACHABLE_TIME. 1240 When the ROS receives a solicited NA message response to its NS 1241 message used for route optimization, it creates or updates an 1242 asymmetric neighbor cache entry for the target network-layer and 1243 link-layer addresses. The ROS then (re)sets ReachableTime for the 1244 neighbor cache entry to REACHABLE_TIME seconds and uses this value to 1245 determine whether packets can be forwarded directly to the target, 1246 i.e., instead of via a default route. The ROS otherwise decrements 1247 ReachableTime while no further solicited NA messages arrive. It is 1248 RECOMMENDED that REACHABLE_TIME be set to the default constant value 1249 30 seconds as specified in [RFC4861]. 1251 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1252 of NS keepalives sent when a correspondent may have gone unreachable, 1253 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1254 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1255 to limit the number of unsolicited NAs that can be sent based on a 1256 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1257 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1258 same as specified in [RFC4861]. 1260 Different values for DEPART_TIME, REPORT_TIME, REACHABLE_TIME, 1261 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1262 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1263 different values are chosen, all nodes on the link MUST consistently 1264 configure the same values. Most importantly, DEPART_TIME and 1265 REPORT_TIME SHOULD be set to a value that is sufficiently longer than 1266 REACHABLE_TIME to avoid packet loss due to stale route optimization 1267 state. 1269 3.5.1. OMNI Neighbor Interface Attributes 1271 OMNI interface IPv6 ND messages include OMNI options 1272 [I-D.templin-6man-omni-interface] with Interface Attributes that 1273 provide Link-Layer Address and QoS Preference information for the 1274 neighbor's underlying interfaces. This information is stored in the 1275 neighbor cache and provides the basis for the forwarding algorithm 1276 specified in Section 3.10. The information is cumulative and 1277 reflects the union of the OMNI information from the most recent ND 1278 messages received from the neighbor; it is therefore not required 1279 that each ND message contain all neighbor information. 1281 The OMNI option Interface Attributes for each underlying interface 1282 includes a two-part "Link-Layer Address" consisting of a simple IP 1283 encapsulation address determined by the FMT and L2ADDR fields and an 1284 OAL ULA determined by the SRT and LHS fields. If the neighbor is 1285 located in the local OMNI link segment (and, if any necessary NAT 1286 state has been established) forwarding via simple IP encapsulation 1287 can be used; otherwise, OAL encapsulation must be used. Underlying 1288 interfaces are further selected based on their associated preference 1289 values "high", "medium", "low" or "disabled". 1291 Note: the OMNI option is distinct from any Source/Target Link-Layer 1292 Address Options (S/TLLAOs) that may appear in an ND message according 1293 to the appropriate IPv6 over specific link layer specification (e.g., 1294 [RFC2464]). If both an OMNI option and S/TLLAO appear, the former 1295 pertains to encapsulation addresses while the latter pertains to the 1296 native L2 address format of the underlying media. 1298 3.5.2. OMNI Neighbor Advertisement Message Flags 1300 As discussed in Section 4.4 of [RFC4861] NA messages include three 1301 flag bits R, S and O. OMNI interface NA messages treat the flags as 1302 follows: 1304 o R: The R ("Router") flag is set to 1 in the NA messages sent by 1305 all AERO/OMNI node types. Simple hosts that would set R to 0 do 1306 not occur on the OMNI link itself, but may occur on the downstream 1307 links of Clients and Relays. 1309 o S: The S ("Solicited") flag is set exactly as specified in 1310 Section 4.4. of [RFC4861], i.e., it is set to 1 for Solicited NAs 1311 and set to 0 for Unsolicited NAs (both unicast and multicast). 1313 o O: The O ("Override") flag is set to 0 for solicited proxy NAs and 1314 set to 1 for all other solicited and unsolicited NAs. For further 1315 study is whether solicited NAs for anycast targets apply for OMNI 1316 links. Since OMNI LLAs must be uniquely assigned to Clients to 1317 support correct ND protocol operation, however, no role is 1318 currently seen for assigning the same OMNI LLA to multiple 1319 Clients. 1321 3.6. OMNI Interface Encapsulation and Re-encapsulation 1323 The OMNI Adaptation Layer (OAL) inserts mid-layer IPv6 headers known 1324 as the OAL/ORH headers when necessary as discussed in the following 1325 sections. After either inserting or omitting the OAL/ORH headers, 1326 the OMNI interface also inserts or omits an outer encapsulation 1327 header as discussed below. 1329 OMNI interfaces avoid outer encapsulation over Direct underlying 1330 interfaces and ANET underlying interfaces for which the Client and 1331 Proxy are connected to the same underlying link. Otherwise, OMNI 1332 interfaces encapsulate packets according to whether they are entering 1333 the OMNI interface from the network layer or if they are being re- 1334 admitted into the same OMNI link they arrived on. This latter form 1335 of encapsulation is known as "re-encapsulation". 1337 For packets entering the OMNI interface from the network layer, the 1338 OMNI interface copies the "TTL/Hop Limit", "Type of Service/Traffic 1339 Class" [RFC2983], "Flow Label"[RFC6438] (for IPv6) and "Congestion 1340 Experienced" [RFC3168] values in the inner packet's IP header into 1341 the corresponding fields in the OAL and outer encapsulation 1342 header(s). 1344 For packets undergoing re-encapsulation, the OMNI interface instead 1345 copies these values from the original encapsulation header into the 1346 new encapsulation header, i.e., the values are transferred between 1347 encapsulation headers and *not* copied from the encapsulated packet's 1348 network-layer header. (Note especially that by copying the TTL/Hop 1349 Limit between encapsulation headers the value will eventually 1350 decrement to 0 if there is a (temporary) routing loop.) 1352 OMNI interfaces configured over ANET underlying interfaces which 1353 employ a different IP protocol version (and/or when the Client and 1354 Proxy may be separated by multiple ANET IP hops) use IP-in-IP 1355 encapsulation so that the inner packet can traverse the ANET without 1356 decrementing the TTL/Hop-Limit. IPv6 underlying ANET interfaces use 1357 [RFC2473] encapsulation, while IPv4 interfaces use the appropriate 1358 encapsulation per one of [RFC5214][RFC2003]. 1360 OMNI interfaces configured over INET underlying interfaces 1361 encapsulate packets in INET headers according to the next hop 1362 determined in the forwarding algorithm in Section 3.10. If the next 1363 hop is reached via a secured tunnel, the OMNI interface uses an 1364 encapsulation format specific to the secured tunnel type (see: 1365 Section 6). If the next hop is reached via an unsecured INET 1366 interface, the OMNI interface instead uses UDP/IP encapsulation per 1367 [RFC4380] and as extended in [RFC6081]. 1369 When UDP/IP encapsulation is used, the OMNI interface next sets the 1370 UDP source port to a constant value that it will use in each 1371 successive packet it sends, and sets the UDP length field to the 1372 length of the encapsulated packet plus 8 bytes for the UDP header 1373 itself plus the length of any included extension headers or trailers. 1374 The encapsulated packet may be either IPv6 or IPv4, as distinguished 1375 by the version number found in the first four bits. 1377 For UDP/IP-encapsulated packets sent to a Server, Relay or Bridge, 1378 the OMNI interface sets the UDP destination port to 8060, i.e., the 1379 IANA-registered port number for AERO. For packets sent to a Client, 1380 the OMNI interface sets the UDP destination port to the port value 1381 stored in the neighbor cache entry for this Client. The OMNI 1382 interface finally includes/omits the UDP checksum according to 1383 [RFC6935][RFC6936]. 1385 3.7. OMNI Interface Decapsulation 1387 OMNI interfaces decapsulate packets destined either to the AERO node 1388 itself or to a destination reached via an interface other than the 1389 OMNI interface the packet was received on. When the encapsulated 1390 packet arrives in multiple OAL fragments, the OMNI interface 1391 reassembles as discussed in Section 3.9. Further decapsulation steps 1392 are performed according to the appropriate encapsulation format 1393 specification. 1395 3.8. OMNI Interface Data Origin Authentication 1397 AERO nodes employ simple data origin authentication procedures. In 1398 particular: 1400 o AERO Bridges, Servers and Proxys accept encapsulated data packets 1401 and control messages received from the (secured) spanning tree. 1403 o AERO Proxys and Clients accept packets that originate from within 1404 the same secured ANET. 1406 o AERO Clients and Relays accept packets from downstream network 1407 correspondents based on ingress filtering. 1409 o AERO Clients, Relays and Servers verify the outer UDP/IP 1410 encapsulation addresses according to [RFC4380]. 1412 AERO nodes silently drop any packets that do not satisfy the above 1413 data origin authentication procedures. Further security 1414 considerations are discussed in Section 6. 1416 3.9. OMNI Adaptation Layer and OMNI Interface MTU 1418 The OMNI interface observes the link nature of tunnels, including the 1419 Maximum Transmission Unit (MTU), Maximum Reassembly Unit (MRU) and 1420 the role of fragmentation and reassembly [I-D.ietf-intarea-tunnels]. 1421 The OMNI interface employs an OMNI Adaptation Layer (OAL) for 1422 accommodating multiple underlying links with diverse MTUs. The 1423 functions of the OAL and the OMNI interface MTU/MRU are specified in 1424 Section 5 of [I-D.templin-6man-omni-interface], with MTU/MRU both set 1425 to the constant value 9180 bytes. 1427 When the network layer presents an IP packet to the OMNI interface, 1428 the OAL encapsulates the packet in OAL/ORH headers. The OAL then 1429 fragments the encapsulated packet if necessary such that the OAL/ORH 1430 headers appear in each fragment while the original IP packet header 1431 appears only in the first fragment. The OAL then transmits each OAL/ 1432 ORH packet/fragment over an underlying linterface connected to either 1433 a physical link such as Ethernet, WiFi and the like or a virtual link 1434 such as an Internet or higher-layer tunnel (see the definition of 1435 link in [RFC8200]). 1437 Note: Although the ORH may be removed by a Bridge on the path (see: 1438 Section 3.10.4), this does not interfere with the destination's 1439 ability to reassemble in the event that the packet was fragmented. 1440 This is due to the fact that the ORH is not included in the 1441 fragmentabe part; therefore, its removal does not invalidate the 1442 offset values in any fragment headers. 1444 3.10. OMNI Interface Forwarding Algorithm 1446 IP packets enter a node's OMNI interface either from the network 1447 layer (i.e., from a local application or the IP forwarding system) or 1448 from the link layer (i.e., from an OMNI interface neighbor). All 1449 packets entering a node's OMNI interface first undergo data origin 1450 authentication as discussed in Section 3.8. Packets that satisfy 1451 data origin authentication are processed further, while all others 1452 are dropped silently. The OMNI interface OAL wraps accepted packets 1453 in OAL/ORH headers if necessary as discussed above. 1455 Packets that enter the OMNI interface from the network layer are 1456 forwarded to an OMNI interface neighbor. Packets that enter the OMNI 1457 interface from the link layer are either re-admitted into the OMNI 1458 link or forwarded to the network layer where they are subject to 1459 either local delivery or IP forwarding. In all cases, the OMNI 1460 interface itself MUST NOT decrement the network layer TTL/Hop-count 1461 since its forwarding actions occur below the network layer. 1463 OMNI interfaces may have multiple underlying interfaces and/or 1464 neighbor cache entries for neighbors with multiple underlying 1465 interfaces (see Section 3.3). The OMNI interface uses interface 1466 attributes and/or traffic classifiers (e.g., DSCP value, port number, 1467 etc.) to select an outgoing underlying interface for each packet 1468 based on the node's own QoS preferences, and also to select a 1469 destination link-layer address based on the neighbor's underlying 1470 interface with the highest preference. AERO implementations SHOULD 1471 allow for QoS preference values to be modified at runtime through 1472 network management. 1474 If multiple outgoing interfaces and/or neighbor interfaces have a 1475 preference of "high", the AERO node replicates the packet and sends 1476 one copy via each of the (outgoing / neighbor) interface pairs; 1477 otherwise, the node sends a single copy of the packet via an 1478 interface with the highest preference. AERO nodes keep track of 1479 which underlying interfaces are currently "reachable" or 1480 "unreachable", and only use "reachable" interfaces for forwarding 1481 purposes. 1483 The following sections discuss the OMNI interface forwarding 1484 algorithms for Clients, Proxys, Servers and Bridges. In the 1485 following discussion, a packet's destination address is said to 1486 "match" if it is the same as a cached address, or if it is covered by 1487 a cached prefix (which may be encoded in an LLA). 1489 3.10.1. Client Forwarding Algorithm 1491 When an IP packet enters a Client's OMNI interface from the network 1492 layer the Client searches for an asymmetric neighbor cache entry that 1493 matches the destination. If there is a match, the Client uses one or 1494 more "reachable" neighbor interfaces in the entry for packet 1495 forwarding. If there is no asymmetric neighbor cache entry, the 1496 Client instead forwards the packet toward a Server (the packet is 1497 intercepted by a Proxy if there is a Proxy on the path). The Client 1498 encapsulates the packet in OAL/ORH headers if necessary and fragments 1499 according to MTU requirements (see: Section 3.9). 1501 When an IP packet enters a Client's OMNI interface from the link- 1502 layer, if the destination matches one of the Client's MNPs or link- 1503 local addresses the Client reassembles and decapsulates as necessary 1504 and delivers the inner packet to the network layer. Otherwise, the 1505 Client drops the packet and MAY return a network-layer ICMP 1506 Destination Unreachable message subject to rate limiting (see: 1507 Section 3.11). 1509 3.10.2. Proxy Forwarding Algorithm 1511 For control messages originating from or destined to a Client, the 1512 Proxy intercepts the message and updates its proxy neighbor cache 1513 entry for the Client. The Proxy then forwards a (proxyed) copy of 1514 the control message. (For example, the Proxy forwards a proxied 1515 version of a Client's NS/RS message to the target neighbor, and 1516 forwards a proxied version of the NA/RA reply to the Client.) 1518 When the Proxy receives a data packet from a Client within the ANET, 1519 the Proxy reassembles and re-fragments if necessary then searches for 1520 an asymmetric neighbor cache entry that matches the destination and 1521 forwards as follows: 1523 o if the destination matches an asymmetric neighbor cache entry, the 1524 Proxy uses one or more "reachable" neighbor interfaces in the 1525 entry for packet forwarding using OAL/ORH encapsulation if 1526 necessary according to the cached link-layer address information. 1527 If the neighbor interface is in the same OMNI link segment, the 1528 Proxy forwards the packet directly to the neighbor; otherwise, it 1529 forwards the packet to a Bridge. 1531 o else, the Proxy uses OAL/ORH encapsulation and forwards the packet 1532 to a Bridge while using the ULA corresponding to the packet's 1533 destination as the destination address. 1535 When the Proxy receives an encapsulated data packet from an INET 1536 neighbor or from a secured tunnel from a Bridge, it accepts the 1537 packet only if data origin authentication succeeds and if there is a 1538 proxy neighbor cache entry that matches the inner destination. Next, 1539 the Proxy reassembles the packet (if necessary) and continues 1540 processing. If the reassembly is complete and the neighbor cache 1541 state is REACHABLE, the Proxy then returns a PTB if necessary (see: 1542 Section 3.9) then either drops or forwards the packet to the Client 1543 while performing OAL/ORH encapsulation and re-fragmentation if 1544 necessary. If the neighbor cache entry state is DEPARTED, the Proxy 1545 instead changes the destination address to the address of the new 1546 Server and forwards it to a Bridge while performing OAL/ORH re- 1547 fragmentation if necessary. 1549 3.10.3. Server/Relay Forwarding Algorithm 1551 For control messages destined to a target Client's LLA that are 1552 received from a secured tunnel, the Server intercepts the message and 1553 sends a Proxyed response on behalf of the Client. (For example, the 1554 Server sends a Proxyed NA message reply in response to an NS message 1555 directed to one of its associated Clients.) If the Client's neighbor 1556 cache entry is in the DEPARTED state, however, the Server instead 1557 forwards the packet to the Client's new Server as discussed in 1558 Section 3.16. 1560 When the Server receives an encapsulated data packet from an INET 1561 neighbor or from a secured tunnel, it accepts the packet only if data 1562 origin authentication succeeds. The Server then continues processing 1563 as follows: 1565 o if the network layer destination matches a symmetric neighbor 1566 cache entry in the REACHABLE state the Server prepares the packet 1567 for forwarding to the destination Client. The Server first 1568 reassembles (if necessary) and forwards the packet (while re- 1569 fragmenting if necessary) as specified in Section 3.9. 1571 o else, if the destination matches a symmetric neighbor cache entry 1572 in the DEPARETED state the Server re-encapsulates the packet and 1573 forwards it using the ULA of the Client's new Server as the 1574 destination. 1576 o else, if the destination matches an asymmetric neighbor cache 1577 entry, the Server uses one or more "reachable" neighbor interfaces 1578 in the entry for packet forwarding via the local INET if the 1579 neighbor is in the same OMNI link segment or using OAL/ORH 1580 encapsulation if necessary with the final destination set to the 1581 neighbor's ULA otherwise. 1583 o else, if the destination matches a non-MNP route in the IP 1584 forwarding table or an LLA assigned to the Server's OMNI 1585 interface, the Server reassembles if necessary, decapsulates the 1586 packet and releases it to the network layer for local delivery or 1587 IP forwarding. 1589 o else, the Server drops the packet. 1591 When the Server's OMNI interface receives a data packet from the 1592 network layer or from a VPNed or Direct Client, it performs OAL/ORH 1593 encapsulation and fragmentation if necessary, then processes the 1594 packet according to the network-layer destination address as follows: 1596 o if the destination matches a symmetric or asymmetric neighbor 1597 cache entry the Server processes the packet as above. 1599 o else, the Server encapsulates the packet in OLA/ORH headers and 1600 forwards it to a Bridge using its own ULA as the source and the 1601 ULA corresponding to the destination as the destination. 1603 3.10.4. Bridge Forwarding Algorithm 1605 Bridges forward OAL/ORH-encapsulated packets over secured tunnels the 1606 same as any IP router. When the Bridge receives an OAL/ORH- 1607 encapsulated packet via a secured tunnel, it removes the outer INET 1608 header and searches for a forwarding table entry that matches the 1609 destination address. The Bridge then processes the packet as 1610 follows: 1612 o if the destination matches its ULA Subnet Router Anycast address, 1613 the Bridge determines if the next header is an ORH. If so, the 1614 Bridge removes the ORH from the packet while decrementing the OAL 1615 header Payload Length field. If the ORH includes a Destination 1616 Suffix the Bridge also writes the ULA formed from the Destination 1617 Suffix into the OAL header destination address; otherwise, it 1618 writes the ULA formed from the SRT/LHS values. Next, the Bridge 1619 examines the FMT to determine if the target is behind a NAT. If 1620 no NAT is indicated, the Bridge forwards the packet directly to 1621 the L2ADDR using link-layer (UDP/IP) encapsulation. If a NAT is 1622 indicated, the Bridge MAY perform NAT traversal procedures by 1623 sending bubbles per [RFC4380]. The Bridge then either applies 1624 AERO route optimization after NAT traversal procedures have 1625 converged, or simply forwards the packet directly to the Server 1626 indicated by SRT/LHS. 1628 o if the destination matches one of the Bridge's own addresses, the 1629 Bridge submits the packet for local delivery. 1631 o else, if the destination matches a forwarding table entry the 1632 Bridge forwards the packet via a secured tunnel to the next hop. 1633 If the destination matches an MSP without matching an MNP, 1634 however, the Bridge instead drops the packet and returns an ICMP 1635 Destination Unreachable message subject to rate limiting (see: 1636 Section 3.11). 1638 o else, the Bridge drops the packet and returns an ICMP Destination 1639 Unreachable as above. 1641 As for any IP router, the Bridge decrements the TTL/Hop Limit when it 1642 forwards the packet. Therefore, when an OAL header is present only 1643 the Hop Limit in the OAL header is decremented and not the TTL/Hop 1644 Limit in the inner packet header. Bridges do not insert OAL/ORH 1645 headers themselves; instead, they act as IPv6 routers and forward 1646 packets based on the destination address found in the headers of 1647 packets they receive. 1649 3.11. OMNI Interface Error Handling 1651 When an AERO node admits a packet into the OMNI interface, it may 1652 receive link-layer or network-layer error indications. 1654 A link-layer error indication is an ICMP error message generated by a 1655 router in the INET on the path to the neighbor or by the neighbor 1656 itself. The message includes an IP header with the address of the 1657 node that generated the error as the source address and with the 1658 link-layer address of the AERO node as the destination address. 1660 The IP header is followed by an ICMP header that includes an error 1661 Type, Code and Checksum. Valid type values include "Destination 1662 Unreachable", "Time Exceeded" and "Parameter Problem" 1663 [RFC0792][RFC4443]. (OMNI interfaces ignore all link-layer IPv4 1664 "Fragmentation Needed" and IPv6 "Packet Too Big" messages since they 1665 only emit packets that are guaranteed to be no larger than the IP 1666 minimum link MTU as discussed in Section 3.9.) 1667 The ICMP header is followed by the leading portion of the packet that 1668 generated the error, also known as the "packet-in-error". For 1669 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1670 much of invoking packet as possible without the ICMPv6 packet 1671 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1672 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1673 "Internet Header + 64 bits of Original Data Datagram", however 1674 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1675 ICMP datagram SHOULD contain as much of the original datagram as 1676 possible without the length of the ICMP datagram exceeding 576 1677 bytes". 1679 The link-layer error message format is shown in Figure 5 (where, "L2" 1680 and "L3" refer to link-layer and network-layer, respectively): 1682 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1683 ~ ~ 1684 | L2 IP Header of | 1685 | error message | 1686 ~ ~ 1687 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1688 | L2 ICMP Header | 1689 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1690 ~ ~ P 1691 | IP and other encapsulation | a 1692 | headers of original L3 packet | c 1693 ~ ~ k 1694 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1695 ~ ~ t 1696 | IP header of | 1697 | original L3 packet | i 1698 ~ ~ n 1699 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1700 ~ ~ e 1701 | Upper layer headers and | r 1702 | leading portion of body | r 1703 | of the original L3 packet | o 1704 ~ ~ r 1705 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1707 Figure 5: OMNI Interface Link-Layer Error Message Format 1709 The AERO node rules for processing these link-layer error messages 1710 are as follows: 1712 o When an AERO node receives a link-layer Parameter Problem message, 1713 it processes the message the same as described as for ordinary 1714 ICMP errors in the normative references [RFC0792][RFC4443]. 1716 o When an AERO node receives persistent link-layer Time Exceeded 1717 messages, the IP ID field may be wrapping before earlier fragments 1718 awaiting reassembly have been processed. In that case, the node 1719 should begin including integrity checks and/or institute rate 1720 limits for subsequent packets. 1722 o When an AERO node receives persistent link-layer Destination 1723 Unreachable messages in response to encapsulated packets that it 1724 sends to one of its asymmetric neighbor correspondents, the node 1725 should process the message as an indication that a path may be 1726 failing, and optionally initiate NUD over that path. If it 1727 receives Destination Unreachable messages over multiple paths, the 1728 node should allow future packets destined to the correspondent to 1729 flow through a default route and re-initiate route optimization. 1731 o When an AERO Client receives persistent link-layer Destination 1732 Unreachable messages in response to encapsulated packets that it 1733 sends to one of its symmetric neighbor Servers, the Client should 1734 mark the path as unusable and use another path. If it receives 1735 Destination Unreachable messages on many or all paths, the Client 1736 should associate with a new Server and release its association 1737 with the old Server as specified in Section 3.16.5. 1739 o When an AERO Server receives persistent link-layer Destination 1740 Unreachable messages in response to encapsulated packets that it 1741 sends to one of its symmetric neighbor Clients, the Server should 1742 mark the underlying path as unusable and use another underlying 1743 path. 1745 o When an AERO Server or Proxy receives link-layer Destination 1746 Unreachable messages in response to an encapsulated packet that it 1747 sends to one of its permanent neighbors, it treats the messages as 1748 an indication that the path to the neighbor may be failing. 1749 However, the dynamic routing protocol should soon reconverge and 1750 correct the temporary outage. 1752 When an AERO Bridge receives a packet for which the network-layer 1753 destination address is covered by an MSP, if there is no more- 1754 specific routing information for the destination the Bridge drops the 1755 packet and returns a network-layer Destination Unreachable message 1756 subject to rate limiting. The Bridge writes the network-layer source 1757 address of the original packet as the destination address and uses 1758 one of its non link-local addresses as the source address of the 1759 message. 1761 When an AERO node receives an encapsulated packet for which the 1762 reassembly buffer it too small, it drops the packet and returns a 1763 network-layer Packet Too Big (PTB) message. The node first writes 1764 the MRU value into the PTB message MTU field, writes the network- 1765 layer source address of the original packet as the destination 1766 address and writes one of its non link-local addresses as the source 1767 address. 1769 3.12. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1771 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1772 coordinated as discussed in the following Sections. 1774 3.12.1. AERO Service Model 1776 Each AERO Server on the OMNI link is configured to facilitate Client 1777 prefix delegation/registration requests. Each Server is provisioned 1778 with a database of MNP-to-Client ID mappings for all Clients enrolled 1779 in the AERO service, as well as any information necessary to 1780 authenticate each Client. The Client database is maintained by a 1781 central administrative authority for the OMNI link and securely 1782 distributed to all Servers, e.g., via the Lightweight Directory 1783 Access Protocol (LDAP) [RFC4511], via static configuration, etc. 1784 Clients receive the same service regardless of the Servers they 1785 select. 1787 AERO Clients and Servers use ND messages to maintain neighbor cache 1788 entries. AERO Servers configure their OMNI interfaces as advertising 1789 NBMA interfaces, and therefore send unicast RA messages with a short 1790 Router Lifetime value (e.g., ReachableTime seconds) in response to a 1791 Client's RS message. Thereafter, Clients send additional RS messages 1792 to keep Server state alive. 1794 AERO Clients and Servers include prefix delegation and/or 1795 registration parameters in RS/RA messages (see 1796 [I-D.templin-6man-omni-interface]). The ND messages are exchanged 1797 between Client and Server according to the prefix management schedule 1798 required by the service. If the Client knows its MNP in advance, it 1799 can employ prefix registration by including its LLA as the source 1800 address of an RS message and with an OMNI option with valid prefix 1801 registration information for the MNP. If the Server (and Proxy) 1802 accept the Client's MNP assertion, they inject the prefix into the 1803 routing system and establish the necessary neighbor cache state. 1805 The following sections specify the Client and Server behavior. 1807 3.12.2. AERO Client Behavior 1809 AERO Clients discover the addresses of Servers in a similar manner as 1810 described in [RFC5214]. Discovery methods include static 1811 configuration (e.g., from a flat-file map of Server addresses and 1812 locations), or through an automated means such as Domain Name System 1813 (DNS) name resolution [RFC1035]. Alternatively, the Client can 1814 discover Server addresses through a layer 2 data link login exchange, 1815 or through a unicast RA response to a multicast/anycast RS as 1816 described below. In the absence of other information, the Client can 1817 resolve the DNS Fully-Qualified Domain Name (FQDN) 1818 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 1819 text string and "[domainname]" is a DNS suffix for the OMNI link 1820 (e.g., "example.com"). 1822 To associate with a Server, the Client acts as a requesting router to 1823 request MNPs. The Client prepares an RS message with prefix 1824 management parameters and includes a Nonce and Timestamp option if 1825 the Client needs to correlate RA replies. If the Client already 1826 knows the Server's LLA, it includes the LLA as the network-layer 1827 destination address; otherwise, it includes (link-local) All-Routers 1828 multicast as the network-layer destination. If the Client already 1829 knows its own LLA, it uses the LLA as the network-layer source 1830 address; otherwise, it uses an OMNI Temporary LLA as the network- 1831 layer source address and includes a DHCP Unique Identifier (DUID) 1832 sub-option in the OMNI option (see: 1833 [I-D.templin-6man-omni-interface]). 1835 The Client next includes an OMNI option in the RS message to register 1836 its link-layer information with the Server. The Client sets the OMNI 1837 option prefix registration information according to the MNP, and 1838 includes Interface Attributes corresponding to the underlying 1839 interface over which the Client will send the RS message. The Client 1840 MAY include additional Interface Attributes specific to other 1841 underlying interfaces. 1843 The Client then sends the RS message (either directly via Direct 1844 interfaces, via a VPN for VPNed interfaces, via a Proxy for ANET 1845 interfaces or via INET encapsulation for INET interfaces) and waits 1846 for an RA message reply (see Section 3.12.3). The Client retries up 1847 to MAX_RTR_SOLICITATIONS times until an RA is received. If the 1848 Client receives no RAs, or if it receives an RA with Router Lifetime 1849 set to 0, the Client SHOULD abandon this Server and try another 1850 Server. Otherwise, the Client processes the prefix information found 1851 in the RA message. 1853 Next, the Client creates a symmetric neighbor cache entry with the 1854 Server's LLA as the network-layer address and the Server's 1855 encapsulation and/or link-layer addresses as the link-layer address. 1856 The Client records the RA Router Lifetime field value in the neighbor 1857 cache entry as the time for which the Server has committed to 1858 maintaining the MNP in the routing system via this underlying 1859 interface, and caches the other RA configuration information 1860 including Cur Hop Limit, M and O flags, Reachable Time and Retrans 1861 Timer. The Client then autoconfigures LLAs for each of the delegated 1862 MNPs and assigns them to the OMNI interface. The Client also caches 1863 any MSPs included in Route Information Options (RIOs) [RFC4191] as 1864 MSPs to associate with the OMNI link, and assigns the MTU value in 1865 the MTU option to the underlying interface. 1867 The Client then registers additional underlying interfaces with the 1868 Server by sending RS messages via each additional interface. The RS 1869 messages include the same parameters as for the initial RS/RA 1870 exchange, but with destination address set to the Server's LLA. 1872 Following autoconfiguration, the Client sub-delegates the MNPs to its 1873 attached EUNs and/or the Client's own internal virtual interfaces as 1874 described in [I-D.templin-v6ops-pdhost] to support the Client's 1875 downstream attached "Internet of Things (IoT)". The Client 1876 subsequently sends additional RS messages over each underlying 1877 interface before the Router Lifetime received for that interface 1878 expires. 1880 After the Client registers its underlying interfaces, it may wish to 1881 change one or more registrations, e.g., if an interface changes 1882 address or becomes unavailable, if QoS preferences change, etc. To 1883 do so, the Client prepares an RS message to send over any available 1884 underlying interface. The RS includes an OMNI option with prefix 1885 registration information specific to its MNP, with Interface 1886 Attributes specific to the selected underlying interface, and with 1887 any additional Interface Attributes specific to other underlying 1888 interfaces. When the Client receives the Server's RA response, it 1889 has assurance that the Server has been updated with the new 1890 information. 1892 If the Client wishes to discontinue use of a Server it issues an RS 1893 message over any underlying interface with an OMNI option with a 1894 prefix release indication. When the Server processes the message, it 1895 releases the MNP, sets the symmetric neighbor cache entry state for 1896 the Client to DEPARTED and returns an RA reply with Router Lifetime 1897 set to 0. After a short delay (e.g., 2 seconds), the Server 1898 withdraws the MNP from the routing system. 1900 3.12.3. AERO Server Behavior 1902 AERO Servers act as IP routers and support a prefix delegation/ 1903 registration service for Clients. Servers arrange to add their LLAs 1904 to a static map of Server addresses for the link and/or the DNS 1905 resource records for the FQDN "linkupnetworks.[domainname]" before 1906 entering service. Server addresses should be geographically and/or 1907 topologically referenced, and made available for discovery by Clients 1908 on the OMNI link. 1910 When a Server receives a prospective Client's RS message on its OMNI 1911 interface, it SHOULD return an immediate RA reply with Router 1912 Lifetime set to 0 if it is currently too busy or otherwise unable to 1913 service the Client. Otherwise, the Server authenticates the RS 1914 message and processes the prefix delegation/registration parameters. 1915 The Server first determines the correct MNPs to provide to the Client 1916 by searching the Client database. When the Server returns the MNPs, 1917 it also creates a forwarding table entry for the ULA corresponding to 1918 each MNP so that the MNPs are propagated into the routing system 1919 (see: Section 3.2.3). For IPv6, the Server creates an IPv6 1920 forwarding table entry for each MNP. For IPv4, the Server creates an 1921 IPv6 forwarding table entry with the IPv4-compatibility ULA prefix 1922 corresponding to the IPv4 address. 1924 The Server next creates a symmetric neighbor cache entry for the 1925 Client using the base LLA as the network-layer address and with 1926 lifetime set to no more than the smallest prefix lifetime. Next, the 1927 Server updates the neighbor cache entry by recording the information 1928 in each Interface Attributes sub-option in the RS OMNI option. The 1929 Server also records the actual OAL/INET addresses in the neighbor 1930 cache entry. 1932 Next, the Server prepares an RA message using its LLA as the network- 1933 layer source address and the network-layer source address of the RS 1934 message as the network-layer destination address. The Server sets 1935 the Router Lifetime to the time for which it will maintain both this 1936 underlying interface individually and the symmetric neighbor cache 1937 entry as a whole. The Server also sets Cur Hop Limit, M and O flags, 1938 Reachable Time and Retrans Timer to values appropriate for the OMNI 1939 link. The Server includes the MNPs, any other prefix management 1940 parameters and an OMNI option with no Interface Attributes. The 1941 Server then includes one or more RIOs that encode the MSPs for the 1942 OMNI link, plus an MTU option (see Section 3.9). The Server finally 1943 forwards the message to the Client using OAL/INET, INET, or NULL 1944 encapsulation as necessary. 1946 After the initial RS/RA exchange, the Server maintains a 1947 ReachableTime timer for each of the Client's underlying interfaces 1948 individually (and for the Client's symmetric neighbor cache entry 1949 collectively) set to expire after ReachableTime seconds. If the 1950 Client (or Proxy) issues additional RS messages, the Server sends an 1951 RA response and resets ReachableTime. If the Server receives an ND 1952 message with a prefix release indication it sets the Client's 1953 symmetric neighbor cache entry to the DEPARTED state and withdraws 1954 the MNP from the routing system after a short delay (e.g., 2 1955 seconds). If ReachableTime expires before a new RS is received on an 1956 individual underlying interface, the Server marks the interface as 1957 DOWN. If ReachableTime expires before any new RS is received on any 1958 individual underlying interface, the Server sets the symmetric 1959 neighbor cache entry state to STALE and sets a 10 second timer. If 1960 the Server has not received a new RS or ND message with a prefix 1961 release indication before the 10 second timer expires, it deletes the 1962 neighbor cache entry and withdraws the MNP from the routing system. 1964 The Server processes any ND messages pertaining to the Client and 1965 returns an NA/RA reply in response to solicitations. The Server may 1966 also issue unsolicited RA messages, e.g., with reconfigure parameters 1967 to cause the Client to renegotiate its prefix delegation/ 1968 registrations, with Router Lifetime set to 0 if it can no longer 1969 service this Client, etc. Finally, If the symmetric neighbor cache 1970 entry is in the DEPARTED state, the Server deletes the entry after 1971 DepartTime expires. 1973 Note: Clients SHOULD notify former Servers of their departures, but 1974 Servers are responsible for expiring neighbor cache entries and 1975 withdrawing routes even if no departure notification is received 1976 (e.g., if the Client leaves the network unexpectedly). Servers 1977 SHOULD therefore set Router Lifetime to ReachableTime seconds in 1978 solicited RA messages to minimize persistent stale cache information 1979 in the absence of Client departure notifications. A short Router 1980 Lifetime also ensures that proactive Client/Server RS/RA messaging 1981 will keep any NAT state alive (see above). 1983 Note: All Servers on an OMNI link MUST advertise consistent values in 1984 the RA Cur Hop Limit, M and O flags, Reachable Time and Retrans Timer 1985 fields the same as for any link, since unpredictable behavior could 1986 result if different Servers on the same link advertised different 1987 values. 1989 3.12.3.1. DHCPv6-Based Prefix Registration 1991 When a Client is not pre-provisioned with an OMNI LLA containing a 1992 MNP, it will need for the Server to select one or more MNPs on its 1993 behalf and set up the correct state in the AERO routing service. (A 1994 Client with a pre-provisioned MNP may also request the Server to 1995 select additional MNPs.) The DHCPv6 service [RFC8415] is used to 1996 support this requirement. 1998 When a Client needs to have the Server select MNPs, it sends an RS 1999 message with an OMNI option that includes a DHCPv6 message suboption 2000 with DHCPv6 Prefix Delegation (DHCPv6-PD) parameters. When the 2001 Server receives the RS message, it extracts the DHCPv6-PD message 2002 from the OMNI option. 2004 The Server then acts as a "Proxy DHCPv6 Client" in a message exchange 2005 with the locally-resident DHCPv6 server, which delegates MNPs and 2006 returns a DHCPv6-PD Reply message. (If the Server wishes to defer 2007 creation of MN state until the DHCPv6-PD Reply is received, it can 2008 instead act as a Lightweight DHCPv6 Relay Agent per [RFC6221] by 2009 encapsulating the DHCPv6-PD message in a Relay-forward/reply exchange 2010 with Relay Message and Interface ID options.) 2012 When the Server receives the DHCPv6-PD Reply, it adds a route to the 2013 routing system and creates an OMNI MN LLA based on the delegated MNP. 2014 The Server then sends an RA back to the Client with the (newly- 2015 created) OMNI MN LLA as the destination address and with the 2016 DHCPv6-PD Reply message coded in the OMNI option. When the Client 2017 receives the RA, it creates a default route, assigns the Subnet 2018 Router Anycast address and sets its OMNI LLA based on the delegated 2019 MNP. 2021 3.13. The AERO Proxy 2023 Clients may connect to protected-spectrum ANETs that employ physical 2024 and/or link-layer security services to facilitate communications to 2025 Servers in outside INETs. In that case, the ANET can employ an AERO 2026 Proxy. The Proxy is located at the ANET/INET border and listens for 2027 RS messages originating from or RA messages destined to ANET Clients. 2028 The Proxy acts on these control messages as follows: 2030 o when the Proxy receives an RS message from a new ANET Client, it 2031 first authenticates the message then examines the network-layer 2032 destination address. If the destination address is a Server's 2033 LLA, the Proxy proceeds to the next step. Otherwise, if the 2034 destination is (link-local) All-Routers multicast, the Proxy 2035 selects a "nearby" Server that is likely to be a good candidate to 2036 serve the Client and replaces the destination address with the 2037 Server's LLA. Next, the Proxy creates a proxy neighbor cache 2038 entry and caches the Client and Server link-layer addresses along 2039 with the OMNI option information and any other identifying 2040 information including Transaction IDs, Client Identifiers, Nonce 2041 values, etc. The Proxy finally encapsulates the (proxyed) RS 2042 message in an OAL header with source set to the Proxy's ULA and 2043 destination set to the Server's ULA. The Proxy also includes an 2044 OMNI header with an Interface Attributes option that includes its 2045 own INET address plus a unique Port Number for this Client, then 2046 forwards the message into the OMNI link spanning tree. 2048 o when the Server receives the RS, it authenticates the message then 2049 creates or updates a symmetric neighbor cache entry for the Client 2050 with the Proxy's ULA, INET address and Port Number as the link- 2051 layer address information. The Server then sends an RA message 2052 back to the Proxy via the spanning tree. 2054 o when the Proxy receives the RA, it authenticates the message and 2055 matches it with the proxy neighbor cache entry created by the RS. 2056 The Proxy then caches the prefix information as a mapping from the 2057 Client's MNPs to the Client's link-layer address, caches the 2058 Server's advertised Router Lifetime and sets the neighbor cache 2059 entry state to REACHABLE. The Proxy then optionally rewrites the 2060 Router Lifetime and forwards the (proxyed) message to the Client. 2061 The Proxy finally includes an MTU option (if necessary) with an 2062 MTU to use for the underlying ANET interface. 2064 After the initial RS/RA exchange, the Proxy forwards any Client data 2065 packets for which there is no matching asymmetric neighbor cache 2066 entry to a Bridge using OAL encapsulation with its own ULA as the 2067 source and the ULA corresponding to the Client as the destination. 2068 The Proxy instead forwards any Client data destined to an asymmetric 2069 neighbor cache target directly to the target according to the OAL/ 2070 link-layer information - the process of establishing asymmetric 2071 neighbor cache entries is specified in Section 3.14. 2073 While the Client is still attached to the ANET, the Proxy sends NS, 2074 RS and/or unsolicited NA messages to update the Server's symmetric 2075 neighbor cache entries on behalf of the Client and/or to convey QoS 2076 updates. This allows for higher-frequency Proxy-initiated RS/RA 2077 messaging over well-connected INET infrastructure supplemented by 2078 lower-frequency Client-initiated RS/RA messaging over constrained 2079 ANET data links. 2081 If the Server ceases to send solicited advertisements, the Proxy 2082 sends unsolicited RAs on the ANET interface with destination set to 2083 (link-local) All-Nodes multicast and with Router Lifetime set to zero 2084 to inform Clients that the Server has failed. Although the Proxy 2085 engages in ND exchanges on behalf of the Client, the Client can also 2086 send ND messages on its own behalf, e.g., if it is in a better 2087 position than the Proxy to convey QoS changes, etc. For this reason, 2088 the Proxy marks any Client-originated solicitation messages (e.g. by 2089 inserting a Nonce option) so that it can return the solicited 2090 advertisement to the Client instead of processing it locally. 2092 If the Client becomes unreachable, the Proxy sets the neighbor cache 2093 entry state to DEPARTED and retains the entry for DepartTime seconds. 2094 While the state is DEPARTED, the Proxy forwards any packets destined 2095 to the Client to a Bridge via OAL encapsulation with the Client's 2096 current Server as the destination. The Bridge in turn forwards the 2097 packets to the Client's current Server. When DepartTime expires, the 2098 Proxy deletes the neighbor cache entry and discards any further 2099 packets destined to this (now forgotten) Client. 2101 In some ANETs that employ a Proxy, the Client's MNP can be injected 2102 into the ANET routing system. In that case, the Client can send data 2103 messages without encapsulation so that the ANET routing system 2104 transports the unencapsulated packets to the Proxy. This can be very 2105 beneficial, e.g., if the Client connects to the ANET via low-end data 2106 links such as some aviation wireless links. 2108 If the first-hop ANET access router is on the same underlying link 2109 and recognizes the AERO/OMNI protocol, the Client can avoid 2110 encapsulation for both its control and data messages. When the 2111 Client connects to the link, it can send an unencapsulated RS message 2112 with source address set to its LLA and with destination address set 2113 to the LLA of the Client's selected Server or to (link-local) All- 2114 Routers multicast. The Client includes an OMNI option formatted as 2115 specified in [I-D.templin-6man-omni-interface]. 2117 The Client then sends the unencapsulated RS message, which will be 2118 intercepted by the AERO-Aware access router. The access router then 2119 encapsulates the RS message in an ANET header with its own address as 2120 the source address and the address of a Proxy as the destination 2121 address. The access router further remembers the address of the 2122 Proxy so that it can encapsulate future data packets from the Client 2123 via the same Proxy. If the access router needs to change to a new 2124 Proxy, it simply sends another RS message toward the Server via the 2125 new Proxy on behalf of the Client. 2127 In some cases, the access router and Proxy may be one and the same 2128 node. In that case, the node would be located on the same physical 2129 link as the Client, but its message exchanges with the Server would 2130 need to pass through a security gateway at the ANET/INET border. The 2131 method for deploying access routers and Proxys (i.e. as a single node 2132 or multiple nodes) is an ANET-local administrative consideration. 2134 3.13.1. Combined Proxy/Servers 2136 Clients may need to connect directly to Servers via INET, Direct and 2137 VPNed interfaces (i.e., non-ANET interfaces). If the Client's 2138 underlying interfaces all connect via the same INET partition, then 2139 it can connect to a single controlling Server via all interfaces. 2141 If some Client interfaces connect via different INET partitions, 2142 however, the Client still selects a set of controlling Servers and 2143 sends RS messages via their directly-connected Servers while using 2144 the LLA of the controlling Server as the destination. 2146 When a Server receives an RS with destination set to the LLA of a 2147 controlling Server, it acts as a Proxy to forward the message to the 2148 controlling Server while forwarding the corresponding RA reply to the 2149 Client. 2151 3.13.2. Detecting and Responding to Server Failures 2153 In environments where fast recovery from Server failure is required, 2154 Proxys SHOULD use proactive Neighbor Unreachability Detection (NUD) 2155 to track Server reachability in a similar fashion as for 2156 Bidirectional Forwarding Detection (BFD) [RFC5880]. Proxys can then 2157 quickly detect and react to failures so that cached information is 2158 re-established through alternate paths. The NUD control messaging is 2159 carried only over well-connected ground domain networks (i.e., and 2160 not low-end aeronautical radio links) and can therefore be tuned for 2161 rapid response. 2163 Proxys perform proactive NUD with Servers for which there are 2164 currently active ANET Clients by sending continuous NS messages in 2165 rapid succession, e.g., one message per second. The Proxy sends the 2166 NS message via the spanning tree with the Proxy's LLA as the source 2167 and the LLA of the Server as the destination. When the Proxy is also 2168 sending RS messages to the Server on behalf of ANET Clients, the 2169 resulting RA responses can be considered as equivalent hints of 2170 forward progress. This means that the Proxy need not also send a 2171 periodic NS if it has already sent an RS within the same period. If 2172 the Server fails (i.e., if the Proxy ceases to receive 2173 advertisements), the Proxy can quickly inform Clients by sending 2174 multicast RA messages on the ANET interface. 2176 The Proxy sends RA messages on the ANET interface with source address 2177 set to the Server's address, destination address set to (link-local) 2178 All-Nodes multicast, and Router Lifetime set to 0. The Proxy SHOULD 2179 send MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small 2180 delays [RFC4861]. Any Clients on the ANET that had been using the 2181 failed Server will receive the RA messages and associate with a new 2182 Server. 2184 3.13.3. Point-to-Multipoint Server Coordination 2186 In environments where Client messaging over ANETs is bandwidth- 2187 limited and/or expensive, Clients can enlist the services of the 2188 Proxy to coordinate with multiple Servers in a single RS/RA message 2189 exchange. The Client can send a single RS message to (link-local) 2190 All-Routers multicast that includes the ID's of multiple Servers in 2191 MS-Register sub-options of the OMNI option. 2193 When the Proxy receives the RS and processes the OMNI option, it 2194 sends a separate RS to each MS-Register Server ID. When the Proxy 2195 receives an RA, it can optionally return an immediate "singleton" RA 2196 to the Client or record the Server's ID for inclusion in a pending 2197 "aggregate" RA message. The Proxy can then return aggregate RA 2198 messages to the Client including multiple Server IDs in order to 2199 conserve bandwidth. Each RA includes a proper subset of the Server 2200 IDs from the original RS message, and the Proxy must ensure that the 2201 message contents of each RA are consistent with the information 2202 received from the (aggregated) Servers. 2204 Clients can thereafter employ efficient point-to-multipoint Server 2205 coordination under the assistance of the Proxy to reduce the number 2206 of messages sent over the ANET while enlisting the support of 2207 multiple Servers for fault tolerance. Clients can further include 2208 MS-Release sub-options in IPv6 ND messages to request the Proxy to 2209 release from former Servers via the procedures discussed in 2210 Section 3.16.5. 2212 The OMNI interface specification [I-D.templin-6man-omni-interface] 2213 provides further discussion of the Client/Proxy RS/RA messaging 2214 involved in point-to-multipoint coordination. 2216 3.14. AERO Route Optimization / Address Resolution 2218 While data packets are flowing between a source and target node, 2219 route optimization SHOULD be used. Route optimization is initiated 2220 by the first eligible Route Optimization Source (ROS) closest to the 2221 source as follows: 2223 o For Clients on VPNed and Direct interfaces, the Server is the ROS. 2225 o For Clients on ANET interfaces, the Proxy is the ROS. 2227 o For Clients on INET interfaces, the Client itself is the ROS. 2229 o For correspondent nodes on INET/EUN interfaces serviced by a 2230 Relay, the Relay is the ROS. 2232 The route optimization procedure is conducted between the ROS and the 2233 target Server/Relay acting as a Route Optimization Responder (ROR) in 2234 the same manner as for IPv6 ND Address Resolution and using the same 2235 NS/NA messaging. The target may either be a MNP Client serviced by a 2236 Server, or a non-MNP correspondent reachable via a Relay. 2238 The procedures are specified in the following sections. 2240 3.14.1. Route Optimization Initiation 2242 While data packets are flowing from the source node toward a target 2243 node, the ROS performs address resolution by sending an NS message 2244 for Address Resolution (NS(AR)) to receive a solicited NA message 2245 from the ROR. When the ROS sends an NS(AR), it includes: 2247 o the LLA of the ROS as the source address. 2249 o the data packet's destination as the Target Address. 2251 o the Solicited-Node multicast address [RFC4291] formed from the 2252 lower 24 bits of the data packet's destination as the destination 2253 address, e.g., for 2001:db8:1:2::10:2000 the NS destination 2254 address is ff02:0:0:0:0:1:ff10:2000. 2256 The NS(AR) message includes an OMNI option with no Interface 2257 Attributes, such that the target will not create a neighbor cache 2258 entry. The Prefix Length in the OMNI option is set to the Prefix 2259 Length associated with the ROS's LLA. 2261 The ROS then encapsulates the NS(AR) message in an OAL header with 2262 source set to its own ULA and destination set to the ULA 2263 corresponding to the target, then sends the message into the spanning 2264 tree without decrementing the network-layer TTL/Hop Limit field. 2265 (When the ROS is a Client, it instead securely sends the NS(AR) to 2266 one of its current Servers by including an Authentication option per 2267 [RFC4380]. The Server then forwards the message into the spanning 2268 tree on behalf of the Client, while setting the IPv6 source address 2269 and the OAL source address to the LLA and ULA of the Client, 2270 respectively.) 2272 3.14.2. Relaying the NS 2274 When the Bridge receives the NS(AR) message from the ROS, it discards 2275 the INET header and determines that the ROR is the next hop by 2276 consulting its standard IPv6 forwarding table for the OAL header 2277 destination address. The Bridge then forwards the message toward the 2278 ROR via the spanning tree the same as for any IPv6 router. The 2279 final-hop Bridge in the spanning tree will deliver the message via a 2280 secured tunnel to the ROR. 2282 3.14.3. Processing the NS and Sending the NA 2284 When the ROR receives the NS(AR) message, it examines the Target 2285 Address to determine whether it has a neighbor cache entry and/or 2286 route that matches the target. If there is no match, the ROR drops 2287 the message. Otherwise, the ROR continues processing as follows: 2289 o if the target belongs to an MNP Client neighbor in the DEPARTED 2290 state the ROR changes the NS(AR) message OAL destination address 2291 to the ULA of the Client's new Server, forwards the message into 2292 the spanning tree and returns from processing. 2294 o If the target belongs to an MNP Client neighbor in the REACHABLE 2295 state, the ROR instead adds the AERO source address to the target 2296 Client's Report List with time set to ReportTime. 2298 o If the target belongs to a non-MNP route, the ROR continues 2299 processing without adding an entry to the Report List. 2301 The ROR then prepares a solicited NA message to send back to the ROS 2302 but does not create a neighbor cache entry. The ROR sets the NA 2303 source address to the LLA corresponding to the target, sets the 2304 Target Address to the target of the solicitation, and sets the 2305 destination address to the source of the solicitation. The ROR then 2306 includes an OMNI option with Prefix Length set to the length 2307 associated with the LLA. 2309 If the target is an MNP Client, the ROR next includes Interface 2310 Attributes in the OMNI option for each of the target Client's 2311 underlying interfaces with current information for each interface and 2312 with the S/T-ifIndex field in the OMNI header set to 0 to indicate 2313 that the message originated from the ROR and not the Client. 2315 For each Interface Attributes sub-option, the ROR sets the L2ADDR 2316 according to its own INET address for VPNed or Direct interfaces, to 2317 the INET address of the Proxy or to the Client's INET address for 2318 INET interfaces. The ROR then includes the lower 32 bits of its own 2319 ULA (or the ULA of the Proxy) as the LHS, encodes the ULA prefix 2320 length code in the SRT field and sets the FMT code accordingly as 2321 specified in Section 3.3. 2323 The ROR then sets the NA message R flag to 1 (as a router), S flag to 2324 1 (as a response to a solicitation), and O flag to 0 (as a proxy). 2325 The ROR finally encapsulates the NA message in an OAL header with 2326 source set to its own ULA and destination set to the source ULA of 2327 the NS(AR) message, then forwards the message into the spanning tree 2328 without decrementing the network-layer TTL/Hop Limit field. 2330 3.14.4. Relaying the NA 2332 When the Bridge receives the NA message from the ROR, it discards the 2333 INET header and determines that the ROS is the next hop by consulting 2334 its standard IPv6 forwarding table for the OAL header destination 2335 address. The Bridge then forwards the OAL-encapsulated NA message 2336 toward the ROS the same as for any IPv6 router. The final-hop Bridge 2337 in the spanning tree will deliver the message via a secured tunnel to 2338 the ROS. 2340 3.14.5. Processing the NA 2342 When the ROS receives the solicited NA message, it processes the 2343 message the same as for standard IPv6 Address Resolution [RFC4861]. 2344 In the process, it caches the source ULA then creates an asymmetric 2345 neighbor cache entry for the target and caches all information found 2346 in the OMNI option. The ROS finally sets the asymmetric neighbor 2347 cache entry lifetime to ReachableTime seconds. (When the ROS is a 2348 Client, the solicited NA message will first be delivered via the 2349 spanning tree to one of its current Servers, which then securely 2350 forwards the message to the Client by including an Authentication 2351 option per [RFC4380]. 2353 3.14.6. Route Optimization Maintenance 2355 Following route optimization, the ROS forwards future data packets 2356 destined to the target via the addresses found in the cached link- 2357 layer information. The route optimization is shared by all sources 2358 that send packets to the target via the ROS, i.e., and not just the 2359 source on behalf of which the route optimization was initiated. 2361 While new data packets destined to the target are flowing through the 2362 ROS, it sends additional NS(AR) messages to the ROR before 2363 ReachableTime expires to receive a fresh solicited NA message the 2364 same as described in the previous sections (route optimization 2365 refreshment strategies are an implementation matter, with a non- 2366 normative example given in Appendix A.1). The ROS uses the cached 2367 ULA of the ROR as the NS(AR) OAL destination address (i.e., instead 2368 of using the ULA corresponding to the target as was the case for the 2369 initial NS(AR)), and sends up to MAX_MULTICAST_SOLICIT NS(AR) 2370 messages separated by 1 second until an NA is received. If no NA is 2371 received, the ROS assumes that the current ROR has become unreachable 2372 and deletes the target neighbor cache entry. Subsequent data packets 2373 will trigger a new route optimization with an NS with OAL destination 2374 address set to the ULA corresponding to the target per Section 3.14.1 2375 to discover a new ROR while initial data packets travel over a 2376 suboptimal route. 2378 If an NA is received, the ROS then updates the asymmetric neighbor 2379 cache entry to refresh ReachableTime, while (for MNP destinations) 2380 the ROR adds or updates the ROS address to the target's Report List 2381 and with time set to ReportTime. While no data packets are flowing, 2382 the ROS instead allows ReachableTime for the asymmetric neighbor 2383 cache entry to expire. When ReachableTime expires, the ROS deletes 2384 the asymmetric neighbor cache entry. Any future data packets flowing 2385 through the ROS will again trigger a new route optimization. 2387 The ROS may also receive unsolicited NA messages from the ROR at any 2388 time (see: Section 3.16). If there is an asymmetric neighbor cache 2389 entry for the target, the ROS updates the link-layer information but 2390 does not update ReachableTime since the receipt of an unsolicited NA 2391 does not confirm that any forward paths are working. If there is no 2392 asymmetric neighbor cache entry, the ROS simply discards the 2393 unsolicited NA. 2395 In this arrangement, the ROS holds an asymmetric neighbor cache entry 2396 for the target via the ROR, but the ROR does not hold an asymmetric 2397 neighbor cache entry for the ROS. The route optimization neighbor 2398 relationship is therefore asymmetric and unidirectional. If the 2399 target node also has packets to send back to the source node, then a 2400 separate route optimization procedure is performed in the reverse 2401 direction. But, there is no requirement that the forward and reverse 2402 paths be symmetric. 2404 3.15. Neighbor Unreachability Detection (NUD) 2406 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2407 [RFC4861] either reactively in response to persistent link-layer 2408 errors (see Section 3.11) or proactively to confirm reachability. 2409 The NUD algorithm is based on periodic control message exchanges. 2410 The algorithm may further be seeded by ND hints of forward progress, 2411 but care must be taken to avoid inferring reachability based on 2412 spoofed information. For example, authentic IPv6 ND message 2413 exchanges may be considered as acceptable hints of forward progress, 2414 while spurious data packets should not be. 2416 AERO Servers, Proxys and Relays can use (OAL-encapsulated) standard 2417 NS/NA NUD exchanges sent over the OMNI link spanning tree to securely 2418 test reachability without risk of DoS attacks from nodes pretending 2419 to be a neighbor (these NS/NA(NUD) messages use the unicast LLAs and 2420 ULAs of the two parties involved in the NUD test the same as for 2421 standard IPv6 ND). Proxys can further perform NUD to securely verify 2422 Server reachability on behalf of their proxyed Clients. However, a 2423 means for an ROS to test the unsecured forward directions of target 2424 route optimized paths is also necessary. 2426 When an ROR directs an ROS to a neighbor with one or more target 2427 link-layer addresses, the ROS can proactively test each such 2428 unsecured route optimized path by sending "loopback" NS(NUD) 2429 messages. While testing the paths, the ROS can optionally continue 2430 to send packets via the spanning tree, maintain a small queue of 2431 packets until target reachability is confirmed, or (optimistically) 2432 allow packets to flow via the route optimized paths. 2434 When the ROS sends a loopback NS(NUD) message, it uses its own LLA as 2435 both the IPv6 source and destination address, and the MNP Subnet- 2436 Router anycast address as the Target Address. The ROS includes a 2437 Nonce and Timestamp option, then encapsulates the message in OAL/INET 2438 headers with its own ULA as the source and the ULA of the route 2439 optimization target as the destination. The ROS then forwards the 2440 message to the target (either directly to the L2ADDR of the target if 2441 the target is in the same OMNI link segment, or via a Bridge if the 2442 target is in a different OMNI link segment). 2444 When the route optimization target receives the NS(NUD) message, it 2445 notices that the IPv6 destination address is the same as the source 2446 address. It then reverses the OAL header source and destination 2447 addresses and returns the message to the ROS (either directly or via 2448 the spanning tree). The route optimization target does not decrement 2449 the NS(NUD) message IPv6 Hop-Limit in the process, since the message 2450 has not exited the OMNI link. 2452 When the ROS receives the NS(NUD) message, it can determine from the 2453 Nonce, Timestamp and Target Address that the message originated from 2454 itself and that it transited the forward path. The ROS need not 2455 prepare an NA response, since the destination of the response would 2456 be itself and testing the route optimization path again would be 2457 redundant. 2459 The ROS marks route optimization target paths that pass these NUD 2460 tests as "reachable", and those that do not as "unreachable". These 2461 markings inform the OMNI interface forwarding algorithm specified in 2462 Section 3.10. 2464 Note that to avoid a DoS vector nodes MUST NOT return loopback 2465 NS(NUD) messages received from an unsecured link-layer source via the 2466 (secured) spanning tree. 2468 3.16. Mobility Management and Quality of Service (QoS) 2470 AERO is a Distributed Mobility Management (DMM) service. Each Server 2471 is responsible for only a subset of the Clients on the OMNI link, as 2472 opposed to a Centralized Mobility Management (CMM) service where 2473 there is a single network mobility collective entity for all Clients. 2474 Clients coordinate with their associated Servers via RS/RA exchanges 2475 to maintain the DMM profile, and the AERO routing system tracks all 2476 current Client/Server peering relationships. 2478 Servers provide default routing and mobility/multilink services for 2479 their dependent Clients. Clients are responsible for maintaining 2480 neighbor relationships with their Servers through periodic RS/RA 2481 exchanges, which also serves to confirm neighbor reachability. When 2482 a Client's underlying interface address and/or QoS information 2483 changes, the Client is responsible for updating the Server with this 2484 new information. Note that when there is a Proxy in the path, the 2485 Proxy can also perform some RS/RA exchanges on the Client's behalf. 2487 Mobility management messaging is based on the transmission and 2488 reception of unsolicited Neighbor Advertisement (uNA) messages. Each 2489 uNA message sets the IPv6 destination address to (link-local) All- 2490 Nodes multicast to convey a general update of Interface Attributes to 2491 (possibly) multiple recipients, or to a specific unicast LLA to 2492 announce a departure event to a specific recipient. Implementations 2493 must therefore examine the destination address to determine the 2494 nature of the mobility event (i.e., update vs departure). 2496 Mobility management considerations are specified in the following 2497 sections. 2499 3.16.1. Mobility Update Messaging 2501 Servers accommodate Client mobility, multilink and/or QoS change 2502 events by sending unsolicited NA (uNA) messages to each ROS in the 2503 target Client's Report List. When a Server sends a uNA message, it 2504 sets the IPv6 source address to the Client's LLA, sets the 2505 destination address to (link-local) All-Nodes multicast and sets the 2506 Target Address to the Client's Subnet-Router anycast address. The 2507 Server also includes an OMNI option with Prefix Length set to the 2508 length associated with the Client's LLA, with Interface Attributes 2509 for the target Client's underlying interfaces and with the OMNI 2510 header S/T-ifIndex set to 0. The Server then sets the NA R flag to 2511 1, the S flag to 0 and the O flag to 1, then encapsulates the message 2512 in an OAL header with source set to its own ULA and destination set 2513 to the ULA of the ROS and sends the message into the spanning tree. 2515 As discussed in Section 7.2.6 of [RFC4861], the transmission and 2516 reception of uNA messages is unreliable but provides a useful 2517 optimization. In well-connected Internetworks with robust data links 2518 uNA messages will be delivered with high probability, but in any case 2519 the Server can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT uNAs 2520 to each ROS to increase the likelihood that at least one will be 2521 received. 2523 When the ROS receives a uNA message prepared as above, it ignores the 2524 message if there is no existing neighbor cache entry for the Client. 2525 Otherwise, it uses the included OMNI option information to update the 2526 neighbor cache entry, but does not reset ReachableTime since the 2527 receipt of an unsolicited NA message from the target Server does not 2528 provide confirmation that any forward paths to the target Client are 2529 working. 2531 If uNA messages are lost, the ROS may be left with stale address and/ 2532 or QoS information for the Client for up to ReachableTime seconds. 2533 During this time, the ROS can continue sending packets according to 2534 its stale neighbor cache information. When ReachableTime is close to 2535 expiring, the ROS will re-initiate route optimization and receive 2536 fresh link-layer address information. 2538 In addition to sending uNA messages to the current set of ROSs for 2539 the Client, the Server also sends uNAs to the ULA associated with the 2540 link-layer address for any underlying interface for which the link- 2541 layer address has changed. These uNA messages update an old Proxy/ 2542 Server that cannot easily detect (e.g., without active probing) when 2543 a formerly-active Client has departed. When the Server sends the 2544 uNA, it sets the IPv6 source address to the Client's LLA, sets the 2545 destination address to the old Proxy/Server's LLA, and sets the 2546 Target Address to the Client's Subnet-Router anycast address. The 2547 Server also includes an OMNI option with Prefix Length set to the 2548 length associated with the Client's LLA, with Interface Attributes 2549 for the changed underlying interface, and with the OMNI header S/ 2550 T-ifIndex set to 0. The Server then sets the NA R flag to 1, the S 2551 flag to 0 and the O flag to 1, then encapsulates the message in an 2552 OAL header with source set to its own ULA and destination set to the 2553 ULA of the old Proxy/Server and sends the message into the spanning 2554 tree. 2556 3.16.2. Announcing Link-Layer Address and/or QoS Preference Changes 2558 When a Client needs to change its underlying interface addresses and/ 2559 or QoS preferences (e.g., due to a mobility event), either the Client 2560 or its Proxys send RS messages to the Server via the spanning tree 2561 with an OMNI option that includes Interface attributes with the new 2562 link quality and address information. 2564 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 2565 sending actual data packets in case one or more RAs are lost. If all 2566 RAs are lost, the Client SHOULD re-associate with a new Server. 2568 When the Server receives the Client's changes, it sends uNA messages 2569 to all nodes in the Report List the same as described in the previous 2570 section. 2572 3.16.3. Bringing New Links Into Service 2574 When a Client needs to bring new underlying interfaces into service 2575 (e.g., when it activates a new data link), it sends an RS message to 2576 the Server via the underlying interface with an OMNI option that 2577 includes Interface Attributes with appropriate link quality values 2578 and with link-layer address information for the new link. 2580 3.16.4. Deactivating Existing Links 2582 When a Client needs to deactivate an existing underlying interface, 2583 it sends an RS or uNA message to its Server with an OMNI option with 2584 appropriate Interface Attribute values - in particular, the link 2585 quality value 0 assures that neighbors will cease to use the link. 2587 If the Client needs to send RS/uNA messages over an underlying 2588 interface other than the one being deactivated, it MUST include 2589 Interface Attributes with appropriate link quality values for any 2590 underlying interfaces being deactivated. 2592 Note that when a Client deactivates an underlying interface, 2593 neighbors that have received the RS/uNA messages need not purge all 2594 references for the underlying interface from their neighbor cache 2595 entries. The Client may reactivate or reuse the underlying interface 2596 and/or its ifIndex at a later point in time, when it will send RS/uNA 2597 messages with fresh Interface Attributes to update any neighbors. 2599 3.16.5. Moving Between Servers 2601 The Client performs the procedures specified in Section 3.12.2 when 2602 it first associates with a new Server or renews its association with 2603 an existing Server. The Client also includes MS-Release identifiers 2604 in the RS message OMNI option per [I-D.templin-6man-omni-interface] 2605 if it wants the new Server to notify any old Servers from which the 2606 Client is departing. 2608 When the new Server receives the Client's RS message, it returns an 2609 RA as specified in Section 3.12.3 and sends up to 2610 MAX_NEIGHBOR_ADVERTIISEMENT uNA messages to any old Servers listed in 2611 OMNI option MS-Release identifiers. When the new Server sends a uNA 2612 message, it sets the IPv6 source address to the Client's LLA, sets 2613 the destination address to the old Server's LLA, and sets the Target 2614 Address to the Client's Subnet-Router anycast address. The new 2615 Server also includes an OMNI option with Prefix Length set to the 2616 length associated with the Client's LLA, with Interface Attributes 2617 for its own underlying interface, and with the OMNI header S/ 2618 T-ifIndex set to 0. The new Server then sets the NA R flag to 1, the 2619 S flag to 0 and the O flag to 1, then encapsulates the message in an 2620 OAL header with source set to its own ULA and destination set to the 2621 ULA of the old Server and sends the message into the spanning tree. 2623 When an old Server receives the uNA, it changes the Client's neighbor 2624 cache entry state to DEPARTED, sets the link-layer address of the 2625 Client to the new Server's ULA, and resets DepartTime. After a short 2626 delay (e.g., 2 seconds) the old Server withdraws the Client's MNP 2627 from the routing system. After DepartTime expires, the old Server 2628 deletes the Client's neighbor cache entry. 2630 The old Server also iteratively forwards a copy of the uNA message to 2631 each ROS in the Client's Report List by changing the OAL destination 2632 address to the ULA of the ROS while leaving all other fields of the 2633 message unmodified. When the ROS receives the uNA, it examines the 2634 Target address to determine the correct asymmetric neighbor cache 2635 entry and verifies that the IPv6 destination address matches the old 2636 Server. The ROS then caches the IPv6 source address as the new 2637 Server for the existing asymmetric neighbor cache entry and marks the 2638 entry as STALE. While in the STALE state, the ROS allows new data 2639 packets to flow according to any existing cached link-layer 2640 information and sends new NS(AR) messages using its own ULA as the 2641 OAL source and the ULA of the new Server as the OAL destination 2642 address to elicit NA messages that reset the asymmetric neighbor 2643 cache entry state to REACHABLE. If no new NA message is received for 2644 10 seconds while in the STALE state, the ROS deletes the neighbor 2645 cache entry. 2647 Clients SHOULD NOT move rapidly between Servers in order to avoid 2648 causing excessive oscillations in the AERO routing system. Examples 2649 of when a Client might wish to change to a different Server include a 2650 Server that has gone unreachable, topological movements of 2651 significant distance, movement to a new geographic region, movement 2652 to a new OMNI link segment, etc. 2654 When a Client moves to a new Server, some of the fragments of a 2655 multiple fragment packet may have already arrived at the old Server 2656 while others are en route to the new Server, however no special 2657 attention in the reassembly algorithm is necessary when re-routed 2658 fragments are simply treated as loss. 2660 3.17. Multicast 2662 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 2663 [RFC3810] proxy service for its EUNs and/or hosted applications 2664 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 2665 underlying interfaces for which group membership is required. The 2666 IGMP/MLD messages may be further forwarded by a first-hop ANET access 2667 router acting as an IGMP/MLD-snooping switch [RFC4541], then 2668 ultimately delivered to an AERO Proxy/Server acting as a Protocol 2669 Independent Multicast - Sparse-Mode (PIM-SM, or simply "PIM") 2670 Designated Router (DR) [RFC7761]. AERO Relays also act as PIM 2671 routers (i.e., the same as AERO Proxys/Servers) on behalf of nodes on 2672 INET/EUN networks. The behaviors identified in the following 2673 sections correspond to Source-Specific Multicast (SSM) and Any-Source 2674 Multicast (ASM) operational modes. 2676 3.17.1. Source-Specific Multicast (SSM) 2678 When an ROS (i.e., an AERO Proxy/Server/Relay) "X" acting as PIM 2679 router receives a Join/Prune message from a node on its downstream 2680 interfaces containing one or more ((S)ource, (G)roup) pairs, it 2681 updates its Multicast Routing Information Base (MRIB) accordingly. 2682 For each S belonging to a prefix reachable via X's non-OMNI 2683 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 2684 on those interfaces per [RFC7761]. 2686 For each S belonging to a prefix reachable via X's OMNI interface, X 2687 originates a separate copy of the Join/Prune for each (S,G) in the 2688 message using its own LLA as the source address and ALL-PIM-ROUTERS 2689 as the destination address. X then encapsulates each message in an 2690 OAL header with source address set to the ULA of X and destination 2691 address set to S then forwards the message into the spanning tree, 2692 which delivers it to AERO Server/Relay "Y" that services S. At the 2693 same time, if the message was a Join, X sends a route-optimization NS 2694 message toward each S the same as discussed in Section 3.14. The 2695 resulting NAs will return the LLA for the prefix that matches S as 2696 the network-layer source address and with an OMNI option with the ULA 2697 corresponding to any underlying interfaces that are currently 2698 servicing S. 2700 When Y processes the Join/Prune message, if S located behind any 2701 INET, Direct, or VPNed interfaces Y acts as a PIM router and updates 2702 its MRIB to list X as the next hop in the reverse path. If S is 2703 located behind any Proxys "Z"*, Y also forwards the message to each 2704 Z* over the spanning tree while continuing to use the LLA of X as the 2705 source address. Each Z* then updates its MRIB accordingly and 2706 maintains the LLA of X as the next hop in the reverse path. Since 2707 the Bridges do not examine network layer control messages, this means 2708 that the (reverse) multicast tree path is simply from each Z* (and/or 2709 Y) to X with no other multicast-aware routers in the path. If any Z* 2710 (and/or Y) is located on the same OMNI link segment as X, the 2711 multicast data traffic sent to X directly using OAL/INET 2712 encapsulation instead of via a Bridge. 2714 Following the initial Join/Prune and NS/NA messaging, X maintains an 2715 asymmetric neighbor cache entry for each S the same as if X was 2716 sending unicast data traffic to S. In particular, X performs 2717 additional NS/NA exchanges to keep the neighbor cache entry alive for 2718 up to t_periodic seconds [RFC7761]. If no new Joins are received 2719 within t_periodic seconds, X allows the neighbor cache entry to 2720 expire. Finally, if X receives any additional Join/Prune messages 2721 for (S,G) it forwards the messages to each Y and Z* in the neighbor 2722 cache entry over the spanning tree. 2724 At some later time, Client C that holds an MNP for source S may 2725 depart from a first Proxy Z1 and/or connect via a new Proxy Z2. In 2726 that case, Y sends an unsolicited NA message to X the same as 2727 specified for unicast mobility in Section 3.16. When X receives the 2728 unsolicited NA message, it updates its asymmetric neighbor cache 2729 entry for the LLA for source S and sends new Join messages to any new 2730 Proxys Z2. There is no requirement to send any Prune messages to old 2731 Proxys Z1 since source S will no longer source any multicast data 2732 traffic via Z1. Instead, the multicast state for (S,G) in Proxy Z1 2733 will soon time out since no new Joins will arrive. 2735 After some later time, C may move to a new Server Y2 and depart from 2736 old Sever Y1. In that case, Y1 sends Join messages for any of C's 2737 active (S,G) groups to Y2 while including its own LLA as the source 2738 address. This causes Y2 to include Y1 in the multicast forwarding 2739 tree during the interim time that Y1's symmetric neighbor cache entry 2740 for C is in the DEPARTED state. At the same time, Y1 sends an 2741 unsolicited NA message to X with an OMNI option with S/T-ifIndex in 2742 the header set to 0 and a release indication to cause X to release 2743 its asymmetric neighbor cache entry. X then sends a new Join message 2744 to S via the spanning tree and re-initiates route optimization the 2745 same as if it were receiving a fresh Join message from a node on a 2746 downstream link. 2748 3.17.2. Any-Source Multicast (ASM) 2750 When an ROS X acting as a PIM router receives a Join/Prune from a 2751 node on its downstream interfaces containing one or more (*,G) pairs, 2752 it updates its Multicast Routing Information Base (MRIB) accordingly. 2753 X then forwards a copy of the message to the Rendezvous Point (RP) R 2754 for each G over the spanning tree. X uses its own LLA as the source 2755 address and ALL-PIM-ROUTERS as the destination address, then 2756 encapsulates each message in an OAL header with source address set to 2757 the ULA of X and destination address set to R, then sends the message 2758 into the spanning tree. At the same time, if the message was a Join 2759 X initiates NS/NA route optimization the same as for the SSM case 2760 discussed in Section 3.17.1. 2762 For each source S that sends multicast traffic to group G via R, the 2763 Proxy/Server Z* for the Client that aggregates S encapsulates the 2764 packets in PIM Register messages and forwards them to R via the 2765 spanning tree, which may then elect to send a PIM Join to Z*. This 2766 will result in an (S,G) tree rooted at Z* with R as the next hop so 2767 that R will begin to receive two copies of the packet; one native 2768 copy from the (S, G) tree and a second copy from the pre-existing (*, 2769 G) tree that still uses PIM Register encapsulation. R can then issue 2770 a PIM Register-stop message to suppress the Register-encapsulated 2771 stream. At some later time, if C moves to a new Proxy/Server Z*, it 2772 resumes sending packets via PIM Register encapsulation via the new 2773 Z*. 2775 At the same time, as multicast listeners discover individual S's for 2776 a given G, they can initiate an (S,G) Join for each S under the same 2777 procedures discussed in Section 3.17.1. Once the (S,G) tree is 2778 established, the listeners can send (S, G) Prune messages to R so 2779 that multicast packets for group G sourced by S will only be 2780 delivered via the (S, G) tree and not from the (*, G) tree rooted at 2781 R. All mobility considerations discussed for SSM apply. 2783 3.17.3. Bi-Directional PIM (BIDIR-PIM) 2785 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 2786 approach to ASM that treats the Rendezvous Point (RP) as a Designated 2787 Forwarder (DF). Further considerations for BIDIR-PIM are out of 2788 scope. 2790 3.18. Operation over Multiple OMNI Links 2792 An AERO Client can connect to multiple OMNI links the same as for any 2793 data link service. In that case, the Client maintains a distinct 2794 OMNI interface for each link, e.g., 'omni0' for the first link, 2795 'omni1' for the second, 'omni2' for the third, etc. Each OMNI link 2796 would include its own distinct set of Bridges, Servers and Proxys, 2797 thereby providing redundancy in case of failures. 2799 Each OMNI link could utilize the same or different ANET connections. 2800 The links can be distinguished at the link-layer via the SRT prefix 2801 in a similar fashion as for Virtual Local Area Network (VLAN) tagging 2802 (e.g., IEEE 802.1Q) and/or through assignment of distinct sets of 2803 MSPs on each link. This gives rise to the opportunity for supporting 2804 multiple redundant networked paths, with each VLAN distinguished by a 2805 different SRT "color" (see: Section 3.2.5). 2807 The Client's IP layer can select the outgoing OMNI interface 2808 appropriate for a given traffic profile while (in the reverse 2809 direction) correspondent nodes must have some way of steering their 2810 packets destined to a target via the correct OMNI link. 2812 In a first alternative, if each OMNI link services different MSPs, 2813 then the Client can receive a distinct MNP from each of the links. 2814 IP routing will therefore assure that the correct Red/Green/Blue/etc. 2815 network is used for both outbound and inbound traffic. This can be 2816 accomplished using existing technologies and approaches, and without 2817 requiring any special supporting code in correspondent nodes or 2818 Bridges. 2820 In a second alternative, if each OMNI link services the same MSP(s) 2821 then each link could assign a distinct "OMNI link Anycast" address 2822 that is configured by all Bridges on the link. Correspondent nodes 2823 can then perform Segment Routing to select the correct SRT, which 2824 will then direct the packet over multiple hops to the target. 2826 3.19. DNS Considerations 2828 AERO Client MNs and INET correspondent nodes consult the Domain Name 2829 System (DNS) the same as for any Internetworking node. When 2830 correspondent nodes and Client MNs use different IP protocol versions 2831 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 2832 A records for IPv4 address mappings to MNs which must then be 2833 populated in Relay NAT64 mapping caches. In that way, an IPv4 2834 correspondent node can send packets to the IPv4 address mapping of 2835 the target MN, and the Relay will translate the IPv4 header and 2836 destination address into an IPv6 header and IPv6 destination address 2837 of the MN. 2839 When an AERO Client registers with an AERO Server, the Server can 2840 return the address(es) of DNS servers in RDNSS options [RFC6106]. 2841 The DNS server provides the IP addresses of other MNs and 2842 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 2844 3.20. Transition Considerations 2846 OAL encapsulation ensures that dissimilar INET partitions can be 2847 joined into a single unified OMNI link, even though the partitions 2848 themselves may have differing protocol versions and/or incompatible 2849 addressing plans. However, a commonality can be achieved by 2850 incrementally distributing globally routable (i.e., native) IP 2851 prefixes to eventually reach all nodes (both mobile and fixed) in all 2852 OMNI link segments. This can be accomplished by incrementally 2853 deploying AERO Relays on each INET partition, with each Relay 2854 distributing its MNPs and/or discovering non-MNP prefixes on its INET 2855 links. 2857 This gives rise to the opportunity to eventually distribute native IP 2858 addresses to all nodes, and to present a unified OMNI link view even 2859 if the INET partitions remain in their current protocol and 2860 addressing plans. In that way, the OMNI link can serve the dual 2861 purpose of providing a mobility/multilink service and a transition 2862 service. Or, if an INET partition is transitioned to a native IP 2863 protocol version and addressing scheme that is compatible with the 2864 OMNI link MNP-based addressing scheme, the partition and OMNI link 2865 can be joined by Relays. 2867 Relays that connect INETs/EUNs with dissimilar IP protocol versions 2868 may need to employ a network address and protocol translation 2869 function such as NAT64 [RFC6146]. 2871 3.21. Detecting and Reacting to Server and Bridge Failures 2873 In environments where rapid failure recovery is required, Servers and 2874 Bridges SHOULD use Bidirectional Forwarding Detection (BFD) 2875 [RFC5880]. Nodes that use BFD can quickly detect and react to 2876 failures so that cached information is re-established through 2877 alternate nodes. BFD control messaging is carried only over well- 2878 connected ground domain networks (i.e., and not low-end radio links) 2879 and can therefore be tuned for rapid response. 2881 Servers and Bridges maintain BFD sessions in parallel with their BGP 2882 peerings. If a Server or Bridge fails, BGP peers will quickly re- 2883 establish routes through alternate paths the same as for common BGP 2884 deployments. Similarly, Proxys maintain BFD sessions with their 2885 associated Bridges even though they do not establish BGP peerings 2886 with them. 2888 Proxys SHOULD use proactive NUD for Servers for which there are 2889 currently active ANET Clients in a manner that parallels BFD, i.e., 2890 by sending unicast NS messages in rapid succession to receive 2891 solicited NA messages. When the Proxy is also sending RS messages on 2892 behalf of ANET Clients, the RS/RA messaging can be considered as 2893 equivalent hints of forward progress. This means that the Proxy need 2894 not also send a periodic NS if it has already sent an RS within the 2895 same period. If a Server fails, the Proxy will cease to receive 2896 advertisements and can quickly inform Clients of the outage by 2897 sending multicast RA messages on the ANET interface. 2899 The Proxy sends multicast RA messages with source address set to the 2900 Server's address, destination address set to (link-local) All-Nodes 2901 multicast, and Router Lifetime set to 0. The Proxy SHOULD send 2902 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2903 [RFC4861]. Any Clients on the ANET interface that have been using 2904 the (now defunct) Server will receive the RA messages and associate 2905 with a new Server. 2907 3.22. AERO Clients on the Open Internet 2909 AERO Clients that connect to the open Internet via INET interfaces 2910 can establish a VPN or direct link to securely connect to a Server in 2911 a "tethered" arrangement with all of the Client's traffic transiting 2912 the Server. Alternatively, the Client can associate with an INET 2913 Server using UDP/IP encapsulation and asymmetric securing services as 2914 discussed in the following sections. 2916 When a Client's OMNI interface enables an INET underlying interface, 2917 it first determines whether the interface is likely to be behind a 2918 NAT. For IPv4, the Client assumes it is on the open Internet if the 2919 INET address is not a special-use IPv4 address per [RFC3330]. 2920 Similarly for IPv6, the Client assumes it is on the open Internet if 2921 the INET address is not a link-local [RFC4291] or unique-local 2922 [RFC4193] IPv6 address. 2924 The Client then prepares a UDP/IP-encapsulated RS message with IPv6 2925 source address set to its LLA, with IPv6 destination set to (link- 2926 local) All-Routers multicast and with an OMNI option with underlying 2927 interface attributes. If the Client believes that it is on the open 2928 Internet, it SHOULD include Interface Attributes with the L2ADDR used 2929 for INET encapsulation (otherwise, it MAY omit L2ADDR). If the 2930 underlying address is IPv4, the Client includes the Port Number and 2931 IPv4 address written in obfuscated form [RFC4380] as discussed in 2932 Section 3.3. If the underlying interface address is IPv6, the Client 2933 instead includes the Port Number and IPv6 address in obfuscated form. 2934 The Client finally includes an Authentication option per [RFC4380] to 2935 provide message authentication, sets the UDP/IP source to its INET 2936 address and UDP port, sets the UDP/IP destination to the Server's 2937 INET address and the AERO service port number (8060), then sends the 2938 message to the Server. 2940 When the Server receives the RS, it authenticates the message and 2941 registers the Client's MNP and INET interface information according 2942 to the OMNI option parameters. If the RS message includes an L2ADDR 2943 in the OMNI option, the Server compares the encapsulation IP address 2944 and UDP port number with the (unobfuscated) values. If the values 2945 are the same, the Server caches the Client's information as "INET" 2946 addresses meaning that the Client is likely to accept direct messages 2947 without requiring NAT traversal exchanges. If the values are 2948 different (or, if the OMNI option did not include an L2ADDR) the 2949 Server instead caches the Client's information as "NAT" addresses 2950 meaning that NAT traversal exchanges may be necessary. 2952 The Server then returns an RA message with IPv6 source and 2953 destination set corresponding to the addresses in the RS, and with an 2954 Authentication option per [RFC4380]. For IPv4, the Server also 2955 includes an Origin option per [RFC4380] with the mapped and 2956 obfuscated Port Number and IPv4 address observed in the encapsulation 2957 headers. For IPv6, the Server instead includes an IPv6 Origin option 2958 per Figure 6 with the mapped and obfuscated observed Port Number and 2959 IPv6 address (note that the value 0x02 in the second octet 2960 differentiates from other [RFC4380] option types). 2962 +--------+--------+-----------------+ 2963 | 0x00 | 0x02 | Origin port # | 2964 +--------+--------+-----------------+ 2965 ~ Origin IPv6 address ~ 2966 +-----------------------------------+ 2968 Figure 6: IPv6 Origin Option 2970 When the Client receives the RA message, it compares the mapped Port 2971 Number and IP address from the Origin option with its own address. 2972 If the addresses are the same, the Client assumes the open Internet / 2973 Cone NAT principle; if the addresses are different, the Client 2974 instead assumes that further qualification procedures are necessary 2975 to detect the type of NAT and proceeds according to standard 2976 [RFC4380] procedures. 2978 After the Client has registered its INET interfaces in such RS/RA 2979 exchanges it sends periodic RS messages to receive fresh RA messages 2980 before the Router Lifetime received on each INET interface expires. 2981 The Client also maintains default routes via its Servers, i.e., the 2982 same as described in earlier sections. 2984 When the Client sends messages to target IP addresses, it also 2985 invokes route optimization per Section 3.14 using IPv6 ND address 2986 resolution messaging. The Client sends the NS(AR) message to the 2987 Server wrapped in a UDP/IP header with an Authentication option with 2988 the NS source address set to the Client's LLA and destination address 2989 set to the target solicited node multicast address. The Server 2990 authenticates the message and sends a corresponding NS(AR) message 2991 over the spanning tree the same as if it were the ROS, but with the 2992 OAL source address set to the Server's ULA and destination set to the 2993 ULA of the target. When the ROR receives the NS(AR), it adds the 2994 Server's ULA and Client's LLA to the target's Report List, and 2995 returns an NA with OMNI option information for the target. The 2996 Server then returns a UDP/IP encapsulated NA message with an 2997 Authentication option to the Client. 2999 Following route optimization for targets in the same OMNI link 3000 segment, if the target's L2ADDR is on the open INET, the Client 3001 forwards data packets directly to the target INET address. If the 3002 target is behind a NAT, the Client first establishes NAT state for 3003 the L2ADDR using the "bubble" mechanisms specified in 3004 [RFC6081][RFC4380]. The Client continues to send data packets via 3005 its Server until NAT state is populated, then begins forwarding 3006 packets via the direct path through the NAT to the target. For 3007 targets in different OMNI link segments, the Client uses OAL/ORH 3008 encapsulation and forwards data packets to the Bridge that returned 3009 the NA message. 3011 The ROR may return uNAs via the Server if the target moves, and the 3012 Server will send corresponding Authentication-protected uNAs to the 3013 Client. The Client can also send "loopback" NS(NUD) messages to test 3014 forward path reachability even though there is no security 3015 association between the Client and the target. 3017 The Client sends UDP/IP encapsulated IPv6 packets no larger than 1280 3018 bytes in one piece. In order to accommodate larger IPv6 packets (up 3019 to the OMNI interface MTU), the Client inserts an OAL header with 3020 source set to its own ULA and destination set to the ULA of the 3021 target and uses IPv6 fragmentation according to Section 3.9. The 3022 Client then encapsulates each fragment in a UDP/IP header and sends 3023 the fragments to the next hop. 3025 3.23. Time-Varying MNPs 3027 In some use cases, it is desirable, beneficial and efficient for the 3028 Client to receive a constant MNP that travels with the Client 3029 wherever it moves. For example, this would allow air traffic 3030 controllers to easily track aircraft, etc. In other cases, however 3031 (e.g., intelligent transportation systems), the MN may be willing to 3032 sacrifice a modicum of efficiency in order to have time-varying MNPs 3033 that can be changed every so often to defeat adversarial tracking. 3035 The DHCPv6 service offers a way for Clients that desire time-varying 3036 MNPs to obtain short-lived prefixes (e.g., on the order of a small 3037 number of minutes). In that case, the identity of the Client would 3038 not be bound to the MNP but rather the Client's identity would be 3039 bound to the DHCPv6 Device Unique Identifier (DUID) and used as the 3040 seed for Prefix Delegation. The Client would then be obligated to 3041 renumber its internal networks whenever its MNP (and therefore also 3042 its LLA) changes. This should not present a challenge for Clients 3043 with automated network renumbering services, however presents limits 3044 for the durations of ongoing sessions that would prefer to use a 3045 constant address. 3047 4. Implementation Status 3049 An early AERO implementation based on OpenVPN (https://openvpn.net/) 3050 was announced on the v6ops mailing list on January 10, 2018 and an 3051 initial public release of the AERO proof-of-concept source code was 3052 announced on the intarea mailing list on August 21, 2015. 3054 AERO Release-3.0.2 was tagged on October 15, 2020, and is undergoing 3055 internal testing. Additional releases expected Q42020, with first 3056 public release expected before year-end. 3058 5. IANA Considerations 3060 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 3061 AERO in the "enterprise-numbers" registry. 3063 The IANA has assigned the UDP port number "8060" for an earlier 3064 experimental version of AERO [RFC6706]. This document obsoletes 3065 [RFC6706] and claims the UDP port number "8060" for all future use. 3067 The IANA is instructed to assign a new type value TBD in the IPv6 3068 Routing Types registry. 3070 No further IANA actions are required. 3072 6. Security Considerations 3074 AERO Bridges configure secured tunnels with AERO Servers, Relays and 3075 Proxys within their local OMNI link segments. Applicable secured 3076 tunnel alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 3077 [RFC6347], WireGuard [WG], etc. The AERO Bridges of all OMNI link 3078 segments in turn configure secured tunnels for their neighboring AERO 3079 Bridges in a spanning tree topology. Therefore, control messages 3080 exchanged between any pair of OMNI link neighbors on the spanning 3081 tree are already secured. 3083 AERO Servers, Relays and Proxys targeted by a route optimization may 3084 also receive data packets directly from arbitrary nodes in INET 3085 partitions instead of via the spanning tree. For INET partitions 3086 that apply effective ingress filtering to defeat source address 3087 spoofing, the simple data origin authentication procedures in 3088 Section 3.8 can be applied. 3090 For INET partitions that require strong security in the data plane, 3091 two options for securing communications include 1) disable route 3092 optimization so that all traffic is conveyed over secured tunnels, or 3093 2) enable on-demand secure tunnel creation between INET partition 3094 neighbors. Option 1) would result in longer routes than necessary 3095 and traffic concentration on critical infrastructure elements. 3096 Option 2) could be coordinated by establishing a secured tunnel on- 3097 demand instead of performing an NS/NA exchange in the route 3098 optimization procedures. Procedures for establishing on-demand 3099 secured tunnels are out of scope. 3101 AERO Clients that connect to secured ANETs need not apply security to 3102 their ND messages, since the messages will be intercepted by a 3103 perimeter Proxy that applies security on its INET-facing interface as 3104 part of the spanning tree (see above). AERO Clients connected to the 3105 open INET can use symmetric network and/or transport layer security 3106 services such as VPNs or can by some other means establish a direct 3107 link. When a VPN or direct link may be impractical, however, an 3108 asymmetric security service such as the Authentication option 3109 specified in [RFC4380] should be applied. The Authentication option 3110 requires a unique Client identifier, which can be obtained per the 3111 Universally Unique IDentifier (UUID) [RFC4122] service and also used 3112 as a DHCP Unique Identifier (DUID) per [RFC6355]. 3114 Application endpoints SHOULD use application-layer security services 3115 such as TLS/SSL, DTLS or SSH [RFC4251] to assure the same level of 3116 protection as for critical secured Internet services. AERO Clients 3117 that require host-based VPN services SHOULD use symmetric network 3118 and/or transport layer security services such as IPsec, TLS/SSL, 3119 DTLS, etc. AERO Proxys and Servers can also provide a network-based 3120 VPN service on behalf of the Client, e.g., if the Client is located 3121 within a secured enclave and cannot establish a VPN on its own 3122 behalf. 3124 AERO Servers and Bridges present targets for traffic amplification 3125 Denial of Service (DoS) attacks. This concern is no different than 3126 for widely-deployed VPN security gateways in the Internet, where 3127 attackers could send spoofed packets to the gateways at high data 3128 rates. This can be mitigated by connecting Servers and Bridges over 3129 dedicated links with no connections to the Internet and/or when 3130 connections to the Internet are only permitted through well-managed 3131 firewalls. Traffic amplification DoS attacks can also target an AERO 3132 Client's low data rate links. This is a concern not only for Clients 3133 located on the open Internet but also for Clients in secured 3134 enclaves. AERO Servers and Proxys can institute rate limits that 3135 protect Clients from receiving packet floods that could DoS low data 3136 rate links. 3138 AERO Relays must implement ingress filtering to avoid a spoofing 3139 attack in which spurious messages with ULA addresses are injected 3140 into an OMNI link from an outside attacker. AERO Clients MUST ensure 3141 that their connectivity is not used by unauthorized nodes on their 3142 EUNs to gain access to a protected network, i.e., AERO Clients that 3143 act as routers MUST NOT provide routing services for unauthorized 3144 nodes. (This concern is no different than for ordinary hosts that 3145 receive an IP address delegation but then "share" the address with 3146 other nodes via some form of Internet connection sharing such as 3147 tethering.) 3149 The MAP list MUST be well-managed and secured from unauthorized 3150 tampering, even though the list contains only public information. 3151 The MAP list can be conveyed to the Client in a similar fashion as in 3152 [RFC5214] (e.g., through layer 2 data link login messaging, secure 3153 upload of a static file, DNS lookups, etc.). 3155 SRH authentication facilities are specified in [RFC8754]. 3157 Security considerations for accepting link-layer ICMP messages and 3158 reflected packets are discussed throughout the document. 3160 Security considerations for IPv6 fragmentation and reassembly are 3161 discussed in [I-D.templin-6man-omni-interface]. 3163 7. Acknowledgements 3165 Discussions in the IETF, aviation standards communities and private 3166 exchanges helped shape some of the concepts in this work. 3167 Individuals who contributed insights include Mikael Abrahamsson, Mark 3168 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 3169 Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian Farrel, Nick Green, 3170 Sri Gundavelli, Brian Haberman, Bernhard Haindl, Joel Halpern, Tom 3171 Herbert, Sascha Hlusiak, Lee Howard, Zdenek Jaron, Andre Kostur, 3172 Hubert Kuenig, Ted Lemon, Andy Malis, Satoru Matsushima, Tomek 3173 Mrugalski, Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal 3174 Skorepa, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, Lloyd 3175 Wood and James Woodyatt. Members of the IESG also provided valuable 3176 input during their review process that greatly improved the document. 3177 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 3178 for their shepherding guidance during the publication of the AERO 3179 first edition. 3181 This work has further been encouraged and supported by Boeing 3182 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 3183 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 3184 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 3185 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 3186 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 3187 Gene MacLean III, Rob Muszkiewicz, Sean O'Sullivan, Vijay 3188 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 3189 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 3190 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 3191 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 3192 Benson, Katie Tran and Eric Yeh are especially acknowledged for 3193 implementing the AERO functions as extensions to the public domain 3194 OpenVPN distribution. 3196 Earlier works on NBMA tunneling approaches are found in 3197 [RFC2529][RFC5214][RFC5569]. 3199 Many of the constructs presented in this second edition of AERO are 3200 based on the author's earlier works, including: 3202 o The Internet Routing Overlay Network (IRON) 3203 [RFC6179][I-D.templin-ironbis] 3205 o Virtual Enterprise Traversal (VET) 3206 [RFC5558][I-D.templin-intarea-vet] 3208 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 3209 [RFC5320][I-D.templin-intarea-seal] 3211 o AERO, First Edition [RFC6706] 3213 Note that these works cite numerous earlier efforts that are not also 3214 cited here due to space limitations. The authors of those earlier 3215 works are acknowledged for their insights. 3217 This work is aligned with the NASA Safe Autonomous Systems Operation 3218 (SASO) program under NASA contract number NNA16BD84C. 3220 This work is aligned with the FAA as per the SE2025 contract number 3221 DTFAWA-15-D-00030. 3223 This work is aligned with the Boeing Commercial Airplanes (BCA) 3224 Internet of Things (IoT) and autonomy programs. 3226 This work is aligned with the Boeing Information Technology (BIT) 3227 MobileNet program. 3229 8. References 3231 8.1. Normative References 3233 [I-D.templin-6man-omni-interface] 3234 Templin, F. and T. Whyman, "Transmission of IP Packets 3235 over Overlay Multilink Network (OMNI) Interfaces", draft- 3236 templin-6man-omni-interface-57 (work in progress), 3237 December 2020. 3239 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3240 DOI 10.17487/RFC0791, September 1981, 3241 . 3243 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 3244 RFC 792, DOI 10.17487/RFC0792, September 1981, 3245 . 3247 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3248 Requirement Levels", BCP 14, RFC 2119, 3249 DOI 10.17487/RFC2119, March 1997, 3250 . 3252 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3253 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3254 December 1998, . 3256 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3257 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3258 DOI 10.17487/RFC3971, March 2005, 3259 . 3261 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3262 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3263 . 3265 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3266 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3267 November 2005, . 3269 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3270 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3271 . 3273 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 3274 Network Address Translations (NATs)", RFC 4380, 3275 DOI 10.17487/RFC4380, February 2006, 3276 . 3278 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3279 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3280 DOI 10.17487/RFC4861, September 2007, 3281 . 3283 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3284 Address Autoconfiguration", RFC 4862, 3285 DOI 10.17487/RFC4862, September 2007, 3286 . 3288 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 3289 DOI 10.17487/RFC6081, January 2011, 3290 . 3292 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3293 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3294 May 2017, . 3296 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3297 (IPv6) Specification", STD 86, RFC 8200, 3298 DOI 10.17487/RFC8200, July 2017, 3299 . 3301 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3302 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3303 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3304 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3305 . 3307 8.2. Informative References 3309 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3310 2016. 3312 [I-D.bonica-6man-comp-rtg-hdr] 3313 Bonica, R., Kamite, Y., Alston, A., Henriques, D., and L. 3314 Jalil, "The IPv6 Compact Routing Header (CRH)", draft- 3315 bonica-6man-comp-rtg-hdr-23 (work in progress), October 3316 2020. 3318 [I-D.bonica-6man-crh-helper-opt] 3319 Li, X., Bao, C., Ruan, E., and R. Bonica, "Compressed 3320 Routing Header (CRH) Helper Option", draft-bonica-6man- 3321 crh-helper-opt-02 (work in progress), October 2020. 3323 [I-D.ietf-intarea-frag-fragile] 3324 Bonica, R., Baker, F., Huston, G., Hinden, R., Troan, O., 3325 and F. Gont, "IP Fragmentation Considered Fragile", draft- 3326 ietf-intarea-frag-fragile-17 (work in progress), September 3327 2019. 3329 [I-D.ietf-intarea-tunnels] 3330 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3331 Architecture", draft-ietf-intarea-tunnels-10 (work in 3332 progress), September 2019. 3334 [I-D.ietf-rtgwg-atn-bgp] 3335 Templin, F., Saccone, G., Dawra, G., Lindem, A., and V. 3336 Moreno, "A Simple BGP-based Mobile Routing System for the 3337 Aeronautical Telecommunications Network", draft-ietf- 3338 rtgwg-atn-bgp-06 (work in progress), June 2020. 3340 [I-D.templin-6man-dhcpv6-ndopt] 3341 Templin, F., "A Unified Stateful/Stateless Configuration 3342 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-10 3343 (work in progress), June 2020. 3345 [I-D.templin-intarea-seal] 3346 Templin, F., "The Subnetwork Encapsulation and Adaptation 3347 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 3348 progress), January 2014. 3350 [I-D.templin-intarea-vet] 3351 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 3352 templin-intarea-vet-40 (work in progress), May 2013. 3354 [I-D.templin-ironbis] 3355 Templin, F., "The Interior Routing Overlay Network 3356 (IRON)", draft-templin-ironbis-16 (work in progress), 3357 March 2014. 3359 [I-D.templin-v6ops-pdhost] 3360 Templin, F., "IPv6 Prefix Delegation and Multi-Addressing 3361 Models", draft-templin-v6ops-pdhost-26 (work in progress), 3362 June 2020. 3364 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3366 [RFC1035] Mockapetris, P., "Domain names - implementation and 3367 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3368 November 1987, . 3370 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3371 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3372 . 3374 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3375 DOI 10.17487/RFC2003, October 1996, 3376 . 3378 [RFC2004] Perkins, C., "Minimal Encapsulation within IP", RFC 2004, 3379 DOI 10.17487/RFC2004, October 1996, 3380 . 3382 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3383 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3384 . 3386 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 3387 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 3388 . 3390 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3391 Domains without Explicit Tunnels", RFC 2529, 3392 DOI 10.17487/RFC2529, March 1999, 3393 . 3395 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3396 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3397 . 3399 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3400 of Explicit Congestion Notification (ECN) to IP", 3401 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3402 . 3404 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 3405 DOI 10.17487/RFC3330, September 2002, 3406 . 3408 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3409 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3410 DOI 10.17487/RFC3810, June 2004, 3411 . 3413 [RFC4122] Leach, P., Mealling, M., and R. Salz, "A Universally 3414 Unique IDentifier (UUID) URN Namespace", RFC 4122, 3415 DOI 10.17487/RFC4122, July 2005, 3416 . 3418 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3419 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3420 January 2006, . 3422 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3423 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3424 DOI 10.17487/RFC4271, January 2006, 3425 . 3427 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3428 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3429 2006, . 3431 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3432 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3433 December 2005, . 3435 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3436 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3437 2006, . 3439 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3440 Control Message Protocol (ICMPv6) for the Internet 3441 Protocol Version 6 (IPv6) Specification", STD 89, 3442 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3443 . 3445 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3446 Protocol (LDAP): The Protocol", RFC 4511, 3447 DOI 10.17487/RFC4511, June 2006, 3448 . 3450 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3451 "Considerations for Internet Group Management Protocol 3452 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3453 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3454 . 3456 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3457 "Internet Group Management Protocol (IGMP) / Multicast 3458 Listener Discovery (MLD)-Based Multicast Forwarding 3459 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3460 August 2006, . 3462 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3463 Algorithms in Cryptographically Generated Addresses 3464 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3465 . 3467 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3468 "Bidirectional Protocol Independent Multicast (BIDIR- 3469 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3470 . 3472 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3473 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3474 DOI 10.17487/RFC5214, March 2008, 3475 . 3477 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3478 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3479 February 2010, . 3481 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3482 Route Optimization Requirements for Operational Use in 3483 Aeronautics and Space Exploration Mobile Networks", 3484 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3485 . 3487 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3488 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3489 . 3491 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3492 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3493 January 2010, . 3495 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 3496 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 3497 . 3499 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3500 "IPv6 Router Advertisement Options for DNS Configuration", 3501 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3502 . 3504 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3505 NAT64: Network Address and Protocol Translation from IPv6 3506 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3507 April 2011, . 3509 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3510 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3511 . 3513 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3514 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3515 DOI 10.17487/RFC6221, May 2011, 3516 . 3518 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3519 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3520 DOI 10.17487/RFC6273, June 2011, 3521 . 3523 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3524 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3525 January 2012, . 3527 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 3528 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, 3529 DOI 10.17487/RFC6355, August 2011, 3530 . 3532 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3533 for Equal Cost Multipath Routing and Link Aggregation in 3534 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3535 . 3537 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3538 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3539 . 3541 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 3542 UDP Checksums for Tunneled Packets", RFC 6935, 3543 DOI 10.17487/RFC6935, April 2013, 3544 . 3546 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 3547 for the Use of IPv6 UDP Datagrams with Zero Checksums", 3548 RFC 6936, DOI 10.17487/RFC6936, April 2013, 3549 . 3551 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3552 Korhonen, "Requirements for Distributed Mobility 3553 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3554 . 3556 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3557 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3558 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3559 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3560 2016, . 3562 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3563 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3564 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3565 July 2018, . 3567 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3568 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3569 . 3571 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 3572 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3573 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 3574 . 3576 [WG] Wireguard, "Wireguard, https://www.wireguard.com", August 3577 2020. 3579 Appendix A. Non-Normative Considerations 3581 AERO can be applied to a multitude of Internetworking scenarios, with 3582 each having its own adaptations. The following considerations are 3583 provided as non-normative guidance: 3585 A.1. Implementation Strategies for Route Optimization 3587 Route optimization as discussed in Section 3.14 results in the route 3588 optimization source (ROS) creating an asymmetric neighbor cache entry 3589 for the target neighbor. The neighbor cache entry is maintained for 3590 at most ReachableTime seconds and then deleted unless updated. In 3591 order to refresh the neighbor cache entry lifetime before the 3592 ReachableTime timer expires, the specification requires 3593 implementations to issue a new NS/NA exchange to reset ReachableTime 3594 while data packets are still flowing. However, the decision of when 3595 to initiate a new NS/NA exchange and to perpetuate the process is 3596 left as an implementation detail. 3598 One possible strategy may be to monitor the neighbor cache entry 3599 watching for data packets for (ReachableTime - 5) seconds. If any 3600 data packets have been sent to the neighbor within this timeframe, 3601 then send an NS to receive a new NA. If no data packets have been 3602 sent, wait for 5 additional seconds and send an immediate NS if any 3603 data packets are sent within this "expiration pending" 5 second 3604 window. If no additional data packets are sent within the 5 second 3605 window, delete the neighbor cache entry. 3607 The monitoring of the neighbor data packet traffic therefore becomes 3608 an asymmetric ongoing process during the neighbor cache entry 3609 lifetime. If the neighbor cache entry expires, future data packets 3610 will trigger a new NS/NA exchange while the packets themselves are 3611 delivered over a longer path until route optimization state is re- 3612 established. 3614 A.2. Implicit Mobility Management 3616 OMNI interface neighbors MAY provide a configuration option that 3617 allows them to perform implicit mobility management in which no ND 3618 messaging is used. In that case, the Client only transmits packets 3619 over a single interface at a time, and the neighbor always observes 3620 packets arriving from the Client from the same link-layer source 3621 address. 3623 If the Client's underlying interface address changes (either due to a 3624 readdressing of the original interface or switching to a new 3625 interface) the neighbor immediately updates the neighbor cache entry 3626 for the Client and begins accepting and sending packets according to 3627 the Client's new address. This implicit mobility method applies to 3628 use cases such as cellphones with both WiFi and Cellular interfaces 3629 where only one of the interfaces is active at a given time, and the 3630 Client automatically switches over to the backup interface if the 3631 primary interface fails. 3633 A.3. Direct Underlying Interfaces 3635 When a Client's OMNI interface is configured over a Direct interface, 3636 the neighbor at the other end of the Direct link can receive packets 3637 without any encapsulation. In that case, the Client sends packets 3638 over the Direct link according to QoS preferences. If the Direct 3639 interface has the highest QoS preference, then the Client's IP 3640 packets are transmitted directly to the peer without going through an 3641 ANET/INET. If other interfaces have higher QoS preferences, then the 3642 Client's IP packets are transmitted via a different interface, which 3643 may result in the inclusion of Proxys, Servers and Bridges in the 3644 communications path. Direct interfaces must be tested periodically 3645 for reachability, e.g., via NUD. 3647 A.4. AERO Critical Infrastructure Considerations 3649 AERO Bridges can be either Commercial off-the Shelf (COTS) standard 3650 IP routers or virtual machines in the cloud. Bridges must be 3651 provisioned, supported and managed by the INET administrative 3652 authority, and connected to the Bridges of other INETs via inter- 3653 domain peerings. Cost for purchasing, configuring and managing 3654 Bridges is nominal even for very large OMNI links. 3656 AERO Servers can be standard dedicated server platforms, but most 3657 often will be deployed as virtual machines in the cloud. The only 3658 requirements for Servers are that they can run the AERO user-level 3659 code and have at least one network interface connection to the INET. 3660 As with Bridges, Servers must be provisioned, supported and managed 3661 by the INET administrative authority. Cost for purchasing, 3662 configuring and managing Servers is nominal especially for virtual 3663 Servers hosted in the cloud. 3665 AERO Proxys are most often standard dedicated server platforms with 3666 one network interface connected to the ANET and a second interface 3667 connected to an INET. As with Servers, the only requirements are 3668 that they can run the AERO user-level code and have at least one 3669 interface connection to the INET. Proxys must be provisioned, 3670 supported and managed by the ANET administrative authority. Cost for 3671 purchasing, configuring and managing Proxys is nominal, and borne by 3672 the ANET administrative authority. 3674 AERO Relays can be any dedicated server or COTS router platform 3675 connected to INETs and/or EUNs. The Relay connects to the OMNI link 3676 and engages in eBGP peering with one or more Bridges as a stub AS. 3677 The Relay then injects its MNPs and/or non-MNP prefixes into the BGP 3678 routing system, and provisions the prefixes to its downstream- 3679 attached networks. The Relay can perform ROS/ROR services the same 3680 as for any Server, and can route between the MNP and non-MNP address 3681 spaces. 3683 A.5. AERO Server Failure Implications 3685 AERO Servers may appear as a single point of failure in the 3686 architecture, but such is not the case since all Servers on the link 3687 provide identical services and loss of a Server does not imply 3688 immediate and/or comprehensive communication failures. Although 3689 Clients typically associate with a single Server at a time, Server 3690 failure is quickly detected and conveyed by Bidirectional Forward 3691 Detection (BFD) and/or proactive NUD allowing Clients to migrate to 3692 new Servers. 3694 If a Server fails, ongoing packet forwarding to Clients will continue 3695 by virtue of the asymmetric neighbor cache entries that have already 3696 been established in route optimization sources (ROSs). If a Client 3697 also experiences mobility events at roughly the same time the Server 3698 fails, unsolicited NA messages may be lost but proxy neighbor cache 3699 entries in the DEPARTED state will ensure that packet forwarding to 3700 the Client's new locations will continue for up to DepartTime 3701 seconds. 3703 If a Client is left without a Server for an extended timeframe (e.g., 3704 greater than ReachableTime seconds) then existing asymmetric neighbor 3705 cache entries will eventually expire and both ongoing and new 3706 communications will fail. The original source will continue to 3707 retransmit until the Client has established a new Server 3708 relationship, after which time continuous communications will resume. 3710 Therefore, providing many Servers on the link with high availability 3711 profiles provides resilience against loss of individual Servers and 3712 assurance that Clients can establish new Server relationships quickly 3713 in event of a Server failure. 3715 A.6. AERO Client / Server Architecture 3717 The AERO architectural model is client / server in the control plane, 3718 with route optimization in the data plane. The same as for common 3719 Internet services, the AERO Client discovers the addresses of AERO 3720 Servers and selects one Server to connect to. The AERO service is 3721 analogous to common Internet services such as google.com, yahoo.com, 3722 cnn.com, etc. However, there is only one AERO service for the link 3723 and all Servers provide identical services. 3725 Common Internet services provide differing strategies for advertising 3726 server addresses to clients. The strategy is conveyed through the 3727 DNS resource records returned in response to name resolution queries. 3728 As of January 2020 Internet-based 'nslookup' services were used to 3729 determine the following: 3731 o When a client resolves the domainname "google.com", the DNS always 3732 returns one A record (i.e., an IPv4 address) and one AAAA record 3733 (i.e., an IPv6 address). The client receives the same addresses 3734 each time it resolves the domainname via the same DNS resolver, 3735 but may receive different addresses when it resolves the 3736 domainname via different DNS resolvers. But, in each case, 3737 exactly one A and one AAAA record are returned. 3739 o When a client resolves the domainname "ietf.org", the DNS always 3740 returns one A record and one AAAA record with the same addresses 3741 regardless of which DNS resolver is used. 3743 o When a client resolves the domainname "yahoo.com", the DNS always 3744 returns a list of 4 A records and 4 AAAA records. Each time the 3745 client resolves the domainname via the same DNS resolver, the same 3746 list of addresses are returned but in randomized order (i.e., 3747 consistent with a DNS round-robin strategy). But, interestingly, 3748 the same addresses are returned (albeit in randomized order) when 3749 the domainname is resolved via different DNS resolvers. 3751 o When a client resolves the domainname "amazon.com", the DNS always 3752 returns a list of 3 A records and no AAAA records. As with 3753 "yahoo.com", the same three A records are returned from any 3754 worldwide Internet connection point in randomized order. 3756 The above example strategies show differing approaches to Internet 3757 resilience and service distribution offered by major Internet 3758 services. The Google approach exposes only a single IPv4 and a 3759 single IPv6 address to clients. Clients can then select whichever IP 3760 protocol version offers the best response, but will always use the 3761 same IP address according to the current Internet connection point. 3762 This means that the IP address offered by the network must lead to a 3763 highly-available server and/or service distribution point. In other 3764 words, resilience is predicated on high availability within the 3765 network and with no client-initiated failovers expected (i.e., it is 3766 all-or-nothing from the client's perspective). However, Google does 3767 provide for worldwide distributed service distribution by virtue of 3768 the fact that each Internet connection point responds with a 3769 different IPv6 and IPv4 address. The IETF approach is like google 3770 (all-or-nothing from the client's perspective), but provides only a 3771 single IPv4 or IPv6 address on a worldwide basis. This means that 3772 the addresses must be made highly-available at the network level with 3773 no client failover possibility, and if there is any worldwide service 3774 distribution it would need to be conducted by a network element that 3775 is reached via the IP address acting as a service distribution point. 3777 In contrast to the Google and IETF philosophies, Yahoo and Amazon 3778 both provide clients with a (short) list of IP addresses with Yahoo 3779 providing both IP protocol versions and Amazon as IPv4-only. The 3780 order of the list is randomized with each name service query 3781 response, with the effect of round-robin load balancing for service 3782 distribution. With a short list of addresses, there is still 3783 expectation that the network will implement high availability for 3784 each address but in case any single address fails the client can 3785 switch over to using a different address. The balance then becomes 3786 one of function in the network vs function in the end system. 3788 The same implications observed for common highly-available services 3789 in the Internet apply also to the AERO client/server architecture. 3790 When an AERO Client connects to one or more ANETs, it discovers one 3791 or more AERO Server addresses through the mechanisms discussed in 3792 earlier sections. Each Server address presumably leads to a fault- 3793 tolerant clustering arrangement such as supported by Linux-HA, 3794 Extended Virtual Synchrony or Paxos. Such an arrangement has 3795 precedence in common Internet service deployments in lightweight 3796 virtual machines without requiring expensive hardware deployment. 3797 Similarly, common Internet service deployments set service IP 3798 addresses on service distribution points that may relay requests to 3799 many different servers. 3801 For AERO, the expectation is that a combination of the Google/IETF 3802 and Yahoo/Amazon philosophies would be employed. The AERO Client 3803 connects to different ANET access points and can receive 1-2 Server 3804 LLAs at each point. It then selects one AERO Server address, and 3805 engages in RS/RA exchanges with the same Server from all ANET 3806 connections. The Client remains with this Server unless or until the 3807 Server fails, in which case it can switch over to an alternate 3808 Server. The Client can likewise switch over to a different Server at 3809 any time if there is some reason for it to do so. So, the AERO 3810 expectation is for a balance of function in the network and end 3811 system, with fault tolerance and resilience at both levels. 3813 Appendix B. Change Log 3815 << RFC Editor - remove prior to publication >> 3817 Changes from draft-templin-intarea-6706bis-61 to draft-templin- 3818 intrea-6706bis-62: 3820 o New sub-section on OMNI Neighbor Interface Attributes 3822 Changes from draft-templin-intarea-6706bis-59 to draft-templin- 3823 intrea-6706bis-60: 3825 o Removed all references to S/TLLAO - all Interface Attributes are 3826 now maintained completely in the OMNI option. 3828 Changes from draft-templin-intarea-6706bis-58 to draft-templin- 3829 intrea-6706bis-59: 3831 o The term "Relay" used in older draft versions is now "Bridge". 3832 "Relay" now refers to what was formally called: "Gateway". 3834 o Fine-grained cleanup of Forwarding Algorithm; IPv6 ND message 3835 addressing; OMNI Prefix Lengths, etc. 3837 Changes from draft-templin-intarea-6706bis-54 to draft-templin- 3838 intrea-6706bis-55: 3840 o Updates on Segment Routing and S/TLLAO contents. 3842 o Various editorials and addressing cleanups. 3844 Changes from draft-templin-intarea-6706bis-52 to draft-templin- 3845 intrea-6706bis-53: 3847 o Normative reference to the OMNI spec, and remove portions that are 3848 already specified in OMNI. 3850 o Renamed "AERO interface/link" to "OMIN interface/link" throughout 3851 the document. 3853 o Truncated obsolete back section matter. 3855 Author's Address 3857 Fred L. Templin (editor) 3858 Boeing Research & Technology 3859 P.O. Box 3707 3860 Seattle, WA 98124 3861 USA 3863 Email: fltemplin@acm.org