idnits 2.17.1 draft-templin-intarea-6706bis-51.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 (May 20, 2020) is 1434 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) -- Looks like a reference, but probably isn't: '0' on line 918 -- Looks like a reference, but probably isn't: '1' on line 1049 -- Looks like a reference, but probably isn't: '2' on line 1055 == Missing Reference: 'N' is mentioned on line 1065, but not defined == Unused Reference: 'RFC2474' is defined on line 3327, but no explicit reference was found in the text == Unused Reference: 'RFC5175' is defined on line 3365, but no explicit reference was found in the text == Unused Reference: 'BGP' is defined on line 3391, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-6man-segment-routing-header' is defined on line 3394, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-gue' is defined on line 3406, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-gue-extensions' is defined on line 3411, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-tunnels' is defined on line 3416, but no explicit reference was found in the text == Unused Reference: 'RFC1191' is defined on line 3466, but no explicit reference was found in the text == Unused Reference: 'RFC2764' is defined on line 3491, but no explicit reference was found in the text == Unused Reference: 'RFC2923' is defined on line 3505, but no explicit reference was found in the text == Unused Reference: 'RFC3819' is defined on line 3527, but no explicit reference was found in the text == Unused Reference: 'RFC4607' is defined on line 3582, but no explicit reference was found in the text == Unused Reference: 'RFC4963' is defined on line 3586, but no explicit reference was found in the text == Unused Reference: 'RFC5991' is defined on line 3632, but no explicit reference was found in the text == Unused Reference: 'RFC6864' is defined on line 3673, but no explicit reference was found in the text == Unused Reference: 'RFC7269' is defined on line 3687, but no explicit reference was found in the text == Unused Reference: 'RFC7421' is defined on line 3697, but no explicit reference was found in the text == Unused Reference: 'RFC8086' is defined on line 3709, but no explicit reference was found in the text == Outdated reference: A later version (-99) exists of draft-templin-6man-omni-interface-21 == Outdated reference: A later version (-13) exists of draft-ietf-intarea-tunnels-10 == Outdated reference: A later version (-26) exists of draft-ietf-rtgwg-atn-bgp-05 == Outdated reference: A later version (-11) exists of draft-templin-6man-dhcpv6-ndopt-09 == Outdated reference: A later version (-27) exists of draft-templin-v6ops-pdhost-25 -- Obsolete informational reference (is this intentional?): RFC 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 (~~), 26 warnings (==), 7 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group F. Templin, Ed. 3 Internet-Draft Boeing Research & Technology 4 Obsoletes: rfc5320, rfc5558, rfc5720, May 20, 2020 5 rfc6179, rfc6706 (if 6 approved) 7 Intended status: Standards Track 8 Expires: November 21, 2020 10 Asymmetric Extended Route Optimization (AERO) 11 draft-templin-intarea-6706bis-51 13 Abstract 15 This document specifies the operation of IP over tunnel virtual links 16 using Asymmetric Extended Route Optimization (AERO). AERO interfaces 17 use an IPv6 link-local address format that supports operation of the 18 IPv6 Neighbor Discovery (ND) protocol and links ND to IP forwarding. 19 Prefix delegation/registration services are employed for network 20 admission and to manage the routing system. Multilink operation, 21 mobility management, quality of service (QoS) signaling and route 22 optimization are naturally supported through dynamic neighbor cache 23 updates. Standard IP multicasting services are also supported. AERO 24 is a widely-applicable mobile internetworking service especially 25 well-suited to aviation services, intelligent transportation systems, 26 mobile Virtual Private Networks (VPNs) and many other applications. 28 Status of This Memo 30 This Internet-Draft is submitted in full conformance with the 31 provisions of BCP 78 and BCP 79. 33 Internet-Drafts are working documents of the Internet Engineering 34 Task Force (IETF). Note that other groups may also distribute 35 working documents as Internet-Drafts. The list of current Internet- 36 Drafts is at https://datatracker.ietf.org/drafts/current/. 38 Internet-Drafts are draft documents valid for a maximum of six months 39 and may be updated, replaced, or obsoleted by other documents at any 40 time. It is inappropriate to use Internet-Drafts as reference 41 material or to cite them other than as "work in progress." 43 This Internet-Draft will expire on November 21, 2020. 45 Copyright Notice 47 Copyright (c) 2020 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents 52 (https://trustee.ietf.org/license-info) in effect on the date of 53 publication of this document. Please review these documents 54 carefully, as they describe your rights and restrictions with respect 55 to this document. Code Components extracted from this document must 56 include Simplified BSD License text as described in Section 4.e of 57 the Trust Legal Provisions and are provided without warranty as 58 described in the Simplified BSD License. 60 Table of Contents 62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 63 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 64 3. Asymmetric Extended Route Optimization (AERO) . . . . . . . . 10 65 3.1. AERO Node Types . . . . . . . . . . . . . . . . . . . . . 10 66 3.2. The AERO Link . . . . . . . . . . . . . . . . . . . . . . 11 67 3.2.1. AERO Link Reference Model . . . . . . . . . . . . . . 11 68 3.2.2. AERO Link-Local Addresses (LLAs) . . . . . . . . . . 14 69 3.2.3. AERO Unique Local Addresses (ULAs) . . . . . . . . . 15 70 3.2.4. AERO Routing System . . . . . . . . . . . . . . . . . 16 71 3.2.5. AERO Link Encapsulation . . . . . . . . . . . . . . . 17 72 3.2.6. Segment Routing Topologies (SRTs) . . . . . . . . . . 19 73 3.2.7. Segment Routing To the AERO Link . . . . . . . . . . 19 74 3.2.8. Segment Routing Within the AERO Link . . . . . . . . 20 75 3.3. AERO Interface Characteristics . . . . . . . . . . . . . 22 76 3.4. AERO Interface Initialization . . . . . . . . . . . . . . 27 77 3.4.1. AERO Server/Relay Behavior . . . . . . . . . . . . . 27 78 3.4.2. AERO Proxy Behavior . . . . . . . . . . . . . . . . . 27 79 3.4.3. AERO Client Behavior . . . . . . . . . . . . . . . . 27 80 3.4.4. AERO Bridge Behavior . . . . . . . . . . . . . . . . 28 81 3.5. AERO Interface Neighbor Cache Maintenance . . . . . . . . 28 82 3.6. AERO Interface Encapsulation and Re-encapsulation . . . . 30 83 3.7. AERO Interface Decapsulation . . . . . . . . . . . . . . 31 84 3.8. AERO Interface Data Origin Authentication . . . . . . . . 31 85 3.9. AERO Interface MTU and Fragmentation . . . . . . . . . . 32 86 3.10. AERO Interface Forwarding Algorithm . . . . . . . . . . . 34 87 3.10.1. Client Forwarding Algorithm . . . . . . . . . . . . 34 88 3.10.2. Proxy Forwarding Algorithm . . . . . . . . . . . . . 35 89 3.10.3. Server/Relay Forwarding Algorithm . . . . . . . . . 36 90 3.10.4. Bridge Forwarding Algorithm . . . . . . . . . . . . 37 91 3.11. AERO Interface Error Handling . . . . . . . . . . . . . . 38 92 3.12. AERO Router Discovery, Prefix Delegation and 93 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 40 94 3.12.1. AERO ND/PD Service Model . . . . . . . . . . . . . . 40 95 3.12.2. AERO Client Behavior . . . . . . . . . . . . . . . . 41 96 3.12.3. AERO Server Behavior . . . . . . . . . . . . . . . . 43 97 3.13. The AERO Proxy . . . . . . . . . . . . . . . . . . . . . 46 98 3.13.1. Detecting and Responding to Server Failures . . . . 48 99 3.13.2. Point-to-Multipoint Server Coordination . . . . . . 49 100 3.14. AERO Route Optimization . . . . . . . . . . . . . . . . . 49 101 3.14.1. Route Optimization Initiation . . . . . . . . . . . 50 102 3.14.2. Relaying the NS . . . . . . . . . . . . . . . . . . 50 103 3.14.3. Processing the NS and Sending the NA . . . . . . . . 50 104 3.14.4. Relaying the NA . . . . . . . . . . . . . . . . . . 52 105 3.14.5. Processing the NA . . . . . . . . . . . . . . . . . 52 106 3.14.6. Route Optimization Maintenance . . . . . . . . . . . 52 107 3.15. Neighbor Unreachability Detection (NUD) . . . . . . . . . 53 108 3.16. Mobility Management and Quality of Service (QoS) . . . . 54 109 3.16.1. Mobility Update Messaging . . . . . . . . . . . . . 55 110 3.16.2. Announcing Link-Layer Address and/or QoS Preference 111 Changes . . . . . . . . . . . . . . . . . . . . . . 56 112 3.16.3. Bringing New Links Into Service . . . . . . . . . . 56 113 3.16.4. Removing Existing Links from Service . . . . . . . . 56 114 3.16.5. Moving to a New Server . . . . . . . . . . . . . . . 57 115 3.17. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 58 116 3.17.1. Source-Specific Multicast (SSM) . . . . . . . . . . 58 117 3.17.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 59 118 3.17.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 60 119 3.18. Operation over Multiple AERO Links (VLANs) . . . . . . . 60 120 3.19. DNS Considerations . . . . . . . . . . . . . . . . . . . 61 121 3.20. Transition Considerations . . . . . . . . . . . . . . . . 62 122 3.21. Detecting and Reacting to Server and Bridge Failures . . 62 123 3.22. AERO Clients on the Open Internet . . . . . . . . . . . . 63 124 3.22.1. Use of SEND and CGA . . . . . . . . . . . . . . . . 65 125 3.23. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . 67 126 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 67 127 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 67 128 6. Security Considerations . . . . . . . . . . . . . . . . . . . 68 129 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 70 130 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 71 131 8.1. Normative References . . . . . . . . . . . . . . . . . . 71 132 8.2. Informative References . . . . . . . . . . . . . . . . . 73 133 Appendix A. AERO Alternate Encapsulations . . . . . . . . . . . 80 134 Appendix B. Non-Normative Considerations . . . . . . . . . . . . 82 135 B.1. Implementation Strategies for Route Optimization . . . . 83 136 B.2. Implicit Mobility Management . . . . . . . . . . . . . . 83 137 B.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 84 138 B.4. Operation on AERO Links with /64 ASPs . . . . . . . . . . 84 139 B.5. AERO Critical Infrastructure Considerations . . . . . . . 85 140 B.6. AERO Server Failure Implications . . . . . . . . . . . . 85 141 B.7. AERO Client / Server Architecture . . . . . . . . . . . . 86 142 Appendix C. Change Log . . . . . . . . . . . . . . . . . . . . . 88 143 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 97 145 1. Introduction 147 Asymmetric Extended Route Optimization (AERO) fulfills the 148 requirements of Distributed Mobility Management (DMM) [RFC7333] and 149 route optimization [RFC5522] for aeronautical networking and other 150 network mobility use cases such as intelligent transportation 151 systems. AERO is based on a Non-Broadcast, Multiple Access (NBMA) 152 virtual link model known as the AERO link. The AERO link is a 153 virtual overlay configured over one or more underlying Internetworks, 154 and nodes on the link can exchange IP packets via tunneling. 155 Multilink operation allows for increased reliability, bandwidth 156 optimization and traffic path diversity. 158 The AERO service comprises Clients, Proxys, Servers and Relays that 159 are seen as AERO link neighbors as well as Bridges that interconnect 160 AERO link segments. Each node's AERO interface uses an IPv6 link- 161 local address format that supports operation of the IPv6 Neighbor 162 Discovery (ND) protocol [RFC4861] and links ND to IP forwarding. A 163 node's AERO interface can be configured over multiple underlying 164 interfaces, and may therefore appear as a single interface with 165 multiple link-layer addresses. Each link-layer address is subject to 166 change due to mobility and/or QoS fluctuations, and link-layer 167 address changes are signaled by ND messaging the same as for any IPv6 168 link. 170 AERO links provide a cloud-based service where mobile nodes may use 171 any Server acting as a Mobility Anchor Point (MAP) and fixed nodes 172 may use any Relay on the link for efficient communications. Fixed 173 nodes forward packets destined to other AERO nodes to the nearest 174 Relay, which forwards them through the cloud. A mobile node's 175 initial packets are forwarded through the Server, while direct 176 routing is supported through asymmetric extended route optimization 177 while data packets are flowing. Both unicast and multicast 178 communications are supported, and mobile nodes may efficiently move 179 between locations while maintaining continuous communications with 180 correspondents and without changing their IP Address. 182 AERO Bridges are interconnected in a secured private BGP overlay 183 routing instance using encapsulation to provide a hybrid routing/ 184 bridging service that joins the underlying Internetworks of multiple 185 disjoint administrative domains into a single unified AERO link. 186 Each AERO link instance is characterized by the set of Mobility 187 Service Prefixes (MSPs) common to all mobile nodes. The link extends 188 to the point where a Relay/Server is on the optimal route from any 189 correspondent node on the link, and provides a conduit between the 190 underlying Internetwork and the AERO link. To the underlying 191 Internetwork, the Relay/Server is the source of a route to the MSP, 192 and hence uplink traffic to the mobile node is naturally routed to 193 the nearest Relay/Server. 195 AERO assumes the use of PIM Sparse Mode in support of multicast 196 communication. In support of Source Specific Multicast (SSM) when a 197 Mobile Node is the source, AERO route optimization ensures that a 198 shortest-path multicast tree is established with provisions for 199 mobility and multilink operation. In all other multicast scenarios 200 there are no AERO dependencies. 202 AERO was designed for aeronautical networking for both manned and 203 unmanned aircraft, where the aircraft is treated as a mobile node 204 that can connect an Internet of Things (IoT). AERO is also 205 applicable to a wide variety of other use cases. For example, it can 206 be used to coordinate the Virtual Private Network (VPN) links of 207 mobile nodes (e.g., cellphones, tablets, laptop computers, etc.) that 208 connect into a home enterprise network via public access networks 209 using services such as OpenVPN [OVPN]. It can also be used to 210 facilitate vehicular and pedestrian communications services for 211 intelligent transportation systems. Other applicable use cases are 212 also in scope. 214 The following numbered sections present the AERO specification. The 215 appendices at the end of the document are non-normative. 217 2. Terminology 219 The terminology in the normative references applies; the following 220 terms are defined within the scope of this document: 222 IPv6 Neighbor Discovery (ND) 223 an IPv6 control message service for coordinating neighbor 224 relationships between nodes connected to a common link. AERO 225 interfaces use the ND service specified in [RFC4861]. 227 IPv6 Prefix Delegation (PD) 228 a networking service for delegating IPv6 prefixes to nodes on the 229 link. The nominal PD service is DHCPv6 [RFC8415], however 230 alternate services (e.g., based on ND messaging) are also in scope 231 [I-D.templin-v6ops-pdhost][I-D.templin-6man-dhcpv6-ndopt]. Most 232 notably, a minimal form of PD known as "prefix registration" can 233 be used if the Client knows its prefix in advance and can 234 represent it in the IPv6 source address of an ND message. 236 Access Network (ANET) 237 a node's first-hop data link service network (e.g., a radio access 238 network, cellular service provider network, corporate enterprise 239 network, etc.) that often provides link-layer security services 240 such as IEEE 802.1X and physical-layer security prevent 241 unauthorized access internally and with border network-layer 242 security services such as firewalls and proxies that prevent 243 unauthorized outside access. 245 ANET interface 246 a node's attachment to a link in an ANET. 248 Internetwork (INET) 249 a connected IP network topology with a coherent routing and 250 addressing plan and that provides a transit backbone service for 251 ANET end systems. INETs also provide an underlay service over 252 which the AERO virtual link is configured. Example INETs include 253 corporate enterprise networks, aviation networks, and the public 254 Internet itself. When there is no administrative boundary between 255 an ANET and the INET, the ANET and INET are one and the same. 257 INET Partition 258 frequently, INETs such as large corporate enterprise networks are 259 sub-divided internally into separate isolated partitions. Each 260 partition is fully connected internally but disconnected from 261 other partitions, and there is no requirement that separate 262 partitions maintain consistent Internet Protocol and/or addressing 263 plans. (Each INET partition is seen as a separate AERO link 264 segment as discussed below.) 266 INET interface 267 a node's attachment to a link in an INET. 269 INET address 270 an IP address assigned to a node's interface connection to an 271 INET. 273 INET encapsulation 274 the encapsulation of a packet in an outer header or headers that 275 can be routed within the scope of the local INET partition. 277 AERO link 278 a Non-Broadcast, Multiple Access (NBMA) virtual overlay over one 279 or more underlying INETs manifested by IPv6 encapsulation 280 [RFC2473]. The AERO link spans underlying INET segments joined by 281 virtual bridges in a spanning tree the same as a bridged campus 282 LAN. Nodes on the AERO link appear as single-hop neighbors even 283 though they may be separated by multiple underlying INET hops, and 284 can use Segment Routing [RFC8402] to cause packets to visit 285 selected waypoints on the link. 287 AERO interface 288 a node's attachment to an AERO link. Since the addresses assigned 289 to an AERO interface are managed for uniqueness, AERO interfaces 290 do not require Duplicate Address Detection (DAD) and therefore set 291 the administrative variable 'DupAddrDetectTransmits' to zero 292 [RFC4862]. 294 underlying interface 295 an ANET or INET interface over which an AERO interface is 296 configured. 298 AERO Link-Local Address (LLA) 299 a link local IPv6 address per [RFC4291] constructed as specified 300 in Section 3.2.2. 302 AERO Unique-Local Address (ULA) 303 a unique local IPv6 address per [RFC4193] constructed as specified 304 in Section 3.2.3. AERO ULAs are statelessly derived from AERO 305 LLAs, and vice-versa. 307 Mobility Service Prefix (MSP) 308 an IP prefix assigned to the AERO link and from which more- 309 specific Mobile Network Prefixes (MNPs) are derived. 311 Mobile Network Prefix (MNP) 312 an IP prefix allocated from an MSP and delegated to an AERO Client 313 or Relay. 315 AERO node 316 a node that is connected to an AERO link, or that provides 317 services to other nodes on an AERO link. 319 AERO Client ("Client") 320 an AERO node that connects over one or more underlying interfaces 321 and requests MNP PDs from AERO Servers. The Client assigns a 322 Client LLA to the AERO interface for use in ND exchanges with 323 other AERO nodes and forwards packets to correspondents according 324 to AERO interface neighbor cache state. 326 AERO Server ("Server") 327 an INET node that configures an AERO interface to provide default 328 forwarding and mobility/multilink services for AERO Clients. The 329 Server assigns an administratively-provisioned LLA to its AERO 330 interface to support the operation of the ND/PD services, and 331 advertises all of its associated MNPs via BGP peerings with 332 Bridges. 334 AERO Relay ("Relay") 335 an AERO Server that also provides forwarding services between 336 nodes reached via the AERO link and correspondents on other links. 337 AERO Relays are provisioned with MNPs (i.e., the same as for an 338 AERO Client) and run a dynamic routing protocol to discover any 339 non-MNP IP routes. In both cases, the Relay advertises the MSP(s) 340 to its downstream networks, and distributes all of its associated 341 MNPs and non-MNP IP routes via BGP peerings with Bridges (i.e., 342 the same as for an AERO Server). 344 AERO Bridge ("Bridge") 345 a node that provides hybrid routing/bridging services (as well as 346 a security trust anchor) for nodes on an AERO link. As a router, 347 the Bridge forwards packets using standard IP forwarding. As a 348 bridge, the Bridge forwards packets over the AERO link without 349 decrementing the IPv6 Hop Limit. AERO Bridges peer with Servers 350 and other Bridges to discover the full set of MNPs for the link as 351 well as any non-MNPs that are reachable via Relays. 353 AERO Proxy ("Proxy") 354 a node that provides proxying services between Clients in an ANET 355 and Servers in external INETs. The AERO Proxy is a conduit 356 between the ANET and external INETs in the same manner as for 357 common web proxies, and behaves in a similar fashion as for ND 358 proxies [RFC4389]. 360 ingress tunnel endpoint (ITE) 361 an AERO interface endpoint that injects encapsulated packets into 362 an AERO link. 364 egress tunnel endpoint (ETE) 365 an AERO interface endpoint that receives encapsulated packets from 366 an AERO link. 368 link-layer address 369 an IP address used as an encapsulation header source or 370 destination address from the perspective of the AERO interface. 371 When an upper layer protocol (e.g., UDP) is used as part of the 372 encapsulation, the port number is also considered as part of the 373 link-layer address. 375 network layer address 376 the source or destination address of an encapsulated IP packet 377 presented to the AERO interface. 379 end user network (EUN) 380 an internal virtual or external edge IP network that an AERO 381 Client or Relay connects to the rest of the network via the AERO 382 interface. The Client/Relay sees each EUN as a "downstream" 383 network, and sees the AERO interface as the point of attachment to 384 the "upstream" network. 386 Mobile Node (MN) 387 an AERO Client and all of its downstream-attached networks that 388 move together as a single unit, i.e., an end system that connects 389 an Internet of Things. 391 Mobile Router (MR) 392 a MN's on-board router that forwards packets between any 393 downstream-attached networks and the AERO link. 395 Route Optimization Source (ROS) 396 the AERO node nearest the source that initiates route 397 optimization. The ROS may be a Server or Proxy acting on behalf 398 of the source Client. 400 Route Optimization responder (ROR) 401 the AERO node nearest the target destination that responds to 402 route optimization requests. The ROR may be a Server acting on 403 behalf of a target MNP Client, or a Relay for a non-MNP 404 destination. 406 MAP List 407 a geographically and/or topologically referenced list of addresses 408 of all Servers within the same AERO link. There is a single MAP 409 list for the entire AERO link. 411 Distributed Mobility Management (DMM) 412 a BGP-based overlay routing service coordinated by Servers and 413 Bridges that tracks all Server-to-Client associations. 415 Mobility Service (MS) 416 the collective set of all Servers, Proxys, Bridges and Relays that 417 provide the AERO Service to Clients. 419 Mobility Service Endpoint MSE) 420 an individual Server, Proxy, Bridge or Relay in the Mobility 421 Service. 423 Throughout the document, the simple terms "Client", "Server", 424 "Bridge", "Proxy" and "Relay" refer to "AERO Client", "AERO Server", 425 "AERO Bridge", "AERO Proxy" and "AERO Relay", respectively. 427 Capitalization is used to distinguish these terms from other common 428 Internetworking uses in which they appear without capitalization. 430 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 431 the names of node variables, messages and protocol constants) is used 432 throughout this document. The terms "All-Routers multicast", "All- 433 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 434 anycast" are defined in [RFC4291] (with Link-Local scope assumed). 435 Also, the term "IP" is used to generically refer to either Internet 436 Protocol version, i.e., IPv4 [RFC0791] or IPv6 [RFC8200]. 438 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 439 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 440 "OPTIONAL" in this document are to be interpreted as described in BCP 441 14 [RFC2119][RFC8174] when, and only when, they appear in all 442 capitals, as shown here. 444 3. Asymmetric Extended Route Optimization (AERO) 446 The following sections specify the operation of IP over Asymmetric 447 Extended Route Optimization (AERO) links: 449 3.1. AERO Node Types 451 AERO Bridges provide hybrid routing/bridging services (as well as a 452 security trust anchor) for nodes on an AERO link. Bridges use 453 standard IPv6 routing to forward packets both within the same INET 454 partitions and between disjoint INET partitions based on a mid-layer 455 IPv6 encapsulation per [RFC2473]. The inner IP layer experiences a 456 virtual bridging service since the inner IP TTL/Hop Limit is not 457 decremented during forwarding. Each Bridge also peers with Servers 458 and other Bridges in a dynamic routing protocol instance to provide a 459 Distributed Mobility Management (DMM) service for the list of active 460 MNPs (see Section 3.2.4). Bridges present the AERO link as a set of 461 one or more Mobility Service Prefixes (MSPs) but as link-layer 462 devices need not connect directly to the AERO link themselves unless 463 an administrative interface is desired. Bridges configure secured 464 tunnels with Servers, Relays, Proxys and other Bridges; they further 465 maintain IP forwarding table entries for each Mobile Network Prefix 466 (MNP) and any other reachable non-MNP prefixes. 468 AERO Servers provide default forwarding and mobility/multilink 469 services for AERO Client Mobile Nodes (MNs). Each Server also peers 470 with Bridges in a dynamic routing protocol instance to advertise its 471 list of associated MNPs (see Section 3.2.4). Servers facilitate PD 472 exchanges with Clients, where each delegated prefix becomes an MNP 473 taken from an MSP. Servers forward packets between AERO interface 474 neighbors and track each Client's mobility profiles. 476 AERO Clients register their MNPs through PD exchanges with AERO 477 Servers over the AERO link, and distribute the MNPs to nodes on EUNs. 478 A Client may also be co-resident on the same physical or virtual 479 platform as a Server; in that case, the Client and Server behave as a 480 single functional unit. 482 AERO Proxys provide a conduit for ANET Clients to associate with 483 Servers in external INETs. Client and Servers exchange control plane 484 messages via the Proxy acting as a bridge between the ANET/INET 485 boundary. The Proxy forwards data packets between Clients and the 486 AERO link according to forwarding information in the neighbor cache. 487 The Proxy function is specified in Section 3.13. 489 AERO Relays are Servers that provide forwarding services between the 490 AERO interface and INET/EUN interfaces. Relays are provisioned with 491 MNPs the same as for an AERO Client, and also run a dynamic routing 492 protocol to discover any non-MNP IP routes. The Relay advertises the 493 MSP(s) to its connected networks, and distributes all of its 494 associated MNPs and non-MNP IP routes via BGP peerings with Bridges. 496 AERO Bridges, Servers, Proxys and Relays are critical infrastructure 497 elements in fixed (i.e., non-mobile) INET deployments and hence have 498 permanent and unchanging INET addresses. AERO Clients are MNs that 499 connect via underlying interfaces with addresses that may change when 500 the Client moves to a new network connection point. 502 3.2. The AERO Link 504 3.2.1. AERO Link Reference Model 506 Figure 1 presents the basic AERO link reference model: 508 +----------------+ 509 | AERO Bridge B1 | 510 | Nbr: S1, S2, P1| 511 |(X1->S1; X2->S2)| 512 | MSP M1 | 513 +-+---------+--+-+ 514 +--------------+ | Secured | | +--------------+ 515 |AERO Server S1| | tunnels | | |AERO Server S2| 516 | Nbr: C1, B1 +-----+ | +-----+ Nbr: C2, B1 | 517 | default->B1 | | | default->B1 | 518 | X1->C1 | | | X2->C2 | 519 +-------+------+ | +------+-------+ 520 | AERO Link | | 521 X===+===+===================+==)===============+===+===X 522 | | | | 523 +-----+--------+ +--------+--+-----+ +--------+-----+ 524 |AERO Client C1| | AERO Proxy P1 | |AERO Client C2| 525 | Nbr: S1 | |(Proxy Nbr Cache)| | Nbr: S2 | 526 | default->S1 | +--------+--------+ | default->S2 | 527 | MNP X1 | | | MNP X2 | 528 +------+-------+ .--------+------. +-----+--------+ 529 | (- Proxyed Clients -) | 530 .-. `---------------' .-. 531 ,-( _)-. ,-( _)-. 532 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 533 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 534 `-(______)-' +-------+ +-------+ `-(______)-' 536 Figure 1: AERO Link Reference Model 538 In this model: 540 o the AERO link is an overlay network service configured over one or 541 more underlying INET partitions which may be managed by different 542 administrative authorities and have incompatible protocols and/or 543 addressing plans. 545 o AERO Bridge B1 aggregates Mobility Service Prefix (MSP) M1, 546 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 547 via BGP peerings over secured tunnels to Servers (S1, S2). 548 Bridges connect the disjoint segments of a partitioned AERO link. 550 o AERO Servers/Relays S1 and S2 configure secured tunnels with 551 Bridge B1 and also provide mobility, multilink and default router 552 services for their associated Clients C1 and C2. 554 o AERO Clients C1 and C2 associate with Servers S1 and S2, 555 respectively. They receive Mobile Network Prefix (MNP) 556 delegations X1 and X2, and also act as default routers for their 557 associated physical or internal virtual EUNs. Simple hosts H1 and 558 H2 attach to the EUNs served by Clients C1 and C2, respectively. 560 o AERO Proxy P1 configures a secured tunnel with Bridge B1 and 561 provides proxy services for AERO Clients in secured enclaves that 562 cannot associate directly with other AERO link neighbors. 564 An AERO link configured over a single INET appears as a single 565 unified link with a consistent underlying network addressing plan. 566 In that case, all nodes on the link can exchange packets via simple 567 INET encapsulation, since the underlying INET is connected. In 568 common practice, however, an AERO link may be partitioned into 569 multiple "segments", where each segment is a distinct INET 570 potentially managed under a different administrative authority (e.g., 571 as for worldwide aviation service providers such as ARINC, SITA, 572 Inmarsat, etc.). Individual INETs may also themselves be partitioned 573 internally, in which case each internal partition is seen as a 574 separate segment. 576 The addressing plan of each segment is consistent internally but will 577 often bear no relation to the addressing plans of other segments. 578 Each segment is also likely to be separated from others by network 579 security devices (e.g., firewalls, proxies, packet filtering 580 gateways, etc.), and in many cases disjoint segments may not even 581 have any common physical link connections. Therefore, nodes can only 582 be assured of exchanging packets directly with correspondents in the 583 same segment, and not with those in other segments. The only means 584 for joining the segments therefore is through inter-domain peerings 585 between AERO Bridges. 587 The same as for traditional campus LANs, multiple AERO link segments 588 can be joined into a single unified link via a virtual bridging 589 service using a mid-layer IPv6 encpasulation per [RFC2473] known as 590 the "SPAN header" that supports inter-segment forwarding (i.e., 591 bridging) without decrementing the network-layer TTL/Hop Limit. This 592 bridging of AERO link segments is shown in Figure 2: 594 . . . . . . . . . . . . . . . . . . . . . . . 595 . . 596 . .-(::::::::) . 597 . .-(::::::::::::)-. +-+ . 598 . (:::: Segment A :::)--|B|---+ . 599 . `-(::::::::::::)-' +-+ | . 600 . `-(::::::)-' | . 601 . | . 602 . .-(::::::::) | . 603 . .-(::::::::::::)-. +-+ | . 604 . (:::: Segment B :::)--|B|---+ . 605 . `-(::::::::::::)-' +-+ | . 606 . `-(::::::)-' | . 607 . | . 608 . .-(::::::::) | . 609 . .-(::::::::::::)-. +-+ | . 610 . (:::: Segment C :::)--|B|---+ . 611 . `-(::::::::::::)-' +-+ | . 612 . `-(::::::)-' | . 613 . | . 614 . ..(etc).. x . 615 . . 616 . . 617 . <- AERO Link Bridged by encapsulation -> . 618 . . . . . . . . . . . . . .. . . . . . . . . 620 Figure 2: Bridging AERO Link Segments 622 Bridges, Servers, Relays and Proxys connect via secured INET tunnels 623 over their respecitve segments in a spanning tree topology rooted at 624 the Bridges. The secured spanning tree supports strong 625 authentication for IPv6 ND control messages and may also be used to 626 convey the initial data packets in a flow. Route optimization can 627 then be employed to cause data packets to take more direct paths 628 between AERO link neighbors without having to strictly follow the 629 spanning tree. 631 3.2.2. AERO Link-Local Addresses (LLAs) 633 Nodes on AERO links use the Link-Local Address (LLA) prefix fe80::/10 634 [RFC4193] to assign LLAs used for network-layer addresses in IPv6 ND 635 and data messages. A Client's LLA is an IPv6 link-local address 636 formed from the Client's delegated MNP. Bridge, Server, Relay and 637 Proxy LLAs are assigned from the range fe80::/96 and include an 638 administratively-provisioned value in the lower 32 bits. 640 IPv6 Client LLAs encode the Subnet-Router anycast address of a MNP 641 (or non-MNP globally routable prefix) within the least-significant 642 112 bits of the IPv6 link-local prefix fe80::/16. For example, for 643 the MNP 2001:db8:1000:2000::/56 the corresponding LLA is 644 fe80:2001:db8:1000:2000::/72. 646 IPv4-compatible Client LLAs are based on an IPv4-mapped IPv6 address 647 [RFC4291] formed from an IPv4 MNP and with a prefix length of 96 plus 648 the MNP prefix length. For example, for the IPv4 MNP 192.0.2.16/28 649 the IPv4-mapped IPv6 MNP is: 651 0:0:0:0:0:ffff:192.0.2.16/124 (also written as 652 0:0:0:0:0:ffff:c000:0210/124) 654 The Client then constructs its LLA with the prefix fe80::/64 and with 655 the lower 64 bits of the IPv4-mapped IPv6 address in the interface 656 identifier as: fe80::ffff:192.0.2.16. 658 Mobility Service (MS) LLAs (used by Bridges, Servers, Relays and 659 Proxys) are allocated from the range fe80::/96, and MUST be managed 660 for uniqueness. The lower 32 bits of the LLA includes a unique 661 integer value between 1 and 0xfeffffff (e.g., fe80::1, fe80::2, 662 fe80::3, etc., fe80::feff:ffff) as assigned by the administrative 663 authority for the link. The address fe80:: is the IPv6 link-local 664 Subnet-Router anycast address, and the address range 665 fe80::ff00:0000/104 is reserved for future use. 667 Finally, the address range fe80::/32 is used as the Teredo service 668 prefix for AERO according to the format in Section 4 of [RFC4380] 669 (see Section 3.22 for further discussion). 671 3.2.3. AERO Unique Local Addresses (ULAs) 673 Nodes on AERO links use the Unique Local Address (ULA) prefix 674 fd80::/10 [RFC4193] to form ULAs used for SPAN header source and 675 desitnation addresses. The prefix length intentionally matches the 676 IPv6 link-local prefix (fe80::/10), and enables a simple stateless 677 translation between LLAs and ULAs. 679 AERO ULAs are formed by simply rewriting the upper bits of the 680 corresponding LLA as follows: 682 o the ULA formed from the IPv6 Client LLA fe80:2001:db8:1000:2000:: 683 is simply fd80:2001:db8:1000:2000:: 685 o the ULA formed from the IPv4-compatible Client LLA 686 fe80::ffff:192.0.2.1 is simply fd80::ffff:192.0.2.1 688 o the ULA formed from the MS LLA fe80::1001 is simply fd80::1001 689 o the ULA prefix fd80::/32 is used as the Teredo service prefix the 690 same as for LLAs above. 692 3.2.4. AERO Routing System 694 The AERO routing system comprises a private instance of the Border 695 Gateway Protocol (BGP) [RFC4271] that is coordinated between Bridges 696 and Servers and does not interact with either the public Internet BGP 697 routing system or any underlying INET routing systems. 699 In a reference deployment, each Server is configured as an Autonomous 700 System Border Router (ASBR) for a stub Autonomous System (AS) using 701 an AS Number (ASN) that is unique within the BGP instance, and each 702 Server further uses eBGP to peer with one or more Bridges but does 703 not peer with other Servers. Each INET of a multi-segment AERO link 704 must include one or more Bridges, which peer with the Servers and 705 Proxys within that INET. All Bridges within the same INET are 706 members of the same hub AS using a common ASN, and use iBGP to 707 maintain a consistent view of all active MNPs currently in service. 708 The Bridges of different INETs peer with one another using eBGP. 710 Bridges advertise the AERO link's MSPs and any non-MNP routes to each 711 of their Servers. This means that any aggregated non-MNPs (including 712 "default") are advertised to all Servers. Each Bridge configures a 713 black-hole route for each of its MSPs. By black-holing the MSPs, the 714 Bridge will maintain forwarding table entries only for the MNPs that 715 are currently active, and packets destined to all other MNPs will 716 correctly incur Destination Unreachable messages due to the black- 717 hole route. In this way, Servers have only partial topology 718 knowledge (i.e., they know only about the MNPs of their directly 719 associated Clients) and they forward all other packets to Bridges 720 which have full topology knowledge. 722 Each AERO link segment assigns a unique sub-prefix of fd80::/96 known 723 as the ULA partition prefix. For example, a first segment could 724 assign fd80::1000/116, a second could assign fd80::2000/116, a third 725 could assign fd80::3000/116, etc. The administrative authorities for 726 each segment must therefore coordinate to assure mutually-exclusive 727 partiton prefix assignments, but internal provisioning of each prefix 728 is an independent local consideration for each administrative 729 authority. 731 ULA partition prefixes are statitcally represented in Bridge 732 forwarding tables. Bridges join multiple segments into a unified 733 AERO link over multiple diverse administrative domains. They support 734 a bridging function by first establishing forwarding table entries 735 for their partiion prefixes either via standard BGP routing or static 736 routes. For example, if three Bridges ('A', 'B' and 'C') from 737 different segments serviced fd80::1000/116, fd80::2000/116 and 738 fd80::3000/116 respectively, then the forwarding tables in each 739 Bridge are as follows: 741 A: fd80::1000/116->local, fd80::2000/116->B, fd80::3000/116->C 743 B: fd80::1000/116->A, fd80::2000/116->local, fd80::3000/116->C 745 C: fd80::1000/116->A, fd80::2000/116->B, fd80::3000/116->local 747 These forwarding table entries are permanent and never change, since 748 they correspond to fixed infrastructure elements in their respective 749 segments. 751 ULA Client prefixes are instead dynamically advertised in the AERO 752 link routing system by Servers and Relays that provide service for 753 their corresponding MNPs. For example, if three Servers ('D', 'E' 754 and 'F') service the MNPs 2001:db8:1000:2000::/56, 755 2001:db8:3000:4000::/56 and 2001:db8:5000:6000::/56 then the routing 756 system would include: 758 D: fd80:2001:db8:1000:2000::/72 760 E: fd80:2001:db8:3000:4000::/72 762 F: fd80:2001:db8:5000:6000::/72 764 A full discussion of the BGP-based routing system used by AERO is 765 found in [I-D.ietf-rtgwg-atn-bgp]. The system provides for 766 Distributed Mobility Management (DMM) per the distributed mobility 767 anchoring architecture [I-D.ietf-dmm-distributed-mobility-anchoring]. 769 3.2.5. AERO Link Encapsulation 771 With the Client and partition prefixes in place in each Bridge's 772 forwarding table, control and data packets sent between AERO nodes in 773 different segments can therefore be carried over the via mid-layer 774 encapsulation using the SPAN header. For example, when a source AERO 775 node forwards a packet with IPv6 address 2001:db8:1:2::1 to a target 776 AERO node with IPv6 address 2001:db8:1000:2000::1, it first 777 encapsulates the packet in a SPAN header with source address set to 778 fd80:2001:db8:1:2:: and destination address set to 779 fd80:2001:db8:1000:2000::. Next, it encapsulates the resulting SPAN 780 packet in an INET header with source address set to its own INET 781 address (e.g., 192.0.2.100) and destination set to the INET address 782 of a Bridge (e.g., 192.0.2.1). 784 SPAN encapsulation is based on Generic Packet Tunneling in IPv6 785 [RFC2473]; the encapsulation format in the above example is shown in 786 Figure 3: 788 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 789 | INET Header | 790 | src = 192.0.2.100 | 791 | dst = 192.0.2.1 | 792 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 793 | SPAN Header | 794 | src = fd80:2001:db8:1:2:: | 795 | dst=fd80:2001:db8:1000:2000:: | 796 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 797 | Inner IP Header | 798 | src = 2001:db8:1:2::1 | 799 | dst = 2001:db8:1000:2000::1 | 800 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 801 | | 802 ~ ~ 803 ~ Inner Packet Body ~ 804 ~ ~ 805 | | 806 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 808 Figure 3: SPAN Encapsulation 810 In this format, the inner IP header and packet body are the original 811 IP packet, the SPAN header is an IPv6 header prepared according to 812 [RFC2473], and the INET header is prepared as discussed in 813 Section 3.6. 815 This gives rise to a routing system that contains both Client prefix 816 routes that may change dynamically due to regional node mobility and 817 partion prefix routes that never change. The Bridges can therefore 818 provide link-layer bridging by sending packets over the SRT instead 819 of network-layer routing according to MNP routes. As a result, 820 opportunities for packet loss due to node mobility between different 821 segments are mitigated. 823 In normal operations, IPv6 ND messages are conveyed over secured 824 paths between AERO link neighbors so that specific Proxys, Servers or 825 Relays can be addressed without being subject to mobility events. 826 Conversely, only the first few packets destined to Clients need to 827 traverse secured paths until route optimization can determine a more 828 direct path. 830 3.2.6. Segment Routing Topologies (SRTs) 832 The 16-bit sub-prefixes of fd80::/10 (e.g., fd80::/16, fd81::/16, 833 fd82::/16, etc.) identify distinct Segment Routing Topologies (SRTs) 834 (see: Section 3.2.6). Each SRT is a mutually-exclusive AERO link 835 overlay instance using a mutually-exclusive set of ULAs, and emulates 836 a Virtual LAN (VLAN) service for the AERO link. In some cases (e.g., 837 when redundant topologies are needed for fault tolerance and 838 reliability) it may be beneficial to deploy multiple SRTs that act as 839 independent overlay instances. A communication failure in one 840 instance therefore will not affect communications in other instances. 842 Each SRT is identified by a distinct value in bits 10-15 of he SSP 843 fd80::10, i.e., as fd80::/16, fd81::/16, fd82::/16, etc. This 844 document asserts that up to four SRTs provide a level of safety 845 sufficient for critical communications such as civil aviation. Each 846 SRT is designated with a color that identifies a different AERO link 847 instance as follows: 849 o Red (default) - corresponds to the SSP fd80::/16 851 o Green - corresponds to the SSP fd81::/16 853 o Blue-1 - corresponds to the SSP fd82::/16 855 o Blue-2 - corresponds to SSP fd83::/16 857 o SSPs fd84::/16 through fdbf::/16 are reserved for future use. 859 Each AERO interface assigns an anycast ULA corresponding to its SRT 860 prefix. For example, the anycast ULA for the Green SRT is simply 861 fd81::. The anycast ULA is used for AERO interface determination in 862 Safety-Based Multilink (SBM) as discussed in 863 [I-D.templin-6man-omni-interface]. Each AERO interface further 864 applies Performance-Based Multilink (PBM) internally. 866 3.2.7. Segment Routing To the AERO Link 868 An original IPv6 source can direct a packet to a specific SRT ingress 869 router for the AERO link by including a Segment Routing Header (SRH) 870 with the anycast ULA for the selected SRT as either the IPv6 871 destination or as an intermediate segment ID within the SRH. This 872 allows the original source to determine the specific topology a 873 packet will traverse when there may be multiple alternatives to 874 choose from. This form of Segment Routing supports Safety-Based 875 Multilink (SBM), and can be exercised through general-purpose SRH 876 types such as [RFC8754]. 878 3.2.8. Segment Routing Within the AERO Link 880 AERO nodes that insert a SPAN header can use Segment Routing within 881 the AERO link to influence the path of packets destined to Clients on 882 INET interfaces without causing all packets to traverse the Server. 883 When a Client, Proxy or Server has a packet to send to a target 884 discovered through route optimization located in the same AERO link 885 segment, it encapsulates the packet in a SPAN header with the ULA of 886 the target as the destination address, then uses the target's Link 887 Layer Address information for INET encapsulation. 889 When a Client, Proxy or Server has a packet to send to a route 890 optimization target located in a different AERO link segment, it 891 encapsulates the packet in a SPAN header with the ULA of the target's 892 Server as the destination. The node also includes a SRH with the MNP 893 corresponding to the target ULA as the penultimate Segment ID (SID) 894 and with the IP encapsulation address of the target as the ultimate 895 SID. The node then forwards the packet via a secured tunnel to a 896 Bridge, which will eventually direct it to a Bridge on the same 897 segment as the target's Server. 899 When a Bridge on the same segment as the target's Server receives a 900 SPAN-encapsulated packet destined to the target Server, it looks 901 ahead into the SRH to determine that the penultimate SID is set to 902 the target's MNP and the ultimate SID is set to the target's Link 903 Layer Address. The Bridge then advances the SPAN destination address 904 to the target's ULA and encapsulates the SPAN packet in an INET 905 header based on the target's Link Layer Address, then forwards the 906 packet to the target directly while bypassing the target's Server. 907 In this way, the Bridge participates in route optimization to greatly 908 reduce traffic load and suboptimal routing through the target's 909 Server. 911 Segment Routing within the AERO link uses the AERO SRH format shown 912 in Figure 4 (note that a similar format is defined by [RFC8138] 913 0 1 2 3 914 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 1 915 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 916 | Next Header | Hdr Ext Len | Routing Type | Segments Left | 917 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 918 |Type[0]| Len[0]| SID[0] (1 to 16 bytes) ..... 919 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 920 |Type[1]| Len[1]| SID[1] (1 to 16 bytes) ..... 921 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 922 |Type[2] Len[2]| SID[2] (1 to 16 bytes) ..... 923 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 924 ... 925 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- -+-+-+-+-+-+-+-+-+ 926 |Type[N]| Len[N]| SID[N] (1 to 16 bytes) ..... zero padding | 927 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- -+-+-+-+-+-+-+-+-+ 929 Figure 4: IPv6 Compressed Routing Header with Variable Length 930 Addresses 932 In this format: 934 o Next Header, Hdr Ext Len and Segments Left have the same meaning 935 as in [RFC8754]. This document assigns a new Routing Type value 936 TBD. 938 o Immediately following are N Type/Length/Value Segment IDs as 939 follows: 941 * Type[i] indicates the type of SID that follows. The following 942 types are currently defined: 944 + 0000 - SID is a 4 byte public IPv4 address followed by a 2 945 byte port number 947 + 0001 - SID is a 4 byte NATed IPv4 address followed by a 2 948 byte port number 950 + 0010 - SID is a 16 byte public IPv6 address followed by a 2 951 byte port number 953 + 0011 - SID is a 16 byte NATed IPv6 address followed by a 2 954 byte port number 956 + 0100 - SID is a 2 byte ID corresponding to an IPv6 router 958 + 0101 - SID is a 4 byte ID corresponding to an IPv6 router 960 + 0110 - 0111 - Reserved 961 + 1000 - SID is a (Len+1) byte IPv4 MNP. Len SHOULD be sized 962 to accommodate the IPv4 prefix in the fewest possible bytes, 963 If the IPv4 prefix does not end on an even byte boundary, 964 the rightmost bits are set to 0. 966 + 1001 - SID is a (Len+1) byte IPv6 MNP. Len SHOULD be sized 967 to accommdate the IPv6 prefix in the fewest possible bytes. 968 If the IPv6 prefix does not end on an even byte boundary, 969 the rightmost bits are set to 0. 971 + 1010 - 1111 - Reserved 973 * Len[i] is a value that, when added to 1, indicates the length 974 of the SID that follows. Note that Len is exclusive of the 2 975 byte port number included for Types 0000 - 0011. For example, 976 for Type 0010 Len encodes the value 15, yet a 2 byte trailing 977 port number follows immediately after the SID. 979 * SID[i] is an ID of (Len +1) bytes in length formatted according 980 to Type. For Type values less than 0100, the SID is followed 981 by a 2 byte port number whose length is NOT included in Len. 983 * Trailing padding added if necessary to create an integral 984 number of 8 octet blocks. 986 3.3. AERO Interface Characteristics 988 AERO interfaces are virtual interfaces configured over one or more 989 underlying interfaces classified as follows: 991 o INET interfaces connect to an INET either natively or through one 992 or several IPv4 Network Address Translators (NATs). Native INET 993 interfaces have global IP addresses that are reachable from any 994 INET correspondent. All Server, Relay and Bridge interfaces are 995 native interfaces, as are INET-facing interfaces of Proxys. NATed 996 INET interfaces connect to a private network behind one or more 997 NATs that provide INET access. Clients that are behind a NAT are 998 required to send periodic keepalive messages to keep NAT state 999 alive when there are no data packets flowing. 1001 o Proxyed interfaces connect to an ANET that is separated from the 1002 open INET by an AERO Proxy. Proxys can actively issue control 1003 messages over the INET on behalf of the Client to reduce ANET 1004 congestion. Clients connected to Proxyed interfaces receive RAs 1005 with the P flag set to 1. 1007 o VPNed interfaces use security encapsulation over the INET to a 1008 Virtual Private Network (VPN) server that also acts as an AERO 1009 Server. Other than the link-layer encapsulation format, VPNed 1010 interfaces behave the same as Direct interfaces. 1012 o Direct interfaces connect a Client directly to a Server without 1013 crossing any ANET/INET paths. An example is a line-of-sight link 1014 between a remote pilot and an unmanned aircraft. The same Client 1015 considerations apply as for VPNed interfaces. 1017 AERO interfaces use SPAN encapsulation as necessary as discussed in 1018 Section 3.2.5. AERO interfaces use link-layer encapsulation (see: 1019 Section 3.6) to exchange packets with AERO link neighbors over INET 1020 or VPNed interfaces. AERO interfaces do not use link-layer 1021 encapsulation over Proxyed and Direct underlying interfaces. 1023 AERO interfaces maintain a neighbor cache for tracking per-neighbor 1024 state the same as for any interface. AERO interfaces use ND messages 1025 including Router Solicitation (RS), Router Advertisement (RA), 1026 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 1027 neighbor cache management. 1029 AERO interfaces send ND messages with an Overlay Multilink Network 1030 Interface (OMNI) option formatted as specified in 1031 [I-D.templin-6man-omni-interface]. The OMNI option includes prefix 1032 registration information and "ifIndex-tuples" containing link 1033 information parameters for the AERO interface's underlying 1034 interfaces. 1036 When encapsulation is used, AERO interface ND messages MAY also 1037 include an AERO Source/Target Link-Layer Address Option (S/TLLAO) 1038 formatted as shown in Figure 5: 1040 0 1 2 3 1041 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 1042 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1043 | Type | Length | ifIndex[1] | SRT | LHS |FMT| 1044 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1045 ~ Segment Routing List [1] ~ 1046 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1047 ~ Link Layer Address [1] ~ 1048 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1049 | Port Number [1] | ifIndex[2] | SRT | LHS |FMT| 1050 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1051 ~ Segment Routing List [2] ~ 1052 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1053 ~ Link Layer Address [2] ~ 1054 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1055 | Port Number [2] | .... ~ 1056 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 1057 ~ ... ~ 1058 ~ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1059 ~ | ifIndex[N] | SRT | LHS |FMT| 1060 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1061 ~ Segment Routing List [N] ~ 1062 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1063 ~ Link Layer Address [N] ~ 1064 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1065 | Port Number [N] | Zero Padding (if necessary) ... 1066 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+--+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1068 Figure 5: AERO Source/Target Link-Layer Address Option (S/TLLAO) 1069 Format 1071 In this format, Type and Length are set the same as specified for S/ 1072 TLLAOs in [RFC4861], with trailing zero padding octets added as 1073 necessary to produce an integral number of 8 octet blocks. The S/ 1074 TLLAO includes N ifIndex-tuples in correspondence to ifIndex-tuples 1075 that appear in the OMNI option. Each ifIndex-tuple includes the 1076 following information: 1078 o ifIndex[i] - the same value as in the corresponding ifIndex-tuple 1079 included in the OMNI option. 1081 o SRT[i] - a 3-bit "Segment Routing Topology" value (see: 1082 Section 3.2.6) coded as follows: 1084 * 000 - Red 1086 * 001 - Green 1087 * 010 - Blue-1 1089 * 011 - Blue-2 1091 * 100 - 111 - Reserved 1093 o LHS[i] - a 3-bit "LookaHead Segments" value that encodes the 1094 number (from 0 to 7) of entries in Segment Routing List [i]. 1096 o FMT[i] - a 2-bit "Format" code. Determines the format of the Link 1097 Layer Address [i] field as follows: 1099 * 00 - Link Layer Address [i] encodes a Teredo-format LLA for a 1100 node behind a NAT. 1102 * 01 - Link Layer Address [i] encodes a Teredo-format LLA for a 1103 node on the open INET. 1105 * 10 - Link Layer Address [i] encodes an IPv6 address for a node 1106 behind a NAT.. 1108 * 11 - Link Layer Address [i] encodes an IPv6 address for a node 1109 on the open INET. 1111 o Segment Routing List [i] - Includes LHS[i]-many 16 byte ULAs 1112 corresponding to the Segment IDs (SIDs) that must be visited prior 1113 to forwarding to Link Layer Address [i]. The ultimate SID appears 1114 first, followed by the penultimate SID second, etc. 1116 o Link Layer Address [i] - Included according to FMT[i], and 1117 identifies the link-layer address of the source/target. The IP 1118 address and port number (for IPv4) are recorded in ones-compliment 1119 "obfuscated" form per [RFC4380]. 1121 o Port Number [i] - Present only for IPv6 (i.e., when FMT[i] is 10 1122 or 11). When present, the field is 2 bytes in length and 1123 immediately follows Link Layer Address [i], with the value 1124 recorded in obfuscated form. 1126 If an S/TLLAO is included, any ifIndex-tuples correspond to a proper 1127 subset of the OMNI option ifIndex-tuples. Any S/TLLAO ifIndex-tuple 1128 having an ifIndex value that does not appear in an OMNI option 1129 ifindex-tuple is ignored. If the same ifIndex value appears in 1130 multiple ifIndex-tuples, the first tuple is processed and the 1131 remaining tuples are ignored. Any S/TLLAO ifIndex-tuples can 1132 therefore be viewed as inter-dependent extensions of their 1133 corresponding OMNI option ifIndex-tuples, i.e., the OMNI option and 1134 S/TLLAO are companions that are interpreted in conjunction with each 1135 other. 1137 A Client's AERO interface may be configured over multiple underlying 1138 interface connections. For example, common mobile handheld devices 1139 have both wireless local area network ("WLAN") and cellular wireless 1140 links. These links are often used "one at a time" with low-cost WLAN 1141 preferred and highly-available cellular wireless as a standby, but a 1142 simultaneous-use capability could provide benefits. In a more 1143 complex example, aircraft frequently have many wireless data link 1144 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1145 directional, etc.) with diverse performance and cost properties. 1147 If a Client's multiple underlying interfaces are used "one at a time" 1148 (i.e., all other interfaces are in standby mode while one interface 1149 is active), then ND message OMNI options include only a single 1150 ifIndex-tuple set to constant values. In that case, the Client would 1151 appear to have a single interface but with a dynamically changing 1152 link-layer address. 1154 If the Client has multiple active underlying interfaces, then from 1155 the perspective of ND it would appear to have multiple link-layer 1156 addresses. In that case, ND message OMNI options MAY include 1157 multiple ifIndex-tuples - each with values that correspond to a 1158 specific interface. Every ND message need not include all OMNI and/ 1159 or S/TLLAO ifIndex-tuples; for any ifIndex-tuple not included, the 1160 neighbor considers the status as unchanged. 1162 Bridge, Server and Proxy AERO interfaces may be configured over one 1163 or more secured tunnel interfaces. The AERO interface configures 1164 both an LLA and its corresponding ULA, while the underlying secured 1165 tunnel interfaces are either unnumbered or configure the same ULA. 1166 The AERO interface encapsulates each IP packet in a SPAN header and 1167 presents the packet to the underlying secured tunnel interface. For 1168 Bridges that do not configure an AERO interface, the secured tunnel 1169 interfaces themselves are exposed to the IP layer with each interface 1170 configuring the Bridge's ULA. Routing protocols such as BGP 1171 therefore run directly over the Bridge's secured tunnel interfaces. 1172 For nodes that configure an AERO interface, routing protocols such as 1173 BGP run over the AERO interface but do not employ SPAN encapsulation. 1174 Instead, the AERO interface presents the routing protocol messages 1175 directly to the underlying secured tunnels without applying 1176 encapsulation and while using the ULA as the source address. This 1177 distinction must be honored consistently according to each node's 1178 configuration so that the IP forwarding table will associate 1179 discovered IP routes with the correct interface. 1181 3.4. AERO Interface Initialization 1183 AERO Servers, Proxys and Clients configure AERO interfaces as their 1184 point of attachment to the AERO link. AERO nodes assign the MSPs for 1185 the link to their AERO interfaces (i.e., as a "route-to-interface") 1186 to ensure that packets with destination addresses covered by an MNP 1187 not explicitly assigned to a non-AERO interface are directed to the 1188 AERO interface. 1190 AERO interface initialization procedures for Servers, Proxys, Clients 1191 and Bridges are discussed in the following sections. 1193 3.4.1. AERO Server/Relay Behavior 1195 When a Server enables an AERO interface, it assigns an LLA/ULA 1196 appropriate for the given AERO link segment. The Server also 1197 configures secured tunnels with one or more neighboring Bridges and 1198 engages in a BGP routing protocol session with each Bridge. 1200 The AERO interface provides a single interface abstraction to the IP 1201 layer, but internally comprises multiple secured tunnels as well as 1202 an NBMA nexus for sending encapsulated data packets to AERO interface 1203 neighbors. The Server further configures a service to facilitate ND/ 1204 PD exchanges with AERO Clients and manages per-Client neighbor cache 1205 entries and IP forwarding table entries based on control message 1206 exchanges. 1208 Relays are simply Servers that run a dynamic routing protocol to 1209 redistribute routes between the AERO interface and INET/EUN 1210 interfaces (see: Section 3.2.4). The Relay provisions MNPs to 1211 networks on the INET/EUN interfaces (i.e., the same as a Client would 1212 do) and advertises the MSP(s) for the AERO link over the INET/EUN 1213 interfaces. The Relay further provides an attachment point of the 1214 AERO link to a non-MNP-based global topology. 1216 3.4.2. AERO Proxy Behavior 1218 When a Proxy enables an AERO interface, it assigns an LLA/ULA and 1219 configures permanent neighbor cache entries the same as for Servers. 1220 The Proxy also configures secured tunnels with one or more 1221 neighboring Bridges and maintains per-Client neighbor cache entries 1222 based on control message exchanges. 1224 3.4.3. AERO Client Behavior 1226 When a Client enables an AERO interface, it sends RS messages with 1227 ND/PD parameters over its underlying interfaces to a Server in the 1228 MAP list, which returns an RA message with corresponding parameters. 1230 (The RS/RA messages may pass through a Proxy in the case of a 1231 Client's Proxyed interface, or through one or more NATs in the case 1232 of a Client's INET interface.) 1234 3.4.4. AERO Bridge Behavior 1236 AERO Bridges need not connect directly to the AERO link, since they 1237 operate as link-layer forwarding devices instead of network layer 1238 routers. Configuration of AERO interfaces on Bridges is therefore 1239 OPTIONAL, e.g., if an administrative interface is needed. Bridges 1240 configure secured tunnels with Servers, Proxys and other Bridges; 1241 they also configure LLAs/ULAs and permanent neighbor cache entries 1242 the same as Servers. Bridges engage in a BGP routing protocol 1243 session with a subset of the Servers and other Bridges on the 1244 spanning tree (see: Section 3.2.4). 1246 3.5. AERO Interface Neighbor Cache Maintenance 1248 Each AERO interface maintains a conceptual neighbor cache that 1249 includes an entry for each neighbor it communicates with on the AERO 1250 link per [RFC4861]. AERO interface neighbor cache entries are said 1251 to be one of "permanent", "symmetric", "asymmetric" or "proxy". 1253 Permanent neighbor cache entries are created through explicit 1254 administrative action; they have no timeout values and remain in 1255 place until explicitly deleted. AERO Bridges maintain permanent 1256 neighbor cache entries for their associated Proxys and Servers (and 1257 vice-versa). Each entry maintains the mapping between the neighbor's 1258 network-layer LLA and corresponding INET address. 1260 Symmetric neighbor cache entries are created and maintained through 1261 RS/RA exchanges as specified in Section 3.12, and remain in place for 1262 durations bounded by ND/PD lifetimes. AERO Servers maintain 1263 symmetric neighbor cache entries for each of their associated 1264 Clients, and AERO Clients maintain symmetric neighbor cache entries 1265 for each of their associated Servers. The list of all Servers on the 1266 AERO link is maintained in the link's MAP list. 1268 Asymmetric neighbor cache entries are created or updated based on 1269 route optimization messaging as specified in Section 3.14, and are 1270 garbage-collected when keepalive timers expire. AERO ROSs maintain 1271 asymmetric neighbor cache entries for active targets with lifetimes 1272 based on ND messaging constants. Asymmetric neighbor cache entries 1273 are unidirectional since only the ROS (and not the ROR) creates an 1274 entry. 1276 Proxy neighbor cache entries are created and maintained by AERO 1277 Proxys when they process Client/Server ND/PD exchanges, and remain in 1278 place for durations bounded by ND/PD lifetimes. AERO Proxys maintain 1279 proxy neighbor cache entries for each of their associated Clients. 1280 Proxy neighbor cache entries track the Client state and the address 1281 of the Client's associated Server(s). 1283 To the list of neighbor cache entry states in Section 7.3.2 of 1284 [RFC4861], Proxy and Server AERO interfaces add an additional state 1285 DEPARTED that applies to symmetric and proxy neighbor cache entries 1286 for Clients that have recently departed. The interface sets a 1287 "DepartTime" variable for the neighbor cache entry to "DEPART_TIME" 1288 seconds. DepartTime is decremented unless a new ND message causes 1289 the state to return to REACHABLE. While a neighbor cache entry is in 1290 the DEPARTED state, packets destined to the target Client are 1291 forwarded to the Client's new location instead of being dropped. 1292 When DepartTime decrements to 0, the neighbor cache entry is deleted. 1293 It is RECOMMENDED that DEPART_TIME be set to the default constant 1294 value REACHABLE_TIME plus 10 seconds (40 seconds by default) to allow 1295 a window for packets in flight to be delivered while stale route 1296 optimization state may be present. 1298 When an ROR receives an authentic NS message used for route 1299 optimization, it searches for a symmetric neighbor cache entry for 1300 the target Client. The ROR then returns a solicited NA message 1301 without creating a neighbor cache entry for the ROS, but creates or 1302 updates a target Client "Report List" entry for the ROS and sets a 1303 "ReportTime" variable for the entry to REPORT_TIME seconds. The ROR 1304 resets ReportTime when it receives a new authentic NS message, and 1305 otherwise decrements ReportTime while no authentic NS messages have 1306 been received. It is RECOMMENDED that REPORT_TIME be set to the 1307 default constant value REACHABLE_TIME plus 10 seconds (40 seconds by 1308 default) to allow a window for route optimization to converge before 1309 ReportTime decrements below REACHABLE_TIME. 1311 When the ROS receives a solicited NA message response to its NS 1312 message used for route optimization, it creates or updates an 1313 asymmetric neighbor cache entry for the target network-layer and 1314 link-layer addresses. The ROS then (re)sets ReachableTime for the 1315 neighbor cache entry to REACHABLE_TIME seconds and uses this value to 1316 determine whether packets can be forwarded directly to the target, 1317 i.e., instead of via a default route. The ROS otherwise decrements 1318 ReachableTime while no further solicited NA messages arrive. It is 1319 RECOMMENDED that REACHABLE_TIME be set to the default constant value 1320 30 seconds as specified in [RFC4861]. 1322 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1323 of NS keepalives sent when a correspondent may have gone unreachable, 1324 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1325 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1326 to limit the number of unsolicited NAs that can be sent based on a 1327 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1328 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1329 same as specified in [RFC4861]. 1331 Different values for DEPART_TIME, REPORT_TIME, REACHABLE_TIME, 1332 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1333 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1334 different values are chosen, all nodes on the link MUST consistently 1335 configure the same values. Most importantly, DEPART_TIME and 1336 REPORT_TIME SHOULD be set to a value that is sufficiently longer than 1337 REACHABLE_TIME to avoid packet loss due to stale route optimization 1338 state. 1340 3.6. AERO Interface Encapsulation and Re-encapsulation 1342 In some instances, AERO interfaces insert a mid-layer IPv6 header 1343 known as the SPAN header as discussed in the following sections. 1344 After either inserting or omitting the SPAN header, the AERO 1345 interface inserts an outer encapsulation header as discussed below. 1347 AERO interfaces avoid outer encapsulation over Direct underlying 1348 interfaces and Proxyed underlying interfaces for which the first-hop 1349 access router is AERO-aware. Other AERO interfaces encapsulate 1350 packets according to whether they are entering the AERO interface 1351 from the network layer or if they are being re-admitted into the same 1352 AERO link they arrived on. This latter form of encapsulation is 1353 known as "re-encapsulation". 1355 For packets entering the AERO interface from the network layer, the 1356 AERO interface copies the "TTL/Hop Limit", "Type of Service/Traffic 1357 Class" [RFC2983], "Flow Label"[RFC6438] (for IPv6) and "Congestion 1358 Experienced" [RFC3168] values in the inner packet's IP header into 1359 the corresponding fields in the SPAN and outer encapsulation 1360 header(s). 1362 For packets undergoing re-encapsulation, the AERO interface instead 1363 copies these values from the original encapsulation header into the 1364 new encapsulation header, i.e., the values are transferred between 1365 encapsulation headers and *not* copied from the encapsulated packet's 1366 network-layer header. (Note especially that by copying the TTL/Hop 1367 Limit between encapsulation headers the value will eventually 1368 decrement to 0 if there is a (temporary) routing loop.) For IPv4 1369 encapsulation/re-encapsulation, the AERO interface sets the DF bit as 1370 discussed in Section 3.9. 1372 AERO interfaces configured over INET underlying interfaces 1373 encapsulate packets in INET headers according to the next hop 1374 determined in the forwarding algorithm in Section 3.10. If the next 1375 hop is reached via a secured tunnel, the AERO interface uses an 1376 encapsulation format specific to the secured tunnel type (see: 1377 Section 6). If the next hop is reached via an unsecured INET 1378 interface, the AERO interface instead uses UDP/IP encapsulation 1379 according to the Teredo format specified in [RFC4380] and as extended 1380 in [RFC6081]. 1382 When Teredo encapsulation is used, the AERO interface next sets the 1383 UDP source port to a constant value that it will use in each 1384 successive packet it sends, and sets the UDP length field to the 1385 length of the encapsulated packet plus 8 bytes for the UDP header 1386 itself plus the length of any included Teredo extension headers or 1387 trailers. The encapsulated packet may be either IPv6 or IPv4, as 1388 distinguished by the version number found in the first four bits. 1390 For Teredo-encapsulated packets sent to a Server, Relay or Bridge, 1391 the AERO interface sets the UDP destination port to 8060, i.e., the 1392 IANA-registered port number for AERO. For packets sent to a Client, 1393 the AERO interface sets the UDP destination port to the port value 1394 stored in the neighbor cache entry for this Client. The AERO 1395 interface finally includes/omits the UDP checksum according to 1396 [RFC6935][RFC6936]. 1398 AERO interfaces observe the packet sizing and fragmentation 1399 considerations found in Section 3.9. 1401 3.7. AERO Interface Decapsulation 1403 AERO interfaces decapsulate packets destined either to the AERO node 1404 itself or to a destination reached via an interface other than the 1405 AERO interface the packet was received on. When the encapsulated 1406 packet arrives in multiple SPAN fragments, the AERO interface 1407 reassembles as discussed in Section 3.9. Further decapsulation steps 1408 are performed according to the appropriate encapsulation format 1409 specification. 1411 3.8. AERO Interface Data Origin Authentication 1413 AERO nodes employ simple data origin authentication procedures. In 1414 particular: 1416 o AERO Bridges, Servers and Proxys accept encapsulated data packets 1417 and control messages received from the spanning tree. 1419 o AERO Proxys and Clients accept packets that originate from within 1420 the same secured ANET. 1422 o AERO Clients and Relays accept packets from downstream network 1423 correspondents based on ingress filtering. 1425 o AERO Clients, Relays and Servers verify the outer UDP/IP 1426 encapsulation addresses according to the Teredo specification 1427 [RFC4380]. 1429 AERO nodes silently drop any packets that do not satisfy the above 1430 data origin authentication procedures. Further security 1431 considerations are discussed in Section 6. 1433 3.9. AERO Interface MTU and Fragmentation 1435 IPv6 underlying interfaces are REQUIRED to configure a minimum 1436 Maximum Transmission Unit (MTU) of 1280 bytes [RFC8200]. The minimum 1437 MTU for IPv4 underlying interfaces is only 68 bytes [RFC1122], 1438 meaning that a packet smaller than the IPv6 MTU may require 1439 fragmentation when IPv4 encapsulation is used. Therefore, the Don't 1440 Fragment (DF) bit in the IPv4 encapsulation header MUST be set to 0. 1442 The AERO interface configures an MTU of 9180 bytes [RFC2492]; the 1443 size is therefore not a reflection of the underlying interface MTUs, 1444 but rather determines the largest packet the AERO interface can 1445 forward or reassemble. The AERO interface therefore accommodates IP 1446 packets up to 9180 bytes while generating IPv6 Path MTU Discovery 1447 (PMTUD) Packet Too Big (PTB) messages [RFC8201] as necessary (see 1448 below). 1450 AERO interfaces employ SPAN encapsulation and fragmentation/ 1451 reassembly per [RFC2473] to accommodate the 9180 byte MTU. The AERO 1452 interface returns internally-generated PTB messages for packets 1453 admitted into the interface that it deems too large (e.g., according 1454 to link performance characteristics, reassembly cost, etc.) while 1455 either dropping or forwarding the packet as necessary. The AERO 1456 interface performs PMTUD even if the destination appears to be on the 1457 same link since intermediate AERO link nodes may return a PTB. This 1458 ensures that the path MTU is adaptive and reflects the current path 1459 used for a given data flow. 1461 AERO nodes perform SPAN encapsulation and fragmentation/reassembly as 1462 follows: 1464 o When a node's AERO interface sends a packet over a Proxyed, VPNed 1465 or Direct underlying interface, it sends without SPAN 1466 encapsulation if the packet is no larger than the underlying 1467 interface MTU. Otherwise, it inserts a SPAN header with source 1468 address set to the node's own ULA and destination set to the ULA 1469 of the link-layer peer Proxy, Server or Client on the underlying 1470 interface. The AERO interface then uses IPv6 fragmentation to 1471 break the packet into a minimum number of non-overlapping 1472 fragments, where the largest fragment size is determined by the 1473 underlying interface MTU and the smallest fragment is no smaller 1474 than 640 bytes. The AERO interface then sends the fragments to 1475 the link-layer peer, which reassembles before forwarding toward 1476 the final destination. 1478 o When a node's AERO interface sends a packet over an INET 1479 underlying interface, it sends encapsulated packets no larger than 1480 1280 bytes without a SPAN header if the destination is reached via 1481 an INET address within the same AERO link segment. Otherwise, it 1482 inserts a SPAN header with source address set to the node's ULA, 1483 destination set to the ULA of the next hop AERO node toward the 1484 final destination and (if necessary) with a SRH with the remaining 1485 Segment IDs on the path to the final destination. The AERO 1486 interface then uses IPv6 fragmentation to break the encapsulated 1487 packet into a minimum number of non-overlapping fragments, where 1488 the largest fragment size (including both SPAN and INET 1489 encapsulation) is 1280 bytes and the smallest fragment is no 1490 smaller than 640 bytes. The AERO interface then encapsulates the 1491 SPAN fragments in INET headers and sends them to the SPAN 1492 destination, which reassembles before forwarding toward the final 1493 destination. 1495 In order to avoid a "tiny fragment" attack, AERO interfaces 1496 unconditionally drop all SPAN fragments smaller than 640 bytes. In 1497 order to set the correct context for reassembly, the AERO interface 1498 that inserts a SPAN header MUST also be the one that inserts the IPv6 1499 Fragment Header Identification value. Although all fragments of the 1500 same fragmented SPAN packet are typically sent via the same 1501 underlying interface, this is not strictly required since all 1502 fragments will arrive at the AERO interface that performs reassembly 1503 even if they travel over different paths. 1505 Note that the AERO interface can forward large packets via 1506 encapsulation and fragmentation while at the same time returning 1507 advisory PTB messages, e.g., subject to rate limiting. The receiving 1508 node that performs reassembly can also send advisory PTB messages if 1509 reassembly conditions become unfavorable. The AERO interface can 1510 therefore continuously forward large packets without loss while 1511 returning advisory messages recommending a smaller size (but no 1512 smaller than 1280). Advisory PTB messages are differentiated from 1513 PTB messages that report loss by setting the Code field in the ICMPv6 1514 message header to the value 1. This document therefore updates 1515 [RFC4443] and [RFC8201]. 1517 3.10. AERO Interface Forwarding Algorithm 1519 IP packets enter a node's AERO interface either from the network 1520 layer (i.e., from a local application or the IP forwarding system) or 1521 from the link layer (i.e., from an AERO interface neighbor). All 1522 packets entering a node's AERO interface first undergo data origin 1523 authentication as discussed in Section 3.8. Packets that satisfy 1524 data origin authentication are processed further, while all others 1525 are dropped silently. 1527 Packets that enter the AERO interface from the network layer are 1528 forwarded to an AERO interface neighbor. Packets that enter the AERO 1529 interface from the link layer are either re-admitted into the AERO 1530 link or forwarded to the network layer where they are subject to 1531 either local delivery or IP forwarding. In all cases, the AERO 1532 interface itself MUST NOT decrement the network layer TTL/Hop-count 1533 since its forwarding actions occur below the network layer. 1535 AERO interfaces may have multiple underlying interfaces and/or 1536 neighbor cache entries for neighbors with multiple ifIndex-tuple 1537 registrations (see Section 3.3). The AERO interface uses traffic 1538 classifiers (e.g., DSCP value, port number, etc.) to select an 1539 outgoing underlying interface for each packet based on the node's own 1540 QoS preferences, and also to select a destination link-layer address 1541 based on the neighbor's underlying interface with the highest 1542 preference. AERO implementations SHOULD allow for QoS preference 1543 values to be modified at runtime through network management. 1545 If multiple outgoing interfaces and/or neighbor interfaces have a 1546 preference of "high", the AERO node replicates the packet and sends 1547 one copy via each of the (outgoing / neighbor) interface pairs; 1548 otherwise, the node sends a single copy of the packet via an 1549 interface with the highest preference. AERO nodes keep track of 1550 which underlying interfaces are currently "reachable" or 1551 "unreachable", and only use "reachable" interfaces for forwarding 1552 purposes. 1554 The following sections discuss the AERO interface forwarding 1555 algorithms for Clients, Proxys, Servers and Bridges. In the 1556 following discussion, a packet's destination address is said to 1557 "match" if it is the same as a cached address, or if it is covered by 1558 a cached prefix (which may be encoded in an LLA). 1560 3.10.1. Client Forwarding Algorithm 1562 When an IP packet enters a Client's AERO interface from the network 1563 layer the Client searches for an asymmetric neighbor cache entry that 1564 matches the destination. If there is a match, the Client uses one or 1565 more "reachable" neighbor interfaces in the entry for packet 1566 forwarding. If there is no asymmetric neighbor cache entry, the 1567 Client instead forwards the packet toward a Server (the packet is 1568 intercepted by a Proxy if there is a Proxy on the path). The Client 1569 encapsulates the packet in a SPAN header and fragments if necessary 1570 according to MTU requirements (see: Section 3.9). 1572 When an IP packet enters a Client's AERO interface from the link- 1573 layer, if the destination matches one of the Client's MNPs or link- 1574 local addresses the Client reassembles and decapsulates as necessary 1575 and delivers the inner packet to the network layer. Otherwise, the 1576 Client drops the packet and MAY return a network-layer ICMP 1577 Destination Unreachable message subject to rate limiting (see: 1578 Section 3.11). 1580 3.10.2. Proxy Forwarding Algorithm 1582 For control messages originating from or destined to a Client, the 1583 Proxy intercepts the message and updates its proxy neighbor cache 1584 entry for the Client. The Proxy then forwards a (proxyed) copy of 1585 the control message. (For example, the Proxy forwards a proxied 1586 version of a Client's NS/RS message to the target neighbor, and 1587 forwards a proxied version of the NA/RA reply to the Client.) 1589 When the Proxy receives a data packet from a Client within the ANET, 1590 the Proxy reassembles and re-fragments if necessary then searches for 1591 an asymmetric neighbor cache entry that matches the destination and 1592 forwards as follows: 1594 o if the destination matches an asymmetric neighbor cache entry, the 1595 Proxy uses one or more "reachable" neighbor interfaces in the 1596 entry for packet forwarding using SPAN encapsulation and including 1597 a SRH if necessary according to the cached TLLAO information. If 1598 the neighbor interface is in the same SPAN segment, the Proxy 1599 forwards the packet directly to the neighbor; otherwise, it 1600 forwards the packet to a Bridge. 1602 o else, the Proxy uses SPAN encapsulation and forwards the packet to 1603 a Bridge while using the ULA corresponding to the packet's 1604 destination as the SPAN destination address. 1606 When the Proxy receives an encapsulated data packet from an INET 1607 neighbor or from a secured tunnel from a Bridge, it accepts the 1608 packet only if data origin authentication succeeds and if there is a 1609 proxy neighbor cache entry that matches the inner destination. Next, 1610 the Proxy reassembles the packet (if necessary) and continues 1611 processing. 1613 Next if reassembly is complete and the neighbor cache state is 1614 REACHABLE, the Proxy returns a PTB if necessary (see: Section 3.9) 1615 then either drops or forwards the packet to the Client while 1616 performing SPAN encapsulation and re-fragmentation to the ANET MTU 1617 size if necessary. If the neighbor cache entry state is DEPARTED, 1618 the Proxy instead changes the SPAN destination address to the address 1619 of the new Server and forwards it to a Bridge while performing re- 1620 fragmentation to 1280 bytes if necessary. 1622 3.10.3. Server/Relay Forwarding Algorithm 1624 For control messages destined to a target Client's LLA that are 1625 received from a secured tunnel, the Server intercepts the message and 1626 sends an appropriate response on behalf of the Client. (For example, 1627 the Server sends an NA message reply in response to an NS message 1628 directed to one of its associated Clients.) If the Client's neighbor 1629 cache entry is in the DEPARTED state, however, the Server instead 1630 forwards the packet to the Client's new Server as discussed in 1631 Section 3.16. 1633 When the Server receives an encapsulated data packet from an INET 1634 neighbor or from a secured tunnel, it accepts the packet only if data 1635 origin authentication succeeds. If the SPAN destination address is 1636 its own address, the Server continues processing as follows: 1638 o if the destination matches a symmetric neighbor cache entry in the 1639 REACHABLE state the Server prepares the packet for forwarding to 1640 the destination Client. The Server first reassembles (if 1641 necessary) and forwards the packet (while re-fragmenting if 1642 necessary) as specified in Section 3.9. 1644 o else, if the destination matches a symmetric neighbor cache entry 1645 in the DEPARETED state the Server re-encapsulates the packet and 1646 forwards it using the ULA of the Client's new Server as the 1647 destination. 1649 o else, if the destination matches an asymmetric neighbor cache 1650 entry, the Server uses one or more "reachable" neighbor interfaces 1651 in the entry for packet forwarding via the local INET if the 1652 neighbor is in the same AERO link segment or using SPAN 1653 encapsulation and Segment Routing if necessary with the final 1654 destination set to the neighbor's ULA otherwise. 1656 o else, if the destination is an LLA that is not assigned on the 1657 AERO interface the Server drops the packet. 1659 o else, the Server (acting as a Relay) reassembles if necessary, 1660 decapsulates the packet and releases it to the network layer for 1661 local delivery or IP forwarding. Based on the information in the 1662 forwarding table, the network layer may return the packet to the 1663 same AERO interface in which case further processing occurs as 1664 below. (Note that this arrangement accommodates common 1665 implementations in which the IP forwarding table is not accessible 1666 from within the AERO interface. If the AERO interface can 1667 directly access the IP forwarding table (such as for in-kernel 1668 implementations) the forwarding table lookup can instead be 1669 performed internally from within the AERO interface itself.) 1671 When the Server's AERO interface receives a data packet from the 1672 network layer or from a VPNed or Direct Client, it performs SPAN 1673 encapsulation and fragmentation if necessary, then processes the 1674 packet according to the network-layer destination address as follows: 1676 o if the destination matches a symmetric or asymmetric neighbor 1677 cache entry the Server processes the packet as above. 1679 o else, the Server encapsulates the packet and forwards it to a 1680 Bridge using its own ULA as the source and the ULA corresponding 1681 to the destination as the destination. 1683 3.10.4. Bridge Forwarding Algorithm 1685 Bridges forward SPAN-encapsulated packets over secured tunnels the 1686 same as any IP router. When the Bridge receives a SPAN-encapsulated 1687 packet via a secured tunnel, it removes the outer INET header and 1688 searches for a forwarding table entry that matches the SPAN 1689 destination address. The Bridge then processes the packet as 1690 follows: 1692 o if the destination is the ULA of a Server located in the local 1693 AERO link segment, the Bridge checks for a SRH. If there is a SRH 1694 with the ULA of the final destination as the penultimate ID and 1695 with a Link Layer Address of the final destination as the ultimate 1696 ID, the Bridge copies the ULA of the final destination into the 1697 destination address. If the Link Layer Address does not indicate 1698 the presence of a NAT, the Bridge then forwards the packet 1699 directly to the Link Layer Address using link-layer (UDP/IP) 1700 encapsulation. Otherwise, the Bridge forwards the packet directly 1701 to the Server. 1703 o if the destination matches one of the Bridge's own addresses, the 1704 Bridge submits the packet for local delivery. 1706 o else, if the destination matches a forwarding table entry the 1707 Bridge forwards the packet via a secured tunnel to the next hop. 1708 If the destination matches an MSP without matching an MNP, 1709 however, the Bridge instead drops the packet and returns an ICMP 1710 Destination Unreachable message subject to rate limiting (see: 1711 Section 3.11). 1713 o else, the Bridge drops the packet and returns an ICMP Destination 1714 Unreachable as above. 1716 As for any IP router, the Bridge decrements the TTL/Hop Limit when it 1717 forwards the packet. Therefore, only the Hop Limit in the SPAN 1718 header is decremented, and not the TTL/Hop Limit in the inner packet 1719 header. 1721 3.11. AERO Interface Error Handling 1723 When an AERO node admits a packet into the AERO interface, it may 1724 receive link-layer or network-layer error indications. 1726 A link-layer error indication is an ICMP error message generated by a 1727 router in the INET on the path to the neighbor or by the neighbor 1728 itself. The message includes an IP header with the address of the 1729 node that generated the error as the source address and with the 1730 link-layer address of the AERO node as the destination address. 1732 The IP header is followed by an ICMP header that includes an error 1733 Type, Code and Checksum. Valid type values include "Destination 1734 Unreachable", "Time Exceeded" and "Parameter Problem" 1735 [RFC0792][RFC4443]. (AERO interfaces ignore all link-layer IPv4 1736 "Fragmentation Needed" and IPv6 "Packet Too Big" messages since they 1737 only emit packets that are guaranteed to be no larger than the IP 1738 minimum link MTU as discussed in Section 3.9.) 1740 The ICMP header is followed by the leading portion of the packet that 1741 generated the error, also known as the "packet-in-error". For 1742 ICMPv6, [RFC4443] specifies that the packet-in-error includes: "As 1743 much of invoking packet as possible without the ICMPv6 packet 1744 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1745 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1746 "Internet Header + 64 bits of Original Data Datagram", however 1747 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1748 ICMP datagram SHOULD contain as much of the original datagram as 1749 possible without the length of the ICMP datagram exceeding 576 1750 bytes". 1752 The link-layer error message format is shown in Figure 6 (where, "L2" 1753 and "L3" refer to link-layer and network-layer, respectively): 1755 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1756 ~ ~ 1757 | L2 IP Header of | 1758 | error message | 1759 ~ ~ 1760 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1761 | L2 ICMP Header | 1762 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1763 ~ ~ P 1764 | IP and other encapsulation | a 1765 | headers of original L3 packet | c 1766 ~ ~ k 1767 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1768 ~ ~ t 1769 | IP header of | 1770 | original L3 packet | i 1771 ~ ~ n 1772 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1773 ~ ~ e 1774 | Upper layer headers and | r 1775 | leading portion of body | r 1776 | of the original L3 packet | o 1777 ~ ~ r 1778 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1780 Figure 6: AERO Interface Link-Layer Error Message Format 1782 The AERO node rules for processing these link-layer error messages 1783 are as follows: 1785 o When an AERO node receives a link-layer Parameter Problem message, 1786 it processes the message the same as described as for ordinary 1787 ICMP errors in the normative references [RFC0792][RFC4443]. 1789 o When an AERO node receives persistent link-layer Time Exceeded 1790 messages, the IP ID field may be wrapping before earlier fragments 1791 awaiting reassembly have been processed. In that case, the node 1792 should begin including integrity checks and/or institute rate 1793 limits for subsequent packets. 1795 o When an AERO node receives persistent link-layer Destination 1796 Unreachable messages in response to encapsulated packets that it 1797 sends to one of its asymmetric neighbor correspondents, the node 1798 should process the message as an indication that a path may be 1799 failing, and optionally initiate NUD over that path. If it 1800 receives Destination Unreachable messages over multiple paths, the 1801 node should allow future packets destined to the correspondent to 1802 flow through a default route and re-initiate route optimization. 1804 o When an AERO Client receives persistent link-layer Destination 1805 Unreachable messages in response to encapsulated packets that it 1806 sends to one of its symmetric neighbor Servers, the Client should 1807 mark the path as unusable and use another path. If it receives 1808 Destination Unreachable messages on many or all paths, the Client 1809 should associate with a new Server and release its association 1810 with the old Server as specified in Section 3.16.5. 1812 o When an AERO Server receives persistent link-layer Destination 1813 Unreachable messages in response to encapsulated packets that it 1814 sends to one of its symmetric neighbor Clients, the Server should 1815 mark the underlying path as unusable and use another underlying 1816 path. 1818 o When an AERO Server or Proxy receives link-layer Destination 1819 Unreachable messages in response to an encapsulated packet that it 1820 sends to one of its permanent neighbors, it treats the messages as 1821 an indication that the path to the neighbor may be failing. 1822 However, the dynamic routing protocol should soon reconverge and 1823 correct the temporary outage. 1825 When an AERO Bridge receives a packet for which the network-layer 1826 destination address is covered by an MSP, if there is no more- 1827 specific routing information for the destination the Bridge drops the 1828 packet and returns a network-layer Destination Unreachable message 1829 subject to rate limiting. The Bridge writes the network-layer source 1830 address of the original packet as the destination address and uses 1831 one of its non link-local addresses as the source address of the 1832 message. 1834 When an AERO node receives an encapsulated packet for which the 1835 reassembly buffer it too small, it drops the packet and returns a 1836 network-layer Packet Too Big (PTB) message. The node first writes 1837 the MRU value into the PTB message MTU field, writes the network- 1838 layer source address of the original packet as the destination 1839 address and writes one of its non link-local addresses as the source 1840 address. 1842 3.12. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1844 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1845 coordinated as discussed in the following Sections. 1847 3.12.1. AERO ND/PD Service Model 1849 Each AERO Server on the link configures a PD service to facilitate 1850 Client requests. Each Server is provisioned with a database of MNP- 1851 to-Client ID mappings for all Clients enrolled in the AERO service, 1852 as well as any information necessary to authenticate each Client. 1853 The Client database is maintained by a central administrative 1854 authority for the AERO link and securely distributed to all Servers, 1855 e.g., via the Lightweight Directory Access Protocol (LDAP) [RFC4511], 1856 via static configuration, etc. Clients receive the same service 1857 regardless of the Servers they select. 1859 AERO Clients and Servers use ND messages to maintain neighbor cache 1860 entries. AERO Servers configure their AERO interfaces as advertising 1861 NBMA interfaces, and therefore send unicast RA messages with a short 1862 Router Lifetime value (e.g., ReachableTime seconds) in response to a 1863 Client's RS message. Thereafter, Clients send additional RS messages 1864 to keep Server state alive. 1866 AERO Clients and Servers include PD parameters in RS/RA messages (see 1867 [I-D.templin-6man-dhcpv6-ndopt] for ND/PD alternatives). The unified 1868 ND/PD messages are exchanged between Client and Server according to 1869 the prefix management schedule required by the PD service. If the 1870 Client knows its MNP in advance, it can instead employ prefix 1871 registration by including its LLA as the source address of an RS 1872 message and with an OMNI option with valid prefix registration 1873 information for the MNP. If the Server (and Proxy) accept the 1874 Client's MNP assertion, they inject the prefix into the routing 1875 system and establish the necessary neighbor cache state. 1877 The following sections specify the Client and Server behavior. 1879 3.12.2. AERO Client Behavior 1881 AERO Clients discover the addresses of Servers in a similar manner as 1882 described in [RFC5214]. Discovery methods include static 1883 configuration (e.g., from a flat-file map of Server addresses and 1884 locations), or through an automated means such as Domain Name System 1885 (DNS) name resolution [RFC1035]. Alternatively, the Client can 1886 discover Server addresses through a layer 2 data link login exchange, 1887 or through a unicast RA response to a multicast/anycast RS as 1888 described below. In the absence of other information, the Client can 1889 resolve the DNS Fully-Qualified Domain Name (FQDN) 1890 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 1891 text string and "[domainname]" is a DNS suffix for the AERO link 1892 (e.g., "example.com"). 1894 To associate with a Server, the Client acts as a requesting router to 1895 request MNPs. The Client prepares an RS message with PD parameters 1896 and includes a Nonce and Timestamp option if the Client needs to 1897 correlate RA replies. If the Client already knows the Server's LLA, 1898 it includes the LLA as the network-layer destination address; 1899 otherwise, it includes the link-scoped All-Routers multicast 1900 (ff02::2) or Subnet-Router anycast (fe80::) address as the network- 1901 layer destination. If the Client already knows its own LLA, it uses 1902 the LLA as the network-layer source address; otherwise, it uses the 1903 unspecified IPv6 address (::/128) as the network-layer source 1904 address. 1906 The Client next includes an OMNI option in the RS message to register 1907 its link-layer information with the Server. The Client sets the OMNI 1908 option prefix registration information according to the MNP, and 1909 includes an ifIndex-tuple with S set to '1' corresponding to the 1910 underlying interface over which the Client will send the RS message. 1911 The Client MAY include additional ifIndex-tuples specific to other 1912 underlying interfaces. The Client MAY also include an SLLAO 1913 corresponding to the OMNI option ifIndex-tuple with S set to '1'. 1915 The Client then sends the RS message (either directly via Direct 1916 interfaces, via a VPN for VPNed interfaces, via a Proxy for proxyed 1917 interfaces or via INET encapsulation for INET interfaces) and waits 1918 for an RA message reply (see Section 3.12.3). The Client retries up 1919 to MAX_RTR_SOLICITATIONS times until an RA is received. If the 1920 Client receives no RAs, or if it receives an RA with Router Lifetime 1921 set to 0, the Client SHOULD abandon this Server and try another 1922 Server. Otherwise, the Client processes the PD information found in 1923 the RA message. 1925 Next, the Client creates a symmetric neighbor cache entry with the 1926 Server's LLA as the network-layer address and the Server's 1927 encapsulation and/or link-layer addresses as the link-layer address. 1928 The Client records the RA Router Lifetime field value in the neighbor 1929 cache entry as the time for which the Server has committed to 1930 maintaining the MNP in the routing system via this underlying 1931 interface, and caches the other RA configuration information 1932 including Cur Hop Limit, M and O flags, Reachable Time and Retrans 1933 Timer. The Client then autoconfigures LLAs for each of the delegated 1934 MNPs and assigns them to the AERO interface. The Client also caches 1935 any MSPs included in Route Information Options (RIOs) [RFC4191] as 1936 MSPs to associate with the AERO link, and assigns the MTU value in 1937 the MTU option to the underlying interface. 1939 The Client then registers additional underlying interfaces with the 1940 Server by sending RS messages via each additional interface. The RS 1941 messages include the same parameters as for the initial RS/RA 1942 exchange, but with destination address set to the Server's LLA. 1944 Following autoconfiguration, the Client sub-delegates the MNPs to its 1945 attached EUNs and/or the Client's own internal virtual interfaces as 1946 described in [I-D.templin-v6ops-pdhost] to support the Client's 1947 downstream attached "Internet of Things (IoT)". The Client 1948 subsequently sends additional RS messages over each underlying 1949 interface before the Router Lifetime received for that interface 1950 expires. 1952 After the Client registers its underlying interfaces, it may wish to 1953 change one or more registrations, e.g., if an interface changes 1954 address or becomes unavailable, if QoS preferences change, etc. To 1955 do so, the Client prepares an RS message to send over any available 1956 underlying interface. The RS includes an OMNI option with prefix 1957 registration information specific to its MNP, with an ifIndex-tuple 1958 specific to the selected underlying interface with S set to '1', and 1959 with any additional ifIndex-tuples specific to other underlying 1960 interfaces. The Client includes fresh ifIndex-tuple values to update 1961 the Server's neighbor cache entry. When the Client receives the 1962 Server's RA response, it has assurance that the Server has been 1963 updated with the new information. 1965 If the Client wishes to discontinue use of a Server it issues an RS 1966 message over any underlying interface with an OMNI option with a 1967 prefix release indication. When the Server processes the message, it 1968 releases the MNP, sets the symmetric neighbor cache entry state for 1969 the Client to DEPARTED and returns an RA reply with Router Lifetime 1970 set to 0. After a short delay (e.g., 2 seconds), the Server 1971 withdraws the MNP from the routing system. 1973 3.12.3. AERO Server Behavior 1975 AERO Servers act as IP routers and support a PD service for Clients. 1976 Servers arrange to add their LLAs to a static map of Server addresses 1977 for the link and/or the DNS resource records for the FQDN 1978 "linkupnetworks.[domainname]" before entering service. Server 1979 addresses should be geographically and/or topologically referenced, 1980 and made available for discovery by Clients on the AERO link. 1982 When a Server receives a prospective Client's RS message on its AERO 1983 interface, it SHOULD return an immediate RA reply with Router 1984 Lifetime set to 0 if it is currently too busy or otherwise unable to 1985 service the Client. Otherwise, the Server authenticates the RS 1986 message and processes the PD parameters. The Server first determines 1987 the correct MNPs to delegate to the Client by searching the Client 1988 database. When the Server delegates the MNPs, it also creates a 1989 forwarding table entry for each MNP so that the MNPs are propagated 1990 into the routing system (see: Section 3.2.4). For IPv6, the Server 1991 creates an IPv6 forwarding table entry for each MNP. For IPv4, the 1992 Server creates an IPv6 forwarding table entry with the SPAN 1993 Compatibility Prefix (SCP) corresponding to the IPv4 address. 1995 The Server next creates a symmetric neighbor cache entry for the 1996 Client using the base LLA as the network-layer address and with 1997 lifetime set to no more than the smallest PD lifetime. Next, the 1998 Server updates the neighbor cache entry by recording the information 1999 in each ifIndex-tuple in the RS OMNI option. The Server also records 2000 the actual SPAN/INET addresses in the neighbor cache entry. 2002 Next, the Server prepares an RA message using its LLA as the network- 2003 layer source address and the network-layer source address of the RS 2004 message as the network-layer destination address. The Server sets 2005 the Router Lifetime to the time for which it will maintain both this 2006 underlying interface individually and the symmetric neighbor cache 2007 entry as a whole. The Server also sets Cur Hop Limit, M and O flags, 2008 Reachable Time and Retrans Timer to values appropriate for the AERO 2009 link. The Server includes the delegated MNPs, any other PD 2010 parameters and an OMNI option with no ifIndex-tuples. The Server 2011 then includes one or more RIOs that encode the MSPs for the AERO 2012 link, plus an MTU option (see Section 3.9). The Server finally 2013 forwards the message to the Client using SPAN/INET, INET, or NULL 2014 encapsulation as necessary. 2016 After the initial RS/RA exchange, the Server maintains a 2017 ReachableTime timer for each of the Client's underlying interfaces 2018 individually (and for the Client's symmetric neighbor cache entry 2019 collectively) set to expire after ReachableTime seconds. If the 2020 Client (or Proxy) issues additional RS messages, the Server sends an 2021 RA response and resets ReachableTime. If the Server receives an ND 2022 message with PD release indication it sets the Client's symmetric 2023 neighbor cache entry to the DEPARTED state and withdraws the MNP from 2024 the routing system after a short delay (e.g., 2 seconds). If 2025 ReachableTime expires before a new RS is received on an individual 2026 underlying interface, the Server marks the interface as DOWN. If 2027 ReachableTime expires before any new RS is received on any individual 2028 underlying interface, the Server sets the symmetric neighbor cache 2029 entry state to STALE and sets a 10 second timer. If the Server has 2030 not received a new RS or ND message with PD release indication before 2031 the 10 second timer expires, it deletes the neighbor cache entry and 2032 withdraws the MNP from the routing system. 2034 The Server processes any ND/PD messages pertaining to the Client and 2035 returns an NA/RA reply in response to solicitations. The Server may 2036 also issue unsolicited RA messages, e.g., with PD reconfigure 2037 parameters to cause the Client to renegotiate its PDs, with Router 2038 Lifetime set to 0 if it can no longer service this Client, etc. 2039 Finally, If the symmetric neighbor cache entry is in the DEPARTED 2040 state, the Server deletes the entry after DepartTime expires. 2042 Note: Clients SHOULD notify former Servers of their departures, but 2043 Servers are responsible for expiring neighbor cache entries and 2044 withdrawing routes even if no departure notification is received 2045 (e.g., if the Client leaves the network unexpectedly). Servers 2046 SHOULD therefore set Router Lifetime to ReachableTime seconds in 2047 solicited RA messages to minimize persistent stale cache information 2048 in the absence of Client departure notifications. A short Router 2049 Lifetime also ensures that proactive Client/Server RS/RA messaging 2050 will keep any NAT state alive (see above). 2052 Note: All Servers on an AERO link MUST advertise consistent values in 2053 the RA Cur Hop Limit, M and O flags, Reachable Time and Retrans Timer 2054 fields the same as for any link, since unpredictable behavior could 2055 result if different Servers on the same link advertised different 2056 values. 2058 3.12.3.1. Lightweight DHCPv6 Relay Agent (LDRA) 2060 When DHCPv6 is used as the ND/PD service back end, AERO Clients and 2061 Servers are always on the same link (i.e., the AERO link) from the 2062 perspective of DHCPv6. However, in some implementations the DHCPv6 2063 server and ND function may be located in separate modules. In that 2064 case, the Server's AERO interface module can act as a Lightweight 2065 DHCPv6 Relay Agent (LDRA)[RFC6221] to relay PD messages to and from 2066 the DHCPv6 server module. 2068 When the LDRA receives an authentic RS message, it extracts the PD 2069 message parameters and uses them to construct an IPv6/UDP/DHCPv6 2070 message. It sets the IPv6 source address to the source address of 2071 the RS message, sets the IPv6 destination address to 2072 'All_DHCP_Relay_Agents_and_Servers' and sets the UDP fields to values 2073 that will be understood by the DHCPv6 server. 2075 The LDRA then wraps the message in a DHCPv6 'Relay-Forward' message 2076 header and includes an 'Interface-Id' option that includes enough 2077 information to allow the LDRA to forward the resulting Reply message 2078 back to the Client (e.g., the Client's link-layer addresses, a 2079 security association identifier, etc.). The LDRA also wraps the OMNI 2080 option and SLLAO into the Interface-Id option, then forwards the 2081 message to the DHCPv6 server. 2083 When the DHCPv6 server prepares a Reply message, it wraps the message 2084 in a 'Relay-Reply' message and echoes the Interface-Id option. The 2085 DHCPv6 server then delivers the Relay-Reply message to the LDRA, 2086 which discards the Relay-Reply wrapper and IPv6/UDP headers, then 2087 uses the DHCPv6 message to construct an RA response to the Client. 2088 The Server uses the information in the Interface-Id option to prepare 2089 the RA message and to cache the link-layer addresses taken from the 2090 OMNI option and SLLAO echoed in the Interface-Id option. 2092 3.13. The AERO Proxy 2094 Clients may connect to ANETs that deploy perimeter security services 2095 to facilitate communications to Servers in outside INETs. In that 2096 case, the ANET can employ an AERO Proxy. The Proxy is located at the 2097 ANET/INET border and listens for RS messages originating from or RA 2098 messages destined to ANET Clients. The Proxy acts on these control 2099 messages as follows: 2101 o when the Proxy receives an RS message from a new ANET Client, it 2102 first authenticates the message then examines the network-layer 2103 destination address. If the destination address is a Server's 2104 LLA, the Proxy proceeds to the next step. Otherwise, if the 2105 destination is All-Routers multicast or Subnet-Router anycast, the 2106 Proxy selects a "nearby" Server that is likely to be a good 2107 candidate to serve the Client and replaces the destination address 2108 with the Server's LLA. Next, the Proxy creates a proxy neighbor 2109 cache entry and caches the Client and Server link-layer addresses 2110 along with the OMNI option information and any other identifying 2111 information including Transaction IDs, Client Identifiers, Nonce 2112 values, etc. The Proxy finally encapsulates the (proxyed) RS 2113 message in a SPAN header with source set to the Proxy's ULA and 2114 destination set to the Server's ULA then forwards the message into 2115 the SPAN. 2117 o when the Server receives the RS, it authenticates the message then 2118 creates or updates a symmetric neighbor cache entry for the Client 2119 with the Proxy's ULA as the link-layer address. The Server then 2120 sends an RA message back to the Proxy via the spanning tree. 2122 o when the Proxy receives the RA, it authenticates the message and 2123 matches it with the proxy neighbor cache entry created by the RS. 2124 The Proxy then caches the PD route information as a mapping from 2125 the Client's MNPs to the Client's link-layer address, caches the 2126 Server's advertised Router Lifetime and sets the neighbor cache 2127 entry state to REACHABLE. The Proxy then sets the P bit in the RA 2128 flags field, optionally rewrites the Router Lifetime and forwards 2129 the (proxyed) message to the Client. The Proxy finally includes 2130 an MTU option (if necessary) with an MTU to use for the underlying 2131 ANET interface. 2133 After the initial RS/RA exchange, the Proxy forwards any Client data 2134 packets for which there is no matching asymmetric neighbor cache 2135 entry to a Bridge using SPAN encapsulation with its own ULA as the 2136 source and the ULA corresponding to the Client as the destination. 2138 The Proxy instead forwards any Client data destined to an asymmetric 2139 neighbor cache target directly to the target according to the SPAN/ 2140 link-layer information - the process of establishing asymmetric 2141 neighbor cache entries is specified in Section 3.14. 2143 While the Client is still attached to the ANET, the Proxy sends NS, 2144 RS and/or unsolicited NA messages to update the Server's symmetric 2145 neighbor cache entries on behalf of the Client and/or to convey QoS 2146 updates. This allows for higher-frequency Proxy-initiated RS/RA 2147 messaging over well-connected INET infrastructure supplemented by 2148 lower-frequency Client-initiated RS/RA messaging over constrained 2149 ANET data links. 2151 If the Server ceases to send solicited advertisements, the Proxy 2152 sends unsolicited RAs on the ANET interface with destination set to 2153 All-Nodes multicast (ff02::1) and with Router Lifetime set to zero to 2154 inform Clients that the Server has failed. Although the Proxy 2155 engages in ND exchanges on behalf of the Client, the Client can also 2156 send ND messages on its own behalf, e.g., if it is in a better 2157 position than the Proxy to convey QoS changes, etc. For this reason, 2158 the Proxy marks any Client-originated solicitation messages (e.g. by 2159 inserting a Nonce option) so that it can return the solicited 2160 advertisement to the Client instead of processing it locally. 2162 If the Client becomes unreachable, the Proxy sets the neighbor cache 2163 entry state to DEPARTED and retains the entry for DepartTime seconds. 2164 While the state is DEPARTED, the Proxy forwards any packets destined 2165 to the Client to a Bridge via SPAN encapsulation with the Client's 2166 current Server as the destination. The Bridge in turn forwards the 2167 packets to the Client's current Server. When DepartTime expires, the 2168 Proxy deletes the neighbor cache entry and discards any further 2169 packets destined to this (now forgotten) Client. 2171 In some ANETs that employ a Proxy, the Client's MNP can be injected 2172 into the ANET routing system. In that case, the Client can send data 2173 messages without encapsulation so that the ANET routing system 2174 transports the unencapsulated packets to the Proxy. This can be very 2175 beneficial, e.g., if the Client connects to the ANET via low-end data 2176 links such as some aviation wireless links. 2178 If the first-hop ANET access router is AERO-aware, the Client can 2179 avoid encapsulation for both its control and data messages. When the 2180 Client connects to the link, it can send an unencapsulated RS message 2181 with source address set to its LLA and with destination address set 2182 to the LLA of the Client's selected Server or to All-Routers 2183 multicast or Subnet-Router anycast. The Client includes an OMNI 2184 option formatted as specified in [I-D.templin-6man-omni-interface]. 2186 The Client then sends the unencapsulated RS message, which will be 2187 intercepted by the AERO-Aware access router. The access router then 2188 encapsulates the RS message in an ANET header with its own address as 2189 the source address and the address of a Proxy as the destination 2190 address. The access router further remembers the address of the 2191 Proxy so that it can encapsulate future data packets from the Client 2192 via the same Proxy. If the access router needs to change to a new 2193 Proxy, it simply sends another RS message toward the Server via the 2194 new Proxy on behalf of the Client. 2196 In some cases, the access router and Proxy may be one and the same 2197 node. In that case, the node would be located on the same physical 2198 link as the Client, but its message exchanges with the Server would 2199 need to pass through a security gateway at the ANET/INET border. The 2200 method for deploying access routers and Proxys (i.e. as a single node 2201 or multiple nodes) is an ANET-local administrative consideration. 2203 3.13.1. Detecting and Responding to Server Failures 2205 In environments where fast recovery from Server failure is required, 2206 Proxys SHOULD use proactive Neighbor Unreachability Detection (NUD) 2207 to track Server reachability in a similar fashion as for 2208 Bidirectional Forwarding Detection (BFD) [RFC5880]. Proxys can then 2209 quickly detect and react to failures so that cached information is 2210 re-established through alternate paths. The NUD control messaging is 2211 carried only over well-connected ground domain networks (i.e., and 2212 not low-end aeronautical radio links) and can therefore be tuned for 2213 rapid response. 2215 Proxys perform proactive NUD with Servers for which there are 2216 currently active ANET Clients by sending continuous NS messages in 2217 rapid succession, e.g., one message per second. The Proxy sends the 2218 NS message via the spanning tree with the Proxy's LLA as the source 2219 and the LLA of the Server as the destination. When the Proxy is also 2220 sending RS messages to the Server on behalf of ANET Clients, the 2221 resulting RA responses can be considered as equivalent hints of 2222 forward progress. This means that the Proxy need not also send a 2223 periodic NS if it has already sent an RS within the same period. If 2224 the Server fails (i.e., if the Proxy ceases to receive 2225 advertisements), the Proxy can quickly inform Clients by sending 2226 multicast RA messages on the ANET interface. 2228 The Proxy sends RA messages on the ANET interface with source address 2229 set to the Server's address, destination address set to All-Nodes 2230 multicast, and Router Lifetime set to 0. The Proxy SHOULD send 2231 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2232 [RFC4861]. Any Clients on the ANET that had been using the failed 2233 Server will receive the RA messages and associate with a new Server. 2235 3.13.2. Point-to-Multipoint Server Coordination 2237 In environments where Client messaging over ANETs is bandwidth- 2238 limited and/or expensive, Clients can enlist the services of the 2239 Proxy to coordinate with multiple Servers in a single RS/RA message 2240 exchange. The Client can send a single RS message to All-Routers 2241 multicast that includes the ID's of multiple Servers in MS-Register 2242 sub-options of the OMNI option. 2244 When the Proxy receives the RS and processes the OMNI option, it 2245 performs a separate RS/RA exchange with each MS-Register Server. 2246 When it has received the RA messages, it creates an "aggregate" RA 2247 message to return to the Client with an OMNI option with each 2248 responding Server's ID recorded in an MS-Register sub-option. 2250 Clients can thereafter employ efficient point-to-multipoint Server 2251 coordination under the assistance of the Proxy to dramatically reduce 2252 the number of messages sent over the ANET while enlisting the support 2253 of multiple Servers for fault tolerance. Clients can further include 2254 MS-Release suboptions in RS messages to request the Proxy to release 2255 from former Servers via the procedures discussed in Section 3.16.5. 2257 The OMNI interface specification [I-D.templin-6man-omni-interface] 2258 provides further discussion of the Client/Proxy RS/RA messaging 2259 involved in point-to-multipoint coordination. 2261 3.14. AERO Route Optimization 2263 While data packets are flowing between a source and target node, 2264 route optimization SHOULD be used. Route optimization is initiated 2265 by the first eligible Route Optimization Source (ROS) closest to the 2266 source as follows: 2268 o For Clients on VPNed and Direct interfaces, the Server is the ROS. 2270 o For Clients on Proxyed interfaces, the Proxy is the ROS. 2272 o For Clients on INET interfaces, the Client itself is the ROS. 2274 o For correspondent nodes on INET/EUN interfaces serviced by a 2275 Relay, the Relay is the ROS. 2277 The route optimization procedure is conducted between the ROS and the 2278 target Server/Relay acting as a Route Optimization Responder (ROR) in 2279 the same manner as for IPv6 ND Address Resolution and using the same 2280 NS/NA messaging. The target may either be a MNP Client serviced by a 2281 Server, or a non-MNP correspondent reachable via a Relay. 2283 The procedures are specified in the following sections. 2285 3.14.1. Route Optimization Initiation 2287 While data packets are flowing from the source node toward a target 2288 node, the ROS performs address resolution by sending an NS message 2289 for Address Resolution (NS(AR)) to receive a solicited NA message 2290 from the ROR. When the ROS sends an NS(AR), it includes: 2292 o the LLA of the ROS as the source address. 2294 o the data packet's destination as the Target Address. 2296 o the Solicited-Node multicast address [RFC4291] formed from the 2297 lower 24 bits of the data packet's destination as the destination 2298 address, e.g., for 2001:db8:1:2::10:2000 the NS destination 2299 address is ff02:0:0:0:0:1:ff10:2000. 2301 The NS(AR) message includes an OMNI option with no ifIndex-tuples and 2302 no SLLAO, such that the target will not create a neighbor cache 2303 entry. 2305 The ROS then encapsulates the NS(AR) message in a SPAN header with 2306 source set to its own ULA and destination set to the ULA 2307 corresponding to the packet's final destination, then sends the 2308 message into the spanning tree without decrementing the network-layer 2309 TTL/Hop Limit field. 2311 3.14.2. Relaying the NS 2313 When the Bridge receives the NS(AR) message from the ROS, it discards 2314 the INET header and determines that the ROR is the next hop by 2315 consulting its standard IPv6 forwarding table for the SPAN header 2316 destination address. The Bridge then forwards the message toward the 2317 ROR via the spanning tree the same as for any IPv6 router. The 2318 final-hop Bridge in the spanning tree will deliver the message via a 2319 secured tunnel to the ROR. 2321 3.14.3. Processing the NS and Sending the NA 2323 When the ROR receives the NS(AR) message, it examines the Target 2324 Address to determine whether it has a neighbor cache entry and/or 2325 route that matches the target. If there is no match, the ROR drops 2326 the message. Otherwise, the ROR continues processing as follows: 2328 o if the target belongs to an MNP Client neighbor in the DEPARTED 2329 state the ROR changes the NS(AR) message SPAN destination address 2330 to the ULA of the Client's new Server, forwards the message into 2331 the spanning tree and returns from processing. 2333 o If the target belongs to an MNP Client neighbor in the REACHABLE 2334 state, the ROR instead adds the AERO source address to the target 2335 Client's Report List with time set to ReportTime. 2337 o If the target belongs to a non-MNP route, the ROR continues 2338 processing without adding an entry to the Report List. 2340 The ROR then prepares a solicited NA message to send back to the ROS 2341 but does not create a neighbor cache entry. The ROR sets the NA 2342 source address to the LLA corresponding to the target, sets the 2343 Target Address to the target of the solicitation, and sets the 2344 destination address to the source of the solicitation. 2346 The ROR then includes an OMNI option with prefix registration length 2347 set to the length of the MNP if the target is an MNP Client; 2348 otherwise, set to the maximum of the non-MNP prefix length and 64. 2349 (Note that a /64 limit is imposed to avoid causing the ROS to set 2350 short prefixes (e.g., "default") that would match destinations for 2351 which the routing system includes more-specific prefixes.) 2353 If the target is an MNP Client, the ROR next includes ifIndex-tuples 2354 in the OMNI option for each of the target Client's underlying 2355 interfaces with current information for each interface and with the S 2356 flag set to 0. The ROR then includes a TLLAO with ifIndex-tuples in 2357 one-to-one correspondence with the tuples that appear in the OMNI 2358 option. 2360 The ROR sets the Link Layer Address and Port Number (if necessary) to 2361 its own INET address for VPNed and Direct interfaces or to the INET 2362 address of the Proxy for Proxyed interface, then includes its own ULA 2363 or the ULA of the Proxy as the ultimate Segment Routing List entry. 2364 For INET interfaces, the ROR instead sets the Link Layer Address and 2365 Port Number (if necessary) to the Client's INET address then sets its 2366 own ULA in the penultimate Segment Routing List entry and sets the 2367 target's ULA in the ultimate Segment Routing List entry. 2369 The ROR then sets the NA message R flag to 1 (as a router), S flag to 2370 1 (as a response to a solicitation), and O flag to 0 (as a proxy). 2371 The ROR finally encapsulates the NA message in a SPAN header with 2372 source set to its own ULA and destination set to the source ULA of 2373 the NS(AR) message, then forwards the message into the spanning tree 2374 without decrementing the network-layer TTL/Hop Limit field. 2376 3.14.4. Relaying the NA 2378 When the Bridge receives the NA message from the ROR, it discards the 2379 INET header and determines that the ROS is the next hop by consulting 2380 its standard IPv6 forwarding table for the SPAN header destination 2381 address. The Bridge then forwards the SPAN-encapsulated NA message 2382 toward the ROS the same as for any IPv6 router. The final-hop Bridge 2383 in the spanning tree will deliver the message via a secured tunnel to 2384 the ROS. 2386 3.14.5. Processing the NA 2388 When the ROS receives the solicited NA message, it processes the 2389 message the same as for standard IPv6 Address Resolution [RFC4861]. 2390 In the process, it caches the source ULA then creates an asymmetric 2391 neighbor cache entry for the ROR and caches all information found in 2392 the OMNI and TLLAO options. The ROS finally sets the asymmetric 2393 neighbor cache entry lifetime to ReachableTime seconds. 2395 3.14.6. Route Optimization Maintenance 2397 Following route optimization, the ROS forwards future data packets 2398 destined to the target via the addresses found in the cached link- 2399 layer information. The route optimization is shared by all sources 2400 that send packets to the target via the ROS, i.e., and not just the 2401 source on behalf of which the route optimization was initiated. 2403 While new data packets destined to the target are flowing through the 2404 ROS, it sends additional NS(AR) messages to the ROR before 2405 ReachableTime expires to receive a fresh solicited NA message the 2406 same as described in the previous sections (route optimization 2407 refreshment strategies are an implementation matter, with a non- 2408 normative example given in Appendix B.1). The ROS uses the cached 2409 ULA of the ROR as the NS(AR) SPAN destination address, and sends up 2410 to MAX_MULTICAST_SOLICIT NS(AR) messages separated by 1 second until 2411 an NA is received. If no NA is received, the ROS assumes that the 2412 current ROR has become unreachable and deletes the neighbor cache 2413 entry. Subsequent data packets will trigger a new route optimization 2414 per Section 3.14.1 to discover a new ROR while initial data packets 2415 travel over a suboptimal route. 2417 If an NA is received, the ROS then updates the asymmetric neighbor 2418 cache entry to refresh ReachableTime, while (for MNP destinations) 2419 the ROR adds or updates the ROS address to the target Client's Report 2420 List and with time set to ReportTime. While no data packets are 2421 flowing, the ROS instead allows ReachableTime for the asymmetric 2422 neighbor cache entry to expire. When ReachableTime expires, the ROS 2423 deletes the asymmetric neighbor cache entry. Any future data packets 2424 flowing through the ROS will again trigger a new route optimization. 2426 The ROS may also receive unsolicited NA messages from the ROR at any 2427 time (see: Section 3.16). If there is an asymmetric neighbor cache 2428 entry for the target, the ROS updates the link-layer information but 2429 does not update ReachableTime since the receipt of an unsolicited NA 2430 does not confirm that any forward paths are working. If there is no 2431 asymmetric neighbor cache entry, the ROS simply discards the 2432 unsolicited NA. 2434 In this arrangement, the ROS holds an asymmetric neighbor cache entry 2435 for the ROR, but the ROR does not hold an asymmetric neighbor cache 2436 entry for the ROS. The route optimization neighbor relationship is 2437 therefore asymmetric and unidirectional. If the target node also has 2438 packets to send back to the source node, then a separate route 2439 optimization procedure is performed in the reverse direction. But, 2440 there is no requirement that the forward and reverse paths be 2441 symmetric. 2443 3.15. Neighbor Unreachability Detection (NUD) 2445 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2446 [RFC4861] either reactively in response to persistent link-layer 2447 errors (see Section 3.11) or proactively to confirm reachability. 2448 The NUD algorithm is based on periodic control message exchanges. 2449 The algorithm may further be seeded by ND hints of forward progress, 2450 but care must be taken to avoid inferring reachability based on 2451 spoofed information. For example, authentic IPv6 ND message 2452 exchanges may be considered as acceptable hints of forward progress, 2453 while spurious data packets should not be. 2455 AERO Servers, Proxys and Relays can use standard NS/NA NUD exchanges 2456 sent over the spanning tree to securely test reachability without 2457 risk of DoS attacks from nodes pretending to be a neighbor; Proxys 2458 can further perform NUD to securely verify Server reachability on 2459 behalf of their proxyed Clients. However, a means for a ROS to test 2460 the unsecured forward directions of target route optimized paths is 2461 also necessary. 2463 When an ROR directs an ROS to a neighbor with one or more target 2464 link-layer addresses, the ROS can proactively test each such 2465 unsecured route optimized path by sending "loopback" NS(NUD) 2466 messages. While testing the paths, the ROS can optionally continue 2467 to send packets via the spanning tree, maintain a small queue of 2468 packets until target reachability is confirmed, or (optimistically) 2469 allow packets to flow via the route optimized paths. 2471 When the ROS sends a loopback NS(NUD) message, it uses its LLA as 2472 both the IPv6 source and destination address, and the MNP Subnet- 2473 Router anycast address as the Target Address. The ROS includes a 2474 Nonce and Timestamp option, then encapsulates the message in SPAN/ 2475 INET headers with its own ULA as the source and the ULA of the route 2476 optimization target as the destination. The ROS then forwards the 2477 message to the target (either directly to the link layer address of 2478 the target if the target is in the same AERO link segment, or via a 2479 Bridge if the target is in a different AERO link segment). 2481 When the route optimization target receives the NS(NUD) message, it 2482 notices that the IPv6 destination address is the same as the source 2483 address. It then reverses the SPAN source and destination addresses 2484 and returns the message to the ROS (either directly or via the 2485 spanning tree). The route optimization target does not decrement the 2486 NS(NUD) message IPv6 Hop-Limit in the process, since the message has 2487 not exited the AERO link. 2489 When the ROS receives the NS(NUD) message, it can determine from the 2490 Nonce, Timestamp and Target Address that the message originated from 2491 itself and that it transited the forward path. The ROS need not 2492 prepare an NA response, since the destination of the response would 2493 be itself and testing the route optimization path again would be 2494 redundant. 2496 The ROS marks route optimization target paths that pass these NUD 2497 tests as "reachable", and those that do not as "unreachable". These 2498 markings inform the AERO interface forwarding algorithm specified in 2499 Section 3.10. 2501 Note that to avoid a DoS vector nodes MUST NOT return loopback 2502 NS(NUD) messages received from an unsecured link-layer source via the 2503 spanning tree. 2505 3.16. Mobility Management and Quality of Service (QoS) 2507 AERO is a Distributed Mobility Management (DMM) service. Each Server 2508 is responsible for only a subset of the Clients on the AERO link, as 2509 opposed to a Centralized Mobility Management (CMM) service where 2510 there is a single network mobility collective entity for all Clients. 2511 Clients coordinate with their associated Servers via RS/RA exchanges 2512 to maintain the DMM profile, and the AERO routing system tracks all 2513 current Client/Server peering relationships. 2515 Servers provide default routing and mobility/multilink services for 2516 their dependent Clients. Clients are responsible for maintaining 2517 neighbor relationships with their Servers through periodic RS/RA 2518 exchanges, which also serves to confirm neighbor reachability. When 2519 a Client's underlying interface address and/or QoS information 2520 changes, the Client is responsible for updating the Server with this 2521 new information. Note that for Proxyed interfaces, however, the 2522 Proxy can also perform some RS/RA exchanges on the Client's behalf. 2524 Mobility management considerations are specified in the following 2525 sections. 2527 3.16.1. Mobility Update Messaging 2529 Servers accommodate Client mobility/multilink and/or QoS change 2530 events by sending unsolicited NA (uNA) messages to each ROS in the 2531 target Client's Report List. When a Server sends a uNA message, it 2532 sets the IPv6 source address to the Client's LLA, sets the 2533 destination address to All-Nodes multicast and sets the Target 2534 Address to the Client's Subnet-Router anycast address. The Server 2535 also includes an OMNI option with prefix registration information and 2536 with ifIndex-tuples for the target Client's remaining interfaces with 2537 S set to 0. The Server then includes a TLLAO with corresponding 2538 ifIndex-tuples prepared the same as for the initial route 2539 optimization event. The Server sets the NA R flag to 1, the S flag 2540 to 0 and the O flag to 0, then encapsulates the message in a SPAN 2541 header with source set to its own ULA and destination set to the ULA 2542 of the ROS and sends the message into the spanning tree. 2544 As discussed in Section 7.2.6 of [RFC4861], the transmission and 2545 reception of uNA messages is unreliable but provides a useful 2546 optimization. In well-connected Internetworks with robust data links 2547 uNA messages will be delivered with high probability, but in any case 2548 the Server can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT uNAs 2549 to each ROS to increase the likelihood that at least one will be 2550 received. 2552 When the ROS receives a uNA message, it ignores the message if there 2553 is no existing neighbor cache entry for the Client. Otherwise, it 2554 uses the included OMNI option and TLLAO information to update the 2555 neighbor cache entry, but does not reset ReachableTime since the 2556 receipt of an unsolicited NA message from the target Server does not 2557 provide confirmation that any forward paths to the target Client are 2558 working. 2560 If uNA messages are lost, the ROS may be left with stale address and/ 2561 or QoS information for the Client for up to ReachableTime seconds. 2562 During this time, the ROS can continue sending packets according to 2563 its stale neighbor cache information. When ReachableTime is close to 2564 expiring, the ROS will re-initiate route optimization and receive 2565 fresh link-layer address information. 2567 In addition to sending uNA messages to the current set of ROSs for 2568 the Client, the Server also sends uNAs to the former link-layer 2569 address for any ifIndex-tuple for which the link-layer address has 2570 changed. The uNA messages update Proxys that cannot easily detect 2571 (e.g., without active probing) when a formerly-active Client has 2572 departed. 2574 3.16.2. Announcing Link-Layer Address and/or QoS Preference Changes 2576 When a Client needs to change its underlying interface addresses and/ 2577 or QoS preferences (e.g., due to a mobility event), either the Client 2578 or its Proxys send RS messages to the Server via the spanning tree 2579 with an OMNI option that includes an ifIndex-tuple with S set to 1 2580 and with the new link quality and address information. 2582 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 2583 sending actual data packets in case one or more RAs are lost. If all 2584 RAs are lost, the Client SHOULD re-associate with a new Server. 2586 When the Server receives the Client's changes, it sends uNA messages 2587 to all nodes in the Report List the same as described in the previous 2588 section. 2590 3.16.3. Bringing New Links Into Service 2592 When a Client needs to bring new underlying interfaces into service 2593 (e.g., when it activates a new data link), it sends an RS message to 2594 the Server via the underlying interface with an OMNI option that 2595 includes an ifIndex-tuple with S set to 1 and appropriate link 2596 quality values and with link-layer address information for the new 2597 link. 2599 3.16.4. Removing Existing Links from Service 2601 When a Client needs to remove existing underlying interfaces from 2602 service (e.g., when it de-activates an existing data link), it sends 2603 an RS or uNA message to its Server with an OMNI option with 2604 appropriate link quality values. 2606 If the Client needs to send RS/uNA messages over an underlying 2607 interface other than the one being removed from service, it MUST 2608 include ifIndex-tuples with appropriate link quality values for any 2609 underlying interfaces being removed from service. 2611 3.16.5. Moving to a New Server 2613 When a Client associates with a new Server, it performs the Client 2614 procedures specified in Section 3.12.2. The Client also includes MS- 2615 Release identifiers in the RS message OMNI option per 2616 [I-D.templin-6man-omni-interface] if it wants the new Server to 2617 notify any old Servers from which the Client is departing. 2619 When the new Server receives the Client's RS message, it returns an 2620 RA as specified in Section 3.12.3 and sends up to 2621 MAX_NEIGHBOR_ADVERTIISEMENT uNA messages to any old Servers listed in 2622 OMNI option MS-Release identifiers. Each uNA message includes the 2623 Client's LLA as the source address, the old Server's LLA as the 2624 destination address, and an OMNI option with the Register/Release bit 2625 set to 0. The new Server wraps the uNA in a SPAN header with its own 2626 ULA as the source and the old Server's ULA as the destination, then 2627 sends the message into the spanning tree. 2629 When an old Server receives the uNA, it changes the Client's neighbor 2630 cache entry state to DEPARTED, sets the link-layer address of the 2631 Client to the new Server's ULA, and resets DepartTime. After a short 2632 delay (e.g., 2 seconds) the old Server withdraws the Client's MNP 2633 from the routing system. After DepartTime expires, the old Server 2634 deletes the Client's neighbor cache entry. 2636 The old Server also sends unsolicited NA messages to all ROSs in the 2637 Client's Report List with an OMNI option with a single ifIndex-tuple 2638 with ifIndex set to 0 and S set to '1', and with the ULA of the new 2639 Server in a companion TLLAO. When the ROS receives the NA, it caches 2640 the address of the new Server in the existing asymmetric neighbor 2641 cache entry and marks the entry as STALE for a period of 10 seconds 2642 after which the cache entry is deleted. While in the STALE state, 2643 subsequent data packets flow according to any existing cached link- 2644 layer information and trigger a new NS(AR)/NA exchange via the new 2645 Server. 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 AERO 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-AERO 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 AERO 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 a 2690 SPAN 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 TLLAOs with the ULA 2697 corresponding to any ifIndex-tuples that are currently servicing S. 2699 When Y processes the Join/Prune message, if S located behind any 2700 INET, Direct, or VPNed interfaces Y acts as a PIM router and updates 2701 its MRIB to list X as the next hop in the reverse path. If S is 2702 located behind any Proxys "Z"*, Y also forwards the message to each 2703 Z* over the spanning tree while continuing to use the LLA of X as the 2704 source address. Each Z* then updates its MRIB accordingly and 2705 maintains the LLA of X as the next hop in the reverse path. Since 2706 the Bridges do not examine network layer control messages, this means 2707 that the (reverse) multicast tree path is simply from each Z* (and/or 2708 Y) to X with no other multicast-aware routers in the path. If any Z* 2709 (and/or Y) is located on the same AERO link segment as X, the 2710 multicast data traffic sent to X directly using SPAN/INET 2711 encapsulation instead of via a Bridge. 2713 Following the initial Join/Prune and NS/NA messaging, X maintains an 2714 asymmetric neighbor cache entry for each S the same as if X was 2715 sending unicast data traffic to S. In particular, X performs 2716 additional NS/NA exchanges to keep the neighbor cache entry alive for 2717 up to t_periodic seconds [RFC7761]. If no new Joins are received 2718 within t_periodic seconds, X allows the neighbor cache entry to 2719 expire. Finally, if X receives any additional Join/Prune messages 2720 for (S,G) it forwards the messages to each Y and Z* in the neighbor 2721 cache entry over the spanning tree. 2723 At some later time, Client C that holds an MNP for source S may 2724 depart from a first Proxy Z1 and/or connect via a new Proxy Z2. In 2725 that case, Y sends an unsolicited NA message to X the same as 2726 specified for unicast mobility in Section 3.16. When X receives the 2727 unsolicited NA message, it updates its asymmetric neighbor cache 2728 entry for the LLA for source S and sends new Join messages to any new 2729 Proxys Z2. There is no requirement to send any Prune messages to old 2730 Proxys Z1 since source S will no longer source any multicast data 2731 traffic via Z1. Instead, the multicast state for (S,G) in Proxy Z1 2732 will soon time out since no new Joins will arrive. 2734 After some later time, C may move to a new Server Y2 and depart from 2735 old Sever Y1. In that case, Y1 sends Join messages for any of C's 2736 active (S,G) groups to Y2 while including its own LLA as the source 2737 address. This causes Y2 to include Y1 in the multicast forwarding 2738 tree during the interim time that Y1's symmetric neighbor cache entry 2739 for C is in the DEPARTED state. At the same time, Y1 sends an 2740 unsolicited NA message to X with an OMNI option and TLLAO with 2741 ifIndex-tuple set to 0 and a release indication to cause X to release 2742 its asymmetric neighbor cache entry. X then sends a new Join message 2743 to S via the spanning tree and re-initiates route optimization the 2744 same as if it were receiving a fresh Join message from a node on a 2745 downstream link. 2747 3.17.2. Any-Source Multicast (ASM) 2749 When an ROS X acting as a PIM router receives a Join/Prune from a 2750 node on its downstream interfaces containing one or more (*,G) pairs, 2751 it updates its Multicast Routing Information Base (MRIB) accordingly. 2752 X then forwards a copy of the message to the Rendezvous Point (RP) R 2753 for each G over the spanning tree. X uses its own LLA as the source 2754 address and ALL-PIM-ROUTERS as the destination address, then 2755 encapsulates each message in a SPAN header with source address set to 2756 the ULA of X and destination address set to R, then sends the message 2757 into the spanning tree. At the same time, if the message was a Join 2758 X initiates NS/NA route optimization the same as for the SSM case 2759 discussed in Section 3.17.1. 2761 For each source S that sends multicast traffic to group G via R, the 2762 Proxy/Server Z* for the Client that aggregates S encapsulates the 2763 packets in PIM Register messages and forwards them to R via the 2764 spanning tree, which may then elect to send a PIM Join to Z*. This 2765 will result in an (S,G) tree rooted at Z* with R as the next hop so 2766 that R will begin to receive two copies of the packet; one native 2767 copy from the (S, G) tree and a second copy from the pre-existing (*, 2768 G) tree that still uses PIM Register encapsulation. R can then issue 2769 a PIM Register-stop message to suppress the Register-encapsulated 2770 stream. At some later time, if C moves to a new Proxy/Server Z*, it 2771 resumes sending packets via PIM Register encapsulation via the new 2772 Z*. 2774 At the same time, as multicast listeners discover individual S's for 2775 a given G, they can initiate an (S,G) Join for each S under the same 2776 procedures discussed in Section 3.17.1. Once the (S,G) tree is 2777 established, the listeners can send (S, G) Prune messages to R so 2778 that multicast packets for group G sourced by S will only be 2779 delivered via the (S, G) tree and not from the (*, G) tree rooted at 2780 R. All mobility considerations discussed for SSM apply. 2782 3.17.3. Bi-Directional PIM (BIDIR-PIM) 2784 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 2785 approach to ASM that treats the Rendezvous Point (RP) as a Designated 2786 Forwarder (DF). Further considerations for BIDIR-PIM are out of 2787 scope. 2789 3.18. Operation over Multiple AERO Links (VLANs) 2791 An AERO Client can connect to multiple AERO links the same as for any 2792 data link service. In that case, the Client maintains a distinct 2793 AERO interface for each link, e.g., 'aero0' for the first link, 2794 'aero1' for the second, 'aero2' for the third, etc. Each AERO link 2795 would include its own distinct set of Bridges, Servers and Proxys, 2796 thereby providing redundancy in case of failures. 2798 The Bridges, Servers and Proxys on each AERO link can assign AERO and 2799 ULAs that use the same or different numberings from those on other 2800 links. Since the links are mutually independent there is no 2801 requirement for avoiding inter-link address duplication, e.g., the 2802 same LLA such as fe80::1000 could be used to number distinct nodes 2803 that connect to different AERO links. 2805 Each AERO link could utilize the same or different ANET connections. 2806 The links can be distinguished at the link-layer via the SSP in a 2807 similar fashion as for Virtual Local Area Network (VLAN) tagging 2808 (e.g., IEEE 802.1Q) and/or through assignment of distinct sets of 2809 MSPs on each link. This gives rise to the opportunity for supporting 2810 multiple redundant networked paths, where each VLAN is distinguished 2811 by a different SRT color (see: Section 3.2.6). In particular, the 2812 Client can tag its RS messages with the appropriate label to cause 2813 the network to select the desired VLAN. 2815 The Client's IP layer can select the outgoing AERO interface 2816 appropriate for a given traffic profile while (in the reverse 2817 direction) correspondent nodes must have some way of steering their 2818 packets destined to a target via the correct AERO link. 2820 In a first alternative, if each AERO link services different MSPs, 2821 then the Client can receive a distinct MNP from each of the links. 2822 IP routing will therefore assure that the correct Red/Green/Blue/etc. 2823 network is used for both outbound and inbound traffic. This can be 2824 accomplished using existing technologies and approaches, and without 2825 requiring any special supporting code in correspondent nodes or 2826 Bridges. 2828 In a second alternative, if each AERO link services the same MSP(s) 2829 then each link could assign a distinct "AERO Link Anycast" address 2830 that is configured by all Bridges on the link. Correspondent nodes 2831 can then perform Segment Routing to select the correct SRT, which 2832 will then direct the packet over multiple hops to the target. 2834 3.19. DNS Considerations 2836 AERO Client MNs and INET correspondent nodes consult the Domain Name 2837 System (DNS) the same as for any Internetworking node. When 2838 correspondent nodes and Client MNs use different IP protocol versions 2839 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 2840 A records for IPv4 address mappings to MNs which must then be 2841 populated in Relay NAT64 mapping caches. In that way, an IPv4 2842 correspondent node can send packets to the IPv4 address mapping of 2843 the target MN, and the Relay will translate the IPv4 header and 2844 destination address into an IPv6 header and IPv6 destination address 2845 of the MN. 2847 When an AERO Client registers with an AERO Server, the Server can 2848 return the address(es) of DNS servers in RDNSS options [RFC6106]. 2849 The DNS server provides the IP addresses of other MNs and 2850 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 2852 3.20. Transition Considerations 2854 SPAN encapsulation ensures that dissimilar INET partitions can be 2855 joined into a single unified AERO link, even though the partitions 2856 themselves may have differing protocol versions and/or incompatible 2857 addressing plans. However, a commonality can be achieved by 2858 incrementally distributing globally routable (i.e., native) IP 2859 prefixes to eventually reach all nodes (both mobile and fixed) in all 2860 AERO link segments. This can be accomplished by incrementally 2861 deploying AERO Relays on each INET partition, with each Relay 2862 distributing its MNPs and/or discovering non-MNP prefixes on its INET 2863 links. 2865 This gives rise to the opportunity to eventually distribute native IP 2866 addresses to all nodes, and to present a unified AERO link view even 2867 if the INET partitions remain in their current protocol and 2868 addressing plans. In that way, the AERO link can serve the dual 2869 purpose of providing a mobility/multilink service and a transition 2870 service. Or, if an INET partition is transitioned to a native IP 2871 protocol version and addressing scheme that is compatible with the 2872 AERO link MNP-based addressing scheme, the partition and AERO link 2873 can be joined by Relays. 2875 Relays that connect INETs/EUNs with dissimilar IP protocol versions 2876 may need to employ a network address and protocol translation 2877 function such as NAT64[RFC6146]. 2879 3.21. Detecting and Reacting to Server and Bridge Failures 2881 In environments where rapid failure recovery is required, Servers and 2882 Bridges SHOULD use Bidirectional Forwarding Detection (BFD) 2883 [RFC5880]. Nodes that use BFD can quickly detect and react to 2884 failures so that cached information is re-established through 2885 alternate nodes. BFD control messaging is carried only over well- 2886 connected ground domain networks (i.e., and not low-end radio links) 2887 and can therefore be tuned for rapid response. 2889 Servers and Bridges maintain BFD sessions in parallel with their BGP 2890 peerings. If a Server or Bridge fails, BGP peers will quickly re- 2891 establish routes through alternate paths the same as for common BGP 2892 deployments. Similarly, Proxys maintain BFD sessions with their 2893 associated Bridges even though they do not establish BGP peerings 2894 with them. 2896 Proxys SHOULD use proactive NUD for Servers for which there are 2897 currently active ANET Clients in a manner that parallels BFD, i.e., 2898 by sending unicast NS messages in rapid succession to receive 2899 solicited NA messages. When the Proxy is also sending RS messages on 2900 behalf of ANET Clients, the RS/RA messaging can be considered as 2901 equivalent hints of forward progress. This means that the Proxy need 2902 not also send a periodic NS if it has already sent an RS within the 2903 same period. If a Server fails, the Proxy will cease to receive 2904 advertisements and can quickly inform Clients of the outage by 2905 sending multicast RA messages on the ANET interface. 2907 The Proxy sends multicast RA messages with source address set to the 2908 Server's address, destination address set to All-Nodes multicast, and 2909 Router Lifetime set to 0. The Proxy SHOULD send 2910 MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated by small delays 2911 [RFC4861]. Any Clients on the ANET interface that have been using 2912 the (now defunct) Server will receive the RA messages and associate 2913 with a new Server. 2915 3.22. AERO Clients on the Open Internet 2917 AERO Clients that connect to the open Internet via INET interfaces 2918 can establish a VPN or direct link to securely connect to a Server in 2919 a "tethered" arrangement with all of the Client's traffic transiting 2920 the Server. Alternatively, the Client can associate with an INET 2921 Server using UDP/IP encapsulation and asymmetric securing services as 2922 discussed in the following sections. 2924 When a Client's AERO interface enables an INET underlying interface, 2925 it first determines whether the interface is likely to be behind a 2926 NAT. For IPv4, the Client might assume it is on the open Internet if 2927 the INET address is not a special-use IPv4 address per [RFC3330]. 2928 Similarly for IPv6, the Client might assume it is on the open 2929 Internet if the INET address is not a link-local [RFC4291] or unique- 2930 local [RFC4193] IPv6 address. 2932 The Client then prepares a UDP/IP-encapsulated RS message with IPv6 2933 source address set to its LLA, with IPv6 destination set to All- 2934 Routers multicast and with an OMNI option with underlying interface 2935 parameters. If the Client believes that it is on the open Internet, 2936 it SHOULD also include an SLLAO set according to the address used for 2937 INET encapsulation (otherwise, it MAY omit the SLLAO). If the 2938 underlying address is IPv4, the Client includes a Teredo address 2939 [RFC4380] using the prefix fe80::/32 with the Server's IPv4 address, 2940 and with the IP address and Port Number used for INET encapsulation 2941 written in obfuscated form and with FMT set to '01' (INET) as 2942 discussed in Section 3.3. If the underlying interface address is 2943 IPv6, the Client instead includes the IPv6 address and Port number in 2944 obfuscated form and sets FMT to '11' (INET). The Client finally 2945 includes a Teredo Authentication option per [RFC4380] to provide 2946 message authentication, sets the UDP/IP source to its INET address 2947 and UDP port, sets the UDP/IP destination to the Server's INET 2948 address and the AERO service port number (8060), then sends the 2949 message to the Server. 2951 When the Server receives the RS, it authenticates the message and 2952 registers the Client's MNP and INET interface information according 2953 to the OMNI option parameters. If the RS message includes an SLLAO, 2954 the Server compares the encapsulation IP address and UDP port number 2955 with the (unobfuscated) SLLAO values. If the values are the same, 2956 the Server caches the Client's information as "INET" addresses 2957 meaning that the Client is likely to accept direct messages without 2958 requiring NAT traversal exchanges. If the values are different (or, 2959 if there was no SLLAO) the Server instead caches the Client's 2960 information as "NAT" addresses meaning that NAT traversal exchanges 2961 may be necessary. 2963 The Server then returns an RA message with IPv6 source and 2964 destination set corresponding to the addresses in the RS, and with a 2965 Teredo Authentication option. For IPv4, the Server also includes an 2966 IPv4 Teredo Origin option per [RFC4380] with the mapped and 2967 obfuscated IPv4 address and port number observed in the encapsulation 2968 headers. For IPv6, the Server instead includes an IPv6 Teredo Origin 2969 option per Figure 7 with the mapped and obfuscated observed IPv6 2970 address and port number (note that the value 0x02 in the second octet 2971 differentiates from other Teredo option types). 2973 +--------+--------+-----------------+ 2974 | 0x00 | 0x02 | Origin port # | 2975 +--------+--------+-----------------+ 2976 ~ Origin IPv6 address ~ 2977 +-----------------------------------+ 2979 Figure 7: IPv6 Teredo Origin Option 2981 When the Client receives the RA message, it compares the mapped IP 2982 address and port from the Teredo Origin option with its own address. 2983 If the addresses are the same, the Client assumes the open Internet / 2984 Cone NAT principle; if the addresses are different, the Client 2985 instead assumes that further Server qualification procedures are 2986 necessary to detect the type of NAT and proceeds according to 2987 standard Teredo procedures. 2989 After the Client has registered its INET interfaces in such RS/RA 2990 exchanges it sends periodic RS messages to receive fresh RA messages 2991 before the Router Lifetime received on each INET interface expires. 2992 The Client also maintains default routes via its Servers, i.e., the 2993 same as described in earlier sections. 2995 When the Client sends messages to target IP addresses, it also 2996 invokes route optimization per Section 3.14 using IPv6 ND address 2997 resolution messaging. The Client sends the NS(AR) message to the 2998 Server wrapped in a UDP/IP header with a Teredo Authentication option 2999 with the NS source address set to the Client's LLA and destination 3000 address set to the target solicited node multicast address. The 3001 Server authenticates the message and sends a corresponding NS(AR) 3002 message over the spanning tree the same as if it were the ROS, but 3003 with the SPAN source address set to the Server's ULA and destination 3004 set to the ULA of the target. When the ROR receives the NS(AR), it 3005 adds the Server's ULA and Client's LLA to the target's Report List, 3006 and returns an NA with OMNI and TLLAO information for the target. 3007 The Server then returns a UDP/IP encapsulated NA message with a 3008 Teredo Authentication option to the Client. 3010 Following route optimization, for targets in the same AERO link 3011 segment if the target's TLLAO addresss is on the open INET, the 3012 Client forwards data packets directly to the target INET address. If 3013 the target's TLLAO address is behind a NAT, the Client first 3014 establishes NAT state for the Link Layer Address using the "bubble" 3015 mechanisms specified in [RFC6081][RFC4380]. The Client continues to 3016 send data packets via its Server until NAT state is populated, then 3017 begins forwarding packets via the direct path through the NAT to the 3018 target. For targets in different AERO link segments, the Client 3019 inserts a Segment Routing header and forwards data packets to the 3020 Bridge that returned the NA message. 3022 The ROR may return uNAs via the Server if the target moves, and the 3023 Server will send corresponding Teredo Authentication-protected uNAs 3024 to the Client. The Client can also send "loopback" NS(NUD) messages 3025 to test forward path reachability even though there is no security 3026 association between the Client and the target. 3028 The Client sends Teredo UDP/IP encapsulated IPv6 packets no larger 3029 than 1280 bytes in one piece. In order to accommodate larger IPv6 3030 packets (up to the AERO interface 9180 MTU), the Client inserts a 3031 SPAN header with source set to its own ULA and destination set to the 3032 ULA of the target and uses IPv6 fragmentation according to 3033 Section 3.9. The Client then encapsulates each fragment in a UDP/IP 3034 header and sends the fragments to the next hop. 3036 3.22.1. Use of SEND and CGA 3038 In some environments, use of the Teredo Authentication option alone 3039 may be sufficient for assuring IPv6 ND message authentication between 3040 Clients and Servers. When additional protection is necessary, nodes 3041 should employ SEcure Neighbor Discovery (SEND) [RFC3971] with 3042 Cryptographically-Generated Addresses (CGA) [RFC3972]. 3044 When SEND/CGA are used, the Client prepares RS messages with its 3045 link-local CGA as the IPv6 source and All-Routers as the IPv6 3046 Destination, includes any SEND options and wraps the message in a 3047 SPAN header. The Client sets the SPAN source address to its own ULA 3048 and sets the SPAN destination address to the "All-Routers" ULA. The 3049 Client then wraps the RS message in UDP/IP headers according to the 3050 Teredo format and sends the message to the Server. 3052 When the Server receives the message, it first verifies the Teredo 3053 Authentication option (if present) then uses the SPAN source address 3054 to determine the MNP of the Client. The Server then processes the 3055 SEND options to authenticate the RS message and prepares an RA 3056 message response. The Server prepares the RA with its own link-local 3057 CGA and the CGA of the Client as the IPv6 source and destination, 3058 includes any SEND options and wraps the message in a SPAN header. 3059 The Server sets the SPAN source address to its own ULA and sets the 3060 SPAN destination address to the Client's ULA. The Server then wraps 3061 the RA message in UDP/IP headers according to the Teredo format and 3062 sends the message to the Client. Thereafter, the Client/Server send 3063 additional RS/RA messages to maintain their association and any NAT 3064 state. 3066 The Client and Server also may exchange NS/NA messages using their 3067 own CGA as the source and with SPAN encapsulation as above. When a 3068 Client sends an NS(AR), it sets the IPv6 source to its CGA and sets 3069 the IPv6 destination to the Solicited-Node Multicast address of the 3070 target. The Client then wraps the message in a SPAN header with its 3071 own ULA as the source and the ULA of the target as the destination 3072 and sends it to the Server. The Server authenticates the message, 3073 then changes the IPv6 source address to the Client's LLA, removes the 3074 SEND options, and sends a corresponding NS(AR) into the spanning 3075 tree. When the Server receives the corresponding SPAN-encapsulated 3076 NA, it changes the IPv6 destination address to the Client's CGA, 3077 inserts SEND options, then wraps the message in UDP/IP headers and 3078 sends it to the Client. 3080 When a Client sends a uNA, it sets the IPv6 source address to its own 3081 CGA and sets the IPv6 destination address to All-Nodes multicast, 3082 includes SEND options, wraps the message in SPAN and UDP/IP headers 3083 and sends the message to the Server. The Server authenticates the 3084 message, then changes the IPv6 address to the Client's LLA, removes 3085 the SEND options and forwards the message the same as discussed in 3086 Section 3.16.1. In the reverse direction, when the Server forwards a 3087 uNA to the Client, it changes the IPv6 address to its own CGA and 3088 inserts SEND options then forwards the message to the Client. 3090 When a Client sends an NS(NUD), it sets both the IPv6 source and 3091 destination address to its own LLA, wraps the message in a SPAN 3092 header and UDP/IP headers, then sends the message directly to the 3093 peer which will loop the message back. In this case alone, the 3094 Client does not use the Server as a trust broker for forwarding the 3095 ND message. 3097 3.23. Time-Varying MNPs 3099 In some use cases, it is desirable, beneficial and efficient for the 3100 Client to receive a constant MNP that travels with the Client 3101 wherever it moves. For example, this would allow air traffic 3102 controllers to easily track aircraft, etc. In other cases, however 3103 (e.g., intelligent transportation systems), the MN may be willing to 3104 sacrifice a modicum of efficiency in order to have time-varying MNPs 3105 that can be changed every so often to defeat adversarial tracking. 3107 The DHCPv6-PD service offers a way for Clients that desire time- 3108 varying MNPs to obtain short-lived prefixes (e.g., on the order of a 3109 small number of minutes). In that case, the identity of the Client 3110 would not be bound to the MNP but rather the Client's identity would 3111 be bound to the DHCPv6 Device Unique Identifier (DUID) and used as 3112 the seed for Prefix Delegation. The Client would then be obligated 3113 to renumber its internal networks whenever its MNP (and therefore 3114 also its LLA) changes. This should not present a challenge for 3115 Clients with automated network renumbering services, however presents 3116 limits for the durations of ongoing sessions that would prefer to use 3117 a constant address. 3119 4. Implementation Status 3121 An AERO implementation based on OpenVPN (https://openvpn.net/) was 3122 announced on the v6ops mailing list on January 10, 2018 and an 3123 initial public release of the AERO proof-of-concept source code was 3124 announced on the intarea mailing list on August 21, 2015. 3126 As of 4/1/2020, more recent updated implementations are under 3127 internal development and testing with plans to release in the near 3128 future. 3130 5. IANA Considerations 3132 The IANA has assigned a 4-octet Private Enterprise Number "45282" for 3133 AERO in the "enterprise-numbers" registry. 3135 The IANA has assigned the UDP port number "8060" for an earlier 3136 experimental version of AERO [RFC6706]. This document obsoletes 3137 [RFC6706] and claims the UDP port number "8060" for all future use. 3139 The IANA is instructed to assign an IPv6 Routing Type code (TBD) in 3140 the ipv6-parameters registry [RFC5871]. 3142 No further IANA actions are required. 3144 6. Security Considerations 3146 AERO Bridges configure secured tunnels with AERO Servers, Realys and 3147 Proxys within their local AERO link segments. Applicable secured 3148 tunnel alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 3149 [RFC6347], WireGuard, etc. The AERO Bridges of all AERO link 3150 segments in turn configure secured tunnels for their neighboring AERO 3151 Bridges in a spanning tree topology. Therefore, control messages 3152 exchanged between any pair of AERO link neighbors on the spanning 3153 tree are already secured. 3155 AERO Servers, Relays and Proxys targeted by a route optimization may 3156 also receive data packets directly from arbitrary nodes in INET 3157 partitions instead of via the spanning tree. For INET partitions 3158 that apply effective ingress filtering to defeat source address 3159 spoofing, the simple data origin authentication procedures in 3160 Section 3.8 can be applied. 3162 For INET partitions that cannot apply effective ingress filtering, 3163 the two options for securing communications include 1) disable route 3164 optimization so that all traffic is conveyed over secured tunnels, or 3165 2) enable on-demand secure tunnel creation between INET partition 3166 neighbors. Option 1) would result in longer routes than necessary 3167 and traffic concentration on critical infrastructure elements. 3168 Option 2) could be coordinated by establishing a secured tunnel on- 3169 demand instead of performing an NS/NA exchange in the route 3170 optimization procedures. Procedures for establishing on-demand 3171 secured tunnels are out of scope. 3173 AERO Clients that connect to secured ANETs need not apply security to 3174 their ND messages, since the messages will be intercepted by a 3175 perimeter Proxy that applies security on its INET-facing interface. 3176 AERO Clients connected to the open INET can use symmetric network 3177 and/or transport layer security services such as VPNs or can by some 3178 other means establish a direct link. When a VPN or direct link may 3179 be impractical, however, an asymmetric security service such as 3180 SEcure Neighbor Discovery (SEND) [RFC3971] with Cryptographically 3181 Generated Addresses (CGAs) [RFC3972] and/or the Teredo Authentication 3182 option [RFC4380] may be necessary. 3184 Application endpoints SHOULD use application-layer security services 3185 such as TLS/SSL, DTLS or SSH [RFC4251] to assure the same level of 3186 protection as for critical secured Internet services. AERO Clients 3187 that require host-based VPN services SHOULD use symmetric network 3188 and/or transport layer security services such as IPsec, TLS/SSL, 3189 DTLS, etc. AERO Proxys and Servers can also provide a network-based 3190 VPN service on behalf of the Client, e.g., if the Client is located 3191 within a secured enclave and cannot establish a VPN on its own 3192 behalf. 3194 AERO Servers and Bridges present targets for traffic amplification 3195 Denial of Service (DoS) attacks. This concern is no different than 3196 for widely-deployed VPN security gateways in the Internet, where 3197 attackers could send spoofed packets to the gateways at high data 3198 rates. This can be mitigated by connecting Servers and Bridges over 3199 dedicated links with no connections to the Internet and/or when 3200 connections to the Internet are only permitted through well-managed 3201 firewalls. Traffic amplification DoS attacks can also target an AERO 3202 Client's low data rate links. This is a concern not only for Clients 3203 located on the open Internet but also for Clients in secured 3204 enclaves. AERO Servers and Proxys can institute rate limits that 3205 protect Clients from receiving packet floods that could DoS low data 3206 rate links. 3208 AERO Relays must implement ingress filtering to avoid a spoofing 3209 attack in which spurious messages with ULA addresses are injected 3210 into an AERO link from an outside attacker. AERO Clients MUST ensure 3211 that their connectivity is not used by unauthorized nodes on their 3212 EUNs to gain access to a protected network, i.e., AERO Clients that 3213 act as routers MUST NOT provide routing services for unauthorized 3214 nodes. (This concern is no different than for ordinary hosts that 3215 receive an IP address delegation but then "share" the address with 3216 other nodes via some form of Internet connection sharing such as 3217 tethering.) 3219 The MAP list MUST be well-managed and secured from unauthorized 3220 tampering, even though the list contains only public information. 3221 The MAP list can be conveyed to the Client in a similar fashion as in 3222 [RFC5214] (e.g., through layer 2 data link login messaging, secure 3223 upload of a static file, DNS lookups, etc.). 3225 Although public domain and commercial SEND implementations exist, 3226 concerns regarding the strength of the cryptographic hash algorithm 3227 have been documented [RFC6273] [RFC4982]. 3229 SRH authentication facilities are specified in [RFC8754]. 3231 Security considerations for accepting link-layer ICMP messages and 3232 reflected packets are discussed throughout the document. 3234 7. Acknowledgements 3236 Discussions in the IETF, aviation standards communities and private 3237 exchanges helped shape some of the concepts in this work. 3238 Individuals who contributed insights include Mikael Abrahamsson, Mark 3239 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 3240 Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian Farrel, Nick Green, 3241 Sri Gundavelli, Brian Haberman, Bernhard Haindl, Joel Halpern, Tom 3242 Herbert, Sascha Hlusiak, Lee Howard, Zdenek Jaron, Andre Kostur, 3243 Hubert Kuenig, Ted Lemon, Andy Malis, Satoru Matsushima, Tomek 3244 Mrugalski, Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal 3245 Skorepa, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, Lloyd 3246 Wood and James Woodyatt. Members of the IESG also provided valuable 3247 input during their review process that greatly improved the document. 3248 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 3249 for their shepherding guidance during the publication of the AERO 3250 first edition. 3252 This work has further been encouraged and supported by Boeing 3253 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 3254 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 3255 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 3256 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 3257 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 3258 Gene MacLean III, Rob Muszkiewicz, Sean O'Sullivan, Vijay 3259 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 3260 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 3261 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 3262 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 3263 Benson, Katie Tran and Eric Yeh are especially acknowledged for 3264 implementing the AERO functions as extensions to the public domain 3265 OpenVPN distribution. 3267 Earlier works on NBMA tunneling approaches are found in 3268 [RFC2529][RFC5214][RFC5569]. 3270 Many of the constructs presented in this second edition of AERO are 3271 based on the author's earlier works, including: 3273 o The Internet Routing Overlay Network (IRON) 3274 [RFC6179][I-D.templin-ironbis] 3276 o Virtual Enterprise Traversal (VET) 3277 [RFC5558][I-D.templin-intarea-vet] 3279 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 3280 [RFC5320][I-D.templin-intarea-seal] 3282 o AERO, First Edition [RFC6706] 3284 Note that these works cite numerous earlier efforts that are not also 3285 cited here due to space limitations. The authors of those earlier 3286 works are acknowledged for their insights. 3288 This work is aligned with the NASA Safe Autonomous Systems Operation 3289 (SASO) program under NASA contract number NNA16BD84C. 3291 This work is aligned with the FAA as per the SE2025 contract number 3292 DTFAWA-15-D-00030. 3294 This work is aligned with the Boeing Commercial Airplanes (BCA) 3295 Internet of Things (IoT) and autonomy programs. 3297 This work is aligned with the Boeing Information Technology (BIT) 3298 MobileNet program. 3300 8. References 3302 8.1. Normative References 3304 [I-D.templin-6man-omni-interface] 3305 Templin, F. and T. Whyman, "Transmission of IPv6 Packets 3306 over Overlay Multilink Network (OMNI) Interfaces", draft- 3307 templin-6man-omni-interface-21 (work in progress), May 3308 2020. 3310 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3311 DOI 10.17487/RFC0791, September 1981, 3312 . 3314 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 3315 RFC 792, DOI 10.17487/RFC0792, September 1981, 3316 . 3318 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3319 Requirement Levels", BCP 14, RFC 2119, 3320 DOI 10.17487/RFC2119, March 1997, 3321 . 3323 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3324 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3325 December 1998, . 3327 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 3328 "Definition of the Differentiated Services Field (DS 3329 Field) in the IPv4 and IPv6 Headers", RFC 2474, 3330 DOI 10.17487/RFC2474, December 1998, 3331 . 3333 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3334 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3335 DOI 10.17487/RFC3971, March 2005, 3336 . 3338 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3339 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3340 . 3342 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3343 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3344 November 2005, . 3346 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3347 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3348 . 3350 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 3351 Network Address Translations (NATs)", RFC 4380, 3352 DOI 10.17487/RFC4380, February 2006, 3353 . 3355 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3356 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3357 DOI 10.17487/RFC4861, September 2007, 3358 . 3360 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3361 Address Autoconfiguration", RFC 4862, 3362 DOI 10.17487/RFC4862, September 2007, 3363 . 3365 [RFC5175] Haberman, B., Ed. and R. Hinden, "IPv6 Router 3366 Advertisement Flags Option", RFC 5175, 3367 DOI 10.17487/RFC5175, March 2008, 3368 . 3370 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 3371 DOI 10.17487/RFC6081, January 2011, 3372 . 3374 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3375 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3376 May 2017, . 3378 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3379 (IPv6) Specification", STD 86, RFC 8200, 3380 DOI 10.17487/RFC8200, July 2017, 3381 . 3383 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3384 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3385 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3386 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3387 . 3389 8.2. Informative References 3391 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3392 2016. 3394 [I-D.ietf-6man-segment-routing-header] 3395 Filsfils, C., Dukes, D., Previdi, S., Leddy, J., 3396 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3397 (SRH)", draft-ietf-6man-segment-routing-header-26 (work in 3398 progress), October 2019. 3400 [I-D.ietf-dmm-distributed-mobility-anchoring] 3401 Chan, A., Wei, X., Lee, J., Jeon, S., and C. Bernardos, 3402 "Distributed Mobility Anchoring", draft-ietf-dmm- 3403 distributed-mobility-anchoring-15 (work in progress), 3404 March 2020. 3406 [I-D.ietf-intarea-gue] 3407 Herbert, T., Yong, L., and O. Zia, "Generic UDP 3408 Encapsulation", draft-ietf-intarea-gue-09 (work in 3409 progress), October 2019. 3411 [I-D.ietf-intarea-gue-extensions] 3412 Herbert, T., Yong, L., and F. Templin, "Extensions for 3413 Generic UDP Encapsulation", draft-ietf-intarea-gue- 3414 extensions-06 (work in progress), March 2019. 3416 [I-D.ietf-intarea-tunnels] 3417 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3418 Architecture", draft-ietf-intarea-tunnels-10 (work in 3419 progress), September 2019. 3421 [I-D.ietf-rtgwg-atn-bgp] 3422 Templin, F., Saccone, G., Dawra, G., Lindem, A., and V. 3423 Moreno, "A Simple BGP-based Mobile Routing System for the 3424 Aeronautical Telecommunications Network", draft-ietf- 3425 rtgwg-atn-bgp-05 (work in progress), January 2020. 3427 [I-D.templin-6man-dhcpv6-ndopt] 3428 Templin, F., "A Unified Stateful/Stateless Configuration 3429 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-09 3430 (work in progress), January 2020. 3432 [I-D.templin-intarea-grefrag] 3433 Templin, F., "GRE Tunnel Level Fragmentation", draft- 3434 templin-intarea-grefrag-04 (work in progress), July 2016. 3436 [I-D.templin-intarea-seal] 3437 Templin, F., "The Subnetwork Encapsulation and Adaptation 3438 Layer (SEAL)", draft-templin-intarea-seal-68 (work in 3439 progress), January 2014. 3441 [I-D.templin-intarea-vet] 3442 Templin, F., "Virtual Enterprise Traversal (VET)", draft- 3443 templin-intarea-vet-40 (work in progress), May 2013. 3445 [I-D.templin-ironbis] 3446 Templin, F., "The Interior Routing Overlay Network 3447 (IRON)", draft-templin-ironbis-16 (work in progress), 3448 March 2014. 3450 [I-D.templin-v6ops-pdhost] 3451 Templin, F., "IPv6 Prefix Delegation and Multi-Addressing 3452 Models", draft-templin-v6ops-pdhost-25 (work in progress), 3453 January 2020. 3455 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3457 [RFC1035] Mockapetris, P., "Domain names - implementation and 3458 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3459 November 1987, . 3461 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 3462 Communication Layers", STD 3, RFC 1122, 3463 DOI 10.17487/RFC1122, October 1989, 3464 . 3466 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 3467 DOI 10.17487/RFC1191, November 1990, 3468 . 3470 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3471 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3472 . 3474 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3475 DOI 10.17487/RFC2003, October 1996, 3476 . 3478 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3479 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3480 . 3482 [RFC2492] Armitage, G., Schulter, P., and M. Jork, "IPv6 over ATM 3483 Networks", RFC 2492, DOI 10.17487/RFC2492, January 1999, 3484 . 3486 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3487 Domains without Explicit Tunnels", RFC 2529, 3488 DOI 10.17487/RFC2529, March 1999, 3489 . 3491 [RFC2764] Gleeson, B., Lin, A., Heinanen, J., Armitage, G., and A. 3492 Malis, "A Framework for IP Based Virtual Private 3493 Networks", RFC 2764, DOI 10.17487/RFC2764, February 2000, 3494 . 3496 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 3497 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 3498 DOI 10.17487/RFC2784, March 2000, 3499 . 3501 [RFC2890] Dommety, G., "Key and Sequence Number Extensions to GRE", 3502 RFC 2890, DOI 10.17487/RFC2890, September 2000, 3503 . 3505 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", 3506 RFC 2923, DOI 10.17487/RFC2923, September 2000, 3507 . 3509 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3510 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3511 . 3513 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3514 of Explicit Congestion Notification (ECN) to IP", 3515 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3516 . 3518 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 3519 DOI 10.17487/RFC3330, September 2002, 3520 . 3522 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3523 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3524 DOI 10.17487/RFC3810, June 2004, 3525 . 3527 [RFC3819] Karn, P., Ed., Bormann, C., Fairhurst, G., Grossman, D., 3528 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 3529 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 3530 RFC 3819, DOI 10.17487/RFC3819, July 2004, 3531 . 3533 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 3534 for IPv6 Hosts and Routers", RFC 4213, 3535 DOI 10.17487/RFC4213, October 2005, 3536 . 3538 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3539 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3540 January 2006, . 3542 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3543 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3544 DOI 10.17487/RFC4271, January 2006, 3545 . 3547 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3548 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3549 2006, . 3551 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3552 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3553 December 2005, . 3555 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3556 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3557 2006, . 3559 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3560 Control Message Protocol (ICMPv6) for the Internet 3561 Protocol Version 6 (IPv6) Specification", STD 89, 3562 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3563 . 3565 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3566 Protocol (LDAP): The Protocol", RFC 4511, 3567 DOI 10.17487/RFC4511, June 2006, 3568 . 3570 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3571 "Considerations for Internet Group Management Protocol 3572 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3573 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3574 . 3576 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3577 "Internet Group Management Protocol (IGMP) / Multicast 3578 Listener Discovery (MLD)-Based Multicast Forwarding 3579 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3580 August 2006, . 3582 [RFC4607] Holbrook, H. and B. Cain, "Source-Specific Multicast for 3583 IP", RFC 4607, DOI 10.17487/RFC4607, August 2006, 3584 . 3586 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 3587 Errors at High Data Rates", RFC 4963, 3588 DOI 10.17487/RFC4963, July 2007, 3589 . 3591 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3592 Algorithms in Cryptographically Generated Addresses 3593 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3594 . 3596 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3597 "Bidirectional Protocol Independent Multicast (BIDIR- 3598 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3599 . 3601 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3602 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3603 DOI 10.17487/RFC5214, March 2008, 3604 . 3606 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3607 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3608 February 2010, . 3610 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3611 Route Optimization Requirements for Operational Use in 3612 Aeronautics and Space Exploration Mobile Networks", 3613 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3614 . 3616 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3617 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3618 . 3620 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3621 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3622 January 2010, . 3624 [RFC5871] Arkko, J. and S. Bradner, "IANA Allocation Guidelines for 3625 the IPv6 Routing Header", RFC 5871, DOI 10.17487/RFC5871, 3626 May 2010, . 3628 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 3629 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 3630 . 3632 [RFC5991] Thaler, D., Krishnan, S., and J. Hoagland, "Teredo 3633 Security Updates", RFC 5991, DOI 10.17487/RFC5991, 3634 September 2010, . 3636 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3637 "IPv6 Router Advertisement Options for DNS Configuration", 3638 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3639 . 3641 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3642 NAT64: Network Address and Protocol Translation from IPv6 3643 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3644 April 2011, . 3646 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3647 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3648 . 3650 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3651 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3652 DOI 10.17487/RFC6221, May 2011, 3653 . 3655 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3656 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3657 DOI 10.17487/RFC6273, June 2011, 3658 . 3660 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3661 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3662 January 2012, . 3664 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3665 for Equal Cost Multipath Routing and Link Aggregation in 3666 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3667 . 3669 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3670 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3671 . 3673 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID Field", 3674 RFC 6864, DOI 10.17487/RFC6864, February 2013, 3675 . 3677 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 3678 UDP Checksums for Tunneled Packets", RFC 6935, 3679 DOI 10.17487/RFC6935, April 2013, 3680 . 3682 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 3683 for the Use of IPv6 UDP Datagrams with Zero Checksums", 3684 RFC 6936, DOI 10.17487/RFC6936, April 2013, 3685 . 3687 [RFC7269] Chen, G., Cao, Z., Xie, C., and D. Binet, "NAT64 3688 Deployment Options and Experience", RFC 7269, 3689 DOI 10.17487/RFC7269, June 2014, 3690 . 3692 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3693 Korhonen, "Requirements for Distributed Mobility 3694 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3695 . 3697 [RFC7421] Carpenter, B., Ed., Chown, T., Gont, F., Jiang, S., 3698 Petrescu, A., and A. Yourtchenko, "Analysis of the 64-bit 3699 Boundary in IPv6 Addressing", RFC 7421, 3700 DOI 10.17487/RFC7421, January 2015, 3701 . 3703 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3704 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3705 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3706 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3707 2016, . 3709 [RFC8086] Yong, L., Ed., Crabbe, E., Xu, X., and T. Herbert, "GRE- 3710 in-UDP Encapsulation", RFC 8086, DOI 10.17487/RFC8086, 3711 March 2017, . 3713 [RFC8138] Thubert, P., Ed., Bormann, C., Toutain, L., and R. Cragie, 3714 "IPv6 over Low-Power Wireless Personal Area Network 3715 (6LoWPAN) Routing Header", RFC 8138, DOI 10.17487/RFC8138, 3716 April 2017, . 3718 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 3719 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 3720 DOI 10.17487/RFC8201, July 2017, 3721 . 3723 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3724 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3725 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3726 July 2018, . 3728 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3729 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3730 . 3732 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 3733 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3734 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 3735 . 3737 Appendix A. AERO Alternate Encapsulations 3739 When GUE encapsulation is not needed, AERO can use common 3740 encapsulations such as IP-in-IP [RFC2003][RFC2473][RFC4213], Generic 3741 Routing Encapsulation (GRE) [RFC2784][RFC2890] and others. The 3742 encapsulation is therefore only differentiated from non-AERO tunnels 3743 through the application of AERO control messaging and not through, 3744 e.g., a well-known UDP port number. 3746 As for GUE encapsulation, alternate AERO encapsulation formats may 3747 require encapsulation layer fragmentation. For simple IP-in-IP 3748 encapsulation, an IPv6 fragment header is inserted directly between 3749 the inner and outer IP headers when needed, i.e., even if the outer 3750 header is IPv4. The IPv6 Fragment Header is identified to the outer 3751 IP layer by its IP protocol number, and the Next Header field in the 3752 IPv6 Fragment Header identifies the inner IP header version. For GRE 3753 encapsulation, a GRE fragment header is inserted within the GRE 3754 header [I-D.templin-intarea-grefrag]. 3756 Figure 8 shows the AERO IP-in-IP encapsulation format before any 3757 fragmentation is applied: 3759 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3760 | Outer IPv4 Header | | Outer IPv6 Header | 3761 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3762 |IPv6 Frag Header (optional)| |IPv6 Frag Header (optional)| 3763 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3764 | Inner IP Header | | Inner IP Header | 3765 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3766 | | | | 3767 ~ ~ ~ ~ 3768 ~ Inner Packet Body ~ ~ Inner Packet Body ~ 3769 ~ ~ ~ ~ 3770 | | | | 3771 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3773 Minimal Encapsulation in IPv4 Minimal Encapsulation in IPv6 3775 Figure 8: Minimal Encapsulation Format using IP-in-IP 3777 Figure 9 shows the AERO GRE encapsulation format before any 3778 fragmentation is applied: 3780 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3781 | Outer IP Header | 3782 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3783 | GRE Header | 3784 | (with checksum, key, etc..) | 3785 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3786 | GRE Fragment Header (optional)| 3787 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3788 | Inner IP Header | 3789 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3790 | | 3791 ~ ~ 3792 ~ Inner Packet Body ~ 3793 ~ ~ 3794 | | 3795 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3797 Figure 9: Minimal Encapsulation Using GRE 3799 Alternate encapsulation may be preferred in environments where GUE 3800 encapsulation would add unnecessary overhead. For example, certain 3801 low-bandwidth wireless data links may benefit from a reduced 3802 encapsulation overhead. 3804 GUE encapsulation can traverse network paths that are inaccessible to 3805 non-UDP encapsulations, e.g., for crossing Network Address 3806 Translators (NATs). More and more, network middleboxes are also 3807 being configured to discard packets that include anything other than 3808 a well-known IP protocol such as UDP and TCP. It may therefore be 3809 necessary to determine the potential for middlebox filtering before 3810 enabling alternate encapsulation in a given environment. 3812 In addition to IP-in-IP, GRE and GUE, AERO can also use security 3813 encapsulations such as IPsec, TLS/SSL, DTLS, etc. In that case, AERO 3814 control messaging and route determination occur before security 3815 encapsulation is applied for outgoing packets and after security 3816 decapsulation is applied for incoming packets. 3818 AERO is especially well suited for use with VPN system encapsulations 3819 such as OpenVPN [OVPN]. 3821 Appendix B. Non-Normative Considerations 3823 AERO can be applied to a multitude of Internetworking scenarios, with 3824 each having its own adaptations. The following considerations are 3825 provided as non-normative guidance: 3827 B.1. Implementation Strategies for Route Optimization 3829 Route optimization as discussed in Section 3.14 results in the route 3830 optimization source (ROS) creating an asymmetric neighbor cache entry 3831 for the target neighbor. The neighbor cache entry is maintained for 3832 at most ReachableTime seconds and then deleted unless updated. In 3833 order to refresh the neighbor cache entry lifetime before the 3834 ReachableTime timer expires, the specification requires 3835 implementations to issue a new NS/NA exchange to reset ReachableTime 3836 while data packets are still flowing. However, the decision of when 3837 to initiate a new NS/NA exchange and to perpetuate the process is 3838 left as an implementation detail. 3840 One possible strategy may be to monitor the neighbor cache entry 3841 watching for data packets for (ReachableTime - 5) seconds. If any 3842 data packets have been sent to the neighbor within this timeframe, 3843 then send an NS to receive a new NA. If no data packets have been 3844 sent, wait for 5 additional seconds and send an immediate NS if any 3845 data packets are sent within this "expiration pending" 5 second 3846 window. If no additional data packets are sent within the 5 second 3847 window, delete the neighbor cache entry. 3849 The monitoring of the neighbor data packet traffic therefore becomes 3850 an asymmetric ongoing process during the neighbor cache entry 3851 lifetime. If the neighbor cache entry expires, future data packets 3852 will trigger a new NS/NA exchange while the packets themselves are 3853 delivered over a longer path until route optimization state is re- 3854 established. 3856 B.2. Implicit Mobility Management 3858 AERO interface neighbors MAY provide a configuration option that 3859 allows them to perform implicit mobility management in which no ND 3860 messaging is used. In that case, the Client only transmits packets 3861 over a single interface at a time, and the neighbor always observes 3862 packets arriving from the Client from the same link-layer source 3863 address. 3865 If the Client's underlying interface address changes (either due to a 3866 readdressing of the original interface or switching to a new 3867 interface) the neighbor immediately updates the neighbor cache entry 3868 for the Client and begins accepting and sending packets according to 3869 the Client's new address. This implicit mobility method applies to 3870 use cases such as cellphones with both WiFi and Cellular interfaces 3871 where only one of the interfaces is active at a given time, and the 3872 Client automatically switches over to the backup interface if the 3873 primary interface fails. 3875 B.3. Direct Underlying Interfaces 3877 When a Client's AERO interface is configured over a Direct interface, 3878 the neighbor at the other end of the Direct link can receive packets 3879 without any encapsulation. In that case, the Client sends packets 3880 over the Direct link according to QoS preferences. If the Direct 3881 interface has the highest QoS preference, then the Client's IP 3882 packets are transmitted directly to the peer without going through an 3883 ANET/INET. If other interfaces have higher QoS preferences, then the 3884 Client's IP packets are transmitted via a different interface, which 3885 may result in the inclusion of Proxys, Servers and Bridges in the 3886 communications path. Direct interfaces must be tested periodically 3887 for reachability, e.g., via NUD. 3889 B.4. Operation on AERO Links with /64 ASPs 3891 IPv6 AERO links typically have MSPs that aggregate many candidate 3892 MNPs of length /64 or shorter. However, in some cases it may be 3893 desirable to use AERO over links that have only a /64 MSP. This can 3894 be accommodated by treating all Clients on the AERO link as simple 3895 hosts that receive /128 prefix delegations. 3897 In that case, the Client sends an RS message to the Server the same 3898 as for ordinary AERO links. The Server responds with an RA message 3899 that includes one or more /128 prefixes (i.e., singleton addresses) 3900 that include the /64 MSP prefix along with an interface identifier 3901 portion to be assigned to the Client. The Client and Server then 3902 configure their LLAs based on the interface identifier portions of 3903 the /128s (i.e., the lower 64 bits) and not based on the /64 prefix 3904 (i.e., the upper 64 bits). 3906 For example, if the MSP for the host-only IPv6 AERO link is 3907 2001:db8:1000:2000::/64, each Client will receive one or more /128 3908 IPv6 prefix delegations such as 2001:db8:1000:2000::1/128, 3909 2001:db8:1000:2000::2/128, etc. When the Client receives the prefix 3910 delegations, it assigns the LLAs fe80::1, fe80::2, etc. to the AERO 3911 interface, and assigns the global IPv6 addresses (i.e., the /128s) to 3912 either the AERO interface or an internal virtual interface such as a 3913 loopback. In this arrangement, the Client conducts route 3914 optimization in the same sense as discussed in Section 3.14. 3916 This specification has applicability for nodes that act as a Client 3917 on an "upstream" AERO link, but also act as a Server on "downstream" 3918 AERO links. More specifically, if the node acts as a Client to 3919 receive a /64 prefix from the upstream AERO link it can then act as a 3920 Server to provision /128s to Clients on downstream AERO links. 3922 B.5. AERO Critical Infrastructure Considerations 3924 AERO Bridges can be either Commercial off-the Shelf (COTS) standard 3925 IP routers or virtual machines in the cloud. Bridges must be 3926 provisioned, supported and managed by the INET administrative 3927 authority, and connected to the Bridges of other INETs via inter- 3928 domain peerings. Cost for purchasing, configuring and managing 3929 Bridges is nominal even for very large AERO links. 3931 AERO Servers can be standard dedicated server platforms, but most 3932 often will be deployed as virtual machines in the cloud. The only 3933 requirements for Servers are that they can run the AERO user-level 3934 code and have at least one network interface connection to the INET. 3935 As with Bridges, Servers must be provisioned, supported and managed 3936 by the INET administrative authority. Cost for purchasing, 3937 configuring and managing Servers is nominal especially for virtual 3938 Servers hosted in the cloud. 3940 AERO Proxys are most often standard dedicated server platforms with 3941 one network interface connected to the ANET and a second interface 3942 connected to an INET. As with Servers, the only requirements are 3943 that they can run the AERO user-level code and have at least one 3944 interface connection to the INET. Proxys must be provisioned, 3945 supported and managed by the ANET administrative authority. Cost for 3946 purchasing, configuring and managing Proxys is nominal, and borne by 3947 the ANET administrative authority. 3949 AERO Relays can be any dedicated server or COTS router platform 3950 connected to INETs and/or EUNs. The Relay connects to the AERO link 3951 and engages in eBGP peering with one or more Bridges as a stub AS. 3952 The Relay then injects its MNPs and/or non-MNP prefixes into the BGP 3953 routing system, and provisions the prefixes to its downstream- 3954 attached networks. The Relay can perform ROS/ROR services the same 3955 as for any Server, and can route between the MNP and non-MNP address 3956 spaces. 3958 B.6. AERO Server Failure Implications 3960 AERO Servers may appear as a single point of failure in the 3961 architecture, but such is not the case since all Servers on the link 3962 provide identical services and loss of a Server does not imply 3963 immediate and/or comprehensive communication failures. Although 3964 Clients typically associate with a single Server at a time, Server 3965 failure is quickly detected and conveyed by Bidirectional Forward 3966 Detection (BFD) and/or proactive NUD allowing Clients to migrate to 3967 new Servers. 3969 If a Server fails, ongoing packet forwarding to Clients will continue 3970 by virtue of the asymmetric neighbor cache entries that have already 3971 been established in route optimization sources (ROSs). If a Client 3972 also experiences mobility events at roughly the same time the Server 3973 fails, unsolicited NA messages may be lost but proxy neighbor cache 3974 entries in the DEPARTED state will ensure that packet forwarding to 3975 the Client's new locations will continue for up to DepartTime 3976 seconds. 3978 If a Client is left without a Server for an extended timeframe (e.g., 3979 greater than ReachableTime seconds) then existing asymmetric neighbor 3980 cache entries will eventually expire and both ongoing and new 3981 communications will fail. The original source will continue to 3982 retransmit until the Client has established a new Server 3983 relationship, after which time continuous communications will resume. 3985 Therefore, providing many Servers on the link with high availability 3986 profiles provides resilience against loss of individual Servers and 3987 assurance that Clients can establish new Server relationships quickly 3988 in event of a Server failure. 3990 B.7. AERO Client / Server Architecture 3992 The AERO architectural model is client / server in the control plane, 3993 with route optimization in the data plane. The same as for common 3994 Internet services, the AERO Client discovers the addresses of AERO 3995 Servers and selects one Server to connect to. The AERO service is 3996 analogous to common Internet services such as google.com, yahoo.com, 3997 cnn.com, etc. However, there is only one AERO service for the link 3998 and all Servers provide identical services. 4000 Common Internet services provide differing strategies for advertising 4001 server addresses to clients. The strategy is conveyed through the 4002 DNS resource records returned in response to name resolution queries. 4003 As of January 2020 Internet-based 'nslookup' services were used to 4004 determine the following: 4006 o When a client resolves the domainname "google.com", the DNS always 4007 returns one A record (i.e., an IPv4 address) and one AAAA record 4008 (i.e., an IPv6 address). The client receives the same addresses 4009 each time it resolves the domainname via the same DNS resolver, 4010 but may receive different addresses when it resolves the 4011 domainname via different DNS resolvers. But, in each case, 4012 exactly one A and one AAAA record are returned. 4014 o When a client resolves the domainname "ietf.org", the DNS always 4015 returns one A record and one AAAA record with the same addresses 4016 regardless of which DNS resolver is used. 4018 o When a client resolves the domainname "yahoo.com", the DNS always 4019 returns a list of 4 A records and 4 AAAA records. Each time the 4020 client resolves the domainname via the same DNS resolver, the same 4021 list of addresses are returned but in randomized order (i.e., 4022 consistent with a DNS round-robin strategy). But, interestingly, 4023 the same addresses are returned (albeit in randomized order) when 4024 the domainname is resolved via different DNS resolvers. 4026 o When a client resolves the domainname "amazon.com", the DNS always 4027 returns a list of 3 A records and no AAAA records. As with 4028 "yahoo.com", the same three A records are returned from any 4029 worldwide Internet connection point in randomized order. 4031 The above example strategies show differing approaches to Internet 4032 resilience and service distribution offered by major Internet 4033 services. The Google approach exposes only a single IPv4 and a 4034 single IPv6 address to clients. Clients can then select whichever IP 4035 protocol version offers the best response, but will always use the 4036 same IP address according to the current Internet connection point. 4037 This means that the IP address offered by the network must lead to a 4038 highly-available server and/or service distribution point. In other 4039 words, resilience is predicated on high availability within the 4040 network and with no client-initiated failovers expected (i.e., it is 4041 all-or-nothing from the client's perspective). However, Google does 4042 provide for worldwide distributed service distribution by virtue of 4043 the fact that each Internet connection point responds with a 4044 different IPv6 and IPv4 address. The IETF approach is like google 4045 (all-or-nothing from the client's perspective), but provides only a 4046 single IPv4 or IPv6 address on a worldwide basis. This means that 4047 the addresses must be made highly-available at the network level with 4048 no client failover possibility, and if there is any worldwide service 4049 distribution it would need to be conducted by a network element that 4050 is reached via the IP address acting as a service distribution point. 4052 In contrast to the Google and IETF philosophies, Yahoo and Amazon 4053 both provide clients with a (short) list of IP addresses with Yahoo 4054 providing both IP protocol versions and Amazon as IPv4-only. The 4055 order of the list is randomized with each name service query 4056 response, with the effect of round-robin load balancing for service 4057 distribution. With a short list of addresses, there is still 4058 expectation that the network will implement high availability for 4059 each address but in case any single address fails the client can 4060 switch over to using a different address. The balance then becomes 4061 one of function in the network vs function in the end system. 4063 The same implications observed for common highly-available services 4064 in the Internet apply also to the AERO client/server architecture. 4065 When an AERO Client connects to one or more ANETs, it discovers one 4066 or more AERO Server addresses through the mechanisms discussed in 4067 earlier sections. Each Server address presumably leads to a fault- 4068 tolerant clustering arrangement such as supported by Linux-HA, 4069 Extended Virtual Synchrony or Paxos. Such an arrangement has 4070 precedence in common Internet service deployments in lightweight 4071 virtual machines without requiring expensive hardware deployment. 4072 Similarly, common Internet service deployments set service IP 4073 addresses on service distribution points that may relay requests to 4074 many different servers. 4076 For AERO, the expectation is that a combination of the Google/IETF 4077 and Yahoo/Amazon philosophies would be employed. The AERO Client 4078 connects to different ANET access points and can receive 1-2 Server 4079 LLAs at each point. It then selects one AERO Server address, and 4080 engages in RS/RA exchanges with the same Server from all ANET 4081 connections. The Client remains with this Server unless or until the 4082 Server fails, in which case it can switch over to an alternate 4083 Server. The Client can likewise switch over to a different Server at 4084 any time if there is some reason for it to do so. So, the AERO 4085 expectation is for a balance of function in the network and end 4086 system, with fault tolerance and resilience at both levels. 4088 Appendix C. Change Log 4090 << RFC Editor - remove prior to publication >> 4092 Changes from draft-templin-intarea-6706bis-48 to draft-templin- 4093 intrea-6706bis-49: 4095 o SPAN Anycast address and SBM/PBM concepts introduced. 4097 Changes from draft-templin-intarea-6706bis-47 to draft-templin- 4098 intrea-6706bis-48: 4100 o SEND/CGA. 4102 Changes from draft-templin-intarea-6706bis-46 to draft-templin- 4103 intrea-6706bis-47: 4105 o Major changes to align with Teredo, including changed AERO "Relay" 4106 to "Bridge", and changed AERO "Gateway" to "Relay". The term 4107 "[Rr]elay" now refers to exactly the same thing in both AERO and 4108 Teredo. 4110 o Changed to use Teredo message authentication instead of SEND. 4112 Changes from draft-templin-intarea-6706bis-42 to draft-templin- 4113 intrea-6706bis-43: 4115 o Segment Routing. 4117 Changes from draft-templin-intarea-6706bis-39 to draft-templin- 4118 intrea-6706bis-40: 4120 o Teredo. 4122 Changes from draft-templin-intarea-6706bis-38 to draft-templin- 4123 intrea-6706bis-39: 4125 o Major clarifications and simplifications of SPAN fragmentation/ 4126 reassembly. 4128 o Revised AERO address format to support prefix lengths up to 112. 4130 o New method for forming SPAN Client Prefixes and population in the 4131 routing system. 4133 o Updates RFC4443 to set a new value in the ICMP PTB Code field. 4135 Changes from draft-templin-intarea-6706bis-35 to draft-templin- 4136 intrea-6706bis-36: 4138 o Clients in the open Internet secured using SEND/CGA. 4140 Changes from draft-templin-intarea-6706bis-32 to draft-templin- 4141 intrea-6706bis-33: 4143 o Updated Proxy discussion with "point-to-multipoint" server 4144 coordination 4146 o Significant updates to Address Resolution and NUD to include 4147 correct addresses in messages 4149 o Differentiate between NS(AR) and NS(NUD) as their addresses and 4150 use cases differ. 4152 Changes from draft-templin-intarea-6706bis-30 to draft-templin- 4153 intrea-6706bis-31: 4155 o Added "advisory PTB messages" under FAA SE2025 contract number 4156 DTFAWA-15-D-00030. 4158 Changes from draft-templin-intarea-6706bis-29 to draft-templin- 4159 intrea-6706bis-30: 4161 o Deprecate "primary" concept. Now, RS/RA keepalives are maintained 4162 over *all* underlying interfaces (i.e., and not just one primary). 4164 Changes from draft-templin-intarea-6706bis-28 to draft-templin- 4165 intrea-6706bis-29: 4167 o Changed OMNI interface citation to "draft-templin-6man-omni- 4168 interface" 4170 o Changed SPAN Service Prefix to fd80::/10. 4172 o Changed S/TLLAO format to include 'S' bit for ifIndex 4173 corresponding to the underlying interface that is Source of ND 4174 message. 4176 o Updated Path MTU 4178 Changes from draft-templin-intarea-6706bis-27 to draft-templin- 4179 intrea-6706bis-28: 4181 o MTU and fragmentation. 4183 Changes from draft-templin-intarea-6706bis-26 to draft-templin- 4184 intrea-6706bis-27: 4186 o MTU and fragmentation. 4188 o SPAN Service Prefix set to fd00::/10 4190 o Client SPAN addresses defined. 4192 Changes from draft-templin-intarea-6706bis-25 to draft-templin- 4193 intrea-6706bis-26: 4195 o MTU and RA configuration information updated. 4197 Changes from draft-templin-intarea-6706bis-24 to draft-templin- 4198 intrea-6706bis-25: 4200 o Added concept of "primary" to allow for proxyed RS/RA over only 4201 selected underlying interfaces. 4203 o General Cleanup. 4205 Changes from draft-templin-intarea-6706bis-23 to draft-templin- 4206 intrea-6706bis-24: 4208 o OMNI interface spec now a normative reference. 4210 o Use REACHABLE_TIME as the nominal Router Lifetime to return in 4211 RAs. 4213 o General cleanup. 4215 Changes from draft-templin-intarea-6706bis-22 to draft-templin- 4216 intrea-6706bis-23: 4218 o Choice of using either RS/RA or unsolicited NA for old Server 4219 notification. 4221 o General cleanup. 4223 Changes from draft-templin-intarea-6706bis-21 to draft-templin- 4224 intrea-6706bis-22: 4226 o Tightened up text on Proxy. 4228 o Removed unnecessarily restrictive texts. 4230 o General cleanup. 4232 Changes from draft-templin-intarea-6706bis-20 to draft-templin- 4233 intrea-6706bis-21: 4235 o Clarified relationship between OMNI and S/TLLAO ifIndex-tuples. 4237 o Important text in Section 13.15.3 on Servers timing out Clients 4238 that have gone silent without sending a departure notification. 4240 o New text on RS/RA as "hints of forward progress" for proactive 4241 NUD. 4243 Changes from draft-templin-intarea-6706bis-19 to draft-templin- 4244 intrea-6706bis-20: 4246 o Included new route optimization source and destination addressing 4247 strategy. Now, route optimization maintenance uses the address of 4248 the existing Server instead of the data packet destination address 4249 so that less pressure is placed on the BGP routing system 4250 convergence time and Server constancy is supported. 4252 o Included new method for releasing from old MSE without requiring 4253 Client messaging. 4255 o Included references to new OMNI interface spec (including the OMNI 4256 option). 4258 o New appendix on AERO Client/Server architecture. 4260 Changes from draft-templin-intarea-6706bis-18 to draft-templin- 4261 intrea-6706bis-19: 4263 o Changed Proxy/Server keepalives to use "proactive NUD" in a manner 4264 that parallels BFD 4266 Changes from draft-templin-intarea-6706bis-17 to draft-templin- 4267 intrea-6706bis-18: 4269 o Discuss how AERO option is used in relation to S/TLLAOs 4271 o New text on Bidirectional Forwarding Detection (BFD) 4273 o Cleaned up usage (and non-usage) of unsolicited NAs 4275 o New appendix on Server failures 4277 Changes from draft-templin-intarea-6706bis-15 to draft-templin- 4278 intrea-6706bis-17: 4280 o S/TLLAO now includes multiple link-layer addresses within a single 4281 option instead of requiring multiple options 4283 o New unsolicited NA message to inform the old link that a Client 4284 has moved to a new link 4286 Changes from draft-templin-intarea-6706bis-14 to draft-templin- 4287 intrea-6706bis-15: 4289 o MTU and fragmentation 4291 o New details in movement to new Server 4293 Changes from draft-templin-intarea-6706bis-13 to draft-templin- 4294 intrea-6706bis-14: 4296 o Security based on secured tunnels, ingress filtering, MAP list and 4297 ROS list 4299 Changes from draft-templin-intarea-6706bis-12 to draft-templin- 4300 intrea-6706bis-13: 4302 o New paragraph in Section 3.6 on AERO interface layering over 4303 secured tunnels 4305 o Removed extraneous text in Section 3.7 4307 o Added new detail to the forwarding algorithm in Section 3.9 4308 o Clarified use of fragmentation 4310 o Route optimization now supported for both MNP and non-MNP-based 4311 prefixes 4313 o Relays are now seen as link-layer elements in the architecture. 4315 o Built out multicast section in detail. 4317 o New Appendix on implementation considerations for route 4318 optimization. 4320 Changes from draft-templin-intarea-6706bis-11 to draft-templin- 4321 intrea-6706bis-12: 4323 o Introduced Gateways as a new AERO element for connecting 4324 Correspondent Nodes on INET links 4326 o Introduced terms "Access Network (ANET)" and "Internetwork (INET)" 4328 o Changed "ASP" to "MSP", and "ACP" to "MNP" 4330 o New figure on the relation of Segments to the SPAN and AERO link 4332 o New "S" bit in S/TLLAO to indicate the "Source" S/TLLAO as opposed 4333 to additional S/TLLAOs 4335 o Changed Interface ID for Servers from 255 to 0xffff 4337 o Significant updates to Route Optimization, NUD, and Mobility 4338 Management 4340 o New Section on Multicast 4342 o New Section on AERO Clients in the open Internetwork 4344 o New Section on Operation over multiple AERO links (VLANs over the 4345 SPAN) 4347 o New Sections on DNS considerations and Transition considerations 4349 o 4351 Changes from draft-templin-intarea-6706bis-10 to draft-templin- 4352 intrea-6706bis-11: 4354 o Added The SPAN 4355 Changes from draft-templin-intarea-6706bis-09 to draft-templin- 4356 intrea-6706bis-10: 4358 o Orphaned packets in flight (e.g., when a neighbor cache entry is 4359 in the DEPARTED state) are now forwarded at the link layer instead 4360 of at the network layer. Forwarding at the network layer can 4361 result in routing loops and/or excessive delays of forwarded 4362 packets while the routing system is still reconverging. 4364 o Update route optimization to clarify the unsecured nature of the 4365 first NS used for route discovery 4367 o Many cleanups and clarifications on ND messaging parameters 4369 Changes from draft-templin-intarea-6706bis-08 to draft-templin- 4370 intrea-6706bis-09: 4372 o Changed PRL to "MAP list" 4374 o For neighbor cache entries, changed "static" to "symmetric", and 4375 "dynamic" to "asymmetric" 4377 o Specified Proxy RS/RA exchanges with Servers on behalf of Clients 4379 o Added discussion of unsolicited NAs in Section 3.16, and included 4380 forward reference to Section 3.18 4382 o Added discussion of AERO Clients used as critical infrastructure 4383 elements to connect fixed networks. 4385 o Added network-based VPN under security considerations 4387 Changes from draft-templin-intarea-6706bis-07 to draft-templin- 4388 intrea-6706bis-08: 4390 o New section on AERO-Aware Access Router 4392 Changes from draft-templin-intarea-6706bis-06 to draft-templin- 4393 intrea-6706bis-07: 4395 o Added "R" bit for release of PDs. Now have a full RS/RA service 4396 that can do PD without requiring DHCPv6 messaging over-the-air 4398 o Clarifications on solicited vs unsolicited NAs 4400 o Clarified use of MAX_NEIGHBOR_ADVERTISEMENT for the purpose of 4401 increase reliability 4403 Changes from draft-templin-intarea-6706bis-05 to draft-templin- 4404 intrea-6706bis-06: 4406 o Major re-work and simplification of Route Optimization function 4408 o Added Distributed Mobility Management (DMM) and Mobility Anchor 4409 Point (MAP) terminology 4411 o New section on "AERO Critical Infrastructure Element 4412 Considerations" demonstrating low overall cost for the service 4414 o minor text revisions and deletions 4416 o removed extraneous appendices 4418 Changes from draft-templin-intarea-6706bis-04 to draft-templin- 4419 intrea-6706bis-05: 4421 o New Appendix E on S/TLLAO Extensions for special-purpose links. 4422 Discussed ATN/IPS as example. 4424 o New sentence in introduction to declare appendices as non- 4425 normative. 4427 Changes from draft-templin-intarea-6706bis-03 to draft-templin- 4428 intrea-6706bis-04: 4430 o Added definitions for Potential Router List (PRL) and secure 4431 enclave 4433 o Included text on mapping transport layer port numbers to network 4434 layer DSCP values 4436 o Added reference to DTLS and DMM Distributed Mobility Anchoring 4437 working group document 4439 o Reworked Security Considerations 4441 o Updated references. 4443 Changes from draft-templin-intarea-6706bis-02 to draft-templin- 4444 intrea-6706bis-03: 4446 o Added new section on SEND. 4448 o Clarifications on "AERO Address" section. 4450 o Updated references and added new reference for RFC8086. 4452 o Security considerations updates. 4454 o General text clarifications and cleanup. 4456 Changes from draft-templin-intarea-6706bis-01 to draft-templin- 4457 intrea-6706bis-02: 4459 o Note on encapsulation avoidance in Section 4. 4461 Changes from draft-templin-intarea-6706bis-00 to draft-templin- 4462 intrea-6706bis-01: 4464 o Remove DHCPv6 Server Release procedures that leveraged the old way 4465 Relays used to "route" between Server link-local addresses 4467 o Remove all text relating to Relays needing to do any AERO-specific 4468 operations 4470 o Proxy sends RS and receives RA from Server using SEND. Use CGAs 4471 as source addresses, and destination address of RA reply is to the 4472 AERO address corresponding to the Client's ACP. 4474 o Proxy uses SEND to protect RS and authenticate RA (Client does not 4475 use SEND, but rather relies on subnetwork security. When the 4476 Proxy receives an RS from the Client, it creates a new RS using 4477 its own addresses as the source and uses SEND with CGAs to send a 4478 new RS to the Server. 4480 o Emphasize distributed mobility management 4482 o AERO address-based RS injection of ACP into underlying routing 4483 system. 4485 Changes from draft-templin-aerolink-82 to draft-templin-intarea- 4486 6706bis-00: 4488 o Document use of NUD (NS/NA) for reliable link-layer address 4489 updates as an alternative to unreliable unsolicited NA. 4490 Consistent with Section 7.2.6 of RFC4861. 4492 o Server adds additional layer of encapsulation between outer and 4493 inner headers of NS/NA messages for transmission through Relays 4494 that act as vanilla IPv6 routers. The messages include the AERO 4495 Server Subnet Router Anycast address as the source and the Subnet 4496 Router Anycast address corresponding to the Client's ACP as the 4497 destination. 4499 o Clients use Subnet Router Anycast address as the encapsulation 4500 source address when the access network does not provide a 4501 topologically-fixed address. 4503 Author's Address 4505 Fred L. Templin (editor) 4506 Boeing Research & Technology 4507 P.O. Box 3707 4508 Seattle, WA 98124 4509 USA 4511 Email: fltemplin@acm.org