idnits 2.17.1 draft-ietf-bess-nsh-bgp-control-plane-14.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 : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (June 2, 2020) is 1418 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: 'SPI' is mentioned on line 604, but not defined == Missing Reference: 'SI' is mentioned on line 604, but not defined == Outdated reference: A later version (-27) exists of draft-ietf-idr-rfc5575bis-25 == Outdated reference: A later version (-22) exists of draft-ietf-idr-tunnel-encaps-15 ** Obsolete normative reference: RFC 5575 (Obsoleted by RFC 8955) ** Downref: Normative reference to an Informational RFC: RFC 7665 ** Downref: Normative reference to an Informational RFC: RFC 8596 == Outdated reference: A later version (-06) exists of draft-dawra-idr-bgp-ls-sr-service-segments-03 Summary: 3 errors (**), 0 flaws (~~), 6 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 BESS Working Group A. Farrel 3 Internet-Draft Old Dog Consulting 4 Intended status: Standards Track J. Drake 5 Expires: December 4, 2020 E. Rosen 6 Juniper Networks 7 J. Uttaro 8 AT&T 9 L. Jalil 10 Verizon 11 June 2, 2020 13 BGP Control Plane for the Network Service Header in Service Function 14 Chaining 15 draft-ietf-bess-nsh-bgp-control-plane-14 17 Abstract 19 This document describes the use of BGP as a control plane for 20 networks that support Service Function Chaining (SFC). The document 21 introduces a new BGP address family called the SFC Address Family 22 Identifier / Subsequent Address Family Identifier (SFC AFI/SAFI) with 23 two route types. One route type is originated by a node to advertise 24 that it hosts a particular instance of a specified service function. 25 This route type also provides "instructions" on how to send a packet 26 to the hosting node in a way that indicates that the service function 27 has to be applied to the packet. The other route type is used by a 28 Controller to advertise the paths of "chains" of service functions, 29 and to give a unique designator to each such path so that they can be 30 used in conjunction with the Network Service Header defined in RFC 31 8300. 33 This document adopts the SFC architecture described in RFC 7665. 35 Status of This Memo 37 This Internet-Draft is submitted in full conformance with the 38 provisions of BCP 78 and BCP 79. 40 Internet-Drafts are working documents of the Internet Engineering 41 Task Force (IETF). Note that other groups may also distribute 42 working documents as Internet-Drafts. The list of current Internet- 43 Drafts is at https://datatracker.ietf.org/drafts/current/. 45 Internet-Drafts are draft documents valid for a maximum of six months 46 and may be updated, replaced, or obsoleted by other documents at any 47 time. It is inappropriate to use Internet-Drafts as reference 48 material or to cite them other than as "work in progress." 49 This Internet-Draft will expire on December 4, 2020. 51 Copyright Notice 53 Copyright (c) 2020 IETF Trust and the persons identified as the 54 document authors. All rights reserved. 56 This document is subject to BCP 78 and the IETF Trust's Legal 57 Provisions Relating to IETF Documents 58 (https://trustee.ietf.org/license-info) in effect on the date of 59 publication of this document. Please review these documents 60 carefully, as they describe your rights and restrictions with respect 61 to this document. Code Components extracted from this document must 62 include Simplified BSD License text as described in Section 4.e of 63 the Trust Legal Provisions and are provided without warranty as 64 described in the Simplified BSD License. 66 Table of Contents 68 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 69 1.1. Requirements Language . . . . . . . . . . . . . . . . . . 5 70 1.2. Terminology . . . . . . . . . . . . . . . . . . . . . . . 5 71 2. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 6 72 2.1. Overview of Service Function Chaining . . . . . . . . . . 6 73 2.2. Control Plane Overview . . . . . . . . . . . . . . . . . 8 74 3. BGP SFC Routes . . . . . . . . . . . . . . . . . . . . . . . 12 75 3.1. Service Function Instance Route (SFIR) . . . . . . . . . 13 76 3.1.1. SFIR Pool Identifier Extended Community . . . . . . . 14 77 3.1.2. MPLS Mixed Swapping/Stacking Extended Community . . . 15 78 3.2. Service Function Path Route (SFPR) . . . . . . . . . . . 16 79 3.2.1. The SFP Attribute . . . . . . . . . . . . . . . . . . 16 80 3.2.2. General Rules For The SFP Attribute . . . . . . . . . 22 81 4. Mode of Operation . . . . . . . . . . . . . . . . . . . . . . 23 82 4.1. Route Targets . . . . . . . . . . . . . . . . . . . . . . 23 83 4.2. Service Function Instance Routes . . . . . . . . . . . . 24 84 4.3. Service Function Path Routes . . . . . . . . . . . . . . 24 85 4.4. Classifier Operation . . . . . . . . . . . . . . . . . . 26 86 4.5. Service Function Forwarder Operation . . . . . . . . . . 26 87 4.5.1. Processing With 'Gaps' in the SI Sequence . . . . . . 27 88 5. Selection within Service Function Paths . . . . . . . . . . . 29 89 6. Looping, Jumping, and Branching . . . . . . . . . . . . . . . 31 90 6.1. Protocol Control of Looping, Jumping, and Branching . . . 31 91 6.2. Implications for Forwarding State . . . . . . . . . . . . 32 92 7. Advanced Topics . . . . . . . . . . . . . . . . . . . . . . . 33 93 7.1. Correlating Service Function Path Instances . . . . . . . 33 94 7.2. Considerations for Stateful Service Functions . . . . . . 34 95 7.3. VPN Considerations and Private Service Functions . . . . 35 96 7.4. Flow Specification for SFC Classifiers . . . . . . . . . 35 97 7.5. Choice of Data Plane SPI/SI Representation . . . . . . . 37 98 7.5.1. MPLS Representation of the SPI/SI . . . . . . . . . . 38 99 7.6. MPLS Label Swapping/Stacking Operation . . . . . . . . . 38 100 7.7. Support for MPLS-Encapsulated NSH Packets . . . . . . . . 39 101 8. Examples . . . . . . . . . . . . . . . . . . . . . . . . . . 39 102 8.1. Example Explicit SFP With No Choices . . . . . . . . . . 41 103 8.2. Example SFP With Choice of SFIs . . . . . . . . . . . . . 41 104 8.3. Example SFP With Open Choice of SFIs . . . . . . . . . . 42 105 8.4. Example SFP With Choice of SFTs . . . . . . . . . . . . . 42 106 8.5. Example Correlated Bidirectional SFPs . . . . . . . . . . 43 107 8.6. Example Correlated Asymmetrical Bidirectional SFPs . . . 43 108 8.7. Example Looping in an SFP . . . . . . . . . . . . . . . . 44 109 8.8. Example Branching in an SFP . . . . . . . . . . . . . . . 45 110 8.9. Examples of SFPs with Stateful Service Functions . . . . 45 111 8.9.1. Forward and Reverse Choice Made at the SFF . . . . . 46 112 8.9.2. Parallel End-to-End SFPs with Shared SFF . . . . . . 47 113 8.9.3. Parallel End-to-End SFPs with Separate SFFs . . . . . 49 114 8.9.4. Parallel SFPs Downstream of the Choice . . . . . . . 51 115 8.10. Examples Using IPv6 Addressing . . . . . . . . . . . . . 54 116 8.10.1. Example Explicit SFP With No Choices . . . . . . . . 56 117 8.10.2. Example SFP With Choice of SFIs . . . . . . . . . . 56 118 8.10.3. Example SFP With Open Choice of SFIs . . . . . . . . 57 119 8.10.4. Example SFP With Choice of SFTs . . . . . . . . . . 57 120 9. Security Considerations . . . . . . . . . . . . . . . . . . . 58 121 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 60 122 10.1. New BGP AF/SAFI . . . . . . . . . . . . . . . . . . . . 60 123 10.2. New BGP Path Attribute . . . . . . . . . . . . . . . . . 60 124 10.3. New SFP Attribute TLVs Type Registry . . . . . . . . . . 61 125 10.4. New SFP Association Type Registry . . . . . . . . . . . 61 126 10.5. New Service Function Type Registry . . . . . . . . . . . 62 127 10.6. New Generic Transitive Experimental Use Extended 128 Community Sub-Types . . . . . . . . . . . . . . . . . . 63 129 10.7. New BGP Transitive Extended Community Type . . . . . . . 63 130 10.8. New SFC Extended Community Sub-Types Registry . . . . . 64 131 10.9. SPI/SI Representation . . . . . . . . . . . . . . . . . 64 132 10.10. SFC SPI/SI Representation Flags Registry . . . . . . . . 64 133 11. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 65 134 12. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 65 135 13. References . . . . . . . . . . . . . . . . . . . . . . . . . 66 136 13.1. Normative References . . . . . . . . . . . . . . . . . . 66 137 13.2. Informative References . . . . . . . . . . . . . . . . . 67 138 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 68 140 1. Introduction 142 As described in [RFC7498], the delivery of end-to-end services can 143 require a packet to pass through a series of Service Functions (SFs) 144 (e.g., WAN and application accelerators, Deep Packet Inspection (DPI) 145 engines, firewalls, TCP optimizers, and server load balancers) in a 146 specified order: this is termed "Service Function Chaining" (SFC). 147 There are a number of issues associated with deploying and 148 maintaining service function chaining in production networks, which 149 are described below. 151 Historically, if a packet needed to travel through a particular 152 service chain, the nodes hosting the service functions of that chain 153 were placed in the network topology in such a way that the packet 154 could not reach its ultimate destination without first passing 155 through all the service functions in the proper order. This need to 156 place the service functions at particular topological locations 157 limited the ability to adapt a service function chain to changes in 158 network topology (e.g., link or node failures), network utilization, 159 or offered service load. These topological restrictions on where the 160 service functions can be placed raised the following issues: 162 1. The process of configuring or modifying a service function chain 163 is operationally complex and may require changes to the network 164 topology. 166 2. Alternate or redundant service functions may need to be co- 167 located with the primary service functions. 169 3. When there is more than one path between source and destination, 170 forwarding may be asymmetric and it may be difficult to support 171 bidirectional service function chains using simple routing 172 methodologies and protocols without adding mechanisms for traffic 173 steering or traffic engineering. 175 In order to address these issues, the SFC architecture describes 176 Service Function Chains that are built in their own overlay network 177 (the service function overlay network), coexisting with other overlay 178 networks, over a common underlay network [RFC7665]. A Service 179 Function Chain is a sequence of Service Functions through which 180 packet flows that satisfy specified criteria will pass. 182 This document describes the use of BGP as a control plane for 183 networks that support Service Function Chaining (SFC). The document 184 introduces a new BGP address family called the SFC AFI/SAFI with two 185 route types. One route type is originated by a node to advertise 186 that it hosts a particular instance of a specified service function. 187 This route type also provides "instructions" on how to send a packet 188 to the hosting node in a way that indicates that the service function 189 has to be applied to the packet. The other route type is used by a 190 Controller (a centralized network component responsible for planning 191 and coordinating Service Function Chaining within the network) to 192 advertise the paths of "chains" of service functions, and to give a 193 unique designator to each such path so that they can be used in 194 conjunction with the Network Service Header [RFC8300]. 196 This document adopts the SFC architecture described in [RFC7665]. 198 1.1. Requirements Language 200 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 201 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 202 "OPTIONAL" in this document are to be interpreted as described in BCP 203 14 [RFC2119] [RFC8174] when, and only when, they appear in all 204 capitals, as shown here. 206 1.2. Terminology 208 This document uses the following terms from [RFC7665]: 210 o Bidirectional Service Function Chain 212 o Classifier 214 o Service Function (SF) 216 o Service Function Chain (SFC) 218 o Service Function Forwarder (SFF) 220 o Service Function Instance (SFI) 222 o Service Function Path (SFP) 224 o SFC branching 226 Additionally, this document uses the following terms from [RFC8300]: 228 o Network Service Header (NSH) 230 o Service Index (SI) 232 o Service Path Identifier (SPI) 234 This document introduces the following terms: 236 o Service Function Instance Route (SFIR). A new BGP Route Type 237 advertised by the node that hosts an SFI to describe the SFI and 238 to announce the way to forward a packet to the node through the 239 underlay network. 241 o Service Function Overlay Network. The logical network comprised 242 of Classifiers, SFFs, and SFIs that are connected by paths or 243 tunnels through underlay transport networks. 245 o Service Function Path Route (SFPR). A new BGP Route Type 246 originated by Controllers to advertise the details of each SFP. 248 o Service Function Type (SFT). An indication of the function and 249 features of an SFI. 251 2. Overview 253 This section provides an overview of Service Function Chaining in 254 general, and the control plane defined in this document. After 255 reading this section, readers may find it helpful to look through 256 Section 8 for some simple worked examples. 258 2.1. Overview of Service Function Chaining 260 In [RFC8300] a Service Function Chain (SFC) is an ordered list of 261 Service Functions (SFs). A Service Function Path (SFP) is an 262 indication of which instances of SFs are acceptable to be traversed 263 in an instantiation of an SFC in a service function overlay network. 264 The Service Path Identifier (SPI) is a 24-bit number that identifies 265 a specific SFP, and a Service Index (SI) is an 8-bit number that 266 identifies a specific point in that path. In the context of a 267 particular SFP (identified by an SPI), an SI represents a particular 268 Service Function, and indicates the order of that SF in the SFP. 270 Within the context of a specific SFP, an SI references a set of one 271 or more SFs. Each of those SFs may be supported by one or more 272 Service Function Instances (SFIs). Thus an SI may represent a choice 273 of SFIs of one or more Service Function Types. By deploying multiple 274 SFIs for a single SF, one can provide load balancing and redundancy. 276 A special functional element, called a Classifier, is located at each 277 ingress point to a service function overlay network. It assigns the 278 packets of a given packet flow to a specific Service Function Path. 279 This may be done by comparing specific fields in a packet's header 280 with local policy, which may be customer/network/service specific. 281 The Classifier picks an SFP and sets the SPI accordingly, it then 282 sets the SI to the value of the SI for the first hop in the SFP, and 283 then prepends a Network Services Header (NSH) [RFC8300] containing 284 the assigned SPI/SI to that packet. Note that the Classifier and the 285 node that hosts the first Service Function in a Service Function Path 286 need not be located at the same point in the service function overlay 287 network. 289 Note that the presence of the NSH can make it difficult for nodes in 290 the underlay network to locate the fields in the original packet that 291 would normally be used to constrain equal cost multipath (ECMP) 292 forwarding. Therefore, it is recommended that the node prepending 293 the NSH also provide some form of entropy indicator that can be used 294 in the underlay network. How this indicator is generated and 295 supplied, and how an SFF generates a new entropy indicator when it 296 forwards a packet to the next SFF, are out of scope of this document. 298 The Service Function Forwarder (SFF) receives a packet from the 299 previous node in a Service Function Path, removes the packet's link 300 layer or tunnel encapsulation and hands the packet and the NSH to the 301 Service Function Instance for processing. The SFI has no knowledge 302 of the SFP. 304 When the SFF receives the packet and the NSH back from the SFI it 305 must select the next SFI along the path using the SPI and SI in the 306 NSH and potentially choosing between multiple SFIs (possibly of 307 different Service Function Types) as described in Section 5. In the 308 normal case the SPI remains unchanged and the SI will have been 309 decremented to indicate the next SF along the path. But other 310 possibilities exist if the SF makes other changes to the NSH through 311 a process of re-classification: 313 o The SI in the NSH may indicate: 315 * A previous SF in the path: known as "looping" (see Section 6). 317 * An SF further down the path: known as "jumping" (see also 318 Section 6). 320 o The SPI and the SI may point to an SF on a different SFP: known as 321 "branching" (see also Section 6). 323 Such modifications are limited to within the same service function 324 overlay network. That is, an SPI is known within the scope of 325 service function overlay network. Furthermore, the new SI value is 326 interpreted in the context of the SFP identified by the SPI. 328 As described in [RFC8300], an unknown or invalid SPI is treated as an 329 error and the SFF drops the packet: such errors should be logged, and 330 such logs are subject to rate limits. 332 Also, as described in [RFC8300], an SFF receiving an SI that is 333 unknown in the context of the SPI can reduce the value to the next 334 meaningful SI value in the SFP indicated by the SPI. If no such 335 value exists or if the SFF does not support reducing the SI, the SFF 336 drops the packet and should log the event: such logs are also subject 337 to rate limits. 339 The SFF then selects an SFI that provides the SF denoted by the SPI/ 340 SI, and forwards the packet to the SFF that supports that SFI. 342 [RFC8300] makes it clear that the intended scope is for use within a 343 single provider's operational domain. 345 This document adopts the SFC architecture described in [RFC7665] and 346 adds a control plane to support the functions as described in 347 Section 2.2. An essential component of this solution is the 348 Controller. This is a network component responsible for planning 349 SFPs within the network. It gathers information about the 350 availability of SFIs and SFFs, instructs the control plane about the 351 SFPs to be programmed, and instructs the Classifiers how to assign 352 traffic flows to individual SFPs. 354 2.2. Control Plane Overview 356 To accomplish the function described in Section 2.1, this document 357 introduces the Service Function Type (SFT) that is the category of SF 358 that is supported by an SFF (such as "firewall"). An IANA registry 359 of Service Function Types is introduced in Section 10 and is 360 consistent with types used in other work such as 361 [I-D.dawra-idr-bgp-ls-sr-service-segments]. An SFF may support SFs 362 of multiple different SFTs, and may support multiple SFIs of each SF. 364 This document also introduces a new BGP AFI/SAFI (values to be 365 assigned by IANA) for "SFC Routes". Two SFC Route Types are defined 366 by this document: the Service Function Instance Route (SFIR), and the 367 Service Function Path Route (SFPR). As detailed in Section 3, the 368 route type is indicated by a sub-field in the NLRI. 370 o The SFIR is advertised by the node hosting the service function 371 instance (i.e., the SFF). The SFIR describes a particular 372 instance of a particular Service Function (i.e., an SFI) and the 373 way to forward a packet to it through the underlay network, i.e., 374 IP address and encapsulation information. 376 o The SFPRs are originated by Controllers. One SFPR is originated 377 for each Service Function Path. The SFPR specifies: 379 A. the SPI of the path 381 B. the sequence of SFTs and/or SFIs of which the path consists 382 C. for each such SFT or SFI, the SI that represents it in the 383 identified path. 385 This approach assumes that there is an underlay network that provides 386 connectivity between SFFs and Controllers, and that the SFFs are 387 grouped to form one or more service function overlay networks through 388 which SFPs are built. We assume the the Controllers have BGP 389 connectivity to all SFFs and all Classifiers within each service 390 function overlay network. 392 When choosing the next SFI in a path, the SFF uses the SPI and SI as 393 well as the SFT to choose among the SFIs, applying, for example, a 394 load balancing algorithm or direct knowledge of the underlay network 395 topology as described in Section 4. 397 The SFF then encapsulates the packet using the encapsulation 398 specified by the SFIR of the selected SFI and forwards the packet. 399 See Figure 1. 401 Thus the SFF can be seen as a portal in the underlay network through 402 which a particular SFI is reached. 404 Figure 1 shows a reference model for the SFC architecture. There are 405 four SFFs (SFF-1 through SFF-4) connected by tunnels across the 406 underlay network. Packets arrive at a Classifier and are channeled 407 along SFPs to destinations reachable through SFF-4. 409 SFF-1 and SFF-4 each have one instance of one SF attached (SFa and 410 SFe). SFF-2 has two types of SF attached: there is one instance of 411 one (SFc), and three instances of the other (SFb). SFF-3 has just 412 one instance of an SF (SFd), but it in this case the type of SFd is 413 the same type as SFb (SFTx). 415 This figure demonstrates how load balancing can be achieved by 416 creating several SFPs that satisfy the same SFC. Suppose an SFC 417 needs to include SFa, an SF of type SFTx, and SFc. A number of SFPs 418 can be constructed using any instance of SFb or using SFd. Load 419 balancing may be applied at two places: 421 o The Classifier may distribute different flows onto different SFPs 422 to share the load in the network and across SFIs. 424 o SFF-2 may distribute different flows (on the same SFP) to 425 different instances of SFb to share the processing load. 427 Note that, for convenience and clarity, Figure 1 shows only a few 428 tunnels between SFFs. There could be a full mesh of such tunnels, or 429 more likely, a selection of tunnels connecting key SFFs to enable the 430 construction of SFPs and to balance load and traffic in the network. 431 Further, the figure does not show any controllers: these would each 432 have BGP connectivity to the Classifier and all of the SFFs. 434 Packets 435 | | | 436 ------------ 437 | | 438 | Classifier | 439 | | 440 ------+----- 441 | 442 ---+--- --------- ------- 443 | | Tunnel | | | | 444 | SFF-1 |===============| SFF-2 |=========| SFF-4 | 445 | | | | | | 446 | | -+-----+- | | 447 | | ,,,,,,,,,,,,,,/,, \ | | 448 | | ' .........../. ' ..\...... | | 449 | | ' : SFb / : ' : \ SFc : | | 450 | | ' : ---+- : ' : --+-- : | | 451 | | ' : -| SFI | : ' : | SFI | : | | 452 | | ' : -| ----- : ' : ----- : | | 453 | | ' : | ----- : ' ......... | | 454 | | ' : ----- : ' | | 455 | | ' ............. ' | |--- Dests 456 | | ' ' | |--- Dests 457 | | ' ......... ' | | 458 | | ' : ----- : ' | | 459 | | ' : | SFI | : ' | | 460 | | ' : --+-- : ' | | 461 | | ' :SFd | : ' | | 462 | | ' ....|.... ' | | 463 | | ' | ' | | 464 | | ' SFTx | ' | | 465 | | ',,,,,,,,|,,,,,,,,' | | 466 | | | | | 467 | | ---+--- | | 468 | | | | | | 469 | |======| SFF-3 |====================| | 470 ---+--- | | ---+--- 471 | ------- | 472 ....|.... ....|.... 473 : | SFa: : | SFe: 474 : --+-- : : --+-- : 475 : | SFI | : : | SFI | : 476 : ----- : : ----- : 477 ......... ......... 479 Figure 1: The SFC Architecture Reference Model 481 As previously noted, [RFC8300] makes it clear that the mechanisms it 482 defines are intended for use within a single provider's operational 483 domain. This reduces the requirements on the control plane function. 485 [RFC7665] sets out the functions provided by a control plane for an 486 SFC network in Section 5.2. The functions are broken down into six 487 items the first four of which are completely covered by the 488 mechanisms described in this document: 490 1. Visiblity of all SFs and the SFFs through which they are reached. 492 2. Computation of SFPs and progrmming into the network. 494 3. Selection of SFIs explicitly in the SFP or dynamically within the 495 network. 497 4. Programming of SFFs with forwarding path information. 499 The fifth and six items in the list in RFC 7665 concern the use of 500 metadata. These are more peripheral to the control plane mechanisms 501 defined in this document, but are discussed in Section 4.4. 503 3. BGP SFC Routes 505 This document defines a new AFI/SAFI for BGP, known as "SFC", with an 506 NLRI that is described in this section. 508 The format of the SFC NLRI is shown in Figure 2. 510 +---------------------------------------+ 511 | Route Type (2 octets) | 512 +---------------------------------------+ 513 | Length (2 octets) | 514 +---------------------------------------+ 515 | Route Type specific (variable) | 516 +---------------------------------------+ 518 Figure 2: The Format of the SFC NLRI 520 The Route Type field determines the encoding of the rest of the route 521 type specific SFC NLRI. 523 The Length field indicates the length in octets of the route type 524 specific field of the SFC NLRI. 526 This document defines the following Route Types: 528 1. Service Function Instance Route (SFIR) 530 2. Service Function Path Route (SFPR) 532 A Service Function Instance Route (SFIR) is used to identify an SFI. 533 A Service Function Path Route (SFPR) defines a sequence of Service 534 Functions (each of which has at least one instance advertised in an 535 SFIR) that form an SFP. 537 The detailed encoding and procedures for these Route Types are 538 described in subsequent sections. 540 The SFC NLRI is carried in BGP [RFC4271] using BGP Multiprotocol 541 Extensions [RFC4760] with an Address Family Identifier (AFI) of TBD1 542 and a Subsequent Address Family Identifier (SAFI) of TBD2. The NLRI 543 field in the MP_REACH_NLRI/MP_UNREACH_NLRI attribute contains the SFC 544 NLRI, encoded as specified above. 546 In order for two BGP speakers to exchange SFC NLRIs, they MUST use 547 BGP Capabilities Advertisements to ensure that they both are capable 548 of properly processing such NLRIs. This is done as specified in 549 [RFC4760], by using capability code 1 (Multiprotocol BGP) with an AFI 550 of TBD1 and a SAFI of TBD2. 552 The nexthop field of the MP_REACH_NLRI attribute of the SFC NLRI MUST 553 be set to a loopback address of the advertising SFF. 555 3.1. Service Function Instance Route (SFIR) 557 Figure 3 shows the Route Type specific NLRI of the SFIR. 559 +--------------------------------------------+ 560 | Route Distinguisher (RD) (8 octets) | 561 +--------------------------------------------+ 562 | Service Function Type (2 octets) | 563 +--------------------------------------------+ 565 Figure 3: SFIR Route Type specific NLRI 567 Per [RFC4364] the RD field comprises a two byte Type field and a six 568 byte Value field. If two SFIRs are originated from different 569 administrative domains (within the same provier's operational 570 domain), they MUST have different RDs. In particular, SFIRs from 571 different VPNs (for different service function overlay networks) MUST 572 have different RDs, and those RDs MUST be different from any non-VPN 573 SFIRs. 575 The Service Function Type identifies the functions/features a service 576 function can offer, e.g., Classifier, firewall, load balancer. There 577 may be several SFIs that can perform a given Service Function. Each 578 node hosting an SFI MUST originate an SFIR for each type of SF that 579 it hosts (as indicated by the SFT value), and it MAY advertise an 580 SFIR for each instance of each type of SF. The minimal advertisement 581 allows construction of valid SFPs and leaves the selection of SFIs to 582 the local SFF; the detailed advertisement may have scaling concerns, 583 but allows a Controller that constructs an SFP to make an explicit 584 choice of SFI. 586 Note that a node may advertise all its SFIs of one SFT in one shot 587 using normal BGP Update packing. That is, all of the SFIRs in an 588 Update share a common Tunnel Encapsulation and Route Target (RT) 589 attribute. See also Section 3.2.1. 591 The SFIR representing a given SFI will contain an NLRI with RD field 592 set to an RD as specified above, and with SFT field set to identify 593 that SFI's Service Function Type. The values for the SFT field are 594 taken from a registry administered by IANA (see Section 10). A BGP 595 Update containing one or more SFIRs MUST also include a Tunnel 596 Encapsulation attribute [I-D.ietf-idr-tunnel-encaps]. If a data 597 packet needs to be sent to an SFI identified in one of the SFIRs, it 598 will be encapsulated as specified by the Tunnel Encapsulation 599 attribute, and then transmitted through the underlay network. 601 Note that the Tunnel Encapsulation attribute MUST contain sufficient 602 information to allow the advertising SFF to identify the overlay or 603 VPN network which a received packet is transiting. This is because 604 the [SPI, SI] in a received packet is specific to a particular 605 overlay or VPN network. 607 3.1.1. SFIR Pool Identifier Extended Community 609 This document defines a new transitive extended community [RFC4360] 610 of type TBD6 called the SFC extended community. When used with Sub- 611 Type TBD7, this is called the SFIR Pool Identifier extended 612 community. It MAY be included in SFIR advertisements, and is used to 613 indicate the identity of a pool of SFIRs to which an SFIR belongs. 614 Since an SFIR may be a member of multiple pools, multiple of these 615 extended communities may be present on a single SFIR advertisement. 617 SFIR pools allow SFIRs to be grouped for any purpose. Possible uses 618 include control plane scalability and stability. A pool identifier 619 may be included in an SFPR to indicate a set of SFIs that are 620 acceptable at a specific point on an SFP (see Section 3.2.1.3 and 621 Section 4.3). 623 The SFIR Pool Identifier extended community is encoded in 8 octets as 624 shown in Figure 4. 626 +--------------------------------------------+ 627 | Type = TBD6 (1 octet) | 628 +--------------------------------------------+ 629 | Sub-Type = TBD7 (1 octet) | 630 +--------------------------------------------+ 631 | SFIR Pool Identifier Value (6 octets) | 632 +--------------------------------------------+ 634 Figure 4: The SFIR Pool Identifier Extended Community 636 The SFIR Pool Identifier Value is encoded in a 6 octet field in 637 network byte order, and the value is unique within the scope of an 638 overlay network. This means that pool identifiers need to be 639 centrally managed, which is consistent with the assignment of SFIs to 640 pools. 642 3.1.2. MPLS Mixed Swapping/Stacking Extended Community 644 As noted in Section 3.1.1, this document defines a new transitive 645 extended community of type TBD6 called the SFC extended community. 646 When used with Sub-Type TBD8, this is called the MPLS Mixed Swapping/ 647 Stacking Labels extended community. The community is encoded as 648 shown in Figure 5. It contains a pair of MPLS labels: an SFC Context 649 Label and an SF Label as described in [RFC8595]. Each label is 20 650 bits encoded in a 3-octet (24 bit) field with 4 trailing bits that 651 MUST be set to zero. 653 +--------------------------------------------+ 654 | Type = TBD6 (1 octet) | 655 +--------------------------------------------| 656 | Sub-Type = TBD8 (1 octet) | 657 +--------------------------------------------| 658 | SFC Context Label (3 octets) | 659 +--------------------------------------------| 660 | SF Label (3 octets) | 661 +--------------------------------------------+ 663 Figure 5: The MPLS Mixed Swapping/Stacking Extended Community 665 Note that it is assumed that each SFF has one or more globally unique 666 SFC Context Labels and that the context label space and the SPI 667 address space are disjoint (i.e., a label value cannot be used both 668 to indicate an SFC context and an SPI, and it can be determined from 669 knowledge of the label spaces whether a label indicates an SFC 670 context or an SPI). 672 If an SFF supports SFP Traversal with an MPLS Label Stack it MUST 673 include this extended community with the SFIRs that it advertises. 675 See Section 7.6 for a description of how this extended community is 676 used. 678 3.2. Service Function Path Route (SFPR) 680 Figure 6 shows the Route Type specific NLRI of the SFPR. 682 +-----------------------------------------------+ 683 | Route Distinguisher (RD) (8 octets) | 684 +-----------------------------------------------+ 685 | Service Path Identifier (SPI) (3 octets) | 686 +-----------------------------------------------+ 688 Figure 6: SFPR Route Type Specific NLRI 690 Per [RFC4364] the RD field comprises a two byte Type field and a six 691 byte Value field. All SFPs MUST be associated with an RD. The 692 association of an SFP with an RD is determined by provisioning. If 693 two SFPRs are originated from different Controllers they MUST have 694 different RDs. Additionally, SFPRs from different VPNs (i.e., in 695 different service function overlay networks) MUST have different RDs, 696 and those RDs MUST be different from any non-VPN SFPRs. 698 The Service Path Identifier is defined in [RFC8300] and is the value 699 to be placed in the Service Path Identifier field of the NSH header 700 of any packet sent on this Service Function Path. It is expected 701 that one or more Controllers will originate these routes in order to 702 configure a service function overlay network. 704 The SFP is described in a new BGP Path attribute, the SFP attribute. 705 Section 3.2.1 shows the format of that attribute. 707 3.2.1. The SFP Attribute 709 [RFC4271] defines BGP Path attributes. This document introduces a 710 new Optional Transitive Path attribute called the SFP attribute with 711 value TBD3 to be assigned by IANA. The first SFP attribute MUST be 712 processed and subsequent instances MUST be ignored. 714 The common fields of the SFP attribute are set as follows: 716 o Optional bit is set to 1 to indicate that this is an optional 717 attribute. 719 o The Transitive bit is set to 1 to indicate that this is a 720 transitive attribute. 722 o The Extended Length bit is set if the length of the SFP attribute 723 is encoded in one octet (set to 0) or two octets (set to 1) as 724 described in [RFC4271]. 726 o The Attribute Type Code is set to TBD3. 728 The content of the SFP attribute is a series of Type-Length-Value 729 (TLV) constructs. Some TLVs may include sub-TLVs. All TLVs and sub- 730 TLVs have a common format that is: 732 o Type: A single octet indicating the type of the SFP attribute TLV. 733 Values are taken from the registry described in Section 10.3. 735 o Length: A two octet field indicating the length of the data 736 following the Length field counted in octets. 738 o Value: The contents of the TLV. 740 The formats of the TLVs defined in this document are shown in the 741 following sections. The presence rules and meanings are as follows. 743 o The SFP attribute contains a sequence of zero or more Association 744 TLVs. That is, the Association TLV is OPTIONAL. Each Association 745 TLV provides an association between this SFPR and another SFPR. 746 Each associated SFPR is indicated using the RD with which it is 747 advertised (we say the SFPR-RD to avoid ambiguity). 749 o The SFP attribute contains a sequence of one or more Hop TLVs. 750 Each Hop TLV contains all of the information about a single hop in 751 the SFP. 753 o Each Hop TLV contains an SI value and a sequence of one or more 754 SFT TLVs. Each SFT TLV contains an SFI reference for each 755 instance of an SF that is allowed at this hop of the SFP for the 756 specific SFT. Each SFI is indicated using the RD with which it is 757 advertised (we say the SFIR-RD to avoid ambiguity). 759 Section 6 of [RFC4271] describes the handling of malformed BGP 760 attributes, or those that are in error in some way. [RFC7606] 761 revises BGP error handling specifically for the UPDATE message, 762 provides guidelines for the authors of documents defining new 763 attributes, and revises the error handling procedures for a number of 764 existing attributes. This document introduces the SFP attribute and 765 so defines error handling as follows: 767 o When parsing a message, an unknown Attribute Type code or a length 768 that suggests that the attribute is longer than the remaining 769 message is treated as a malformed message and the "treat-as- 770 withdraw" approach used as per [RFC7606]. 772 o When parsing a message that contains an SFP attribute, the 773 following cases constitute errors: 775 1. Optional bit is set to 0 in SFP attribute. 777 2. Transitive bit is set to 0 in SFP attribute. 779 3. Unknown TLV type field found in SFP attribute. 781 4. TLV length that suggests the TLV extends beyond the end of the 782 SFP attribute. 784 5. Association TLV contains an unknown SFPR-RD. 786 6. No Hop TLV found in the SFP attribute. 788 7. No sub-TLV found in a Hop TLV. 790 8. Unknown SFIR-RD found in an SFT TLV. 792 o The errors listed above are treated as follows: 794 1., 2., 4., 6., 7.: The attribute MUST be treated as malformed 795 and the "treat-as-withdraw" approach used as per [RFC7606]. 797 3.: Unknown TLVs MUST be ignored, and message processing MUST 798 continue. 800 5., 8.: The absence of an RD with which to correlate is nothing 801 more than a soft error. The receiver SHOULD store the 802 information from the SFP attribute until a corresponding 803 advertisement is received. 805 3.2.1.1. The Association TLV 807 The Association TLV is an optional TLV in the SFP attribute. It MAY 808 be present multiple times. Each occurrence provides an association 809 with another SFP as advertised in another SFPR. The format of the 810 Association TLV is shown in Figure 7 812 +--------------------------------------------+ 813 | Type = 1 (1 octet) | 814 +--------------------------------------------| 815 | Length (2 octets) | 816 +--------------------------------------------| 817 | Association Type (1 octet) | 818 +--------------------------------------------| 819 | Associated SFPR-RD (8 octets) | 820 +--------------------------------------------| 821 | Associated SPI (3 octets) | 822 +--------------------------------------------+ 824 Figure 7: The Format of the Association TLV 826 The fields are as follows: 828 Type is set to 1 to indicate an Association TLV. 830 Length indicates the length in octets of the Association Type and 831 Associated SFPR-RD fields. The value of the Length field is 12. 833 The Association Type field indicate the type of association. The 834 values are tracked in an IANA registry (see Section 10.4). Only 835 one value is defined in this document: type 1 indicates 836 association of two unidirectional SFPs to form a bidirectional 837 SFP. An SFP attribute SHOULD NOT contain more than one 838 Association TLV with Association Type 1: if more than one is 839 present, the first one MUST be processed and subsequent instances 840 MUST be ignored. Note that documents that define new Association 841 Types must also define the presence rules for Association TLVs of 842 the new type. 844 The Associated SFPR-RD contains the RD of the associated SFP as 845 advertised in an SFPR. 847 The Associated SPI contains the SPI of the associated SFP as 848 advertised in an SFPR. 850 Association TLVs with unknown Association Type values SHOULD be 851 ignored. Association TLVs that contain an Associated SFPR-RD value 852 equal to the RD of the SFPR in which they are contained SHOULD be 853 ignored. If the Associated SPI is not equal to the SPI advertised in 854 the SFPR indicated by the Associated SFPR-RD then the Association TLV 855 SHOULD be ignored. In all three of these cases an implementation MAY 856 reject the SFP attribute as malformed and use the "treat-as-withdraw" 857 approach per [RFC7606], however implementers are cautioned that such 858 an approach may make an implementation less flexible in the event of 859 future extensions to this protocol. 861 Note that when two SFPRs reference each other using the Association 862 TLV, one SFPR advertisement will be received before the other. 863 Therefore, processing of an association MUST NOT be rejected simply 864 because the Associated SFPR-RD is unknown. 866 Further discussion of correlation of SFPRs is provided in 867 Section 7.1. 869 3.2.1.2. The Hop TLV 871 There is one Hop TLV in the SFP attribute for each hop in the SFP. 872 The format of the Hop TLV is shown in Figure 8. At least one Hop TLV 873 MUST be present in an SFP attribute. 875 +--------------------------------------------+ 876 | Type = 2 (1 octet) | 877 +--------------------------------------------| 878 | Length (2 octets) | 879 +--------------------------------------------| 880 | Service Index (1 octet) | 881 +--------------------------------------------| 882 | Hop Details (variable) | 883 +--------------------------------------------+ 885 Figure 8: The Format of the Hop TLV 887 The fields are as follows: 889 Type is set to 2 to indicate a Hop TLV. 891 Length indicates the length in octets of the Service Index and Hop 892 Details fields. 894 The Service Index is defined in [RFC8300] and is the value found 895 in the Service Index field of the NSH header that an SFF will use 896 to lookup to which next SFI a packet is to be sent. 898 The Hop Details field consists of a sequence of one or more sub- 899 TLVs. 901 Each hop of the SFP may demand that a specific type of SF is 902 executed, and that type is indicated in sub-TLVs of the Hop TLV. At 903 least one sub-TLV MUST be present. This document defines the SFT 904 Sub-TLV (see Section 3.2.1.3 and the MPLS Swapping/Stacking Sub-TLV 905 (see Section Section 3.2.1.4: other sub-TLVs may be defined in 906 future. This provides a list of which types of SF are acceptable at 907 a specific hop, and for each type it allows a degree of control to be 908 imposed on the choice of SFIs of that particular type. 910 If no Hop TLV is present in an SFP Attribute, it is a malformed 911 attribute 913 3.2.1.3. The SFT Sub-TLV 915 The SFT Sub-TLV MAY be included in the list of sub-TLVs of the Hop 916 TLV. The format of the SFT Sub-TLV is shown in Figure 9. The Sub- 917 TLV contains a list of SFIR-RD values each taken from the 918 advertisement of an SFI. Together they form a list of acceptable 919 SFIs of the indicated type. 921 +--------------------------------------------+ 922 | Type = 3 (1 octet) | 923 +--------------------------------------------| 924 | Length (2 octets) | 925 +--------------------------------------------| 926 | Service Function Type (2 octets) | 927 +--------------------------------------------| 928 | SFIR-RD List (variable) | 929 +--------------------------------------------+ 931 Figure 9: The Format of the SFT Sub-TLV 933 The fields are as follows: 935 Type is set to 3 to indicate an SFT Sub-TLV. 937 Length indicates the length in octets of the Service Function Type 938 and SFIR-RD List fields. 940 The Service Function Type value indicates the category (type) of 941 SF that is to be executed at this hop. The types are as 942 advertised for the SFs supported by the SFFs. SFT values in the 943 range 1-31 are Special Purpose SFT values and have meanings 944 defined by the documents that describe them - the value 'Change 945 Sequence' is defined in Section 6.1 of this document. 947 The hop description is further qualified beyond the specification 948 of the SFTs by listing, for each SFT in each hop, the SFIs that 949 may be used at the hop. The SFIs are identified using the SFIR- 950 RDs from the advertisements of the SFIs in the SFIRs. Note that 951 if the list contains one or more SFIR Pool Identifiers, then for 952 each the SFIR-RD list is effectively expanded to include the SFIR- 953 RD of each SFIR advertised with that SFIR Pool Identifier. An 954 SFIR-RD of value zero has special meaning as described in 955 Section 5. Each entry in the list is eight octets long, and the 956 number of entries in the list can be deduced from the value of the 957 Length field. 959 3.2.1.4. MPLS Swapping/Stacking Sub-TLV 961 The MPLS Swapping/Stacking Sub-TLV (Type value 4) is a zero length 962 sub-TLV that is OPTIONAL in the Hop TLV and is used when the data 963 representation is MPLS (see Section 7.5). When present it indicates 964 to the Classifier imposing an MPLS label stack that the current hop 965 is to use an {SFC Context Label, SF label} rather than an {SPI, SF} 966 label pair. See Section 7.6 for more details. 968 3.2.1.5. SFP Traversal With MPLS Label Stack TLV 970 The SFP Traversal With MPLS Label Stack TLV (Type value 5) is a zero 971 length TLV that can be carried in the SFP Attribute and indicates to 972 the Classifier and the SFFs on the SFP that an MPLS label stack with 973 label swapping/stacking is to be used for packets traversing the SFP. 974 All of the SFF specified at each the SFP's hops MUST have advertised 975 an MPLS Mixed Swapping/Stacking Extended Community (see 976 Section 3.1.2) for the SFP to be considered usable. 978 3.2.2. General Rules For The SFP Attribute 980 It is possible for the same SFI, as described by an SFIR, to be used 981 in multiple SFPRs. 983 When two SFPRs have the same SPI but different SFPR-RDs there can be 984 three cases: 986 o Two or more Controllers are originating SFPRs for the same SFP. 987 In this case the content of the SFPRs is identical and the 988 duplication is to ensure receipt and to provide Controller 989 redundancy. 991 o There is a transition in content of the advertised SFP and the 992 advertisements may originate from one or more Controllers. In 993 this case the content of the SFPRs will be different. 995 o The reuse of an SPI may result from a configuration error. 997 In all cases, there is no way for the receiving SFF to know which 998 SFPR to process, and the SFPRs could be received in any order. At 999 any point in time, when multiple SFPRs have the same SPI but 1000 different SFPR-RDs, the SFF MUST use the SFPR with the numerically 1001 lowest SFPR-RD when interpretting the RDs as 8-octet integers in 1002 network byte order. The SFF SHOULD log this occurrence to assist 1003 with debugging. 1005 Furthermore, a Controller that wants to change the content of an SFP 1006 is RECOMMENDED to use a new SPI and so create a new SFP onto which 1007 the Classifiers can transition packet flows before the SFPR for the 1008 old SFP is withdrawn. This avoids any race conditions with SFPR 1009 advertisements. 1011 Additionally, a Controller SHOULD NOT re-use an SPI after it has 1012 withdrawn the SFPR that used it until at least a configurable amount 1013 of time has passed. This timer SHOULD have a default of one hour. 1015 4. Mode of Operation 1017 This document describes the use of BGP as a control plane to create 1018 and manage a service function overlay network. 1020 4.1. Route Targets 1022 The main feature introduced by this document is the ability to create 1023 multiple service function overlay networks through the use of Route 1024 Targets (RTs) [RFC4364]. 1026 Every BGP UPDATE containing an SFIR or SFPR carries one or more RTs. 1027 The RT carried by a particular SFIR or SFPR is determined by the 1028 provisioning of the route's originator. 1030 Every node in a service function overlay network is configured with 1031 one or more import RTs. Thus, each SFF will import only the SFPRs 1032 with matching RTs allowing the construction of multiple service 1033 function overlay networks or the instantiation of Service Function 1034 Chains within an L3VPN or EVPN instance (see Section 7.3). An SFF 1035 that has a presence in multiple service function overlay networks 1036 (i.e., imports more than one RT) will usually maintain separate 1037 forwarding state for each overlay network. 1039 4.2. Service Function Instance Routes 1041 The SFIR (see Section 3.1) is used to advertise the existence and 1042 location of a specific Service Function Instance and consists of: 1044 o The RT as just described. 1046 o A Service Function Type (SFT) that is the type of service function 1047 that is provided (such as "firewall"). 1049 o A Route Distinguisher (RD) that is unique to a specific overlay. 1051 4.3. Service Function Path Routes 1053 The SFPR (see Section 3.2) describes a specific path of a Service 1054 Function Chain. The SFPR contains the Service Path Identifier (SPI) 1055 used to identify the SFP in the NSH in the data plane. It also 1056 contains a sequence of Service Indexes (SIs). Each SI identifies a 1057 hop in the SFP, and each hop is a choice between one of more SFIs. 1059 As described in this document, each Service Function Path Route is 1060 identified in the service function overlay network by an RD and an 1061 SPI. The SPI is unique within a single VPN instance supported by the 1062 underlay network. 1064 The SFPR advertisement comprises: 1066 o An RT as described in Section 4.1. 1068 o A tuple that identifies the SFPR 1070 * An RD that identifies an advertisement of an SFPR. 1072 * The SPI that uniquely identifies this path within the VPN 1073 instance distinguished by the RD. This SPI also appears in the 1074 NSH. 1076 o A series of Service Indexes. Each SI is used in the context of a 1077 particular SPI and identifies one or more SFs (distinguished by 1078 their SFTs) and for each SF a set of SFIs that instantiate the SF. 1079 The values of the SI indicate the order in which the SFs are to be 1080 executed in the SFP that is represented by the SPI. 1082 o The SI is used in the NSH to identify the entries in the SFP. 1083 Note that the SI values have meaning only relative to a specific 1084 path. They have no semantic other than to indicate the order of 1085 Service Functions within the path and are assumed to be 1086 monotonically decreasing from the start to the end of the path 1087 [RFC8300]. 1089 o Each Service Index is associated with a set of one or more Service 1090 Function Instances that can be used to provide the indexed Service 1091 Function within the path. Each member of the set comprises: 1093 * The RD used in an SFIR advertisement of the SFI. 1095 * The SFT that indicates the type of function as used in the same 1096 SFIR advertisement of the SFI. 1098 This may be summarized as follows where the notations "SFPR-RD" and 1099 "SFIR-RD" are used to distinguish the two different RDs, and where 1100 "*" indicates a multiplier: 1102 RT, {SFPR-RD, SPI}, m * {SI, {n * {SFT, p * SFIR-RD} } } 1104 Where: 1106 RT: Route Target 1108 SFPR-RD: The Route Descriptor of the Service Function Path Route 1109 advertisement 1111 SPI: Service Path Identifier used in the NSH 1113 m: The number of hops in the Service Function Path 1115 n: The number of choices of Service Function Type for a specific 1116 hop 1118 p: The number of choices of Service Function Instance for given 1119 Service Function Type in a specific hop 1121 SI: Service Index used in the NSH to indicate a specific hop 1123 SFT: The Service Function Type used in the same advertisement of 1124 the Service Function Instance Route 1126 SFIR-RD: The Route Descriptor used in an advertisement of the 1127 Service Function Instance Route 1129 That is, there can be multiple SFTs at a given hop as described in 1130 Section 5. 1132 Note that the values of SI are from the set {255, ..., 1} and are 1133 monotonically decreasing within the SFP. SIs MUST appear in order 1134 within the SFPR (i.e., monotonically decreasing) and MUST NOT appear 1135 more than once. Gaps MAY appear in the sequence as described in 1136 Section 4.5.1. Malformed SFPRs MUST be discarded and MUST cause any 1137 previous instance of the SFPR (same SFPR-RD and SPI) to be discarded. 1139 Note that if the SFIR-RD list in an SFT TLV contains one or more SFIR 1140 Pool identifiers, then in the above expression, 'p' is the sum of the 1141 number of individual SFIR-RD values and the sum for each SFIR Pool 1142 Identifier of the number of SFIRs advertised with that SFIR Pool 1143 Identifier. I.e., the list of SFIR-RD values is effectively expanded 1144 to include the SFIR-RD of each SFIR advertised with each SFIR Pool 1145 Identifier in the SFIR-RD list. 1147 The choice of SFI is explained further in Section 5. Note that an 1148 SFIR-RD value of zero has special meaning as described in that 1149 Section. 1151 4.4. Classifier Operation 1153 As shown in Figure 1, the Classifier is a component that is used to 1154 assign packets to an SFP. 1156 The Classifier is responsible for determining to which packet flow a 1157 packet belongs. The mechanism it uses to achieve that classification 1158 is out of scope of this document, but might include inspection of the 1159 packet header. The Classifier has been instructed (by the Controller 1160 or through some other configuration mechanism - see Section 7.4) 1161 which flows are to be assigned to which SFPs, and so it can impose an 1162 NSH on each packet and initialize the NSH with the SPI of the 1163 selected SFP and the SI of its first hop. 1165 Note that instructions delivered to the Classifier may include 1166 information about the metadata to encode (and the format for that 1167 encoding) on packets that are classified by the Classifier to a 1168 particular SFP. As mentioned in Section 2.2, this corresponds to the 1169 fifth element of control plane functionality described in [RFC7665]. 1170 Such instructions fall outside the scope of this specification 1171 (although, see Section 7.4), as do instructions to other SFC elements 1172 on how to interpret metadata (as described in the sixth element of 1173 control plane functionality described in [RFC7665]. 1175 4.5. Service Function Forwarder Operation 1177 Each packet sent to an SFF is transmitted encapsulated in an NSH. 1178 The NSH includes an SPI and SI: the SPI indicates the SFPR 1179 advertisement that announced the Service Function Path; the tuple 1180 SPI/SI indicates a specific hop in a specific path and maps to the 1181 RD/SFT of a particular SFIR advertisement. 1183 When an SFF gets an SFPR advertisement it will first determine 1184 whether to import the route by examining the RT. If the SFPR is 1185 imported the SFF then determines whether it is on the SFP by looking 1186 for its own SFIR-RDs or any SFIR-RD with value zero in the SFPR. For 1187 each occurrence in the SFP, the SFF creates forwarding state for 1188 incoming packets and forwarding state for outgoing packets that have 1189 been processed by the specified SFI. 1191 The SFF creates local forwarding state for packets that it receives 1192 from other SFFs. This state makes the association between the SPI/SI 1193 in the NSH of the received packet and one or more specific local SFIs 1194 as identified by the SFIR-RD/SFT. If there are multiple local SFIs 1195 that match this is because a single advertisement was made for a set 1196 of equivalent SFIs and the SFF may use local policy (such as load 1197 balancing) to determine to which SFI to forward a received packet. 1199 The SFF also creates next hop forwarding state for packets received 1200 back from the local SFI that need to be forwarded to the next hop in 1201 the SFP. There may be a choice of next hops as described in 1202 Section 4.3. The SFF could install forwarding state for all 1203 potential next hops, or it could choose to only install forwarding 1204 state to a subset of the potential next hops. If a choice is made 1205 then it will be as described in Section 5. 1207 The installed forwarding state may change over time reacting to 1208 changes in the underlay network and the availability of particular 1209 SFIs. Note that the forwarding state describes how one SFF send 1210 packets to another SFF, but not how those packets are routed through 1211 the underlay network. SFFs may be connected by tunnels across the 1212 underlay, or packets may be sent addressed to the next SFF and routed 1213 through the underlay. In any case, transmission across the underlay 1214 requires encapsulation of packets with a header for transport in the 1215 underlay network. 1217 Note that SFFs only create and store forwarding state for the SFPs on 1218 which they are included. They do not retain state for all SFPs 1219 advertised. 1221 An SFF may also install forwarding state to support looping, jumping, 1222 and branching. The protocol mechanism for explicit control of 1223 looping, jumping, and branching uses a specific reserved SFT value at 1224 a given hop of an SFPR and is described in Section 6.1. 1226 4.5.1. Processing With 'Gaps' in the SI Sequence 1228 The behavior of an SF as described in [RFC8300] is to decrement the 1229 value of the SI field in the NSH by one before returning a packet to 1230 the local SFF for further processing. This means that there is a 1231 good reason to assume that the SFP is composed of a series of SFs 1232 each indicated by an SI value one less than the previous. 1234 However, there is an advantage to having non-successive SIs in an 1235 SPI. Consider the case where an SPI needs to be modified by the 1236 insertion or removal of an SF. In the latter case this would lead to 1237 a "gap" in the sequence of SIs, and in the former case, this could 1238 only be achieved if a gap already existed into which the new SF with 1239 its new SI value could be inserted. Otherwise, all "downstream" SFs 1240 would need to be renumbered. 1242 Now, of course, such renumbering could be performed, but would lead 1243 to a significant disruption to the SFC as all the SFFs along the SFP 1244 were "reprogrammed". Thus, to achieve dynamic modification of an SFP 1245 (and even, in-service modification) it is desirable to be able to 1246 make these modifications without changing the SIs of the elements 1247 that were present before the modification. This will produce much 1248 more consistent/predictable behavior during the convergence period 1249 where otherwise the change would need to be fully propagated. 1251 Another approach says that any change to an SFP simply creates a new 1252 SFP that can be assigned a new SPI. All that would be needed would 1253 be to give a new instruction to the Classifier and traffic would be 1254 switched to the new SFP that contains the new set of SFs. This 1255 approach is practical, but neglects to consider that the SFP may be 1256 referenced by other SFPs (through "branch" instructions) and used by 1257 many Classifiers. In those cases the corresponding configuration 1258 resulting from a change in SPI may have wide ripples and give scope 1259 for errors that are hard to trace. 1261 Therefore, while this document requires that the SI values in an SFP 1262 are monotonic decreasing, it makes no assumption that the SI values 1263 are sequential. Configuration tools may apply that rule, but they 1264 are not required to. To support this, an SFF SHOULD process as 1265 follows when it receives a packet: 1267 o If the SI indicates a known entry in the SFP, the SFF MUST process 1268 the packet as normal, looking up the SI and determining to which 1269 local SFI to deliver the packet. 1271 o If the SI does not match an entry in the SFP, the SFF MUST reduce 1272 the SI value to the next (smaller) value present in the SFP and 1273 process the packet using that SI. 1275 o If there is no smaller SI (i.e., if the end of the SFP has been 1276 reached) the SFF MUST treat the SI value as invalid as described 1277 in [RFC8300]. 1279 This makes the behavior described in this document a superset of the 1280 function in [RFC8300]. That is, an implementation that strictly 1281 follows RFC 8300 in performing SI decrements in units of one, is 1282 perfectly in line with the mechanisms defined in this document. 1284 SFF implementations MAY choose to only support contiguous SI values 1285 in an SFP. Such an implementation will not support receiving an SI 1286 value that is not present in the SFP and will discard the packets as 1287 described in [RFC8300]. 1289 5. Selection within Service Function Paths 1291 As described in Section 2 the SPI/SI in the NSH passed back from an 1292 SFI to the SFF may leave the SFF with a choice of next hop SFTs, and 1293 a choice of SFIs for each SFT. That is, the SPI indicates an SFPR, 1294 and the SI indicates an entry in that SFPR. Each entry in an SFPR is 1295 a set of one or more SFT/SFIR-RD pairs. The SFF MUST choose one of 1296 these, identify the SFF that supports the chosen SFI, and send the 1297 packet to that next hop SFF. 1299 The choice be may offered for load balancing across multiple SFIs, or 1300 for discrimination between different actions necessary at a specific 1301 hop in the SFP. Different SFT values may exist at a given hop in an 1302 SFP to support several cases: 1304 o There may be multiple instances of similar service functions that 1305 are distinguished by different SFT values. For example, firewalls 1306 made by vendor A and vendor B may need to be identified by 1307 different SFT values because, while they have similar 1308 functionality, their behavior is not identical. Then, some SFPs 1309 may limit the choice of SF at a given hop by specifying the SFT 1310 for vendor A, but other SFPs might not need to control which 1311 vendor's SF is used and so can indicate that either SFT can be 1312 used. 1314 o There may be an obvious branch needed in an SFP such as the 1315 processing after a firewall where admitted packets continue along 1316 the SFP, but suspect packets are diverted to a "penalty box". In 1317 this case, the next hop in the SFP will be indicated with two 1318 different SFT values. 1320 In the typical case, the SFF chooses a next hop SFF by looking at the 1321 set of all SFFs that support the SFs identified by the SI (that set 1322 having been advertised in individual SFIR advertisements), finding 1323 the one or more that are "nearest" in the underlay network, and 1324 choosing between next hop SFFs using its own load-balancing 1325 algorithm. 1327 An SFI may influence this choice process by passing additional 1328 information back along with the packet and NSH. This information may 1329 influence local policy at the SFF to cause it to favor a next hop SFF 1330 (perhaps selecting one that is not nearest in the underlay), or to 1331 influence the load-balancing algorithm. 1333 This selection applies to the normal case, but also applies in the 1334 case of looping, jumping, and branching (see Section 6). 1336 Suppose an SFF in a particular service overlay network (identified by 1337 a particular import RT, RT-z) needs to forward an NSH-encapsulated 1338 packet whose SPI is SPI-x and whose SI is SI-y. It does the 1339 following: 1341 1. It looks for an installed SFPR that carries RT-z and that has 1342 SPI-x in its NLRI. If there is none, then such packets cannot be 1343 forwarded. 1345 2. From the SFP attribute of that SFPR, it finds the Hop TLV with SI 1346 value set to SI-y. If there is no such Hop TLV, then such 1347 packets cannot be forwarded. 1349 3. It then finds the "relevant" set of SFIRs by going through the 1350 list of SFT TLVs contained in the Hop TLV as follows: 1352 A. An SFIR is relevant if it carries RT-z, the SFT in its NLRI 1353 matches the SFT value in one of the SFT TLVs, and the RD 1354 value in its NLRI matches an entry in the list of SFIR-RDs in 1355 that SFT TLV. 1357 B. If an entry in the SFIR-RD list of an SFT TLV contains the 1358 value zero, then an SFIR is relevant if it carries RT-z and 1359 the SFT in its NLRI matches the SFT value in that SFT TLV. 1360 I.e., any SFIR in the service function overlay network 1361 defined by RT-z and with the correct SFT is relevant. 1363 C. If a pool identifier is in use then an SFIR is relevant if it 1364 is a member of the pool. 1366 Each of the relevant SFIRs identifies a single SFI, and contains a 1367 Tunnel Encapsulation attribute that specifies how to send a packet to 1368 that SFI. For a particular packet, the SFF chooses a particular SFI 1369 from the set of relevant SFIRs. This choice is made according to 1370 local policy. 1372 A typical policy might be to figure out the set of SFIs that are 1373 closest, and to load balance among them. But this is not the only 1374 possible policy. 1376 Thus, at any point in time when an SFF selects its next hop, it 1377 chooses from the intersection of the set of next hop RDs contained in 1378 the SFPR and the RDs contained in the SFF's local set of SFIRs (i.e., 1379 according to the determination of "relevance", above). If the 1380 intersection is null, the SFPR is unusable. Similarly, when this 1381 condition applies on the controller that originated the SFPR, it 1382 SHOULD either withdraw the SFPR or re-advertise it with a new set of 1383 RDs for the affected hop. 1385 6. Looping, Jumping, and Branching 1387 As described in Section 2 an SFI or an SFF may cause a packet to 1388 "loop back" to a previous SF on a path in order that a sequence of 1389 functions may be re-executed. This is simply achieved by replacing 1390 the SI in the NSH with a higher value instead of decreasing it as 1391 would normally be the case to determine the next hop in the path. 1393 Section 2 also describes how an SFI or an SFF may cause a packets to 1394 "jump forward" to an SF on a path that is not the immediate next SF 1395 in the SFP. This is simply achieved by replacing the SI in the NSH 1396 with a lower value than would be achieved by decreasing it by the 1397 normal amount. 1399 A more complex option to move packets from one SFP to another is 1400 described in [RFC8300] and Section 2 where it is termed "branching". 1401 This mechanism allows an SFI or SFF to make a choice of downstream 1402 treatments for packets based on local policy and output of the local 1403 SF. Branching is achieved by changing the SPI in the NSH to indicate 1404 the new path and setting the SI to indicate the point in the path at 1405 which the packets enter. 1407 Note that the NSH does not include a marker to indicate whether a 1408 specific packet has been around a loop before. Therefore, the use of 1409 NSH metadata ([RFC8300]) may be required in order to prevent infinite 1410 loops. 1412 6.1. Protocol Control of Looping, Jumping, and Branching 1414 If the SFT value in an SFT TLV in an SFPR has the Special Purpose SFT 1415 value "Change Sequence" (see Section 10) then this is an indication 1416 that the SFF may make a loop, jump, or branch according to local 1417 policy and information returned by the local SFI. 1419 In this case, the SPI and SI of the next hop are encoded in the eight 1420 bytes of an entry in the SFIR-RD list as follows: 1422 3 bytes SPI 1423 1 bytes SI 1425 4 bytes Reserved (SHOULD be set to zero and ignored) 1427 If the SI in this encoding is not part of the SFPR indicated by the 1428 SPI in this encoding, then this is an explicit error that SHOULD be 1429 detected by the SFF when it parses the SFPR. The SFPR SHOULD NOT 1430 cause any forwarding state to be installed in the SFF and packets 1431 received with the SPI that indicates this SFPR SHOULD be silently 1432 discarded. 1434 If the SPI in this encoding is unknown, the SFF SHOULD NOT install 1435 any forwarding state for this SFPR, but MAY hold the SFPR pending 1436 receipt of another SFPR that does use the encoded SPI. 1438 If the SPI matches the current SPI for the path, this is a loop or 1439 jump. In this case, if the SI is greater than to the current SI it 1440 is a loop. If the SPI matches and the SI is less than the next SI, 1441 it is a jump. 1443 If the SPI indicates another path, this is a branch and the SI 1444 indicates the point at which to enter that path. 1446 The Change Sequence SFT is just another SFT that may appear in a set 1447 of SFI/SFT tuples within an SI and is selected as described in 1448 Section 5. 1450 Note that Special Purpose SFTs MUST NOT be advertised in SFIRs. If 1451 such an SFIR is received it SHOULD be ignored. 1453 6.2. Implications for Forwarding State 1455 Support for looping and jumping requires that the SFF has forwarding 1456 state established to an SFF that provides access to an instance of 1457 the appropriate SF. This means that the SFF must have seen the 1458 relevant SFIR advertisements and known that it needed to create the 1459 forwarding state. This is a matter of local configuration and 1460 implementation: for example, an implementation could be configured to 1461 install forwarding state for specific looping/jumping. 1463 Support for branching requires that the SFF has forwarding state 1464 established to an SFF that provides access to an instance of the 1465 appropriate entry SF on the other SFP. This means that the SFF must 1466 have seen the relevant SFIR and SFPR advertisements and known that it 1467 needed to create the forwarding state. This is a matter of local 1468 configuration and implementation: for example, an implementation 1469 could be configured to install forwarding state for specific 1470 branching (identified by SPI and SI). 1472 7. Advanced Topics 1474 This section highlights several advanced topics introduced elsewhere 1475 in this document. 1477 7.1. Correlating Service Function Path Instances 1479 It is often useful to create bidirectional SFPs to enable packet 1480 flows to traverse the same set of SFs, but in the reverse order. 1481 However, packets on SFPs in the data plane (per [RFC8300]) do not 1482 contain a direction indicator, so each direction must use a different 1483 SPI. 1485 As described in Section 3.2.1.1 an SFPR can contain one or more 1486 correlators encoded in Association TLVs. If the Association Type 1487 indicates "Bidirectional SFP" then the SFP advertised in the SFPR is 1488 one direction of a bidirectional pair of SFPs where the other in the 1489 pair is advertised in the SFPR with RD as carried in the Associated 1490 SFPR-RD field of the Association TLV. The SPI carried in the 1491 Associated SPI field of the Association TLV provides a cross-check 1492 against the SPI advertised in the SFPR with RD as carried in the 1493 Associated SFPR-RD field of the Association TLV. 1495 As noted in Section 3.2.1.1, when SFPRs reference each other, one 1496 SFPR advertisement will be received before the other. Therefore, 1497 processing of an association will require that the first SFPR is not 1498 rejected simply because the Associated SFPR-RD it carries is unknown. 1499 However, the SFP defined by the first SFPR is valid and SHOULD be 1500 available for use as a unidirectional SFP even in the absence of an 1501 advertisement of its partner. 1503 Furthermore, in error cases where SFPR-a associates with SFPR-b, but 1504 SFPR-b associates with SFPR-c such that a bidirectional pair of SFPs 1505 cannot be formed, the individual SFPs are still valid and SHOULD be 1506 available for use as unidirectional SFPs. An implementation SHOULD 1507 log this situation because it represents a Controller error. 1509 Usage of a bidirectional SFP may be programmed into the Classifiers 1510 by the Controller. Alternatively, a Classifier may look at incoming 1511 packets on a bidirectional packet flow, extract the SPI from the 1512 received NSH, and look up the SFPR to find the reverse direction SFP 1513 to use when it sends packets. 1515 See Section 8 for an example of how this works. 1517 7.2. Considerations for Stateful Service Functions 1519 Some service functions are stateful. That means that they build and 1520 maintain state derived from configuration or from the packet flows 1521 that they handle. In such cases it can be important or necessary 1522 that all packets from a flow continue to traverse the same instance 1523 of a service function so that the state can be leveraged and does not 1524 need to be regenerated. 1526 In the case of bidirectional SFPs, it may be necessary to traverse 1527 the same instances of a stateful service function in both directions. 1528 A firewall is a good example of such a service function. 1530 This issue becomes a concern where there are multiple parallel 1531 instances of a service function and a determination of which one to 1532 use could normally be left to the SFF as a load-balancing or local 1533 policy choice. 1535 For the forward direction SFP, the concern is that the same choice of 1536 service function is made for all packets of a flow under normal 1537 network conditions. It may be possible to guarantee that the load 1538 balancing functions applied in the SFFs are stable and repeatable, 1539 but a controller that constructs SFPs might not want to trust to 1540 this. The controller can, in these cases, build a number of more 1541 specific SFPs each traversing a specific instance of the stateful 1542 SFs. In this case, the load balancing choice can be left up to the 1543 Classifier. Thus the Classifier selects which instance of a stateful 1544 SF is used by a particular flow by selecting the SFP that the flow 1545 uses. 1547 For bidirectional SFPs where the same instance of a stateful SF must 1548 be traversed in both directions, it is not enough to leave the choice 1549 of service function instance as a local choice even if the load 1550 balancing is stable because coordination would be required between 1551 the decision points in the forward and reverse directions and this 1552 may be hard to achieve in all cases except where it is the same SFF 1553 that makes the choice in both directions. 1555 Note that this approach necessarily increases the amount of SFP state 1556 in the network (i.e., there are more SFPs). It is possible to 1557 mitigate this effect by careful construction of SFPs built from a 1558 concatenation of other SFPs. 1560 Section 8.9 includes some simple examples of SFPs for stateful 1561 service functions. 1563 7.3. VPN Considerations and Private Service Functions 1565 Likely deployments include reserving specific instances of Service 1566 Functions for specific customers or allowing customers to deploy 1567 their own Service Functions within the network. Building Service 1568 Functions in such environments requires that suitable identifiers are 1569 used to ensure that SFFs distinguish which SFIs can be used and which 1570 cannot. 1572 This problem is similar to how VPNs are supported and is solved in a 1573 similar way. The RT field is used to indicate a set of Service 1574 Functions from which all choices must be made. 1576 7.4. Flow Specification for SFC Classifiers 1578 [RFC5575] and [I-D.ietf-idr-rfc5575bis] define a set of BGP routes 1579 that can be used to identify the packets in a given flow using fields 1580 in the header of each packet, and a set of actions, encoded as 1581 extended communities, that can be used to disposition those packets. 1582 This document enables the use of these mechanisms by SFC Classifiers 1583 by defining a new action extended community called "Flow 1584 Specification for SFC Classifiers" identified by the value TBD4. 1585 Note that implementation of this specification MUST NOT include other 1586 action extended communities at the same time as an SFC Classifier: 1587 the inclusion of the "Flow Specification for SFC Classifiers" action 1588 extended community along with any other action MUST be treated by 1589 implementation of this specification as an error which SHOULD result 1590 in the Flow Specification UPDATE message being handled as Treat-as- 1591 withdraw according to [RFC7606] Section 2. 1593 To put the Flow Specification into context when multiple SFC overlays 1594 are present in one network, each FlowSpec update MUST be tagged with 1595 the route target of the overlay or VPN network for which it is 1596 intended. 1598 This extended community is encoded as an 8-octet value, as shown in 1599 Figure 10. 1601 1 2 3 1602 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 1603 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1604 | Type=0x80 | Sub-Type=TBD4 | SPI | 1605 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1606 | SPI (cont.) | SI | SFT | 1607 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1609 Figure 10: The Format of the Flow Specification for SFC Classifiers 1610 Extended Community 1612 The extended community contains the Service Path Identifier (SPI), 1613 Service Index (SI), and Service Function Type (SFT) as defined 1614 elsewhere in this document. Thus, each action extended community 1615 defines the entry point (not necessarily the first hop) into a 1616 specific service function path. This allows, for example, different 1617 flows to enter the same service function path at different points. 1619 Note that a given Flow Specification update according to [RFC5575] 1620 and [I-D.ietf-idr-rfc5575bis] may include multiple of these action 1621 extended communities, and that if a given action extended community 1622 does not contain an installed SFPR with the specified {SPI, SI, SFT} 1623 it MUST NOT be used for dispositioning the packets of the specified 1624 flow. 1626 The normal case of packet classification for SFC will see a packet 1627 enter the SFP at its first hop. In this case the SI in the extended 1628 community is superfluous and the SFT may also be unnecessary. To 1629 allow these cases to be handled, a special meaning is assigned to a 1630 Service Index of zero (not a valid value) and an SFT of zero (a 1631 reserved value in the registry - see Section 10.5). 1633 o If an SFC Classifiers Extended Community is received with SI = 0 1634 then it means that the first hop of the SFP indicated by the SPI 1635 MUST be used. 1637 o If an SFC Classifiers Extended Community is received with SFT = 0 1638 then there are two sub-cases: 1640 * If there is a choice of SFT in the hop indicated by the value 1641 of the SI (including SI = 0) then SFT = 0 means there is a free 1642 choice according to local policy of which SFT to use). 1644 * If there is no choice of SFT in the hop indicated by the value 1645 of SI, then SFT = 0 means that the value of the SFT at that hop 1646 as indicated in the SFPR for the indicated SPI MUST be used. 1648 One of the filters that the Flow Specification may describe is the 1649 VPN to which the traffic belongs. Additionally, as noted above, to 1650 put the indicated SPI into context when multiple SFC overlays are 1651 present in one network, each FlowSpec update MUST be tagged with the 1652 route target of the overlay or VPN network for which it is intended. 1654 Note that future extensions might be made to the Flow Specification 1655 for SFC Classifiers Extended Community to provide instruction to the 1656 Classifier about what metadata to add to packets that it classifies 1657 for forwarding on a specific SFP, but that is outside the scope of 1658 this document. 1660 7.5. Choice of Data Plane SPI/SI Representation 1662 This document ties together the control and data planes of an SFC 1663 overlay network through the use of the SPI/SI which is nominally 1664 carried in the NSH of a given packet. However, in order to handle 1665 situations in which the NSH is not ubiquitously deployed, it is also 1666 possible to use alternative data plane representations of the SPI/SI 1667 by carrying the identical semantics in other protocol fields such as 1668 MPLS labels [RFC8595]. 1670 This document defines a new sub-TLV for the Tunnel Encapsulation 1671 attribute [I-D.ietf-idr-tunnel-encaps], the SPI/SI Representation 1672 sub-TLV of type TBD5. This sub-TLV MAY be present in each Tunnel TLV 1673 contained in a Tunnel Encapsulation attribute when the attribute is 1674 carried by an SFIR. The value field of this sub-TLV is a two octet 1675 field of flags numbered counting from the the most significant bit, 1676 each of which describes how the originating SFF expects to see the 1677 SPI/SI represented in the data plane for packets carried in the 1678 tunnels described by the Tunnel TLV. 1680 The following bits are defined by this document and are tracked in an 1681 IANA registry desribed in Section 10.10: 1683 Bit TBD9: If this bit is set the NSH is to be used to carry the SPI/ 1684 SI in the data plane. 1686 Bit TBD10: If this bit is set two labels in an MPLS label stack are 1687 to be used as described in Section 7.5.1. 1689 If a given Tunnel TLV does not contain an SPI/SI Representation sub- 1690 TLV then it MUST be processed as if such a sub-TLV is present with 1691 Bit TBD9 set and no other bits set. That is, the absence of the sub- 1692 TLV SHALL be interpreted to mean that the NSH is to be used. 1694 If a given Tunnel TLV contains an SPI/SI Representation sub-TLV with 1695 value field that has no flag set then the tunnel indicated by the 1696 Tunnel TLV MUST NOT be used for forwarding SFC packets. If a given 1697 Tunnel TLV contains an SPI/SI Representation sub-TLV with both bit 1698 TBD9 and bit TBD10 set then the tunnel indicated by the Tunnel TLV 1699 MUST NOT be used for forwarding SFC packets. The meaning and rules 1700 for presence of other bits is to be defined in future documents, but 1701 implementations of this specification MUST set other bits to zero and 1702 ignore them on receipt. 1704 If a given Tunnel TLV contains more than one SPI/SI Representation 1705 sub-TLV then the first one MUST be considered and subsequent 1706 instances MUST be ignored. 1708 Note that the MPLS representation of the logical NSH may be used even 1709 if the tunnel is not an MPLS tunnel. Conversely, MPLS tunnels may be 1710 used to carry other encodings of the logical NSH (specifically, the 1711 NSH itself). It is a requirement that both ends of a tunnel over the 1712 underlay network know that the tunnel is used for SFC and know what 1713 form of NSH representation is used. The signaling mechanism 1714 described here allows coordination of this information. 1716 7.5.1. MPLS Representation of the SPI/SI 1718 If bit TBD10 is set in the in the SPI/SI Representation sub-TLV then 1719 labels in the MPLS label stack are used to indicate SFC forwarding 1720 and processing instructions to achieve the semantics of a logical 1721 NSH. The label stack is encoded as shown in [RFC8595]. 1723 7.6. MPLS Label Swapping/Stacking Operation 1725 When a Classifier constructs an MPLS label stack for an SFP it starts 1726 with that SFP's last hop. If the last hop requires an {SPI, SI} 1727 label pair for label swapping, it pushes the SI (set to the SI value 1728 of the last hop) and the SFP's SPI onto the MPLS label stack. If the 1729 last hop requires a {context label, SFI label} label pair for label 1730 stacking it selects a specific SFIR and pushes that SFIR's SFI label 1731 and context label onto the MPLS label stack. 1733 The Classifier then moves sequentially back through the SFP one hop 1734 at a time. For each hop, if the hop requires an {SPI, SI]} and there 1735 is an {SPI, SI} at the top of the MPLS label stack, the SI is set to 1736 the SI value of the current hop. If there is not an {SPI, SI} at the 1737 top of the MPLS label stack, it pushes the SI (set to the SI value of 1738 the current hop) and the SFP's SPI onto the MPLS label stack. 1740 If the hop requires a {context label, SFI label}, it selects a 1741 specific SFIR and pushes that SFIR's SFI label and context label onto 1742 the MPLS label stack. 1744 7.7. Support for MPLS-Encapsulated NSH Packets 1746 [RFC8596] describes how to transport SFC packets using the NSH over 1747 an MPLS transport network. Signaling MPLS encapsulation of SFC 1748 packets using the NSH is also supported by this document by using the 1749 "BGP Tunnel Encapsulation Attribute Sub-TLV" with the codepoint 10 1750 (representing "MPLS Label Stack") from the "BGP Tunnel Encapsulation 1751 Attribute Sub-TLVs" registry defined in [I-D.ietf-idr-tunnel-encaps], 1752 and also using the "SFP Traversal With MPLS Label Stack TLV" and the 1753 "SPI/SI Representation sub-TLV" with bit TBD9 set and bit TBD10 1754 cleared. 1756 In this case the MPLS label stack constructed by the SFF to forward a 1757 packet to the next SFF on the SFP will consist of the labels needed 1758 to reach that SFF, and if label stacking is used it will also include 1759 the labels advertised in the MPLS Label Stack sub-TLV and the labels 1760 remaining in the stack needed to traverse the remainder of the SFP. 1762 8. Examples 1764 Most of the examples in this section use IPv4 addressing. But there 1765 is nothing special about IPv4 in the mechanisms described in this 1766 document, and they are equally applicable to IPv6. A few examples 1767 using IPv6 addressing are provided in Section 8.10. 1769 Assume we have a service function overlay network with four SFFs 1770 (SFF1, SFF3, SFF3, and SFF4). The SFFs have addresses in the 1771 underlay network as follows: 1773 SFF1 192.0.2.1 1774 SFF2 192.0.2.2 1775 SFF3 192.0.2.3 1776 SFF4 192.0.2.4 1778 Each SFF provides access to some SFIs from the four Service Function 1779 Types SFT=41, SFT=42, SFT=43, and SFT=44 as follows: 1781 SFF1 SFT=41 and SFT=42 1782 SFF2 SFT=41 and SFT=43 1783 SFF3 SFT=42 and SFT=44 1784 SFF4 SFT=43 and SFT=44 1786 The service function network also contains a Controller with address 1787 198.51.100.1. 1789 This example service function overlay network is shown in Figure 11. 1791 -------------- 1792 | Controller | 1793 | 198.51.100.1 | ------ ------ ------ ------ 1794 -------------- | SFI | | SFI | | SFI | | SFI | 1795 |SFT=41| |SFT=42| |SFT=41| |SFT=43| 1796 ------ ------ ------ ------ 1797 \ / \ / 1798 --------- --------- 1799 ---------- | SFF1 | | SFF2 | 1800 Packet --> | | |192.0.2.1| |192.0.2.2| 1801 Flows --> |Classifier| --------- --------- -->Dest 1802 | | --> 1803 ---------- --------- --------- 1804 | SFF3 | | SFF4 | 1805 |192.0.2.3| |192.0.2.4| 1806 --------- --------- 1807 / \ / \ 1808 ------ ------ ------ ------ 1809 | SFI | | SFI | | SFI | | SFI | 1810 |SFT=42| |SFT=44| |SFT=43| |SFT=44| 1811 ------ ------ ------ ------ 1813 Figure 11: Example Service Function Overlay Network 1815 The SFFs advertise routes to the SFIs they support. So we see the 1816 following SFIRs: 1818 RD = 192.0.2.1/1, SFT = 41 1819 RD = 192.0.2.1/2, SFT = 42 1820 RD = 192.0.2.2/1, SFT = 41 1821 RD = 192.0.2.2/2, SFT = 43 1822 RD = 192.0.2.3/7, SFT = 42 1823 RD = 192.0.2.3/8, SFT = 44 1824 RD = 192.0.2.4/5, SFT = 43 1825 RD = 192.0.2.4/6, SFT = 44 1827 Note that the addressing used for communicating between SFFs is taken 1828 from the Tunnel Encapsulation attribute of the SFIR and not from the 1829 SFIR-RD. 1831 8.1. Example Explicit SFP With No Choices 1833 Consider the following SFPR. 1835 SFP1: RD = 198.51.100.1/101, SPI = 15, 1836 [SI = 255, SFT = 41, RD = 192.0.2.1/1], 1837 [SI = 250, SFT = 43, RD = 192.0.2.2/2] 1839 The Service Function Path consists of an SF of type 41 located at 1840 SFF1 followed by an SF of type 43 located at SFF2. This path is 1841 fully explicit and each SFF is offered no choice in forwarding 1842 packets along the path. 1844 SFF1 will receive packets on the path from the Classifier and will 1845 identify the path from the SPI (15). The initial SI will be 255 and 1846 so SFF1 will deliver the packets to the SFI for SFT 41. 1848 When the packets are returned to SFF1 by the SFI the SI will be 1849 decreased to 250 for the next hop. SFF1 has no flexibility in the 1850 choice of SFF to support the next hop SFI and will forward the packet 1851 to SFF2 which will send the packets to the SFI that supports SFT 43 1852 before forwarding the packets to their destinations. 1854 8.2. Example SFP With Choice of SFIs 1856 SFP2: RD = 198.51.100.1/102, SPI = 16, 1857 [SI = 255, SFT = 41, RD = 192.0.2.1/1], 1858 [SI = 250, SFT = 43, {RD = 192.0.2.2/2, 1859 RD = 192.0.2.4/5 } ] 1861 In this example the path also consists of an SF of type 41 located at 1862 SFF1 and this is followed by an SF of type 43, but in this case the 1863 SI = 250 contains a choice between the SFI located at SFF2 and the 1864 SFI located at SFF4. 1866 SFF1 will receive packets on the path from the Classifier and will 1867 identify the path from the SPI (16). The initial SI will be 255 and 1868 so SFF1 will deliver the packets to the SFI for SFT 41. 1870 When the packets are returned to SFF1 by the SFI the SI will be 1871 decreased to 250 for the next hop. SFF1 now has a choice of next hop 1872 SFF to execute the next hop in the path. It can either forward 1873 packets to SFF2 or SFF4 to execute a function of type 43. It uses 1874 its local load balancing algorithm to make this choice. The chosen 1875 SFF will send the packets to the SFI that supports SFT 43 before 1876 forwarding the packets to their destinations. 1878 8.3. Example SFP With Open Choice of SFIs 1880 SFP3: RD = 198.51.100.1/103, SPI = 17, 1881 [SI = 255, SFT = 41, RD = 192.0.2.1/1], 1882 [SI = 250, SFT = 44, RD = 0] 1884 In this example the path also consists of an SF of type 41 located at 1885 SFF1 and this is followed by an SI with an RD of zero and SF of type 1886 44. This means that a choice can be made between any SFF that 1887 supports an SFI of type 44. 1889 SFF1 will receive packets on the path from the Classifier and will 1890 identify the path from the SPI (17). The initial SI will be 255 and 1891 so SFF1 will deliver the packets to the SFI for SFT 41. 1893 When the packets are returned to SFF1 by the SFI the SI will be 1894 decreased to 250 for the next hop. SFF1 now has a free choice of 1895 next hop SFF to execute the next hop in the path selecting between 1896 all SFFs that support SFs of type 44. Looking at the SFIRs it has 1897 received, SFF1 knows that SF type 44 is supported by SFF3 and SFF4. 1898 SFF1 uses its local load balancing algorithm to make this choice. 1899 The chosen SFF will send the packets to the SFI that supports SFT 44 1900 before forwarding the packets to their destinations. 1902 8.4. Example SFP With Choice of SFTs 1904 SFP4: RD = 198.51.100.1/104, SPI = 18, 1905 [SI = 255, SFT = 41, RD = 192.0.2.1/1], 1906 [SI = 250, {SFT = 43, RD = 192.0.2.2/2, 1907 SFT = 44, RD = 192.0.2.3/8 } ] 1909 This example provides a choice of SF type in the second hop in the 1910 path. The SI of 250 indicates a choice between SF type 43 located at 1911 SF2 and SF type 44 located at SF3. 1913 SFF1 will receive packets on the path from the Classifier and will 1914 identify the path from the SPI (18). The initial SI will be 255 and 1915 so SFF1 will deliver the packets to the SFI for SFT 41. 1917 When the packets are returned to SFF1 by the SFI the SI will be 1918 decreased to 250 for the next hop. SFF1 now has a free choice of 1919 next hop SFF to execute the next hop in the path selecting between 1920 all SFFs that support an SF of type 43 and SFF3 that supports an SF 1921 of type 44. These may be completely different functions that are to 1922 be executed dependent on specific conditions, or may be similar 1923 functions identified with different type identifiers (such as 1924 firewalls from different vendors). SFF1 uses its local policy and 1925 load balancing algorithm to make this choice, and may use additional 1926 information passed back from the local SFI to help inform its 1927 selection. The chosen SFF will send the packets to the SFI that 1928 supports the chose SFT before forwarding the packets to their 1929 destinations. 1931 8.5. Example Correlated Bidirectional SFPs 1933 SFP5: RD = 198.51.100.1/105, SPI = 19, 1934 Assoc-Type = 1, Assoc-RD = 198.51.100.1/106, Assoc-SPI = 20, 1935 [SI = 255, SFT = 41, RD = 192.0.2.1/1], 1936 [SI = 250, SFT = 43, RD = 192.0.2.2/2] 1938 SFP6: RD = 198.51.100.1/106, SPI = 20, 1939 Assoc-Type = 1, Assoc-RD = 198.51.100.1/105, Assoc-SPI = 19, 1940 [SI = 254, SFT = 43, RD = 192.0.2.2/2], 1941 [SI = 249, SFT = 41, RD = 192.0.2.1/1] 1943 This example demonstrates correlation of two SFPs to form a 1944 bidirectional SFP as described in Section 7.1. 1946 Two SFPRs are advertised by the Controller. They have different SPIs 1947 (19 and 20) so they are known to be separate SFPs, but they both have 1948 Association TLVs with Association Type set to 1 indicating 1949 bidirectional SFPs. Each has an Associated SFPR-RD field containing 1950 the value of the other SFPR-RD to correlated the two SFPs as a 1951 bidirectional pair. 1953 As can be seen from the SFPRs in this example, the paths are 1954 symmetric: the hops in SFP5 appear in the reverse order in SFP6. 1956 8.6. Example Correlated Asymmetrical Bidirectional SFPs 1957 SFP7: RD = 198.51.100.1/107, SPI = 21, 1958 Assoc-Type = 1, Assoc-RD = 198.51.100.1/108, Assoc-SPI = 22, 1959 [SI = 255, SFT = 41, RD = 192.0.2.1/1], 1960 [SI = 250, SFT = 43, RD = 192.0.2.2/2] 1962 SFP8: RD = 198.51.100.1/108, SPI = 22, 1963 Assoc-Type = 1, Assoc-RD = 198.51.100.1/107, Assoc-SPI = 21, 1964 [SI = 254, SFT = 44, RD = 192.0.2.4/6], 1965 [SI = 249, SFT = 41, RD = 192.0.2.1/1] 1967 Asymmetric bidirectional SFPs can also be created. This example 1968 shows a pair of SFPs with distinct SPIs (21 and 22) that are 1969 correlated in the same way as in the example in Section 8.5. 1971 However, unlike in that example, the SFPs are different in each 1972 direction. Both paths include a hop of SF type 41, but SFP7 includes 1973 a hop of SF type 43 supported at SFF2 while SFP8 includes a hop of SF 1974 type 44 supported at SFF4. 1976 8.7. Example Looping in an SFP 1978 SFP9: RD = 198.51.100.1/109, SPI = 23, 1979 [SI = 255, SFT = 41, RD = 192.0.2.1/1], 1980 [SI = 250, SFT = 44, RD = 192.0.2.4/5], 1981 [SI = 245, {SFT = 1, RD = {SPI=23, SI=255, Rsv=0}, 1982 SFT = 42, RD = 192.0.2.3/7 } ] 1984 Looping and jumping are described in Section 6. This example shows 1985 an SFP that contains an explicit loop-back instruction that is 1986 presented as a choice within an SFP hop. 1988 The first two hops in the path (SI = 255 and SI = 250) are normal. 1989 That is, the packets will be delivered to SFF1 and SFF4 in turn for 1990 execution of SFs of type 41 and 44 respectively. 1992 The third hop (SI = 245) presents SFF4 with a choice of next hop. It 1993 can either forward the packets to SFF3 for an SF of type 42 (the 1994 second choice), or it can loop back. 1996 The loop-back entry in the SFPR for SI = 245 is indicated by the 1997 special purpose SFT value 1 ("Change Sequence"). Within this hop, 1998 the RD is interpreted as encoding the SPI and SI of the next hop (see 1999 Section 6.1. In this case the SPI is 23 which indicates that this is 2000 loop or branch: i.e., the next hop is on the same SFP. The SI is set 2001 to 255: this is a higher number than the current SI (245) indicating 2002 a loop. 2004 SFF4 must make a choice between these two next hops. Either the 2005 packets will be forwarded to SFF3 with the NSH SI decreased to 245 or 2006 looped back to SFF1 with the NSH SI reset to 255. This choice will 2007 be made according to local policy, information passed back by the 2008 local SFI, and details in the packets' metadata that are used to 2009 prevent infinite looping. 2011 8.8. Example Branching in an SFP 2013 SFP10: RD = 198.51.100.1/110, SPI = 24, 2014 [SI = 254, SFT = 42, RD = 192.0.2.3/7], 2015 [SI = 249, SFT = 43, RD = 192.0.2.2/2] 2017 SFP11: RD = 198.51.100.1/111, SPI = 25, 2018 [SI = 255, SFT = 41, RD = 192.0.2.1/1], 2019 [SI = 250, SFT = 1, RD = {SPI=24, SI=254, Rsv=0}] 2021 Branching follows a similar procedure to that for looping (and 2022 jumping) as shown in Section 8.7 however there are two SFPs involved. 2024 SFP10 shows a normal path with packets forwarded to SFF3 and SFF2 for 2025 execution of service functions of type 42 and 43 respectively. 2027 SFP11 starts as normal (SFF1 for an SF of type 41), but then SFF1 2028 processes the next hop in the path and finds a "Change Sequence" 2029 Special Purpose SFT. The SFIR-RD field includes an SPI of 24 which 2030 indicates SFP10, not the current SFP. The SI in the SFIR-RD is 254, 2031 so SFF1 knows that it must set the SPI/SI in the NSH to 24/254 and 2032 send the packets to the appropriate SFF as advertised in the SFPR for 2033 SFP10 (that is, SFF3). 2035 8.9. Examples of SFPs with Stateful Service Functions 2037 This section provides some examples to demonstrate establishing SFPs 2038 when there is a choice of service functions at a particular hop, and 2039 where consistency of choice is required in both directions. The 2040 scenarios that give rise to this requirement are discussed in 2041 Section 7.2. 2043 8.9.1. Forward and Reverse Choice Made at the SFF 2045 Consider the topology shown in Figure 12. There are three SFFs 2046 arranged neatly in a line, and the middle one (SFF2) supports three 2047 SFIs all of SFT 42. These three instances can be used by SFF2 to 2048 load balance so that no one instance is swamped. 2050 ------ ------ ------ ------ ------ 2051 | SFI | | SFIa | | SFIb | | SFIc | | SFI | 2052 |SFT=41| |SFT=42| |SFT=42| |SFT=42| |SFT=43| 2053 ------ ------\ ------ /------ ------ 2054 \ \ | / / 2055 --------- --------- --------- 2056 ---------- | SFF1 | | SFF2 | | SFF3 | 2057 --> | |..|192.0.2.1|...|192.0.2.2|...|192.0.2.3|--> 2058 --> |Classifier| --------- --------- --------- 2059 | | 2060 ---------- 2062 Figure 12: Example Where Choice is Made at the SFF 2064 This leads to the following SFIRs being advertised. 2066 RD = 192.0.2.1/11, SFT = 41 2067 RD = 192.0.2.2/11, SFT = 42 (for SFIa) 2068 RD = 192.0.2.2/12, SFT = 42 (for SFIb) 2069 RD = 192.0.2.2/13, SFT = 42 (for SFIc) 2070 RD = 192.0.2.3/11, SFT = 43 2072 The controller can create a single forward SFP (SFP12) giving SFF2 2073 the choice of which SFI to use to provide function of SFT 42 as 2074 follows. The load-balancing choice between the three available SFIs 2075 is assumed to be within the capabilities of the SFF and if the SFs 2076 are stateful it is assumed that the SFF knows this and arranges load 2077 balancing in a stable, flow-dependent way. 2079 SFP12: RD = 198.51.100.1/112, SPI = 26, 2080 Assoc-Type = 1, Assoc-RD = 198.51.100.1/113, Assoc-SPI = 27, 2081 [SI = 255, SFT = 41, RD = 192.0.2.1/11], 2082 [SI = 254, SFT = 42, {RD = 192.0.2.2/11, 2083 192.0.2.2/12, 2084 192.0.2.2/13 }], 2085 [SI = 253, SFT = 43, RD = 192.0.2.3/11] 2087 The reverse SFP (SFP13) in this case may also be created as shown 2088 below using association with the forward SFP and giving the load- 2089 balancing choice to SFF2. This is safe, even in the case that the 2090 SFs of type 42 are stateful because SFF2 is doing the load balancing 2091 in both directions and can apply the same algorithm to ensure that 2092 packets associated with the same flow use the same SFI regardless of 2093 the direction of travel. 2095 SFP13: RD = 198.51.100.1/113, SPI = 27, 2096 Assoc-Type = 1, Assoc-RD = 198.51.100.1/112, Assoc-SPI = 26, 2097 [SI = 255, SFT = 43, RD = 192.0.2.3/11], 2098 [SI = 254, SFT = 42, {RD = 192.0.2.2/11, 2099 192.0.2.2/12, 2100 192.0.2.2/13 }], 2101 [SI = 253, SFT = 41, RD = 192.0.2.1/11] 2103 How an SFF knows that an attached SFI is stateful is out of scope of 2104 this document. It is assumed that this will form part of the process 2105 by which SFIs are registered as local to SFFs. Section 7.2 provides 2106 additional observations about the coordination of the use of stateful 2107 SFIs in the case of bidirectional SFPs. 2109 In general, the problems of load balancing and the selection of the 2110 same SFIs in both directions of a bidirectional SFP can be addressed 2111 by using sufficiently precisely specified SFPs (specifying the exact 2112 SFIs to use) and suitable programming of the Classifiers at each end 2113 of the SFPs to make sure that the matching pair of SFPs are used. 2115 8.9.2. Parallel End-to-End SFPs with Shared SFF 2117 The mechanism described in Section 8.9.1 might not be desirable 2118 because of the functional assumptions it places on SFF2 to be able to 2119 load balance with suitable flow identification, stability, and 2120 equality in both directions. Instead, it may be desirable to place 2121 the responsibility for flow classification in the Classifier and let 2122 it determine load balancing with the implied choice of SFIs. 2124 Consider the network graph as shown in Figure 12 and with the same 2125 set of SFIRs as listed in Section 8.9.1. In this case the controller 2126 could specify three forward SFPs with their corresponding associated 2127 reverse SFPs. Each bidirectional pair of SFPs uses a different SFI 2128 for the SF of type 42. The controller can instruct the Classifier 2129 how to place traffic on the three bidirectional SFPs, or can treat 2130 them as a group leaving the Classifier responsible for balancing the 2131 load. 2133 SFP14: RD = 198.51.100.1/114, SPI = 28, 2134 Assoc-Type = 1, Assoc-RD = 198.51.100.1/117, Assoc-SPI = 31, 2135 [SI = 255, SFT = 41, RD = 192.0.2.1/11], 2136 [SI = 254, SFT = 42, RD = 192.0.2.2/11], 2137 [SI = 253, SFT = 43, RD = 192.0.2.3/11] 2139 SFP15: RD = 198.51.100.1/115, SPI = 29, 2140 Assoc-Type = 1, Assoc-RD = 198.51.100.1/118, Assoc-SPI = 32, 2141 [SI = 255, SFT = 41, RD = 192.0.2.1/11], 2142 [SI = 254, SFT = 42, RD = 192.0.2.2/12], 2143 [SI = 253, SFT = 43, RD = 192.0.2.3/11] 2145 SFP16: RD = 198.51.100.1/116, SPI = 30, 2146 Assoc-Type = 1, Assoc-RD = 198.51.100.1/119, Assoc-SPI = 33, 2147 [SI = 255, SFT = 41, RD = 192.0.2.1/11], 2148 [SI = 254, SFT = 42, RD = 192.0.2.2/13], 2149 [SI = 253, SFT = 43, RD = 192.0.2.3/11] 2151 SFP17: RD = 198.51.100.1/117, SPI = 31, 2152 Assoc-Type = 1, Assoc-RD = 198.51.100.1/114, Assoc-SPI = 28, 2153 [SI = 255, SFT = 43, RD = 192.0.2.3/11], 2154 [SI = 254, SFT = 42, RD = 192.0.2.2/11], 2155 [SI = 253, SFT = 41, RD = 192.0.2.1/11] 2157 SFP18: RD = 198.51.100.1/118, SPI = 32, 2158 Assoc-Type = 1, Assoc-RD = 198.51.100.1/115, Assoc-SPI = 29, 2159 [SI = 255, SFT = 43, RD = 192.0.2.3/11], 2160 [SI = 254, SFT = 42, RD = 192.0.2.2/12], 2161 [SI = 253, SFT = 41, RD = 192.0.2.1/11] 2163 SFP19: RD = 198.51.100.1/119, SPI = 33, 2164 Assoc-Type = 1, Assoc-RD = 198.51.100.1/116, Assoc-SPI = 30, 2165 [SI = 255, SFT = 43, RD = 192.0.2.3/11], 2166 [SI = 254, SFT = 42, RD = 192.0.2.2/13], 2167 [SI = 253, SFT = 41, RD = 192.0.2.1/11] 2169 8.9.3. Parallel End-to-End SFPs with Separate SFFs 2171 While the examples in Section 8.9.1 and Section 8.9.2 place the 2172 choice of SFI as subtended from the same SFF, it is also possible 2173 that the SFIs are each subtended from a different SFF as shown in 2174 Figure 13. In this case it is harder to coordinate the choices for 2175 forward and reverse paths without some form of coordination between 2176 SFF1 and SFF3. Therefore it would be normal to consider end-to-end 2177 parallel SFPs as described in Section 8.9.2. 2179 ------ 2180 | SFIa | 2181 |SFT=42| 2182 ------ 2183 ------ | 2184 | SFI | --------- 2185 |SFT=41| | SFF5 | 2186 ------ ..|192.0.2.5|.. 2187 | ..: --------- :.. 2188 ---------.: :.--------- 2189 ---------- | SFF1 | --------- | SFF3 | 2190 --> | |..|192.0.2.1|....| SFF6 |....|192.0.2.3| --> 2191 --> |Classifier| ---------: |192.0.2.6| :--------- 2192 | | : --------- : | 2193 ---------- : | : ------ 2194 : ------ : | SFI | 2195 :.. | SFIb | ..: |SFT=43| 2196 :.. |SFT=42| ..: ------ 2197 : ------ : 2198 :.---------.: 2199 | SFF7 | 2200 |192.0.2.7| 2201 --------- 2202 | 2203 ------ 2204 | SFIc | 2205 |SFT=42| 2206 ------ 2208 Figure 13: Second Example With Parallel End-to-End SFPs 2210 In this case, five SFIRs are advertised as follows: 2212 RD = 192.0.2.1/11, SFT = 41 2213 RD = 192.0.2.5/11, SFT = 42 (for SFIa) 2214 RD = 192.0.2.6/11, SFT = 42 (for SFIb) 2215 RD = 192.0.2.7/11, SFT = 42 (for SFIc) 2216 RD = 192.0.2.3/11, SFT = 43 2218 In this case the controller could specify three forward SFPs with 2219 their corresponding associated reverse SFPs. Each bidirectional pair 2220 of SFPs uses a different SFF and SFI for middle hop (for an SF of 2221 type 42). The controller can instruct the Classifier how to place 2222 traffic on the three bidirectional SFPs, or can treat them as a group 2223 leaving the Classifier responsible for balancing the load. 2225 SFP20: RD = 198.51.100.1/120, SPI = 34, 2226 Assoc-Type = 1, Assoc-RD = 198.51.100.1/123, Assoc-SPI = 37, 2227 [SI = 255, SFT = 41, RD = 192.0.2.1/11], 2228 [SI = 254, SFT = 42, RD = 192.0.2.5/11], 2229 [SI = 253, SFT = 43, RD = 192.0.2.3/11] 2231 SFP21: RD = 198.51.100.1/121, SPI = 35, 2232 Assoc-Type = 1, Assoc-RD = 198.51.100.1/124, Assoc-SPI = 38, 2233 [SI = 255, SFT = 41, RD = 192.0.2.1/11], 2234 [SI = 254, SFT = 42, RD = 192.0.2.6/11], 2235 [SI = 253, SFT = 43, RD = 192.0.2.3/11] 2237 SFP22: RD = 198.51.100.1/122, SPI = 36, 2238 Assoc-Type = 1, Assoc-RD = 198.51.100.1/125, Assoc-SPI = 39, 2239 [SI = 255, SFT = 41, RD = 192.0.2.1/11], 2240 [SI = 254, SFT = 42, RD = 192.0.2.7/11], 2241 [SI = 253, SFT = 43, RD = 192.0.2.3/11] 2243 SFP23: RD = 198.51.100.1/123, SPI = 37, 2244 Assoc-Type = 1, Assoc-RD = 198.51.100.1/120, Assoc-SPI = 34, 2245 [SI = 255, SFT = 43, RD = 192.0.2.3/11], 2246 [SI = 254, SFT = 42, RD = 192.0.2.5/11], 2247 [SI = 253, SFT = 41, RD = 192.0.2.1/11] 2249 SFP24: RD = 198.51.100.1/124, SPI = 38, 2250 Assoc-Type = 1, Assoc-RD = 198.51.100.1/121, Assoc-SPI = 35, 2251 [SI = 255, SFT = 43, RD = 192.0.2.3/11], 2252 [SI = 254, SFT = 42, RD = 192.0.2.6/11], 2253 [SI = 253, SFT = 41, RD = 192.0.2.1/11] 2255 SFP25: RD = 198.51.100.1/125, SPI = 39, 2256 Assoc-Type = 1, Assoc-RD = 198.51.100.1/122, Assoc-SPI = 36, 2257 [SI = 255, SFT = 43, RD = 192.0.2.3/11], 2258 [SI = 254, SFT = 42, RD = 192.0.2.7/11], 2259 [SI = 253, SFT = 41, RD = 192.0.2.1/11] 2261 8.9.4. Parallel SFPs Downstream of the Choice 2263 The mechanism of parallel SFPs demonstrated in Section 8.9.3 is 2264 perfectly functional and may be practical in many environments. 2265 However, there may be scaling concerns because of the large amount of 2266 state (knowledge of SFPs, i.e., SFPR advertisements retained) if 2267 there is a very large amount of choice of SFIs (for example, tens of 2268 instances of the same stateful SF), or if there are multiple choices 2269 of stateful SF along a path. This situation may be mitigated using 2270 SFP fragments that are combined to form the end to end SFPs. 2272 The example presented here is necessarily simplistic, but should 2273 convey the basic principle. The example presented in Figure 14 is 2274 similar to that in Section 8.9.3 but with an additional first hop. 2276 ------ 2277 | SFIa | 2278 |SFT=43| 2279 ------ 2280 ------ ------ | 2281 | SFI | | SFI | --------- 2282 |SFT=41| |SFT=42| | SFF5 | 2283 ------ ------ ..|192.0.2.5|.. 2284 | | ..: --------- :.. 2285 --------- ---------.: :.--------- 2286 ------ | SFF1 | | SFF2 | --------- | SFF3 | 2287 -->|Class-|.|192.0.2.1|.|192.0.2.2|....| SFF6 |....|192.0.2.3|--> 2288 -->| ifier| --------- ---------: |192.0.2.6| :--------- 2289 ------ : --------- : | 2290 : | : ------ 2291 : ------ : | SFI | 2292 :.. | SFIb | ..: |SFT=44| 2293 :.. |SFT=43| ..: ------ 2294 : ------ : 2295 :.---------.: 2296 | SFF7 | 2297 |192.0.2.7| 2298 --------- 2299 | 2300 ------ 2301 | SFIc | 2302 |SFT=43| 2303 ------ 2305 Figure 14: Example With Parallel SFPs Downstream of Choice 2307 The six SFIs are advertised as follows: 2309 RD = 192.0.2.1/11, SFT = 41 2310 RD = 192.0.2.2/11, SFT = 42 2311 RD = 192.0.2.5/11, SFT = 43 (for SFIa) 2312 RD = 192.0.2.6/11, SFT = 43 (for SFIb) 2313 RD = 192.0.2.7/11, SFT = 43 (for SFIc) 2314 RD = 192.0.2.3/11, SFT = 44 2316 SFF2 is the point at which a load balancing choice must be made. So 2317 "tail-end" SFPs are constructed as follows. Each takes in a 2318 different SFF that provides access to an SF of type 43. 2320 SFP26: RD = 198.51.100.1/126, SPI = 40, 2321 Assoc-Type = 1, Assoc-RD = 198.51.100.1/130, Assoc-SPI = 44, 2322 [SI = 255, SFT = 43, RD = 192.0.2.5/11], 2323 [SI = 254, SFT = 44, RD = 192.0.2.3/11] 2325 SFP27: RD = 198.51.100.1/127, SPI = 41, 2326 Assoc-Type = 1, Assoc-RD = 198.51.100.1/131, Assoc-SPI = 45, 2327 [SI = 255, SFT = 43, RD = 192.0.2.6/11], 2328 [SI = 254, SFT = 44, RD = 192.0.2.3/11] 2330 SFP28: RD = 198.51.100.1/128, SPI = 42, 2331 Assoc-Type = 1, Assoc-RD = 198.51.100.1/132, Assoc-SPI = 46, 2332 [SI = 255, SFT = 43, RD = 192.0.2.7/11], 2333 [SI = 254, SFT = 44, RD = 192.0.2.3/11] 2335 Now an end-to-end SFP with load balancing choice can be constructed 2336 as follows. The choice made by SFF2 is expressed in terms of 2337 entering one of the three "tail end" SFPs. 2339 SFP29: RD = 198.51.100.1/129, SPI = 43, 2340 [SI = 255, SFT = 41, RD = 192.0.2.1/11], 2341 [SI = 254, SFT = 42, RD = 192.0.2.2/11], 2342 [SI = 253, {SFT = 1, RD = {SPI=40, SI=255, Rsv=0}, 2343 RD = {SPI=41, SI=255, Rsv=0}, 2344 RD = {SPI=42, SI=255, Rsv=0} } ] 2346 Now, despite the load balancing choice being made other than at the 2347 initial Classifier, it is possible for the reverse SFPs to be well- 2348 constructed without any ambiguity. The three reverse paths appear as 2349 follows. 2351 SFP30: RD = 198.51.100.1/130, SPI = 44, 2352 Assoc-Type = 1, Assoc-RD = 198.51.100.1/126, Assoc-SPI = 40, 2353 [SI = 255, SFT = 44, RD = 192.0.2.4/11], 2354 [SI = 254, SFT = 43, RD = 192.0.2.5/11], 2355 [SI = 253, SFT = 42, RD = 192.0.2.2/11], 2356 [SI = 252, SFT = 41, RD = 192.0.2.1/11] 2358 SFP31: RD = 198.51.100.1/131, SPI = 45, 2359 Assoc-Type = 1, Assoc-RD = 198.51.100.1/127, Assoc-SPI = 41, 2360 [SI = 255, SFT = 44, RD = 192.0.2.4/11], 2361 [SI = 254, SFT = 43, RD = 192.0.2.6/11], 2362 [SI = 253, SFT = 42, RD = 192.0.2.2/11], 2363 [SI = 252, SFT = 41, RD = 192.0.2.1/11] 2365 SFP32: RD = 198.51.100.1/132, SPI = 46, 2366 Assoc-Type = 1, Assoc-RD = 198.51.100.1/128, Assoc-SPI = 42, 2367 [SI = 255, SFT = 44, RD = 192.0.2.4/11], 2368 [SI = 254, SFT = 43, RD = 192.0.2.7/11], 2369 [SI = 253, SFT = 42, RD = 192.0.2.2/11], 2370 [SI = 252, SFT = 41, RD = 192.0.2.1/11] 2372 8.10. Examples Using IPv6 Addressing 2374 This section provides several examples using IPv6 addressing. As 2375 will be seen from the examples, there is nothing special or clever 2376 about using IPv6 addressing rather than IPv4 addressing. 2378 The reference network for these IPv6 examples is based on that 2379 described at the top of Section 8 and shown in Figure 11. 2381 Assume we have a service function overlay network with four SFFs 2382 (SFF1, SFF3, SFF3, and SFF4). The SFFs have addresses in the 2383 underlay network as follows: 2385 SFF1 2001:db8::192:0:2:1 2386 SFF2 2001:db8::192:0:2:2 2387 SFF3 2001:db8::192:0:2:3 2388 SFF4 2001:db8::192:0:2:4 2390 Each SFF provides access to some SFIs from the four Service Function 2391 Types SFT=41, SFT=42, SFT=43, and SFT=44 just as before: 2393 SFF1 SFT=41 and SFT=42 2394 SFF2 SFT=41 and SFT=43 2395 SFF3 SFT=42 and SFT=44 2396 SFF4 SFT=43 and SFT=44 2398 The service function network also contains a Controller with address 2399 2001:db8::198:51:100:1. 2401 This example service function overlay network is shown in Figure 15. 2403 ------------------------ 2404 | Controller | 2405 | 2001:db8::198:51:100:1 | 2406 ------------------------ 2407 ------ ------ ------ ------ 2408 | SFI | | SFI | | SFI | | SFI | 2409 |SFT=41| |SFT=42| |SFT=41| |SFT=43| 2410 ------ ------ ------ ------ 2411 \ / \ / 2412 ------------------- ------------------- 2413 | SFF1 | | SFF2 | 2414 |2001:db8::192:0:2:1| |2001:db8::192:0:2:2| 2415 ------------------- ------------------- 2416 ---------- 2417 Packet --> | | --> 2418 Flows --> |Classifier| -->Dest 2419 | | --> 2420 ---------- 2421 ------------------- ------------------- 2422 | SFF3 | | SFF4 | 2423 |2001:db8::192:0:2:3| |2001:db8::192:0:2:4| 2424 ------------------- ------------------- 2425 / \ / \ 2426 ------ ------ ------ ------ 2427 | SFI | | SFI | | SFI | | SFI | 2428 |SFT=42| |SFT=44| |SFT=43| |SFT=44| 2429 ------ ------ ------ ------ 2431 Figure 15: Example Service Function Overlay Network 2433 The SFFs advertise routes to the SFIs they support. So we see the 2434 following SFIRs: 2436 RD = 2001:db8::192:0:2:1/1, SFT = 41 2437 RD = 2001:db8::192:0:2:1/2, SFT = 42 2438 RD = 2001:db8::192:0:2:2/1, SFT = 41 2439 RD = 2001:db8::192:0:2:2/2, SFT = 43 2440 RD = 2001:db8::192:0:2:3/7, SFT = 42 2441 RD = 2001:db8::192:0:2:3/8, SFT = 44 2442 RD = 2001:db8::192:0:2:4/5, SFT = 43 2443 RD = 2001:db8::192:0:2:4/6, SFT = 44 2445 Note that the addressing used for communicating between SFFs is taken 2446 from the Tunnel Encapsulation attribute of the SFIR and not from the 2447 SFIR-RD. 2449 8.10.1. Example Explicit SFP With No Choices 2451 Consider the following SFPR similar to that in Section 8.1. 2453 SFP1: RD = 2001:db8::198:51:100:1/101, SPI = 15, 2454 [SI = 255, SFT = 41, RD = 2001:db8::192:0:2:1/1], 2455 [SI = 250, SFT = 43, RD = 2001:db8::192:0:2:2/2] 2457 The Service Function Path consists of an SF of type 41 located at 2458 SFF1 followed by an SF of type 43 located at SFF2. This path is 2459 fully explicit and each SFF is offered no choice in forwarding packet 2460 along the path. 2462 SFF1 will receive packets on the path from the Classifier and will 2463 identify the path from the SPI (15). The initial SI will be 255 and 2464 so SFF1 will deliver the packets to the SFI for SFT 41. 2466 When the packets are returned to SFF1 by the SFI the SI will be 2467 decreased to 250 for the next hop. SFF1 has no flexibility in the 2468 choice of SFF to support the next hop SFI and will forward the packet 2469 to SFF2 which will send the packets to the SFI that supports SFT 43 2470 before forwarding the packets to their destinations. 2472 8.10.2. Example SFP With Choice of SFIs 2474 SFP2: RD = 2001:db8::198:51:100:1/102, SPI = 16, 2475 [SI = 255, SFT = 41, RD = 2001:db8::192:0:2:1/1], 2476 [SI = 250, SFT = 43, {RD = 2001:db8::192:0:2:2/2, 2477 RD = 2001:db8::192:0:2:4/5 } ] 2479 In this example, like that in Section 8.2, the path also consists of 2480 an SF of type 41 located at SFF1 and this is followed by an SF of 2481 type 43, but in this case the SI = 250 contains a choice between the 2482 SFI located at SFF2 and the SFI located at SFF4. 2484 SFF1 will receive packets on the path from the Classifier and will 2485 identify the path from the SPI (16). The initial SI will be 255 and 2486 so SFF1 will deliver the packets to the SFI for SFT 41. 2488 When the packets are returned to SFF1 by the SFI the SI will be 2489 decreased to 250 for the next hop. SFF1 now has a choice of next hop 2490 SFF to execute the next hop in the path. It can either forward 2491 packets to SFF2 or SFF4 to execute a function of type 43. It uses 2492 its local load balancing algorithm to make this choice. The chosen 2493 SFF will send the packets to the SFI that supports SFT 43 before 2494 forwarding the packets to their destinations. 2496 8.10.3. Example SFP With Open Choice of SFIs 2498 SFP3: RD = 2001:db8::198:51:100:1/103, SPI = 17, 2499 [SI = 255, SFT = 41, RD = 2001:db8::192:0:2:1/1], 2500 [SI = 250, SFT = 44, RD = 0] 2502 In this example, like that in Section 8.3 the path also consists of 2503 an SF of type 41 located at SFF1 and this is followed by an SI with 2504 an RD of zero and SF of type 44. This means that a choice can be 2505 made between any SFF that supports an SFI of type 44. 2507 SFF1 will receive packets on the path from the Classifier and will 2508 identify the path from the SPI (17). The initial SI will be 255 and 2509 so SFF1 will deliver the packets to the SFI for SFT 41. 2511 When the packets are returned to SFF1 by the SFI the SI will be 2512 decreased to 250 for the next hop. SFF1 now has a free choice of 2513 next hop SFF to execute the next hop in the path selecting between 2514 all SFFs that support SFs of type 44. Looking at the SFIRs it has 2515 received, SFF1 knows that SF type 44 is supported by SFF3 and SFF4. 2516 SFF1 uses its local load balancing algorithm to make this choice. 2517 The chosen SFF will send the packets to the SFI that supports SFT 44 2518 before forwarding the packets to their destinations. 2520 8.10.4. Example SFP With Choice of SFTs 2521 SFP4: RD = 2001:db8::198:51:100:1/104, SPI = 18, 2522 [SI = 255, SFT = 41, RD = 2001:db8::192:0:2:1/1], 2523 [SI = 250, {SFT = 43, RD = 2001:db8::192:0:2:2/2, 2524 SFT = 44, RD = 2001:db8::192:0:2:3/8 } ] 2526 This example, similar to that in Section 8.4 provides a choice of SF 2527 type in the second hop in the path. The SI of 250 indicates a choice 2528 between SF type 43 located through SF2 and SF type 44 located at SF3. 2530 SFF1 will receive packets on the path from the Classifier and will 2531 identify the path from the SPI (18). The initial SI will be 255 and 2532 so SFF1 will deliver the packets to the SFI for SFT 41. 2534 When the packets are returned to SFF1 by the SFI the SI will be 2535 decreased to 250 for the next hop. SFF1 now has a free choice of 2536 next hop SFF to execute the next hop in the path selecting between 2537 all SFF2 that support an SF of type 43 and SFF3 that supports an SF 2538 of type 44. These may be completely different functions that are to 2539 be executed dependent on specific conditions, or may be similar 2540 functions identified with different type identifiers (such as 2541 firewalls from different vendors). SFF1 uses its local policy and 2542 load balancing algorithm to make this choice, and may use additional 2543 information passed back from the local SFI to help inform its 2544 selection. The chosen SFF will send the packets to the SFI that 2545 supports the chose SFT before forwarding the packets to their 2546 destinations. 2548 9. Security Considerations 2550 The mechanisms in this document use BGP for the control plane. 2551 Hence, techniques such as those discussed in [RFC5925]] can be used 2552 to help authenticate BGP sessions and thus the messages between BGP 2553 peers, making it harder to spoof updates (which could be used to 2554 install bogus SFPs or to advertise false SIs) or withdrawals. 2556 Further discussion of security considerations for BGP may be found in 2557 the BGP specification itself [RFC4271] and in the security analysis 2558 for BGP [RFC4272]. The original discussion of the use of the TCP MD5 2559 signature option to protect BGP sessions is found in [RFC5925], while 2560 [RFC6952] includes an analysis of BGP keying and authentication 2561 issues. 2563 Additionally, this document depends on other documents that specify 2564 BGP Multiprotocol Extensions and the documents that define the 2565 attributes that are carried by BGP UPDATEs of the SFC AFI/SAFI. 2566 Relevant additional security measures are considered in [RFC4760] and 2567 [I-D.ietf-idr-tunnel-encaps]. 2569 This document does not fundamentally change the security behavior of 2570 BGP deployments which depend considerably on the network operator's 2571 perception of risk in their network. It may be observed that the 2572 application of the mechanisms described in this document are scoped 2573 to a single domain as implied by [RFC8300] noted in Section 2.1. 2574 Applicability of BGP within a single domain may enable a network 2575 operator to make easier and more consistent decisions about what 2576 security measures to apply, and the domain boundary, which BGP 2577 enforces by definition, provides a safeguard that prevents leakage of 2578 SFC programming in either direction at the boundary. 2580 Service Function Chaining provides a significant attack opportunity: 2581 packets can be diverted from their normal paths through the network, 2582 packets can be made to execute unexpected functions, and the 2583 functions that are instantiated in software can be subverted. 2584 However, this specification does not change the existence of Service 2585 Function Chaining and security issues specific to Service Function 2586 Chaining are covered in [RFC7665] and [RFC8300]. 2588 This document defines a control plane for Service Function Chaining. 2589 Clearly, this provides an attack vector for a Service Function 2590 Chaining system as an attack on this control plane could be used to 2591 make the system misbehave. Thus, the security of the BGP system is 2592 critically important to the security of the whole Service Function 2593 Chaining system. The control plane mechanisms are very similar to 2594 those used for BGP/MPLS IP VPNs as described in [RFC4364], and so the 2595 security considerations in that document (Section 13) provide good 2596 guidance for securing SFC systems reliant on this specification. Of 2597 particular relevance is the need to securely distinguish between 2598 messages intended for the control of different SFC overlays which is 2599 similar to the need to distinguish between different VPNs. 2600 Section 19 of [RFC7432] also provides useful guidance on the use of 2601 BGP in a similar environment. 2603 Note that a component of an SFC system that uses the procedures 2604 described in this document also requires communications between a 2605 controller and the SFC network elements. This communication covers 2606 instructing the Classifiers using BGP mechanisms (see Section 7.4), 2607 thus the use of BGP security is strongly recommended.. But it also 2608 covers other mechanisms for programming the Classifier and 2609 instructing the SFFs and SFs (for example, to bind SFs to an SFF, and 2610 to cause the establishment of tunnels between SFFs). This document 2611 does not cover these latter mechanisms and so their security is out 2612 of scope, but it should be noted that these communications provide an 2613 attack vector on the SFC system and so attention must be paid to 2614 ensuring that they are secure. 2616 There is an intrinsic assumption in SFC systems that nodes that 2617 announce support for specific SFs actually offer those functions, and 2618 that SFs are not, themselves, attacked or subverted. This is 2619 particularly important when the SFs are implemented as software that 2620 can be updated. Protection against this sort of concern forms part 2621 of the security of any SFC system and so is outside the scope of the 2622 control plane mechanisms described in this document. 2624 Similarly, there is a vulnerablity if a rogue or subverted controller 2625 announces SFPs especially if that controller "takes over" an existing 2626 SFP and changes its contents. This is corresponds to a rogue BGP 2627 speaker entering a routing system, or even to a Route Reflector 2628 becoming subverted. Protection mechanisms, as above, include 2629 securing BGP sessions and protecting software loads on the 2630 controllers. 2632 Lastly, note that Section 3.2.2 makes two operational suggestions 2633 that have implications for the stability and security of the 2634 mechanisms described in this document: 2636 o That modifications to active SFPs not be made. 2638 o That SPIs not be immediately re-used. 2640 10. IANA Considerations 2642 10.1. New BGP AF/SAFI 2644 IANA maintains a registry of "Address Family Numbers". IANA is 2645 requested to assign a new Address Family Number from the "Standards 2646 Action" range called "BGP SFC" (TBD1 in this document) with this 2647 document as a reference. 2649 IANA maintains a registry of "Subsequent Address Family Identifiers 2650 (SAFI) Parameters". IANA is requested to assign a new SAFI value 2651 from the "Standards Action" range called "BGP SFC" (TBD2 in this 2652 document) with this document as a reference. 2654 10.2. New BGP Path Attribute 2656 IANA maintains a registry of "Border Gateway Protocol (BGP) 2657 Parameters" with a subregistry of "BGP Path Attributes". IANA is 2658 requested to assign a new Path attribute called "SFP attribute" (TBD3 2659 in this document) with this document as a reference. 2661 10.3. New SFP Attribute TLVs Type Registry 2663 IANA maintains a registry of "Border Gateway Protocol (BGP) 2664 Parameters". IANA is request to create a new subregistry called the 2665 "SFP Attribute TLVs" registry. 2667 Valid values are in the range 0 to 65535. 2669 o Values 0 and 65535 are to be marked "Reserved, not to be 2670 allocated". 2672 o Values 1 through 65534 are to be assigned according to the "First 2673 Come First Served" policy [RFC8126]. 2675 This document should be given as a reference for this registry. 2677 The new registry should track: 2679 o Type 2681 o Name 2683 o Reference Document or Contact 2685 o Registration Date 2687 The registry should initially be populated as follows: 2689 Type | Name | Reference | Date 2690 ------+-------------------------+---------------+--------------- 2691 1 | Association TLV | [This.I-D] | Date-to-be-set 2692 2 | Hop TLV | [This.I-D] | Date-to-be-set 2693 3 | SFT TLV | [This.I-D] | Date-to-be-set 2694 4 | MPLS Swapping/Stacking | [This.I-D] | Date-to-be-set 2695 5 | SFP Traversal With MPLS | [This.I-D] | Date-to-be-set 2697 10.4. New SFP Association Type Registry 2699 IANA maintains a registry of "Border Gateway Protocol (BGP) 2700 Parameters". IANA is request to create a new subregistry called the 2701 "SFP Association Type" registry. 2703 Valid values are in the range 0 to 65535. 2705 o Values 0 and 65535 are to be marked "Reserved, not to be 2706 allocated". 2708 o Values 1 through 65534 are to be assigned according to the "First 2709 Come First Served" policy [RFC8126]. 2711 This document should be given as a reference for this registry. 2713 The new registry should track: 2715 o Association Type 2717 o Name 2719 o Reference Document or Contact 2721 o Registration Date 2723 The registry should initially be populated as follows: 2725 Association Type | Name | Reference | Date 2726 -----------------+--------------------+------------+--------------- 2727 1 | Bidirectional SFP | [This.I-D] | Date-to-be-set 2729 10.5. New Service Function Type Registry 2731 IANA is request to create a new top-level registry called "Service 2732 Function Chaining Service Function Types". 2734 Valid values are in the range 0 to 65535. 2736 o Values 0 and 65535 are to be marked "Reserved, not to be 2737 allocated". 2739 o Values 1 through 31 are to be assigned by "Standards Action" 2740 [RFC8126] and are referred to as the Special Purpose SFT values. 2742 o Other values (32 through 65534) are to be assigned according to 2743 the "First Come First Served" policy [RFC8126]. 2745 This document should be given as a reference for this registry. 2747 The new registry should track: 2749 o Value 2751 o Name 2753 o Reference Document or Contact 2754 o Registration Date 2756 The registry should initially be populated as follows: 2758 Value | Name | Reference | Date 2759 ------+-------------------------------+------------+--------------- 2760 0 | Reserved, not to be allocated | [This.I-D] | Date-to-be-set 2761 1 | Change Sequence | [This.I-D] | Date-to-be-set 2762 2-31 | Unassigned | | 2763 32 | Classifier | [This.I-D] | Date-to-be-set 2764 33 | Firewall | [This.I-D] | Date-to-be-set 2765 34 | Load balancer | [This.I-D] | Date-to-be-set 2766 35 | Deep packet inspection engine | [This.I-D] | Date-to-be-set 2767 36 | Penalty box | [This.I-D] | Date-to-be-set 2768 37 | WAN accelerator | [This.I-D] | Date-to-be-set 2769 38 | Application accelerator | [This.I-D] | Date-to-be-set 2770 39 | TCP optimizer | [This.I-D] | Date-to-be-set 2771 40 | Network Address Translator | [This.I-D] | Date-to-be-set 2772 41 | NAT44 | [This.I-D] | Date-to-be-set 2773 42 | NAT64 | [This.I-D] | Date-to-be-set 2774 43 | NPTv6 | [This.I-D] | Date-to-be-set 2775 44 | Lawful intercept | [This.I-D] | Date-to-be-set 2776 45 | HOST_ID injection | [This.I-D] | Date-to-be-set 2777 46 | HTTP header enrichment | [This.I-D] | Date-to-be-set 2778 47 | Caching engine | [This.I-D] | Date-to-be-set 2779 48- | | | 2780 -65534|Unassigned | | 2781 65535 | Reserved, not to be allocated | [This.I-D] | Date-to-be-set 2783 10.6. New Generic Transitive Experimental Use Extended Community Sub- 2784 Types 2786 IANA maintains a registry of "Border Gateway Protocol (BGP) 2787 Parameters" with a subregistry of "Generic Transitive Experimental 2788 Use Extended Community Sub-Type". IANA is requested to assign a new 2789 sub-type as follows: 2791 "Flow Specification for SFC Classifiers" (TBD4 in this document) 2792 with this document as the reference. 2794 10.7. New BGP Transitive Extended Community Type 2796 IANA maintains a registry of "Border Gateway Protocol (BGP) 2797 Parameters" with a subregistry of "BGP Transitive Extended Community 2798 Types". IANA is requested to assign a new type as follows: 2800 o SFC (Sub-Types are defined in the "SFC Extended Community Sub- 2801 Types" registry) (TBD6 in this document) with this document as the 2802 reference. 2804 10.8. New SFC Extended Community Sub-Types Registry 2806 IANA maintains a registry of "Border Gateway Protocol (BGP) 2807 Parameters". IANA is requested to create a new sub-registry called 2808 the "SFC Extended Community Sub-Types Registry". 2810 IANA should include the following note replacing the string "TBD6" 2811 with the value assigned for Section 10.7: 2813 This registry contains values of the second octet (the "Sub-Type" 2814 field) of an extended community when the value of the first octet 2815 (the "Type" field) is set to TBD6. 2817 The allocation policy for this registry should be First Come First 2818 Served. 2820 IANA is requested to populate this registry with the following 2821 entries: 2823 Sub-Type | | | 2824 Value | Name | Reference | Date 2825 ---------+----------------------+-------------+--------------- 2826 TBD7 | SFIR Pool Identifier | [This.I-D] | Date-to-be-set 2827 TBD8 | MPLS Label Stack | [This.I-D] | Date-to-be-set 2828 | Mixed Swapping/ | | 2829 | Stacking Labels | | 2831 All other values should be marked "Unassigned". 2833 10.9. SPI/SI Representation 2835 IANA is requested to assign a codepoint from the "BGP Tunnel 2836 Encapsulation Attribute Sub-TLVs" registry for the "SPI/SI 2837 Representation Sub-TLV" (TBD5 in this document) with this document 2838 being the reference. 2840 10.10. SFC SPI/SI Representation Flags Registry 2842 IANA maintains the "BGP Tunnel Encapsulation Attribute Sub-TLVs" 2843 registry and is requested to create an associated registry called the 2844 "SFC SPI/SI Representation Flags" registry. 2846 Bits are to be assigned by Standards Action. The field is 16 bits 2847 long, and bits are counted from the the most significant bit as bit 2848 zero. 2850 IANA is requested to populate the registry as follows: 2852 Bit number | Name | Reference 2853 -----------+----------------------+----------- 2854 TBD9 | NSH data plane | [This.I-D] 2855 TBD10 | MPLS data plane | [This.I-D] 2857 11. Contributors 2859 Stuart Mackie 2860 Juniper Networks 2862 Email: wsmackie@juinper.net 2864 Keyur Patel 2865 Arrcus, Inc. 2867 Email: keyur@arrcus.com 2869 Avinash Lingala 2870 AT&T 2872 Email: ar977m@att.com 2874 12. Acknowledgements 2876 Thanks to Tony Przygienda, Jeff Haas, and Andy Malis for helpful 2877 comments, and to Joel Halpern for discussions that improved this 2878 document. Yuanlong Jiang provided a useful review and caught some 2879 important issues. Stephane Litkowski did an exceptionally good and 2880 detailed document shepherd review. 2882 Andy Malis contributed text that formed the basis of Section 7.7. 2884 Brian Carpenter and Martin Vigoureux provided useful reviews during 2885 IETF last call. Thanks also to Sheng Jiang, Ravi Singh, Benjamin 2886 Kaduk, Roman Danyliw, Adam Roach, and Barry Leiba for review 2887 comments. 2889 13. References 2891 13.1. Normative References 2893 [I-D.ietf-idr-rfc5575bis] 2894 Loibl, C., Hares, S., Raszuk, R., McPherson, D., and M. 2895 Bacher, "Dissemination of Flow Specification Rules", 2896 draft-ietf-idr-rfc5575bis-25 (work in progress), May 2020. 2898 [I-D.ietf-idr-tunnel-encaps] 2899 Patel, K., Velde, G., and S. Ramachandra, "The BGP Tunnel 2900 Encapsulation Attribute", draft-ietf-idr-tunnel-encaps-15 2901 (work in progress), December 4019. 2903 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2904 Requirement Levels", BCP 14, RFC 2119, 2905 DOI 10.17487/RFC2119, March 1997, 2906 . 2908 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 2909 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 2910 DOI 10.17487/RFC4271, January 2006, 2911 . 2913 [RFC4360] Sangli, S., Tappan, D., and Y. Rekhter, "BGP Extended 2914 Communities Attribute", RFC 4360, DOI 10.17487/RFC4360, 2915 February 2006, . 2917 [RFC4364] Rosen, E. and Y. Rekhter, "BGP/MPLS IP Virtual Private 2918 Networks (VPNs)", RFC 4364, DOI 10.17487/RFC4364, February 2919 2006, . 2921 [RFC4760] Bates, T., Chandra, R., Katz, D., and Y. Rekhter, 2922 "Multiprotocol Extensions for BGP-4", RFC 4760, 2923 DOI 10.17487/RFC4760, January 2007, 2924 . 2926 [RFC5575] Marques, P., Sheth, N., Raszuk, R., Greene, B., Mauch, J., 2927 and D. McPherson, "Dissemination of Flow Specification 2928 Rules", RFC 5575, DOI 10.17487/RFC5575, August 2009, 2929 . 2931 [RFC7432] Sajassi, A., Ed., Aggarwal, R., Bitar, N., Isaac, A., 2932 Uttaro, J., Drake, J., and W. Henderickx, "BGP MPLS-Based 2933 Ethernet VPN", RFC 7432, DOI 10.17487/RFC7432, February 2934 2015, . 2936 [RFC7606] Chen, E., Ed., Scudder, J., Ed., Mohapatra, P., and K. 2937 Patel, "Revised Error Handling for BGP UPDATE Messages", 2938 RFC 7606, DOI 10.17487/RFC7606, August 2015, 2939 . 2941 [RFC7665] Halpern, J., Ed. and C. Pignataro, Ed., "Service Function 2942 Chaining (SFC) Architecture", RFC 7665, 2943 DOI 10.17487/RFC7665, October 2015, 2944 . 2946 [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for 2947 Writing an IANA Considerations Section in RFCs", BCP 26, 2948 RFC 8126, DOI 10.17487/RFC8126, June 2017, 2949 . 2951 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2952 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 2953 May 2017, . 2955 [RFC8300] Quinn, P., Ed., Elzur, U., Ed., and C. Pignataro, Ed., 2956 "Network Service Header (NSH)", RFC 8300, 2957 DOI 10.17487/RFC8300, January 2018, 2958 . 2960 [RFC8595] Farrel, A., Bryant, S., and J. Drake, "An MPLS-Based 2961 Forwarding Plane for Service Function Chaining", RFC 8595, 2962 DOI 10.17487/RFC8595, June 2019, 2963 . 2965 [RFC8596] Malis, A., Bryant, S., Halpern, J., and W. Henderickx, 2966 "MPLS Transport Encapsulation for the Service Function 2967 Chaining (SFC) Network Service Header (NSH)", RFC 8596, 2968 DOI 10.17487/RFC8596, June 2019, 2969 . 2971 13.2. Informative References 2973 [I-D.dawra-idr-bgp-ls-sr-service-segments] 2974 Dawra, G., Filsfils, C., Talaulikar, K., Clad, F., 2975 daniel.bernier@bell.ca, d., Uttaro, J., Decraene, B., 2976 Elmalky, H., Xu, X., Guichard, J., and C. Li, "BGP-LS 2977 Advertisement of Segment Routing Service Segments", draft- 2978 dawra-idr-bgp-ls-sr-service-segments-03 (work in 2979 progress), January 2020. 2981 [RFC4272] Murphy, S., "BGP Security Vulnerabilities Analysis", 2982 RFC 4272, DOI 10.17487/RFC4272, January 2006, 2983 . 2985 [RFC5925] Touch, J., Mankin, A., and R. Bonica, "The TCP 2986 Authentication Option", RFC 5925, DOI 10.17487/RFC5925, 2987 June 2010, . 2989 [RFC6952] Jethanandani, M., Patel, K., and L. Zheng, "Analysis of 2990 BGP, LDP, PCEP, and MSDP Issues According to the Keying 2991 and Authentication for Routing Protocols (KARP) Design 2992 Guide", RFC 6952, DOI 10.17487/RFC6952, May 2013, 2993 . 2995 [RFC7498] Quinn, P., Ed. and T. Nadeau, Ed., "Problem Statement for 2996 Service Function Chaining", RFC 7498, 2997 DOI 10.17487/RFC7498, April 2015, 2998 . 3000 Authors' Addresses 3002 Adrian Farrel 3003 Old Dog Consulting 3005 Email: adrian@olddog.co.uk 3007 John Drake 3008 Juniper Networks 3010 Email: jdrake@juniper.net 3012 Eric Rosen 3013 Juniper Networks 3015 Email: erosen52@gmail.com 3017 Jim Uttaro 3018 AT&T 3020 Email: ju1738@att.com 3022 Luay Jalil 3023 Verizon 3025 Email: luay.jalil@verizon.com