idnits 2.17.1 draft-ietf-sfc-nsh-10.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 (September 20, 2016) is 2768 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) ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) Summary: 1 error (**), 0 flaws (~~), 1 warning (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Service Function Chaining P. Quinn, Ed. 3 Internet-Draft Cisco Systems, Inc. 4 Intended status: Standards Track U. Elzur, Ed. 5 Expires: March 24, 2017 Intel 6 September 20, 2016 8 Network Service Header 9 draft-ietf-sfc-nsh-10.txt 11 Abstract 13 This document describes a Network Service Header (NSH) inserted onto 14 packets or frames to realize service function paths. NSH also 15 provides a mechanism for metadata exchange along the instantiated 16 service path. NSH is the SFC encapsulation required to support the 17 Service Function Chaining (SFC) Architecture (defined in RFC7665). 19 1. Requirements Language 21 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 22 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 23 document are to be interpreted as described in RFC 2119 [RFC2119]. 25 Status of this Memo 27 This Internet-Draft is submitted in full conformance with the 28 provisions of BCP 78 and BCP 79. 30 Internet-Drafts are working documents of the Internet Engineering 31 Task Force (IETF). Note that other groups may also distribute 32 working documents as Internet-Drafts. The list of current Internet- 33 Drafts is at http://datatracker.ietf.org/drafts/current/. 35 Internet-Drafts are draft documents valid for a maximum of six months 36 and may be updated, replaced, or obsoleted by other documents at any 37 time. It is inappropriate to use Internet-Drafts as reference 38 material or to cite them other than as "work in progress." 40 This Internet-Draft will expire on March 24, 2017. 42 Copyright Notice 44 Copyright (c) 2016 IETF Trust and the persons identified as the 45 document authors. All rights reserved. 47 This document is subject to BCP 78 and the IETF Trust's Legal 48 Provisions Relating to IETF Documents 49 (http://trustee.ietf.org/license-info) in effect on the date of 50 publication of this document. Please review these documents 51 carefully, as they describe your rights and restrictions with respect 52 to this document. Code Components extracted from this document must 53 include Simplified BSD License text as described in Section 4.e of 54 the Trust Legal Provisions and are provided without warranty as 55 described in the Simplified BSD License. 57 Table of Contents 59 1. Requirements Language . . . . . . . . . . . . . . . . . . . . 2 60 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 61 2.1. Definition of Terms . . . . . . . . . . . . . . . . . . . 4 62 2.2. Problem Space . . . . . . . . . . . . . . . . . . . . . . 5 63 2.3. NSH-based Service Chaining . . . . . . . . . . . . . . . . 5 64 3. Network Service Header . . . . . . . . . . . . . . . . . . . . 7 65 3.1. Network Service Header Format . . . . . . . . . . . . . . 7 66 3.2. NSH Base Header . . . . . . . . . . . . . . . . . . . . . 7 67 3.3. Service Path Header . . . . . . . . . . . . . . . . . . . 10 68 3.4. NSH MD Type 1 . . . . . . . . . . . . . . . . . . . . . . 10 69 3.5. NSH MD Type 2 . . . . . . . . . . . . . . . . . . . . . . 11 70 3.5.1. Optional Variable Length Metadata . . . . . . . . . . 12 71 4. NSH Actions . . . . . . . . . . . . . . . . . . . . . . . . . 14 72 5. NSH Encapsulation . . . . . . . . . . . . . . . . . . . . . . 16 73 6. Fragmentation Considerations . . . . . . . . . . . . . . . . . 17 74 7. Service Path Forwarding with NSH . . . . . . . . . . . . . . . 18 75 7.1. SFFs and Overlay Selection . . . . . . . . . . . . . . . . 18 76 7.2. Mapping NSH to Network Transport . . . . . . . . . . . . . 20 77 7.3. Service Plane Visibility . . . . . . . . . . . . . . . . . 21 78 7.4. Service Graphs . . . . . . . . . . . . . . . . . . . . . . 21 79 8. Policy Enforcement with NSH . . . . . . . . . . . . . . . . . 22 80 8.1. NSH Metadata and Policy Enforcement . . . . . . . . . . . 22 81 8.2. Updating/Augmenting Metadata . . . . . . . . . . . . . . . 24 82 8.3. Service Path Identifier and Metadata . . . . . . . . . . . 25 83 9. Security Considerations . . . . . . . . . . . . . . . . . . . 27 84 10. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 28 85 11. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 31 86 12. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 32 87 12.1. NSH EtherType . . . . . . . . . . . . . . . . . . . . . . 32 88 12.2. Network Service Header (NSH) Parameters . . . . . . . . . 32 89 12.2.1. NSH Base Header Reserved Bits . . . . . . . . . . . . 32 90 12.2.2. NSH Version . . . . . . . . . . . . . . . . . . . . . 32 91 12.2.3. MD Type Registry . . . . . . . . . . . . . . . . . . . 32 92 12.2.4. MD Class Registry . . . . . . . . . . . . . . . . . . 33 93 12.2.5. NSH Base Header Next Protocol . . . . . . . . . . . . 33 94 13. References . . . . . . . . . . . . . . . . . . . . . . . . . . 35 95 13.1. Normative References . . . . . . . . . . . . . . . . . . . 35 96 13.2. Informative References . . . . . . . . . . . . . . . . . . 35 97 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 37 99 2. Introduction 101 Service functions are widely deployed and essential in many networks. 102 These service functions provide a range of features such as security, 103 WAN acceleration, and server load balancing. Service functions may 104 be instantiated at different points in the network infrastructure 105 such as the wide area network, data center, campus, and so forth. 107 Prior to development of the SFC architecture [RFC7665] and the 108 protocol specified in this document, current service function 109 deployment models have been relatively static, and bound to topology 110 for insertion and policy selection. Furthermore, they do not adapt 111 well to elastic service environments enabled by virtualization. 113 New data center network and cloud architectures require more flexible 114 service function deployment models. Additionally, the transition to 115 virtual platforms requires an agile service insertion model that 116 supports dynamic and elastic service delivery; the movement of 117 service functions and application workloads in the network and the 118 ability to easily bind service policy to granular information such as 119 per-subscriber state and steer traffic to the requisite service 120 function(s) are necessary. 122 NSH defines a new service plane protocol specifically for the 123 creation of dynamic service chains and is composed of the following 124 elements: 126 1. Service Function Path identification 128 2. Transport independent service function chain 130 3. Per-packet network and service metadata or optional variable 131 type-length-value (TLV) metadata. 133 NSH is designed to be easy to implement across a range of devices, 134 both physical and virtual, including hardware platforms. 136 An NSH-aware control plane is outside the scope of this document. 138 [RFC7665] provides an overview of a service chaining architecture 139 that clearly defines the roles of the various elements and the scope 140 of a service function chaining encapsulation. NSH is the SFC 141 encapsulation referenced in RFC7665. 143 2.1. Definition of Terms 144 Classification: Defined in [RFC7665]. 146 Classifier: Defined in [RFC7665]. 148 Metadata: Defined in [RFC7665]. 150 Network Locator: dataplane address, typically IPv4 or IPv6, used to 151 send and receive network traffic. 153 Network Node/Element: Device that forwards packets or frames based 154 on outer header (i.e. transport) information. 156 Network Overlay: Logical network built on top of existing network 157 (the underlay). Packets are encapsulated or tunneled to create 158 the overlay network topology. 160 Service Classifier: Logical entity providing classification 161 function. Since they are logical, classifiers may be co-resident 162 with SFC elements such as SFs or SFFs. Service classifiers 163 perform classification and impose NSH. The initial classifier 164 imposes the initial NSH and sends the NSH packet to the first SFF 165 in the path. Non-initial (i.e. subsequent) classification can 166 occur as needed and can alter, or create a new service path. 168 Service Function (SF): Defined in [RFC7665]. 170 Service Function Chain (SFC): Defined in [RFC7665]. 172 Service Function Forwarder (SFF): Defined in [RFC7665]. 174 Service Function Path (SFP): Defined in [RFC7665]. 176 SFC Proxy: Defined in [RFC7665]. 178 2.2. Problem Space 180 Network Service Header (NSH) addresses several limitations associated 181 with service function deployments. [RFC7498] provides a 182 comprehensive review of those issues. 184 2.3. NSH-based Service Chaining 186 The NSH creates a dedicated service plane, more specifically, NSH 187 enables: 189 1. Topological Independence: Service forwarding occurs within the 190 service plane, the underlying network topology does not require 191 modification. NSH provides an identifier used to select the 192 network overlay for network forwarding. 194 2. Service Chaining: NSH enables service chaining per [RFC7665]. 195 NSH contains path identification information needed to realize a 196 service path. Furthermore, NSH provides the ability to monitor 197 and troubleshoot a service chain, end-to-end via service-specific 198 OAM messages. The NSH fields can be used by administrators (via, 199 for example a traffic analyser) to verify (account, ensure 200 correct chaining, provide reports, etc.) the path specifics of 201 packets being forwarded along a service path. 203 3. NSH provides a mechanism to carry shared metadata between 204 participating entities and service functions. The semantics of 205 the shared metadata is communicated via a control plane, which is 206 outside the scope of this document, to participating nodes. 207 [SFC-CP] provides an example of such in section 3.3. Examples of 208 metadata include classification information used for policy 209 enforcement and network context for forwarding post service 210 delivery. 212 4. Classification and re-classification: sharing the metadata allows 213 service functions to share initial and intermediate 214 classification results with downstream service functions saving 215 re-classification, where enough information was enclosed. 217 5. NSH offers a common and standards-based header for service 218 chaining to all network and service nodes. 220 6. Transport Agnostic: NSH is transport independent. An appropriate 221 (for a given deployment) network transport protocol can be used 222 to transport NSH-encapsulated traffic. This transport may form 223 an overlay network and if an existing overlay topology provides 224 the required service path connectivity, that existing overlay may 225 be used. 227 3. Network Service Header 229 A Network Service Header (NSH) contains service path information and 230 optionally metadata that are added to a packet or frame and used to 231 create a service plane. An outer transport header is imposed, on NSH 232 and the original packet/frame, for network forwarding. 234 A Service Classifier adds the NSH. The NSH is removed by the last 235 SFF in the service chain or by a SF that consumes the packet. 237 3.1. Network Service Header Format 239 An NSH is composed of a 4-byte (all references to bytes in this draft 240 refer to 8-bit bytes, or octets) Base Header, a 4-byte Service Path 241 Header and Context Headers, as shown in Figure 1 below. 243 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 1 244 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 245 | Base Header | 246 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 247 | Service Path Header | 248 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 249 | | 250 ~ Context Headers ~ 251 | | 252 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 254 Figure 1: Network Service Header 256 Base header: provides information about the service header and the 257 payload protocol. 259 Service Path Header: provide path identification and location within 260 a service path. 262 Context headers: carry metadata (i.e. context data) along a service 263 path. 265 3.2. NSH Base Header 267 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 1 268 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 269 |Ver|O|C|R|R|R|R|R|R| Length | MD Type | Next Protocol | 270 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 271 Figure 2: NSH Base Header 273 Base Header Field Descriptions: 275 Version: The version field is used to ensure backward compatibility 276 going forward with future NSH updates. It MUST be set to 0x0 by the 277 sender, in this first revision of NSH. Given the widespread 278 implementation of existing hardware that uses the first nibble after 279 an MPLS label stack for ECMP decision processing, this document 280 reserves version 01 and this value MUST NOT be used in future 281 versions of the protocol. Please see [RFC7325] for further 282 discussion of MPLS-related forwarding requirements. 284 O bit: Setting this bit indicates an Operations, Administration, and 285 Maintenance (OAM) packet. The actual packet format and processing of 286 SFC OAM messages is outside the scope of this specification (see [I- 287 D.ietf-sfc-oam-framework]). 289 SF/SFF/SFC Proxy/Classifer implementations, which do not support SFC 290 OAM procedures, SHALL discard packets with O-bit set. 292 SF/SFF/SFC Proxy/Classifer implementations MAY support a configurable 293 parameter to enable forwarding received SFC OAM packets unmodified to 294 the next element in the chain. Such behavior may be acceptable for a 295 subset of OAM functions, but can result in unexpected outcomes for 296 others, thus it is recommended to analyze the impact of forwarding an 297 OAM packet for all OAM functions prior to enabling this behavior. 298 The configurable parameter MUST be disabled by default. 300 For non OAM packets, the O-bit MUST be cleared and MUST NOT be 301 modified along the SFP. 303 C bit: Indicates that a critical metadata TLV is present. This bit 304 acts as an indication for hardware implementers to decide how to 305 handle the presence of a critical TLV without necessarily needing to 306 parse all TLVs present. For an MD Type of 0x1 (i.e. no variable 307 length metadata is present), the C bit MUST be set to 0x0. 309 All other flag fields are reserved for future use. Reserved bits 310 MUST be set to zero when sent and MUST be ignored upon receipt. 312 Length: total length, in 4-byte words, of NSH including the Base 313 Header, the Service Path Header and the context headers or optional 314 variable length metadata. The Length MUST be of value 0x6 for MD 315 Type equal to 0x1 and MUST be of value 0x2 or greater for MD Type 316 equal to 0x2. The NSH header length MUST be an integer number of 4 317 bytes. The length field indicates the "end" of NSH and where the 318 original packet/frame begins. 320 MD Type: indicates the format of NSH beyond the mandatory Base Header 321 and the Service Path Header. MD Type defines the format of the 322 metadata being carried. Please see IANA Considerations section 323 below. 325 NSH defines two MD types: 327 0x1 - which indicates that the format of the header includes fixed 328 length context headers (see Figure 4 below). 330 0x2 - which does not mandate any headers beyond the Base Header and 331 Service Path Header, but may contain optional variable length context 332 information. 334 The format of the base header and the service path header is 335 invariant, and not affected by MD Type. 337 NSH implementations MUST support MD Type = 0x1, and SHOULD support MD 338 Type = 0x2. There exists, however, a middle ground, wherein a device 339 will support MD Type 0x1 (as per the MUST) metadata, yet be deployed 340 in a network with MD Type 0x2 metadata packets. In that case, the MD 341 Type 0x1 node, MUST utilize the base header length field to determine 342 the original payload offset if it requires access to the original 343 packet/frame. 345 Next Protocol: indicates the protocol type of the encapsulated data. 346 NSH does not alter the inner payload, and the semantics on the inner 347 protocol remain unchanged due to NSH service function chaining. 348 Please see IANA Considerations section below. 350 This draft defines the following Next Protocol values: 352 0x1 : IPv4 353 0x2 : IPv6 354 0x3 : Ethernet 355 0x4: NSH 356 0x5: MPLS 357 0x6-0xFD: Unassigned 358 0xFE-0xFF: Experimental 360 3.3. Service Path Header 362 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 1 363 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 364 | Service Path Identifier (SPI) | Service Index | 365 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 367 Service Path Identifier (SPI): 24 bits 368 Service Index (SI): 8 bits 370 Figure 3: NSH Service Path Header 372 Service Path Identifier (SPI): identifies a service path. 373 Participating nodes MUST use this identifier for Service Function 374 Path selection. The initial classifier MUST set the appropriate SPI 375 for a given classification result. 377 Service Index (SI): provides location within the SFP. The initial 378 classifier MUST set the appropriate SI value for a given 379 classification result. The initial SI value SHOULD default to 255. 380 However, the classifier MUST allow configuration of other SI values. 382 Service Index MUST be decremented by Service Functions or by SFC 383 Proxy nodes after performing required services and the new 384 decremented SI value MUST be used in the egress NSH packet. The 385 initial Classifier MUST send the packet to the first SFF in the 386 identified SFP for forwarding along an SFP. If re-classification 387 occurs, and that re-classification results in a new SPI, the 388 (re)classifier is, in effect, the initial classifier for the 389 resultant SPI. 391 SI SHOULD be used in conjunction with SPI for SFP selection and, 392 consequently, determining the next SFF/SF in the path. Service Index 393 (SI) is also valuable when troubleshooting/ reporting service paths. 394 When an SPI and SI do not correspond to a valid next hop in a SFP, it 395 is an error and the SFF SHOULD generate an error/log message. The 396 value zero for SI is not valid and indicates a broken SFC or 397 malfunctioning SF. In addition to indicating the location within a 398 Service Function Path, SI can be used for service plane loop 399 detection. 401 3.4. NSH MD Type 1 403 When the Base Header specifies MD Type = 0x1, four Context Headers, 404 4-byte each, MUST be added immediately following the Service Path 405 Header, as per Figure 4. Context Headers that carry no metadata MUST 406 be set to zero. 408 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 1 409 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 410 |Ver|O|C|R|R|R|R|R|R| Length | MD type=0x1 | Next Protocol | 411 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 412 | Service Path Identifer | Service Index | 413 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 414 | Mandatory Context Header | 415 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 416 | Mandatory Context Header | 417 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 418 | Mandatory Context Header | 419 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 420 | Mandatory Context Header | 421 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 423 Figure 4: NSH MD Type=0x1 425 [dcalloc] and [broadalloc] provide specific examples of how metadata 426 can be allocated. 428 3.5. NSH MD Type 2 430 When the base header specifies MD Type= 0x2, zero or more Variable 431 Length Context Headers MAY be added, immediately following the 432 Service Path Header. Therefore, Length = 0x2, indicates that only 433 the Base Header followed by the Service Path Header are present. The 434 optional Variable Length Context Headers MUST be of an integer number 435 of 4-bytes. The base header length field MUST be used to determine 436 the offset to locate the original packet or frame for SFC nodes that 437 require access to that information. 439 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 1 440 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 441 |Ver|O|C|R|R|R|R|R|R| Length | MD Type=0x2 | Next Protocol | 442 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 443 | Service Path Identifier | Service Index | 444 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 445 | | 446 ~ Variable Length Context Headers (opt.) ~ 447 | | 448 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 450 Figure 5: NSH MD Type=0x2 452 3.5.1. Optional Variable Length Metadata 454 The format of the optional variable length context headers, is as 455 described below. 457 0 1 2 3 458 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 1 459 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 460 | Metadata Class |C| Type |R| Len | 461 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 462 | Variable Metadata | 463 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 465 Figure 6: Variable Context Headers 467 Metadata Class (MD Class): The MD Class defines the scope of the 468 'Type' field to provide a hierarchical namespace. The IANA 469 Considerations section defines how the MD Class values can be 470 allocated to standards bodies, vendors, and others. 472 Type: the Type field is split into two ranges - 0 to 127 for non- 473 critical options and 128-255 for critical options. While the value 474 allocation is the responsibility of the MD Class owner, critical 475 options MUST NOT be allocated from the 0 to 127 range and non- 476 critical options MUST NOT be allocated from the 128-255 range. 478 Figure 7 below illustrates the placement of the Critical bit within 479 the Type field. 481 +-+-+-+-+-+-+-+-+ 482 |C| Type | 483 +-+-+-+-+-+-+-+-+ 485 Figure 7: Critical Bit Placement Within the TLV Type Field 487 If an NSH-aware node receives an encapsulated packet containing a TLV 488 with the Critical bit set to 0x1 in the Type field and it does not 489 understand how to process the Type, it MUST drop the packet. Transit 490 devices (i.e. network nodes that do not participate in the service 491 plane) MUST NOT drop packets based on the setting of this bit. 493 Reserved bit: one reserved bit is present for future use. The 494 reserved bits MUST be set to 0x0. 496 Length: Length of the variable metadata, in single byte words. In 497 case the metadata length is not an integer number of 4-byte words, 498 the sender MUST add pad bytes immediately following the last metadata 499 byte to extend the metadata to an integer number of 4-byte words. 500 The receiver MUST round up the length field to the nearest 4-byte 501 word boundary, to locate and process the next field in the packet. 502 The receiver MUST access only those bytes in the metadata indicated 503 by the length field (i.e. actual number of single byte words) and 504 MUST ignore the remaining bytes up to the nearest 4-byte word 505 boundary. A value of 0x0 or higher can be used. 507 A value of 0x0 denotes a TLV header without a Variable Metadata 508 field. 510 4. NSH Actions 512 NSH-aware nodes are the only nodes that MAY alter the content of the 513 NSH headers. NSH-aware nodes include: service classifiers, SFF, SF 514 and SFC proxies. These nodes have several possible header related 515 actions: 517 1. Insert or remove NSH: These actions can occur at the start and 518 end respectively of a service path. Packets are classified, and 519 if determined to require servicing, NSH will be imposed. A 520 service classifier MUST insert NSH at the start of an SFP. An 521 imposed NSH MUST contain valid Base Header and Service Path 522 Header. At the end of a service function path, a SFF, MUST be 523 the last node operating on the service header and MUST remove it. 525 Multiple logical classifiers may exist within a given service 526 path. Non-initial classifiers may re-classify data and that re- 527 classification MAY result in a new Service Function Path. When 528 the logical classifier performs re-classification that results in 529 a change of service path, it MUST remove the existing NSH and 530 MUST impose a new NSH with the Base Header and Service Path 531 Header reflecting the new service path information and set the 532 initial SI. Metadata MAY be preserved in the new NSH. 534 2. Select service path: The Service Path Header provides service 535 chain information and is used by SFFs to determine correct 536 service path selection. SFFs MUST use the Service Path Header 537 for selecting the next SF or SFF in the service path. 539 3. Update NSH: NSH-aware service functions (SF) MUST decrement the 540 service index. If an SFF receives a packet with an SPI and SI 541 that do not correspond to a valid next hop in a valid Service 542 Function Path, that packet MUST be dropped by the SFF. 544 Classifier(s) MAY update Context Headers if new/updated context 545 is available. 547 If an SFC proxy is in use (acting on behalf of a non-NSH-aware 548 service function for NSH actions), then the proxy MUST update 549 Service Index and MAY update contexts. When an SFC proxy 550 receives an NSH-encapsulated packet, it MUST remove the NSH 551 headers before forwarding it to an NSH unaware SF. When the SFC 552 Proxy receives a packet back from an NSH unaware SF, it MUST re- 553 encapsulates it with the correct NSH, and MUST decrement the 554 Service Index. 556 4. Service policy selection: Service Function instances derive 557 policy (i.e. service actions such as permit or deny) selection 558 and enforcement from the service header. Metadata shared in the 559 service header can provide a range of service-relevant 560 information such as traffic classification. Service functions 561 SHOULD use NSH to select local service policy. 563 Figure 8 maps each of the four actions above to the components in the 564 SFC architecture that can perform it. 566 +---------------+------------------+-------+----------------+---------+ 567 | | Insert |Select | Update |Service | 568 | | or remove NSH |Service| NSH |policy | 569 | | |Function| |selection| 570 | Component +--------+--------+Path +----------------+ | 571 | | | | | Dec. |Update | | 572 | | Insert | Remove | |Service |Context| | 573 | | | | | Index |Header | | 574 +----------------+--------+--------+-------+--------+-------+---------+ 575 | | + | + | | | + | | 576 |Classifier | | | | | | | 577 +--------------- +--------+--------+-------+--------+-------+---------+ 578 |Service Function| | + | + | | | | 579 |Forwarder(SFF) | | | | | | | 580 +--------------- +--------+--------+-------+--------+-------+---------+ 581 |Service | | | | + | + | + | 582 |Function (SF) | | | | | | | 583 +--------------- +--------+--------+-------+--------+-------+---------+ 584 |SFC Proxy | + | + | | + | | | 585 +----------------+--------+--------+-------+--------+-------+---------+ 587 Figure 8: NSH Action and Role Mapping 589 5. NSH Encapsulation 591 Once NSH is added to a packet, an outer encapsulation is used to 592 forward the original packet and the associated metadata to the start 593 of a service chain. The encapsulation serves two purposes: 595 1. Creates a topologically independent services plane. Packets are 596 forwarded to the required services without changing the 597 underlying network topology 599 2. Transit network nodes simply forward the encapsulated packets as 600 is. 602 The service header is independent of the encapsulation used and is 603 encapsulated in existing transports. The presence of NSH is 604 indicated via protocol type or other indicator in the outer 605 encapsulation. 607 6. Fragmentation Considerations 609 NSH and the associated transport header are "added" to the 610 encapsulated packet/frame. This additional information increases the 611 size of the packet. In order to ensure proper forwarding of NSH 612 packets, several options for handling fragmentation and re-assembly 613 exist: 615 As discussed in [encap-considerations], within an administrative 616 domain, an operator can ensure that the underlay MTU is sufficient to 617 carry SFC traffic without requiring fragmentation. 619 However, there will be cases where the underlay MTU is not large 620 enough to carry the NSH traffic. Since NSH does not provide 621 fragmentation support at the service plane, the transport/overlay 622 layer MUST provide the requisite fragmentation handling. Section 9 623 of [encap-considerations] provides guidance for those scenarios. 625 7. Service Path Forwarding with NSH 627 7.1. SFFs and Overlay Selection 629 As described above, NSH contains a Service Path Identifier (SPI) and 630 a Service Index (SI). The SPI is, as per its name, an identifier. 631 The SPI alone cannot be used to forward packets along a service path. 632 Rather the SPI provide a level of indirection between the service 633 path/topology and the network transport. Furthermore, there is no 634 requirement, or expectation of an SPI being bound to a pre-determined 635 or static network path. 637 The Service Index provides an indication of location within a service 638 path. The combination of SPI and SI provides the identification of a 639 logical SF and its order within the service plane, and is used to 640 select the appropriate network locator(s) for overlay forwarding. 641 The logical SF may be a single SF, or a set of eligible SFs that are 642 equivalent. In the latter case, the SFF provides load distribution 643 amongst the collection of SFs as needed. 645 SI can also serve as a mechanism for loop detection within a service 646 path since each SF in the path decrements the index; an Service Index 647 of 0 indicates that a loop occurred and the packet must be discarded. 649 This indirection -- path ID to overlay -- creates a true service 650 plane. That is the SFF/SF topology is constructed without impacting 651 the network topology but more importantly service plane only 652 participants (i.e. most SFs) need not be part of the network overlay 653 topology and its associated infrastructure (e.g. control plane, 654 routing tables, etc.). As mentioned above, an existing overlay 655 topology may be used provided it offers the requisite connectivity. 657 The mapping of SPI to transport occurs on an SFF (as discussed above, 658 the first SFF in the path gets a NSH encapsulated packet from the 659 Classifier). The SFF consults the SPI/ID values to determine the 660 appropriate overlay transport protocol (several may be used within a 661 given network) and next hop for the requisite SF. Figure 9 below 662 depicts an example of a single next-hop SPI/SI to network overlay 663 network locator mapping. 665 +-------------------------------------------------------+ 666 | SPI | SI | Next hop(s) | Transport | 667 +-------------------------------------------------------+ 668 | 10 | 255 | 192.0.2.1 | VXLAN-gpe | 669 | 10 | 254 | 198.51.100.10 | GRE | 670 | 10 | 251 | 198.51.100.15 | GRE | 671 | 40 | 251 | 198.51.100.15 | GRE | 672 | 50 | 200 | 01:23:45:67:89:ab | Ethernet | 673 | 15 | 212 | Null (end of path) | None | 674 +-------------------------------------------------------+ 676 Figure 9: SFF NSH Mapping Example 678 Additionally, further indirection is possible: the resolution of the 679 required SF network locator may be a localized resolution on an SFF, 680 rather than a service function chain control plane responsibility, as 681 per figures 10 and 11 below. 683 Please note: VXLAN-gpe and GRE in the above table refer to 684 [VXLAN-gpe] and [RFC2784], respectively. 686 +----------------------------+ 687 | SPI | SI | Next hop(s) | 688 +----------------------------+ 689 | 10 | 3 | SF2 | 690 | 245 | 12 | SF34 | 691 | 40 | 9 | SF9 | 692 +----------------------------+ 694 Figure 10: NSH to SF Mapping Example 696 +----------------------------------------+ 697 | SF | Next hop(s) | Transport | 698 +----------------------------------------| 699 | SF2 | 192.0.2.2 | VXLAN-gpe | 700 | SF34| 198.51.100.34 | UDP | 701 | SF9 | 2001:db8::1 | GRE | 702 +--------------------------+------------- 703 = 704 Figure 11: SF Locator Mapping Example 706 Since the SPI is a representation of the service path, the lookup may 707 return more than one possible next-hop within a service path for a 708 given SF, essentially a series of weighted (equally or otherwise) 709 paths to be used (for load distribution, redundancy or policy), see 710 Figure 12. The metric depicted in Figure 12 is an example to help 711 illustrated weighing SFs. In a real network, the metric will range 712 from a simple preference (similar to routing next- hop), to a true 713 dynamic composite metric based on some service function-centric state 714 (including load, sessions state, capacity, etc.) 716 +----------------------------------+ 717 | SPI | SI | NH | Metric | 718 +----------------------------------+ 719 | 10 | 3 | 203.0.113.1 | 1 | 720 | | | 203.0.113.2 | 1 | 721 | | | | | 722 | 20 | 12 | 192.0.2.1 | 1 | 723 | | | 203.0.113.4 | 1 | 724 | | | | | 725 | 30 | 7 | 192.0.2.10 | 10 | 726 | | | 198.51.100.1| 5 | 727 +----------------------------------+ 728 (encapsulation type omitted for formatting) 730 Figure 12: NSH Weighted Service Path 732 7.2. Mapping NSH to Network Transport 734 As described above, the mapping of SPI to network topology may result 735 in a single path, or it might result in a more complex topology. 736 Furthermore, the SPI to overlay mapping occurs at each SFF 737 independently. Any combination of topology selection is possible. 738 Please note, there is no requirement to create a new overlay topology 739 if a suitable one already existing. NSH packets can use any (new or 740 existing) overlay provided the requisite connectivity requirements 741 are satisfied. 743 Examples of mapping for a topology: 745 1. Next SF is located at SFFb with locator 2001:db8::1 746 SFFa mapping: SPI=10 --> VXLAN-gpe, dst-ip: 2001:db8::1 748 2. Next SF is located at SFFc with multiple network locators for 749 load distribution purposes: 750 SFFb mapping: SPI=10 --> VXLAN-gpe, dst_ip:203.0.113.1, 751 203.0.113.2, 203.0.113.3, equal cost 753 3. Next SF is located at SFFd with two paths from SFFc, one for 754 redundancy: 755 SFFc mapping: SPI=10 --> VXLAN-gpe, dst_ip:192.0.2.10 cost=10, 756 203.0.113.10, cost=20 758 In the above example, each SFF makes an independent decision about 759 the network overlay path and policy for that path. In other words, 760 there is no a priori mandate about how to forward packets in the 761 network (only the order of services that must be traversed). 763 The network operator retains the ability to engineer the network 764 paths as required. For example, the overlay path between SFFs may 765 utilize traffic engineering, QoS marking, or ECMP, without requiring 766 complex configuration and network protocol support to be extended to 767 the service path explicitly. In other words, the network operates as 768 expected, and evolves as required, as does the service plane. 770 7.3. Service Plane Visibility 772 The SPI and SI serve an important function for visibility into the 773 service topology. An operator can determine what service path a 774 packet is "on", and its location within that path simply by viewing 775 the NSH information (packet capture, IPFIX, etc.). The information 776 can be used for service scheduling and placement decisions, 777 troubleshooting and compliance verification. 779 7.4. Service Graphs 781 While a given realized service function path is a specific sequence 782 of service functions, the service as seen by a user can actually be a 783 collection of service function paths, with the interconnection 784 provided by classifiers (in-service path, non-initial re- 785 classification). These internal re-classifiers examine the packet at 786 relevant points in the network, and, if needed, SPI and SI are 787 updated (whether this update is a re-write, or the imposition of a 788 new NSH with new values is implementation specific) to reflect the 789 "result" of the classification. These classifiers may also of course 790 modify the metadata associated with the packet. 791 RFC7665, section 2.1 describes Service Graphs in detail. 793 8. Policy Enforcement with NSH 795 8.1. NSH Metadata and Policy Enforcement 797 As described in Section 3, NSH provides the ability to carry metadata 798 along a service path. This metadata may be derived from several 799 sources, common examples include: 801 Network nodes/devices: Information provided by network nodes can 802 indicate network-centric information (such as VRF or tenant) that 803 may be used by service functions, or conveyed to another network 804 node post service path egress. 806 External (to the network) systems: External systems, such as 807 orchestration systems, often contain information that is valuable 808 for service function policy decisions. In most cases, this 809 information cannot be deduced by network nodes. For example, a 810 cloud orchestration platform placing workloads "knows" what 811 application is being instantiated and can communicate this 812 information to all NSH nodes via metadata carried in the context 813 header(s). 815 Service Functions: A classifier co-resident with Service Functions 816 often perform very detailed and valuable classification. In some 817 cases they may terminate, and be able to inspect encrypted 818 traffic. 820 Regardless of the source, metadata reflects the "result" of 821 classification. The granularity of classification may vary. For 822 example, a network switch, acting as a classifier, might only be able 823 to classify based on a 5-tuple, whereas, a service function may be 824 able to inspect application information. Regardless of granularity, 825 the classification information can be represented in NSH. 827 Once the data is added to NSH, it is carried along the service path, 828 NSH-aware SFs receive the metadata, and can use that metadata for 829 local decisions and policy enforcement. The following two examples 830 highlight the relationship between metadata and policy: 832 +-------+ +-------+ +-------+ 833 | SFF )------->( SFF |------->| SFF | 834 +---^---+ +---|---+ +---|---+ 835 ,-|-. ,-|-. ,-|-. 836 / \ / \ / \ 837 ( Class ) SF1 ) ( SF2 ) 838 \ ify / \ / \ / 839 `---' `---' `---' 840 5-tuple: Permit Inspect 841 Tenant A Tenant A AppY 842 AppY 844 Figure 13: Metadata and Policy 846 +-----+ +-----+ +-----+ 847 | SFF |---------> | SFF |----------> | SFF | 848 +--+--+ +--+--+ +--+--+ 849 ^ | | 850 ,-+-. ,-+-. ,-+-. 851 / \ / \ / \ 852 ( Class ) ( SF1 ) ( SF2 ) 853 \ ify / \ / \ / 854 `-+-' `---' `---' 855 | Permit Deny AppZ 856 +---+---+ employees 857 | | 858 +-------+ 859 external 860 system: 861 Employee 862 AppZ 864 Figure 14: External Metadata and Policy 866 In both of the examples above, the service functions perform policy 867 decisions based on the result of the initial classification: the SFs 868 did not need to perform re-classification, rather they rely on a 869 antecedent classification for local policy enforcement. 871 Depending on the information carried in the metadata, data privacy 872 considerations may need to be considered. For example, if the 873 metadata conveys tenant information, that information may need to be 874 authenticated and/or encrypted between the originator and the 875 intended recipients (which may include intended SFs only) . NSH 876 itself does not provide privacy functions, rather it relies on the 877 transport/overlay layer. An operator can select the appropriate 878 transport to ensure the confidentiality (and other security) 879 considerations are met. 881 8.2. Updating/Augmenting Metadata 883 Post-initial metadata imposition (typically performed during initial 884 service path determination), metadata may be augmented or updated: 886 1. Metadata Augmentation: Information may be added to NSH's existing 887 metadata, as depicted in Figure 15. For example, if the initial 888 classification returns the tenant information, a secondary 889 classification (perhaps co-resident with DPI or SLB) may augment 890 the tenant classification with application information, and 891 impose that new information in the NSH metadata. The tenant 892 classification is still valid and present, but additional 893 information has been added to it. 895 2. Metadata Update: Subsequent classifiers may update the initial 896 classification if it is determined to be incorrect or not 897 descriptive enough. For example, the initial classifier adds 898 metadata that describes the traffic as "internet" but a security 899 service function determines that the traffic is really "attack". 900 Figure 16 illustrates an example of updating metadata. 902 +-----+ +-----+ +-----+ 903 | SFF |---------> | SFF |----------> | SFF | 904 +--+--+ +--+--+ +--+--+ 905 ^ | | 906 ,---. ,---. ,---. 907 / \ / \ / \ 908 ( Class ) ( SF1 ) ( SF2 ) 909 \ / \ / \ / 910 `-+-' `---' `---' 911 | Inspect Deny 912 +---+---+ employees employee+ 913 | | Class=AppZ appZ 914 +-------+ 915 external 916 system: 917 Employee 919 Figure 15: Metadata Augmentation 921 +-----+ +-----+ +-----+ 922 | SFF |---------> | SFF |----------> | SFF | 923 +--+--+ +--+--+ +--+--+ 924 ^ | | 925 ,---. ,---. ,---. 926 / \ / \ / \ 927 ( Class ) ( SF1 ) ( SF2 ) 928 \ / \ / \ / 929 `---' `---' `---' 930 5-tuple: Inspect Deny 931 Tenant A Tenant A attack 932 --> attack 934 Figure 16: Metadata Update 936 8.3. Service Path Identifier and Metadata 938 Metadata information may influence the service path selection since 939 the Service Path Identifier and Service Index values can represent 940 the result of classification. A given SPI and SI can be defined 941 based on classification results (including metadata classification). 942 The imposition of the SPI/SI (new or an change of existing) reflect, 943 as previously described, the new SFP. 945 This relationship provides the ability to create a dynamic service 946 plane based on complex classification without requiring each node to 947 be capable of such classification, or requiring a coupling to the 948 network topology. This yields service graph functionality as 949 described in Section 7.4. Figure 17 illustrates an example of this 950 behavior. 952 +-----+ +-----+ +-----+ 953 | SFF |---------> | SFF |------+---> | SFF | 954 +--+--+ +--+--+ | +--+--+ 955 | | | | 956 ,---. ,---. | ,---. 957 / \ / SF1 \ | / \ 958 ( SCL ) ( + ) | ( SF2 ) 959 \ / \SCL2 / | \ / 960 `---' `---' +-----+ `---' 961 5-tuple: Inspect | SFF | Original 962 Tenant A Tenant A +--+--+ next SF 963 --> DoS | 964 V 965 ,-+-. 966 / \ 967 ( SF10 ) 968 \ / 969 `---' 970 DoS 971 "Scrubber" 973 Figure 17: Path ID and Metadata 975 Specific algorithms for mapping metadata to an SPI are outside the 976 scope of this document. 978 9. Security Considerations 980 As with many other protocols, NSH data can be spoofed or otherwise 981 modified. As noted in the descriptive text in [sfc-security- 982 requirements], in many deployments, NSH will be used in a controlled 983 environment, with trusted devices (e.g. a data center) thus 984 mitigating the risk of unauthorized header manipulation. As noted 985 there, far fewer protection mechanisms are needed in these 986 environments, which are the primary design target of NSH. 988 NSH is always encapsulated in a transport protocol and therefore, 989 when required, existing security protocols that provide authenticity 990 (e.g. [ [RFC6071]) can be used between SFF or even to SF. Similarly 991 if confidentiality is required, existing encryption protocols can be 992 used in conjunction with encapsulated NSH. 994 Further, existing best practices, such as [RFC2827] should be 995 deployed at the network layer to ensure that traffic entering the 996 service path is indeed "valid". [encap-considerations] provides 997 additional transport encapsulation considerations. 999 NSH metadata authenticity and confidentiality must be considered as 1000 well. In order to protect the metadata, an operator can leverage the 1001 aforementioned mechanisms provided the transport layer, authenticity 1002 and/or confidentiality. An operator MUST carefully select the 1003 transport/underlay services to ensure end to end security services, 1004 when those are sought after. For example, if RFC6071 is used, the 1005 operator MUST ensure it can be supported by the transport/underlay of 1006 all relevant network segments as well as SFF and SFs. Further, as 1007 described in [section 8.1], operators can and should use indirect 1008 identification for personally identifying information, thus 1009 significantly mitigating the risk of privacy violation. 1011 Further, the extensibility of MD Type 2 to add information to 1012 packets, and where needed to mark that data as critical, allows for 1013 attaching signatures or even encryption keying information to the NSH 1014 header in the future. Based on the learnings from the work on [nsh- 1015 sec], it appears likely that this can provide any needed NSH-specific 1016 security mechanisms in the future. 1018 Lastly, SF security, although out of scope of this document, should 1019 be considered, particularly if an SF needs to access, authenticate or 1020 update NSH metadata. 1022 Further security considerations are discussed in [nsh-sec]. 1024 10. Contributors 1026 This WG document originated as draft-quinn-sfc-nsh and had the 1027 following co-authors and contributors. The editors of this document 1028 would like to thank and recognize them and their contributions. 1029 These co-authors and contributors provided invaluable concepts and 1030 content for this document's creation. 1032 Surendra Kumar 1033 Cisco Systems 1034 smkumar@cisco.com 1036 Michael Smith 1037 Cisco Systems 1038 michsmit@cisco.com 1040 Jim Guichard 1041 Cisco Systems 1042 jguichar@cisco.com 1044 Rex Fernando 1045 Cisco Systems 1046 Email: rex@cisco.com 1048 Navindra Yadav 1049 Cisco Systems 1050 Email: nyadav@cisco.com 1052 Wim Henderickx 1053 Alcatel-Lucent 1054 wim.henderickx@alcatel-lucent.com 1056 Andrew Dolganow 1057 Alcaltel-Lucent 1058 Email: andrew.dolganow@alcatel-lucent.com 1060 Praveen Muley 1061 Alcaltel-Lucent 1062 Email: praveen.muley@alcatel-lucent.com 1064 Tom Nadeau 1065 Brocade 1066 tnadeau@lucidvision.com 1068 Puneet Agarwal 1069 puneet@acm.org 1071 Rajeev Manur 1072 Broadcom 1073 rmanur@broadcom.com 1075 Abhishek Chauhan 1076 Citrix 1077 Abhishek.Chauhan@citrix.com 1079 Joel Halpern 1080 Ericsson 1081 joel.halpern@ericsson.com 1083 Sumandra Majee 1084 F5 1085 S.Majee@f5.com 1087 David Melman 1088 Marvell 1089 davidme@marvell.com 1091 Pankaj Garg 1092 Microsoft 1093 pankajg@microsoft.com 1095 Brad McConnell 1096 Rackspace 1097 bmcconne@rackspace.com 1099 Chris Wright 1100 Red Hat Inc. 1101 chrisw@redhat.com 1103 Kevin Glavin 1104 Riverbed 1105 kevin.glavin@riverbed.com 1107 Hong (Cathy) Zhang 1108 Huawei US R&D 1109 cathy.h.zhang@huawei.com 1111 Louis Fourie 1112 Huawei US R&D 1113 louis.fourie@huawei.com 1115 Ron Parker 1116 Affirmed Networks 1117 ron_parker@affirmednetworks.com 1119 Myo Zarny 1120 Goldman Sachs 1121 myo.zarny@gs.com 1123 11. Acknowledgments 1125 The authors would like to thank Sunil Vallamkonda, Nagaraj Bagepalli, 1126 Abhijit Patra, Peter Bosch, Darrel Lewis, Pritesh Kothari, Tal 1127 Mizrahi and Ken Gray for their detailed review, comments and 1128 contributions. 1130 A special thank you goes to David Ward and Tom Edsall for their 1131 guidance and feedback. 1133 Additionally the authors would like to thank Carlos Pignataro and 1134 Larry Kreeger for their invaluable ideas and contributions which are 1135 reflected throughout this document. 1137 Loa Andersson provided a thorough review and valuable comments, we 1138 thank him for that. 1140 Lastly, Reinaldo Penno deserves a particular thank you for his 1141 architecture and implementation work that helped guide the protocol 1142 concepts and design. 1144 12. IANA Considerations 1146 12.1. NSH EtherType 1148 An IEEE EtherType, 0x894F, has been allocated for NSH. 1150 12.2. Network Service Header (NSH) Parameters 1152 IANA is requested to create a new "Network Service Header (NSH) 1153 Parameters" registry. The following sub-sections request new 1154 registries within the "Network Service Header (NSH) Parameters " 1155 registry. 1157 12.2.1. NSH Base Header Reserved Bits 1159 There are ten bits at the beginning of the NSH Base Header. New bits 1160 are assigned via Standards Action [RFC5226]. 1162 Bits 0-1 - Version 1163 Bit 2 - OAM (O bit) 1164 Bit 3 - Critical TLV (C bit) 1165 Bits 4-9 - Reserved 1167 12.2.2. NSH Version 1169 IANA is requested to setup a registry of "NSH Version". New values 1170 are assigned via Standards Action [RFC5226]. 1172 Version 00: This protocol version. This document. 1173 Version 01: Reserved. This document. 1174 Version 10: Unassigned. 1175 Version 11: Unassigned. 1177 12.2.3. MD Type Registry 1179 IANA is requested to set up a registry of "MD Types". These are 1180 8-bit values. MD Type values 0, 1, 2, 254, and 255 are specified in 1181 this document. Registry entries are assigned by using the "IETF 1182 Review" policy defined in RFC 5226 [RFC5226]. 1184 +---------+--------------+---------------+ 1185 | MD Type | Description | Reference | 1186 +---------+--------------+---------------+ 1187 | 0 | Reserved | This document | 1188 | | | | 1189 | 1 | NSH | This document | 1190 | | | | 1191 | 2 | NSH | This document | 1192 | | | | 1193 | 3..253 | Unassigned | | 1194 | | | | 1195 | 254 | Experiment 1 | This document | 1196 | | | | 1197 | 255 | Experiment 2 | This document | 1198 +---------+--------------+---------------+ 1200 Table 1 1202 12.2.4. MD Class Registry 1204 IANA is requested to set up a registry of "MD Class". These are 16- 1205 bit values. MD Classes defined by this document are assigned as 1206 follows: 1208 0x0000 to 0x01ff: IETF Review 1209 0x0200 to 0xfff5: Expert Review 1210 0xfff6 to 0xfffe: Experimental 1211 0xffff: Reserved 1213 12.2.5. NSH Base Header Next Protocol 1215 IANA is requested to set up a registry of "Next Protocol". These are 1216 8-bit values. Next Protocol values 0, 1, 2, 3, 4 and 5 are defined 1217 in this draft. New values are assigned via Standards Action 1218 [RFC5226]. 1220 +---------------+--------------+---------------+ 1221 | Next Protocol | Description | Reference | 1222 +---------------+--------------+---------------+ 1223 | 0 | Reserved | This document | 1224 | | | | 1225 | 1 | IPv4 | This document | 1226 | | | | 1227 | 2 | IPv6 | This document | 1228 | | | | 1229 | 3 | Ethernet | This document | 1230 | | | | 1231 | 4 | NSH | This document | 1232 | | | | 1233 | 5 | MPLS | This document | 1234 | | | | 1235 | 6..253 | Unassigned | | 1236 | | | | 1237 | 254 | Experiment 1 | This document | 1238 | | | | 1239 | 255 | Experiment 2 | This document | 1240 +---------------+--------------+---------------+ 1242 Table 2 1244 13. References 1246 13.1. Normative References 1248 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1249 Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/ 1250 RFC2119, March 1997, 1251 . 1253 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 1254 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 1255 DOI 10.17487/RFC5226, May 2008, 1256 . 1258 13.2. Informative References 1260 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 1261 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 1262 DOI 10.17487/RFC2784, March 2000, 1263 . 1265 [RFC2827] Ferguson, P. and D. Senie, "Network Ingress Filtering: 1266 Defeating Denial of Service Attacks which employ IP Source 1267 Address Spoofing", BCP 38, RFC 2827, DOI 10.17487/RFC2827, 1268 May 2000, . 1270 [RFC6071] Frankel, S. and S. Krishnan, "IP Security (IPsec) and 1271 Internet Key Exchange (IKE) Document Roadmap", RFC 6071, 1272 DOI 10.17487/RFC6071, February 2011, 1273 . 1275 [RFC7325] Villamizar, C., Ed., Kompella, K., Amante, S., Malis, A., 1276 and C. Pignataro, "MPLS Forwarding Compliance and 1277 Performance Requirements", RFC 7325, DOI 10.17487/RFC7325, 1278 August 2014, . 1280 [RFC7498] Quinn, P., Ed. and T. Nadeau, Ed., "Problem Statement for 1281 Service Function Chaining", RFC 7498, DOI 10.17487/ 1282 RFC7498, April 2015, 1283 . 1285 [RFC7665] Halpern, J., Ed. and C. Pignataro, Ed., "Service Function 1286 Chaining (SFC) Architecture", RFC 7665, DOI 10.17487/ 1287 RFC7665, October 2015, 1288 . 1290 [SFC-CP] Boucadair, M., "Service Function Chaining (SFC) Control 1291 Plane Components & Requirements", 2016, . 1294 [VXLAN-gpe] 1295 Quinn, P., Manur, R., Agarwal, P., Kreeger, L., Lewis, D., 1296 Maino, F., Smith, M., Yong, L., Xu, X., Elzur, U., Garg, 1297 P., and D. Melman, "Generic Protocol Extension for VXLAN", 1298 . 1301 [broadalloc] 1302 Napper, J., Kumar, S., Muley, P., and W. Hendericks, "NSH 1303 Context Header Allocation -- Mobility", 2016, . 1307 [dcalloc] Guichard, J., Smith, M., and et al., "Network Service 1308 Header (NSH) Context Header Allocation (Data Center)", 1309 2016, . 1312 [encap-considerations] 1313 Nordmark, E., Tian, A., Gross, J., Hudson, J., Kreeger, 1314 L., Garg, P., Thaler, P., and T. Herbert, "Encapsulation 1315 Considerations", . 1318 [nsh-sec] Reddy, T., Migault, D., Pignataro, C., Quinn, P., and C. 1319 Inacio, "NSH Security and Privacy requirements", 2016, . 1323 Authors' Addresses 1325 Paul Quinn (editor) 1326 Cisco Systems, Inc. 1328 Email: paulq@cisco.com 1330 Uri Elzur (editor) 1331 Intel 1333 Email: uri.elzur@intel.com