idnits 2.17.1 draft-kaliraj-idr-bgp-classful-transport-planes-13.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 : ---------------------------------------------------------------------------- == There are 29 instances of lines with non-RFC6890-compliant IPv4 addresses in the document. If these are example addresses, they should be changed. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (19 January 2022) is 829 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: 'TBD' is mentioned on line 806, but not defined == Missing Reference: 'RR26' is mentioned on line 868, but not defined == Missing Reference: 'RR27' is mentioned on line 868, but not defined == Missing Reference: 'RR16' is mentioned on line 868, but not defined == Missing Reference: 'CE41' is mentioned on line 873, but not defined == Missing Reference: 'PE25' is mentioned on line 873, but not defined == Missing Reference: 'P28' is mentioned on line 873, but not defined == Missing Reference: 'P29' is mentioned on line 873, but not defined == Missing Reference: 'P15' is mentioned on line 873, but not defined == Missing Reference: 'CE31' is mentioned on line 873, but not defined -- Possible downref: Non-RFC (?) normative reference: ref. 'MPLS-NAMESPACES' -- Possible downref: Non-RFC (?) normative reference: ref. 'PCEP-RSVP-COLOR' -- Possible downref: Non-RFC (?) normative reference: ref. 'RTC-Ext' -- Possible downref: Non-RFC (?) normative reference: ref. 'Seamless-SR' == Outdated reference: A later version (-26) exists of draft-ietf-idr-segment-routing-te-policy-08 -- Possible downref: Non-RFC (?) normative reference: ref. 'SRV6-INTER-DOMAIN' -- Possible downref: Non-RFC (?) normative reference: ref. 'SRV6-MPLS-AGRWL' -- Possible downref: Non-RFC (?) normative reference: ref. 'SRV6-SERVICES' Summary: 0 errors (**), 0 flaws (~~), 13 warnings (==), 8 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group K. Vairavakkalai, Ed. 3 Internet-Draft N. Venkataraman 4 Intended status: Standards Track B. Rajagopalan 5 Expires: 23 July 2022 Juniper Networks, Inc. 6 G. Mishra 7 Verizon Communications Inc. 8 M. Khaddam 9 Cox Communications Inc. 10 X. Xu 11 Capitalonline. 12 R. Szarecki 13 Google. 14 J. Gowda 15 Extreme Networks 16 19 January 2022 18 BGP Classful Transport Planes 19 draft-kaliraj-idr-bgp-classful-transport-planes-13 21 Abstract 23 This document specifies a mechanism, referred to as "service 24 mapping", to express association of overlay routes with underlay 25 routes satisfying a certain SLA, using BGP. The document describes a 26 framework for classifying underlay routes into transport classes, and 27 mapping service routes to specific transport class. 29 The "Transport class" construct maps to a desired SLA, and can be 30 used to realize the "Topology Slice" in 5G Network slicing 31 architecture. 33 This document specifies BGP protocol procedures that enable 34 dissemination of such service mapping information that may span 35 multiple co-operating administrative domains. These domains may be 36 administetered by the same provider or closely co-ordinating provider 37 networks. 39 It makes it possible to advertise multiple tunnels to the same 40 destination address, thus avoiding need of multiple loopbacks on the 41 egress node. 43 A new BGP transport layer address family (SAFI 76) is defined for 44 this purpose that uses RFC-4364 technology and follows RFC-8277 NLRI 45 encoding. This new address family is called "BGP Classful 46 Transport", aka BGP CT. 48 It carries transport prefixes across tunnel domain boundaries (e.g. 49 in Inter-AS Option-C networks), parallel to BGP LU (SAFI 4) . It 50 disseminates "Transport class" information for the transport prefixes 51 across the participating domains, which is not possible with BGP LU. 52 This makes the end-to-end network a "Transport Class" aware tunneled 53 network. 55 Requirements Language 57 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 58 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 59 document are to be interpreted as described in RFC 2119 [RFC2119]. 61 Status of This Memo 63 This Internet-Draft is submitted in full conformance with the 64 provisions of BCP 78 and BCP 79. 66 Internet-Drafts are working documents of the Internet Engineering 67 Task Force (IETF). Note that other groups may also distribute 68 working documents as Internet-Drafts. The list of current Internet- 69 Drafts is at https://datatracker.ietf.org/drafts/current/. 71 Internet-Drafts are draft documents valid for a maximum of six months 72 and may be updated, replaced, or obsoleted by other documents at any 73 time. It is inappropriate to use Internet-Drafts as reference 74 material or to cite them other than as "work in progress." 76 This Internet-Draft will expire on 23 July 2022. 78 Copyright Notice 80 Copyright (c) 2022 IETF Trust and the persons identified as the 81 document authors. All rights reserved. 83 This document is subject to BCP 78 and the IETF Trust's Legal 84 Provisions Relating to IETF Documents (https://trustee.ietf.org/ 85 license-info) in effect on the date of publication of this document. 86 Please review these documents carefully, as they describe your rights 87 and restrictions with respect to this document. Code Components 88 extracted from this document must include Revised BSD License text as 89 described in Section 4.e of the Trust Legal Provisions and are 90 provided without warranty as described in the Revised BSD License. 92 Table of Contents 94 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 95 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 96 3. Transport Class . . . . . . . . . . . . . . . . . . . . . . . 6 97 4. "Transport Class" Route Target Extended Community . . . . . . 7 98 5. Transport RIB . . . . . . . . . . . . . . . . . . . . . . . . 9 99 6. Transport Routing Instance . . . . . . . . . . . . . . . . . 9 100 7. Nexthop Resolution Scheme . . . . . . . . . . . . . . . . . . 9 101 8. BGP Classful Transport Family NLRI . . . . . . . . . . . . . 10 102 9. Comparison with other families using RFC-8277 encoding . . . 11 103 10. Protocol Procedures . . . . . . . . . . . . . . . . . . . . . 12 104 11. Scaling considerations . . . . . . . . . . . . . . . . . . . 15 105 11.1. Avoiding unintended spread of CT routes across 106 domains. . . . . . . . . . . . . . . . . . . . . . . . . 15 107 11.2. Constrained distribution of PNHs to SNs (On Demand 108 Nexthop) . . . . . . . . . . . . . . . . . . . . . . . . 16 109 11.3. Limiting scope of visibility of PE loopback as PNHs . . 17 110 12. OAM considerations . . . . . . . . . . . . . . . . . . . . . 17 111 13. Applicability to Network Slicing . . . . . . . . . . . . . . 18 112 14. SRv6 support . . . . . . . . . . . . . . . . . . . . . . . . 19 113 15. Illustration of procedures with example topology . . . . . . 19 114 15.1. Topology . . . . . . . . . . . . . . . . . . . . . . . . 19 115 15.2. Service Layer route exchange . . . . . . . . . . . . . . 21 116 15.3. Transport Layer route propagation . . . . . . . . . . . 21 117 15.4. Data plane view . . . . . . . . . . . . . . . . . . . . 23 118 15.4.1. Steady state . . . . . . . . . . . . . . . . . . . . 23 119 15.4.2. Absorbing failure of primary path . . . . . . . . . 24 120 16. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 25 121 16.1. New BGP SAFI . . . . . . . . . . . . . . . . . . . . . . 25 122 16.2. New Format for BGP Extended Community . . . . . . . . . 25 123 16.2.1. Existing registries to be modified . . . . . . . . . 25 124 16.2.2. New registries to be created . . . . . . . . . . . . 26 125 16.3. MPLS OAM code points . . . . . . . . . . . . . . . . . . 27 126 17. Security Considerations . . . . . . . . . . . . . . . . . . . 27 127 18. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 27 128 19. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 28 129 20. Normative References . . . . . . . . . . . . . . . . . . . . 28 130 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 30 132 1. Introduction 134 To facilitate service mapping, the tunnels in a network can be 135 grouped by the purpose they serve into a "Transport Class". The 136 tunnels could be created using any signaling protocol, such as LDP, 137 RSVP, BGP LU or SPRING. The tunnels could also use native IP or 138 IPv6, as long as they can carry MPLS payload. Tunnels may exist 139 between different pair of end points. Multiple tunnels may exist 140 between the same pair of end points. 142 Thus, a Transport Class consists of tunnels created by various 143 protocols that satisfy the properties of the class. For example, a 144 "Gold" transport class may consist of tunnels that traverse the 145 shortest path with fast re-route protection, a "Silver" transport 146 class may hold tunnels that traverse shortest paths without 147 protection, a "To NbrAS Foo" transport class may hold tunnels that 148 exit to neighboring AS Foo, and so on. 150 The extensions specified in this document can be used to create a BGP 151 transport tunnel that potentially spans domains, while preserving its 152 Transport Class. Examples of domain are Autonomous System (AS), or 153 IGP area. Within each domain, there is a second level underlay 154 tunnel used by BGP to cross the domain. The second level underlay 155 tunnels could be hetrogeneous: Each domain may use a different type 156 of tunnel (e.g. MPLS, IP, GRE), or use a differnet signaling 157 protocol. A domain boundary is demarcated by a rewrite of BGP 158 nexthop to 'self' while re-advertising tunnel routes in BGP. 159 Examples of domain boundary are inter-AS links and inter-region ABRs. 160 The path uses MPLS label-switching when crossing domain boundary and 161 uses the native intra-AS tunnel of the desired transport class when 162 traversing within a domain. 164 Overlay routes carry sufficient indication of the Transport Classes 165 they should be encapsulated over, in form of BGP community called the 166 "Mapping community". Based on the mapping community, "route 167 resolution" procedure on the ingress node selects from the 168 corresponding Transport Class an appropriate tunnel whose destination 169 matches (LPM) the nexthop of the overlay route. If the overlay route 170 is carried in BGP, the protocol nexthop (or, PNH) is generally 171 carried as an attribute of the route. 173 The PNH of the overlay route is also referred to as "service 174 endpoint" (SEP). The service endpoint may exist in the same domain 175 as the service ingress node or lie in a different domain, adjacent or 176 non-adjacent. In the former case, reachability to the SEP is 177 provided by an intra-domain tunneling protocol, and in the latter 178 case, reachability to the SEP is via BGP transport families. 180 In this architecture, the intra-domain transport protocols (e.g. 181 RSVP, SRTE) are also "Transport Class aware", and they publish 182 ingress routes in Transport RIB associated with the Transport Class, 183 at the tunnel ingress node. These routes are then redistributed into 184 BGP CT to be advertised to adjacent domains. It is outside the scope 185 of this document how exactly the transport protocols are made 186 transport class aware, though configuration on the tunnel ingress 187 node is a simple mechanism to achieve it. 189 This document describes mechanisms to: 191 Model a "Transport Class" as "Transport RIB" on a router, 192 consisting of tunnel ingress routes of a certain class. 194 Enable service routes to resolve over an intended Transport Class 195 by virtue of carrying the appropriate "Mapping community". Which 196 results in using the corresponding Transport RIB for finding 197 nexthop reachability. 199 Advertise tunnel ingress routes in a Transport RIB via BGP without 200 any path hiding, using BGP VPN technology and Add-path. Such that 201 overlay routes in the receiving domains can also resolve over 202 tunnels of associated Transport Class. 204 Provide a way for co-operating domains to reconcile any 205 differences in extended community namespaces, and interoperate 206 between different transport signaling protocols in each domain. 208 In this document we focus mainly on MPLS as the intra-domain 209 transport tunnel forwarding, but the mechanisms described here would 210 work in similar manner for non-MPLS (e.g. IP, GRE, UDP) transport 211 tunnel forwarding technologies too. 213 This document assumes MPLS forwarding when crossing domain 214 boundaries, as that is the defacto standard in deployed networks 215 today. But mechanisms specified in this document can also support 216 different forwarding technologies (e.g. SRv6). 217 Section [SRV6-INTER-DOMAIN]in this document describes adaptation of 218 BGP CT over SRv6 data plane. 220 The document Seamless Segment Routing [Seamless-SR] describes various 221 use cases and applications of procedures described in this document. 223 2. Terminology 225 LSP: Label Switched Path. 227 TE : Traffic Engineering. 229 SN : Service Node. 231 BN : Border Node. 233 TN : Transport Node, P-router. 235 BGP-VPN : VPNs built using RFC4364 mechanisms. 237 RT : Route-Target extended community. 239 RD : Route-Distinguisher. 241 PNH : Protocol-Nexthop address carried in a BGP Update message. 243 SEP : Service End point, the PNH of a Service route. 245 LPM : Longest Prefix Match. 247 Service Family : BGP address family used for advertising routes for 248 "data traffic", as opposed to tunnels. 250 Transport Family : BGP address family used for advertising tunnels, 251 which are in turn used by service routes for resolution. 253 Transport Tunnel : A tunnel over which a service may place traffic. 254 These tunnels can be GRE, UDP, LDP, RSVP, or SR-TE. 256 Tunnel Domain : A domain of the network containing SN and BN, under a 257 single administrative control that has a tunnel between SN and BN. 258 An end-to-end tunnel spanning several adjacent tunnel domains can be 259 created by "stitching" them together using labels. 261 Transport Class : A group of transport tunnels offering the same type 262 of service. 264 Transport Class RT : A Route-Target extended community used to 265 identify a specific Transport Class. 267 Transport RIB : At the SN and BN, a Transport Class has an associted 268 Transport RIB that holds its tunnel routes. 270 Transport Plane : An end to end plane comprising of transport tunnels 271 belonging to same transport class. Tunnels of same transport class 272 are stitched together by BGP route readvertisements with nexthop- 273 self, to span across domain boundaries using Label-Swap forwarding 274 mechanism similar to Inter-AS option-b. 276 Mapping Community : BGP Community/Extended-community on a service 277 route, that maps it to resolve over a Transport Class. 279 3. Transport Class 281 A Transport Class is defined as a set of transport tunnels that share 282 certain characteristics useful for underlay selection. 284 On the wire, a transport class is represented as the Transport Class 285 RT, which is a new Route-Target extended community. 287 A Transport Class is configured at SN and BN, along with attributes 288 like RD and Route-Target. Creation of a Transport Class instantiates 289 the associated Transport RIB and a Transport routing instance to 290 contain them all. 292 The operator may configure a SN/BN to classify a tunnel into an 293 appropriate Transport Class, which causes the tunnel's ingress routes 294 to be installed in the corresponding Transport RIB. At a BN, these 295 tunnel routes may then be advertised into BGP CT. 297 Alternatively, a router receiving the transport routes in BGP with 298 appropriate signaling information can associate those ingress routes 299 to the appropriate Transport Class. E.g. for Classful Transport 300 family (SAFI 76) routes, the Transport Class RT indicates the 301 Transport Class. For BGP LU family(SAFI 4) routes, import processing 302 based on Communities or inter-AS source-peer may be used to place the 303 route in the desired Transport Class. 305 When the ingress route is received via SRTE [SRTE], which encodes the 306 Transport Class as an integer 'Color' in the NLRI as 307 "Color:Endpoint", the 'Color' is mapped to a Transport Class during 308 import processing. SRTE ingress route for 'Endpoint' is installed in 309 that transport class. The SRTE route when advertised out to BGP 310 speakers will then be advertised in Classful Transport family with 311 Transport Class RT and a new label. The MPLS swap route thus 312 installed for the new label will pop the label and deliver 313 decapsulated traffic into the path determined by SRTE route. 315 RFC8664 [RFC8664] extends PCEP to carry SRTE Color. This color 316 association thus learnt is also mapped to a Transport Class thus 317 associating the PCEP signaled SRTE LSP with the desired Transport 318 Class. 320 Similarly, PCEP-RSVP-COLOR [PCEP-RSVP-COLOR] extends PCEP to carry 321 RSVP Color. This color association thus learnt is also mapped to a 322 Transport Class thus associating the PCEP signaled RSVP LSP with the 323 desired Transport Class. 325 4. "Transport Class" Route Target Extended Community 327 This document defines a new type of Route Target, called "Transport 328 Class" Route Target Extended Community. 330 "Transport Class" Route Target extended community is a transitive 331 extended community EXT-COMM [RFC4360] of extended-type, with a new 332 Format (Type high = 0xa) and SubType as 0x2 (Route Target). 334 This new Route Target Format has the following encoding: 336 0 1 2 3 337 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 1 338 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 339 | Type= 0xa | SubType= 0x02 | Reserved | 340 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 341 | Transport Class ID | 342 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 344 "Transport Class" Route Target Extended Community 346 Type: 1 octet 348 Type field contains value 0xa. 350 SubType: 1 octet 352 Subtype field contain 0x2. This indicates 'Route Target'. 354 Transport Class ID: 4 octets 356 The least significant 32-bits of the value field contain the 357 "Transport Class" identifier, which is a 32-bit integer. 359 The remaining 2 octets after SubType field are Reserved, they MUST 360 be set to zero by originator, and ignored, left unaltered by 361 receiver. 363 The "Transport class" Route Target Extended community follows the 364 mechanisms for VPN route import, export as specified in BGP-VPN 365 [RFC4364], and follows the Route Target Contrain mechanisms as 366 specified in VPN-RTC [RFC4684] 368 A BGP speaker that implements RT Constraint VPN-RTC [RFC4684] MUST 369 apply the RT Constraint procedures to the "Transport class" Route 370 Target Extended community as-well. 372 The Transport Class Route Target Extended community is carried on 373 Classful Transport family routes, and allows associating them with 374 appropriate Transport RIBs at receiving BGP speakers. 376 Use of the Transport Class Route Target Extended community with a new 377 Type code avoids conflicts with any VPN Route Target assignments 378 already in use for service families. 380 5. Transport RIB 382 A Transport RIB is a routing-only RIB that is not installed in 383 forwarding path. However, the routes in this RIB are used to resolve 384 reachability of overlay routes' PNH. Transport RIB is created when 385 the Transport Class it represents is configured. 387 Overlay routes that want to use a specific Transport Class confine 388 the scope of nexthop resolution to the set of routes contained in the 389 corresponding Transport RIB. This Transport RIB is the "Routing 390 Table" referred in Section 9.1.2.1 RFC4271 (https://www.rfc- 391 editor.org/rfc/rfc4271#section-9.1.2.1) 393 Routes in a Transport RIB are exported out in 'Classful Transport' 394 address family. 396 6. Transport Routing Instance 398 A BGP VPN routing instance that is a container for the Transport RIB. 399 It imports, and exports routes in this RIB with Transport Class RT. 400 Tunnel destination addresses in this routing instance's context come 401 from the "provider namespace". This is different from user VRFs for 402 e.g., which contain prefixes in "customer namespace" 404 The Transport Routing instance uses the RD and RT configured for the 405 Transport Class. 407 7. Nexthop Resolution Scheme 409 An implementation may provide an option for the service route to 410 resolve over less preferred Transport Classes, should the resolution 411 over preferred, or "primary" Transport Class fail. 413 To accomplish this, the set of service routes may be associated with 414 a user-configured "resolution scheme", which consists of the primary 415 Transport Class, and optionally, an ordered list of fallback 416 Transport Classes. 418 A community called as "Mapping Community" is configured for a 419 "resolution scheme". A Mapping community maps to exactly one 420 resolution scheme. A resolution scheme comprises of one primary 421 transport class and optionally one or more fallback transport 422 classes. 424 A BGP route is associated with a resolution scheme during import 425 processing. The first community on the route that matches a mapping 426 community of a locally configured resolution scheme is considered the 427 effective mapping community for the route. The resolution scheme 428 thus found is used when resolving the route's PNH. If a route 429 contains more than one mapping community, it indicates that the route 430 considers these multiple mapping communities as equivalent. So the 431 first community that maps to a resolution scheme is chosen. 433 A transport route received in BGP Classful Transport family SHOULD 434 use a resolution scheme that contains the primary Transport Class 435 without any fallback to best effort tunnels. The primary Transport 436 Class is identified by the Transport Class RT carried on the route. 437 Thus Transport Class RT serves as the Mapping Community for Classful 438 Transport routes. 440 A service route received in a BGP service family MAY map to a 441 resolution scheme that contains the primary Transport Class 442 identified by the mapping community on the route, and a fallback to 443 best effort tunnels transport class. The primary Transport Class is 444 identified by the Mapping community carried on the route. For e.g. 445 the Extended Color community may serve as the Mapping Community for 446 service routes. Color:0: MAY map to a resolution scheme that has 447 primary transport class , and a fallback to best-effort transport 448 class. 450 8. BGP Classful Transport Family NLRI 452 The Classful Transport (CT) family will use the existing AFI of IPv4 453 or IPv6, and a new SAFI 76 "Classful Transport" that will apply to 454 both IPv4 and IPv6 AFIs. These AFI, SAFI pair of values MUST be 455 negotiated in Multiprotocol Extensions capability described in 456 [RFC4760] to be able to send and receive BGP CT routes. 458 The "Classful Transport" SAFI NLRI itself is encoded as specified in 459 https://tools.ietf.org/html/rfc8277#section-2 [RFC8277]. 461 When AFI is IPv4 the "Prefix" portion of Classful Transport family 462 NLRI consists of an 8-byte RD followed by an IPv4 prefix. When AFI 463 is IPv6 the "Prefix" consists of an 8-byte RD followed by an IPv6 464 prefix. 466 Attributes on a Classful Transport route include the Transport Class 467 Route-Target extended community, which is used to leak the route into 468 the right Transport RIBs on SNs and BNs in the network. 470 SAFI 76 routes can be sent with either IPv4 or IPv6 nexthop. The 471 type of nexthop is inferred from the length of nexthop. 473 When the length of Next Hop Address field is 24 (or 48) the nexthop 474 address is of type VPN-IPv6 with 8-octet RD set to zero (potentially 475 followed by the link-local VPN-IPv6 address of the next hop with an 476 8-octet RD set to zero). 478 When the length of Next Hop Address field is 12 the nexthop address 479 is of type VPN-IPv4 with 8-octet RD set to zero. 481 9. Comparison with other families using RFC-8277 encoding 483 SAFI 128 (Inet-VPN) is a RF8277 encoded family that carries service 484 prefixes in the NLRI, where the prefixes come from the customer 485 namespaces, and are contexualized into separate user virtual service 486 RIBs called VRFs, using RFC4364 procedures. 488 SAFI 4 (BGP LU) is a RFC8277 encoded family that carries transport 489 prefixes in the NLRI, where the prefixes come from the provider 490 namespace. 492 SAFI 76 (Classful Transport) is a RFC8277 encoded family that carries 493 transport prefixes in the NLRI, where the prefixes come from the 494 provider namespace, but are contexualized into separate Transport 495 RIBs, using RFC4364 procedures. 497 It is worth noting that SAFI 128 has been used to carry transport 498 prefixes in "L3VPN Inter-AS Carrier's carrier" scenario, where BGP 499 LU/LDP prefixes in CsC VRF are advertised in SAFI 128 towards the 500 remote-end baby carrier. 502 In this document a new AFI/SAFI is used instead of reusing SAFI 128 503 to carry these transport routes, because it is operationally 504 advantageous to segregate transport and service prefixes into 505 separate address families, RIBs. E.g. It allows to safely enable 506 "per-prefix" label allocation scheme for Classful Transport prefixes 507 without affecting SAFI 128 service prefixes which may have huge 508 scale. "per prefix" label allocation scheme keeps the routing churn 509 local during topology changes. 511 A new family also facilitates having a different readvertisement path 512 of the transport family routes in a network than the service route 513 readvertisement path. viz. Service routes (Inet-VPN) are exchanged 514 over an EBGP multihop session between Autonomous systems with nexthop 515 unchanged; whereas Classful Transport routes are readvertised over 516 EBGP single hop sessions with "nexthop-self" rewrite over inter-AS 517 links. 519 The Classful Transport family is similar in vein to BGP LU, in that 520 it carries transport prefixes. The only difference is, it also 521 carries in Route Target an indication of which Transport Class the 522 transport prefix belongs to, and uses RD to disambiguate multiple 523 instances of the same transport prefix in a BGP Update. 525 10. Protocol Procedures 527 This section summarizes the procedures followed by various nodes 528 speaking Classful Transport family 530 Preparing the network for deploying Classful Transport planes 532 Operator decides on the Transport Classes that exist in the 533 network, and allocates a Route-Target to identify each Transport 534 Class. 536 Operator configures Transport Classes on the SNs and BNs in the 537 network with unique Route-Distinguishers and Route-Targets. 539 Implementations may provide automatic generation and assignment of 540 RD, RT values for a transport routing instance; they MAY also 541 provide a way to manually override the automatic mechanism, in 542 order to deal with any conflicts that may arise with existing RD, 543 RT values in the different network domains participating in a 544 deployment. 546 Origination of Classful Transport route: 548 At the ingress node of the tunnel's home domain, the tunneling 549 protocols install routes in the Transport RIB associated with the 550 Transport Class the tunnel belongs to. 552 The ingress node then advertises this tunnel destination into BGP 553 as a Classful Transport family route with NLRI RD:TunnelEndpoint, 554 attaching a 'Transport Class' Route Target that identifies the 555 Transport Class. This BGP CT route is advertised to EBGP peers 556 and IBGP peers which are RR-clients. This route MUST NOT be 557 advertised to the IBGP peers who are not RR-clients. 559 Alternatively, the egress node of the tunnel i.e. the tunnel 560 endpoint can originate the same BGP Classful Transport route, with 561 NLRI RD:TunnelEndpoint and PNH TunnelEndpoint, which will resolve 562 over the tunnel route at the ingress node. When the tunnel is up, 563 the Classful Transport BGP route will become usable and get re- 564 advertised. 566 Unique RD SHOULD be used by the originator of a Classful Transport 567 route to disambiguate the multiple BGP advertisements for a 568 transport end point. 570 Ingress node receiving Classful Transport route 572 On receiving a BGP Classful Transport route with a PNH that is not 573 directly connected, e.g. an IBGP-route, a mapping community on the 574 route (the Transport Class RT) indicates which Transport Class 575 this route maps to. The routes in the associated Transport RIB 576 are used to resolve the received PNH. If there does not exist a 577 route in the Transport RIB matching the PNH, the Classful 578 Transport route is considered unusable, and MUST NOT be re- 579 advertised further. 581 Border node readvertising Classful Transport route with nexthop self: 583 The BN allocates an MPLS label to advertise upstream in Classful 584 Transport NLRI. The BN also installs an MPLS swap-route for that 585 label that swaps the incoming label with a label received from the 586 downstream BGP speaker, or pops the incoming label. And then 587 pushes received traffic to the transport tunnel or direct 588 interface that the Classful Transport route's PNH resolved over. 590 The label SHOULD be allocated with "per-prefix" label allocation 591 semantics. RD is stripped from the BGP CT NLRI prefix when a BGP 592 CT route is leaked to a Transport RIB. The IP prefix in the 593 transport RIB context (IP-prefix, Transport-Class) is used as the 594 key to do per-prefix label allocation. This helps in avoiding BGP 595 CT route churn through out the CT network when a failure happens 596 in a domain. The failure is not propagated further than the BN 597 closest to the failure. 599 The value of advertised MPLS label is locally significant, and is 600 dynamic by default. The BN may provide option to allocate a value 601 from a statically carved out range. This can be achieved using 602 locally configured export policy, or via mechanisms described in 603 BGP Prefix-SID [RFC8669]. 605 Border node receiving Classful Transport route on EBGP : 607 If the route is received with PNH that is known to be directly 608 connected, e.g. EBGP single-hop peering address, the directly 609 connected interface is checked for MPLS forwarding capability. No 610 other nexthop resolution process is performed, as the inter-AS 611 link can be used for any Transport Class. 613 If the inter-AS links should honor Transport Class, then the BN 614 SHOULD follow procedures of an Ingress node described above, and 615 perform nexthop resolution process. The interface routes SHOULD 616 be installed in the Transport RIB belonging to the associated 617 Transport Class. 619 Avoiding path-hiding through Route Reflectors 621 When multiple BNs exist that advertise a RDn:PEn prefix to RRs, 622 the RRs may hide all but one of the BNs, unless ADDPATH [RFC7911] 623 is used for the Classful Transport family. This is similar to 624 L3VPN option-B scenarios. Hence ADDPATH SHOULD be used for 625 Classful Transport family, to avoid path-hiding through RRs. 627 Avoiding loop between Route Reflectors in forwarding path 629 Pair of redundant ABRs acting as RR with nexthop-self may chose 630 each other as best path instead of the upstream ASBR, causing a 631 traffic forwarding loop. 633 Implementations SHOULD provide a way to alter the tie-breaking 634 rule specified in BGP RR [RFC4456] to tie-break on CLUSTER_LIST 635 step before ROUTER-ID step, when performing path selection for BGP 636 CT routes. RFC4456 considers pure RR which is not in forwarding 637 path. When RR is in forwarding path and reflects routes with 638 nexthop-self, which is the case for ABR BNs in a BGP transport 639 network, this rule may cause loops. This document suggests the 640 following modification to the BGP Decision Process Tie Breaking 641 rules (Sect. 9.1.2.2, [RFC4271]) when doing path selection for BGP 642 CT family routes: 644 The following rule SHOULD be inserted between Steps e) and f): a 645 BGP Speaker SHOULD prefer a route with the shorter CLUSTER_LIST 646 length. The CLUSTER_LIST length is zero if a route does not carry 647 the CLUSTER_LIST attribute. 649 Some deployment considerations can also help in avoiding this 650 problem: 652 - IGP metric should be assigned such that "ABR to redundant ABR" 653 cost is inferior than "ABR to upstream ASBR" cost. 655 - Tunnels belonging to special Transport classes SHOULD NOT be 656 provisioned between ABR to ABRs. This will ensure that the 657 route received from an ABR with nexthop-self will not be usable 658 at a redundant ABR. 660 This avoids possibility of such loops altogether, irrespective of 661 whether the path selection modification mentioned above is 662 implemented. 664 Ingress node receiving service route with mapping community 666 Service routes received with mapping community resolve using 667 Transport RIBs determined by the resolution scheme. If the 668 resolution process does not find an usable Classful Transport 669 route or tunnel route in any of the Transport RIBs, the service 670 route MUST be considered unusable for forwarding purpose. 672 Coordinating between domains using different community namespaces. 674 Cooperating option-C domains may sometimes not agree on RT, RD, 675 Mapping-community or Transport Route Target values because of 676 differences in community namespaces; e.g. during network mergers 677 or renumbering for expansion. Such deployments may deploy 678 mechanisms to map and rewrite the Route-target values on domain 679 boundaries, using per ASBR import policies. This is no different 680 than any other BGP VPN family. Mechanisms employed in inter-AS 681 VPN deployments may be used with the Classful Transport family 682 also. 684 The resolution schemes SHOULD allow association with multiple 685 mapping communities. This helps with renumbering, network 686 mergers, or transitions. 688 Though RD can also be rewritten on domain boundaries, deploying 689 unique RDs is strongly RECOMMENDED, because it helps in trouble 690 shooting by uniquely identifying originator of a route, and avoids 691 path-hiding. 693 This document defines a new format of Route-Target extended- 694 community to carry Transport Class, this avoids collision with 695 regular Route Target namespace used by service routes. 697 11. Scaling considerations 699 11.1. Avoiding unintended spread of CT routes across domains. 701 RFC8212 [RFC8212] suggests BGP speakers require explicit 702 configuration of both BGP Import and Export Policies for any EBGP 703 sessions, in order to receive or send routes on EBGP sessions. 705 It is recommended to follow this for BGP CT routes. It will prohibit 706 unintended advertisement of transport routes through out the BGP CT 707 transport domain which may span multiple AS. This will conserve 708 usage of MPLS label and nexthop resources in the network. An ASBR of 709 a domain can be provisioned to allow routes with only the Transport 710 targets that are required by SNs in the domain. 712 11.2. Constrained distribution of PNHs to SNs (On Demand Nexthop) 714 This section describes how the number of Protocol Nexthops 715 advertised to a SN or BN can be constrained using BGP Classsful 716 Transport and VPN RTC [RFC4684] 718 An egress SN MAY advertise BGP CT route for RD:eSN with two Route 719 Targets: transport-target:0: and a RT carrying :. 720 Where TC is the Transport Class identifier, and eSN is the IP- 721 address used by SN as BGP nexthop in it's service route 722 advertisements. 724 transport-target:0: is the new type of route target (Transport 725 Class RT) defined in this document. It is carried in BGP extended 726 community attribute (BGP attribute code 16). 728 The RT carrying : MAY be an IP-address specific regular 729 RT (BGP attribute code 16), IPv6-address specific RT (BGP 730 attribute code 25), or a Wide-communities based RT (BGP attribute 731 code 34) as described in RTC-Ext [RTC-Ext] 733 An ingress SN MAY import BGP CT routes with Route Target carrying: 734 :. The ingress SN MAY learn the eSN values either by 735 configuration, or it MAY discover them from the BGP nexthop field 736 in the BGP VPN service routes received from eSN. A BGP ingress SN 737 receiving a BGP service route with nexthop of eSN SHOULD generate 738 a RTC/Extended-RTC route for Route Target prefix :/[80|176] in order to learn BGP CT transport routes to 740 reach eSN. This allows constrained distribution of the transport 741 routes to the PNHs actually required by iSN. 743 When path of route propogation of BGP CT routes is same as the RTC 744 routes, a BN would learn the RTC routes advertised by ingress SNs 745 and propagate further. This will allow constraining distribution 746 of BGP CT routes for a PNH to only the necessary BNs in the 747 network, closer to the egress SN. 749 This mechanism provides "On Demand Nexthop" of BGP CT routes, 750 which help with scaling of MPLS forwarding state at SN and BN. 752 But the amount of state carried in RTC family may become 753 proportional to number of PNHs in the network. To strike a 754 balance, the RTC route advertisements for :/[80|176] MAY be confined to the BNs in home region of 756 ingress-SN, or the BNs of a super core. 758 Such a BN in the core of the network SHOULD import BGP CT routes 759 with Transport Class Route Target: 0:, and generate a RTC 760 route for :0:/96, while not propagating the more 761 specific RTC requests for specific PNHs. This will let the BN 762 learn transport routes to all eSN nodes. But confine their 763 propagation to ingress-SNs. 765 11.3. Limiting scope of visibility of PE loopback as PNHs 767 It may be even more desirable to limit the number of PNHs that are 768 globaly visible in the network. This is possible using mechanism 769 described in MPLS Namespaces [MPLS-NAMESPACES] 771 Such that advertisement of PE loopback addresses as next-hop in BGP 772 service routes is confined to the region they belong to. An anycast 773 IP-address called "Context Protocol Nexthop Address" abstracts the 774 PEs in a region from other regions in the network, swapping the PE 775 scoped service label with a CPNH scoped private namespace label. 777 This provides much greater advantage in terms of scaling and 778 convergence. Changes to implement this feature are required only on 779 the region's BNs and RR. 781 12. OAM considerations 783 Standard MPLS OAM procedures specified in [RFC8029] also apply to BGP 784 Classful Transport. 786 The 'Target FEC Stack' sub-TLV for IPv4 Classful Transport has a Sub- 787 Type of [TBD], and a length of 13. The Value field consists of the 788 RD advertised with the Classful Transport prefix, the IPv4 prefix 789 (with trailing 0 bits to make 32 bits in all), and a prefix length, 790 encoded as follows: 792 0 1 2 3 793 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 1 794 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 795 | Route Distinguisher | 796 | (8 octets) | 797 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 798 | IPv4 prefix | 799 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 800 | Prefix Length | Must Be Zero | 801 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 803 Figure 1: Classful Transport IPv4 FEC 805 The 'Target FEC Stack' sub-TLV for IPv6 Classful Transport has a Sub- 806 Type of [TBD], and a length of 25. The Value field consists of the 807 RD advertised with the Classful Transport prefix, the IPv6 prefix 808 (with trailing 0 bits to make 128 bits in all), and a prefix length, 809 encoded as follows: 811 0 1 2 3 812 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 1 813 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 814 | Route Distinguisher | 815 | (8 octets) | 816 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 817 | IPv6 prefix | 818 | | 819 | | 820 | | 821 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 822 | Prefix Length | Must Be Zero | 823 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 825 Figure 2: Classful Transport IPv6 FEC 827 13. Applicability to Network Slicing 829 In Network Slicing, the Transport Slice Controller (TSC) sets up the 830 Topology (e.g. RSVP, SR-TE tunnels with desired characteristics) and 831 resources (e.g. polices/shapers) in a transport network to create a 832 Transport slice. The Transport class construct described in this 833 document represents the "Topology Slice" portion of this equation. 835 The TSC can use the Transport Class Identifier (Color value) to 836 provision a transport tunnel in a specific Topology Slice. 838 Further, Network slice controller can use the Mapping community on 839 the service route to map traffic to the desired Transport slice. 841 14. SRv6 support 843 This section describes how BGP CT may be used to set up inter domain 844 tunnels of a certain Transport Class, when using Segment Routing over 845 IPv6 (SRv6) data plane on the inter AS links or as intra-AS tunneling 846 mechanism. 848 RFC8986, [SRV6-INTER-DOMAIN] specify the SRv6 Endpoint behaviors (End 849 USD, End.BM, End.B6.Encaps and End.Replace, End.ReplaceB6, 850 respectively). These are leveraged for BGP CT with SRv6 data plane. 852 The BGP Classful Transport route update for SRv6 MUST include the BGP 853 Prefix-SID attribute along with SRv6 SID information as specified in 854 [SRV6-SERVICES]. It may also include SRv6 SID structure for 855 Transposition as specified in [SRV6-SERVICES]. It should be noted 856 that prefixes carried in BGP CT family are transport layer end- 857 points, e.g. PE loopback addresses. Thus the SRv6 SID carried in a 858 BGP CT route is also a transport layer identifier. 860 This document extends the usage of "SRv6 label route tunnel" TLV to 861 AFI=1/2 SAFI 76. "SRv6 label route tunnel" is the TLV of the BGP 862 Prefix-SID Attribute as specified in [SRV6-MPLS-AGRWL]. 864 15. Illustration of procedures with example topology 866 15.1. Topology 868 [RR26] [RR27] [RR16] 869 | | | 870 | | | 871 |+-[ABR23]--+|+--[ASBR21]---[ASBR13]-+|+--[PE11]--+ 872 || ||| ` / ||| | 873 [CE41]--[PE25]--[P28] [P29] `/ [P15] [CE31] 874 | | | /` | | | 875 | | | / ` | | | 876 | | | / ` | | | 877 +--[ABR24]--+ +--[ASBR22]---[ASBR14]-+ +--[PE12]--+ 879 | | | | 880 + + + + 881 CE | region-1 | region-2 | |CE 882 AS4 ...AS2... AS1 AS3 884 41.41.41.41 ------------ Traffic Direction ----------> 31.31.31.31 885 This example shows a provider network that comprises of two 886 Autonomous systems, AS1, AS2. They are serving customers AS3, AS4 887 respectively. Traffic direction being described is CE41 to CE31. 888 CE31 may request a specific SLA, e.g. Gold for this traffic, when 889 traversing these provider networks. 891 AS2 is further divided into two regions. So there are three tunnel 892 domains in provider space. AS1 uses ISIS Flex-Algo intra-domain 893 tunnels, whereas AS2 uses RSVP intra-domain tunnels. 895 The network has two Transport classes: Gold with transport class id 896 100, Bronze with transport class id 200. These transport classes are 897 provisioned at the PEs and the Border nodes (ABRs, ASBRs) in the 898 network. 900 Following tunnels exist for Gold transport class. 902 PE25_to_ABR23_gold - RSVP tunnel 904 PE25_to_ABR24_gold - RSVP tunnel 906 ABR23_to_ASBR22_gold - RSVP tunnel 908 ASBR13_to_PE11_gold - ISIS FlexAlgo tunnel 910 ASBR14_to_PE11_gold - ISIS FlexAlgo tunnel 912 Following tunnels exist for Bronze transport class. 914 PE25_to_ABR23_bronze - RSVP tunnel 916 ABR23_to_ASBR21_bronze - RSVP tunnel 918 ABR23_to_ASBR22_bronze - RSVP tunnel 920 ABR24_to_ASBR21_bronze - RSVP tunnel 922 ASBR13_to_PE12_bronze - ISIS FlexAlgo tunnel 924 ASBR14_to_PE11_bronze - ISIS FlexAlgo tunnel 926 These tunnels are either provisioned or auto-discovered to belong to 927 transport class 100 or 200. 929 15.2. Service Layer route exchange 931 Service nodes PE11, PE12 negotiate service families (SAFI 1, 128) on 932 the BGP session with RR16. Service helpers RR16, RR26 have multihop 933 EBGP session to exchange service routes between the two AS. 934 Similarly PE25 negotiates service families with RR26. 936 Forwarding happens using service routes at service nodes PE25, PE11, 937 PE12 only. Routes received from CEs are not present in any other 938 nodes' FIB in the network. 940 CE31 advertises a route for example prefix 31.31.31.31 with nexthop 941 self to PE11, PE12. CE31 can attach a mapping community Color:0:100 942 on this route, to indicate its request for Gold SLA. Or, PE11 can 943 attach the same using locally configured policies. Let us assume 944 CE31 is getting VPN service from PE25. 946 The 31.31.31.31 route is readvertised in SAFI 128 by PE11 with 947 nexthop self (1.1.1.1) and label V-L1, to RR16 with the mapping 948 community Color:0:100 attached. This SAFI 128 route reaches PE25 via 949 RR16, RR26 with the nexthop unchanged, as PE11 and label V-L1. Now 950 PE25 can resolve the PNH 1.1.1.1 using transport routes received in 951 BGP CT or BGP LU. 953 The IP FIB at PE25 will have a route for 31.31.31.31 with a nexthop 954 thus found, that points to a Gold tunnel in ingress domain. 956 15.3. Transport Layer route propagation 958 ASBR13 negotiates BGP CT family with transport ASBRs ASBR21, ASBR22. 959 They negotiate BGP CT family with RR27 in region 2. ABR23, ABR24 960 negotiate BGP CT family with RR27 in region 2 and RR26 in region 1. 961 PE25 receives BGP CT routes from RR26. BGP LU family is also 962 negotiated on these sessions alongside BGP CT family. BGP LU carries 963 "best effort" transport class routes, BGP CT carries gold, bronze 964 transport class routes. 966 ASBR13 is provisioned with transport class 100, RD value 1.1.1.3:10 967 and a transport route target 0:100. And a Transport class 200 with 968 RD value 1.1.1.3:20, and transport route target 0:200. 970 Similarly, these transport classes are also configured on ASBRs, ABRs 971 and PEs, with same transport route target, but unique RDs. 973 Ingress route for ASBR13_to_PE11_gold is advertised by ASBR13 in BGP 974 CT family to ASBRs ASBR21, ASBR22. This route is sent with a NLRI 975 containing RD prefix 1.1.1.3:10:1.1.1.1, Label B-L1 and a route 976 target extended community transport-target:0:100. MPLS swap route is 977 installed at ASBR13 for B-L1 with a nexthop pointing to 978 ASBR13_to_PE11_gold tunnel. 980 Ingress route for ASBR13_to_PE11_bronze is advertised by ASBR13 in 981 BGP CT family to ASBRs ASBR21, ASBR22. This route is sent with a 982 NLRI containing RD prefix 1.1.1.3:20:1.1.1.1, Label B-L2 and a route 983 target extended community transport-target:0:200. MPLS swap route is 984 installed at ASBR13 for label B-L2 with a nexthop pointing to 985 ASBR13_to_PE11_bronze tunnel 987 ASBR21 receives BGP CT route 1.1.1.3:10:1.1.1.1 over the single hop 988 EBGP sesion, and readvertises with nexthop self (loopback adderss 989 2.2.2.1) to RR27, advertising a new label B-L3. MPLS swap route is 990 installed for label B-L3 at ASBR21 to swap to received label B-L1 and 991 forwards to ASBR13. RR27 readvertises this BGP CT route to ABR23, 992 ABR24. 994 ASBR22 receives BGP CT route 1.1.1.3:10:1.1.1.1 over the single hop 995 EBGP sesion, and readvertises with nexthop self (loopback adderss 996 2.2.2.2) to RR27, advertising a new label B-L4. MPLS swap route is 997 installed for label B-L4 at ASBR22 to swap to received label B-L2 and 998 forwards to ASBR13. RR27 readvertises this BGP CT route to ABR23, 999 ABR24. 1001 Addpath is enabled for BGP CT family on the sessions between RR27 and 1002 ASBRs, ABRs. Such that routes for 1.1.1.3:10:1.1.1.1 with the 1003 nexthops ASBR21 and ASBR22 are reflected to ABR23, ABR24 without any 1004 path hiding. Thus giving ABR23 visibiity of both available nexthops 1005 for Gold SLA. 1007 ABR23 receives the route with nexthop 2.2.2.1, label B-L3 from RR27. 1008 The route target "transport-target:0:100" on this route acts as 1009 mapping community, and instructs ABR23 to strictly resolve the 1010 nexthop using transport class 100 routes only. ABR23 is unable to 1011 find a route for 2.2.2.1 with transport class 100. Thus it considers 1012 this route unusable and does not propagate it further. This prunes 1013 ASBR21 from Gold SLA tunneled path. 1015 ABR23 also receives the route with nexthop 2.2.2.2, label B-L4 from 1016 RR27. The route target "transport-target:0:100" on this route acts 1017 as mapping community, and instructs ABR23 to strictly resolve the 1018 nexthop using transport class 100 routes only. ABR23 successfully 1019 resolves the nexthop to point to ABR23_to_ASBR22_gold tunnel. ABR23 1020 readvertises this route with nexthop self (loopback address 2.2.2.3) 1021 and a new label B-L5 to RR26. Swap route for B-L5 is installed by 1022 ABR23 to swap to label B-L4, and forward into ABR23_to_ASBR22_gold 1023 tunnel. 1025 RR26 reflects the route from ABR23 to PE25. PE25 receives the BGP CT 1026 route for prefix 1.1.1.3:10:1.1.1.1 with label B-L5, nexthop 2.2.2.3 1027 and transport-target:0:100 from RR26. And it similarly resolves the 1028 nexthop 2.2.2.3 over transport class 100, pushing labels associated 1029 with PE25_to_ABR23_gold tunnel. 1031 In this manner, the Gold transport LSP "ASBR13_to_PE11_gold" in 1032 egress-domain is extended by BGP CT until the ingress-node PE25 in 1033 ingress domain, to create an end-to-end Gold SLA path. MPLS swap 1034 routes are installed at ASBR13, ASBR22 and ABR23, when propagating 1035 the PE11 BGP CT Gold transport class route 1.1.1.3:10:1.1.1.1 with 1036 nexthop self towards PE25. 1038 The BGP CT LSP thus formed, originates in PE25, and terminates in 1039 ASBR13, traversing over the Gold underlay LSPs in each domain. 1040 ASBR13 uses UHP to stitch the BGP CT LSP into the 1041 "ASBR13_to_PE11_gold" LSP to traverse the last domain, thus 1042 satisfying Gold SLA end-to-end. 1044 When PE25 receives service route with nexthop 1.1.1.1 and mapping 1045 community Color:0:100, it resolves over this BGP CT route 1046 1.1.1.3:10:1.1.1.1. Thus pushing label B-L5, and pushing as top 1047 label the labels associated with PE25_to_ABR23_gold tunnel. 1049 15.4. Data plane view 1051 15.4.1. Steady state 1053 This section describes how the data plane looks like in steady state. 1055 CE41 transmits an IP packet with destination as 31.31.31.31. On 1056 receiving this packet PE25 performs a lookup in the IP FIB associated 1057 with the CE41 interface. This lookup yeids the service route that 1058 pushes the VPN service label V-L1, BGP CT label B-L5, and labels for 1059 PE25_to_ABR23_gold tunnel. Thus PE25 encapsulates the IP packet in 1060 MPLS packet with label V-L1(innermost), B-L5, and top label as 1061 PE25_to_ABR23_gold tunnel. This MPLS packet is thus transmitted to 1062 ABR23 using Gold SLA. 1064 ABR23 decapsulates the packet received on PE25_to_ABR23_gold tunnel 1065 as required, and finds the MPLS packet with label B-L5. It performs 1066 lookup for label B-L5 in the global MPLS FIB. This yields the route 1067 that swaps label B-L5 with label B-L4, and pushes top label provided 1068 by ABR23_to_ASBR22_gold tunnel. Thus ABR23 transmits the MPLS packet 1069 with label B-L4 to ASBR22, on a tunnel that satisfies Gold SLA. 1071 ASBR22 similarly performs a lookup for label B-L4 in global MPLS FIB, 1072 finds the route that swaps label B-L4 with label B-L2, and forwards 1073 to ASBR13 over the directly connected MPLS enabled interface. This 1074 interface is a common resource not dedicated to any specific 1075 transport class, in this example. 1077 ASBR13 receives the MPLS packet with label B-L2, and performs a 1078 lookup in MPLS FIB, finds the route that pops label B-L2, and pushes 1079 labels associated with ASBR13_to_PE11_gold tunnel. This transmits 1080 the MPLS packet with VPN label V-L1 to PE11, using a tunnel that 1081 preserves Gold SLA in AS 1. 1083 PE11 receives the MPLS packet with V-L1, and performs VPN forwarding. 1084 Thus transmitting the original IP payload from CE41 to CE31. The 1085 payload has traversed path satisfying Gold SLA end-to-end. 1087 15.4.2. Absorbing failure of primary path 1089 This section describes how the data plane reacts when gold path 1090 experiences a failure. 1092 Let us assume tunnel ABR23_to_ASBR22_gold goes down, such that now 1093 end-to-end Gold path does not exist in the network. This makes the 1094 BGP CT route for RD prefix 1.1.1.1:10:1.1.1.1 unusable at ABR23. 1095 This makes ABR23 send a BGP withdrawal for 1.1.1.1:10:1.1.1.1 to 1096 RR26, which then withdraws the prefix from PE25. 1098 Withdrawal for 1.1.1.1:10:1.1.1.1 allows PE25 to react to the loss of 1099 gold path to 1.1.1.1. Let us assume PE25 is provisioned to use best- 1100 effort transport class as the backup path. This withdrawal of BGP CT 1101 route allows PE25 to adjust the nexthop of the VPN Service-route to 1102 push the labels provided by the BGP LU route. That repairs the 1103 traffic to go via best effort path. PE25 can also be provisioned to 1104 use Bronze transport class as the backup path. The repair will 1105 happen in similar manner in that case as-well. 1107 Traffic repair to absorb the failure happens at ingress node PE25, in 1108 a service prefix scale independent manner. This is called PIC 1109 (Prefix scale Independent Convergence). The repair time will be 1110 proportional to time taken for withdrawing the BGP CT route. 1112 16. IANA Considerations 1114 This document makes following requests of IANA. 1116 16.1. New BGP SAFI 1118 New BGP SAFI code for "Classful Transport". Value 76. 1120 This will be used to create new AFI,SAFI pairs for IPv4, IPv6 1121 Classful Transport families. viz: 1123 * "Inet, Classful Transport". AFI/SAFI = "1/76" for carrying IPv4 1124 Classful Transport prefixes. 1126 * "Inet6, Classful Transport". AFI/SAFI = "2/76" for carrying IPv6 1127 Classful Transport prefixes. 1129 16.2. New Format for BGP Extended Community 1131 Please assign a new Format (Type high = 0xa) of extended community 1132 EXT-COMM [RFC4360] called "Transport Class" from the following 1133 registries: 1135 the "BGP Transitive Extended Community Types" registry, and 1137 the "BGP Non-Transitive Extended Community Types" registry. 1139 Please assign the same low-order six bits for both allocations. 1141 This document uses this new Format with subtype 0x2 (route target), 1142 as a transitive extended community. 1144 The Route Target thus formed is called "Transport Class" route target 1145 extended community. 1147 Taking reference of RFC7153 [RFC7153] , following requests are made: 1149 16.2.1. Existing registries to be modified 1151 16.2.1.1. Registries for the "Type" Field 1153 16.2.1.1.1. Transitive Types 1155 This registry contains values of the high-order octet (the "Type" 1156 field) of a Transitive Extended Community. 1158 Registry Name: BGP Transitive Extended Community Types 1160 TYPE VALUE NAME 1161 + 0x0a Transitive Transport Class Extended 1162 + Community (Sub-Types are defined in the 1163 + "Transitive Transport Class Extended 1164 + Community Sub-Types" registry) 1166 16.2.1.1.2. Non-Transitive Types 1168 This registry contains values of the high-order octet (the "Type" 1169 field) of a Non-transitive Extended Community. 1171 Registry Name: BGP Non-Transitive Extended Community Types 1173 TYPE VALUE NAME 1175 + 0x4a Non-Transitive Transport Class Extended 1176 + Community (Sub-Types are defined in the 1177 + "Non-Transitive Transport Class Extended 1178 + Community Sub-Types" registry) 1180 16.2.2. New registries to be created 1182 16.2.2.1. Transitive "Transport Class" Extended Community Sub-Types 1183 Registry 1185 This registry contains values of the second octet (the "Sub-Type" 1186 field) of an extended community when the value of the first octet 1187 (the "Type" field) is 0x07. 1189 Registry Name: Transitive Transport Class Extended 1190 Community Sub-Types 1192 RANGE REGISTRATION PROCEDURE 1194 0x00-0xBF First Come First Served 1195 0xC0-0xFF IETF Review 1197 SUB-TYPE VALUE NAME 1199 0x02 Route Target 1201 16.2.2.2. Non-Transitive "Transport Class" Extended Community Sub-Types 1202 Registry 1204 This registry contains values of the second octet (the "Sub-Type" 1205 field) of an extended community when the value of the first octet 1206 (the "Type" field) is 0x47. 1208 Registry Name: Non-Transitive Transport Class Extended 1209 Community Sub-Types 1211 RANGE REGISTRATION PROCEDURE 1213 0x00-0xBF First Come First Served 1214 0xC0-0xFF IETF Review 1216 SUB-TYPE VALUE NAME 1218 0x02 Route Target 1220 16.3. MPLS OAM code points 1222 The following two code points are sought for Target FEC Stack sub- 1223 TLVs: 1225 * IPv4 BGP Classful Transport 1227 * IPv6 BGP Classful Transport 1229 17. Security Considerations 1231 Mechanisms described in this document carry Transport routes in a new 1232 BGP address family. That minimizes possibility of these routes 1233 leaking outside the expected domain or mixing with service routes. 1235 When redistributing between SAFI 4 and SAFI 76 Classful Transport 1236 routes, there is a possibility of SAFI 4 routes mixing with SAFI 1 1237 service routes. To avoid such scenarios, it is RECOMMENDED that 1238 implementations support keeping SAFI 4 routes in a separate transport 1239 RIB, distinct from service RIB that contain SAFI 1 service routes. 1241 18. Contributors 1243 Rajesh M 1244 Juniper Networks, Inc. 1245 Electra, Exora Business Park~Marathahalli - Sarjapur Outer Ring Road, 1246 Bangalore 560103 1247 KA 1248 India 1249 Email: mrajesh@juniper.net 1251 19. Acknowledgements 1253 The authors thank Jeff Haas, John Scudder, Navaneetha Krishnan, Ravi 1254 M R, Chandrasekar Ramachandran, Shradha Hegde, Richard Roberts, 1255 Krzysztof Szarkowicz, John E Drake, Srihari Sangli, Vijay Kestur, 1256 Santosh Kolenchery, Robert Raszuk, Ahmed Darwish for the valuable 1257 discussions and review comments. 1259 The decision to not reuse SAFI 128 and create a new address-family to 1260 carry these transport-routes was based on suggestion made by Richard 1261 Roberts and Krzysztof Szarkowicz. 1263 20. Normative References 1265 [MPLS-NAMESPACES] 1266 Vairavakkalai, Ed., "BGP signalled MPLS-namespaces", 11 1267 June 2021, . 1270 [PCEP-RSVP-COLOR] 1271 Rajagopalan, Ed., "Path Computation Element Protocol(PCEP) 1272 Extension for RSVP Color", 15 January 2021, 1273 . 1276 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1277 Requirement Levels", BCP 14, RFC 2119, 1278 DOI 10.17487/RFC2119, March 1997, 1279 . 1281 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 1282 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 1283 DOI 10.17487/RFC4271, January 2006, 1284 . 1286 [RFC4360] Sangli, S., Tappan, D., and Y. Rekhter, "BGP Extended 1287 Communities Attribute", RFC 4360, DOI 10.17487/RFC4360, 1288 February 2006, . 1290 [RFC4364] Rosen, E. and Y. Rekhter, "BGP/MPLS IP Virtual Private 1291 Networks (VPNs)", RFC 4364, DOI 10.17487/RFC4364, February 1292 2006, . 1294 [RFC4456] Bates, T., Chen, E., and R. Chandra, "BGP Route 1295 Reflection: An Alternative to Full Mesh Internal BGP 1296 (IBGP)", RFC 4456, DOI 10.17487/RFC4456, April 2006, 1297 . 1299 [RFC4684] Marques, P., Bonica, R., Fang, L., Martini, L., Raszuk, 1300 R., Patel, K., and J. Guichard, "Constrained Route 1301 Distribution for Border Gateway Protocol/MultiProtocol 1302 Label Switching (BGP/MPLS) Internet Protocol (IP) Virtual 1303 Private Networks (VPNs)", RFC 4684, DOI 10.17487/RFC4684, 1304 November 2006, . 1306 [RFC4760] Bates, T., Chandra, R., Katz, D., and Y. Rekhter, 1307 "Multiprotocol Extensions for BGP-4", RFC 4760, 1308 DOI 10.17487/RFC4760, January 2007, 1309 . 1311 [RFC7153] Rosen, E. and Y. Rekhter, "IANA Registries for BGP 1312 Extended Communities", RFC 7153, DOI 10.17487/RFC7153, 1313 March 2014, . 1315 [RFC7911] Walton, D., Retana, A., Chen, E., and J. Scudder, 1316 "Advertisement of Multiple Paths in BGP", RFC 7911, 1317 DOI 10.17487/RFC7911, July 2016, 1318 . 1320 [RFC8029] Kompella, K., Swallow, G., Pignataro, C., Ed., Kumar, N., 1321 Aldrin, S., and M. Chen, "Detecting Multiprotocol Label 1322 Switched (MPLS) Data-Plane Failures", RFC 8029, 1323 DOI 10.17487/RFC8029, March 2017, 1324 . 1326 [RFC8212] Mauch, J., Snijders, J., and G. Hankins, "Default External 1327 BGP (EBGP) Route Propagation Behavior without Policies", 1328 RFC 8212, DOI 10.17487/RFC8212, July 2017, 1329 . 1331 [RFC8277] Rosen, E., "Using BGP to Bind MPLS Labels to Address 1332 Prefixes", RFC 8277, DOI 10.17487/RFC8277, October 2017, 1333 . 1335 [RFC8664] Sivabalan, S., Filsfils, C., Tantsura, J., Henderickx, W., 1336 and J. Hardwick, "Path Computation Element Communication 1337 Protocol (PCEP) Extensions for Segment Routing", RFC 8664, 1338 DOI 10.17487/RFC8664, December 2019, 1339 . 1341 [RFC8669] Previdi, S., Filsfils, C., Lindem, A., Ed., Sreekantiah, 1342 A., and H. Gredler, "Segment Routing Prefix Segment 1343 Identifier Extensions for BGP", RFC 8669, 1344 DOI 10.17487/RFC8669, December 2019, 1345 . 1347 [RTC-Ext] Zhang, Z., Ed., "Route Target Constrain Extension", 12 1348 July 2020, . 1351 [Seamless-SR] 1352 Hegde, Ed., "Seamless Segment Routing", 17 November 2020, 1353 . 1356 [SRTE] Previdi, S., Ed., "Advertising Segment Routing Policies in 1357 BGP", 18 November 2019, . 1360 [SRV6-INTER-DOMAIN] 1361 K A, Ed., "SRv6 inter-domain mapping SIDs", 10 January 1362 2021, . 1365 [SRV6-MPLS-AGRWL] 1366 Agrawal, Ed., "SRv6 and MPLS interworking", 22 February 1367 2021, . 1370 [SRV6-SERVICES] 1371 Dawra, Ed., "SRv6 BGP based Overlay Services", 11 April 1372 2021, . 1375 Authors' Addresses 1377 Kaliraj Vairavakkalai (editor) 1378 Juniper Networks, Inc. 1379 1133 Innovation Way, 1380 Sunnyvale, CA 94089 1381 United States of America 1383 Email: kaliraj@juniper.net 1385 Natrajan Venkataraman 1386 Juniper Networks, Inc. 1387 1133 Innovation Way, 1388 Sunnyvale, CA 94089 1389 United States of America 1391 Email: natv@juniper.net 1392 Balaji Rajagopalan 1393 Juniper Networks, Inc. 1394 Electra, Exora Business Park~Marathahalli - Sarjapur Outer Ring Road, 1395 Bangalore 560103 1396 KA 1397 India 1399 Email: balajir@juniper.net 1401 Gyan Mishra 1402 Verizon Communications Inc. 1403 13101 Columbia Pike 1404 Silver Spring, MD 20904 1405 United States of America 1407 Email: gyan.s.mishra@verizon.com 1409 Mazen Khaddam 1410 Cox Communications Inc. 1411 Atlanta, GA 1412 United States of America 1414 Email: mazen.khaddam@cox.com 1416 Xiaohu Xu 1417 Capitalonline. 1418 Beijing 1419 China 1421 Email: xiaohu.xu@capitalonline.net 1423 Rafal Jan Szarecki 1424 Google. 1425 1160 N Mathilda Ave, Bldg 5, 1426 Sunnyvale,, CA 94089 1427 United States of America 1429 Email: szarecki@google.com 1430 Deepak J Gowda 1431 Extreme Networks 1432 55 Commerce Valley Drive West, Suite 300, 1433 Thornhill, Toronto, Ontario L3T 7V9 1434 Canada 1436 Email: dgowda@extremenetworks.com