idnits 2.17.00 (12 Aug 2021) /tmp/idnits44278/draft-ietf-lsvr-bgp-spf-02.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 seems to contain a disclaimer for pre-RFC5378 work, but was first submitted on or after 10 November 2008. The disclaimer is usually necessary only for documents that revise or obsolete older RFCs, and that take significant amounts of text from those RFCs. If you can contact all authors of the source material and they are willing to grant the BCP78 rights to the IETF Trust, you can and should remove the disclaimer. Otherwise, the disclaimer is needed and you can ignore this comment. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (August 6, 2018) is 1384 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: 'RFC2328' is mentioned on line 762, but not defined == Missing Reference: 'RFC5286' is mentioned on line 796, but not defined == Missing Reference: 'RFC4456' is mentioned on line 766, but not defined == Missing Reference: 'RFC4915' is mentioned on line 791, but not defined == Missing Reference: 'RFC5549' is mentioned on line 801, but not defined ** Obsolete undefined reference: RFC 5549 (Obsoleted by RFC 8950) == Missing Reference: 'RFC4790' is mentioned on line 786, but not defined == Missing Reference: 'RFC5880' is mentioned on line 806, but not defined == Missing Reference: 'RFC4760' is mentioned on line 781, but not defined == Missing Reference: 'RFC4750' is mentioned on line 776, but not defined == Missing Reference: 'RFC4724' is mentioned on line 771, but not defined == Outdated reference: draft-ietf-idr-bgpls-segment-routing-epe has been published as RFC 9086 ** Downref: Normative reference to an Informational RFC: RFC 7938 Summary: 2 errors (**), 0 flaws (~~), 13 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group K. Patel 3 Internet-Draft Arrcus, Inc. 4 Intended status: Standards Track A. Lindem 5 Expires: February 7, 2019 Cisco Systems 6 S. Zandi 7 Linkedin 8 W. Henderickx 9 Nokia 10 August 6, 2018 12 Shortest Path Routing Extensions for BGP Protocol 13 draft-ietf-lsvr-bgp-spf-02.txt 15 Abstract 17 Many Massively Scaled Data Centers (MSDCs) have converged on 18 simplified layer 3 routing. Furthermore, requirements for 19 operational simplicity have lead many of these MSDCs to converge on 20 BGP as their single routing protocol for both their fabric routing 21 and their Data Center Interconnect (DCI) routing. This document 22 describes a solution which leverages BGP Link-State distribution and 23 the Shortest Path First (SPF) algorithm similar to Internal Gateway 24 Protocols (IGPs) such as OSPF. 26 Status of This Memo 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF). Note that other groups may also distribute 33 working documents as Internet-Drafts. The list of current Internet- 34 Drafts is at http://datatracker.ietf.org/drafts/current/. 36 Internet-Drafts are draft documents valid for a maximum of six months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on February 7, 2019. 43 Copyright Notice 45 Copyright (c) 2018 IETF Trust and the persons identified as the 46 document authors. All rights reserved. 48 This document is subject to BCP 78 and the IETF Trust's Legal 49 Provisions Relating to IETF Documents 50 (http://trustee.ietf.org/license-info) in effect on the date of 51 publication of this document. Please review these documents 52 carefully, as they describe your rights and restrictions with respect 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 This document may contain material from IETF Documents or IETF 59 Contributions published or made publicly available before November 60 10, 2008. The person(s) controlling the copyright in some of this 61 material may not have granted the IETF Trust the right to allow 62 modifications of such material outside the IETF Standards Process. 63 Without obtaining an adequate license from the person(s) controlling 64 the copyright in such materials, this document may not be modified 65 outside the IETF Standards Process, and derivative works of it may 66 not be created outside the IETF Standards Process, except to format 67 it for publication as an RFC or to translate it into languages other 68 than English. 70 Table of Contents 72 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 73 1.1. BGP Shortest Path First (SPF) Motivation . . . . . . . . 4 74 1.2. Requirements Language . . . . . . . . . . . . . . . . . . 5 75 2. BGP Peering Models . . . . . . . . . . . . . . . . . . . . . 5 76 2.1. BGP Single-Hop Peering on Network Node Connections . . . 5 77 2.2. BGP Peering Between Directly Connected Network Nodes . . 5 78 2.3. BGP Peering in Route-Reflector or Controller Topology . . 6 79 3. BGP-LS Shortest Path Routing (SPF) SAFI . . . . . . . . . . . 6 80 4. Extensions to BGP-LS . . . . . . . . . . . . . . . . . . . . 6 81 4.1. Node NLRI Usage and Modifications . . . . . . . . . . . . 7 82 4.2. Link NLRI Usage . . . . . . . . . . . . . . . . . . . . . 7 83 4.2.1. BGP-LS Link NLRI Attribute Prefix-Length TLVs . . . . 8 84 4.3. Prefix NLRI Usage . . . . . . . . . . . . . . . . . . . . 8 85 4.4. BGP-LS Attribute Sequence-Number TLV . . . . . . . . . . 8 86 5. Decision Process with SPF Algorithm . . . . . . . . . . . . . 9 87 5.1. Phase-1 BGP NLRI Selection . . . . . . . . . . . . . . . 10 88 5.2. Dual Stack Support . . . . . . . . . . . . . . . . . . . 11 89 5.3. SPF Calculation based on BGP-LS NLRI . . . . . . . . . . 11 90 5.4. NEXT_HOP Manipulation . . . . . . . . . . . . . . . . . . 13 91 5.5. IPv4/IPv6 Unicast Address Family Interaction . . . . . . 14 92 5.6. NLRI Advertisement and Convergence . . . . . . . . . . . 14 93 5.7. Error Handling . . . . . . . . . . . . . . . . . . . . . 14 94 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 15 95 7. Security Considerations . . . . . . . . . . . . . . . . . . . 15 96 7.1. Acknowledgements . . . . . . . . . . . . . . . . . . . . 15 97 7.2. Contributors . . . . . . . . . . . . . . . . . . . . . . 15 98 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 15 99 8.1. Normative References . . . . . . . . . . . . . . . . . . 16 100 8.2. Information References . . . . . . . . . . . . . . . . . 16 101 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 18 103 1. Introduction 105 Many Massively Scaled Data Centers (MSDCs) have converged on 106 simplified layer 3 routing. Furthermore, requirements for 107 operational simplicity have lead many of these MSDCs to converge on 108 BGP [RFC4271] as their single routing protocol for both their fabric 109 routing and their Data Center Interconnect (DCI) routing. 110 Requirements and procedures for using BGP are described in [RFC7938]. 111 This document describes an alternative solution which leverages BGP- 112 LS [RFC7752] and the Shortest Path First algorithm similar to 113 Internal Gateway Protocols (IGPs) such as OSPF [RFC2328]. 115 [RFC4271] defines the Decision Process that is used to select routes 116 for subsequent advertisement by applying the policies in the local 117 Policy Information Base (PIB) to the routes stored in its Adj-RIBs- 118 In. The output of the Decision Process is the set of routes that are 119 announced by a BGP speaker to its peers. These selected routes are 120 stored by a BGP speaker in the speaker's Adj-RIBs-Out according to 121 policy. 123 [RFC7752] describes a mechanism by which link-state and TE 124 information can be collected from networks and shared with external 125 components using BGP. This is achieved by defining NLRI advertised 126 within the BGP-LS/BGP-LS-SPF AFI/SAFI. The BGP-LS extensions defined 127 in [RFC7752] makes use of the Decision Process defined in [RFC4271]. 129 This document augments [RFC7752] by replacing its use of the existing 130 Decision Process. Rather than reusing the BGP-LS SAFI, the BGP-LS- 131 SPF SAFI is introduced to insure backward compatibility. The Phase 1 132 and 2 decision functions of the Decision Process are replaced with 133 the Shortest Path First (SPF) algorithm also known as the Dijkstra 134 algorithm. The Phase 3 decision function is also simplified since it 135 is no longer dependent on the previous phases. This solution avails 136 the benefits of both BGP and SPF-based IGPs. These include TCP based 137 flow-control, no periodic link-state refresh, and completely 138 incremental NLRI advertisement. These advantages can reduce the 139 overhead in MSDCs where there is a high degree of Equal Cost Multi- 140 Path (ECMPs) and the topology is very stable. Additionally, using a 141 SPF-based computation can support fast convergence and the 142 computation of Loop-Free Alternatives (LFAs) [RFC5286] in the event 143 of link failures. Furthermore, a BGP based solution lends itself to 144 multiple peering models including those incorporating route- 145 reflectors [RFC4456] or controllers. 147 Support for Multiple Topology Routing (MTR) as described in [RFC4915] 148 is an area for further study dependent on deployment requirements. 150 1.1. BGP Shortest Path First (SPF) Motivation 152 Given that [RFC7938] already describes how BGP could be used as the 153 sole routing protocol in an MSDC, one might question the motivation 154 for defining an alternate BGP deployment model when a mature solution 155 exists. For both alternatives, BGP offers the operational benefits 156 of a single routing protocol. However, BGP SPF offers some unique 157 advantages above and beyond standard BGP distance-vector routing. 159 A primary advantage is that all BGP speakers in the BGP SPF routing 160 domain will have a complete view of the topology. This will allow 161 support for ECMP, IP fast-reroute (e.g., Loop-Free Alternatives), 162 Shared Risk Link Groups (SRLGs), and other routing enhancements 163 without advertisement of addition BGP paths or other extensions. In 164 short, the advantages of an IGP such as OSPF [RFC2328] are availed in 165 BGP. 167 With the simplified BGP decision process as defined in Section 5.1, 168 NLRI changes can be disseminated throughout the BGP routing domain 169 much more rapidly (equivalent to IGPs with the proper 170 implementation). 172 Another primary advantage is a potential reduction in NLRI 173 advertisement. With standard BGP distance-vector routing, a single 174 link failure may impact 100s or 1000s prefixes and result in the 175 withdrawal or re-advertisement of the attendant NLRI. With BGP SPF, 176 only the BGP speakers corresponding to the link NLRI need withdraw 177 the corresponding BGP-LS Link NLRI. This advantage will contribute 178 to both faster convergence and better scaling. 180 With controller and route-reflector peering models, BGP SPF 181 advertisement and distributed computation require a minimal number of 182 sessions and copies of the NLRI since only the latest version of the 183 NLRI from the originator is required. Given that verification of the 184 adjacencies is done outside of BGP (see Section 2), each BGP speaker 185 will only need as many sessions and copies of the NLRI as required 186 for redundancy (e.g., one for the SPF computation and another for 187 backup). Functions such as Optimized Route Reflection (ORR) are 188 supported without extension by virtue of the primary advantages. 189 Additionally, a controller could inject topology that is learned 190 outside the BGP routing domain. 192 Given that controllers are already consuming BGP-LS NLRI [RFC7752], 193 reusing for the BGP-LS SPF leverages the existing controller 194 implementations. 196 Another potential advantage of BGP SPF is that both IPv6 and IPv4 can 197 be supported in the same address family using the same topology. 198 Although not described in this version of the document, multi- 199 topology extensions can be used to support separate IPv4, IPv6, 200 unicast, and multicast topologies while sharing the same NLRI. 202 Finally, the BGP SPF topology can be used as an underlay for other 203 BGP address families (using the existing model) and realize all the 204 above advantages. A simplified peering model using IPv6 link-local 205 addresses as next-hops can be deployed similar to [RFC5549]. 207 1.2. Requirements Language 209 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 210 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 211 "OPTIONAL" in this document are to be interpreted as described in BCP 212 14 [RFC2119] [RFC8174] when, and only when, they appear in all 213 capitals, as shown here. 215 2. BGP Peering Models 217 Depending on the requirements, scaling, and capabilities of the BGP 218 speakers, various peering models are supported. The only requirement 219 is that all BGP speakers in the BGP SPF routing domain receive link- 220 state NLRI on a timely basis, run an SPF calculation, and update 221 their data plane appropriately. The content of the Link NLRI is 222 described in Section 4.2. 224 2.1. BGP Single-Hop Peering on Network Node Connections 226 The simplest peering model is the one described in section 5.2.1 of 227 [RFC7938]. In this model, EBGP single-hop sessions are established 228 over direct point-to-point links interconnecting the SPF domain 229 nodes. For the purposes of BGP SPF, Link NLRI is only advertised if 230 a single-hop BGP session has been established and the Link-State/SPF 231 address family capability has been exchanged [RFC4790] on the 232 corresponding session. If the session goes down, the corresponding 233 Link NLRI will be withdrawn. 235 2.2. BGP Peering Between Directly Connected Network Nodes 237 In this model, BGP speakers peer with all directly connected network 238 nodes but the sessions may be multi-hop and the direct connection 239 discovery and liveliness detection for those connections are 240 independent of the BGP protocol. How this is accomplished is outside 241 the scope of this document. Consequently, there will be a single 242 session even if there are multiple direct connections between BGP 243 speakers. For the purposes of BGP SPF, Link NLRI is advertised as 244 long as a BGP session has been established, the Link-State/SPF 245 address family capability has been exchanged [RFC4790] and the 246 corresponding link is considered is up and considered operational. 248 2.3. BGP Peering in Route-Reflector or Controller Topology 250 In this model, BGP speakers peer solely with one or more Route 251 Reflectors [RFC4456] or controllers. As in the previous model, 252 direct connection discovery and liveliness detection for those 253 connections are done outside the BGP protocol. More specifically, 254 the Liveliness detection is done using BFD protocol described in 255 [RFC5880]. For the purposes of BGP SPF, Link NLRI is advertised as 256 long as the corresponding link is up and considered operational. 258 3. BGP-LS Shortest Path Routing (SPF) SAFI 260 In order to replace the Phase 1 and 2 decision functions of the 261 existing Decision Process with an SPF-based Decision Process and 262 streamline the Phase 3 decision functions in a backward compatible 263 manner, this draft introduces the BGP-LS-SFP SAFI for BGP-LS SPF 264 operation. The BGP-LS-SPF (AF 16388 / SAFI TBD1) [RFC4790] is 265 allocated by IANA as specified in the Section 6. A BGP speaker using 266 the BGP-LS SPF extensions described herein MUST exchange the AFI/SAFI 267 using Multiprotocol Extensions Capability Code [RFC4760] with other 268 BGP speakers in the SPF routing domain. 270 4. Extensions to BGP-LS 272 [RFC7752] describes a mechanism by which link-state and TE 273 information can be collected from networks and shared with external 274 components using BGP protocol. It describes both the definition of 275 BGP-LS NLRI that describes links, nodes, and prefixes comprising IGP 276 link-state information and the definition of a BGP path attribute 277 (BGP-LS attribute) that carries link, node, and prefix properties and 278 attributes, such as the link and prefix metric or auxiliary Router- 279 IDs of nodes, etc. 281 The BGP protocol will be used in the Protocol-ID field specified in 282 table 1 of [I-D.ietf-idr-bgpls-segment-routing-epe]. The local and 283 remote node descriptors for all NLRI will be the BGP Router-ID (TLV 284 516) and either the AS Number (TLV 512) [RFC7752] or the BGP 285 Confederation Member (TLV 517) [RFC8402]. However, if the BGP 286 Router-ID is known to be unique within the BGP Routing domain, it can 287 be used as the sole descriptor. 289 4.1. Node NLRI Usage and Modifications 291 The SPF capability is a new Node Attribute TLV that will be added to 292 those defined in table 7 of [RFC7752]. The new attribute TLV will 293 only be applicable when BGP is specified in the Node NLRI Protocol ID 294 field. The TBD TLV type will be defined by IANA. The new Node 295 Attribute TLV will contain a single-octet SPF algorithm as defined in 296 [RFC8402]. 298 0 1 2 3 299 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 1 300 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 301 | Type | Length | 302 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 303 | SPF Algorithm | 304 +-+-+-+-+-+-+-+-+ 306 The SPF Algorithm may take the following values: 308 0 - Normal Shortest Path First (SPF) algorithm based on link 309 metric. This is the standard shortest path algorithm as 310 computed by the IGP protocol. Consistent with the deployed 311 practice for link-state protocols, Algorithm 0 permits any 312 node to overwrite the SPF path with a different path based on 313 its local policy. 314 1 - Strict Shortest Path First (SPF) algorithm based on link 315 metric. The algorithm is identical to Algorithm 0 but Algorithm 316 1 requires that all nodes along the path will honor the SPF 317 routing decision. Local policy at the node claiming support for 318 Algorithm 1 MUST NOT alter the SPF paths computed by Algorithm 1. 320 Note that usage of Strict Shortest Path First (SPF) algorithm is 321 defined in the IGP algorithm registry but usage is restricted to 322 [I-D.ietf-idr-bgpls-segment-routing-epe]. Hence, its usage for BGP- 323 LS SPF is out of scope. 325 When computing the SPF for a given BGP routing domain, only BGP nodes 326 advertising the SPF capability attribute will be included the 327 Shortest Path Tree (SPT). 329 4.2. Link NLRI Usage 331 The criteria for advertisement of Link NLRI are discussed in 332 Section 2. 334 Link NLRI is advertised with local and remote node descriptors as 335 described above and unique link identifiers dependent on the 336 addressing. For IPv4 links, the links local IPv4 (TLV 259) and 337 remote IPv4 (TLV 260) addresses will be used. For IPv6 links, the 338 local IPv6 (TLV 261) and remote IPv6 (TLV 262) addresses will be 339 used. For unnumbered links, the link local/remote identifiers (TLV 340 258) will be used. For links supporting having both IPv4 and IPv6 341 addresses, both sets of descriptors may be included in the same Link 342 NLRI. The link identifiers are described in table 5 of [RFC7752]. 344 The link IGP metric attribute TLV (TLV 1095) as well as any others 345 required for non-SPF purposes SHOULD be advertised. Algorithms such 346 as setting the metric inversely to the link speed as done in the OSPF 347 MIB [RFC4750] MAY be supported. However, this is beyond the scope of 348 this document. 350 4.2.1. BGP-LS Link NLRI Attribute Prefix-Length TLVs 352 Two BGP-LS Attribute TLVs to BGP-LS Link NLRI are defined to 353 advertise the prefix length associated with the IPv4 and IPv6 link 354 prefixes. The prefix length is used for the optional installation of 355 prefixes corresponding to Link NLRI as defined in Section 5.3. 357 0 1 2 3 358 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 1 359 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 360 | TBD IPv4 or IPv6 Type | Length | 361 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 362 | Prefix-Length | 363 +-+-+-+-+-+-+-+-+ 365 Prefix-length - A one-octet length restricted to 1-32 for IPv4 366 Link NLIR endpoint prefixes and 1-128 for IPv6 367 Link NLRI endpoint prefixes. 369 4.3. Prefix NLRI Usage 371 Prefix NLRI is advertised with a local node descriptor as described 372 above and the prefix and length used as the descriptors (TLV 265) as 373 described in [RFC7752]. The prefix metric attribute TLV (TLV 1155) 374 as well as any others required for non-SPF purposes SHOULD be 375 advertised. For loopback prefixes, the metric should be 0. For non- 376 loopback prefixes, the setting of the metric is a local matter and 377 beyond the scope of this document. 379 4.4. BGP-LS Attribute Sequence-Number TLV 381 A new BGP-LS Attribute TLV to BGP-LS NLRI types is defined to assure 382 the most recent version of a given NLRI is used in the SPF 383 computation. The TBD TLV type will be defined by IANA. The new BGP- 384 LS Attribute TLV will contain an 8-octet sequence number. The usage 385 of the Sequence Number TLV is described in Section 5.1. 387 0 1 2 3 388 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 1 389 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 390 | Type | Length | 391 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 392 | Sequence Number (High-Order 32 Bits) | 393 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 394 | Sequence Number (Low-Order 32 Bits) | 395 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 397 Sequence Number 399 The 64-bit strictly increasing sequence number is incremented for 400 every version of BGP-LS NLRI originated. BGP speakers implementing 401 this specification MUST use available mechanisms to preserve the 402 sequence number's strictly increasing property for the deployed life 403 of the BGP speaker (including cold restarts). One mechanism for 404 accomplishing this would be to use the high-order 32 bits of the 405 sequence number as a wrap/boot count that is incremented anytime the 406 BGP router loses its sequence number state or the low-order 32 bits 407 wrap. 409 When incrementing the sequence number for each self-originated NLRI, 410 the sequence number should be treated as an unsigned 64-bit value. 411 If the lower-order 32-bit value wraps, the higher-order 32-bit value 412 should be incremented and saved in non-volatile storage. If by some 413 chance the BGP Speaker is deployed long enough that there is a 414 possibility that the 64-bit sequence number may wrap or a BGP Speaker 415 completely loses its sequence number state (e.g., the BGP speaker 416 hardware is replaced or experiences a cold-start), the phase 1 417 decision function (see Section 5.1) rules will insure convergence, 418 albeit, not immediately. 420 5. Decision Process with SPF Algorithm 422 The Decision Process described in [RFC4271] takes place in three 423 distinct phases. The Phase 1 decision function of the Decision 424 Process is responsible for calculating the degree of preference for 425 each route received from a BGP speaker's peer. The Phase 2 decision 426 function is invoked on completion of the Phase 1 decision function 427 and is responsible for choosing the best route out of all those 428 available for each distinct destination, and for installing each 429 chosen route into the Loc-RIB. The combination of the Phase 1 and 2 430 decision functions is characterized as a Path Vector algorithm. 432 The SPF based Decision process replaces the BGP best-path Decision 433 process described in [RFC4271]. This process starts with selecting 434 only those Node NLRI whose SPF capability TLV matches with the local 435 BGP speaker's SPF capability TLV value. Since Link-State NLRI always 436 contains the local descriptor [RFC7752], it will only be originated 437 by a single BGP speaker in the BGP routing domain. These selected 438 Node NLRI and their Link/Prefix NLRI are used to build a directed 439 graph during the SPF computation. The best paths for BGP prefixes 440 are installed as a result of the SPF process. 442 When BGP-LS-SPF NLRI is received, all that is required is to 443 determine whether it is the best-path by examining the Node-ID and 444 sequence number as described in Section 5.1. If the received best- 445 path NLRI had changed, it will be advertised to other BGP-LS-SPF 446 peers. If the attributes have changed (other than the sequence 447 number), a BGP SPF calculation will be scheduled. However, a changed 448 NLRI MAY be advertised to other peers almost immediately and 449 propagation of changes can approach IGP convergence times. To 450 accomplish this, the MinRouteAdvertisementIntervalTimer and 451 MinRouteAdvertisementIntervalTimer [RFC4271] are not applicable to 452 the BGP-LS-SPF SAFI. 454 The Phase 3 decision function of the Decision Process [RFC4271] is 455 also simplified since under normal SPF operation, a BGP speaker would 456 advertise the NLRI selected for the SPF to all BGP peers with the 457 BGP-LS/BGP-LS-SPF AFI/SAFI. Application of policy would not be 458 prevented however its usage to best-path process would be limited as 459 the SPF relies solely on link metrics. 461 5.1. Phase-1 BGP NLRI Selection 463 The rules for NLRI selection are greatly simplified from [RFC4271]. 465 1. If the NLRI is received from the BGP speaker originating the NLRI 466 (as determined by the comparing BGP Router ID in the NLRI Node 467 identifiers with the BGP speaker Router ID), then it is preferred 468 over the same NLRI from non-originators. This rule will assure 469 that stale NLRI is updated even if a BGP-LS router loses its 470 sequence number state due to a cold-start. 472 2. If the Sequence-Number TLV is present in the BGP-LS Attribute, 473 then the NLRI with the most recent, i.e., highest sequence number 474 is selected. BGP-LS NLRI with a Sequence-Number TLV will be 475 considered more recent than NLRI without a BGP-LS Attribute or a 476 BGP-LS Attribute that doesn't include the Sequence-Number TLV. 478 3. The final tie-breaker is the NLRI from the BGP Speaker with the 479 numerically largest BGP Router ID. 481 The modified SPF Decision Process performs an SPF calculation rooted 482 at the BGP speaker using the metrics from Link and Prefix NLRI 483 Attribute TLVs [RFC7752]. As a result, any attributes that would 484 influence the Decision process defined in [RFC4271] like ORIGIN, 485 MULTI_EXIT_DISC, and LOCAL_PREF attributes are ignored by the SPF 486 algorithm. Furthermore, the NEXT_HOP attribute value is preserved 487 but otherwise ignored during the SPF or best-path. 489 5.2. Dual Stack Support 491 The SPF-based decision process operates on Node, Link, and Prefix 492 NLRIs that support both IPv4 and IPv6 addresses. Whether to run a 493 single SPF instance or multiple SPF instances for separate AFs is a 494 matter of a local implementation. Normally, IPv4 next-hops are 495 calculated for IPv4 prefixes and IPv6 next-hops are calculated for 496 IPv6 prefixes. However, an interesting use-case is deployment of 497 [RFC5549] where IPv6 next-hops are calculated for both IPv4 and IPv6 498 prefixes. As stated in Section 1, support for Multiple Topology 499 Routing (MTR) is an area for future study. 501 5.3. SPF Calculation based on BGP-LS NLRI 503 This section details the BGP-LS SPF local routing information base 504 (RIB) calculation. The router will use BGP-LS Node, Link, and Prefix 505 NLRI to populate the local RIB using the following algorithm. This 506 calculation yields the set of intra-area routes associated with the 507 BGP-LS domain. A router calculates the shortest-path tree using 508 itself as the root. Variations and optimizations of the algorithm 509 are valid as long as it yields the same set of routes. The algorithm 510 below supports Equal Cost Multi-Path (ECMP) routes. Weighted Unequal 511 Cost Multi-Path are out of scope. The organization of this section 512 owes heavily to section 16 of [RFC2328]. 514 The following abstract data structures are defined in order to 515 specify the algorithm. 517 o Local Route Information Base (RIB) - This is abstract contains 518 reachability information (i.e., next hops) for all prefixes (both 519 IPv4 and IPv6) as well as the Node NLRI reachability. 520 Implementations may choose to implement this as separate RIBs for 521 each address family and/or Node NLRI. 523 o Link State NLRI Database (LSNDB) - Database of BGP-LS NLRI that 524 facilitates access to all Node, Link, and Prefix NLRI as well as 525 all the Link and Prefix NLRI corresponding to a given Node NLRI. 526 Other optimization, such as, resolving bi-directional connectivity 527 associations between Link NLRI are possible but of scope of this 528 document. 530 o Candidate List - This is a list of candidate Node NLRI with the 531 lowest cost Node NLRI at the front of the list. It is typically 532 implemented as a heap but other concrete data structures have also 533 been used. 535 The algorithm is comprised of the steps below: 537 1. The current local RIB is invalidated. The local RIB is built 538 again from scratch. The existing routing entries are preserved 539 for comparision to determine changes that need to be installed in 540 the global RIB. 542 2. The computing router's Node NLRI is installed in the local RIB 543 with a cost of 0 and as as the sole entry in the candidate list. 545 3. The Node NLRI with the lowest cost is removed from the candidate 546 list for processing. The Node corresponding to this NLRI will be 547 referred to as the Current Node. If the candidate list is empty, 548 the SPF calculation has completed and the algorithm proceeds to 549 step 6. 551 4. All the Prefix NLRI with the same Node Identifiers as the Current 552 Node will be considered for installation. The cost for each 553 prefix is the metric advertised in the Prefix NLRI added to the 554 cost to reach the Current Node. 556 * If the prefix is not in the local RIB, the prefix is installed 557 and will inherit the Current Node's next hops. 559 * If the prefix is in the local RIB and the cost is greater than 560 the Current route's metric, the Prefix NLRI does not 561 contribute to the route and is ignored. 563 * If the prefix is in the local RIB and the cost is less than 564 the current route's metric, the Prefix is installed with the 565 Current Node's next-hops replacing the local RIB route's next- 566 hops and the metric being updated. 568 * If the prefix is in the local RIB and the cost is same as the 569 current route's metric, the Prefix is installed with the 570 Current Node's next-hops being merged with local RIB route's 571 next-hops. 573 5. All the Link NLRI with the same Node Identifiers as the Current 574 Node will be considered for installation. Each link will be 575 examined and will be referred to in the following text as the 576 Current Link. The cost of the Current Link is the advertised 577 metric in the Link NLRI added to the cost to reach the Current 578 Node. 580 * Optionally, the prefix(es) associated with the Current Link 581 are installed into the local RIB using the same rules as were 582 used for Prefix NLRI in the previous steps. 584 * The Current Link's endpoint Node NLRI is accessed (i.e., the 585 Node NLRI with the same Node identifiers as the Link 586 endpoint). If it exists, it will be referred to as the 587 Endpoint Node NLRI and the algorithm will proceed as follows: 589 + All the Link NLRI corresponding the Endpoint Node NLRI will 590 be searched for a back-link NLRI pointing to the current 591 node. Both the Node identifiers and the Link endpoint 592 identifiers in the Endpoint Node's Link NLRI must match for 593 a match. If there is no corresponding Link NLRI 594 corresponding to the Endpoint Node NLRI, the Endpoint Node 595 NLIR fails the bi-directional connectivity test and is not 596 processed further. 598 + If the Endpoint Node NLRI is not on the candidate list, it 599 is inserted based on the link cost and BGP Identifier (the 600 latter being used as a tie-breaker). 602 + If the Endpoint Node NLRI is already on the candidate list 603 with a lower cost, it need not be inserted again. 605 + If the Endpoint Node NLRI is already on the candidate list 606 with a higher cost, it must be removed and reinserted with 607 a lower cost. 609 * Return to step 3 to process the next lowest cost Node NLRI on 610 the candidate list. 612 6. The local RIB is examined and changes (adds, deletes, 613 modifications) are installed into the global RIB. 615 5.4. NEXT_HOP Manipulation 617 A BGP speaker that supports SPF extensions MAY interact with peers 618 that don't support SPF extensions. If the BGP-LS address family is 619 advertised to a peer not supporting the SPF extensions described 620 herein, then the BGP speaker MUST conform to the NEXT_HOP rules 621 specified in [RFC4271] when announcing the Link-State address family 622 routes to those peers. 624 All BGP peers that support SPF extensions would locally compute the 625 Loc-RIB next-hops as a result of the SPF process. Consequently, the 626 NEXT_HOP attribute is always ignored on receipt. However, BGP 627 speakers SHOULD set the NEXT_HOP address according to the NEXT_HOP 628 attribute rules specified in [RFC4271]. 630 5.5. IPv4/IPv6 Unicast Address Family Interaction 632 While the BGP-LS SPF address family and the IPv4/IPv6 unicast address 633 families install routes into the same device routing tables, they 634 will operate independently much the same as OSPF and IS-IS would 635 operate today (i.e., "Ships-in-the-Night" mode). There will be no 636 implicit route redistribution between the BGP address families. 637 However, implementation specific redistribution mechanisms SHOULD be 638 made available with the restriction that redistribution of BGP-LS SPF 639 routes into the IPv4 address family applies only to IPv4 routes and 640 redistribution of BGP-LS SPF route into the IPv6 address family 641 applies only to IPv6 routes. 643 Given the fact that SPF algorithms are based on the assumption that 644 all routers in the routing domain calculate the precisely the same 645 SPF tree and install the same set of routes, it is RECOMMENDED that 646 BGP-LS SPF IPv4/IPv6 routes be given priority by default when 647 installed into their respective RIBs. In common implementations the 648 prioritization is governed by route preference or administrative 649 distance with lower being more preferred. 651 5.6. NLRI Advertisement and Convergence 653 A local failure will prevent a link from being used in the SPF 654 calculation due to the IGP bi-directional connectivity requirement. 655 Consequently, local link failures should always be given priority 656 over updates (e.g., withdrawing all routes learned on a session) in 657 order to ensure the highest priority propagation and optimal 658 convergence. 660 Delaying the withdrawal of non-local routes is an area for further 661 study as more IGP-like mechanisms would be required to prevent usage 662 of stale NLRI. 664 5.7. Error Handling 666 When a BGP speaker receives a BGP Update containing a malformed SPF 667 Capability TLV in the Node NLRI BGP-LS Attribute [RFC7752], it MUST 668 ignore the received TLV and the Node NLRI and not pass it to other 669 BGP peers as specified in [RFC7606]. When discarding a Node NLRI 670 with malformed TLV, a BGP speaker SHOULD log an error for further 671 analysis. 673 6. IANA Considerations 675 This document defines an AFI/SAFI for BGP-LS SPF operation and 676 requests IANA to assign the BGP-LS/BGP-LS-SPF (AFI 16388 / SAFI TBD1) 677 as described in [RFC4750]. 679 This document also defines four attribute TLVs for BGP LS NLRI. We 680 request IANA to assign TLVs for the SPF capability, Sequence Number, 681 IPv4 Link Prefix-Length, and IPv6 Link Prefix-Length from the "BGP-LS 682 Node Descriptor, Link Descriptor, Prefix Descriptor, and Attribute 683 TLVs" Registry. 685 7. Security Considerations 687 This extension to BGP does not change the underlying security issues 688 inherent in the existing [RFC4724] and [RFC4271]. 690 7.1. Acknowledgements 692 The authors would like to thank Sue Hares, Jorge Rabadan, Boris 693 Hassanov, and Fred Baker for their review and comments. 695 7.2. Contributors 697 In addition to the authors listed on the front page, the following 698 co-authors have contributed to the document. 700 Derek Yeung 701 Arrcus, Inc. 702 derek@arrcus.com 704 Gunter Van De Velde 705 Nokia 706 gunter.van_de_velde@nokia.com 708 Abhay Roy 709 Cisco Systems 710 akr@cisco.com 712 Venu Venugopal 713 Cisco Systems 714 venuv@cisco.com 716 8. References 717 8.1. Normative References 719 [I-D.ietf-idr-bgpls-segment-routing-epe] 720 Previdi, S., Filsfils, C., Patel, K., Ray, S., and J. 721 Dong, "BGP-LS extensions for Segment Routing BGP Egress 722 Peer Engineering", draft-ietf-idr-bgpls-segment-routing- 723 epe-14 (work in progress), December 2017. 725 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 726 Requirement Levels", BCP 14, RFC 2119, 727 DOI 10.17487/RFC2119, March 1997, . 730 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 731 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 732 DOI 10.17487/RFC4271, January 2006, . 735 [RFC7606] Chen, E., Ed., Scudder, J., Ed., Mohapatra, P., and K. 736 Patel, "Revised Error Handling for BGP UPDATE Messages", 737 RFC 7606, DOI 10.17487/RFC7606, August 2015, 738 . 740 [RFC7752] Gredler, H., Ed., Medved, J., Previdi, S., Farrel, A., and 741 S. Ray, "North-Bound Distribution of Link-State and 742 Traffic Engineering (TE) Information Using BGP", RFC 7752, 743 DOI 10.17487/RFC7752, March 2016, . 746 [RFC7938] Lapukhov, P., Premji, A., and J. Mitchell, Ed., "Use of 747 BGP for Routing in Large-Scale Data Centers", RFC 7938, 748 DOI 10.17487/RFC7938, August 2016, . 751 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 752 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 753 May 2017, . 755 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 756 Decraene, B., Litkowski, S., and R. Shakir, "Segment 757 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 758 July 2018, . 760 8.2. Information References 762 [RFC2328] Moy, J., "OSPF Version 2", STD 54, RFC 2328, 763 DOI 10.17487/RFC2328, April 1998, . 766 [RFC4456] Bates, T., Chen, E., and R. Chandra, "BGP Route 767 Reflection: An Alternative to Full Mesh Internal BGP 768 (IBGP)", RFC 4456, DOI 10.17487/RFC4456, April 2006, 769 . 771 [RFC4724] Sangli, S., Chen, E., Fernando, R., Scudder, J., and Y. 772 Rekhter, "Graceful Restart Mechanism for BGP", RFC 4724, 773 DOI 10.17487/RFC4724, January 2007, . 776 [RFC4750] Joyal, D., Ed., Galecki, P., Ed., Giacalone, S., Ed., 777 Coltun, R., and F. Baker, "OSPF Version 2 Management 778 Information Base", RFC 4750, DOI 10.17487/RFC4750, 779 December 2006, . 781 [RFC4760] Bates, T., Chandra, R., Katz, D., and Y. Rekhter, 782 "Multiprotocol Extensions for BGP-4", RFC 4760, 783 DOI 10.17487/RFC4760, January 2007, . 786 [RFC4790] Newman, C., Duerst, M., and A. Gulbrandsen, "Internet 787 Application Protocol Collation Registry", RFC 4790, 788 DOI 10.17487/RFC4790, March 2007, . 791 [RFC4915] Psenak, P., Mirtorabi, S., Roy, A., Nguyen, L., and P. 792 Pillay-Esnault, "Multi-Topology (MT) Routing in OSPF", 793 RFC 4915, DOI 10.17487/RFC4915, June 2007, 794 . 796 [RFC5286] Atlas, A., Ed. and A. Zinin, Ed., "Basic Specification for 797 IP Fast Reroute: Loop-Free Alternates", RFC 5286, 798 DOI 10.17487/RFC5286, September 2008, . 801 [RFC5549] Le Faucheur, F. and E. Rosen, "Advertising IPv4 Network 802 Layer Reachability Information with an IPv6 Next Hop", 803 RFC 5549, DOI 10.17487/RFC5549, May 2009, 804 . 806 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 807 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 808 . 810 Authors' Addresses 812 Keyur Patel 813 Arrcus, Inc. 815 Email: keyur@arrcus.com 817 Acee Lindem 818 Cisco Systems 819 301 Midenhall Way 820 Cary, NC 27513 821 USA 823 Email: acee@cisco.com 825 Shawn Zandi 826 Linkedin 827 222 2nd Street 828 San Francisco, CA 94105 829 USA 831 Email: szandi@linkedin.com 833 Wim Henderickx 834 Nokia 835 Antwerp 836 Belgium 838 Email: wim.henderickx@nokia.com