idnits 2.17.1 draft-litkowski-pce-state-sync-03.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (April 26, 2018) is 2186 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(See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Outdated reference: A later version (-15) exists of draft-ietf-pce-association-diversity-03 == Outdated reference: A later version (-15) exists of draft-ietf-pce-stateful-hpce-04 Summary: 0 errors (**), 0 flaws (~~), 3 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 PCE Working Group S. Litkowski 3 Internet-Draft Orange 4 Intended status: Standards Track S. Sivabalan 5 Expires: October 28, 2018 Cisco 6 D. Dhody 7 Huawei 8 April 26, 2018 10 Inter Stateful Path Computation Element communication procedures 11 draft-litkowski-pce-state-sync-03 13 Abstract 15 The Path Computation Element Communication Protocol (PCEP) provides 16 mechanisms for Path Computation Elements (PCEs) to perform path 17 computations in response to Path Computation Clients (PCCs) requests. 18 The stateful PCE extensions allow stateful control of Multi-Protocol 19 Label Switching (MPLS) Traffic Engineering Label Switched Paths (TE 20 LSPs) using PCEP. 22 A Path Computation Client (PCC) can synchronize an LSP state 23 information to a Stateful Path Computation Element (PCE). The 24 stateful PCE extension allows a redundancy scenario where a PCC can 25 have redundant PCEP sessions towards multiple PCEs. In such a case, 26 a PCC gives control on a LSP to only a single PCE, and only one PCE 27 is responsible for path computation for this delegated LSP. The 28 document does not state the procedures related to an inter-PCE 29 stateful communication. 31 There are some use cases, where an inter-PCE stateful communication 32 can bring additional resiliency in the design for instance when some 33 PCC-PCE sessions fails. The inter-PCE stateful communication may 34 also provide a faster update of the LSP states when an event occurs. 35 Finally, when, in a redundant PCE scenario, there is a need to 36 compute a set of paths that are part of a group (so there is a 37 dependency between the paths), there may be some cases where the 38 computation of all paths in the group is not handled by the same PCE: 39 this situation is called a split-brain. This split-brain scenario 40 may lead to computation loops between PCEs or suboptimal paths 41 computation. 43 This document describes the procedures to allow a stateful 44 communication between PCEs for various use-cases and also the 45 procedures to prevent computations loops. 47 Requirements Language 49 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 50 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 51 document are to be interpreted as described in [RFC2119]. 53 Status of This Memo 55 This Internet-Draft is submitted in full conformance with the 56 provisions of BCP 78 and BCP 79. 58 Internet-Drafts are working documents of the Internet Engineering 59 Task Force (IETF). Note that other groups may also distribute 60 working documents as Internet-Drafts. The list of current Internet- 61 Drafts is at https://datatracker.ietf.org/drafts/current/. 63 Internet-Drafts are draft documents valid for a maximum of six months 64 and may be updated, replaced, or obsoleted by other documents at any 65 time. It is inappropriate to use Internet-Drafts as reference 66 material or to cite them other than as "work in progress." 68 This Internet-Draft will expire on October 28, 2018. 70 Copyright Notice 72 Copyright (c) 2018 IETF Trust and the persons identified as the 73 document authors. All rights reserved. 75 This document is subject to BCP 78 and the IETF Trust's Legal 76 Provisions Relating to IETF Documents 77 (https://trustee.ietf.org/license-info) in effect on the date of 78 publication of this document. Please review these documents 79 carefully, as they describe your rights and restrictions with respect 80 to this document. Code Components extracted from this document must 81 include Simplified BSD License text as described in Section 4.e of 82 the Trust Legal Provisions and are provided without warranty as 83 described in the Simplified BSD License. 85 Table of Contents 87 1. Introduction and problem statement . . . . . . . . . . . . . 3 88 1.1. Reporting LSP changes . . . . . . . . . . . . . . . . . . 3 89 1.2. Split-brain . . . . . . . . . . . . . . . . . . . . . . . 4 90 1.3. Applicability to H-PCE . . . . . . . . . . . . . . . . . 11 91 2. Proposed solution . . . . . . . . . . . . . . . . . . . . . . 11 92 2.1. State-sync session . . . . . . . . . . . . . . . . . . . 11 93 2.2. Master/Slave relationship between PCE . . . . . . . . . . 13 94 3. Procedures and protocol extensions . . . . . . . . . . . . . 13 95 3.1. Opening a state-sync session . . . . . . . . . . . . . . 13 96 3.1.1. Capability advertisement . . . . . . . . . . . . . . 13 97 3.2. State synchronization . . . . . . . . . . . . . . . . . . 14 98 3.3. Incremental updates and report forwarding rules . . . . . 15 99 3.4. Maintaining LSP states from different sources . . . . . . 16 100 3.5. Computation priority between PCEs and sub-delegation . . 17 101 3.6. Passive stateful procedures . . . . . . . . . . . . . . . 18 102 3.7. PCE initiation procedures . . . . . . . . . . . . . . . . 19 103 4. Examples . . . . . . . . . . . . . . . . . . . . . . . . . . 19 104 4.1. Example 1 . . . . . . . . . . . . . . . . . . . . . . . . 19 105 4.2. Example 2 . . . . . . . . . . . . . . . . . . . . . . . . 20 106 4.3. Example 3 . . . . . . . . . . . . . . . . . . . . . . . . 22 107 5. Using Master/Slave computation and state-sync sessions to 108 increase scaling . . . . . . . . . . . . . . . . . . . . . . 23 109 6. PCEP-PATH-VECTOR-TLV . . . . . . . . . . . . . . . . . . . . 25 110 7. Security Considerations . . . . . . . . . . . . . . . . . . . 26 111 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 26 112 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 26 113 9.1. PCEP-Error Object . . . . . . . . . . . . . . . . . . . . 26 114 9.2. PCEP TLV Type Indicators . . . . . . . . . . . . . . . . 26 115 9.3. STATEFUL-PCE-CAPABILITY TLV . . . . . . . . . . . . . . . 27 116 10. References . . . . . . . . . . . . . . . . . . . . . . . . . 27 117 10.1. Normative References . . . . . . . . . . . . . . . . . . 27 118 10.2. Informative References . . . . . . . . . . . . . . . . . 27 119 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 28 121 1. Introduction and problem statement 123 1.1. Reporting LSP changes 125 When using a stateful PCE ([RFC8231]), a Path Computation Client 126 (PCC) can synchronize an LSP state information to the stateful Path 127 Computation Element (PCE). If the PCC grants the control on the LSP 128 to the PCE, the PCE can update the LSP parameters at any time. 130 In a multi PCE deployment (redundancy, loadbalancing...), with the 131 current specification defined in [RFC8231], the PCC will be in charge 132 of reporting the other PCEs of the LSP parameter change which brings 133 additional hops and delays in notifying the overall network of the 134 LSP parameter change. 136 This delay may affect the reaction time of the other PCEs, if they 137 need to take action after being notified of the LSP parameter change. 139 Apart from the synchronization from the PCC, it is also useful if 140 there is synchronization mechanism between the stateful PCEs. As 141 stateful PCE make changes to its delegated LSPs, these changes 142 (pending LSPs and the sticky resources [RFC7399]) can be synchronized 143 immediately to the other PCEs. 145 +----------+ 146 | PCC1 | LSP1 147 +----------+ 148 / \ 149 / \ 150 +---------+ +---------+ 151 | PCE1 | | PCE2 | 152 +---------+ +---------+ 153 \ / 154 \ / 155 +----------+ 156 | PCC2 | LSP2 157 +----------+ 159 In the figure above, we consider a loadbalanced PCE architecture, so 160 PCE1 is responsible to compute paths for PCC1 and PCE2 is responsible 161 to compute paths for PCC2. When PCE1 triggers an LSP update for 162 LSP1, it sends a PCUpdate message to PCC1 for LSP1 containing the new 163 parameters. PCC1 will take the parameters into account and will send 164 a PCReport to PCE1 and PCE2 reflecting the changes. PCE2 will so be 165 notified of the change only after receiving the PCReport from PCC1. 167 Let's consider that the LSP1 parameters changed in such a way that 168 LSP1 will take over resources from LSP2 with a higher priority. 169 After receiving the report from PCC1, PCE2 will therefore try to find 170 a new path for LSP2. If we consider that there is a round trip delay 171 of about 150msec between the PCEs and PCC1 and a round trip delay of 172 10msec between the two PCEs, if will take more than 150msec for PCE2 173 to be notified of the change. 175 Adding a PCEP session between PCE1 and PCE2 may allow to reduce the 176 notification time, so PCE2 can react more quickly by taking the 177 pending LSPs and attached resources into account during path 178 computation and reoptimization. 180 1.2. Split-brain 182 In a resiliency case, a PCC has redundant PCEP sessions towards 183 multiple PCEs. In such a case, a PCC gives control on an LSP to a 184 single PCE only, and only this PCE is responsible for the path 185 computation for the delegated LSP: the PCC achieves this by setting 186 the D flag only to the active PCE. The election of the active PCE to 187 delegate an LSP is controlled by each PCC. The PCC usually elects 188 the active PCE by a local configured policy (by setting a priority). 190 Upon PCEP session failure, or active PCE failure, PCC may decide to 191 elect a new active PCE by sending new PCRpt message with D flag set 192 to this new active PCE. When the failed PCE or PCEP session comes 193 back online, it will be up to the vendor to implement preemption. 194 Doing preemption may lead to some traffic disruption on the existing 195 path if path results from both PCEs are not exactly the same. By 196 considering a network with multiple PCCs and implementing multiple 197 stateful PCEs for redundancy purpose, there is no guarantee that at 198 any time all the PCCs delegate their LSPs to the same PCE. 200 +----------+ 201 | PCC1 | LSP1 202 +----------+ 203 / \ 204 / \ 205 +---------+ +---------+ 206 | PCE1 | | PCE2 | 207 +---------+ +---------+ 208 \ / 209 *fail* \ / 210 +----------+ 211 | PCC2 | LSP2 212 +----------+ 214 In the example above, we consider that by configuration, both PCCs 215 will firstly delegate their LSP to PCE1. So PCE1 is responsible for 216 computing a path for LSP1 and LSP2. If the PCEP session between PCC2 217 and PCE1 fails, PCC2 will delegate LSP2 to PCE2. So PCE1 becomes 218 responsible only for LSP1 path computation while PCE2 is responsible 219 for the path computation of LSP2. When the PCC2-PCE1 session is back 220 online, PCC2 will keep using PCE2 as active PCE (no preemption in 221 this example). So the result is a permanent situation where each PCE 222 is responsible for a subset of path computation. 224 We call this situation a split-brain scenario as there are multiple 225 computation brains running at the same time while a central 226 computation unit was required in some deployments. 228 Further, there are use cases where a particular LSP path computation 229 is linked to another LSP path computation: the most common use case 230 is path disjointness (see [I-D.ietf-pce-association-diversity]). The 231 set of LSPs that are dependant to each other may start from a 232 different head-end. 234 _________________________________________ 235 / \ 236 / +------+ +------+ \ 237 | | PCE1 | | PCE2 | | 238 | +------+ +------+ | 239 | | 240 | +------+ +------+ | 241 | | PCC1 | ----------------------> | PCC2 | | 242 | +------+ +------+ | 243 | | 244 | | 245 | +------+ +------+ | 246 | | PCC3 | ----------------------> | PCC4 | | 247 | +------+ +------+ | 248 | | 249 \ / 250 \_________________________________________/ 252 _________________________________________ 253 / \ 254 / +------+ +------+ \ 255 | | PCE1 | | PCE2 | | 256 | +------+ +------+ | 257 | | 258 | +------+ 10 +------+ | 259 | | PCC1 | ----- R1 ---- R2 ------- | PCC2 | | 260 | +------+ | | +------+ | 261 | | | | 262 | | | | 263 | +------+ | | +------+ | 264 | | PCC3 | ----- R3 ---- R4 ------- | PCC4 | | 265 | +------+ +------+ | 266 | | 267 \ / 268 \_________________________________________/ 270 In the figure above, we want to create two link-disjoint LSPs: 271 PCC1->PCC2 and PCC3->PCC4. In the topology, all link metrics are 272 equal to 1 except the link R1-R2 which has a metric of 10. The PCEs 273 are responsible for the path computation and PCE1 is the active PCE 274 for all PCCs in the nominal case. 276 Scenario 1: 278 In the nominal case (PCE1 as active PCE), we first configure 279 PCC1->PCC2 LSP, as the only constraint is path disjointness, PCE1 280 sends a PCUpdate message to PCC1 with the ERO: R1->R3->R4->R2->PCC2 281 (shortest path). PCC1 signals and installs the path. When 282 PCC3->PCC4 is configured, the PCE already knows the path of 283 PCC1->PCC2 and can compute a link-disjoint path : the solution 284 requires to move PCC1->PCC2 onto a new path to let room for the new 285 LSP. PCE1 sends a PCUpdate message to PCC1 with the new ERO: 286 R1->R2->PCC2 and a PCUpdate to PCC3 with the following ERO: 287 R3->R4->PCC4. In the nominal case, there is no issue for PCE1 to 288 compute a link-disjoint path. 290 Scenario 2: 292 Now we consider that PCC1 losts its PCEP session with PCE1 (all other 293 PCEP sessions are UP). PCC1 delegates its LSP to PCE2. 295 +----------+ 296 | PCC1 | LSP: PCC1->PCC2 297 +----------+ 298 \ 299 \ D=1 300 +---------+ +---------+ 301 | PCE1 | | PCE2 | 302 +---------+ +---------+ 303 D=1 \ / D=0 304 \ / 305 +----------+ 306 | PCC3 | LSP: PCC3->PCC4 307 +----------+ 309 We first configure PCC1->PCC2 LSP, as the only constraint is path 310 disjointness, PCE2 (which is the new active PCE for PCC1) sends a 311 PCUpdate message to PCC1 with the ERO: R1->32->R4->R2->PCC2 (shortest 312 path). When PCC3->PCC4 is configured, PCE1 is not aware anymore of 313 LSPs from PCC1, so it cannot compute a disjoint path for PCC3->PCC4 314 and will send a PCUpdate message to PCC2 with a shortest path ERO: 315 R3->R4->PCC4. When PCC3->PCC4 LSP will be reported to PCE2 by PCC2, 316 PCE2 will ensure disjointness computation and will correctly move 317 PCC1->PCC2 (as it owns delegation for this LSP) on the following 318 path: R1->R2->PCC2. With this sequence of event and this PCEP 319 session topology, disjointness is ensured. 321 Scenario 3: 323 +----------+ 324 | PCC1 | LSP: PCC1->PCC2 325 +----------+ 326 / \ 327 D=1 / \ D=0 328 +---------+ +---------+ 329 | PCE1 | | PCE2 | 330 +---------+ +---------+ 331 / D=1 332 / 333 +----------+ 334 | PCC3 | LSP: PCC3->PCC4 335 +----------+ 337 With this new PCEP session topology, we first configure PCC1->PCC2, 338 PCE1 computes the shortest path as it is the only LSP in the 339 disjoint-group that it is aware of: R1->R3->R4->R2->PCC2 (shortest 340 path). When PCC3->PCC4 is configured, PCE2 must compute a disjoint 341 path for this LSP. The only solution found is to move PCC1->PCC2 LSP 342 on another path, but PCE2 cannot do it as it does not have delegation 343 for this LSP. In this setup, PCEs are not able to find a disjoint 344 path. 346 Scenario 4: 348 +----------+ 349 | PCC1 | LSP: PCC1->PCC2 350 +----------+ 351 / \ 352 D=1 / \ D=0 353 +---------+ +---------+ 354 | PCE1 | | PCE2 | 355 +---------+ +---------+ 356 D=0 \ / D=1 357 \ / 358 +----------+ 359 | PCC3 | LSP: PCC3->PCC4 360 +----------+ 362 With this new PCEP session topology, we consider that PCEs are 363 configured to fallback to shortest path if disjointness cannot be 364 found. We first configure PCC1->PCC2, PCE1 computes shortest path as 365 it is the only LSP in the disjoint-group that it is aware of: 366 R1->R3->R4->R2->PCC2 (shortest path). When PCC3->PCC4 is configured, 367 PCE2 must compute a disjoint path for this LSP. The only solution 368 found is to move PCC1->PCC2 LSP on another path, but PCE2 cannot do 369 it as it does not have delegation for this LSP. PCE2 then provides 370 shortest path for PCC3->PCC4: R3->R4->PCC4. When PCC3 receives the 371 ERO, it reports it back to both PCEs. When PCE1 becomes aware of 372 PCC3->PCC4 path, it recomputes the CSPF and provides a new path for 373 PCC1->PCC2: R1->R2->PCC2. The new path is reported back to all PCEs 374 by PCC1. PCE2 recomputes also CSPF to take into account the new 375 reported path. The new computation does not lead to any path update. 377 Scenario 5: 379 _____________________________________ 380 / \ 381 / +------+ +------+ \ 382 | | PCE1 | | PCE2 | | 383 | +------+ +------+ | 384 | | 385 | +------+ 100 +------+ | 386 | | | -------------------- | | | 387 | | PCC1 | ----- R1 ----------- | PCC2 | | 388 | +------+ | +------+ | 389 | | | | | 390 | 6 | | 2 | 2 | 391 | | | | | 392 | +------+ | +------+ | 393 | | PCC3 | ----- R3 ----------- | PCC4 | | 394 | +------+ 10 +------+ | 395 | | 396 \ / 397 \_____________________________________/ 399 Now we consider a new network topology with the same PCEP session 400 topology as the previous example. We configure both LSPs almost at 401 the same time. PCE1 will compute a path for PCC1->PCC2 while PCE2 402 will compute a path for PCC3->PCC4. As each other is not aware of 403 the path of the second LSP in the group (not reported yet), each PCE 404 is computing shortest path for the LSP. PCE1 computes ERO: R1->PCC2 405 for PCC1->PCC2 and PCE2 computes ERO: R3->R1->PCC2->PCC4 for 406 PCC3->PCC4. When these shortest paths will be reported to each PCE. 407 Each PCE will recompute disjointness. PCE1 will provide a new path 408 for PCC1->PCC2 with ERO: PCC1->PCC2. PCE2 will provide also a new 409 path for PCC3->PCC4 with ERO: R3->PCC4. When those new paths will be 410 reported to both PCEs, this will trigger CSPF again. PCE1 will 411 provide a new more optimal path for PCC1->PCC2 with ERO: R1->PCC2 and 412 PCE2 will also provide a more optimal path for PCC3->PCC4 with ERO: 413 R3->R1->PCC2->PCC4. So we come back to the initial state. When 414 those paths will be reported to both PCEs, this will trigger CSPF 415 again. An infinite loop of CSPF computation is then happening with a 416 permanent flap of paths because of the split-brain situation. 418 This permanent computation loop comes from the inconsistency between 419 the state of the LSPs as seen by each PCE due to the split-brain: 420 each PCE is trying to modify at the same time its delegated path 421 based on the last received path information which defacto invalidates 422 this receives path information. 424 Scenario 6: multi-domain 426 Domain/Area 1 Domain/Area 2 427 ________________ ________________ 428 / \ / \ 429 / +------+ | | +------+ \ 430 | | PCE1 | | | | PCE3 | | 431 | +------+ | | +------+ | 432 | | | | 433 | +------+ | | +------+ | 434 | | PCE2 | | | | PCE4 | | 435 | +------+ | | +------+ | 436 | | | | 437 | +------+ | | +------+ | 438 | | PCC1 | | | | PCC2 | | 439 | +------+ | | +------+ | 440 | | | | 441 | | | | 442 | +------+ | | +------+ | 443 | | PCC3 | | | | PCC4 | | 444 | +------+ | | +------+ | 445 \ | | | 446 \_______________/ \________________/ 448 In the example above, we want to create disjoint LSPs from PCC1 to 449 PCC2 and from PCC4 to PCC3. All the PCEs have the knowledge of both 450 domain topologies (e.g. using BGP-LS). For operation/management 451 reason, each domain uses its own group of redundant PCEs. PCE1/PCE2 452 in domain 1 have PCEP sessions with PCC1 and PCC3 while PCE3/PCE4 in 453 domain 2 have PCEP sessions with PCC2 and PCC4. As PCE1/2 do not 454 know about LSPs from PCC2/4 and PCE3/4 do not know about LSPs from 455 PCC1/3, there is no possibility to compute the disjointness 456 constraint. This scenario can also be seen as a split-brain 457 scenario. This multi-domain architecture (with multiple groups of 458 PCEs) can also be used in a single domain, where an operator wants to 459 limit the failure domain by creating multiple groups of PCEs 460 maintaining a subset of PCCs. As for the multi-domain example, there 461 will be no possibility to compute disjoint path starting from head- 462 ends managed by different PCE groups. 464 In this document, we will propose a solution that address the 465 possibility to compute LSP association based constraints (like 466 disjointness) in split-brain scenarios while preventing computation 467 loops. 469 1.3. Applicability to H-PCE 471 [I-D.ietf-pce-stateful-hpce] describes general considerations and use 472 cases for the deployment of Stateful PCE(s) using the Hierarchical 473 PCE [RFC6805] architecture. In this architecture there is a clear 474 need to communicate between a child stateful PCE and a parent 475 stateful PCE. The procedures and extensions as described in 476 Section 3 are equally applicable to H-PCE. 478 2. Proposed solution 480 Our solution is based on : 482 o The creation of the inter-PCE stateful PCEP session with specific 483 procedures. 485 o A Master/Slave relationship between PCEs. 487 2.1. State-sync session 489 We propose to create a PCEP session between the stateful PCEs. 490 Creating such session is already authorized by multiple scenarios 491 like the one described in [RFC4655] (multiple PCEs that are handling 492 part of the path computation) and [RFC6805] (hierarchical PCE) but 493 was only focused on stateless PCEP sessions. As stateful PCE brings 494 additional features (LSP state synchronization, path update ...), 495 thus some new behaviors need to be defined. 497 This inter-PCE PCEP session will allow exchange of LSP states between 498 PCEs that would help some scenario where PCEP sessions are lost 499 between PCC and PCE. This inter-PCE PCEP session is called a state- 500 sync session. 502 For example, in the scenario below, there is no possibility to 503 compute disjointness as there is no PCE aware of both LSPs. 505 +----------+ 506 | PCC1 | LSP: PCC1->PCC2 507 +----------+ 508 / 509 D=1 / 510 +---------+ +---------+ 511 | PCE1 | | PCE2 | 512 +---------+ +---------+ 513 / D=1 514 / 515 +----------+ 516 | PCC3 | LSP: PCC3->PCC4 517 +----------+ 519 If we add a state-sync session, PCE1 will be able to send PCReport 520 messages for its LSP to PCE2 and PCE2 will do the same. All the PCEs 521 will be aware of all LSPs even if PCC->PCE session are down. PCEs 522 will then be able to compute disjoint paths. 524 +----------+ 525 | PCC1 | LSP : PCC1->PCC2 526 +----------+ 527 / 528 D=1 / 529 +---------+ PCEP +---------+ 530 | PCE1 | ----- | PCE2 | 531 +---------+ +---------+ 532 / D=1 533 / 534 +----------+ 535 | PCC3 | LSP : PCC3->PCC4 536 +----------+ 538 The procedures associated with this state-sync session are defined in 539 Section 3. 541 Adding this state-sync session does not ensure that a path with LSP 542 association based constraints can always been computed and does not 543 prevent computation loop, but it increases resiliency and ensures 544 that PCEs will have the state information for all LSPs. In addition, 545 this session will allow for a PCE to update the other PCEs providing 546 a faster synchronization mechanism than relying on PCCs only. 548 2.2. Master/Slave relationship between PCE 550 As seen in Section 1, performing a path computation in a split-brain 551 scenario (multiple PCEs responsible for computation) may provide a 552 non optimal LSP placement, no path or computation loops. To provide 553 the best efficiency, an LSP association constraint based computation 554 requires that a single PCE performs the path computation for all LSPs 555 in the association group. Note that, it could be all LSPs belonging 556 to a particular association group, or all LSPs from a particular PCC, 557 or all LSPs in the network that need to be delegated to a single PCE 558 based on the deployment scenarios. 560 We propose to add a priority mechanism between PCEs to elect a single 561 computing PCE. Using this priority mechanism, PCEs can agree on the 562 PCE that will be responsible for the computation for a particular 563 association group, or set of LSPs. The priority could be set per 564 association, per PCC, or for all LSPs. How this priority is set or 565 advertised is out of scope of this document. The rest of the text 566 consider association group as an example. 568 When a single PCE is performing the computation for a particular 569 association group, no computation loop can happen and an optimal 570 placement will be provided. The other PCEs will only act as state 571 collectors and forwarders. 573 In the scenario described in Section 2.1, PCE1 and PCE2 will decide 574 that PCE1 will be responsible for the path computation of both LSPs. 575 If we first configure PCC1->PCC2, PCE1 computes shortest path at it 576 is the only LSP in the disjoint-group that it is aware of: 577 R1->R3->R4->R2->PCC2 (shortest path). When PCC3->PCC4 is configured, 578 PCE2 will not perform computation even if it has delegation but 579 forwards the PCRpt to PCE1 through the state-sync session. PCE1 will 580 then perform disjointness computation and will move PCC1->PCC2 onto 581 R1->R2->PCC2 and provides an ERO to PCE2 for PCC3->PCC4: 582 R3->R4->PCC4. 584 3. Procedures and protocol extensions 586 3.1. Opening a state-sync session 588 3.1.1. Capability advertisement 590 A PCE indicates its support of state-sync procedures during the PCEP 591 Initialization phase. The Open object in the Open message MUST 592 contains the "Stateful PCE Capability" TLV defined in [RFC8231]. A 593 new P (INTER-PCE-CAPABILITY) flag is introduced to indicate the 594 support of state-sync. 596 The format of the STATEFUL-PCE-CAPABILITY TLV is shown in the 597 following figure: 599 0 1 2 3 600 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 1 601 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 602 | Type | Length=4 | 603 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 604 | Flags |P|F|D|T|I|S|U| 605 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 607 This document only updates the Flags field with : 609 P (INTER-PCE-CAPABILITY - 1 bit): If set to 1 by a PCEP Speaker, 610 the PCEP speaker indicates that the session MUST follow the state- 611 sync procedures as described in this document. The P bit MUST be 612 set by both speakers: if a PCEP Speaker receives a STATEFUL-PCE- 613 CAPABILITY TLV with P=0 while it advertised P=1 or if both set P 614 flag to 0, the session SHOULD open but the state-sync procedures 615 MUST NOT be applied on this session. 617 The U flag MUST be set when sending the STATEFUL-PCE-CAPABILITY TLV 618 with the P flag set. S flag MAY be set if optimized synchronization 619 is required as per [RFC8232]. 621 3.2. State synchronization 623 When the INTER-PCE-CAPABILITY has been negotiated, each PCEP speaker 624 will behave as a PCE and as a PCC at the same time regarding the 625 state synchronization as defined in [RFC8231]. This means that each 626 PCEP Speaker: 628 o MUST send a PCRpt message towards its neighbor with S flag set for 629 each LSP in its LSP database learned from a PCC. (PCC role) 631 o MUST send the End Of Synchronization Marker towards its neighbor 632 when all LSPs have been reported. (PCC role) 634 o MUST wait for the LSP synchronization from its neighbor to end 635 (receiving an End Of Synchronization Marker). (PCE role) 637 The process of synchronization runs in parallel on each PCE (no 638 defined order). 640 Optimized synchronization MAY be used as defined in [RFC8232]. 642 When a PCEP Speaker sends a PCReport on a state-sync session, it MUST 643 add the SPEAKER-IDENTITY-TLV (defined in [RFC8232]) in the LSP 644 Object, the value used will refer to the PCC owner of the LSP. If a 645 PCEP Speaker receives a PCReport on a state-sync session without this 646 TLV, it MUST discard the PCReport and it MUST reply with a PCErr 647 message using error-type=6 (Mandatory Object missing) and error- 648 value=TBD1 (SPEAKER-IDENTITY-TLV missing). 650 3.3. Incremental updates and report forwarding rules 652 During the life of an LSP, its state may change (path, constraints, 653 operational state...) and a PCC will advertise a new PCReport to the 654 PCE for each such change. 656 When propagating LSP state changes from a PCE to other PCEs, it is 657 mandatory to ensure that a PCE always uses the freshest state coming 658 from the PCC. 660 When a PCE receives a new PCReport from a PCC with the LSP-DB- 661 VERSION, the PCE MUST forward the PCReport to all its state-sync 662 sessions and MUST add the appropriate SPEAKER-IDENTITY-TLV in the 663 PCReport. In addition, it MUST add a new ORIGINAL-LSP-DB-VERSION TLV 664 (described below). The ORIGINAL-LSP-DB-VERSION should contain the 665 LSP-DB-VERSION coming from the PCC. 667 When a PCE receives a new PCReport from a PCC without the LSP-DB- 668 VERSION, it SHOULD NOT forward the PCReport on any state-sync 669 sessions. 671 When a PCE receives a new PCReport from a PCC with the R flag set and 672 a LSP-DB-VERSION TLV, the PCE MUST forward the PCReport to all its 673 state-sync sessions keeping the R flag set (Remove) and MUST add the 674 appropriate SPEAKER-IDENTITY-TLV and ORIGINAL-LSP-DB-VERSION TLV in 675 the PCReport. 677 When a PCE receives a PCReport from a state-sync session, it MUST NOT 678 forward the PCReport to other state-sync sessions. This helps to 679 prevent message loops between PCEs. As a consequence, a full mesh of 680 PCEP sessions between PCEs is required. 682 When a PCReport is forwarded, all the original objects and values are 683 kept. As an example, the PLSP-ID used in the forwarded PCReport will 684 be the same as the original one used by the PCC. Thus an 685 implementation supporting this document MUST consider SPEAKER- 686 IDENTITY-TLV and PLSP-ID together to uniquely identify an LSP on the 687 state-sync session. 689 The ORIGINAL-LSP-DB-VERSION TLV is encoded as follows and SHOULD 690 always contain the LSP-DB-VERSION received from the PCC owner of the 691 LSP: 693 0 1 2 3 694 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 1 695 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 696 | Type=TBD2 | Length=8 | 697 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 698 | LSP State DB Version Number | 699 | | 700 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 702 Using the ORIGINAL-LSP-DB-VERSION TLV allows a PCE to keep using 703 optimized synchronization ([RFC8232]) with another PCE. In such a 704 case, the PCE will send a PCReport to another PCE with both ORIGINAL- 705 LSP-DB-VERSION TLV and LSP-DB-VERSION TLV. The ORIGINAL-LSP-DB- 706 VERSION TLV will contain the version number as allocated by the PCC 707 while the LSP-DB-VERSION will contain the version number allocated by 708 the local PCE. 710 3.4. Maintaining LSP states from different sources 712 When a PCE receives a PCReport on a state-sync session, it stores the 713 LSP information into the original PCC address context (as the LSP 714 belongs to the PCC). A PCE SHOULD maintain a single state for a 715 particular LSP and SHOULD maintain the list of sources it learned a 716 particular state from. 718 A PCEP speaker may receive a state information for a particular LSP 719 from different sources: the PCC that owns the LSP (through a regular 720 PCEP session) and some PCEs (through PCEP state-sync sessions). A 721 PCEP speaker MUST always keep the freshest state in its LSP database, 722 overriding the previously received information. 724 A PCE, receiving a PCReport from a PCC, updates the state of the LSP 725 in its LSPDB with the new received information. When receiving a 726 PCReport from another PCE, a PCE SHOULD update the LSP state only if 727 the ORIGINAL-LSP-DB-VERSION present in the PCReport is greater than 728 the current ORIGINAL-LSP-DB-VERSION of the stored LSP state. This 729 ensures that a PCE never tries to update its stored LSP state with an 730 old information. Each time a PCE updates an LSP state in its LSPDB, 731 it SHOULD reset the source list associated with the LSP state and 732 SHOULD add the source speaker address in the source list. When a PCE 733 receives a PCReport which has an ORIGINAL-LSP-DB-VERSION (if coming 734 from a PCE) or an LSP-DB-VERSION (if coming from the PCC) equals to 735 the current ORIGINAL-LSP-DB-VERSION of the stored LSP state, it 736 SHOULD add the source speaker address in the source list. 738 When a PCE receives a PCReport requesting an LSP deletion from a 739 particular source, it SHOULD remove this particular source from the 740 list of sources associated with this LSP. 742 When the list of sources becomes empty for a particular LSP, the LSP 743 state MUST be removed. This means that all the sources must send a 744 PCReport with R=1 for an LSP to make the PCE removing the LSP state. 746 3.5. Computation priority between PCEs and sub-delegation 748 A computation priority is necessary to ensure that a single PCE will 749 perform the computation for all the LSPs in an association group: 750 this will allow for a more optimized LSP placement and will prevent 751 computation loops. 753 All PCEs in the network that are handling LSPs in a common LSP 754 association group SHOULD be aware of each other including the 755 computation priority of each PCE. Note that there is no need for PCC 756 to be aware of this. The computation priority is a number and the 757 PCE having the highest priority SHOULD be responsible for the 758 computation. If several PCEs have the same priority value, their IP 759 address SHOULD be used as a tie-breaker to provide a rank: the 760 highest IP address as more priority. How PCEs are aware of the 761 priority of each other is out of scope of this document, but as 762 example learning priorities could be done through IGP informations or 763 local configuration. 765 The definition of the priority MAY be global so the highest priority 766 PCE will handle all path computations or more granular, so a PCE may 767 have highest priority for only a subset of LSPs or association- 768 groups. 770 A PCEP Speaker receiving a PCReport from a PCC with D flag set that 771 does not have the highest computation priority, SHOULD forward the 772 PCReport on all state-sync sessions (as per Section 3.3) and SHOULD 773 set D flag on the state-sync session towards the highest priority 774 PCE, D flag will be unset to all other state-sync sessions. This 775 behavior is similar to the delegation behavior handled at PCC side 776 and is called a sub-delegation (the PCE subdelegates the control of 777 the LSP to another PCE). When a PCEP Speaker sub-delegates a LSP to 778 another PCE, it looses the control on the LSP and cannot update it 779 anymore by its own decision. When a PCE receives a PCReport with D 780 flag set on a state-sync session, as a regular PCE, it becomes 781 granted to update the LSP. 783 If the highest priority PCE is failing or if the state-sync session 784 between the local PCE and the highest priority PCE failed, the local 785 PCE MAY decide to delegate the LSP to the next highest priority PCE 786 or to take back control on the LSP. It is a local policy decision. 788 When a PCE has the delegation for an LSP and needs to update this 789 LSP, it MUST send a PCUpdate message to all state-sync sessions and 790 to the PCC session on which it received the delegation. The D-Flag 791 would be unset in the PCUpdate for state-sync sessions where as 792 D-Flag would be set for the PCC. In case of subdelegation, the 793 computing PCE will send the PCUpdate only to all state-sync sessions 794 (as it has no direct delegation from a PCC). The D-Flag would be set 795 for the state-sync session to the PCE that sub-delegated this LSP and 796 the D-Flag would be unset for other state-sync sessions. 798 The PCUpdate sent over a state-sync session MUST contain the SPEAKER- 799 IDENTITY-TLV in the LSP Object (the value used must identify the 800 target PCC). The PLSP-ID used is the original PLSP-ID generated by 801 the PCC and learned from the forwarded PCReport. If a PCE receives a 802 PCUpdate on a state-sync session without the SPEAKER-IDENTITY-TLV, it 803 MUST discard the PCUpdate and MUST reply with a PCError message using 804 error-type=6 (Mandatory Object missing) and error-value=TBD1 805 (SPEAKER-IDENTITY-TLV missing). 807 When a PCE receives a valid PCUpdate on a state-sync session, it 808 SHOULD forward the PCUpdate to the appropriate PCC (identified based 809 on the SPEAKER-IDENTITY-TLV value) that delegated the LSP originally 810 and SHOULD remove the SPEAKER-IDENTITY-TLV from the LSP Object. The 811 acknowlegment of the PCUpdate is done through a cascaded mechanism, 812 and the PCC is the only responsible of triggering the acknowledgment: 813 when the PCC receives the PCUpdate from the local PCE, it 814 acknowledges it with a PCReport as per [RFC8231]. When receiving the 815 new PCReport from the PCC, the local PCE uses the defined forwarding 816 rules on the state-sync session so the acknowledgment is relayed to 817 the computing PCE. 819 A PCE SHOULD NOT compute a path using an association-group constraint 820 if it has delegation for only a subset of LSPs in the group. In this 821 case, an implementation MAY use a local policy on PCE to decide if 822 PCE does not compute path at all for this set of LSP or if it can 823 compute a path by relaxing the association-group constraint. 825 3.6. Passive stateful procedures 827 In the passive stateful PCE architecture, the PCC is responsible of 828 triggering a path computation request using a PCRequest message to 829 its PCE. Similarly to PCReports which remains unchanged for passive 830 mode, if a PCE receives a PCRequest for an LSP and if this PCE finds 831 that it does not have the highest computation priority of this LSP, 832 or groups..., it MUST forward the PCRequest to the highest priority 833 PCE over the state-sync session. When the highest priority PCE 834 receives the PCRequest, it computes the path and generates a PCReply 835 only to the PCE that is received the PCRequest from. This PCE will 836 then forward the PCReply to the requesting PCC. The handling of LSP 837 object and the SPEAKER-IDENTITY-TLV in PCRequest and PCReply is 838 similar to PCReport/PCUpdate. 840 3.7. PCE initiation procedures 842 TBD 844 4. Examples 846 4.1. Example 1 848 _________________________________________ 849 / \ 850 / +------+ +------+ \ 851 | | PCE1 | | PCE2 | | 852 | +------+ +------+ | 853 | | 854 | +------+ 10 +------+ | 855 | | PCC1 | ----- R1 ---- R2 ------- | PCC2 | | 856 | +------+ | | +------+ | 857 | | | | 858 | | | | 859 | +------+ | | +------+ | 860 | | PCC3 | ----- R3 ---- R4 ------- | PCC4 | | 861 | +------+ +------+ | 862 | | 863 \ / 864 \_________________________________________/ 866 +----------+ 867 | PCC1 | LSP : PCC1->PCC2 868 +----------+ 869 / 870 D=1 / 871 +---------+ +---------+ 872 | PCE1 |----| PCE2 | 873 +---------+ +---------+ 874 / D=1 875 / 876 +----------+ 877 | PCC3 | LSP : PCC3->PCC4 878 +----------+ 880 PCE1 computation priority 100 881 PCE2 computation priority 200 882 With this PCEP session topology where computation priority is global 883 for all LSPs, we still want to have link disjoint LSPs PCC1->PCC2 and 884 PCC3->PCC4. 886 We first configure PCC1->PCC2, PCC1 delegates the LSP to PCE1, but as 887 PCE1 does not have the highest computation priority, it will sub- 888 delegate the LSP to PCE2 by sending a PCReport with D=1 and including 889 the SPEAKER-IDENTITY-TLV over the state-sync session. PCE2 receives 890 the PCReport and as it has delegation for this LSP, it computes the 891 shortest path: R1->R3->R4->R2->PCC2. It then sends a PCUpdate to 892 PCE1 (including the SPEAKER-IDENTITY-TLV) with the computed ERO. 893 PCE1 forwards the PCUpdate to PCC1 (removing the SPEAKER-IDENTITY- 894 TLV). PCC1 acknowledges the PCUpdate by a PCReport to PCE1. PCE1 895 forwards the PCReport to PCE2. 897 When PCC3->PCC4 is configured, PCC3 delegates the LSP to PCE2, PCE2 898 can compute a disjoint path as it has knowledge of both LSPs and has 899 delegation also for both. The only solution found is to move 900 PCC1->PCC2 LSP on another path, PCE2 can move PCC3->PCC4 as it has 901 delegation for it. It creates a new PCUpdate with new ERO: 902 R1->R2-PCC2 towards PCE1 which forwards to PCC1. PCE2 sends a 903 PCUpdate to PCC3 with the path: R3->R4->PCC4. 905 In this setup, PCEs are able to find a disjoint path while without 906 state-sync and computation priority they could not. 908 4.2. Example 2 909 _____________________________________ 910 / \ 911 / +------+ +------+ \ 912 | | PCE1 | | PCE2 | | 913 | +------+ +------+ | 914 | | 915 | +------+ 100 +------+ | 916 | | | -------------------- | | | 917 | | PCC1 | ----- R1 ----------- | PCC2 | | 918 | +------+ | +------+ | 919 | | | | | 920 | 6 | | 2 | 2 | 921 | | | | | 922 | +------+ | +------+ | 923 | | PCC3 | ----- R3 ----------- | PCC4 | | 924 | +------+ 10 +------+ | 925 | | 926 \ / 927 \_____________________________________/ 929 +----------+ 930 | PCC1 | LSP : PCC1->PCC2 931 +----------+ 932 / \ 933 D=1 / \ D=0 934 +---------+ +---------+ 935 | PCE1 |----| PCE2 | 936 +---------+ +---------+ 937 D=0 \ / D=1 938 \ / 939 +----------+ 940 | PCC3 | LSP : PCC3->PCC4 941 +----------+ 943 PCE1 computation priority 200 944 PCE2 computation priority 100 946 In this example, we configure both LSPs almost at the same time. 947 PCE1 sub-delegates PCC1->PCC2 to PCE2 while PCE2 keeps delegation for 948 PCC3->PCC4, PCE2 computes a path for PCC1->PCC2 and PCC3->PCC4 and 949 can achieve disjointness computation easily. No computation loop 950 happens in this case. 952 4.3. Example 3 954 _________________________________________ 955 / \ 956 / +------+ +------+ \ 957 | | PCE1 | | PCE2 | | 958 | +------+ +------+ | 959 | | 960 | +------+ 10 +------+ | 961 | | PCC1 | ----- R1 ---- R2 ------- | PCC2 | | 962 | +------+ | | +------+ | 963 | | | | 964 | | | | 965 | +------+ | | +------+ | 966 | | PCC3 | ----- R3 ---- R4 ------- | PCC4 | | 967 | +------+ +------+ | 968 | | 969 \ / 970 \_________________________________________/ 972 +----------+ 973 | PCC1 | LSP : PCC1->PCC2 974 +----------+ 975 / 976 D=1 / 977 +---------+ +---------+ +---------+ 978 | PCE1 |----| PCE2 |----| PCE3 | 979 +---------+ +---------+ +---------+ 980 / D=1 981 / 982 +----------+ 983 | PCC3 | LSP : PCC3->PCC4 984 +----------+ 986 PCE1 computation priority 100 987 PCE2 computation priority 200 988 PCE2 computation priority 300 990 With this PCEP session topology, we still want to have link disjoint 991 LSPs PCC1->PCC2 and PCC3->PCC4. 993 We first configure PCC1->PCC2, PCC1 delegates the LSP to PCE1, but as 994 PCE1 does not have the highest computation priority, it will sub- 995 delegate the LSP to PCE2 (as it cannot reach PCE3 through a state- 996 sync session). PCE2 cannot compute a path for PCC1->PCC2 as it does 997 not have the highest priority and cannot sub-delegate the LSP again 998 towards PCE3. 1000 When PCC3->PCC4 is configured, PCC3 delegates the LSP to PCE2 that 1001 performs sub-delegation to PCE3. As PCE3 will have knowledge of only 1002 one LSP in the group, it cannot compute disjointness and can decide 1003 to fallback to a less constrained computation to provide a path for 1004 PCC3->PCC4. In this case, it will send a PCUpdate to PCE2 that will 1005 be forwarded to PCC3. 1007 Disjointness cannot be achieved in this scenario because of lack of 1008 state-sync session between PCE1 and PCE3, but no computation loop 1009 happens. Thus it is advised for all PCEs that support state-sync to 1010 have a full mesh sessions between each other. 1012 5. Using Master/Slave computation and state-sync sessions to increase 1013 scaling 1015 The Primary/Backup computation and state-sync sessions architecture 1016 can be used to increase the scaling of the PCE architecture. If the 1017 number of PCCs is really high, it may be too resource consuming for a 1018 single PCE to maintain all the PCEP sessions while at the same time 1019 performing all path computations. Using master/slave computation and 1020 state-sync sessions may allow to create groups of PCEs that manage a 1021 subset of the PCCs and perform some or no path computations. 1022 Decoupling PCEP session maintenance and computation will allow to 1023 increase scaling of the PCE architecture. 1025 +----------+ 1026 | PCC500 | 1027 +----------+-+ 1028 | PCC1 | 1029 +----------+ 1030 / \ 1031 / \ 1032 +---------+ +---------+ 1033 | PCE1 |---| PCE2 | 1034 +---------+ +---------+ 1035 | \ / | 1036 | \/ | 1037 | /\ | 1038 | / \ | 1039 +---------+ +---------+ 1040 | PCE3 |---| PCE4 | 1041 +---------+ +---------+ 1042 \ / 1043 \ / 1044 +----------+ 1045 | PCC501 | 1046 +----------+-+ 1047 | PCC1000 | 1048 +----------+ 1050 In the figure above, two groups of PCEs are created: PCE1/2 maintain 1051 PCEP sessions with PCC1 up to PCC500, while PCE3/4 maintain PCEP 1052 sessions with PCC501 up to PCC1000. A granular master/slave policy 1053 is setup as follows to loadshare computation between PCEs: 1055 o PCE1 has priority 200 for association ID 1 up to 300, association 1056 source 0.0.0.0. All other PCEs have a decreasing priority for 1057 those associations. 1059 o PCE3 has priority 200 for association ID 301 up to 500, 1060 association source 0.0.0.0. All other PCEs have a decreasing 1061 priority for those associations. 1063 If some PCCs delegate LSPs with association ID 1 up to 300 and 1064 association source 0.0.0.0, the receiving PCE (if not PCE1) will sub- 1065 delegate the LSPs to PCE1. PCE1 becomes responsible for the 1066 computation of these LSP associations while PCE3 is responsible for 1067 the computation of another set of associations. 1069 6. PCEP-PATH-VECTOR-TLV 1071 This document allows PCEP messages to be propagated among PCEP 1072 speaker. It may be useful to track informations about the 1073 propagation of the messages. One of the use case is a message loop 1074 detection mechanism, but other use cases like hop by hop information 1075 recording may also be implemented. 1077 This document introduces the PCEP-PATH-VECTOR-TLV (type TBD2) with 1078 the following format: 1080 0 1 2 3 1081 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 1082 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1083 | Type=TBD3 | Length (variable) | 1084 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1085 | PCEP-SPEAKER-INFORMATION#1 | 1086 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1087 | ... | 1088 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1089 | PCEP-SPEAKER-INFORMATION#2 | 1090 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1091 | ... | 1092 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1094 The TLV format and padding rules are as per [RFC5440]. 1096 The PCEP-SPEAKER-INFORMATION field has the following format: 1098 0 1 2 3 1099 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 1100 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1101 | Length (variable) | ID Length (variable) | 1102 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1103 | Speaker Entity identity (variable) | 1104 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1105 | SubTLVs (optional) | 1106 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1108 Length: defines the total length of the PCEP-SPEAKER-INFORMATION 1109 field. 1111 ID Length: defines the length of the Speaker identity actual field 1112 (non-padded). 1114 Speaker Entity identity: same possible values as the SPEAKER- 1115 IDENTIFIER-TLV. Padded with trailing zeroes to a 4-byte boundary. 1117 The PCEP-SPEAKER-INFORMATION may also carry some optional subTLVs 1118 so each PCEP speaker can add local informations that could be 1119 recorded. This document does not define any subTLV. 1121 The PCEP-PATH-VECTOR-TLV MAY be added in the LSP-Object. Its usage 1122 is purely optional. 1124 The list of speakers within the PCEP-PATH-VECTOR-TLV MUST be ordered. 1125 When sending a PCEP message (PCReport, PCUpdate or PCInitiate), a 1126 PCEP Speaker MAY add the PCEP-PATH-VECTOR-TLV with a PCEP-SPEAKER- 1127 INFORMATION containing its own informations. If the PCEP message 1128 sent is the result of a previously received PCEP message, and if the 1129 PCEP-PATH-VECTOR-TLV was already present in the initial message, the 1130 PCEP speaker MAY append a new PCEP-SPEAKER-INFORMATION containing its 1131 own informations. 1133 7. Security Considerations 1135 TBD. 1137 8. Acknowledgements 1139 TBD. 1141 9. IANA Considerations 1143 This document requests IANA actions to allocate code points for the 1144 protocol elements defined in this document. 1146 9.1. PCEP-Error Object 1148 IANA is requested to allocate a new Error Value for the Error Type 9. 1150 Error-Type Meaning Reference 1151 6 Mandatory Object Missing [RFC5440] 1152 Error-value=TBD1: SPEAKER-IDENTITY-TLV This document 1153 missing 1155 9.2. PCEP TLV Type Indicators 1157 IANA is requested to allocate new TLV Type Indicator values within 1158 the "PCEP TLV Type Indicators" sub-registry of the PCEP Numbers 1159 registry, as follows: 1161 Value Meaning Reference 1162 TBD2 ORIGINAL-LSP-DB-VERSION-TLV This document 1163 TBD3 PCEP-PATH-VECTOR-TLV This document 1165 9.3. STATEFUL-PCE-CAPABILITY TLV 1167 IANA is requested to allocate a new bit value in the STATEFUL-PCE- 1168 CAPABILITY TLV Flag Field sub-registry. 1170 Bit Description Reference 1171 TBD INTER-PCE-CAPABILITY This document 1173 10. References 1175 10.1. Normative References 1177 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1178 Requirement Levels", BCP 14, RFC 2119, 1179 DOI 10.17487/RFC2119, March 1997, 1180 . 1182 [RFC5440] Vasseur, JP., Ed. and JL. Le Roux, Ed., "Path Computation 1183 Element (PCE) Communication Protocol (PCEP)", RFC 5440, 1184 DOI 10.17487/RFC5440, March 2009, 1185 . 1187 [RFC8231] Crabbe, E., Minei, I., Medved, J., and R. Varga, "Path 1188 Computation Element Communication Protocol (PCEP) 1189 Extensions for Stateful PCE", RFC 8231, 1190 DOI 10.17487/RFC8231, September 2017, 1191 . 1193 [RFC8232] Crabbe, E., Minei, I., Medved, J., Varga, R., Zhang, X., 1194 and D. Dhody, "Optimizations of Label Switched Path State 1195 Synchronization Procedures for a Stateful PCE", RFC 8232, 1196 DOI 10.17487/RFC8232, September 2017, 1197 . 1199 10.2. Informative References 1201 [I-D.ietf-pce-association-diversity] 1202 Litkowski, S., Sivabalan, S., Barth, C., and D. Dhody, 1203 "Path Computation Element communication Protocol extension 1204 for signaling LSP diversity constraint", draft-ietf-pce- 1205 association-diversity-03 (work in progress), February 1206 2018. 1208 [I-D.ietf-pce-stateful-hpce] 1209 Dhody, D., Lee, Y., Ceccarelli, D., Shin, J., King, D., 1210 and O. Dios, "Hierarchical Stateful Path Computation 1211 Element (PCE).", draft-ietf-pce-stateful-hpce-04 (work in 1212 progress), March 2018. 1214 [RFC4655] Farrel, A., Vasseur, J., and J. Ash, "A Path Computation 1215 Element (PCE)-Based Architecture", RFC 4655, 1216 DOI 10.17487/RFC4655, August 2006, 1217 . 1219 [RFC6805] King, D., Ed. and A. Farrel, Ed., "The Application of the 1220 Path Computation Element Architecture to the Determination 1221 of a Sequence of Domains in MPLS and GMPLS", RFC 6805, 1222 DOI 10.17487/RFC6805, November 2012, 1223 . 1225 [RFC7399] Farrel, A. and D. King, "Unanswered Questions in the Path 1226 Computation Element Architecture", RFC 7399, 1227 DOI 10.17487/RFC7399, October 2014, 1228 . 1230 Authors' Addresses 1232 Stephane Litkowski 1233 Orange 1235 Email: stephane.litkowski@orange.com 1237 Siva Sivabalan 1238 Cisco 1240 Email: msiva@cisco.com 1242 Dhruv Dhody 1243 Huawei 1244 Divyashree Techno Park, Whitefield 1245 Bangalore, Karnataka 560066 1246 India 1248 Email: dhruv.ietf@gmail.com