idnits 2.17.00 (12 Aug 2021) /tmp/idnits62496/draft-ietf-mpls-rfc6374-sfl-10.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (March 05, 2021) is 435 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) No issues found here. Summary: 0 errors (**), 0 flaws (~~), 1 warning (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 MPLS Working Group S. Bryant (Ed) 3 Internet-Draft Futurewei Technologies Inc. 4 Intended status: Standards Track G. Swallow 5 Expires: September 6, 2021 Southend Technical Center 6 M. Chen 7 Huawei 8 G. Fioccola 9 Huawei Technologies 10 G. Mirsky 11 ZTE Corp. 12 March 05, 2021 14 RFC6374 Synonymous Flow Labels 15 draft-ietf-mpls-rfc6374-sfl-10 17 Abstract 19 RFC 6374 describes methods of making loss and delay measurements on 20 Label Switched Paths (LSPs) primarily as used in MPLS Transport 21 Profile (MPLS-TP) networks. This document describes a method of 22 extending RFC 6374 performance measurements from flows carried over 23 MPLS-TP to flows carried over generic MPLS LSPs. In particular, it 24 extends the technique to allow loss and delay measurements to be made 25 on multi-point to point LSPs and introduces some additional 26 techniques to allow more sophisticated measurements to be made in 27 both MPLS-TP and generic MPLS networks. 29 Status of This Memo 31 This Internet-Draft is submitted in full conformance with the 32 provisions of BCP 78 and BCP 79. 34 Internet-Drafts are working documents of the Internet Engineering 35 Task Force (IETF). Note that other groups may also distribute 36 working documents as Internet-Drafts. The list of current Internet- 37 Drafts is at https://datatracker.ietf.org/drafts/current/. 39 Internet-Drafts are draft documents valid for a maximum of six months 40 and may be updated, replaced, or obsoleted by other documents at any 41 time. It is inappropriate to use Internet-Drafts as reference 42 material or to cite them other than as "work in progress." 44 This Internet-Draft will expire on September 6, 2021. 46 Copyright Notice 48 Copyright (c) 2021 IETF Trust and the persons identified as the 49 document authors. All rights reserved. 51 This document is subject to BCP 78 and the IETF Trust's Legal 52 Provisions Relating to IETF Documents 53 (https://trustee.ietf.org/license-info) in effect on the date of 54 publication of this document. Please review these documents 55 carefully, as they describe your rights and restrictions with respect 56 to this document. Code Components extracted from this document must 57 include Simplified BSD License text as described in Section 4.e of 58 the Trust Legal Provisions and are provided without warranty as 59 described in the Simplified BSD License. 61 Table of Contents 63 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 64 2. Requirements Language . . . . . . . . . . . . . . . . . . . . 4 65 3. RFC6374 Packet Loss Measurement with SFL . . . . . . . . . . 4 66 4. RFC6374 Single Packet Delay Measurement . . . . . . . . . . . 4 67 5. Data Service Packet Delay Measurement . . . . . . . . . . . . 5 68 6. Some Simplifying Rules . . . . . . . . . . . . . . . . . . . 6 69 7. Multiple Packet Delay Characteristics . . . . . . . . . . . . 7 70 7.1. Method 1: Time Buckets . . . . . . . . . . . . . . . . . 7 71 7.2. Method 2 Classic Standard Deviation . . . . . . . . . . . 9 72 7.2.1. Multi-Packet Delay Measurement Message Format . . . . 10 73 7.3. Per Packet Delay Measurement . . . . . . . . . . . . . . 11 74 7.4. Average Delay . . . . . . . . . . . . . . . . . . . . . . 11 75 8. Sampled Measurement . . . . . . . . . . . . . . . . . . . . . 13 76 9. Carrying RFC6374 Packets over an LSP using an SFL . . . . . . 13 77 9.1. RFC6374 SFL TLV . . . . . . . . . . . . . . . . . . . . . 15 78 10. RFC6374 Combined Loss-Delay Measurement . . . . . . . . . . . 16 79 11. Privacy Considerations . . . . . . . . . . . . . . . . . . . 17 80 12. Security Considerations . . . . . . . . . . . . . . . . . . . 17 81 13. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 17 82 13.1. Allocation of MPLS Generalized Associated Channel 83 (G-ACh) Types . . . . . . . . . . . . . . . . . . . . . 17 84 13.2. Allocation of MPLS Loss/Delay TLV Object . . . . . . . . 18 85 14. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 18 86 15. Contributing Authors . . . . . . . . . . . . . . . . . . . . 18 87 16. References . . . . . . . . . . . . . . . . . . . . . . . . . 18 88 16.1. Normative References . . . . . . . . . . . . . . . . . . 18 89 16.2. Informative References . . . . . . . . . . . . . . . . . 20 90 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 20 92 1. Introduction 94 [RFC6374] was originally designed for use as an Operations, 95 Administration, and Maintenance (OAM) protocol for use with MPLS 96 Transport Profile (MPLS-TP) [RFC5921] LSPs. MPLS-TP only supports 97 point-to-point and point-to-multi-point LSPs. This document 98 describes how to use RFC6374 in the generic MPLS case, and also 99 introduces a number of more sophisticated measurements of 100 applicability to both cases. 102 [RFC8372] describes the requirement for introducing flow identities 103 when using RFC6374 [RFC6374] packet Loss Measurements (LM). In 104 summary RFC6374 uses the loss-measurement (LM) packet as the packet 105 accounting demarcation point. Unfortunately this gives rise to a 106 number of problems that may lead to significant packet accounting 107 errors in certain situations. For example: 109 1. Where a flow is subjected to Equal Cost Multi-Path (ECMP) 110 treatment packets can arrive out of order with respect to the LM 111 packet. 113 2. Where a flow is subjected to ECMP treatment, packets can arrive 114 at different hardware interfaces, thus requiring reception of an 115 LM packet on one interface to trigger a packet accounting action 116 on a different interface which may not be co-located with it. 117 This is a difficult technical problem to address with the 118 required degree of accuracy. 120 3. Even where there is no ECMP (for example on RSVP-TE, MPLS-TP LSPs 121 and pseudowires(PWs)) local processing may be distributed over a 122 number of processor cores, leading to synchronization problems. 124 4. Link aggregation techniques [RFC7190] may also lead to 125 synchronization issues. 127 5. Some forwarder implementations have a long pipeline between 128 processing a packet and incrementing the associated counter, 129 again leading to synchronization difficulties. 131 An approach to mitigating these synchronization issue is described in 132 [RFC8321] in which packets are batched by the sender and each batch 133 is marked in some way such that adjacent batches can be easily 134 recognized by the receiver. 136 An additional problem arises where the LSP is a multi-point to point 137 LSP, since MPLS does not include a source address in the packet. 138 Network management operations require the measurement of packet loss 139 between a source and destination. It is thus necessary to introduce 140 some source specific information into the packet to identify packet 141 batches from a specific source. 143 [RFC8957] describes a method of encoding per flow instructions in an 144 MPLS label stack using a technique called Synonymous Flow Labels 145 (SFL) in which labels which mimic the behavior of other labels 146 provide the packet batch identifiers and enable the per batch packet 147 accounting. This memo specifies how SFLs are used to perform RFC6374 148 packet loss and RFC6374 delay measurements. 150 2. Requirements Language 152 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 153 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 154 "OPTIONAL" in this document are to be interpreted as described in BCP 155 14 [RFC2119] [RFC8174] when, and only when, they appear in all 156 capitals, as shown here. 158 3. RFC6374 Packet Loss Measurement with SFL 160 The data service packets of the flow being instrumented are grouped 161 into batches, and all the packets within a batch are marked with the 162 SFL [RFC8372] corresponding to that batch. The sender counts the 163 number of packets in the batch. When the batch has completed and the 164 sender is confident that all of the packets in that batch will have 165 been received, the sender issues an RFC6374 Query message to 166 determine the number actually received and hence the number of 167 packets lost. The RFC6374 Query message is sent using the same SFL 168 as the corresponding batch of data service packets. The format of 169 the Query and Response packets is described in Section 9. 171 4. RFC6374 Single Packet Delay Measurement 173 RFC6374 describes how to measure the packet delay by measuring the 174 transit time of an RFC6374 packet over an LSP. Such a packet may not 175 need to be carried over an SFL since the delay over a particular LSP 176 should be a function of the Traffic Class (TC) bits. 178 However, where SFLs are being used to monitor packet loss or where 179 label inferred scheduling is used [RFC3270] then the SFL would be 180 REQUIRED to ensure that the RFC6374 packet which was being used as a 181 proxy for a data service packet experienced a representative delay. 182 The format of an RFC6374 packet carried over the LSP using an SFL is 183 shown in Section 9. 185 5. Data Service Packet Delay Measurement 187 Where it is desired to more thoroughly instrument a packet flow and 188 to determine the delay of a number of packets it is undesirable to 189 send a large number of RFC6374 packets acting as a proxy data service 190 packets (see Section 4). A method of directly measuring the delay 191 characteristics of a batch of packets is therefore needed. 193 Given the long intervals over which it is necessary to measure packet 194 loss, it is not necessarily the case that the batch times for the two 195 measurement types would be identical. Thus, we use a technique that 196 permits the two measurements are made concurrently and yet relatively 197 independent from each other. The notion that they are relatively 198 independent arises from the potential for the two batches to overlap 199 in time, in which case either the delay batch time will need to be 200 cut short or the loss time will need to be extended to allow correct 201 reconciliation of the various counters. 203 The problem is illustrated in Figure 1 below: 205 (1) AAAAAAAAAABBBBBBBBBBAAAAAAAAAABBBBBBBBBB 207 SFL Marking of a packet batch for loss measurement 209 (2) AADDDDAAAABBBBBBBBBBAAAAAAAAAABBBBBBBBBB 211 SFL Marking of a subset of the packets for delay 213 (3) AAAAAAAADDDDBBBBBBBBAAAAAAAAAABBBBBBBBBB 215 SFL Marking of a subset of the packets across a 216 packet loss measurement boundary 218 (4) AACDCDCDAABBBBBBBBBBAAAAAAAAAABBBBBBBBBB 220 The case of multiple delay measurements within 221 a packet loss measurement 223 A & B are packets where loss is being measured 224 C & D are pacekts where loss and delay is being measured 226 Figure 1: RFC6734 Query Packet with SFL 228 In case 1 of Figure 1 we show the case where loss measurement alone 229 is being carried out on the flow under analysis. For illustrative 230 purposes consider that 10 packets are used in each flow in the time 231 interval being analyzed. 233 Now consider case 2 of Figure 1 where a small batch of packets need 234 to be analyzed for delay. These are marked with a different SFL type 235 indicating that they are to be monitored for both loss and delay. 236 The SFL=A indicates loss batch A, SFL=D indicates a batch of packets 237 that are to be instrumented for delay, but SFL D is synonymous with 238 SFL A, which in turn is synonymous with the underlying Forwarding 239 Equivalence Class (FEC). Thus, a packet marked D will be accumulated 240 into the A loss batch, into the delay statistics and will be 241 forwarded as normal. Whether the packet is actually counted twice 242 (for loss and delay) or whether the two counters are reconciled 243 during reporting is a local matter. 245 Now consider case 3 of Figure 1 where a small batch of packets are 246 marked for delay across a loss batch boundary. These packets need to 247 be considered as part of batch A or a part of batch B, and any 248 RFC6374 Query needs to take place after all the packets A or D 249 (whichever option is chosen) have arrived at the receiving LSR. 251 Now consider case 4 of Figure 1. Here we have a case where it is 252 required to take a number of delay measurements within a batch of 253 packets that we are measuring for loss. To do this we need two SFLs 254 for delay (C and D) and alternate between them (on a delay batch by 255 delay batch basis) for the purposes of measuring the delay 256 characteristics of the different batches of packets. 258 6. Some Simplifying Rules 260 It is possible to construct a large set of overlapping measurement 261 types, in terms of loss, delay, loss and delay and batch overlap. If 262 we allow all combinations of cases, this leads to configuration, 263 testing and implementation complexity and hence increased costs. The 264 following simplifying rules represent the default case: 266 1. Any system that needs to measure delay MUST be able to measure 267 loss. 269 2. Any system that is to measure delay MUST be configured to measure 270 loss. Whether the loss statistics are collected or not is a 271 local matter. 273 3. A delay measurement MAY start at any point during a loss 274 measurement batch, subject to rule 4. 276 4. A delay measurement interval MUST be short enough that it will 277 complete before the enclosing loss batch completes. 279 5. The duration of a second delay (D in Figure 1 batch must be such 280 that all packets from the packets belonging to a first delay 281 batch (C in Figure 1)will have been received before the second 282 delay batch completes. This condition is satisfied when the time 283 to send a batch is long compared to the network propagation time, 284 and is a parameter that can be established by the network 285 operator. 287 Given that the sender controls both the start and duration of a loss 288 and a delay packet batch, these rules are readily implemented in the 289 control plane. 291 7. Multiple Packet Delay Characteristics 293 A number of methods are described which add to the set of 294 measurements originally specified in [RFC6374]. Each of these 295 methods has different characteristics and different processing 296 demands on the packet forwarder. The choice of method will depend on 297 the type of diagnostic that the operator seeks. 299 Three Methods are discussed: 301 1. Time Buckets 303 2. Classic Standard Deviation 305 3. Average Delay 307 7.1. Method 1: Time Buckets 309 In this method the receiving LSR measures the inter-packet gap, 310 classifies the delay into a number of delay buckets and records the 311 number of packets in each bucket. As an example, if the operator 312 were concerned about packets with a delay of up to 1us, 2us, 4us, 313 8us, and over 8us then there would be five buckets and packets that 314 arrived up to 1us would cause the 1us bucket counter to increase, 315 between 1us and 2us the 2us bucket counter would increase etc. In 316 practice it might be better in terms of processing and potential 317 parallelism if, when a packet had a delay relative to its predecessor 318 of 2us, then both the up to 1us and the 2us counter were incremented, 319 and any more detailed information was calculated in the analytics 320 system. 322 This method allows the operator to see more structure in the jitter 323 characteristics than simply measuring the average jitter, and avoids 324 the complication of needing to perform a per packet multiply, but 325 will probably need the time intervals between buckets to be 326 programmable by the operator. 328 The packet format of a Time Bucket Jitter Measurement Message is 329 shown below: 331 0 1 2 3 332 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 1 333 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 334 |Version| Flags | Control Code | Message Length | 335 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 336 | QTF | RTF | RPTF | Reserved | 337 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 338 | Session Identifier | DS | 339 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 340 | Number of | Reserved 1 | 341 | Buckets | | 342 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 343 | Interval in 10ns units | 344 | | 345 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 346 | Number pkts in Bucket | 347 | | 348 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 349 ~ ~ 350 ~ ~ 351 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 352 ~ ~ 353 ~ TLV Block ~ 354 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 356 Figure 2: Time Bucket Jitter Measurement Message Format 358 The Version, Flags, Control Code, Message Length, QTF, RTF, RPTF, 359 Session Identifier, Reserved and DS Fields are as defined in section 360 3.2 of RFC6374. The remaining fields, which are unsigned integers, 361 are as follows: 363 o Number of Buckets in the measurement 365 o Reserved 1 must be sent as zero and ignored on receipt 367 o Interval in 10ns units is the inter-packet interval for 368 this bucket 370 o Number Pkts in Bucket is the number of packets found in 371 this bucket. 373 There will be a number of Interval/Number pairs depending on the 374 number of buckets being specified by the Querier. If an RFC6374 375 message is being used to configure the buckets, (i.e. the responder 376 is creating or modifying the buckets according to the intervals in 377 the Query message), then the Responder MUST respond with 0 packets in 378 each bucket until it has been configured for a full measurement 379 period. This indicates that it was configured at the time of the 380 last response message, and thus the response is valid for the whole 381 interval. As per the [RFC6374] convention the Number of pkts in 382 Bucket fields are included in the Query message and set to zero. 384 Out of band configuration is permitted by this mode of operation. 386 Note this is a departure from the normal fixed format used in 387 RFC6374. 389 The time bucket jitter measurement message is carried over an LSP in 390 the way described in [RFC6374] and over an LSP with an SFL as 391 described in Section 9. 393 7.2. Method 2 Classic Standard Deviation 395 In this method, provision is made for reporting the following delay 396 characteristics: 398 1. Number of packets in the batch (n). 400 2. Sum of delays in a batch (S) 402 3. Maximum Delay. 404 4. Minimum Delay. 406 5. Sum of squares of Inter-packet delay (SS). 408 Characteristics 1 and 2 give the mean delay. Measuring the delay of 409 each pair in the batch is discussed in Section 7.3. 411 Characteristics 3 and 4 give the outliers. 413 Characteristics 1, 2 and 5 can be used to calculate the variance of 414 the inter-packet gap and hence the standard deviation giving a view 415 of the distribution of packet delays and hence the jitter. The 416 equation for the variance (var) is given by: 418 var = (SS - S*S/n)/(n-1) 420 There is some concern over the use of this algorithm for measuring 421 variance, because SS and S*S/n can be similar numbers, particularly 422 where variance is low. However the method commends it self by not 423 requiring a division in the hardware. 425 7.2.1. Multi-Packet Delay Measurement Message Format 427 The packet format of a Multi-Packet Delay Measurement Message is 428 shown below: 430 0 1 2 3 431 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 1 432 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 433 |Version| Flags | Control Code | Message Length | 434 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 435 | QTF | RTF | RPTF | Reserved | 436 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 437 | Session Identifier | DS | 438 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 439 | Number of Packets | 440 | | 441 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 442 | Sum of Delays for Batch | 443 | | 444 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 445 | Minimum Delay | 446 | | 447 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 448 | Maximum Delay | 449 | | 450 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 451 | Sum of squares of Inter-packet delay | 452 | | 453 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 454 ~ ~ 455 ~ TLV Block ~ 456 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 458 Figure 3: Multi-packet Delay Measurement Message Format 460 The Version, Flags, Control Code, Message Length, QTF, RTF, RPTF, 461 Session Identifier, Reserved and DS Fields are as defined in section 462 3.2 of RFC6374. The remaining fields are as follows: 464 o Number of Packets is the number of packets in this batch 466 o Sum of Delays for Batch is the duration of the batch in the 467 time measurement format specified in the RTF field. 469 o Minimum Delay is the minimum inter-packet gap observed during 470 the batch in the time format specified in the RTF field. 472 o Maximum Delay is the maximum inter-packet gap observed during 473 the batch in the time format specified in the RTF field. 475 The multi-packet delay measurement message is carried over an LSP in 476 the way described in [RFC6374] and over an LSP with an SFL as 477 described in Section 9. 479 7.3. Per Packet Delay Measurement 481 If detailed packet delay measurement is required then it might be 482 possible to record the inter-packet gap for each packet pair. In 483 other than exception cases of slow flows or small batch sizes, this 484 would create a large (per packet) demand on storage in the 485 instrumentation system, a large bandwidth to such a storage system 486 and large bandwidth to the analytics system. Such a measurement 487 technique is outside the scope of this document. 489 7.4. Average Delay 491 Introduced in [RFC8321] is the concept of a one way delay measurement 492 in which the average time of arrival of a set of packets is measured. 493 In this approach the packet is time-stamped at arrival and the 494 Responder returns the sum of the time-stamps and the number of times- 495 tamps. From this the analytics engine can determine the mean delay. 496 An alternative model is that the Responder returns the time stamp of 497 the first and last packet and the number of packets. This later 498 method has the advantage of allowing the average delay to be 499 determined at a number of points along the packet path and allowing 500 the components of the delay to be characterized. Unless specifically 501 configured otherwise, the responder may return either or both types 502 of response and the analytics engine should process the response 503 appropriately. 505 The packet format of an Average Delay Measurement Message is shown 506 below: 508 0 1 2 3 509 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 1 510 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 511 |Version| Flags | Control Code | Message Length | 512 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 513 | QTF | RTF | RPTF | Reserved | 514 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 515 | Session Identifier | DS | 516 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 517 | Number of Packets | 518 | | 519 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 520 | Time of First Packet | 521 | | 522 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 523 | Time of Last Packet | 524 | | 525 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 526 | Sum of Timestamps of Batch | 527 | | 528 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 530 ~ ~ 531 ~ TLV Block ~ 532 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 534 Figure 4: Average Delay Measurement Message Format 536 The Version, Flags, Control Code, Message Length, QTF, RTF, RPTF, 537 Session Identifier, and DS Fields are as defined in section 3.2 of 538 RFC6374. The remaining fields are as follows: 540 o Number of Packets is the number of packets in this batch. 542 o Time of First Packet is the time of arrival of the first 543 packet in the batch. 545 o Time of Last Packet is the time of arrival of the last 546 packet in the batch. 548 o Sum of Timestamps of Batch. 550 The average delay measurement message is carried over an LSP in the 551 way described in [RFC6374] and over an LSP with an SFL as described 552 in Section 9. As is the convention with RFC6374, the Query message 553 contains placeholders for the Response message. The placeholders are 554 sent as zero. 556 8. Sampled Measurement 558 In the discussion so far it has been assumed that we would measure 559 the delay characteristics of every packet in a delay measurement 560 interval defined by an SFL of constant color. In [RFC8321] the 561 concept of a sampled measurement is considered. That is the 562 Responder only measures a packet at the start of a group of packets 563 being marked for delay measurement by a particular color, rather than 564 every packet in the marked batch. A measurement interval is not 565 defined by the duration of a marked batch of packets but the interval 566 between a pair of RFC6374 packets taking a readout of the delay 567 characteristic. This approach has the advantage that the measurement 568 is not impacted by ECMP effects. 570 This sampled approach may be used if supported by the Responder and 571 configured by the opertor. 573 9. Carrying RFC6374 Packets over an LSP using an SFL 575 We illustrate the packet format of an RFC6374 Query message using 576 SFLs for the case of an MPLS direct loss measurement in Figure 5. 578 +-------------------------------+ 579 | | 580 | LSP | 581 | Label | 582 +-------------------------------+ 583 | | 584 | Synonymous Flow | 585 | Label | 586 +-------------------------------+ 587 | | 588 | GAL | 589 | | 590 +-------------------------------+ 591 | | 592 | ACH Type = 0xA | 593 | | 594 +-------------------------------+ 595 | | 596 | RFC6374 Measurement Message | 597 | | 598 | +-------------------------+ | 599 | | | | 600 | | Fixed-format | | 601 | | portion of msg | | 602 | | | | 603 | +-------------------------+ | 604 | | | | 605 | | Optional SFL TLV | | 606 | | | | 607 | +-------------------------+ | 608 | | | | 609 | | Optional Return | | 610 | | Information | | 611 | | | | 612 | +-------------------------+ | 613 | | 614 +-------------------------------+ 616 Figure 5: RFC6734 Query Packet with SFL 618 The MPLS label stack is exactly the same as that used for the user 619 data service packets being instrumented except for the inclusion of 620 the Generic Associated Channel Label (GAL) [RFC5586] to allow the 621 receiver to distinguish between normal data packets and OAM packets. 622 Since the packet loss measurements are being made on the data service 623 packets, an RFC6374 direct loss measurement is being made, and which 624 is indicated by the type field in the ACH (Type = 0x000A). 626 The RFC6374 measurement message consists of the three components, the 627 RFC6374 fixed-format portion of the message as specified in [RFC6374] 628 carried over the ACH channel type specified the type of measurement 629 being made (currently: loss, delay or loss and delay) as specified in 630 RFC6374. 632 Two optional TLVs MAY also be carried if needed. The first is the 633 SFL TLV specified in Section 9.1. This is used to provide the 634 implementation with a reminder of the SFL that was used to carry the 635 RFC6374 message. This is needed because a number of MPLS 636 implementations do not provide the MPLS label stack to the MPLS OAM 637 handler. This TLV is required if RFC6374 messages are sent over UDP 638 [RFC7876]. This TLV MUST be included unless, by some method outside 639 the scope of this document, it is known that this information is not 640 needed by the RFC6374 Responder. 642 The second set of information that may be needed is the return 643 information that allows the responder send the RFC6374 response to 644 the Querier. This is not needed if the response is requested in-band 645 and the MPLS construct being measured is a point to point LSP, but 646 otherwise MUST be carried. The return address TLV is defined in 647 [RFC6374] and the optional UDP Return Object is defined in [RFC7876]. 649 Where a measurement other than an MPLS direct loss measurement is to 650 be made, the appropriate RFC6374 measurement message is used (for 651 example, one of the new types defined in this document) and this is 652 indicated to the receiver by the use of the corresponding ACH type. 654 9.1. RFC6374 SFL TLV 656 The RFC6374 SFL TLV is shown in Figure 6. This contains the SFL that 657 was carried in the label stack, the FEC that was used to allocate the 658 SFL and the index into the batch of SLs that were allocated for the 659 FEC that corresponds to this SFL. 661 0 1 2 3 662 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 1 663 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 664 | Type | Length |MBZ| SFL Batch | SFL Index | 665 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 666 | SFL | Reserved | 667 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 668 | FEC | 669 . . 670 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 672 Figure 6: SFL TLV 674 Where: 676 Type Type is set to Synonymous Flow Label (SFL-TLV). 678 Length The length of the TLV as specified in RFC6374. 680 MBZ MUST be sent as zero and ignored on receive. 682 SFL Batch The SFL batch that this SFL was allocated as part 683 of see [I-D.bryant-mpls-sfl-control] 685 SPL Index The index into the list of SFLs that were assigned 686 against the FEC that corresponds to the SFL. 688 Multiple SFLs can be assigned to a FEC each 689 with different actions. This index is an optional 690 convenience for use in mapping between the TLV 691 and the associated data structures in the LSRs. 692 The use of this feature is agreed between the 693 two parties during configuration. It is not required, 694 but is a convenience for the receiver if both parties 695 support the facility, 697 SFL The SFL used to deliver this packet. This is an MPLS 698 label which is a component of a label stack entry as 699 defined in Section 2.1 of [RFC3032]. 701 Reserved MUST be sent as zero and ignored on receive. 703 FEC The Forwarding Equivalence Class that was used to 704 request this SFL. This is encoded as per 705 Section 3.4.1 of [RFC5036] 707 This information is needed to allow for operation with hardware that 708 discards the MPLS label stack before passing the remainder of the 709 stack to the OAM handler. By providing both the SFL and the FEC plus 710 index into the array of allocated SFLs a number of implementation 711 types are supported. 713 10. RFC6374 Combined Loss-Delay Measurement 715 This mode of operation is not currently supported by this 716 specification. 718 11. Privacy Considerations 720 The inclusion of originating and/or flow information in a packet 721 provides more identity information and hence potentially degrades the 722 privacy of the communication. Whilst the inclusion of the additional 723 granularity does allow greater insight into the flow characteristics 724 it does not specifically identify which node originated the packet 725 other than by inspection of the network at the point of ingress, or 726 inspection of the control protocol packets. This privacy threat may 727 be mitigated by encrypting the control protocol packets, regularly 728 changing the synonymous labels and by concurrently using a number of 729 such labels. 731 12. Security Considerations 733 The security considerations documented in [RFC6374] and [RFC8372] 734 (which in turn calls up [RFC7258] and [RFC5920]) are applicable to 735 this protocol. 737 The issue noted in Section 11 is a security consideration. There are 738 no other new security issues associated with the MPLS dataplane. Any 739 control protocol used to request SFLs will need to ensure the 740 legitimacy of the request. 742 An attacker that manages to corrupt the RFC6374 SFL TLV Section 9.1 743 could disrupt the measurements in a way that the RFC6374 responder is 744 unable to detect. However, the network opertator is likely to notice 745 the anomalous network performance measurements, and in any case 746 normal MPLS network security proceedures make this type of attack 747 extremely unlikley. 749 13. IANA Considerations 751 13.1. Allocation of MPLS Generalized Associated Channel (G-ACh) Types 753 As per the IANA considerations in [RFC5586] updated by [RFC7026] and 754 [RFC7214], IANA is requested to allocate the following codeponts in 755 the "MPLS Generalized Associated Channel (G-ACh) Type" registry, in 756 the "Generic Associated Channel (G-ACh) Parameters" name space: 758 Value Description Reference 759 ----- --------------------------------- ----------- 760 TBD RFC6374 Time Bucket Jitter Measurement This 762 TBD RFC6374 Multi-Packet Delay This 763 Measurement 765 TBD RFC6374 Average Delay Measurement This 767 13.2. Allocation of MPLS Loss/Delay TLV Object 769 IANA is requested to allocate a new TLV from the 0-127 range of the 770 MPLS Loss/Delay Measurement TLV Object Registry in the "Generic 771 Associated Channel (G-ACh) Parameters" namespace: 773 Type Description Reference 774 ---- --------------------------------- --------- 775 TBD Synonymous Flow Label This 777 A value of 4 is recommended. 779 RFC Editor please delete this para 780 [RFC3032][I-D.bryant-mpls-sfl-control][RFC5036] 782 14. Acknowledgments 784 The authors thank Benjamin Kaduk and Elwyn Davies for their thorough 785 and thoughtful review of this document. 787 15. Contributing Authors 789 Zhenbin Li 790 Huawei 791 Email: lizhenbin@huawei.com 793 Siva Sivabalan 794 Ciena Corporation 795 Email: ssivabal@ciena.com 797 16. References 799 16.1. Normative References 801 [I-D.bryant-mpls-sfl-control] 802 Bryant, S., Swallow, G., and S. Sivabalan, "A Simple 803 Control Protocol for MPLS SFLs", draft-bryant-mpls-sfl- 804 control-09 (work in progress), December 2020. 806 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 807 Requirement Levels", BCP 14, RFC 2119, 808 DOI 10.17487/RFC2119, March 1997, 809 . 811 [RFC3032] Rosen, E., Tappan, D., Fedorkow, G., Rekhter, Y., 812 Farinacci, D., Li, T., and A. Conta, "MPLS Label Stack 813 Encoding", RFC 3032, DOI 10.17487/RFC3032, January 2001, 814 . 816 [RFC5036] Andersson, L., Ed., Minei, I., Ed., and B. Thomas, Ed., 817 "LDP Specification", RFC 5036, DOI 10.17487/RFC5036, 818 October 2007, . 820 [RFC5586] Bocci, M., Ed., Vigoureux, M., Ed., and S. Bryant, Ed., 821 "MPLS Generic Associated Channel", RFC 5586, 822 DOI 10.17487/RFC5586, June 2009, 823 . 825 [RFC6374] Frost, D. and S. Bryant, "Packet Loss and Delay 826 Measurement for MPLS Networks", RFC 6374, 827 DOI 10.17487/RFC6374, September 2011, 828 . 830 [RFC7026] Farrel, A. and S. Bryant, "Retiring TLVs from the 831 Associated Channel Header of the MPLS Generic Associated 832 Channel", RFC 7026, DOI 10.17487/RFC7026, September 2013, 833 . 835 [RFC7214] Andersson, L. and C. Pignataro, "Moving Generic Associated 836 Channel (G-ACh) IANA Registries to a New Registry", 837 RFC 7214, DOI 10.17487/RFC7214, May 2014, 838 . 840 [RFC7876] Bryant, S., Sivabalan, S., and S. Soni, "UDP Return Path 841 for Packet Loss and Delay Measurement for MPLS Networks", 842 RFC 7876, DOI 10.17487/RFC7876, July 2016, 843 . 845 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 846 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 847 May 2017, . 849 [RFC8957] Bryant, S., Chen, M., Swallow, G., Sivabalan, S., and G. 850 Mirsky, "Synonymous Flow Label Framework", RFC 8957, 851 DOI 10.17487/RFC8957, January 2021, 852 . 854 16.2. Informative References 856 [RFC3270] Le Faucheur, F., Wu, L., Davie, B., Davari, S., Vaananen, 857 P., Krishnan, R., Cheval, P., and J. Heinanen, "Multi- 858 Protocol Label Switching (MPLS) Support of Differentiated 859 Services", RFC 3270, DOI 10.17487/RFC3270, May 2002, 860 . 862 [RFC5920] Fang, L., Ed., "Security Framework for MPLS and GMPLS 863 Networks", RFC 5920, DOI 10.17487/RFC5920, July 2010, 864 . 866 [RFC5921] Bocci, M., Ed., Bryant, S., Ed., Frost, D., Ed., Levrau, 867 L., and L. Berger, "A Framework for MPLS in Transport 868 Networks", RFC 5921, DOI 10.17487/RFC5921, July 2010, 869 . 871 [RFC7190] Villamizar, C., "Use of Multipath with MPLS and MPLS 872 Transport Profile (MPLS-TP)", RFC 7190, 873 DOI 10.17487/RFC7190, March 2014, 874 . 876 [RFC7258] Farrell, S. and H. Tschofenig, "Pervasive Monitoring Is an 877 Attack", BCP 188, RFC 7258, DOI 10.17487/RFC7258, May 878 2014, . 880 [RFC8321] Fioccola, G., Ed., Capello, A., Cociglio, M., Castaldelli, 881 L., Chen, M., Zheng, L., Mirsky, G., and T. Mizrahi, 882 "Alternate-Marking Method for Passive and Hybrid 883 Performance Monitoring", RFC 8321, DOI 10.17487/RFC8321, 884 January 2018, . 886 [RFC8372] Bryant, S., Pignataro, C., Chen, M., Li, Z., and G. 887 Mirsky, "MPLS Flow Identification Considerations", 888 RFC 8372, DOI 10.17487/RFC8372, May 2018, 889 . 891 Authors' Addresses 893 Stewart Bryant 894 Futurewei Technologies Inc. 896 Email: sb@stewartbryant.com 897 George Swallow 898 Southend Technical Center 900 Email: swallow.ietf@gmail.com 902 Mach Chen 903 Huawei 905 Email: mach.chen@huawei.com 907 Giuseppe Fioccola 908 Huawei Technologies 910 Email: giuseppe.fioccola@huawei.com 912 Gregory Mirsky 913 ZTE Corp. 915 Email: gregimirsky@gmail.com