idnits 2.17.1 draft-ietf-mpls-flow-ident-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 date (August 25, 2016) is 2800 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- 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 3 Internet-Draft M. Chen 4 Intended status: Informational Z. Li 5 Expires: February 26, 2017 Huawei 6 C. Pignataro 7 Cisco Systems 8 G. Mirsky 9 Ericsson 10 August 25, 2016 12 MPLS Flow Identification Considerations 13 draft-ietf-mpls-flow-ident-02 15 Abstract 17 This memo discusses the desired capabilities for MPLS flow 18 identification. The key application that needs this is in-band 19 performance monitoring of user data packets. 21 Status of This Memo 23 This Internet-Draft is submitted in full conformance with the 24 provisions of BCP 78 and BCP 79. 26 Internet-Drafts are working documents of the Internet Engineering 27 Task Force (IETF). Note that other groups may also distribute 28 working documents as Internet-Drafts. The list of current Internet- 29 Drafts is at http://datatracker.ietf.org/drafts/current/. 31 Internet-Drafts are draft documents valid for a maximum of six months 32 and may be updated, replaced, or obsoleted by other documents at any 33 time. It is inappropriate to use Internet-Drafts as reference 34 material or to cite them other than as "work in progress." 36 This Internet-Draft will expire on February 26, 2017. 38 Copyright Notice 40 Copyright (c) 2016 IETF Trust and the persons identified as the 41 document authors. All rights reserved. 43 This document is subject to BCP 78 and the IETF Trust's Legal 44 Provisions Relating to IETF Documents 45 (http://trustee.ietf.org/license-info) in effect on the date of 46 publication of this document. Please review these documents 47 carefully, as they describe your rights and restrictions with respect 48 to this document. Code Components extracted from this document must 49 include Simplified BSD License text as described in Section 4.e of 50 the Trust Legal Provisions and are provided without warranty as 51 described in the Simplified BSD License. 53 Table of Contents 55 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 56 2. Requirements Language . . . . . . . . . . . . . . . . . . . . 3 57 3. Loss Measurement Considerations . . . . . . . . . . . . . . . 3 58 4. Delay Measurement Considerations . . . . . . . . . . . . . . 4 59 5. Units of identification . . . . . . . . . . . . . . . . . . . 4 60 6. Types of LSP . . . . . . . . . . . . . . . . . . . . . . . . 5 61 7. Network Scope . . . . . . . . . . . . . . . . . . . . . . . . 6 62 8. Backwards Compatibility . . . . . . . . . . . . . . . . . . . 7 63 9. Dataplane . . . . . . . . . . . . . . . . . . . . . . . . . . 7 64 10. Control Plane . . . . . . . . . . . . . . . . . . . . . . . . 8 65 11. Privacy Considerations . . . . . . . . . . . . . . . . . . . 8 66 12. Security Considerations . . . . . . . . . . . . . . . . . . . 9 67 13. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9 68 14. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 9 69 15. References . . . . . . . . . . . . . . . . . . . . . . . . . 9 70 15.1. Normative References . . . . . . . . . . . . . . . . . . 9 71 15.2. Informative References . . . . . . . . . . . . . . . . . 9 72 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 10 74 1. Introduction 76 This memo discusses the desired capabilities for MPLS flow 77 identification. The key application that needs this is in-band 78 performance monitoring of user data packets. 80 There is a need to identify flows in MPLS networks for applications 81 such as packet loss and packet delay measurement. A method of loss 82 and delay measurement in MPLS networks was defined in [RFC6374]. 83 When used to measure packet loss [RFC6374] depends on the use of the 84 injected Operations, Administration, and Maintenance (OAM) packets to 85 designate the beginning and the end of the packet group over which 86 packet loss is being measured. Where the misordering of packets from 87 one group relative to the following group, or misordering of one of 88 the packets being counted relative to the [RFC6374] packet occurs, 89 then an error will occur in the packet loss measurement. In 90 addition, this packet performance measurement system needs to be 91 extended to deal with different granularities of flow and to address 92 a number of the multi-point cases in which two or more ingress Label 93 Switching Routers (LSRs) could send packets to one or more 94 destinations. 96 Improvements in link and transmission technologies mean that it may 97 be difficult to assess packet loss using active performance 98 measurement methods with synthetic traffic, due to the very low loss 99 rate in normal operation. That, together with more demanding service 100 level requirements, mean that network operators need to be able to 101 measure the loss of the actual user data traffic by using passive 102 performance measurement methods. Any technique deployed needs to be 103 transparent to the end user, and it needs to be assumed that they 104 will not take any active part in the measurement process. Indeed it 105 is important that any flow identification technique be invisible to 106 them and that no remnant of the identification of measurement process 107 leak into their network. 109 Additionally where there are multiple traffic sources, such as in 110 multi-point to point and multi-point to multi-point network 111 environments there needs to be a method whereby the sink can 112 distinguish between packets from the various sources, that is to say, 113 that a multi-point to multi-point measurement model needs to be 114 developed. 116 2. Requirements Language 118 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 119 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 120 document are to be interpreted as described in RFC2119 [RFC2119]. 122 3. Loss Measurement Considerations 124 Modern networks, if not oversubscribed, normally drop very few 125 packets, thus packet loss measurement is highly sensitive to counter 126 errors. Without some form of coloring or batch marking such as that 127 proposed in [I-D.tempia-ippm-p3m] it may not be possible to achieve 128 the required accuracy in the loss measurement of customer data 129 traffic. Thus where accuracy better than the data link loss 130 performance of a modern optical network is required, it may be 131 economically advantageous, or even a technical requirement, to 132 include some form of marking in the packets to assign each packet to 133 a particular counter. 135 Where this level of accuracy is required and the traffic between a 136 source-destination pair is subject to Equal-Cost Multipath (ECMP) a 137 demarcation mechanism is needed to group the packets into batches. 138 Once a batch is correlated at both ingress and egress, the packet 139 accounting mechanism is then able to operate on the batch of packets 140 which can be accounted for at both the packet ingress and the packet 141 egress. Errors in the accounting are particularly acute in Label 142 Switched Paths (LSPs) subjected to ECMP because the network transit 143 time will be different for the various ECMP paths since: 145 1. The packets may traverse different sets of LSRs. 147 2. The packets may depart from different interfaces on different 148 line cards on LSRs 150 3. The packets may arrive at different interfaces on different line 151 cards on LSRs. 153 A consideration in modifying the identity label (the MPLS label 154 ordinarily used to identify the LSP, Virtual Private Network, 155 Pseudowire etc) to indicate the batch is the impact that this has on 156 the path chosen by the ECMP mechanism. When the member of the ECMP 157 path set is chosen by deep packet inspection a change of batch 158 represented by a change of identity label will have no impact on the 159 ECMP path. Where the path member is chosen by reference to an 160 entropy label [RFC6790] then changing the batch identifier will not 161 result in a change to the chosen ECMP path. ECMP is so pervasive in 162 multi-point to (multi-) point networks that some method of avoiding 163 accounting errors introduced by ECMP needs to be supported. 165 4. Delay Measurement Considerations 167 Most of the existing delay measurement methods are active measurement 168 that depend on the extra injected test packet to evaluate the delay 169 of a path. With the active measurement method, the rate, numbers and 170 interval between the injected packets may affect the accuracy of the 171 results. Also, for injected test packets, these may not be co-routed 172 with the data traffic due to ECMP. Thus there exists a requirement 173 to measure the delay of the real traffic. 175 For combined loss-delay measurements, both the loss and the delay 176 considerations apply. 178 5. Units of identification 180 The most basic unit of identification is the identity of the node 181 that processed the packet on its entry to the MPLS network. However, 182 the required unit of identification may vary depending on the use 183 case for accounting, performance measurement or other types of packet 184 observations. In particular note that there may be a need to impose 185 identify at several different layers of the MPLS label stack. 187 This document considers following units of identifications: 189 o Per source LSR - everything from one source is aggregated. 191 o Per group of LSPs chosen by an ingress LSR - an ingress LSP 192 aggregates group of LSPs (ex: all LSPs of a tunnel). 194 o Per LSP - the basic form. 196 o Per flow [RFC6790] within an LSP - fine graining method. 198 Note that a fine grained identity resolution is needed when there is 199 a need to perform these operations on a flow not readily identified 200 by some other element in the label stack. Such fine grained 201 resolution may be possible by deep packet inspection, but this may 202 not always be possible, or it may be desired to minimise processing 203 costs by doing this only in entry to the network, and adding a 204 suitable identifier to the packet for reference by other network 205 elements. An example of such a fine grained case might be traffic 206 from a specific application, or from a specific application from a 207 specific source, particularly if matters related to service level 208 agreement or application performance were being investigated. 210 We can thus characterize the identification requirement in the 211 following broad terms: 213 o There needs to be some way for an egress LSR to identify the 214 ingress LSR with an appropriate degree of scope. This concept is 215 discussed further in Section 7. 217 o There needs to be a way to identify a specific LSP at the egress 218 node. This allows for the case of instrumenting multiple LSPs 219 operate between the same pair of nodes. In such cases the 220 identity of the ingress LSR is insufficient. 222 o In order to conserve resources such as labels, counters and/or 223 compute cycles it may be desirable to identify an LSP group so 224 that a operation can be performed on the group as an aggregate. 226 o There needs to be a way to identify a flow within an LSP. This is 227 necessary when investigating a specific flow that has been 228 aggregated into an LSP. 230 The unit of identification and the method of determining which 231 packets constitute a flow will be application or use-case specific 232 and is out of scope of this memo. 234 6. Types of LSP 236 We need to consider a number of types of LSP. The two simplest types 237 to monitor are point to point LSPs and point to multi-point LSPs. 238 The ingress LSR for a point to point LSP, such as those created using 239 the Resource Reservation Protocol - Traffic Engineering (RSVP-TE) 240 [RFC5420] signalling protocol, or those that conform to the MPLS 241 Transport Profile (MPLS-TP) [RFC5654] may be identified by inspection 242 of the top label in the stack, since at any provider-edge (PE) or 243 provider (P) router on the path this is unique to the ingress-egress 244 pair at every hop at a given layer in the LSP hierarchy. Provided 245 that penultimate hop popping is disabled, the identity of the ingress 246 LSR of a point to point LSP is available at the egress LSR and thus 247 determining the identity of the ingress LSR must be regarded as a 248 solved problem. Note however that the identity of a flow cannot to 249 be determined without further information being carried in the 250 packet, or gleaned from some aspect of the packet payload. 252 In the case of a point to multi-point LSP, and in the absence of 253 Penultimate Hop Popping (PHP) the identity of the ingress LSR may 254 also be inferred from the top label. However, it may not possible to 255 adequately identify the flow from the top label alone, and thus 256 further information may need to be carried in the packet, or gleaned 257 from some aspect of the packet payload. In designing any solution it 258 is desirable that a common flow identity solution be used for both 259 point to point and point to multi-point LSP types. Similarly it is 260 desirable that a common method of LSP group identification be used. 261 In the above cases, a context label [RFC5331] needs to be used to 262 provide the required identity information. This is widely supported 263 MPLS feature. 265 A more interesting case is the case of a multi-point to point LSP. 266 In this case the same label is normally used by multiple ingress or 267 upstream LSRs and hence source identification is not possible by 268 inspection of the top label by the egress LSRs. It is therefore 269 necessary for a packet to be able to explicitly convey any of the 270 identity types described in Section 5. 272 Similarly, in the case of a multi-point to multi-point LSP the same 273 label is normally used by multiple ingress or upstream LSRs and hence 274 source identification is not possible by inspection of the top label 275 by egress LSRs. The various types of identity described in Section 5 276 are again needed. Note however, that the scope of the identity may 277 be constrained to be unique within the set of multi-point to multi- 278 point LSPs terminating on any common node. 280 7. Network Scope 282 The scope of identification can be constrained to the set of flows 283 that are uniquely identifiable at an ingress LSR, or some aggregation 284 thereof. There is no question of an ingress LSR seeking assistance 285 from outside the MPLS protocol domain. 287 In any solution that constrains itself to carrying the required 288 identity in the MPLS label stack rather than in some different 289 associated data structure, constraints on the label stack size imply 290 that the scope of identity reside within that MPLS domain. For 291 similar reasons the identity scope of a component of an LSP should be 292 constrained to the scope of that LSP. 294 8. Backwards Compatibility 296 In any network it is unlikely that all LSRs will have the same 297 capability to support the methods of identification discussed in this 298 memo. It is therefore an important constraint on any flow identity 299 solution that it is backwards compatible with deployed MPLS equipment 300 to the extent that deploying the new feature will not disable 301 anything that currently works on a legacy equipment. 303 This is particularly the case when the deployment is incremental or 304 when the feature is not required for all LSRs or all LSPs. Thus in 305 broad the flow identification design MUST support the co-existence of 306 both LSRs that can, and cannot, identify the traffic components 307 described in Section 5. In addition the identification of the 308 traffic components described in Section 5 MUST be an optional feature 309 that is disabled by default. As a design simplification, a solution 310 MAY require that all egress LSRs of a point to multipoint or a multi- 311 point to multipoint LSP support the identification type in use so 312 that a single packet can be correctly processed by all egress 313 devices. The corollary of this last point is that either all egress 314 LSRs are enabled to support the required identity type, or none of 315 them are. 317 9. Dataplane 319 There is a huge installed base of MPLS equipment, typically this type 320 of equipment remains in service for an extended period of time, and 321 in many cases hardware constraints mean that it is not possible to 322 upgrade its dataplane functionality. Changes to the MPLS data plane 323 are therefore expensive to implement, add complexity to the network, 324 and may significantly impact the deployability of a solution that 325 requires such changes. For these reasons, the MPLS designers have 326 set a very high bar to changes to the MPLS data plane, and only a 327 very small number have been adopted. Hence, it is important that the 328 method of identification must minimize changes to the MPLS data 329 plane. Ideally method(s) of identification that require no changes 330 to the MPLS data plane should be given preferential consideration. 331 If a method of identification makes a change to the data plane is 332 chosen it will need to have a significant advantage over any method 333 that makes no change, and the advantage of the approach will need to 334 be carefully evaluated and documented. If a change is necessary to 335 the MPLS data plane proves necessary, it should be (a) be as small a 336 change as possible and (b) be a general purpose method so as to 337 maximise its use for future applications. It is imperative that, as 338 far as can be foreseen, any necessary change made to the MPLS data 339 plane does not impose any foreseeable future limitation on the MPLS 340 data plane. 342 Stack size is an issue with many MPLS implementations both as a 343 result of hardware limitations, and due to the impact on networks and 344 applications where a large number of small payloads need to be 345 transported In particular one MPLS payload may be carried inside 346 another. For example one LSP may be carried over another LSP, or a 347 PW or similar multiplexing construct may be carried over an LSP and 348 identification may be required at both layers. Of particular concern 349 is the implementation of low cost edge LSRs that for cost reasons 350 have a significant limit on the number of Label Stack Elements (LSEs) 351 that they can impose or dispose. Therefore, any method of identity 352 MUST NOT consume an excessive number of unique labels, and MUST NOT 353 result in an excessive increase in the size of the label stack. 355 The MPLS data plane design provides two types of special purpose 356 labels: the original 16 reserved labels and the much larger set of 357 special purpose labels defined in [RFC7274]. The original reserved 358 labels need one LSE, and the newer [RFC7274] special purpose labels 359 need two LSEs. Given the tiny number of original reserved labels, it 360 is core to the MPLS design philosophy that this scarce resource is 361 only used when it is absolutely necessary. Using a single LSE 362 reserved or special purpose label to encode flow identity thus 363 requires two stack entries, one for the reserved label and one for 364 the flow identity. The larger set of [RFC7274] labels requires two 365 labels stack entries for the special purpose label itself and hence a 366 total of three label stack entries to encode the flow identity. 368 The use of special purpose labels (SPL) [RFC7274] as part of a method 369 to encode the identity information therefore has a number of 370 undesirable implications for the data plane and hence whilst a 371 solution may use SPL(s), methods that do not require SPLs need to be 372 carefully considered. 374 10. Control Plane 376 Any flow identity design should both seek to minimise the complexity 377 of the control plane and should minimise the amount of label co- 378 ordination needed amongst LSRs. 380 11. Privacy Considerations 382 The inclusion of originating and/or flow information in a packet 383 provides more identity information and hence potentially degrades the 384 privacy of the communication. Recent IETF concerns on pervasive 385 monitoring [RFC7258] would lead it to prefer a solution that does not 386 degrade the privacy of user traffic below that of an MPLS network not 387 implementing the flow identification feature. The minimizing the 388 scope of the identity indication can be useful in minimizing the 389 observability of the flow characteristics. 391 12. Security Considerations 393 Any solution to the flow identification needs must not degrade the 394 security of the MPLS network below that of an equivalent network not 395 deploying the specified identity solution. Propagation of 396 identification information outside the MPLS network imposing it must 397 be disabled by default. Any solution should provide for the 398 restriction of the identity information to those components of the 399 network that need to know it. It is thus desirable to limit the 400 knowledge of the identify of an endpoint to only those LSRs that need 401 to participate in traffic flow. 403 13. IANA Considerations 405 This memo has no IANA considerations. 407 14. Acknowledgements 409 The authors thank Nobo Akiya (nobo@cisco.com), Nagendra Kumar Nainar 410 (naikumar@cisco.com) and George Swallow (swallow@cisco.com) for their 411 comments. 413 15. References 415 15.1. Normative References 417 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 418 Requirement Levels", BCP 14, RFC 2119, 419 DOI 10.17487/RFC2119, March 1997, 420 . 422 15.2. Informative References 424 [I-D.tempia-ippm-p3m] 425 Capello, A., Cociglio, M., Fioccola, G., Castaldelli, L., 426 and A. Bonda, "A packet based method for passive 427 performance monitoring", draft-tempia-ippm-p3m-03 (work in 428 progress), March 2016. 430 [RFC5331] Aggarwal, R., Rekhter, Y., and E. Rosen, "MPLS Upstream 431 Label Assignment and Context-Specific Label Space", 432 RFC 5331, DOI 10.17487/RFC5331, August 2008, 433 . 435 [RFC5420] Farrel, A., Ed., Papadimitriou, D., Vasseur, JP., and A. 436 Ayyangarps, "Encoding of Attributes for MPLS LSP 437 Establishment Using Resource Reservation Protocol Traffic 438 Engineering (RSVP-TE)", RFC 5420, DOI 10.17487/RFC5420, 439 February 2009, . 441 [RFC5654] Niven-Jenkins, B., Ed., Brungard, D., Ed., Betts, M., Ed., 442 Sprecher, N., and S. Ueno, "Requirements of an MPLS 443 Transport Profile", RFC 5654, DOI 10.17487/RFC5654, 444 September 2009, . 446 [RFC6374] Frost, D. and S. Bryant, "Packet Loss and Delay 447 Measurement for MPLS Networks", RFC 6374, 448 DOI 10.17487/RFC6374, September 2011, 449 . 451 [RFC6790] Kompella, K., Drake, J., Amante, S., Henderickx, W., and 452 L. Yong, "The Use of Entropy Labels in MPLS Forwarding", 453 RFC 6790, DOI 10.17487/RFC6790, November 2012, 454 . 456 [RFC7258] Farrell, S. and H. Tschofenig, "Pervasive Monitoring Is an 457 Attack", BCP 188, RFC 7258, DOI 10.17487/RFC7258, May 458 2014, . 460 [RFC7274] Kompella, K., Andersson, L., and A. Farrel, "Allocating 461 and Retiring Special-Purpose MPLS Labels", RFC 7274, 462 DOI 10.17487/RFC7274, June 2014, 463 . 465 Authors' Addresses 467 Stewart Bryant 468 Huawei 470 Email: stewart.bryant@gmail.com 472 Mach Chen 473 Huawei 475 Email: mach.chen@huawei.com 477 Zhenbin Li 478 Huawei 480 Email: lizhenbin@huawei.com 481 Carlos Pignataro 482 Cisco Systems 484 Email: cpignata@cisco.com 486 Gregory Mirsky 487 Ericsson 489 Email: gregory.mirsky@eicsson.com