idnits 2.17.1 draft-komolafe-pim-igmp-mld-survey-report-00.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 : ---------------------------------------------------------------------------- ** The document seems to lack a Security Considerations section. ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (September 29, 2020) is 1298 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- No issues found here. Summary: 2 errors (**), 0 flaws (~~), 1 warning (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 PIM Working Group O. Komolafe 3 Internet-Draft Arista Networks 4 Intended status: Informational September 29, 2020 5 Expires: April 2, 2021 7 IGMPv3 and MLDv2 Survey Report 8 draft-komolafe-pim-igmp-mld-survey-report-00 10 Abstract 12 The PIM WG intends to progress IGMPv3 and MLDv2 from Proposed 13 Standards to Internet Standards. The WG decided to conduct a survey 14 of operators, vendors and implementors of these and related protocols 15 to gather information about their implementation and deployment. 16 This document presents the results of the survey and briefly 17 summarizes the key findings. The survey indicates that there is 18 widespread deployment and usage of of IGMPv3 and MLDv2, with numerous 19 independent implementations interoperating successfully. No major 20 issues with either protocol were identified and, similarly, no major 21 unused features in the specifications were highlighted. These 22 findings suggest that IGMPv3 and MLDv2 are indeed ready for 23 progression to Internet Standards. 25 Status of This Memo 27 This Internet-Draft is submitted in full conformance with the 28 provisions of BCP 78 and BCP 79. 30 Internet-Drafts are working documents of the Internet Engineering 31 Task Force (IETF). Note that other groups may also distribute 32 working documents as Internet-Drafts. The list of current Internet- 33 Drafts is at https://datatracker.ietf.org/drafts/current/. 35 Internet-Drafts are draft documents valid for a maximum of six months 36 and may be updated, replaced, or obsoleted by other documents at any 37 time. It is inappropriate to use Internet-Drafts as reference 38 material or to cite them other than as "work in progress." 40 This Internet-Draft will expire on April 2, 2021. 42 Copyright Notice 44 Copyright (c) 2020 IETF Trust and the persons identified as the 45 document authors. All rights reserved. 47 This document is subject to BCP 78 and the IETF Trust's Legal 48 Provisions Relating to IETF Documents 49 (https://trustee.ietf.org/license-info) in effect on the date of 50 publication of this document. Please review these documents 51 carefully, as they describe your rights and restrictions with respect 52 to this document. Code Components extracted from this document must 53 include Simplified BSD License text as described in Section 4.e of 54 the Trust Legal Provisions and are provided without warranty as 55 described in the Simplified BSD License. 57 Table of Contents 59 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 60 2. Approach . . . . . . . . . . . . . . . . . . . . . . . . . . 3 61 2.1. Methodology . . . . . . . . . . . . . . . . . . . . . . . 3 62 3. Responses for Vendors or Host Implementors . . . . . . . . . 4 63 3.1. Protocols Implemented . . . . . . . . . . . . . . . . . . 4 64 3.2. Features Supported . . . . . . . . . . . . . . . . . . . 4 65 3.3. Issues Identified . . . . . . . . . . . . . . . . . . . . 4 66 3.4. Suggestions . . . . . . . . . . . . . . . . . . . . . . . 4 67 4. Responses for Network Operators . . . . . . . . . . . . . . . 5 68 4.1. Protocols Deployed . . . . . . . . . . . . . . . . . . . 5 69 4.2. Features Enabled . . . . . . . . . . . . . . . . . . . . 5 70 4.3. Interoperability Issues . . . . . . . . . . . . . . . . . 5 71 4.4. Fallback Mechanism . . . . . . . . . . . . . . . . . . . 5 72 4.5. Strengths and Weaknesses of IGMPv3 and MLDv2 . . . . . . 5 73 5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 6 74 6. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 6 75 7. References . . . . . . . . . . . . . . . . . . . . . . . . . 6 76 7.1. Normative References . . . . . . . . . . . . . . . . . . 6 77 7.2. Informative References . . . . . . . . . . . . . . . . . 7 78 Appendix A. Questionnaire . . . . . . . . . . . . . . . . . . . 7 79 A.1. Questionnaire for Vendors or Host Implementors . . . . . 7 80 A.1.1. Implementation Status . . . . . . . . . . . . . . . . 7 81 A.1.2. Implementation Specifics . . . . . . . . . . . . . . 7 82 A.1.3. Implementation Perspectives . . . . . . . . . . . . . 8 83 A.2. Questionnaire for Network Operators . . . . . . . . . . . 8 84 A.2.1. Deployment Status . . . . . . . . . . . . . . . . . . 8 85 A.2.2. Deployment Specifics . . . . . . . . . . . . . . . . 9 86 A.2.3. Deployment Perspectives . . . . . . . . . . . . . . . 10 87 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 10 89 1. Introduction 91 Internet Group Management Protocol Version 3 (IGMPv3) [RFC3376] and 92 Multicast Listener Discovery Version 2 (MLDv2) for IPv6 [RFC3810] are 93 currently Proposed Standards. Given the fact that multiple 94 independent implementations of these protocols exist and they have 95 been successfully and widely used operationally, the PIM WG is keen 96 to progress these protocols to Internet Standards. As such, it is 97 critical to establish if there are features specified in [RFC3376] 98 and [RFC3810] that have not been widely used and also to determine 99 any interoperability issues that have arisen from using the 100 protocols. 102 Following approach taken for PIM-SM, documented in [RFC7063], the PIM 103 WG has decided that conducting a comprehensive survey on 104 implementations and deployment of IGMPv3 and MLDv2 will provide 105 valuable information to facilitate their progression to Internet 106 Standard. 108 This document summarizes the findings of the survey. 110 2. Approach 112 2.1. Methodology 114 The raw survey questions are shown in Appendix A. In order to make 115 the submission and processing of responses as convenient as possible, 116 Tim Chown kindly formatted and posted the survey online using the 117 JISC online surveys tool. The PIM WG chairs subsequently announced 118 the survey, publicizing the URL at which the survey could be 119 completed. In addition to announcing the survey on the relevant IETF 120 WG mailing lists, effort was made to distribute the survey to other 121 forums such as NANOG. 123 The survey was targeted at: 125 Network operators 127 Router vendors 129 Switch vendors 131 Host implementors 133 Once the deadline for the survey elapsed, Tim Chown collated the 134 responses, anonymizing the data so the responses from a specific 135 operator, vendor or implementor could not be identified. 137 The questions targeted at vendors or host implementors were answered 138 by 10 respondents. The network operators questions were answered by 139 14 respondents. (These numbers are comparable with the number of 140 responses to the PIM-SM survey [RFC7063].) 142 3. Responses for Vendors or Host Implementors 144 3.1. Protocols Implemented 146 80% or more of the respondents had implemented each of IGMPv1, 147 IGMPv2, IGMPv3, MLDv1 and MLDv2, with IGMPv3 being the only protocol 148 that had been implemented by all the respondents. In contrast, 149 Lightweight IGMPv3 and Lightweight MLDv2 had been implemented by only 150 20% of the respondents. 152 3.2. Features Supported 154 All the respondents supported source filtering with include list. 155 Snooping querier was also a popular feature, with 80% of respondents 156 supporting it. Source filtering with exclude list, snooping proxy, 157 snooping filtering, L2 report flooding, host proxy were moderately 158 popular, with 40%-70% of respondents supporting each of these 159 features. Unicast queries/reports were supported by only 20% of the 160 respondents. 162 3.3. Issues Identified 164 No ambiguities or inconsistencies in [RFC3376] and [RFC3810] that 165 made the implementation challenging were identified by any 166 respondent. 168 3.4. Suggestions 170 A number of respondents made suggestions to the PIM WG regarding 171 progressing IGMPv3 and MLDv2 to full standards: 173 o Add source discovery mechanism to SSM in addition to existing 174 application-based source discovery 176 o Improve scalability of query/response messages 178 o Deprecate older versions and streamline IGMPv3 180 o Allow reports to be sent without a querier 182 o Remove source filtering with exclude list as it is not widely used 183 and makes state machine unnecessarily complicated 185 Each of these points was raised by a different respondent, apart from 186 the last point which was raised by two separate respondents. 188 4. Responses for Network Operators 190 4.1. Protocols Deployed 192 IGMPv2 was the most widely deployed protocol, with 86% of respondents 193 indicating it is running in their network. Next was IGMPv3 with 79% 194 of respondents indicating it is deployed. However, between only 20% 195 and 36% of respondents indicated they had deployed IGMPv1, MLDv1 and 196 MLDv2. Lightweight IGMPv3 and Lightweight MLDv2 were undeployed. 198 4.2. Features Enabled 200 Between 20% and 30% of respondents indicated that had enabled Source 201 filtering with include list, source filtering with exclude list, 202 snooping querier, snooping filtering or unicast queries/reports. 203 Snooping proxy and L2 report flooding were enabled by 7% of 204 respondents. No respondent was using host proxy. 206 4.3. Interoperability Issues 208 Half the respondents indicated they were using equipment with multi- 209 vendor implementations in their network. No interoperability issues 210 were identified. 212 4.4. Fallback Mechanism 214 36% of respondents indicated there are dependent on the fallback 215 mechanisms between the different protocol versions. 7% of 216 respondents have experienced issues related to this fallback 217 mechanism. 219 4.5. Strengths and Weaknesses of IGMPv3 and MLDv2 221 A respondent indicated that a significant strength of IGMPv3 was the 222 simplicity introduced by using SSM, avoiding the complexities 223 associated with ASM. The weaknesses associated with IGMPv3 which 224 were identified were: 226 o No CPE implementations 228 o Automatic fallback makes deployments challenging 230 o ASM provides better source filtering (by potentially restricting 231 the acceptance of register messages at the RP) whereas SSM allows 232 only data plane filtering using multicast boundary 234 5. Conclusions 236 There were a total of 24 respondents to the survey which asked 237 vendors/implementors and network operators questions about IGMPv1, 238 IGMPv2, IGMPv3, Lightweight IGMPv3, MLDv1, MLDv2 and Lightweight 239 MLDv2. A reasonable number of responses were gathered to the survey, 240 allowing some interesting observations to be made. Firstly, and 241 perhaps unsurprisingly, operators use a lower number of protocols and 242 protocol features than have been implemented. Furthermore, there is 243 a relatively lower deployment of the different MLD versions, 244 suggesting that IPv6 multicast is less widely used than IPv4 245 multicast. No major flaws, inconsistencies or ambiguity in the 246 IGMPv3 [RFC3376] and MLDv2 [RFC3810] specifications were identified. 247 However, a number of issues were raised about the usage of these 248 protocols, notably concerns about the automatic fallback from IGMPv3 249 to IGMPv2 being sometimes problematic and the loss of certain useful 250 features offered by the ASM control plane with the transition to SSM. 252 These findings suggest that IGMPv3 and MLDv2 are indeed ready for 253 progression to Internet Standards. 255 6. Acknowledgements 257 The authors are grateful to Tim Chown for posting the survey online, 258 and for collating and anonymizing the responses. 260 7. References 262 7.1. Normative References 264 [RFC1112] Deering, S., "Host Extensions for IP Multicasting", 265 RFC 1112, August 1989. 267 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 268 2", RFC 2236, November 1997. 270 [RFC3376] Cain, B., Deering, S., Kouvelas, I., Fenner, B., and A. 271 Thyagarajan, "Internet Group Management Protocol, Version 272 3", RFC 3376, October 2002. 274 [RFC2710] Deering, S., Fenner, W., and B. Haberman, "Multicast 275 Listener Discovery (MLD) for IPv6", RFC 2710, October 276 1999. 278 [RFC3810] Vida, R. and L. Costa, "Multicast Listener Discovery 279 Version 2 (MLDv2) for IPv6", RFC 3810, June 2004. 281 [RFC5790] Liu, H., Cao, W., and H. Asaeda, "Lightweight Internet 282 Group Management Protocol Version 3 (IGMPv3) and Multicast 283 Listener Discovery Version 2 (MLDv2) Protocols", RFC 5790, 284 February 2010. 286 7.2. Informative References 288 [RFC7063] Zheng, L., Zhang, Z., and R. Parekh, "Survey Report on 289 Protocol Independent Multicast - Sparse Mode (PIM-SM) 290 Implementations and Deployments", RFC 7063, December 2013. 292 Appendix A. Questionnaire 294 A.1. Questionnaire for Vendors or Host Implementors 296 Name: 298 Affiliation/Organization: 300 Contact Email: 302 Do you wish to complete the survey anonymously?: Y/N 304 A.1.1. Implementation Status 306 Which of the following have you implemented? 308 1. IGMPv1 [RFC1112]? 310 2. IGMPv2 [RFC2236]? 312 3. IGMPv3 [RFC3376]? 314 4. Lightweight IGMPv3 [RFC5790]? 316 5. MLDv1 [RFC2710]? 318 6. MLDv2 [RFC3810]? 320 7. Lightweight MLDv2 [RFC5790]? 322 A.1.2. Implementation Specifics 324 1. Which IGMPv3 and MLDv2 features have you implemented? 326 A. Source filtering with include list? 328 B. Source filtering with exclude list? 329 C. Snooping proxy? 331 D. Snooping querier? 333 E. Snooping filtering? 335 F. L2 Report flooding? 337 G. Host proxy? 339 H. Unicast queries/reports? 341 2. Have you carried out IGMPv3 or MLDv2 interoperability tests with 342 other implementations? 344 A. What issues, if any, arose during these tests? 346 B. How could [RFC3376] and [RFC3810] have helped minimize these 347 issues? 349 A.1.3. Implementation Perspectives 351 1. Which ambiguities or inconsistencies in RFC 3376 or RFC 3810 made 352 the implementation challenging? 354 2. What suggestions would you make to the PIM WG as it seeks to 355 progress IGMPv3 and MLDv2 to Internet Standard? 357 A.2. Questionnaire for Network Operators 359 Name: 361 Affiliation/Organization: 363 Contact Email: 365 Do you wish to complete the survey anonymously?: Y/N: 367 A.2.1. Deployment Status 369 Which of the following have you deployed in your network? 371 1. IGMPv1 [RFC1112]? 373 2. IGMPv2 [RFC2236]? 375 3. IGMPv3 [RFC3376]? 376 4. Lightweight IGMPv3 [RFC5790]? 378 5. MLDv1 [RFC2710]? 380 6. MLDv2 [RFC3810]? 382 7. Lightweight MLDv2 [RFC5790]? 384 A.2.2. Deployment Specifics 386 1. Which IGMPv3 and MLDv2 features do you use? 388 A. Source filtering with include list? 390 B. Source filtering with exclude list? 392 C. Snooping proxy? 394 D. Snooping querier? 396 E. Snooping filtering? 398 F. L2 Report flooding? 400 G. Host proxy? 402 H. Unicast queries/reports? 404 2. Are you using equipment with multi-vendor implementations in your 405 IGMPv3/MLDv2 deployment? 407 A. What inter-operability issues, if any, have you experienced? 409 B. How could [RFC3376] and [RFC3810] have helped minimize these 410 issues? 412 3. Are you using different IGMP versions or different MLD versions 413 in your network? 415 A. Are you dependent on the fallback mechanism between the 416 different versions? 418 B. Have you experienced any issues related to the fallback 419 mechanism between the different versions? 421 C. How could [RFC3376] and [RFC3810] have helped minimize these 422 issues? 424 A.2.3. Deployment Perspectives 426 1. Based on your operational experience, What have you found to be 427 the strengths of IGMPv3 or MLDv2? 429 2. What have you found to be the weaknesses of IGMPv3 or MLDv2? 431 3. What suggestions would you make to the PIM WG as it seeks to 432 progress IGMPv3 and MLDv2 to Internet Standard? 434 Author's Address 436 Olufemi Komolafe 437 Arista Networks 438 UK 440 Email: femi@arista.com