idnits 2.17.1 draft-learmonth-pearg-safe-internet-measurement-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 (May 16, 2019) is 1806 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 Network Working Group I. Learmonth 3 Internet-Draft Tor Project 4 Intended status: Informational May 16, 2019 5 Expires: November 17, 2019 7 Guidelines for Performing Safe Measurement on the Internet 8 draft-learmonth-pearg-safe-internet-measurement-02 10 Abstract 12 Researchers from industry and academia will often use Internet 13 measurements as a part of their work. While these measurements can 14 give insight into the functioning and usage of the Internet, they can 15 come at the cost of user privacy. This document describes guidelines 16 for ensuring that such measurements can be carried out safely. 18 Note 20 Comments are solicited and should be addressed to the research 21 group's mailing list at pearg@irtf.org and/or the author(s). 23 The sources for this draft are at: 25 https://github.com/irl/draft-safe-internet-measurement 27 Status of This Memo 29 This Internet-Draft is submitted in full conformance with the 30 provisions of BCP 78 and BCP 79. 32 Internet-Drafts are working documents of the Internet Engineering 33 Task Force (IETF). Note that other groups may also distribute 34 working documents as Internet-Drafts. The list of current Internet- 35 Drafts is at https://datatracker.ietf.org/drafts/current/. 37 Internet-Drafts are draft documents valid for a maximum of six months 38 and may be updated, replaced, or obsoleted by other documents at any 39 time. It is inappropriate to use Internet-Drafts as reference 40 material or to cite them other than as "work in progress." 42 This Internet-Draft will expire on November 17, 2019. 44 Copyright Notice 46 Copyright (c) 2019 IETF Trust and the persons identified as the 47 document authors. All rights reserved. 49 This document is subject to BCP 78 and the IETF Trust's Legal 50 Provisions Relating to IETF Documents 51 (https://trustee.ietf.org/license-info) in effect on the date of 52 publication of this document. Please review these documents 53 carefully, as they describe your rights and restrictions with respect 54 to this document. 56 1. Introduction 58 When performing research using the Internet, as opposed to an 59 isolated testbed or simulation platform, means that you research co- 60 exists in a space with other users. This document outlines 61 guidelines for academic and industry researchers that might use the 62 Internet as part of scientific experiementation. 64 1.1. Scope of this document 66 Following the guidelines contained within this document is not a 67 substitute for any institutional ethics review process you may have, 68 although these guidelines could help to inform that process. 69 Similarly, these guidelines are not legal advice and local laws must 70 also be considered before starting any experiment that could have 71 adverse impacts on user privacy. 73 1.2. Active and passive measurements 75 Internet measurement studies can be broadly categorized into two 76 groups: active measurements and passive measurements. Active 77 measurements generate traffic. Performance measurements such as TCP 78 throughput testing [RFC6349] or functional measurements such as the 79 feature-dependent connectivity failure tests performed by 80 [PATHspider] both fall into this category. Performing passive 81 measurements requires existing traffic. Passive measurements can 82 help to inform new developments in Internet protocols but can also 83 carry risk. 85 The type of measurement is not truly binary and many studies will 86 include both active and passive components. Each of the 87 considerations in this document must be carefully considered for 88 their applicability regardless of the type of measurement. 90 2. Consent 92 Ideally, informed consent would be collected from all users of a 93 shared network before measurements were performed on them. In cases 94 where it is practical to do so, this should be done. 96 For consent to be informed, all possible risks must be presented to 97 the users. The considerations in this document can be used to 98 provide a starting point although other risks may be present 99 depending on the nature of the measurements to be performed. 101 2.1. Proxy Consent 103 In cases where it is not practical to collect informed consent from 104 all users of a shared network, it may be possible to obtain proxy 105 consent. Proxy consent may be given by a network operator or 106 employer that would be more familiar with the expectations of users 107 of a network than the researcher. 109 2.2. Implied consent 111 In larger scale measurements, even proxy consent collection may not 112 be practical. In this case, implied consent may be presumed from 113 users for some measurements. Consider that users of a network will 114 have certain expectations of privacy and those expectations may not 115 align with the privacy guarantees offered by the technologies they 116 are using. As a thought experiment, consider how users might respond 117 if you asked for their informed consent for the measurements you'd 118 like to perform. 120 For example, the operator of a web server that is exposed to the 121 Internet hosting a popular website would have the expectation that it 122 may be included in surveys that look at supported protocols or 123 extensions but would not expect that attempts be made to degrade the 124 service with large numbers of simultaneous connections. 126 If practical, attempt to obtain informed consent or proxy consent 127 from a sample of users to better understand the expectations of other 128 users. 130 3. Safety Considerations 132 3.1. Use a testbed 134 Wherever possible, use a testbed. An isolated network means that 135 there are no other users sharing the infrastructure you are using for 136 your experiments. 138 When measuring performance, competing traffic can have negative 139 effects on the performance of your test traffic and so the testbed 140 approach can also produce more accurate and repeatable results than 141 experiments using the public Internet. 143 WAN link conditions can be emulated through artificial delays and/or 144 packet loss using a tool like [netem]. Competing traffic can also be 145 emulated using traffic generators. 147 3.2. Only record your own traffic 149 When performing measurements be sure to only capture traffic that you 150 have generated. Traffic may be identified by IP ranges or by some 151 token that is unlikely to be used by other users. 153 Again, this can help to improve the accuracy and repeatability of 154 your experiment. [RFC2544], for performance benchmarking, requires 155 that any frames received that were not part of the test traffic are 156 discarded and not counted in the results. 158 3.3. Be respectful of other's infrastructure 160 If your experiment is designed to trigger a response from 161 infrastructure that is not your own, consider what the negative 162 consequences of that may be. At the very least your experiment will 163 consume bandwidth that may have to be paid for. 165 In more extreme circumstances, you could cause traffic to be 166 generated that causes legal trouble for the owner of that 167 infrastructure. The Internet is a global network crossing many legal 168 jurisdictions and so what may be legal for you is not necessarily 169 legal for everyone. 171 If you are sending a lot of traffic quickly, or otherwise generally 172 deviate from typical client behaviour, a network may identify this as 173 an attack which means that you will not be collecting results that 174 are representative of what a typical client would see. 176 3.3.1. Maintain a "Do Not Scan" list 178 When performing active measurements on a shared network, maintain a 179 list of hosts that you will never scan regardless of whether they 180 appear in your target lists. When developing tools for performing 181 active measurement, or traffic generation for use in a larger 182 measurement system, ensure that the tool will support the use of a 183 "Do Not Scan" list. 185 If complaints are made that request you do not generate traffic 186 towards a host or network, you must add that host or network to your 187 "Do Not Scan" list, even if no explanation is given or the request is 188 automated. 190 You may ask the requester for their reasoning if it would be useful 191 to your experiment. This can also be an oppertunity to explain your 192 research and offer to share any results that may be of interest. If 193 you plan to share the reasoning when publishing your measurement 194 results, e.g. in an academic paper, you must seek consent for this 195 from the requester. 197 Be aware that in publishing your measurement results, it may be 198 possible to infer your "Do Not Scan" list from those results. For 199 example, if you measured a well-known list of popular websites then 200 it would be possible to correlate the results with that list to 201 determine which are missing. 203 3.4. Only collect data that is safe to make public 205 When deciding on the data to collect, assume that any data collected 206 might become public. There are many ways that this could happen, 207 through operation security mistakes or compulsion by a judicial 208 system. 210 3.5. Minimization 212 For all data collected, consider whether or not it is really needed. 214 3.6. Aggregation 216 When collecting data, consider if the granularity can be limited by 217 using bins or adding noise. XXX: Differential privacy. 219 3.7. Source Aggregation 221 Do this at the source, definitely do it before you write to disk. 223 [Tor.2017-04-001] presents a case-study on the in-memory statistics 224 in the software used by the Tor network, as an example. 226 4. Risk Analysis 228 The benefits should outweigh the risks. Consider auxiliary data 229 (e.g. third-party data sets) when assessing the risks. 231 5. Security Considerations 233 Take reasonable security precautions, e.g. about who has access to 234 your data sets or experimental systems. 236 6. IANA Considerations 238 This document has no actions for IANA. 240 7. Acknowledgements 242 Many of these considerations are based on those from the 243 [TorSafetyBoard] adapted and generalised to be applied to Internet 244 research. 246 Other considerations are taken from the Menlo Report [MenloReport] 247 and its companion document [MenloReportCompanion]. 249 8. Informative References 251 [MenloReport] 252 Dittrich, D. and E. Kenneally, "The Menlo Report: Ethical 253 Principles Guiding Information and Communication 254 Technology Research", August 2012, 255 . 258 [MenloReportCompanion] 259 Bailey, M., Dittrich, D., and E. Kenneally, "Applying 260 Ethical Principles to Information and Communication 261 Technology Research", October 2013, 262 . 265 [netem] Stephen, H., "Network emulation with NetEm", April 2005. 267 [PATHspider] 268 Learmonth, I., Trammell, B., Kuehlewind, M., and G. 269 Fairhurst, "PATHspider: A tool for active measurement of 270 path transparency", DOI 10.1145/2959424.2959441, July 271 2016, 272 . 274 [RFC2544] Bradner, S. and J. McQuaid, "Benchmarking Methodology for 275 Network Interconnect Devices", RFC 2544, 276 DOI 10.17487/RFC2544, March 1999, 277 . 279 [RFC6349] Constantine, B., Forget, G., Geib, R., and R. Schrage, 280 "Framework for TCP Throughput Testing", RFC 6349, 281 DOI 10.17487/RFC6349, August 2011, 282 . 284 [Tor.2017-04-001] 285 Herm, K., "Privacy analysis of Tor's in-memory 286 statistics", Tor Tech Report 2017-04-001, April 2017, 287 . 290 [TorSafetyBoard] 291 Tor Project, "Tor Research Safety Board", 292 . 294 Author's Address 296 Iain R. Learmonth 297 Tor Project 299 Email: irl@torproject.org