idnits 2.17.1 draft-bortzmeyer-dnsop-dns-privacy-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 : ---------------------------------------------------------------------------- ** 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.) ** The abstract seems to contain references ([RFC6973]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (April 27, 2014) is 3623 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- -- Looks like a reference, but probably isn't: '1' on line 338 == Unused Reference: 'RFC2119' is defined on line 446, but no explicit reference was found in the text == Unused Reference: 'RFC2181' is defined on line 456, but no explicit reference was found in the text == Unused Reference: 'RFC5246' is defined on line 463, but no explicit reference was found in the text -- Obsolete informational reference (is this intentional?): RFC 5246 (Obsoleted by RFC 8446) == Outdated reference: A later version (-03) exists of draft-wijngaards-dnsop-confidentialdns-00 == Outdated reference: A later version (-01) exists of draft-hzhwm-start-tls-for-dns-00 Summary: 2 errors (**), 0 flaws (~~), 6 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group S. Bortzmeyer 3 Internet-Draft AFNIC 4 Intended status: Informational April 27, 2014 5 Expires: October 29, 2014 7 DNS privacy considerations 8 draft-bortzmeyer-dnsop-dns-privacy-02 10 Abstract 12 This document describes the privacy issues associated with the use of 13 the DNS by Internet users. It is intended to be mostly an analysis 14 of the present situation, in the spirit of section 8 of [RFC6973] and 15 it does not prescribe solutions. 17 Discussions of the document should take place on the dns-privacy 18 mailing list [dns-privacy]. 20 Status of This Memo 22 This Internet-Draft is submitted in full conformance with the 23 provisions of BCP 78 and BCP 79. 25 Internet-Drafts are working documents of the Internet Engineering 26 Task Force (IETF). Note that other groups may also distribute 27 working documents as Internet-Drafts. The list of current Internet- 28 Drafts is at http://datatracker.ietf.org/drafts/current/. 30 Internet-Drafts are draft documents valid for a maximum of six months 31 and may be updated, replaced, or obsoleted by other documents at any 32 time. It is inappropriate to use Internet-Drafts as reference 33 material or to cite them other than as "work in progress." 35 This Internet-Draft will expire on October 29, 2014. 37 Copyright Notice 39 Copyright (c) 2014 IETF Trust and the persons identified as the 40 document authors. All rights reserved. 42 This document is subject to BCP 78 and the IETF Trust's Legal 43 Provisions Relating to IETF Documents 44 (http://trustee.ietf.org/license-info) in effect on the date of 45 publication of this document. Please review these documents 46 carefully, as they describe your rights and restrictions with respect 47 to this document. Code Components extracted from this document must 48 include Simplified BSD License text as described in Section 4.e of 49 the Trust Legal Provisions and are provided without warranty as 50 described in the Simplified BSD License. 52 Table of Contents 54 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 55 2. Risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 56 2.1. The alleged public nature of DNS data . . . . . . . . . . 4 57 2.2. Data in the DNS request . . . . . . . . . . . . . . . . . 4 58 2.3. Cache snooping . . . . . . . . . . . . . . . . . . . . . 5 59 2.4. On the wire . . . . . . . . . . . . . . . . . . . . . . . 6 60 2.5. In the servers . . . . . . . . . . . . . . . . . . . . . 7 61 2.5.1. In the resolvers . . . . . . . . . . . . . . . . . . 8 62 2.5.2. In the authoritative name servers . . . . . . . . . . 8 63 2.5.3. Rogue servers . . . . . . . . . . . . . . . . . . . . 9 64 3. Actual "attacks" . . . . . . . . . . . . . . . . . . . . . . 9 65 4. Legalities . . . . . . . . . . . . . . . . . . . . . . . . . 9 66 5. Security considerations . . . . . . . . . . . . . . . . . . . 9 67 6. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 10 68 7. References . . . . . . . . . . . . . . . . . . . . . . . . . 10 69 7.1. Normative References . . . . . . . . . . . . . . . . . . 10 70 7.2. Informative References . . . . . . . . . . . . . . . . . 10 71 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 12 73 1. Introduction 75 The Domain Name System is specified in [RFC1034] and [RFC1035]. It 76 is one of the most important infrastructure components of the 77 Internet and one of the most often ignored or misunderstood. Almost 78 every activity on the Internet starts with a DNS query (and often 79 several). Its use has many privacy implications and we try to give 80 here a comprehensive and accurate list. 82 Let us start with a small reminder of the way the DNS works (with 83 some simplifications). A client, the stub resolver, issues a DNS 84 query to a server, the resolver (also called caching resolver or full 85 resolver or recursive name server). For instance, the query is "What 86 are the AAAA records for www.example.com?". AAAA is the qtype (Query 87 Type) and www.example.com the qname (Query Name). To get the answer, 88 the resolver will query first the root nameservers, which will, most 89 of the times, send a referral. Here, the referral will be to .com 90 nameservers. In turn, they will send a referral to the example.com 91 nameservers, which will provide the answer. The root name servers, 92 the name servers of .com and those of example.com are called 93 authoritative name servers. It is important, when analyzing the 94 privacy issues, to remember that the question asked to all these name 95 servers is always the original question, not a derived question. 96 Unlike what many "DNS for dummies" articles say, the question sent to 97 the root name servers is "What are the AAAA records for 98 www.example.com?", not "What are the name servers of .com?". So, the 99 DNS leaks more information than it should. 101 Because the DNS uses caching heavily, not all questions are sent to 102 the authoritative name servers. If the stub resolver, a few seconds 103 later, asks to the resolver "What are the SRV records of _xmpp- 104 server._tcp.example.com?", the resolver will remember that it knows 105 the name servers of example.com and will just query them, bypassing 106 the root and .com. Because there is typically no caching in the stub 107 resolver, the resolver, unlike the authoritative servers, sees 108 everything. 110 Almost all the DNS queries are today sent over UDP, and this has 111 practical consequences if someone thinks of encrypting this traffic 112 (some encryption solutions are typically done for TCP, not UDP). 114 I should be noted to that DNS resolvers sometimes forward requests to 115 bigger machines, with a larger and more shared cache, the forwarders. 116 From the point of view of privacy, forwarders are like resolvers, 117 except that the caching in the resolver before them decreases the 118 amount of data they can see. 120 Another important point to keep in mind when analyzing the privacy 121 issues of DNS is the mix of many sort of DNS requests received by a 122 server. Let's assume the eavesdropper want to know which Web page is 123 visited by an user. For a typical Web page displayed by the user, 124 there are three sorts of DNS requests: 126 Primary request: this is the domain name that the user typed or 127 selected from a bookmark or choosed by clicking on an hyperklink. 128 Presumably, this is what is of interest for the eavesdropper. 130 Secondary requests: these are the requests performed by the user 131 agent (here, the Web browser) without any direct involvment or 132 knowledge of the user. For the Web, they are triggered by 133 included content, CSS sheets, JavaScript code, embedded images, 134 etc. In some cases, there can be dozens of domain names in a 135 single page. 137 Tertiary requests: these are the requests performed by the DNS 138 system itself. For instance, if the answer to a query is a 139 referral to a set of name servers, and the glue is not returned, 140 the resolver will have to do tertiary requests to turn name 141 servers' named into IP addresses. 143 For privacy-related terms, we will use here the terminology of 144 [RFC6973]. 146 2. Risks 148 This draft focuses mostly on the study of privacy risks for the end- 149 user (the one performing DNS requests). Privacy risks for the holder 150 of a zone (the risk that someone gets the data) are discussed in 151 [RFC5936]. Non-privacy risks (such as cache poisoning) are out of 152 scope. 154 2.1. The alleged public nature of DNS data 156 It has long been claimed that "the data in the DNS is public". While 157 this sentence makes sense for an Internet wide lookup system, there 158 are multiple facets to data and meta data that deserve a more 159 detailed look. First, access control lists and private name spaces 160 nonwithstanding, the DNS operates under the assumption that public 161 facing authoritative name servers will respond to "usual" DNS queries 162 for any zone they are authoritative for without further 163 authentication or authorization of the client (resolver). Due to the 164 lack of search capabilities, only a given qname will reveal the 165 resource records associated with that name (or that name's non 166 existence). In other words: one needs to know what to ask for to 167 receive a response. The zone transfer qtype [RFC5936] is often 168 blocked or restricted to authenticated/authorized access to enforce 169 this difference (and maybe for other, more dubious reasons). 171 Another differentiation to be applied is between the DNS data as 172 mentioned above and a particular transaction, most prominently but 173 not limited to a DNS name lookup. The fact that the results of a DNS 174 query are public within the boundaries described in the previous 175 paragraph and therefore might have no confidentiality requirements 176 does not imply the same for a single or a sequence of transactions. 177 A typical example from outside the DNS world: the Web site of 178 Alcoholics Anonymous is public, the fact that you visit it should not 179 be. 181 2.2. Data in the DNS request 183 The DNS request includes many fields but two of them seem specially 184 relevant for the privacy issues, the qname and the source IP address. 185 "source IP address" is used in a loose sense of "source IP address + 186 may be source port", because the port is also in the request and can 187 be used to sort out several users sharing an IP address (CGN for 188 instance). 190 The qname is the full name sent by the original user. It gives 191 information about what the user does ("What are the MX records of 192 example.net?" means he probably wants to send email to someone at 193 example.net, which may be a domain used by only a few persons and 194 therefore very revealing). Some qnames are more sensitive than 195 others. For instance, querying the A record of google-analytics.com 196 reveals very little (everybody visits Web sites which use Google 197 Analytics) but querying the A record of www.verybad.example where 198 verybad.example is the domain of an illegal or very offensive 199 organization may create more problems for the user. Another example 200 is when the qname embeds the software one uses. For instance, 201 _ldap._tcp.Default-First-Site-Name._sites.gc._msdcs.example.org. Or 202 some BitTorrent clients that query a SRV record for _bittorrent- 203 tracker._tcp.domain.example. 205 Another important thing about the privacy of the qname is the future 206 usages. Today, the lack of privacy is an obstacle to putting 207 interesting data in the DNS. At the moment your DNS traffic might 208 reveal that you are doing email but not who with. If your MUA starts 209 looking up PGP keys in the DNS [I-D.wouters-dane-openpgp] then 210 privacy becomes a lot more important. And email is just an example, 211 there will be other really interesting uses for a more secure (in the 212 sense of privacy) DNS. 214 For the communication between the stub resolver and the resolver, the 215 source IP address is the one of the user's machine. Therefore, all 216 the issues and warnings about collection of IP addresses apply here. 217 For the communication between the resolver and the authoritative name 218 servers, the source IP address has a different meaning, it does not 219 have the same status as the source address in a HTTP connection. It 220 is now the IP address of the resolver which, in a way "hides" the 221 real user. However, it does not always work. Sometimes 222 [I-D.vandergaast-edns-client-subnet] is used. Sometimes the end user 223 has a personal resolver on her machine. In that case, the IP address 224 is as sensitive as it is for HTTP. 226 A note about IP addresses: there is currently no IETF document which 227 describes in detail the privacy issues of IP addressing. In the mean 228 time, the discussion here is intended to include both IPv4 and IPv6 229 source addresses. For a number of reasons their assignment and 230 utilization characteristics are different, which may have 231 implications for details of information leakage associated with the 232 collection of source addresses. (For example, a specific IPv6 source 233 address seen on the public Internet is less likely than an IPv4 234 address to originate behind a CGN or other NAT.) However, for both 235 IPv4 and IPv6 addresses, it's important to note that source addresses 236 are propagated with queries and comprise metadata about the host, 237 user, or application that originated them. 239 2.3. Cache snooping 240 The content of resolvers can reveal data about the clients using it. 241 This information can sometimes be examined by sending DNS queries 242 with RD=0 to inspect cache content, particularly looking at the DNS 243 TTLs. Since this also is a reconnaissance technique for subsequent 244 cache poisoning attacks, some counter measures have already been 245 developed and deployed. 247 2.4. On the wire 249 DNS traffic can be seen by an eavesdropper like any other traffic. 250 It is typically not encrypted. (DNSSEC, specified in [RFC4033] 251 explicitely excludes confidentiality from its goals.) So, if an 252 initiator starts a HTTPS communication with a recipient, while the 253 HTTP traffic will be encrypted, the DNS exchange prior to it will not 254 be. When the other protocols will become more or more privacy-aware 255 and secured against surveillance, the DNS risks to become "the 256 weakest link" in privacy. 258 What also makes the DNS traffic different is that it may take a 259 different path than the communication between the initiator and the 260 recipient. For instance, an eavesdropper may be unable to tap the 261 wire between the initiator and the recipient but may have access to 262 the wire going to the resolver, or to the authoritative name servers. 264 The best place, from an eavesdropper's point of view, is clearly 265 between the stub resolvers and the resolvers, because he is not 266 limited by DNS caching. 268 The attack surface between the stub resolver and the rest of the 269 world can vary widely depending upon how the end user's computer is 270 configured. By order of increasing attack surface: 272 The resolver can be on the end user's computer. In (currently) a 273 small number of cases, individuals may choose to operate their own 274 DNS resolver on their local machine. In this case the attack surface 275 for the stub resolver to caching resolver connection is limited to 276 that single machine. 278 The resolver can be in the IAP (Internet Access Provider) premises. 279 For most residential users and potentially other networks the typical 280 case is for the end user's computer to be configured (typically 281 automatically through DHCP) with the addresses of the DNS resolver at 282 the IAP. The attack surface for on-the-wire attacks is therefore 283 from the end user system across the local network and across the IAP 284 network to the IAP's resolvers. 286 The resolver may also be at the local network edge. For many/most 287 enterprise networks and for some residential users the caching 288 resolver may exist on a server at the edge of the local network. In 289 this case the attack surface is the local network. Note that in 290 large enterprise networks the DNS resolver may not be located at the 291 edge of the local network but rather at the edge of the overall 292 enterprise network. In this case the enterprise network could be 293 thought of as similar to the IAP network referenced above. 295 The resolver can be a public DNS service. Some end users may be 296 configured to use public DNS resolvers such as those operated by 297 Google Public DNS or OpenDNS. The end user may have configured their 298 machine to use these DNS resolvers themselves - or their IAP may 299 choose to use the public DNS resolvers rather than operating their 300 own resolvers. In this case the attack surface is the entire public 301 Internet between the end user's connection and the public DNS 302 service. 304 2.5. In the servers 306 Using the terminology of [RFC6973], the DNS servers (resolvers and 307 authoritative servers) are enablers: they facilitate communication 308 between an initiator and a recipient without being directly in the 309 communications path. As a result, they are often forgotten in risk 310 analysis. But, to quote again [RFC6973], "Although [...] enablers 311 may not generally be considered as attackers, they may all pose 312 privacy threats (depending on the context) because they are able to 313 observe, collect, process, and transfer privacy-relevant data." In 314 [RFC6973] parlance, enablers become observers when they start 315 collecting data. 317 Many programs exist to collect and analyze DNS data at the servers. 318 From the "query log" of some programs like BIND, to tcpdump and more 319 sophisticated programs like PacketQ [packetq] reference and DNSmezzo 320 [dnsmezzo]. The organization managing the DNS server can use this 321 data itself or it can be part of a surveillance program like PRISM 322 [prism] and pass data to an outside attacker. 324 Sometimes, these data are kept for a long time and/or distributed to 325 third parties, for research purposes [ditl], for security analysis, 326 or for surveillance tasks. Also, there are observation points in the 327 network which gather DNS data and then make it accessible to third- 328 parties for research or security purposes ("passive DNS 329 [passive-dns]"). 331 2.5.1. In the resolvers 333 The resolvers see the entire traffic since there is typically no 334 caching before them. They are therefore well situated to observe the 335 traffic. To summarize: your resolver knows a lot about you. The 336 resolver of a large IAP, or a large public resolver can collect data 337 from many users. You may get an idea of the data collected by 338 reading the privacy policy of a big public resolver [1]. 340 2.5.2. In the authoritative name servers 342 Unlike the resolvers, they are limited by caching. They see only a 343 part of the requests. For aggregated statistics ("what is the 344 percentage of LOC queries?"), it is sufficient but it may prevent an 345 observer to observe everything. Nevertheless, the authoritative name 346 servers sees a part of the traffic and this sample may be sufficient 347 to defeat some privacy expectations. 349 Also, the end user has typically some legal/contractual link with the 350 resolver (he has chosen the IAP, or he has chosen to use a given 351 public resolver) while he is often not even aware of the role of the 352 authoritative name servers and their observation abilities. 354 It is an interesting question whether the privacy issues are bigger 355 in the root or in a large TLD. The root sees the traffic for all the 356 TLDs (and the huge amount of traffic for non-existing TLD) but a 357 large TLD has less caching before it. 359 As noted before, using a local resolver or a resolver close to the 360 machine decreases the attack surface for an on-the-wire eavesdropper. 361 But it may decrease privacy against an observer located on an 362 authoritative name server since the authoritative name server will 363 see the IP address of the end client, and not the address of a big 364 resolver shared by many users. This is no longer true if 365 [I-D.vandergaast-edns-client-subnet] is used because, in this case, 366 the authoritative name server sees the original IP prefix or address 367 (depending on the setup). 369 As of today, all the instances of one root name server, L-root, 370 receive together around 20 000 queries per second. While most of it 371 is junk (errors on the TLD name), it gives an idea of the amount of 372 big data which pours into name servers. 374 Many domains, including TLD, are partially hosted by third-party 375 servers, sometimes in a different country. The contracts between the 376 domain manager and these servers may or may not take privacy into 377 account. But it may be surprising for an end-user that requests to a 378 given ccTLD may go to servers managed by organisations outside of the 379 country. 381 2.5.3. Rogue servers 383 A rogue DHCP server can direct you to a rogue resolver. Most of the 384 times, it seems to be done to divert traffic, by providing lies for 385 some domain names. But it could be used just to capture the traffic 386 and gather information about you. Same thing for malwares like 387 DNSchanger[dnschanger] which changes the resolver in the machine's 388 configuration. 390 3. Actual "attacks" 392 A very quick examination of DNS traffic may lead to the false 393 conclusion that extracting the needle from the haystack is difficult. 394 "Interesting" primary DNS requests are mixed with useless (for the 395 eavesdropper) second and tertiary requests (see the terminology in 396 Section 1). But, in this time of "big data" processing, powerful 397 techniques now exist to get from the raw data to what you're actually 398 interested in. 400 Many research papers about malware detection use DNS traffic to 401 detect "abnormal" behaviour that can be traced back to the activity 402 of malware on infected machines. Yes, this research was done for the 403 good but, technically, it is a privacy attack and it demonstrates the 404 power of the observation of DNS traffic. See [dns-footprint], 405 [dagon-malware] and [darkreading-dns]. 407 Passive DNS systems [passive-dns] allow reconstruction of the data of 408 sometimes an entire zone. It is used for many reasons, some good, 409 some bad. It is an example of privacy issue even when no source IP 410 address is kept. 412 4. Legalities 414 To our knowledge, there are no specific privacy laws for DNS data. 415 Interpreting general privacy laws like [data-protection-directive] 416 (European Union) in the context of DNS traffic data is not an easy 417 task and it seems there is no court precedent here. 419 5. Security considerations 420 This document is entirely about security, more precisely privacy. 421 Possible solutions to the issues described here are discussed in 422 [I-D.bortzmeyer-dnsop-privacy-sol] (qname minimization, local caching 423 resolvers), [I-D.hzhwm-start-tls-for-dns] (encryption of traffic) or 424 in [I-D.wijngaards-dnsop-confidentialdns] (encryption also). 425 Attempts have been made to encrypt the resource record data 426 [I-D.timms-encrypt-naptr]. 428 6. Acknowledgments 430 Thanks to Nathalie Boulvard and to the CENTR members for the original 431 work which leaded to this draft. Thanks to Ondrej Sury for the 432 interesting discussions. Thanks to Mohsen Souissi for proofreading. 433 Thanks to Dan York, Suzanne Woolf, Tony Finch, Peter Koch and Frank 434 Denis for good written contributions. 436 7. References 438 7.1. Normative References 440 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 441 STD 13, RFC 1034, November 1987. 443 [RFC1035] Mockapetris, P., "Domain names - implementation and 444 specification", STD 13, RFC 1035, November 1987. 446 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 447 Requirement Levels", BCP 14, RFC 2119, March 1997. 449 [RFC6973] Cooper, A., Tschofenig, H., Aboba, B., Peterson, J., 450 Morris, J., Hansen, M., and R. Smith, "Privacy 451 Considerations for Internet Protocols", RFC 6973, July 452 2013. 454 7.2. Informative References 456 [RFC2181] Elz, R. and R. Bush, "Clarifications to the DNS 457 Specification", RFC 2181, July 1997. 459 [RFC4033] Arends, R., Austein, R., Larson, M., Massey, D., and S. 460 Rose, "DNS Security Introduction and Requirements", RFC 461 4033, March 2005. 463 [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security 464 (TLS) Protocol Version 1.2", RFC 5246, August 2008. 466 [RFC5936] Lewis, E. and A. Hoenes, "DNS Zone Transfer Protocol 467 (AXFR)", RFC 5936, June 2010. 469 [I-D.vandergaast-edns-client-subnet] 470 Contavalli, C., Gaast, W., Leach, S., and E. Lewis, 471 "Client Subnet in DNS Requests", draft-vandergaast-edns- 472 client-subnet-02 (work in progress), July 2013. 474 [I-D.bortzmeyer-dnsop-privacy-sol] 475 Bortzmeyer, S., "Possible solutions to DNS privacy 476 issues", draft-bortzmeyer-dnsop-privacy-sol-00 (work in 477 progress), December 2013. 479 [I-D.wijngaards-dnsop-confidentialdns] 480 Wijngaards, W., "Confidential DNS", draft-wijngaards- 481 dnsop-confidentialdns-00 (work in progress), November 482 2013. 484 [I-D.timms-encrypt-naptr] 485 Timms, B., Reid, J., and J. Schlyter, "IANA Registration 486 for Encrypted ENUM", draft-timms-encrypt-naptr-01 (work in 487 progress), July 2008. 489 [I-D.hzhwm-start-tls-for-dns] 490 Zi, Z., Zhu, L., Heidemann, J., Mankin, A., and D. 491 Wessels, "Starting TLS over DNS", draft-hzhwm-start-tls- 492 for-dns-00 (work in progress), February 2014. 494 [I-D.wouters-dane-openpgp] 495 Wouters, P., "Using DANE to Associate OpenPGP public keys 496 with email addresses", draft-wouters-dane-openpgp-02 (work 497 in progress), February 2014. 499 [dns-privacy] 500 IETF, , "The dns-privacy mailing list", March 2014. 502 [dnsop] IETF, , "The dnsop mailing list", October 2013. 504 [dagon-malware] 505 Dagon, D., "Corrupted DNS Resolution Paths: The Rise of a 506 Malicious Resolution Authority", 2007. 508 [dns-footprint] 509 Stoner, E., "DNS footprint of malware", October 2010. 511 [darkreading-dns] 512 Lemos, R., "Got Malware? Three Signs Revealed In DNS 513 Traffic", May 2013. 515 [dnschanger] 516 Wikipedia, , "DNSchanger", November 2011. 518 [dnscrypt] 519 Denis, F., "DNSCrypt", . 521 [dnscurve] 522 Bernstein, D., "DNScurve", . 524 [packetq] , "PacketQ, a simple tool to make SQL-queries against 525 PCAP-files", 2011. 527 [dnsmezzo] 528 Bortzmeyer, S., "DNSmezzo", 2009. 530 [prism] NSA, , "PRISM", 2007. 532 [crime] Rizzo, J. and T. Dong, "The CRIME attack against TLS", 533 2012. 535 [ditl] , "A Day in the Life of the Internet (DITL)", 2002. 537 [data-protection-directive] 538 , "European directive 95/46/EC on the protection of 539 individuals with regard to the processing of personal data 540 and on the free movement of such data", November 1995. 542 [passive-dns] 543 Weimer, F., "Passive DNS Replication", April 2005. 545 [tor-leak] 546 , "DNS leaks in Tor", 2013. 548 Author's Address 550 Stephane Bortzmeyer 551 AFNIC 552 Immeuble International 553 Saint-Quentin-en-Yvelines 78181 554 France 556 Phone: +33 1 39 30 83 46 557 Email: bortzmeyer+ietf@nic.fr 558 URI: http://www.afnic.fr/