idnits 2.17.1 draft-ietf-6lo-privacy-considerations-04.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 (October 31, 2016) is 2735 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- -- Obsolete informational reference (is this intentional?): RFC 4941 (Obsoleted by RFC 8981) == Outdated reference: A later version (-08) exists of draft-ietf-6lo-6lobac-05 == Outdated reference: A later version (-09) exists of draft-ietf-6lo-dect-ule-07 == Outdated reference: A later version (-22) exists of draft-ietf-6lo-nfc-05 Summary: 0 errors (**), 0 flaws (~~), 4 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group D. Thaler 3 Internet-Draft Microsoft 4 Intended status: Informational October 31, 2016 5 Expires: May 4, 2017 7 Privacy Considerations for IPv6 Adaptation Layer Mechanisms 8 draft-ietf-6lo-privacy-considerations-04 10 Abstract 12 This document discusses how a number of privacy threats apply to 13 technologies designed for IPv6 over various link layer protocols, and 14 provides advice to protocol designers on how to address such threats 15 in adaptation layer specifications for IPv6 over such links. 17 Status of This Memo 19 This Internet-Draft is submitted in full conformance with the 20 provisions of BCP 78 and BCP 79. 22 Internet-Drafts are working documents of the Internet Engineering 23 Task Force (IETF). Note that other groups may also distribute 24 working documents as Internet-Drafts. The list of current Internet- 25 Drafts is at http://datatracker.ietf.org/drafts/current/. 27 Internet-Drafts are draft documents valid for a maximum of six months 28 and may be updated, replaced, or obsoleted by other documents at any 29 time. It is inappropriate to use Internet-Drafts as reference 30 material or to cite them other than as "work in progress." 32 This Internet-Draft will expire on May 4, 2017. 34 Copyright Notice 36 Copyright (c) 2016 IETF Trust and the persons identified as the 37 document authors. All rights reserved. 39 This document is subject to BCP 78 and the IETF Trust's Legal 40 Provisions Relating to IETF Documents 41 (http://trustee.ietf.org/license-info) in effect on the date of 42 publication of this document. Please review these documents 43 carefully, as they describe your rights and restrictions with respect 44 to this document. Code Components extracted from this document must 45 include Simplified BSD License text as described in Section 4.e of 46 the Trust Legal Provisions and are provided without warranty as 47 described in the Simplified BSD License. 49 Table of Contents 51 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 52 2. Amount of Entropy Needed in Global Addresses . . . . . . . . 3 53 3. Potential Approaches . . . . . . . . . . . . . . . . . . . . 4 54 3.1. IEEE-Identifier-Based Addresses . . . . . . . . . . . . . 5 55 3.2. Short Addresses . . . . . . . . . . . . . . . . . . . . . 5 56 4. Recommendations . . . . . . . . . . . . . . . . . . . . . . . 6 57 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 7 58 6. Security Considerations . . . . . . . . . . . . . . . . . . . 7 59 7. Informative References . . . . . . . . . . . . . . . . . . . 7 60 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 9 62 1. Introduction 64 RFC 6973 [RFC6973] discusses privacy considerations for Internet 65 protocols, and Section 5.2 of that document covers a number of 66 privacy-specific threats. In the context of IPv6 addresses, 67 Section 3 of [RFC7721] provides further elaboration on the 68 applicability of the privacy threats. 70 When interface identifiers (IIDs) are generated without sufficient 71 entropy compared to the link lifetime, devices and users can become 72 vulnerable to the various threats discussed there, including: 74 o Correlation of activities over time, if the same identifier is 75 used for traffic over period of time 77 o Location tracking, if the same interface identifier is used with 78 different prefixes as a device moves between different networks 80 o Device-specific vulnerability exploitation, if the identifier 81 helps identify a vendor or version or protocol and hence suggests 82 what types of attacks to try 84 o Address scanning, which enables all of the above attacks by off- 85 link attackers. (On some Non-Broadcast Multi-Access (NBMA) links 86 where all nodes aren't already privy to all on-link addresses, 87 address scans might also be done by on-link attackers, but in most 88 cases address scans are not an interesting threat from on-link 89 attackers and thus address scans generally apply only to routable 90 addresses.) 92 For example, for links that may last for years, "enough" bits of 93 entropy means at least 46 or so bits (see Section 2 for why) in a 94 routable address; ideally all 64 bits of the IID should be used, 95 although historically some bits have been excluded for reasons 96 discussed in [RFC7421]. Link-local addresses can also be susceptible 97 to the same privacy threats from off-link attackers, since experience 98 shows they are often leaked by upper-layer protocols such as SMTP, 99 SIP, or DNS. 101 For these reasons, [I-D.ietf-6man-default-iids] recommends using an 102 address generation scheme in [RFC7217], rather than addresses 103 generated from a fixed link-layer address. 105 Furthermore, to mitigate the threat of correlation of activities over 106 time on long-lived links, [RFC4941] specifies the notion of a 107 "temporary" address to be used for transport sessions (typically 108 locally-initiated outbound traffic to the Internet) that should not 109 be linkable to a more permanent identifier such as a DNS name, user 110 name, or fixed link-layer address. Indeed, the default address 111 selection rules [RFC6724] now prefer temporary addresses by default 112 for outgoing connections. If a device needs to simultaneously 113 support unlinkable traffic as well as traffic that is linkable to 114 such a stable identifier, this necessitates supporting simultaneous 115 use of multiple addresses per device. 117 2. Amount of Entropy Needed in Global Addresses 119 In terms of privacy threats discussed in [RFC7721], the one with the 120 need for the most entropy is address scans of routable addresses. To 121 mitigate address scans, one needs enough entropy to make the 122 probability of a successful address probe be negligible. Typically 123 this is measured in the length of time it would take to have a 50% 124 probability of getting at least one hit. Address scans often rely on 125 sending a packet such as a TCP SYN or ICMP Echo Request, and 126 determining whether the reply is an ICMP unreachable error (if no 127 host exists with that address) or a TCP response or ICMP Echo Reply 128 (if a host exists), or neither in which case nothing is known for 129 certain. 131 Many privacy-sensitive devices support a "stealth mode" as discussed 132 in Section 5 of [RFC7288] or are behind a network firewall that will 133 drop unsolicited inbound traffic (e.g., TCP SYNs, ICMP Echo Requests, 134 etc.) and thus no TCP RST or ICMP Echo Reply will be sent. In such 135 cases, and when the device does not listen on a well-known TCP or UDP 136 port known to the scanner, the effectiveness of an address scan is 137 limited by the ability to get ICMP unreachable errors, since the 138 attacker can only infer the presence of a host based on the absense 139 of an ICMP unreachable error. 141 Generation of ICMP unreachable errors is typically rate limited to 2 142 per second (the default in routers such as Cisco routers running IOS 143 12.0 or later). Such a rate results in taking about a year to 144 completely scan 26 bits of space. 146 The actual math is as follows. Let 2^N be the number of devices on 147 the subnet. Let 2^M be the size of the space to scan (i.e., M bits 148 of entropy). Let S be the number of scan attempts. The formula for 149 a 50% chance of getting at least one hit in S attempts is: P(at least 150 one success) = 1 - (1 - 2^N/2^M)^S = 1/2. Assuming 2^M >> S, this 151 simplifies to: S * 2^N/2^M = 1/2, giving S = 2^(M-N-1), or M = N + 1 152 + log_2(S). Using a scan rate of 2 per second, this results in the 153 following rule of thumb: 155 Bits of entropy needed = log_2(# devices per link) + log_2(seconds 156 of link lifetime) + 2 158 For example, for a network with at most 2^16 devices on the same 159 long-lived link, and the average lifetime of a link being 8 years 160 (2^28 seconds) or less, this results in a need for at least 46 bits 161 of entropy (16+28+2) so that an address scan would need to be 162 sustained for longer than the lifetime of the link to have a 50% 163 chance of getting a hit. 165 Although 46 bits of entropy may be enough to provide privacy in such 166 cases, 59 or more bits of entropy would be needed if addresses are 167 used to provide security against attacks such as spoofing, as CGAs 168 [RFC3972] and HBAs [RFC5535] do, since attacks are not limited by 169 ICMP rate limiting but by the processing power of the attacker. See 170 those RFCs for more discussion. 172 If, on the other hand, the devices being scanned for respond to 173 unsolicited inbound packets, then the address scan is not limited by 174 the ICMP unreachable rate limit in routers, since an adversary can 175 determine the presence of a host without them. In such cases, more 176 bits of entropy would be needed to provide the same level of 177 protection. 179 3. Potential Approaches 181 The table below shows the number of bits of entropy currently 182 available in various technologies: 184 +---------------+--------------------------+--------------------+ 185 | Technology | Reference | Bits of Entropy | 186 +---------------+--------------------------+--------------------+ 187 | 802.15.4 | [RFC4944] | 16+ or any EUI-64 | 188 | Bluetooth LE | [RFC7668] | 48 | 189 | DECT ULE | [I-D.ietf-6lo-dect-ule] | 40 or any EUI-48 | 190 | MS/TP | [I-D.ietf-6lo-6lobac] | 7 or 64 | 191 | ITU-T G.9959 | [RFC7428] | 8 | 192 | NFC | [I-D.ietf-6lo-nfc] | 5 | 193 +---------------+--------------------------+--------------------+ 195 Such technologies generally support either IEEE identifiers or so 196 called "Short Addresses", or both, as link layer addresses. We 197 discuss each in turn. 199 3.1. IEEE-Identifier-Based Addresses 201 Some technologies allow the use of IEEE EUI-48 or EUI-64 identifiers, 202 or allow using an arbitrary 64-bit identifier. Using such an 203 identifier to construct IPv6 addresses makes it easy to use the 204 normal LOWPAN_IPHC encoding with stateless compression, allowing such 205 IPv6 addresses to be fully elided in common cases. 207 Global addresses with interface identifiers formed from IEEE 208 identifiers can have insufficient entropy to mitigate address scans 209 unless the IEEE identifier itself has sufficient entropy, and enough 210 bits of entropy are carried over into the IPv6 address to 211 sufficiently mitigate the threats. Privacy threats other than 212 "Correlation over time" can be mitigated using per-network randomized 213 link-layer addresses with enough entropy compared to the link 214 lifetime. A number of such proposals can be found at 215 , and Section 10.8 of 216 [BTCorev4.1] specifies one for Bluetooth. Using routable IPv6 217 addresses derived from such link-layer addresses would be roughly 218 equivalent to those specified in [RFC7217]. 220 Correlation over time (for all addresses, not just routable 221 addresses) can be mitigated if the link-layer address itself changes 222 often enough, such as each time the link is established, if the link 223 lifetime is short. For further discussion, see 224 [I-D.huitema-6man-random-addresses]. 226 Another potential concern is that of efficiency, such as avoiding 227 Duplicate Address Detection (DAD) all together when IPv6 addresses 228 are IEEE-identifier-based. Appendix A of [RFC4429] provides an 229 analysis of address collision probability based on the number of bits 230 of entropy. A simple web search on "duplicate MAC addresses" will 231 show that collisions do happen with MAC addresses, and thus based on 232 the analysis in [RFC4429], using sufficient bits of entropy in random 233 addresses can provide greater protection against collision than using 234 MAC addresses. 236 3.2. Short Addresses 238 A routable IPv6 address with an interface identifier formed from the 239 combination of a "Short Address" and a set of well-known constant 240 bits (such as padding with 0's) lacks sufficient entropy to mitigate 241 address scanning unless the link lifetime is extremely short. 242 Furthermore, an adversary could also use statistical methods to 243 determine the size of the L2 address space and thereby make some 244 inference regarding the underlying technology on a given link, and 245 target further attacks accordingly. 247 When Short Addresses are desired on links that are not guaranteed to 248 have a short enough lifetime, the mechanism for constructing an IPv6 249 interface identifier from a Short Address could be designed to 250 sufficiently mitigate the problem. For example, if all nodes on a 251 given L2 network have a shared secret (such as the key needed to get 252 on the layer-2 network), the 64-bit IID might be generated using a 253 one-way hash that includes (at least) the shared secret together with 254 the Short Address. The use of such a hash would result in the IIDs 255 being spread out among the full range of IID address space, thus 256 mitigating address scans, while still allowing full stateless 257 compression/elision. 259 For long-lived links, "temporary" addresses might even be generated 260 in the same way by (for example) also including in the hash the 261 Version Number from the Authoritative Border Router Option 262 (Section 4.3 of [RFC6775]), if any. This would allow changing 263 temporary addresses whenever the Version Number is changed, even if 264 the set of prefix or context information is unchanged. 266 In summary, any specification using Short Addresses should carefully 267 construct an IID generation mechanism so as to provide sufficient 268 entropy compared to the link lifetime. 270 4. Recommendations 272 The following are recommended for adaptation layer specifications: 274 o Security (privacy) sections should say how address scans are 275 mitigated. An address scan might be mitigated by having a link 276 always be short-lived, or might be mitigated by having a large 277 number of bits of entropy in routable addresses, or some 278 combination. Thus, a specification should explain what the 279 maximum lifetime of a link is in practice, and show how the number 280 of bits of entropy is sufficient given that lifetime. 282 o Technologies should define a way to include sufficient bits of 283 entropy in the IPv6 interface identifier, based on the maximum 284 link lifetime. Specifying that randomized link-layer addresses 285 can be used is one easy way to do so, for technologies that 286 support such identifiers. 288 o Specifications should not simply construct an IPv6 interface 289 identifier by padding a short address with a set of other well- 290 known constant bits, unless the link lifetime is guaranteed to be 291 extremely short or the short address is allocated by the network 292 (rather than being constant in the node). This also applies to 293 link-local addresses if the same short address is used independent 294 of network and is unique enough to allow location tracking. 296 o Specifications should make sure that an IPv6 address can change 297 over long periods of time. For example, the interface identifier 298 might change each time a device connects to the network (if 299 connections are short), or might change each day (if connections 300 can be long). This is necessary to mitigate correlation over 301 time. 303 o If a device can roam between networks, and more than a few bits of 304 entropy exist in the IPv6 interface identifier, then make sure 305 that the interface identifier can vary per network as the device 306 roams. This is necessary to mitigate location tracking. 308 5. IANA Considerations 310 This document has no actions for IANA. 312 6. Security Considerations 314 This entire document is about security considerations and how to 315 specify possible mitigations. 317 7. Informative References 319 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 320 RFC 3972, DOI 10.17487/RFC3972, March 2005, 321 . 323 [RFC4429] Moore, N., "Optimistic Duplicate Address Detection (DAD) 324 for IPv6", RFC 4429, DOI 10.17487/RFC4429, April 2006, 325 . 327 [RFC4941] Narten, T., Draves, R., and S. Krishnan, "Privacy 328 Extensions for Stateless Address Autoconfiguration in 329 IPv6", RFC 4941, DOI 10.17487/RFC4941, September 2007, 330 . 332 [RFC4944] Montenegro, G., Kushalnagar, N., Hui, J., and D. Culler, 333 "Transmission of IPv6 Packets over IEEE 802.15.4 334 Networks", RFC 4944, DOI 10.17487/RFC4944, September 2007, 335 . 337 [RFC5535] Bagnulo, M., "Hash-Based Addresses (HBA)", RFC 5535, 338 DOI 10.17487/RFC5535, June 2009, 339 . 341 [RFC6724] Thaler, D., Ed., Draves, R., Matsumoto, A., and T. Chown, 342 "Default Address Selection for Internet Protocol Version 6 343 (IPv6)", RFC 6724, DOI 10.17487/RFC6724, September 2012, 344 . 346 [RFC6775] Shelby, Z., Ed., Chakrabarti, S., Nordmark, E., and C. 347 Bormann, "Neighbor Discovery Optimization for IPv6 over 348 Low-Power Wireless Personal Area Networks (6LoWPANs)", 349 RFC 6775, DOI 10.17487/RFC6775, November 2012, 350 . 352 [RFC6973] Cooper, A., Tschofenig, H., Aboba, B., Peterson, J., 353 Morris, J., Hansen, M., and R. Smith, "Privacy 354 Considerations for Internet Protocols", RFC 6973, 355 DOI 10.17487/RFC6973, July 2013, 356 . 358 [RFC7217] Gont, F., "A Method for Generating Semantically Opaque 359 Interface Identifiers with IPv6 Stateless Address 360 Autoconfiguration (SLAAC)", RFC 7217, 361 DOI 10.17487/RFC7217, April 2014, 362 . 364 [RFC7288] Thaler, D., "Reflections on Host Firewalls", RFC 7288, 365 DOI 10.17487/RFC7288, June 2014, 366 . 368 [RFC7421] Carpenter, B., Ed., Chown, T., Gont, F., Jiang, S., 369 Petrescu, A., and A. Yourtchenko, "Analysis of the 64-bit 370 Boundary in IPv6 Addressing", RFC 7421, 371 DOI 10.17487/RFC7421, January 2015, 372 . 374 [RFC7428] Brandt, A. and J. Buron, "Transmission of IPv6 Packets 375 over ITU-T G.9959 Networks", RFC 7428, 376 DOI 10.17487/RFC7428, February 2015, 377 . 379 [RFC7668] Nieminen, J., Savolainen, T., Isomaki, M., Patil, B., 380 Shelby, Z., and C. Gomez, "IPv6 over BLUETOOTH(R) Low 381 Energy", RFC 7668, DOI 10.17487/RFC7668, October 2015, 382 . 384 [RFC7721] Cooper, A., Gont, F., and D. Thaler, "Security and Privacy 385 Considerations for IPv6 Address Generation Mechanisms", 386 RFC 7721, DOI 10.17487/RFC7721, March 2016, 387 . 389 [I-D.ietf-6man-default-iids] 390 Gont, F., Cooper, A., Thaler, D., and S. LIU, 391 "Recommendation on Stable IPv6 Interface Identifiers", 392 draft-ietf-6man-default-iids-16 (work in progress), 393 September 2016. 395 [I-D.ietf-6lo-6lobac] 396 Lynn, K., Martocci, J., Neilson, C., and S. Donaldson, 397 "Transmission of IPv6 over MS/TP Networks", draft-ietf- 398 6lo-6lobac-05 (work in progress), June 2016. 400 [I-D.ietf-6lo-dect-ule] 401 Mariager, P., Petersen, J., Shelby, Z., Logt, M., and D. 402 Barthel, "Transmission of IPv6 Packets over DECT Ultra Low 403 Energy", draft-ietf-6lo-dect-ule-07 (work in progress), 404 October 2016. 406 [I-D.ietf-6lo-nfc] 407 Choi, Y., Youn, J., and Y. Hong, "Transmission of IPv6 408 Packets over Near Field Communication", draft-ietf-6lo- 409 nfc-05 (work in progress), October 2016. 411 [I-D.huitema-6man-random-addresses] 412 Huitema, C., "Implications of Randomized Link Layers 413 Addresses for IPv6 Address Assignment", draft-huitema- 414 6man-random-addresses-03 (work in progress), March 2016. 416 [BTCorev4.1] 417 Bluetooth Special Interest Group, "Bluetooth Core 418 Specification Version 4.1", December 2013, 419 . 422 Author's Address 424 Dave Thaler 425 Microsoft 426 One Microsoft Way 427 Redmond, WA 98052 428 USA 430 Email: dthaler@microsoft.com