idnits 2.17.1 draft-gont-6man-rfc4941bis-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 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 'Obsoletes: ' line in the draft header should list only the _numbers_ of the RFCs which will be obsoleted by this document (if approved); it should not include the word 'RFC' in the list. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (March 5, 2018) is 2244 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'MD5' is defined on line 729, but no explicit reference was found in the text == Unused Reference: 'RFC7136' is defined on line 767, but no explicit reference was found in the text == Unused Reference: 'DDNS' is defined on line 782, but no explicit reference was found in the text == Unused Reference: 'DHCP' is defined on line 786, but no explicit reference was found in the text ** Downref: Normative reference to an Informational RFC: RFC 1321 (ref. 'MD5') ** Obsolete normative reference: RFC 4941 (Obsoleted by RFC 8981) -- Obsolete informational reference (is this intentional?): RFC 2965 (ref. 'COOKIES') (Obsoleted by RFC 6265) -- Obsolete informational reference (is this intentional?): RFC 3315 (ref. 'DHCPV6') (Obsoleted by RFC 8415) == Outdated reference: A later version (-12) exists of draft-ietf-dnsop-ipv6-dns-issues-10 == Outdated reference: A later version (-04) exists of draft-gont-6man-non-stable-iids-01 == Outdated reference: A later version (-01) exists of draft-gont-taps-address-usage-problem-statement-00 -- Duplicate reference: RFC1321, mentioned in 'RFC1321', was also mentioned in 'MD5'. Summary: 3 errors (**), 0 flaws (~~), 9 warnings (==), 4 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 IPv6 Maintenance (6man) Working Group F. Gont 3 Internet-Draft SI6 Networks / UTN-FRH 4 Obsoletes: rfc4941 (if approved) S. Krishnan 5 Intended status: Standards Track Ericsson Research 6 Expires: September 6, 2018 March 5, 2018 8 Privacy Extensions for Stateless Address Autoconfiguration in IPv6 9 draft-gont-6man-rfc4941bis-00 11 Abstract 13 Nodes use IPv6 stateless address autoconfiguration to generate 14 addresses using a combination of locally available information and 15 information advertised by routers. Addresses are formed by combining 16 network prefixes with an interface identifier. This document 17 describes an extension that causes nodes to generate global scope 18 addresses from interface identifiers that change over time. Changing 19 the interface identifier (and the global scope addresses generated 20 from it) over time makes it more difficult for eavesdroppers and 21 other information collectors to identify when different addresses 22 used in different transactions actually correspond to the same node. 24 Status of this Memo 26 This Internet-Draft is submitted in full conformance with the 27 provisions of BCP 78 and BCP 79. 29 Internet-Drafts are working documents of the Internet Engineering 30 Task Force (IETF). Note that other groups may also distribute 31 working documents as Internet-Drafts. The list of current Internet- 32 Drafts is at http://datatracker.ietf.org/drafts/current/. 34 Internet-Drafts are draft documents valid for a maximum of six months 35 and may be updated, replaced, or obsoleted by other documents at any 36 time. It is inappropriate to use Internet-Drafts as reference 37 material or to cite them other than as "work in progress." 39 This Internet-Draft will expire on September 6, 2018. 41 Copyright Notice 43 Copyright (c) 2018 IETF Trust and the persons identified as the 44 document authors. All rights reserved. 46 This document is subject to BCP 78 and the IETF Trust's Legal 47 Provisions Relating to IETF Documents 48 (http://trustee.ietf.org/license-info) in effect on the date of 49 publication of this document. Please review these documents 50 carefully, as they describe your rights and restrictions with respect 51 to this document. Code Components extracted from this document must 52 include Simplified BSD License text as described in Section 4.e of 53 the Trust Legal Provisions and are provided without warranty as 54 described in the Simplified BSD License. 56 Table of Contents 58 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 59 1.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . 3 60 1.2. Problem Statement . . . . . . . . . . . . . . . . . . . . 3 61 2. Background . . . . . . . . . . . . . . . . . . . . . . . . . . 5 62 2.1. Extended Use of the Same Identifier . . . . . . . . . . . 5 63 2.2. Possible Approaches . . . . . . . . . . . . . . . . . . . 6 64 3. Protocol Description . . . . . . . . . . . . . . . . . . . . . 8 65 3.1. Assumptions . . . . . . . . . . . . . . . . . . . . . . . 8 66 3.2. Generation Of Randomized Interface Identifiers . . . . . . 9 67 3.2.1. Randomized IIDs . . . . . . . . . . . . . . . . . . . 9 68 3.2.2. Hash-based generation of randomized interface 69 identifiers . . . . . . . . . . . . . . . . . . . . . 9 70 3.3. Generating Temporary Addresses . . . . . . . . . . . . . . 11 71 3.4. Expiration of Temporary Addresses . . . . . . . . . . . . 12 72 3.5. Regeneration of Randomized Interface Identifiers . . . . . 13 73 3.6. Deployment Considerations . . . . . . . . . . . . . . . . 14 74 4. Implications of Changing Interface Identifiers . . . . . . . . 16 75 5. Defined Constants . . . . . . . . . . . . . . . . . . . . . . 17 76 6. Future Work . . . . . . . . . . . . . . . . . . . . . . . . . 18 77 7. Security Considerations . . . . . . . . . . . . . . . . . . . 19 78 8. Significant Changes from RFC RFC4941 . . . . . . . . . . . . . 20 79 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 21 80 10. References . . . . . . . . . . . . . . . . . . . . . . . . . . 22 81 10.1. Normative References . . . . . . . . . . . . . . . . . . . 22 82 10.2. Informative References . . . . . . . . . . . . . . . . . . 23 83 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 25 85 1. Introduction 87 Stateless address autoconfiguration [RFC4862] defines how an IPv6 88 node generates addresses without the need for a DHCPv6 server. 90 The security and privacy implications of such addresses have been 91 discussed in great detail in [RFC7721],[RFC7217], and RFC7707. 93 Section 2 provides background information on the issue. Section 3 94 describes a procedure for generating alternate interface identifiers 95 and global scope addresses. Section 4 discusses implications of 96 changing interface identifiers. The term "global scope addresses" is 97 used in this document to collectively refer to "Global unicast 98 addresses" as defined in [RFC4291] and "Unique local addresses" as 99 defined in [RFC4193]. 101 1.1. Terminology 103 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 104 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 105 document are to be interpreted as described in [RFC2119]. 107 The terms "Public address", "stable address", "temporary address", 108 "constant IID", "Stable IID", and "Temporary IID" are to be 109 interpreted as specified in [RFC7721]. 111 1.2. Problem Statement 113 Addresses generated using Stateless address autoconfiguration 114 [RFC4862]contain an embedded interface identifier, which remains 115 stable over time. Anytime a fixed identifier is used in multiple 116 contexts, it becomes possible to correlate seemingly unrelated 117 activity using this identifier. 119 The correlation can be performed by 121 o An attacker who is in the path between the node in question and 122 the peer(s) it is communicating to, and can view the IPv6 123 addresses present in the datagrams. 125 o An attacker who can access the communication logs of the peers 126 with which the node has communicated. 128 Since the identifier is embedded within the IPv6 address, which is a 129 fundamental requirement of communication, it cannot be easily hidden. 130 This document proposes a solution to this issue by generating 131 interface identifiers which vary over time. 133 Note that an attacker, who is on path, may be able to perform 134 significant correlation based on 136 o The payload contents of the packets on the wire 138 o The characteristics of the packets such as packet size and timing 140 Use of temporary addresses will not prevent such payload based 141 correlation. 143 2. Background 145 This section discusses the problem in more detail, provides context 146 for evaluating the significance of the concerns in specific 147 environments and makes comparisons with existing practices. 149 2.1. Extended Use of the Same Identifier 151 The use of a non-changing interface identifier to form addresses is a 152 specific instance of the more general case where a constant 153 identifier is reused over an extended period of time and in multiple 154 independent activities. Anytime the same identifier is used in 155 multiple contexts, it becomes possible for that identifier to be used 156 to correlate seemingly unrelated activity. For example, a network 157 sniffer placed strategically on a link across which all traffic to/ 158 from a particular host crosses could keep track of which destinations 159 a node communicated with and at what times. Such information can in 160 some cases be used to infer things, such as what hours an employee 161 was active, when someone is at home, etc. Although it might appear 162 that changing an address regularly in such environments would be 163 desirable to lessen privacy concerns, it should be noted that the 164 network prefix portion of an address also serves as a constant 165 identifier. All nodes at (say) a home, would have the same network 166 prefix, which identifies the topological location of those nodes. 167 This has implications for privacy, though not at the same granularity 168 as the concern that this document addresses. Specifically, all nodes 169 within a home could be grouped together for the purposes of 170 collecting information. If the network contains a very small number 171 of nodes, say just one, changing just the interface identifier will 172 not enhance privacy at all, since the prefix serves as a constant 173 identifier. 175 One of the requirements for correlating seemingly unrelated 176 activities is the use (and reuse) of an identifier that is 177 recognizable over time within different contexts. IP addresses 178 provide one obvious example, but there are more. Many nodes also 179 have DNS names associated with their addresses, in which case the DNS 180 name serves as a similar identifier. Although the DNS name 181 associated with an address is more work to obtain (it may require a 182 DNS query) the information is often readily available. In such 183 cases, changing the address on a machine over time would do little to 184 address the concerns raised in this document, unless the DNS name is 185 changed as well (see Section 4). 187 Web browsers and servers typically exchange "cookies" with each other 188 [COOKIES]. Cookies allow web servers to correlate a current activity 189 with a previous activity. One common usage is to send back targeted 190 advertising to a user by using the cookie supplied by the browser to 191 identify what earlier queries had been made (e.g., for what type of 192 information). Based on the earlier queries, advertisements can be 193 targeted to match the (assumed) interests of the end-user. 195 The use of a constant identifier within an address is of special 196 concern because addresses are a fundamental requirement of 197 communication and cannot easily be hidden from eavesdroppers and 198 other parties. Even when higher layers encrypt their payloads, 199 addresses in packet headers appear in the clear. Consequently, if a 200 mobile host (e.g., laptop) accessed the network from several 201 different locations, an eavesdropper might be able to track the 202 movement of that mobile host from place to place, even if the upper 203 layer payloads were encrypted. 205 The security and privacy implications of IPv6 addresses are discussed 206 in detail in [RFC7721], [RFC7707], and [RFC7217]. 208 2.2. Possible Approaches 210 One way to avoid having a stable non-changing address is to use 211 DHCPv6[DHCPV6] for obtaining addresses. Section 12 of [DHCPV6] 212 discusses the use of DHCPv6 for the assignment and management of 213 "temporary addresses", which are never renewed and provide the same 214 property of temporary addresses described in this document with 215 regards to the privacy concern. 217 Another approach, compatible with the stateless address 218 autoconfiguration architecture, would be to change the interface 219 identifier portion of an address over time. Changing the interface 220 identifier can make it more difficult to look at the IP addresses in 221 independent transactions and identify which ones actually correspond 222 to the same node, both in the case where the routing prefix portion 223 of an address changes and when it does not. 225 Many machines function as both clients and servers. In such cases, 226 the machine would need a DNS name for its use as a server. Whether 227 the address stays fixed or changes has little privacy implication 228 since the DNS name remains constant and serves as a constant 229 identifier. When acting as a client (e.g., initiating 230 communication), however, such a machine may want to vary the 231 addresses it uses. In such environments, one may need multiple 232 addresses: a "stable" address and public address registered in the 233 DNS, that is used to accept incoming connection requests from other 234 machines, and a "temporary" address used to shield the identity of 235 the client when it initiates communication. These two cases are 236 roughly analogous to telephone numbers and caller ID, where a user 237 may list their telephone number in the public phone book, but disable 238 the display of its number via caller ID when initiating calls. 240 On the other hand, a machine that functions only as a client may want 241 to employ only temporary addresses for public communication. 243 To make it difficult to make educated guesses as to whether two 244 different interface identifiers belong to the same node, the 245 algorithm for generating alternate identifiers must include input 246 that has an unpredictable component from the perspective of the 247 outside entities that are collecting information. 249 [I-D.gont-6man-non-stable-iids] specifies requirements for temporary 250 addresses. This document specifies a number of algorithms for 251 generating temporary addresses that comply with the aforementioned 252 requirements. 254 3. Protocol Description 256 The goal of this section is to define procedures that: 258 1. Do not result in any changes to the basic behavior of addresses 259 generated via stateless address autoconfiguration [RFC4862]. 261 2. Create temporary addresses based on an unpredictable interface 262 identifier for the purpose of initiating outgoing sessions. 263 These temporary addresses would be used for a short period of 264 time (hours to days) and would then be deprecated. Deprecated 265 address can continue to be used for already established 266 connections, but are not used to initiate new connections. New 267 temporary addresses are generated periodically to replace 268 temporary addresses that expire, with the exact time between 269 address generation a matter of local policy. 271 3. Produce a sequence of temporary global scope addresses from a 272 sequence of interface identifiers that appear to be random in the 273 sense that it is difficult for an outside observer to predict a 274 future address (or identifier) based on a current one and it is 275 difficult to determine previous addresses (or identifiers) 276 knowing only the present one. 278 4. By default, generate one address for each prefix to be employed 279 for stateless address autoconfiguration. 281 3.1. Assumptions 283 The following algorithm assumes that for a given temporary address, 284 an implementation can determine the prefix from which it was 285 generated. When a temporary address is deprecated, a new temporary 286 address is generated. The specific valid and preferred lifetimes for 287 the new address are dependent on the corresponding lifetime values 288 set for the prefix from which it was generated. 290 Finally, this document assumes that when a node initiates outgoing 291 communication, temporary addresses can be given preference over 292 stable addresses (if available), when the device is configured to do 293 so. [RFC6724] mandates implementations to provide a mechanism, which 294 allows an application to configure its preference for temporary 295 addresses over stable addresses. It also allows for an 296 implementation to prefer temporary addresses by default, so that the 297 connections initiated by the node can use temporary addresses without 298 requiring application-specific enablement. This document also 299 assumes that an API will exist that allows individual applications to 300 indicate whether they prefer to use temporary or stable addresses and 301 override the system defaults. 303 3.2. Generation Of Randomized Interface Identifiers 305 The following subsections specificy some possible algorithms for 306 generating temporary interface identifiers that comply with the 307 requirements in [I-D.gont-6man-non-stable-iids]. 309 3.2.1. Randomized IIDs 311 One possible approach would be to select a pseudorandom number of the 312 appropriate length. A node employing this algorithm should generate 313 IIDs as follows: 315 1. Obtain a random number (see [RFC4086] for randomness requirements 316 for security) 318 2. The Interface Identifier is obtained by taking as many bits from 319 the aforementioned random number (obtained in the previous step) 320 as necessary. 322 We note that [RFC4291] requires that the Interface IDs of all 323 unicast addresses (except those that start with the binary 324 value 000) be 64 bits long. However, the method discussed in 325 this document could be employed for generating Interface IDs 326 of any arbitrary length, albeit at the expense of reduced 327 entropy (when employing Interface IDs smaller than 64 bits). 329 3. The resulting Interface Identifier SHOULD be compared against the 330 reserved IPv6 Interface Identifiers [RFC5453] [IANA-RESERVED-IID] 331 and against those Interface Identifiers already employed in an 332 address of the same network interface and the same network 333 prefix. In the event that an unacceptable identifier has been 334 generated, a new interface identifier should be generated, by 335 repeating the algorithm from the first step. 337 3.2.2. Hash-based generation of randomized interface identifiers 339 The algorithm in [RFC7217] can be augmented for the generation of 340 temporary addresses. The benefit of this would be that a node could 341 employ a single algorithm for generating stable and temporary 342 addresses, by employing appropriate parameters. 344 Nodes would employ the following algorithm for generating the 345 temporary IID: 347 1. Compute a random identifier with the expression: 349 RID = F(Prefix, MAC_Address, Network_ID, Time, DAD_Counter, 350 secret_key) 351 Where: 353 RID: 354 Random Identifier 356 F(): 357 A pseudorandom function (PRF) that MUST NOT be computable from 358 the outside (without knowledge of the secret key). F() MUST 359 also be difficult to reverse, such that it resists attempts to 360 obtain the secret_key, even when given samples of the output 361 of F() and knowledge or control of the other input parameters. 362 F() SHOULD produce an output of at least 64 bits. F() could 363 be implemented as a cryptographic hash of the concatenation of 364 each of the function parameters. SHA-1 [FIPS-SHS] and SHA-256 365 are two possible options for F(). Note: MD5 [RFC1321] is 366 considered unacceptable for F() [RFC6151]. 368 Prefix: 369 The prefix to be used for SLAAC, as learned from an ICMPv6 370 Router Advertisement message. 372 MAC_Address: 373 The MAC address corresponding to the underlying network 374 interface card. Employing the MAC address in this expression 375 (in replacement of the Net_Iface parameter of the expression 376 in RFC7217) means that the re-generation of a randomized MAC 377 address will result in a different temporary address. 379 Network_ID: 380 Some network-specific data that identifies the subnet to which 381 this interface is attached -- for example, the IEEE 802.11 382 Service Set Identifier (SSID) corresponding to the network to 383 which this interface is associated. Additionally, Simple DNA 384 [RFC6059] describes ideas that could be leveraged to generate 385 a Network_ID parameter. This parameter is SHOULD be employed 386 if some form of "Network_ID" is available. 388 Time: 389 An implementation-dependent representation of time. One 390 possible example is the representation in UNIX-like systems 391 [OPEN-GROUP], that measure time in terms of the number of 392 seconds elapsed since the Epoch (00:00:00 Coordinated 393 Universal Time (UTC), 1 January 1970). 395 DAD_Counter: 396 A counter that is employed to resolve Duplicate Address 397 Detection (DAD) conflicts. 399 secret_key: 400 A secret key that is not known by the attacker. The secret 401 key SHOULD be of at least 128 bits. It MUST be initialized to 402 a pseudo-random number (see [RFC4086] for randomness 403 requirements for security) when the operating system is 404 installed or when the IPv6 protocol stack is "bootstrapped" 405 for the first time. 407 2. The Interface Identifier is finally obtained by taking as many 408 bits from the RID value (computed in the previous step) as 409 necessary, starting from the least significant bit. The 410 resulting Interface Identifier SHOULD be compared against the 411 reserved IPv6 Interface Identifiers [RFC5453] [IANA-RESERVED-IID] 412 and against those Interface Identifiers already employed in an 413 address of the same network interface and the same network 414 prefix. In the event that an unacceptable identifier has been 415 generated, the value DAD_Counter should be incremented by 1, and 416 the algorithm should be restarted from the first step. 418 3.3. Generating Temporary Addresses 420 [RFC4862] describes the steps for generating a link-local address 421 when an interface becomes enabled as well as the steps for generating 422 addresses for other scopes. This document extends [RFC4862] as 423 follows. When processing a Router Advertisement with a Prefix 424 Information option carrying a global scope prefix for the purposes of 425 address autoconfiguration (i.e., the A bit is set), the node 426 implementing this specification MUST perform the following steps: 428 1. Process the Prefix Information Option as defined in [RFC4862], 429 either creating a new stable address or adjusting the lifetimes 430 of existing addresses, both stable and temporary. If a received 431 option will extend the lifetime of a stable address, the 432 lifetimes of temporary addresses should be extended, subject to 433 the overall constraint that no temporary addresses should ever 434 remain "valid" or "preferred" for a time longer than 435 (TEMP_VALID_LIFETIME - DESYNC_FACTOR) or (TEMP_PREFERRED_LIFETIME 436 - DESYNC_FACTOR) respectively. The configuration variables 437 TEMP_VALID_LIFETIME and TEMP_PREFERRED_LIFETIME correspond to 438 approximate target lifetimes for temporary addresses. 440 2. One way an implementation can satisfy the above constraints is to 441 associate with each temporary address a creation time (called 442 CREATION_TIME) that indicates the time at which the address was 443 created. When updating the preferred lifetime of an existing 444 temporary address, it would be set to expire at whichever time is 445 earlier: the time indicated by the received lifetime or 446 (CREATION_TIME + TEMP_PREFERRED_LIFETIME - DESYNC_FACTOR). A 447 similar approach can be used with the valid lifetime. 449 3. When a new stable address is created as described in [RFC4862], 450 or if the node has not configured any temporary address for the 451 corresponding prefix, the node SHOULD create a new temporary 452 address for such prefix. 454 4. When creating a temporary address, the lifetime values MUST be 455 derived from the corresponding prefix as follows: 457 * Its Valid Lifetime is the lower of the Valid Lifetime of the 458 stable address (if available) or TEMP_VALID_LIFETIME 460 * Its Preferred Lifetime is the lower of the Preferred Lifetime 461 of the stable address (if available) or 462 TEMP_PREFERRED_LIFETIME - DESYNC_FACTOR. 464 5. A temporary address is created only if this calculated Preferred 465 Lifetime is greater than REGEN_ADVANCE time units. In 466 particular, an implementation MUST NOT create a temporary address 467 with a zero Preferred Lifetime. 469 6. New temporary addresses MUST be created by appending the 470 interface's current randomized interface identifier to the prefix 471 that was received. 473 7. The node MUST Perform duplicate address detection (DAD) on the 474 generated temporary address. If DAD indicates the address is 475 already in use, the node MUST generate a new randomized interface 476 identifier, and repeat the previous steps as appropriate up to 477 TEMP_IDGEN_RETRIES times. If after TEMP_IDGEN_RETRIES 478 consecutive attempts no non-unique address was generated, the 479 node MUST log a system error and MUST NOT attempt to generate 480 temporary addresses for that interface. Note that DAD MUST be 481 performed on every unicast address generated from this randomized 482 interface identifier. 484 3.4. Expiration of Temporary Addresses 486 When a temporary address becomes deprecated, a new one MUST be 487 generated. This is done by repeating the actions described in 488 Section 3.3, starting at step 3). Note that, except for the 489 transient period when a temporary address is being regenerated, in 490 normal operation at most one temporary address per prefix should be 491 in a non-deprecated state at any given time on a given interface. 492 Note that if a temporary address becomes deprecated as result of 493 processing a Prefix Information Option with a zero Preferred 494 Lifetime, then a new temporary address MUST NOT be generated. To 495 ensure that a preferred temporary address is always available, a new 496 temporary address SHOULD be regenerated slightly before its 497 predecessor is deprecated. This is to allow sufficient time to avoid 498 race conditions in the case where generating a new temporary address 499 is not instantaneous, such as when duplicate address detection must 500 be run. The node SHOULD start the address regeneration process 501 REGEN_ADVANCE time units before a temporary address would actually be 502 deprecated. 504 As an optional optimization, an implementation MAY remove a 505 deprecated temporary address that is not in use by applications or 506 upper-layers as detailed in Section 6. 508 3.5. Regeneration of Randomized Interface Identifiers 510 The frequency at which temporary addresses changes depends on how a 511 device is being used (e.g., how frequently it initiates new 512 communication) and the concerns of the end user. The most egregious 513 privacy concerns appear to involve addresses used for long periods of 514 time (weeks to months to years). The more frequently an address 515 changes, the less feasible collecting or coordinating information 516 keyed on interface identifiers becomes. Moreover, the cost of 517 collecting information and attempting to correlate it based on 518 interface identifiers will only be justified if enough addresses 519 contain non-changing identifiers to make it worthwhile. Thus, having 520 large numbers of clients change their address on a daily or weekly 521 basis is likely to be sufficient to alleviate most privacy concerns. 523 There are also client costs associated with having a large number of 524 addresses associated with a node (e.g., in doing address lookups, the 525 need to join many multicast groups, etc.). Thus, changing addresses 526 frequently (e.g., every few minutes) may have performance 527 implications. 529 Nodes following this specification SHOULD generate new temporary 530 addresses on a periodic basis. This can be achieved automatically by 531 generating a new randomized interface identifier at least once every 532 (TEMP_PREFERRED_LIFETIME - REGEN_ADVANCE - DESYNC_FACTOR) time units. 533 As described above, generating a new temporary address REGEN_ADVANCE 534 time units before a temporary address becomes deprecated produces 535 addresses with a preferred lifetime no larger than 536 TEMP_PREFERRED_LIFETIME. The value DESYNC_FACTOR is a random value 537 (different for each client) that ensures that clients don't 538 synchronize with each other and generate new addresses at exactly the 539 same time. When the preferred lifetime expires, a new temporary 540 address MUST be generated using the new randomized interface 541 identifier. 543 Because the precise frequency at which it is appropriate to generate 544 new addresses varies from one environment to another, implementations 545 SHOULD provide end users with the ability to change the frequency at 546 which addresses are regenerated. The default value is given in 547 TEMP_PREFERRED_LIFETIME and is one day. In addition, the exact time 548 at which to invalidate a temporary address depends on how 549 applications are used by end users. Thus, the suggested default 550 value of one week (TEMP_VALID_LIFETIME) may not be appropriate in all 551 environments. Implementations SHOULD provide end users with the 552 ability to override both of these default values. 554 Finally, when an interface connects to a new link, a new set of 555 temporary addresses MUST be generated immediately. If a device moves 556 from one ethernet to another, generating a new set of temporary 557 addresses ensures that the device uses different randomized interface 558 identifiers for the temporary addresses associated with the two 559 links, making it more difficult to correlate addresses from the two 560 different links as being from the same node. The node MAY follow any 561 process available to it, to determine that the link change has 562 occurred. One such process is described by Detecting Network 563 Attachment [DNA]. 565 3.6. Deployment Considerations 567 Devices implementing this specification MUST provide a way for the 568 end user to explicitly enable or disable the use of temporary 569 addresses. In addition, a site might wish to disable the use of 570 temporary addresses in order to simplify network debugging and 571 operations. Consequently, implementations SHOULD provide a way for 572 trusted system administrators to enable or disable the use of 573 temporary addresses. 575 Additionally, sites might wish to selectively enable or disable the 576 use of temporary addresses for some prefixes. For example, a site 577 might wish to disable temporary address generation for "Unique local" 578 [RFC4193] prefixes while still generating temporary addresses for all 579 other global prefixes. Another site might wish to enable temporary 580 address generation only for the prefixes 2001::/16 and 2002::/16 581 while disabling it for all other prefixes. To support this behavior, 582 implementations SHOULD provide a way to enable and disable generation 583 of temporary addresses for specific prefix subranges. This per- 584 prefix setting SHOULD override the global settings on the node with 585 respect to the specified prefix subranges. Note that the pre-prefix 586 setting can be applied at any granularity, and not necessarily on a 587 per subnet basis. 589 The use of temporary addresses may cause unexpected difficulties with 590 some applications. As described below, some servers refuse to accept 591 communications from clients for which they cannot map the IP address 592 into a DNS name. In addition, some applications may not behave 593 robustly if temporary addresses are used and an address expires 594 before the application has terminated, or if it opens multiple 595 sessions, but expects them to all use the same addresses. 597 If a very small number of nodes (say only one) use a given prefix for 598 extended periods of time, just changing the interface identifier part 599 of the address may not be sufficient to ensure privacy, since the 600 prefix acts as a constant identifier. The procedures described in 601 this document are most effective when the prefix is reasonably non 602 static or is used by a fairly large number of nodes. 604 4. Implications of Changing Interface Identifiers 606 The desires of protecting individual privacy versus the desire to 607 effectively maintain and debug a network can conflict with each 608 other. Having clients use addresses that change over time will make 609 it more difficult to track down and isolate operational problems. 610 For example, when looking at packet traces, it could become more 611 difficult to determine whether one is seeing behavior caused by a 612 single errant machine, or by a number of them. 614 Some servers refuse to grant access to clients for which no DNS name 615 exists. That is, they perform a DNS PTR query to determine the DNS 616 name, and may then also perform an AAAA query on the returned name to 617 verify that the returned DNS name maps back into the address being 618 used. Consequently, clients not properly registered in the DNS may 619 be unable to access some services. As noted earlier, however, a 620 node's DNS name (if non-changing) serves as a constant identifier. 621 The wide deployment of the extension described in this document could 622 challenge the practice of inverse-DNS-based "authentication," which 623 has little validity, though it is widely implemented. In order to 624 meet server challenges, nodes could register temporary addresses in 625 the DNS using random names (for example a string version of the 626 random address itself). 628 Use of the extensions defined in this document may complicate 629 debugging and other operational troubleshooting activities. 630 Consequently, it may be site policy that temporary addresses should 631 not be used. Consequently, implementations MUST provide a method for 632 the end user or trusted administrator to override the use of 633 temporary addresses. 635 5. Defined Constants 637 Constants defined in this document include: 639 TEMP_VALID_LIFETIME -- Default value: 1 week. Users should be able 640 to override the default value. 642 TEMP_PREFERRED_LIFETIME -- Default value: 1 day. Users should be 643 able to override the default value. 645 REGEN_ADVANCE -- 5 seconds 647 MAX_DESYNC_FACTOR -- 10 minutes. Upper bound on DESYNC_FACTOR. 649 DESYNC_FACTOR -- A random value within the range 0 - 650 MAX_DESYNC_FACTOR. It is computed once at system start (rather than 651 each time it is used) and must never be greater than 652 (TEMP_VALID_LIFETIME - REGEN_ADVANCE). 654 TEMP_IDGEN_RETRIES -- Default value: 3 656 6. Future Work 658 An implementation might want to keep track of which addresses are 659 being used by upper layers so as to be able to remove a deprecated 660 temporary address from internal data structures once no upper layer 661 protocols are using it (but not before). This is in contrast to 662 current approaches where addresses are removed from an interface when 663 they become invalid [RFC4862], independent of whether or not upper 664 layer protocols are still using them. For TCP connections, such 665 information is available in control blocks. For UDP-based 666 applications, it may be the case that only the applications have 667 knowledge about what addresses are actually in use. Consequently, an 668 implementation generally will need to use heuristics in deciding when 669 an address is no longer in use. 671 The determination as to whether to use stable versus temporary 672 addresses can in some cases only be made by an application. For 673 example, some applications may always want to use temporary 674 addresses, while others may want to use them only in some 675 circumstances or not at all. Suitable API extensions will likely 676 need to be developed to enable individual applications to indicate 677 with sufficient granularity their needs with regards to the use of 678 temporary addresses. See 679 [I-D.gont-taps-address-usage-problem-statement] for further details. 680 Recommendations on DNS practices to avoid the problem described in 681 Section 4 when reverse DNS lookups fail may be needed. [DNSOP] 682 contains a more detailed discussion of the DNS related issues. 684 While this document discusses ways of obscuring a user's IP address, 685 the method described is believed to be ineffective against 686 sophisticated forms of traffic analysis. To increase effectiveness, 687 one may need to consider use of more advanced techniques, such as 688 Onion Routing [ONION]. 690 7. Security Considerations 692 Ingress filtering has been and is being deployed as a means of 693 preventing the use of spoofed source addresses in Distributed Denial 694 of Service (DDoS) attacks. In a network with a large number of 695 nodes, new temporary addresses are created at a fairly high rate. 696 This might make it difficult for ingress filtering mechanisms to 697 distinguish between legitimately changing temporary addresses and 698 spoofed source addresses, which are "in-prefix" (They use a 699 topologically correct prefix and non-existent interface ID). This 700 can be addressed by using access control mechanisms on a per address 701 basis on the network egress point. 703 The security and privacy implications of IPv6 addresses are discussed 704 in great detail in [RFC7721] and [RFC7217]. 706 8. Significant Changes from RFC RFC4941 708 This section summarizes the changes in this document relative to RFC 709 4941 that an implementer of RFC 4941 should be aware of. 711 1. The algorithm to generate randomized interface identifiers was 712 replaced by two possible alternative algorithms. 714 2. Generation of stable addresses is not implied or required by this 715 document. 717 3. Temporary addresses are *not* disabled by default. 719 9. Acknowledgements 721 This document is based on [RFC4941] (authored by T. Narten, R. 722 Draves, and S. Krishnan) and [I-D.gont-6man-non-stable-iids] 723 (authored by F. Gont, C. Huitema, G. Gont, and M. Garcia Corbo). 725 10. References 727 10.1. Normative References 729 [MD5] Rivest, R., "The MD5 Message-Digest Algorithm", RFC 1321, 730 April 1992. 732 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 733 Requirement Levels", RFC 2119, March 1997. 735 [RFC4086] Eastlake 3rd, D., Schiller, J., and S. Crocker, 736 "Randomness Requirements for Security", BCP 106, RFC 4086, 737 DOI 10.17487/RFC4086, June 2005, 738 . 740 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 741 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 742 . 744 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 745 Architecture", RFC 4291, DOI 10.17487/RFC4291, 746 February 2006, . 748 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 749 Address Autoconfiguration", RFC 4862, DOI 10.17487/ 750 RFC4862, September 2007, 751 . 753 [RFC4941] Narten, T., Draves, R., and S. Krishnan, "Privacy 754 Extensions for Stateless Address Autoconfiguration in 755 IPv6", RFC 4941, DOI 10.17487/RFC4941, September 2007, 756 . 758 [RFC5453] Krishnan, S., "Reserved IPv6 Interface Identifiers", 759 RFC 5453, DOI 10.17487/RFC5453, February 2009, 760 . 762 [RFC6724] Thaler, D., Ed., Draves, R., Matsumoto, A., and T. Chown, 763 "Default Address Selection for Internet Protocol Version 6 764 (IPv6)", RFC 6724, DOI 10.17487/RFC6724, September 2012, 765 . 767 [RFC7136] Carpenter, B. and S. Jiang, "Significance of IPv6 768 Interface Identifiers", RFC 7136, DOI 10.17487/RFC7136, 769 February 2014, . 771 [RFC7217] Gont, F., "A Method for Generating Semantically Opaque 772 Interface Identifiers with IPv6 Stateless Address 773 Autoconfiguration (SLAAC)", RFC 7217, DOI 10.17487/ 774 RFC7217, April 2014, 775 . 777 10.2. Informative References 779 [COOKIES] Kristol, D. and L. Montulli, "HTTP State Management 780 Mechanism", RFC 2965, October 2000. 782 [DDNS] Vixie, P., Thomson, S., Rekhter, Y., and J. Bound, 783 "Dynamic Updates in the Domain Name System (DNS UPDATE)", 784 RFC 2136, April 1997. 786 [DHCP] Droms, R., "Dynamic Host Configuration Protocol", 787 RFC 2131, March 1997. 789 [DHCPV6] Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C., 790 and M. Carney, "Dynamic Host Configuration Protocol for 791 IPv6 (DHCPv6)", RFC 3315, July 2003. 793 [DNA] Choi, J. and G. Daley, "Detecting Network Attachment in 794 IPv6 Goals", draft-ietf-dna-goals-04 (work in progress), 795 December 2004. 797 [DNSOP] Durand, A., Ihren, J., and P. Savola, "Operational 798 Considerations and Issues with IPv6 DNS", 799 draft-ietf-dnsop-ipv6-dns-issues-10 (work in progress), 800 October 2004. 802 [FIPS-SHS] 803 NIST, "Secure Hash Standard (SHS)", FIPS 804 Publication 180-4, March 2012, . 807 [I-D.gont-6man-non-stable-iids] 808 Gont, F., Huitema, C., Gont, G., and M. Corbo, 809 "Recommendation on Temporary IPv6 Interface Identifiers", 810 draft-gont-6man-non-stable-iids-01 (work in progress), 811 March 2017. 813 [I-D.gont-taps-address-usage-problem-statement] 814 Gont, F., Gont, G., Corbo, M., and C. Huitema, "Problem 815 Statement Regarding IPv6 Address Usage", 816 draft-gont-taps-address-usage-problem-statement-00 (work 817 in progress), February 2018. 819 [IANA-RESERVED-IID] 820 IANA, "Reserved IPv6 Interface Identifiers", 821 . 823 [ONION] Reed, MGR., Syverson, PFS., and DMG. Goldschlag, "Proxies 824 for Anonymous Routing", Proceedings of the 12th Annual 825 Computer Security Applications Conference, San Diego, CA, 826 December 1996. 828 [OPEN-GROUP] 829 The Open Group, "The Open Group Base Specifications Issue 830 7 / IEEE Std 1003.1-2008, 2016 Edition", Section 831 4.16 Seconds Since the Epoch, 2016, . 835 [RFC1321] Rivest, R., "The MD5 Message-Digest Algorithm", RFC 1321, 836 DOI 10.17487/RFC1321, April 1992, 837 . 839 [RFC6059] Krishnan, S. and G. Daley, "Simple Procedures for 840 Detecting Network Attachment in IPv6", RFC 6059, 841 DOI 10.17487/RFC6059, November 2010, 842 . 844 [RFC6151] Turner, S. and L. Chen, "Updated Security Considerations 845 for the MD5 Message-Digest and the HMAC-MD5 Algorithms", 846 RFC 6151, DOI 10.17487/RFC6151, March 2011, 847 . 849 [RFC7707] Gont, F. and T. Chown, "Network Reconnaissance in IPv6 850 Networks", RFC 7707, DOI 10.17487/RFC7707, March 2016, 851 . 853 [RFC7721] Cooper, A., Gont, F., and D. Thaler, "Security and Privacy 854 Considerations for IPv6 Address Generation Mechanisms", 855 RFC 7721, DOI 10.17487/RFC7721, March 2016, 856 . 858 Authors' Addresses 860 Fernando Gont 861 SI6 Networks / UTN-FRH 862 Evaristo Carriego 2644 863 Haedo, Provincia de Buenos Aires 1706 864 Argentina 866 Phone: +54 11 4650 8472 867 Email: fgont@si6networks.com 868 URI: http://www.si6networks.com 870 Suresh Krishnan 871 Ericsson Research 872 8400 Decarie Blvd. 873 Town of Mount Royal, QC 874 Canada 876 Email: suresh.krishnan@ericsson.com