idnits 2.17.1 draft-ietf-dprive-phase2-requirements-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Using lowercase 'not' together with uppercase 'MUST', 'SHALL', 'SHOULD', or 'RECOMMENDED' is not an accepted usage according to RFC 2119. Please use uppercase 'NOT' together with RFC 2119 keywords (if that is what you mean). Found 'MUST not' in this paragraph: 11. For secure transports using TLS, TLS 1.3 (or later versions) MUST be supported and downgrades from TLS 1.3 to prior versions MUST not occur. -- The document date (November 02, 2020) is 1271 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- -- Looks like a reference, but probably isn't: '1' on line 439 -- Looks like a reference, but probably isn't: '2' on line 441 ** Obsolete normative reference: RFC 8499 (Obsoleted by RFC 9499) -- Obsolete informational reference (is this intentional?): RFC 7816 (Obsoleted by RFC 9156) Summary: 1 error (**), 0 flaws (~~), 2 warnings (==), 4 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 DPRIVE J. Livingood 3 Internet-Draft Comcast 4 Intended status: Informational A. Mayrhofer 5 Expires: May 6, 2021 nic.at GmbH 6 B. Overeinder 7 NLnet Labs 8 November 02, 2020 10 DNS Privacy Requirements for Exchanges between Recursive Resolvers and 11 Authoritative Servers 12 draft-ietf-dprive-phase2-requirements-02 14 Abstract 16 This document describes requirements and considerations for adding 17 confidentiality to DNS exchanges between recursive resolvers and 18 authoritative servers. The intent of this document is to guide 19 Internet Drafts in the DNS Private Exchange (DPRIVE) Working Group 20 pertaining to recursive to authorized name servers, with the stated 21 requirements and considerations. 23 Status of This Memo 25 This Internet-Draft is submitted in full conformance with the 26 provisions of BCP 78 and BCP 79. 28 Internet-Drafts are working documents of the Internet Engineering 29 Task Force (IETF). Note that other groups may also distribute 30 working documents as Internet-Drafts. The list of current Internet- 31 Drafts is at https://datatracker.ietf.org/drafts/current/. 33 Internet-Drafts are draft documents valid for a maximum of six months 34 and may be updated, replaced, or obsoleted by other documents at any 35 time. It is inappropriate to use Internet-Drafts as reference 36 material or to cite them other than as "work in progress." 38 This Internet-Draft will expire on May 6, 2021. 40 Copyright Notice 42 Copyright (c) 2020 IETF Trust and the persons identified as the 43 document authors. All rights reserved. 45 This document is subject to BCP 78 and the IETF Trust's Legal 46 Provisions Relating to IETF Documents 47 (https://trustee.ietf.org/license-info) in effect on the date of 48 publication of this document. Please review these documents 49 carefully, as they describe your rights and restrictions with respect 50 to this document. Code Components extracted from this document must 51 include Simplified BSD License text as described in Section 4.e of 52 the Trust Legal Provisions and are provided without warranty as 53 described in the Simplified BSD License. 55 Table of Contents 57 1. Introduction & Scope . . . . . . . . . . . . . . . . . . . . 2 58 2. Document Work Via GitHub . . . . . . . . . . . . . . . . . . 3 59 3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 60 4. Threat Model and Problem Statement . . . . . . . . . . . . . 3 61 5. Features to Provide Confidentiality . . . . . . . . . . . . . 4 62 5.1. Requirements . . . . . . . . . . . . . . . . . . . . . . 4 63 5.2. Optional Features . . . . . . . . . . . . . . . . . . . . 5 64 6. Security Considerations . . . . . . . . . . . . . . . . . . . 5 65 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 6 66 8. Changelog . . . . . . . . . . . . . . . . . . . . . . . . . . 6 67 9. APPENDIX: Perspectives and Use Cases . . . . . . . . . . . . 6 68 9.1. The User Perspective and Use Cases . . . . . . . . . . . 6 69 9.2. The Operator Perspective and Use Cases . . . . . . . . . 7 70 9.3. The Implementor / Software Vendor Perspective and Use 71 Cases . . . . . . . . . . . . . . . . . . . . . . . . . . 9 72 10. References . . . . . . . . . . . . . . . . . . . . . . . . . 9 73 10.1. Normative References . . . . . . . . . . . . . . . . . . 9 74 10.2. Informative References . . . . . . . . . . . . . . . . . 9 75 10.3. URIs . . . . . . . . . . . . . . . . . . . . . . . . . . 10 76 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . 10 77 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 10 79 1. Introduction & Scope 81 The 2018 approved charter of the IETF DPRIVE Working Group [1] 82 contains milestones related to confidentiality aspects of DNS 83 transactions between the recursive resolver and authoritative name 84 servers. 86 This is also reflected in the DPRIVE milestones [2], which (as of 87 October 2019) contains two relevant milestones: 89 Develop requirements for adding confidentiality to DNS exchanges 90 between recursive resolvers and authoritative servers (unpublished 91 document). 93 Investigate potential solutions for adding confidentiality to DNS 94 exchanges involving authoritative servers (Experimental). 96 This document intends to cover the first milestone for defining 97 requirements for adding confidentiality to DNS exchanges between 98 recursive resolvers and authoritative servers. This may in turn lead 99 to progress in investigating, developing and standardizing potential 100 experimental methods of meeting those requirements. 102 The motivation for this work is to extend the confidentiality methods 103 used between a user's stub resolver and a recursive resolver to the 104 recursive queries sent by recursive resolvers in response to a DNS 105 lookup (when a cache miss occurs and the server must perform 106 recursion to obtain a response to the query). A recursive resolver 107 will send queries to root servers, to Top Level Domain (TLD) servers, 108 to authoritative second level domain servers and potentially to other 109 authoritative DNS servers and each of these query/response 110 transactions presents an opportunity to extend the confidentiality of 111 user DNS queries. 113 2. Document Work Via GitHub 115 The authors are working on this document via GitHub at 116 https://github.com/alex-nicat/ietf-dprive-phase2-requirements. 117 Feedback via pull requests and issues are invited there. 119 3. Terminology 121 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 122 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 123 "OPTIONAL" in this document are to be interpreted as described in BCP 124 14 [RFC2119] [RFC8174] when, and only when, they appear in all 125 capitals, as shown here. 127 This document also makes use of DNS Terminology defined in [RFC8499] 129 4. Threat Model and Problem Statement 131 Currently, protocols such as DoT provide encryption between the 132 user's stub resolver and a recursive resolver. This potentially 133 provides (1) protection from observation of end user DNS queries and 134 responses, (2) protection from on-the-wire modification DNS queries 135 or responses (including potentially forcing a downgrade to an 136 unencrypted communication). Of course, observation and modification 137 are still possible when performed by the recursive resolver, which 138 decrypts queries, serves a response from cache or performs recursion 139 to obtain a response (or synthesizes a response), and then encrypts 140 the response and sends it back to the user's stub resolver. 142 But observation and modification threats still exist when a recursive 143 resolver must perform DNS recursion, from the root to TLD to 144 authoritative servers. This document specifies requirements for 145 filling those gaps. 147 5. Features to Provide Confidentiality 149 Confidentialty can be provided using a combination of techniques. 150 This section describes the protocol implementation requirements and 151 optional features that can be used to provide confidentiality. 153 5.1. Requirements 155 1. Each implementing party MUST be able to independently take 156 incremental steps to meet requirements without the need for 157 close coordination (e.g. loosely coupled) 159 2. A recursive resolver that supports recursive-to-authoritative 160 DNS encryption MUST be able to determine whether or not a given 161 authoritative name server to which it intends to connect also 162 supports recursive-to-authoritative DNS encryption. 164 3. An authoritative name server that supports recursive-to- 165 authoritative DNS encryption MUST be able to indicate that it 166 supports recursive-to-authoritative DNS encryption in a way that 167 facilitates (2). 169 4. An authoritative name server that does not support recursive-to- 170 authoritative MUST NOT have to make any changes to facilitate 171 (2). 173 5. The secure transport MUST only be established when referential 174 integrity can be verified, MUST NOT have circular dependencies, 175 and MUST be easily analyzed for diagnostic purposes. 177 6. Each implementing party MUST be able to negotiate use of a 178 secure transport protocol or other DNS privacy protections in a 179 manner that enables operators to perform appropriate performance 180 and security monitoring, conduct relevant research, etc. 182 7. The authoritative domain owner or their administrator MUST have 183 the option to specify their secure transport preferences (e.g. 184 what specific protocols are supported). This SHALL include a 185 method to publish a list of secure transport protocols (e.g. 186 DoH, DoT and other future protocols not yet developed). In 187 addition this SHALL include whether a secure transport protocol 188 MUST always be used (non-downgradable) or whether a secure 189 transport protocol MAY be used on an opportunistic (not strict) 190 basis in recognition that some servers for a domain might use a 191 secure transport protocol and others might not. 193 8. The authoritative domain owner or their administrator MUST have 194 the option to vary their preferences on an authoritative 195 nameserver to nameserver basis, due to the fact that 196 administration of a particular DNS zone may be delegated to 197 multiple parties (such as several CDNs), each of which may have 198 different technical capabilities. This includes that some 199 servers for a domain may use secure transport and others may 200 not, as it is common for a given name server to be authoritative 201 for multiple zones. 203 9. A given name server may be authoritative for multiple zones. As 204 such, a name server MAY support use of a secure transport 205 protocol for one zone, but not for another. 207 10. The specification of secure transport preferences MUST be 208 performed using the DNS and MUST NOT depend on non-DNS 209 protocols. 211 11. For secure transports using TLS, TLS 1.3 (or later versions) 212 MUST be supported and downgrades from TLS 1.3 to prior versions 213 MUST not occur. 215 5.2. Optional Features 217 1. QNAME minimisation SHOULD be implemented in all steps of 218 recursion 220 2. DNSSEC validation SHOULD be performed 222 3. If an authoritative domain owner or their administrator indicates 223 that (1) multiple secure transport protocols are available, or 224 that (2) a secure transport and insecure transport are available, 225 or that (3) no secure transport is available, then a recursive 226 server SHOULD negotiate selection of an available transport 227 protocol. 229 6. Security Considerations 231 Authoritative name servers will need to perform additional processing 232 steps, such as completing key exchanges and maintaining persistent 233 connections, when responding to queries from a recursive resolver 234 that requests use of a secure transport protocol. These additional 235 processing steps can have an impact on server availability if they 236 are abused. As such, negotiation and use of a secure transport 237 protocol should be done in a manner that does not increase the risk 238 of an authoritative name server outage or lead a recursive server to 239 fail to communicate with an authoritative name server. 241 7. IANA Considerations 243 This document has no actions for IANA. 245 8. Changelog 247 Version 00: Updated prior individual draft following IETF-106 248 feedback Version 01: Small editorial changes Version 02: Incorporate 249 feedback and suggestions from Scott Hollenbeck, Duane Wessels and 250 email discussions. 252 9. APPENDIX: Perspectives and Use Cases 254 The DNS resolving process involves several entities. These entities 255 have different interests/requirements, and hence it does make sense 256 to examine the interests of those entities separately - though in 257 many cases their interests are aligned. Four different entities can 258 be identified, and their interests are described in the following 259 sections: 261 o Users 263 o Operators 265 o Implementors / Software Developers 267 o Researchers 269 9.1. The User Perspective and Use Cases 271 The privacy and confidentiality of Users (that is, users as in 272 clients of recursive resolvers, which in turn forward/resolve the 273 user's DNS requests by contacting authoritative servers) can be 274 improved in several ways. We call this "minimisation of exposure", 275 and there are currently three ways to reduce that exposure: 277 o Qname minimisation [RFC7816], reducing the amount of information 278 to what is absolutely necessary to resolve a query 280 o Aggressive NSEC/local auth cache [RFC8198], reducing the amount of 281 outgoing queries in the first place 283 o Encryption, removing exposure of information while in transit 285 As recursors typically forwards queries received from the user to 286 authoritative servers. This creates a transitive trust between the 287 user and the recursor, as well as the authoritative server, since 288 information created by the user is exposed to the authoritative 289 server. However, the user never has a chance to identify what data 290 was exposed to which authoritative party (via which path). 292 Also, Users would want to be informed about the status of the 293 connections which were made on their behalf, which adds a fourth 294 point 296 Encryption/privacy status signaling 298 *TODO*: Actual requirements - what do users "want"? Start below: 300 9.2. The Operator Perspective and Use Cases 302 Operators of authoritative services have to provide stable and fast 303 DNS services, and interact with a wide range of clients, not all of 304 them authoritative servers. The operator side actually consists of 305 two sides: 307 o The "upstream" facing side of recursive resolvers 309 o The "downstream" side of authoritative servers 311 Those two sides are typically operated by different entities, but 312 many entities operate "both sides". Even though that is discouraged 313 (*TODO* source), the two sides might even be operated on the same 314 nameserver. 316 o Maybe different technical perspectives for operators 318 * Intelligence (sharing information) 320 * SLD popularity for marketing 322 o Focus initially on Second Level Domains (SLDs) initially 324 * Is there a difference for TLDs vs. SLDs from a "protocol" 325 perspective? 327 o Monitoring and aggregated data analysis 329 o Signaling provisioning information 331 * New record type for finding authoritative server key and 332 authentication? Use SRV? (Being able to use different servers 333 for serving up DNS-over-{TCP,UDP} vs DNS-over-TLS responses may 334 be valuable. 336 * Signal secure transport details (DNS-over-TLS, DNS-over-QUIC, 337 EncryptedSNI, connectionless, etc.), perhaps in an extensible 338 manner? Minimize RTTs and reduce need for trials. 340 * Large provider use cases where the NS names are out of 341 bailiwick for the zone (e.g. small number of distinct NS 342 records serving 100k+ zones) 344 o EDNS client subnet (JL: Not sure ECS crosses the cost/benefit 345 threshold to be included as a requirement and many CDNs that run 346 auth servers will likely say ECS is quite operationally important) 348 o Decide between TLS and connectionless (such as COSE-based 349 messages) 351 o Costs of TLS connection vs. connectionless 353 * Technical solution, e.g. encryption of the DNS query, shouldn't 354 enable an attack vector for DDoS or resource exhaustion. For 355 example, only if the client uses DNS-over-TLS, the upstream 356 query to the authoritative will be over DNS-over-TLS also. If 357 the client uses UDP, the resolver won't invest resources in 358 DNS-over-TLS to prevent a potential resource exhaustion attack. 360 * Reuse connection state (if any) and examine resumption 361 considerations 363 * Minimize server-side state (eg, with session tickets) 365 * Need empirical studies on capacity, traffic, attack vectors 367 * Evaluate impact on architecture and footprint expansion 369 * Analyze optimal persistent connection time/time-out 371 * Analyze optimal number of persistent connections recursive 372 resolvers should maintain 374 * Consider operational concerns with respect to capabilities 375 signaling 377 * Develop a profile that has operational advantages for operators 379 *TODO*: Actual requirements - what do operators "want"? 381 9.3. The Implementor / Software Vendor Perspective and Use Cases 383 Implementer requirements follows requirements from user and operator 384 perspectives: 386 o Non-functional requirements, e.g. diversity of implementations 388 o Horizontal vs. vertical scaling, for example similar to http 389 servers 391 o Use of DANE [RFC6698] for authentication: strict vs. opportunistic 393 o Incremental deployment 395 o Cache reuse vs. downgrade? Does the cache need to be partitioned? 396 When can an in-cache answer retrieved via cleartext be served 397 encrypted to a recursive query? 399 o (Use of TCP fast open) - but this might be a requirement for the 400 actual encryption protocol 402 *TODO*: Actual requirements of implementors - essentially, they 403 follow what Operators need? 405 10. References 407 10.1. Normative References 409 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 410 Requirement Levels", BCP 14, RFC 2119, 411 DOI 10.17487/RFC2119, March 1997, 412 . 414 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 415 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 416 May 2017, . 418 [RFC8499] Hoffman, P., Sullivan, A., and K. Fujiwara, "DNS 419 Terminology", BCP 219, RFC 8499, DOI 10.17487/RFC8499, 420 January 2019, . 422 10.2. Informative References 424 [RFC6698] Hoffman, P. and J. Schlyter, "The DNS-Based Authentication 425 of Named Entities (DANE) Transport Layer Security (TLS) 426 Protocol: TLSA", RFC 6698, DOI 10.17487/RFC6698, August 427 2012, . 429 [RFC7816] Bortzmeyer, S., "DNS Query Name Minimisation to Improve 430 Privacy", RFC 7816, DOI 10.17487/RFC7816, March 2016, 431 . 433 [RFC8198] Fujiwara, K., Kato, A., and W. Kumari, "Aggressive Use of 434 DNSSEC-Validated Cache", RFC 8198, DOI 10.17487/RFC8198, 435 July 2017, . 437 10.3. URIs 439 [1] https://datatracker.ietf.org/doc/charter-ietf-dprive/ 441 [2] https://datatracker.ietf.org/wg/dprive/about/ 443 Acknowledgments 445 The authors would like to thank Scott Hollenbeck for his early 446 feedback and providing text for the Internet Draft. We would also 447 like to thank Duane Wessels for the feedback on the mailing list, and 448 Peter van Dijk for his comments in personal conversations. 450 Authors' Addresses 452 Jason Livingood 453 Comcast 455 Email: Jason_Livingood@comcast.com 457 Alexander Mayrhofer 458 nic.at GmbH 460 Email: alex.mayrhofer.ietf@gmail.com 462 Benno Overeinder 463 NLnet Labs 465 Email: benno@NLnetLabs.nl