idnits 2.17.1 draft-ietf-dnsop-kskroll-sentinel-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 : ---------------------------------------------------------------------------- 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 (December 11, 2017) is 2328 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) No issues found here. Summary: 0 errors (**), 0 flaws (~~), 1 warning (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 DNSOP G. Huston 3 Internet-Draft J. Damas 4 Intended status: Standards Track APNIC 5 Expires: June 14, 2018 W. Kumari 6 Google 7 December 11, 2017 9 A Sentinel for Detecting Trusted Keys in DNSSEC 10 draft-ietf-dnsop-kskroll-sentinel-00.txt 12 Abstract 14 The DNS Security Extensions (DNSSEC) were developed to provide origin 15 authentication and integrity protection for DNS data by using digital 16 signatures. These digital signatures can be verified by building a 17 chain of trust starting from a trust anchor and proceeding down to a 18 particular node in the DNS. This document specifies a mechanism that 19 will allow an end user to determine the trusted key state of the 20 resolvers that handle the user's DNS queries. 22 Status of This Memo 24 This Internet-Draft is submitted in full conformance with the 25 provisions of BCP 78 and BCP 79. 27 Internet-Drafts are working documents of the Internet Engineering 28 Task Force (IETF). Note that other groups may also distribute 29 working documents as Internet-Drafts. The list of current Internet- 30 Drafts is at https://datatracker.ietf.org/drafts/current/. 32 Internet-Drafts are draft documents valid for a maximum of six months 33 and may be updated, replaced, or obsoleted by other documents at any 34 time. It is inappropriate to use Internet-Drafts as reference 35 material or to cite them other than as "work in progress." 37 This Internet-Draft will expire on June 14, 2018. 39 Copyright Notice 41 Copyright (c) 2017 IETF Trust and the persons identified as the 42 document authors. All rights reserved. 44 This document is subject to BCP 78 and the IETF Trust's Legal 45 Provisions Relating to IETF Documents 46 (https://trustee.ietf.org/license-info) in effect on the date of 47 publication of this document. Please review these documents 48 carefully, as they describe your rights and restrictions with respect 49 to this document. Code Components extracted from this document must 50 include Simplified BSD License text as described in Section 4.e of 51 the Trust Legal Provisions and are provided without warranty as 52 described in the Simplified BSD License. 54 Table of Contents 56 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 57 1.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . 3 58 2. Sentinel Mechanism . . . . . . . . . . . . . . . . . . . . . 3 59 3. Sentinel Processing . . . . . . . . . . . . . . . . . . . . . 4 60 4. Sentinel Test Result Considerations . . . . . . . . . . . . . 6 61 5. Security Considerations . . . . . . . . . . . . . . . . . . . 7 62 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 7 63 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 7 64 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 7 65 8.1. Normative References . . . . . . . . . . . . . . . . . . 7 66 8.2. Informative References . . . . . . . . . . . . . . . . . 8 67 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 8 69 1. Introduction 71 The DNS Security Extensions (DNSSEC) [RFC4033], [RFC4034] and 72 [RFC4035] were developed to provide origin authentication and 73 integrity protection for DNS data by using digital signatures. 74 DNSSEC uses Key Tags to efficiently match signatures to the keys from 75 which they are generated. The Key Tag is a 16-bit value computed 76 from the RDATA portion of a DNSKEY RR using a formula not unlike a 77 ones-complement checksum. RRSIG RRs contain a Key Tag field whose 78 value is equal to the Key Tag of the DNSKEY RR that validates the 79 signature. 81 This document specifies how validating resolvers can respond to 82 certain queries in a manner that allows a querier to deduce whether a 83 particular key has been loaded into that resolver's trusted key 84 store. In particular, this response mechanism can be used to 85 determine whether a certain Root Zone KSK is ready to be used as a 86 trusted key within the context of a key roll by this resolver. 88 This new mechanism is OPTIONAL to implement and use, although for 89 reasons of supporting broad-based measurement techniques, it is 90 strongly preferred if configurations of DNSSEC-validating resolvers 91 enabled this mechanism by default, allowing for local configuration 92 directives to disable this mechanism if desired. 94 1.1. Terminology 96 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 97 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 98 document are to be interpreted as described in RFC 2119. 100 2. Sentinel Mechanism 102 DNSSEC-Validating resolvers that implement this mechanism MUST be 103 performing validation of responses in accordance with the DNSSEC 104 response validation specification [RFC4035]. 106 This sentinel mechanism makes use of 2 special labels, "_is-ta-." (intended to be used in a query where the response can 108 answer the question: Is this the key tag a trust anchor which the 109 validating DNS resolver is currently trusting?) and "_not-ta-." (intended to be used in a query where the response can 111 answer the question: Is this the key tag of a key that is NOT in the 112 resolver's current trust store?). The use of the positive question 113 and its inverse allows for queries to detect whether resolvers 114 support this sentinel mechanism. 116 If the outcome of the DNSSEC validation process on the response RRset 117 indicates that the response RRset is authentic, and if the left-most 118 label of the original query name matches the template "_is-ta-.", then the following rule should be applied to the response: 120 If the resolver has placed a Root Zone Key Signing Key with tag index 121 value matching the value specified in the query into the local 122 resolver's store of trusted keys, then the resolver should return a 123 response indicating that the response contains authenticated data 124 according to section 5.8 of [RFC6840]. Otherwise, the resolver MUST 125 return RCODE 2 (server failure). Note that the is 126 specified in the DNS label using hexadecimal notation. 128 If the outcome of the DNSSEC validation process aplied to the 129 response RRset indicates that the response RRset is authentic, and if 130 the left-most label of the original query name matches the template 131 "_not-ta-.", then the following rule should be applied to 132 the response: If the resolver has not placed a Root Zone Key Signing 133 Key with tag index value matching the value specified in the query 134 into the local resolver's store of trusted keys, then the resolver 135 should return a response indicating that the response contains 136 authenticated data according to section 5.8 of [RFC6840]. Otherwise, 137 the resolver MUST return RCODE 2 (server failure). Note that the 138 is specified in the DNS label using hexadecimal notation. 140 In all other cases the resolver MUST NOT alter the outcome of the DNS 141 response validation process. 143 This mechanism is to be applied only by resolvers that are performing 144 DNSSEC validation, and applies only to RRset responses to an A or 145 AAAA query (Query Type value 1 or 28) where the resolver has 146 authenticated the response RRset according to the DNSSEC validation 147 process and where the query name contains either of the labels 148 described in this section as its left-most label. In this case, the 149 resolver is to perform an additional test following the conventional 150 validation function, as described in this section. The result of 151 this additional test determines whether the resolver will alter its 152 response that would've indicated that the RRset is authentic to a 153 response that indicates DNSSEC validation failure via the use of 154 RCODE 2. 156 3. Sentinel Processing 158 This proposed test that uses the sentinel detection mechanism 159 described in this document is based on the use of three DNS names 160 that have three distinct DNS resolution behaviours. The test is 161 intended to allow a user to determine the state of their DNS 162 resolution system, and, in particular, whether or not they are using 163 validating DNS resolvers that have picked up an incoming trust anchor 164 as a trusted key in a root zone KSK roll scenario. 166 The name format can be defined in a number of ways, and no name form 167 is intrinsically better than any other in terms of the test itself. 168 The critical aspect of the DNS names used in any such test is that 169 they contain the specified label for either the positive and negative 170 test as the left-most label in the query name. 172 The sentinel detection process is envisaged to use a test with three 173 query names: 175 a. a query name containing the left-most label "_is-ta-.". This corresponds to a a validly-signed RRset in the 177 zone, so that responses associated with queried names in this 178 zone can be authenticated by a DNSSEC-validating resolver. Any 179 validly-signed DNS zone can be used for this test. 181 b. a query name containing the left-most label "_not-ta-.". This is also a validly-signed name. Any validly- 183 signed DNS zone can be used for this test. 185 c. a third query name that is signed with a DNSSEC signature that 186 cannot be validated (i.e. the corresponding RRset is not signed 187 with a valid RRSIG record). 189 The responses received from queries to resolve each of these names 190 would allow us to infer a trust key state of the resolution 191 environment. To describe this process of classification, we can 192 classify resolvers into four distinct behavior types, for which we 193 will use the labels: "Vnew", "Vold", "Vleg", and "nonV". These 194 labels correspond to resolver behaviour types as follows: 196 o Vnew: A DNSSEC-Validating resolver that is configured to implement 197 this mechanism has loaded the nominated key into its local trusted 198 key store will respond with an A or AAAA RRset response for "_is- 199 ta" queries, SERVFAIL for "_not-ta" queries and SERVFAIL for the 200 invalidly signed name queries. 202 o Vold: A DNSSEC-Validating resolver that is configured to implement 203 this mechanism that has not loaded the nominated key into its 204 local trusted key store will respond with an SERVFAIL for "_is-ta" 205 queries, an A or AAAA RRset response for "_not-ta" queries and 206 SERVFAIL for the invalidly signed name queries. 208 o Vleg: A DNSSEC-Validating resolver that does not implement this 209 mechanism will respond with an A or AAAA RRSET response for "_is- 210 ta", an A record response for "_not-ta" and SERVFAIL for the 211 invalid name. 213 o nonV: A non-DNSSEC-Validating resolver will respond with an A 214 record response for "_is-ta", an A record response for "_not-ta" 215 and an A record response for the invalid name. 217 Given the clear delineation amongst these three cases, if a client 218 directs these three queries to a simple resolver, the variation in 219 response to the three queries should allow the client to determine 220 the category of the resolver, and if it supports this mechanism, 221 whether or not it has loaded a particular key into its local trusted 222 key stash. 224 +-------------+----------+-----------+------------+ 225 | Type\Query | _is-ta | _not-ta | invalid | 226 +-------------+----------+-----------+------------+ 227 | Vnew | A | SERVFAIL | SERVFAIL | 228 | Vold | SERVFAIL | A | SERVFAIL | 229 | Vleg | A | A | SERVFAIL | 230 | nonV | A | A | A | 231 +-------------+----------+-----------+------------+ 233 A "Vnew" response pattern says that the nominated key is trusted by 234 the resolver and has been loaded into its local trusted key stash. A 235 "Vold" response pattern says that the nominated key is not yet 236 trusted by the resolver in its own right. A "Vleg" response pattern 237 is indeterminate, and a "nonV" response pattern indicates that the 238 resolver does not perform DNSSEC validation. 240 4. Sentinel Test Result Considerations 242 The description in the previous section describes a simple situation 243 where the test queries were being passed to a single recursive 244 resolver that directly queried authoritative name servers, including 245 the root servers. 247 There is also the common case where the end client is configured to 248 use multiple resolvers. In these cases the SERVFAIL responses from 249 one resolver will prompt the end client to repeat the query against 250 one of the other configured resolvers. 252 If any of the client's resolvers are non-validating resolvers, the 253 tests will result in the client reporting that it has a non- 254 validating DNS environment ("nonV"), which is effectively the case. 256 If all of the client resolvers are DNSSEC-validating resolvers, but 257 some do not support this trusted key mechanism, then the result will 258 be indeterminate with respect to trusted key status ("Vleg"). 259 Simlarly, if all the client's resolvers support this mechanism, but 260 some have loaded the key into the trusted key stash and some have 261 not, then the result is indeterminate ("Vleg"). 263 There is also the common case of a recursive resolver using a 264 forwarder. 266 If the resolver is non-validating, and it has a single forwarder 267 clause, then the resolver will presumably mirror the capabilities of 268 the forwarder target resolver. If this non-validating resolver it 269 has multiple forwarders, then the above considerations will apply. 271 If the validating resolver has a forwarding configuration, and uses 272 the CD flag on all forwarded queries, then this resolver is acting in 273 a manner that is identical to a standalone resolver. The same 274 consideration applies if any one one of the forwarder targets is a 275 non-validating resolver. Similarly, if all the forwarder targets do 276 not apply this trusted key mechanism, the same considerations apply. 278 A more complex case is where the following conditions all hold: 280 both the validating resolver and the forwarder target resolver 281 support this trusted key sentinel mechanism, and 283 the local resolver's queries do not have the CD bit set, and 284 the trusted key state differs between the forwarding resolver and 285 the forwarder target resolver 287 then either the outcome is indeterminate validating ("Vleg"), or a 288 case of mixed signals (SERVFAIL in all three responses), which is 289 similarly an indeterminate response with respect to the trusted key 290 state. 292 5. Security Considerations 294 This document describes a mechanism to allow users to determine the 295 trust state of root zone key signing keys in the DNS resolution 296 system that they use. 298 The mechanism does not require resolvers to set otherwise 299 unauthenticated responses to be marked as authenticated, and does not 300 alter the security properties of DNSSEC with respect to the 301 interpretation of the authenticity of responses that are so marked. 303 The mechanism does not require any further significant processing of 304 DNS responses, and queries of the form described in this document do 305 not impose any additional load that could be exploited in an attack 306 over the the normal DNSSEC validation processing load. 308 6. IANA Considerations 310 [Note to IANA, to be removed prior to publication: there are no IANA 311 considerations stated in this version of the document.] 313 7. Acknowledgements 315 This document has borrowed extensively from [RFC8145] for the 316 introductory text, and the authors would like to acknowledge and 317 thank the authors of that document both for some text excerpts and 318 for the more general stimulation of thoughts about monitoring the 319 progress of a roll of the Key Signing Key of the Root Zone of the 320 DNS. 322 8. References 324 8.1. Normative References 326 [RFC4033] Arends, R., Austein, R., Larson, M., Massey, D., and S. 327 Rose, "DNS Security Introduction and Requirements", 328 RFC 4033, DOI 10.17487/RFC4033, March 2005, 329 . 331 [RFC4034] Arends, R., Austein, R., Larson, M., Massey, D., and S. 332 Rose, "Resource Records for the DNS Security Extensions", 333 RFC 4034, DOI 10.17487/RFC4034, March 2005, 334 . 336 [RFC4035] Arends, R., Austein, R., Larson, M., Massey, D., and S. 337 Rose, "Protocol Modifications for the DNS Security 338 Extensions", RFC 4035, DOI 10.17487/RFC4035, March 2005, 339 . 341 [RFC6840] Weiler, S., Ed. and D. Blacka, Ed., "Clarifications and 342 Implementation Notes for DNS Security (DNSSEC)", RFC 6840, 343 DOI 10.17487/RFC6840, February 2013, 344 . 346 8.2. Informative References 348 [RFC8145] Wessels, D., Kumari, W., and P. Hoffman, "Signaling Trust 349 Anchor Knowledge in DNS Security Extensions (DNSSEC)", 350 RFC 8145, DOI 10.17487/RFC8145, April 2017, 351 . 353 Authors' Addresses 355 Geoff Huston 357 Email: gih@apnic.net 358 URI: http://www.apnic.net 360 Joao Silva Damas 362 Email: joao@apnic.net 363 URI: http://www.apnic.net 365 Warren Kumari 367 Email: warren@kumari.net